{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.2841243512226443,
  "eval_steps": 3000,
  "global_step": 12000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 10.742589855194092,
      "epoch": 0.0005350741077639254,
      "grad_norm": 4.28125,
      "learning_rate": 2e-06,
      "loss": 10.7541,
      "mean_token_accuracy": 0.0,
      "num_tokens": 10038.0,
      "step": 5
    },
    {
      "entropy": 10.742583560943604,
      "epoch": 0.0010701482155278507,
      "grad_norm": 4.5625,
      "learning_rate": 4.5e-06,
      "loss": 10.7779,
      "mean_token_accuracy": 0.0001301236217841506,
      "num_tokens": 21916.0,
      "step": 10
    },
    {
      "entropy": 10.74256067276001,
      "epoch": 0.0016052223232917758,
      "grad_norm": 4.53125,
      "learning_rate": 7e-06,
      "loss": 10.7579,
      "mean_token_accuracy": 0.0,
      "num_tokens": 33434.0,
      "step": 15
    },
    {
      "entropy": 10.742603588104249,
      "epoch": 0.0021402964310557014,
      "grad_norm": 4.4375,
      "learning_rate": 9.5e-06,
      "loss": 10.7083,
      "mean_token_accuracy": 0.0009079659474082291,
      "num_tokens": 44216.0,
      "step": 20
    },
    {
      "entropy": 10.742497825622559,
      "epoch": 0.0026753705388196263,
      "grad_norm": 4.1875,
      "learning_rate": 1.2e-05,
      "loss": 10.6241,
      "mean_token_accuracy": 0.011269020987674594,
      "num_tokens": 54874.0,
      "step": 25
    },
    {
      "entropy": 10.742123413085938,
      "epoch": 0.0032104446465835517,
      "grad_norm": 3.578125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 10.5235,
      "mean_token_accuracy": 0.036691234819591044,
      "num_tokens": 65085.0,
      "step": 30
    },
    {
      "entropy": 10.740595722198487,
      "epoch": 0.003745518754347477,
      "grad_norm": 3.296875,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 10.4,
      "mean_token_accuracy": 0.03980725221335888,
      "num_tokens": 75601.0,
      "step": 35
    },
    {
      "entropy": 10.73592758178711,
      "epoch": 0.004280592862111403,
      "grad_norm": 2.59375,
      "learning_rate": 1.95e-05,
      "loss": 10.305,
      "mean_token_accuracy": 0.04284566193819046,
      "num_tokens": 86193.0,
      "step": 40
    },
    {
      "entropy": 10.72725419998169,
      "epoch": 0.004815666969875327,
      "grad_norm": 2.1875,
      "learning_rate": 2.2e-05,
      "loss": 10.2038,
      "mean_token_accuracy": 0.04451024234294891,
      "num_tokens": 96957.0,
      "step": 45
    },
    {
      "entropy": 10.716819572448731,
      "epoch": 0.005350741077639253,
      "grad_norm": 1.9921875,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 10.1336,
      "mean_token_accuracy": 0.046419617533683774,
      "num_tokens": 108408.0,
      "step": 50
    },
    {
      "entropy": 10.70680923461914,
      "epoch": 0.005885815185403178,
      "grad_norm": 1.5625,
      "learning_rate": 2.7e-05,
      "loss": 10.0839,
      "mean_token_accuracy": 0.04571915306150913,
      "num_tokens": 119273.0,
      "step": 55
    },
    {
      "entropy": 10.699491024017334,
      "epoch": 0.006420889293167103,
      "grad_norm": 1.765625,
      "learning_rate": 2.95e-05,
      "loss": 10.0328,
      "mean_token_accuracy": 0.045481448620557786,
      "num_tokens": 130817.0,
      "step": 60
    },
    {
      "entropy": 10.691621112823487,
      "epoch": 0.006955963400931029,
      "grad_norm": 1.65625,
      "learning_rate": 3.2e-05,
      "loss": 9.9967,
      "mean_token_accuracy": 0.04715100526809692,
      "num_tokens": 142599.0,
      "step": 65
    },
    {
      "entropy": 10.683260250091553,
      "epoch": 0.007491037508694954,
      "grad_norm": 1.734375,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 9.9367,
      "mean_token_accuracy": 0.04639097899198532,
      "num_tokens": 153617.0,
      "step": 70
    },
    {
      "entropy": 10.678304195404053,
      "epoch": 0.00802611161645888,
      "grad_norm": 1.546875,
      "learning_rate": 3.7e-05,
      "loss": 9.9037,
      "mean_token_accuracy": 0.046589919924736024,
      "num_tokens": 165280.0,
      "step": 75
    },
    {
      "entropy": 10.674316120147704,
      "epoch": 0.008561185724222806,
      "grad_norm": 1.421875,
      "learning_rate": 3.95e-05,
      "loss": 9.8497,
      "mean_token_accuracy": 0.045006733387708664,
      "num_tokens": 176896.0,
      "step": 80
    },
    {
      "entropy": 10.66965456008911,
      "epoch": 0.00909625983198673,
      "grad_norm": 1.515625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 9.7994,
      "mean_token_accuracy": 0.04748752862215042,
      "num_tokens": 188000.0,
      "step": 85
    },
    {
      "entropy": 10.65854778289795,
      "epoch": 0.009631333939750655,
      "grad_norm": 1.6640625,
      "learning_rate": 4.45e-05,
      "loss": 9.7526,
      "mean_token_accuracy": 0.04719592090696097,
      "num_tokens": 199360.0,
      "step": 90
    },
    {
      "entropy": 10.643428230285645,
      "epoch": 0.01016640804751458,
      "grad_norm": 1.4375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 9.719,
      "mean_token_accuracy": 0.04324237890541553,
      "num_tokens": 210214.0,
      "step": 95
    },
    {
      "entropy": 10.625392246246339,
      "epoch": 0.010701482155278505,
      "grad_norm": 1.59375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 9.5936,
      "mean_token_accuracy": 0.04677215106785297,
      "num_tokens": 220567.0,
      "step": 100
    },
    {
      "entropy": 10.598260879516602,
      "epoch": 0.011236556263042432,
      "grad_norm": 1.4765625,
      "learning_rate": 5.2e-05,
      "loss": 9.501,
      "mean_token_accuracy": 0.05159270279109478,
      "num_tokens": 230585.0,
      "step": 105
    },
    {
      "entropy": 10.57242784500122,
      "epoch": 0.011771630370806356,
      "grad_norm": 1.5234375,
      "learning_rate": 5.45e-05,
      "loss": 9.4414,
      "mean_token_accuracy": 0.06068843938410282,
      "num_tokens": 241996.0,
      "step": 110
    },
    {
      "entropy": 10.52662353515625,
      "epoch": 0.012306704478570282,
      "grad_norm": 1.4296875,
      "learning_rate": 5.7e-05,
      "loss": 9.4061,
      "mean_token_accuracy": 0.0571962833404541,
      "num_tokens": 253348.0,
      "step": 115
    },
    {
      "entropy": 10.50029878616333,
      "epoch": 0.012841778586334207,
      "grad_norm": 1.515625,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 9.3061,
      "mean_token_accuracy": 0.05508606694638729,
      "num_tokens": 264239.0,
      "step": 120
    },
    {
      "entropy": 10.464248943328858,
      "epoch": 0.013376852694098133,
      "grad_norm": 1.203125,
      "learning_rate": 6.2e-05,
      "loss": 9.2082,
      "mean_token_accuracy": 0.05605855137109757,
      "num_tokens": 274522.0,
      "step": 125
    },
    {
      "entropy": 10.388753318786621,
      "epoch": 0.013911926801862057,
      "grad_norm": 1.5390625,
      "learning_rate": 6.450000000000001e-05,
      "loss": 9.0975,
      "mean_token_accuracy": 0.05725604817271233,
      "num_tokens": 284492.0,
      "step": 130
    },
    {
      "entropy": 10.300678062438966,
      "epoch": 0.014447000909625984,
      "grad_norm": 1.2265625,
      "learning_rate": 6.7e-05,
      "loss": 9.0275,
      "mean_token_accuracy": 0.05757311806082725,
      "num_tokens": 294796.0,
      "step": 135
    },
    {
      "entropy": 10.245001888275146,
      "epoch": 0.014982075017389908,
      "grad_norm": 1.0625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 8.9879,
      "mean_token_accuracy": 0.054746798053383824,
      "num_tokens": 305252.0,
      "step": 140
    },
    {
      "entropy": 10.183552169799805,
      "epoch": 0.015517149125153834,
      "grad_norm": 1.09375,
      "learning_rate": 7.2e-05,
      "loss": 8.863,
      "mean_token_accuracy": 0.0523694459348917,
      "num_tokens": 315505.0,
      "step": 145
    },
    {
      "entropy": 10.06383113861084,
      "epoch": 0.01605222323291776,
      "grad_norm": 1.0625,
      "learning_rate": 7.45e-05,
      "loss": 8.8077,
      "mean_token_accuracy": 0.05920022949576378,
      "num_tokens": 326997.0,
      "step": 150
    },
    {
      "entropy": 9.937602138519287,
      "epoch": 0.016587297340681685,
      "grad_norm": 0.9921875,
      "learning_rate": 7.7e-05,
      "loss": 8.7117,
      "mean_token_accuracy": 0.05458217263221741,
      "num_tokens": 338089.0,
      "step": 155
    },
    {
      "entropy": 9.84811716079712,
      "epoch": 0.01712237144844561,
      "grad_norm": 0.9375,
      "learning_rate": 7.950000000000001e-05,
      "loss": 8.6529,
      "mean_token_accuracy": 0.054475641995668414,
      "num_tokens": 348536.0,
      "step": 160
    },
    {
      "entropy": 9.70736255645752,
      "epoch": 0.017657445556209534,
      "grad_norm": 0.984375,
      "learning_rate": 8.2e-05,
      "loss": 8.6324,
      "mean_token_accuracy": 0.056421706452965736,
      "num_tokens": 359776.0,
      "step": 165
    },
    {
      "entropy": 9.479524993896485,
      "epoch": 0.01819251966397346,
      "grad_norm": 0.8203125,
      "learning_rate": 8.450000000000001e-05,
      "loss": 8.526,
      "mean_token_accuracy": 0.05786490514874458,
      "num_tokens": 370810.0,
      "step": 170
    },
    {
      "entropy": 9.431317043304443,
      "epoch": 0.018727593771737387,
      "grad_norm": 0.8828125,
      "learning_rate": 8.7e-05,
      "loss": 8.4965,
      "mean_token_accuracy": 0.05679596289992332,
      "num_tokens": 381788.0,
      "step": 175
    },
    {
      "entropy": 9.251888465881347,
      "epoch": 0.01926266787950131,
      "grad_norm": 0.78125,
      "learning_rate": 8.95e-05,
      "loss": 8.4589,
      "mean_token_accuracy": 0.05575237199664116,
      "num_tokens": 391984.0,
      "step": 180
    },
    {
      "entropy": 9.08928623199463,
      "epoch": 0.019797741987265235,
      "grad_norm": 0.78125,
      "learning_rate": 9.2e-05,
      "loss": 8.4473,
      "mean_token_accuracy": 0.055337863974273205,
      "num_tokens": 403430.0,
      "step": 185
    },
    {
      "entropy": 9.020852279663085,
      "epoch": 0.02033281609502916,
      "grad_norm": 0.703125,
      "learning_rate": 9.45e-05,
      "loss": 8.4221,
      "mean_token_accuracy": 0.05334745533764362,
      "num_tokens": 414467.0,
      "step": 190
    },
    {
      "entropy": 8.932736873626709,
      "epoch": 0.020867890202793088,
      "grad_norm": 0.671875,
      "learning_rate": 9.7e-05,
      "loss": 8.3858,
      "mean_token_accuracy": 0.05892428457736969,
      "num_tokens": 425752.0,
      "step": 195
    },
    {
      "entropy": 8.8809063911438,
      "epoch": 0.02140296431055701,
      "grad_norm": 0.66796875,
      "learning_rate": 9.95e-05,
      "loss": 8.3505,
      "mean_token_accuracy": 0.06053958684206009,
      "num_tokens": 435908.0,
      "step": 200
    },
    {
      "entropy": 8.834921646118165,
      "epoch": 0.021938038418320937,
      "grad_norm": 0.7421875,
      "learning_rate": 0.000102,
      "loss": 8.3934,
      "mean_token_accuracy": 0.05831273719668388,
      "num_tokens": 446424.0,
      "step": 205
    },
    {
      "entropy": 8.796385478973388,
      "epoch": 0.022473112526084863,
      "grad_norm": 0.73046875,
      "learning_rate": 0.00010449999999999999,
      "loss": 8.3548,
      "mean_token_accuracy": 0.06081846654415131,
      "num_tokens": 456854.0,
      "step": 210
    },
    {
      "entropy": 8.77077407836914,
      "epoch": 0.02300818663384879,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000107,
      "loss": 8.3721,
      "mean_token_accuracy": 0.06371170245110988,
      "num_tokens": 467899.0,
      "step": 215
    },
    {
      "entropy": 8.727867889404298,
      "epoch": 0.023543260741612712,
      "grad_norm": 0.875,
      "learning_rate": 0.0001095,
      "loss": 8.3147,
      "mean_token_accuracy": 0.05818791352212429,
      "num_tokens": 478683.0,
      "step": 220
    },
    {
      "entropy": 8.737598705291749,
      "epoch": 0.02407833484937664,
      "grad_norm": 0.7265625,
      "learning_rate": 0.000112,
      "loss": 8.3471,
      "mean_token_accuracy": 0.05725232698023319,
      "num_tokens": 488745.0,
      "step": 225
    },
    {
      "entropy": 8.692263221740722,
      "epoch": 0.024613408957140565,
      "grad_norm": 0.69140625,
      "learning_rate": 0.0001145,
      "loss": 8.4263,
      "mean_token_accuracy": 0.05620986633002758,
      "num_tokens": 499448.0,
      "step": 230
    },
    {
      "entropy": 8.758592700958252,
      "epoch": 0.02514848306490449,
      "grad_norm": 0.70703125,
      "learning_rate": 0.00011700000000000001,
      "loss": 8.3538,
      "mean_token_accuracy": 0.06146201193332672,
      "num_tokens": 510148.0,
      "step": 235
    },
    {
      "entropy": 8.709297561645508,
      "epoch": 0.025683557172668414,
      "grad_norm": 0.80859375,
      "learning_rate": 0.00011949999999999999,
      "loss": 8.2859,
      "mean_token_accuracy": 0.054982627555727956,
      "num_tokens": 522093.0,
      "step": 240
    },
    {
      "entropy": 8.692859649658203,
      "epoch": 0.02621863128043234,
      "grad_norm": 0.84375,
      "learning_rate": 0.000122,
      "loss": 8.3126,
      "mean_token_accuracy": 0.061951137706637384,
      "num_tokens": 532598.0,
      "step": 245
    },
    {
      "entropy": 8.706014442443848,
      "epoch": 0.026753705388196266,
      "grad_norm": 0.796875,
      "learning_rate": 0.0001245,
      "loss": 8.2807,
      "mean_token_accuracy": 0.05960803255438805,
      "num_tokens": 542652.0,
      "step": 250
    },
    {
      "entropy": 8.617518043518066,
      "epoch": 0.027288779495960192,
      "grad_norm": 0.6875,
      "learning_rate": 0.000127,
      "loss": 8.3078,
      "mean_token_accuracy": 0.06188668571412563,
      "num_tokens": 552526.0,
      "step": 255
    },
    {
      "entropy": 8.682402229309082,
      "epoch": 0.027823853603724115,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0001295,
      "loss": 8.3288,
      "mean_token_accuracy": 0.06049223951995373,
      "num_tokens": 563143.0,
      "step": 260
    },
    {
      "entropy": 8.659179401397704,
      "epoch": 0.02835892771148804,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000132,
      "loss": 8.2969,
      "mean_token_accuracy": 0.060585920512676236,
      "num_tokens": 573814.0,
      "step": 265
    },
    {
      "entropy": 8.617346477508544,
      "epoch": 0.028894001819251967,
      "grad_norm": 0.6875,
      "learning_rate": 0.00013450000000000002,
      "loss": 8.2583,
      "mean_token_accuracy": 0.0664129201322794,
      "num_tokens": 585367.0,
      "step": 270
    },
    {
      "entropy": 8.633615016937256,
      "epoch": 0.02942907592701589,
      "grad_norm": 2.453125,
      "learning_rate": 0.00013700000000000002,
      "loss": 8.3346,
      "mean_token_accuracy": 0.06159670241177082,
      "num_tokens": 595902.0,
      "step": 275
    },
    {
      "entropy": 8.68937931060791,
      "epoch": 0.029964150034779816,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001395,
      "loss": 8.229,
      "mean_token_accuracy": 0.05921008288860321,
      "num_tokens": 607178.0,
      "step": 280
    },
    {
      "entropy": 8.568243312835694,
      "epoch": 0.030499224142543743,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00014199999999999998,
      "loss": 8.2176,
      "mean_token_accuracy": 0.06554121486842632,
      "num_tokens": 618604.0,
      "step": 285
    },
    {
      "entropy": 8.536308574676514,
      "epoch": 0.03103429825030767,
      "grad_norm": 0.84375,
      "learning_rate": 0.0001445,
      "loss": 8.2832,
      "mean_token_accuracy": 0.06573930382728577,
      "num_tokens": 628111.0,
      "step": 290
    },
    {
      "entropy": 8.680528736114502,
      "epoch": 0.03156937235807159,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000147,
      "loss": 8.3211,
      "mean_token_accuracy": 0.06876323223114014,
      "num_tokens": 639358.0,
      "step": 295
    },
    {
      "entropy": 8.595854473114013,
      "epoch": 0.03210444646583552,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0001495,
      "loss": 8.1939,
      "mean_token_accuracy": 0.0666448600590229,
      "num_tokens": 649504.0,
      "step": 300
    },
    {
      "entropy": 8.553579330444336,
      "epoch": 0.032639520573599444,
      "grad_norm": 0.96875,
      "learning_rate": 0.000152,
      "loss": 8.2016,
      "mean_token_accuracy": 0.06504374705255031,
      "num_tokens": 659924.0,
      "step": 305
    },
    {
      "entropy": 8.563442516326905,
      "epoch": 0.03317459468136337,
      "grad_norm": 0.953125,
      "learning_rate": 0.00015450000000000001,
      "loss": 8.2419,
      "mean_token_accuracy": 0.06233549639582634,
      "num_tokens": 671120.0,
      "step": 310
    },
    {
      "entropy": 8.586988162994384,
      "epoch": 0.033709668789127296,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000157,
      "loss": 8.2807,
      "mean_token_accuracy": 0.06455633044242859,
      "num_tokens": 681956.0,
      "step": 315
    },
    {
      "entropy": 8.551072597503662,
      "epoch": 0.03424474289689122,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0001595,
      "loss": 8.2646,
      "mean_token_accuracy": 0.0649121094495058,
      "num_tokens": 693226.0,
      "step": 320
    },
    {
      "entropy": 8.662254047393798,
      "epoch": 0.03477981700465514,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000162,
      "loss": 8.2671,
      "mean_token_accuracy": 0.06917282193899155,
      "num_tokens": 704691.0,
      "step": 325
    },
    {
      "entropy": 8.417178916931153,
      "epoch": 0.03531489111241907,
      "grad_norm": 0.90625,
      "learning_rate": 0.00016450000000000001,
      "loss": 8.2299,
      "mean_token_accuracy": 0.05973314344882965,
      "num_tokens": 716000.0,
      "step": 330
    },
    {
      "entropy": 8.528615379333496,
      "epoch": 0.035849965220182994,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00016700000000000002,
      "loss": 8.1406,
      "mean_token_accuracy": 0.06505260914564133,
      "num_tokens": 726723.0,
      "step": 335
    },
    {
      "entropy": 8.503752326965332,
      "epoch": 0.03638503932794692,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00016950000000000003,
      "loss": 8.1503,
      "mean_token_accuracy": 0.06767830736935139,
      "num_tokens": 736740.0,
      "step": 340
    },
    {
      "entropy": 8.526687812805175,
      "epoch": 0.03692011343571085,
      "grad_norm": 1.6953125,
      "learning_rate": 0.00017199999999999998,
      "loss": 8.211,
      "mean_token_accuracy": 0.06581731289625167,
      "num_tokens": 747567.0,
      "step": 345
    },
    {
      "entropy": 8.503904342651367,
      "epoch": 0.03745518754347477,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00017449999999999999,
      "loss": 8.1534,
      "mean_token_accuracy": 0.06844821311533451,
      "num_tokens": 757937.0,
      "step": 350
    },
    {
      "entropy": 8.430353832244872,
      "epoch": 0.0379902616512387,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000177,
      "loss": 8.1759,
      "mean_token_accuracy": 0.07079985812306404,
      "num_tokens": 768003.0,
      "step": 355
    },
    {
      "entropy": 8.572574520111084,
      "epoch": 0.03852533575900262,
      "grad_norm": 0.796875,
      "learning_rate": 0.0001795,
      "loss": 8.2185,
      "mean_token_accuracy": 0.06302542686462402,
      "num_tokens": 779098.0,
      "step": 360
    },
    {
      "entropy": 8.457394027709961,
      "epoch": 0.039060409866766545,
      "grad_norm": 0.859375,
      "learning_rate": 0.000182,
      "loss": 8.1942,
      "mean_token_accuracy": 0.06275034993886948,
      "num_tokens": 790002.0,
      "step": 365
    },
    {
      "entropy": 8.509715270996093,
      "epoch": 0.03959548397453047,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0001845,
      "loss": 8.1721,
      "mean_token_accuracy": 0.06606314815580845,
      "num_tokens": 800444.0,
      "step": 370
    },
    {
      "entropy": 8.493050575256348,
      "epoch": 0.0401305580822944,
      "grad_norm": 0.75390625,
      "learning_rate": 0.000187,
      "loss": 8.1172,
      "mean_token_accuracy": 0.06942100189626217,
      "num_tokens": 810912.0,
      "step": 375
    },
    {
      "entropy": 8.490777015686035,
      "epoch": 0.04066563219005832,
      "grad_norm": 0.75390625,
      "learning_rate": 0.0001895,
      "loss": 8.2116,
      "mean_token_accuracy": 0.0633488953113556,
      "num_tokens": 822809.0,
      "step": 380
    },
    {
      "entropy": 8.404875659942627,
      "epoch": 0.04120070629782225,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000192,
      "loss": 8.1156,
      "mean_token_accuracy": 0.06941530928015709,
      "num_tokens": 833362.0,
      "step": 385
    },
    {
      "entropy": 8.491305160522462,
      "epoch": 0.041735780405586176,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0001945,
      "loss": 8.1304,
      "mean_token_accuracy": 0.0708375744521618,
      "num_tokens": 844239.0,
      "step": 390
    },
    {
      "entropy": 8.351639652252198,
      "epoch": 0.0422708545133501,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00019700000000000002,
      "loss": 8.1313,
      "mean_token_accuracy": 0.06419948525726796,
      "num_tokens": 855178.0,
      "step": 395
    },
    {
      "entropy": 8.44887342453003,
      "epoch": 0.04280592862111402,
      "grad_norm": 0.90625,
      "learning_rate": 0.00019950000000000002,
      "loss": 8.1241,
      "mean_token_accuracy": 0.06882764175534248,
      "num_tokens": 866120.0,
      "step": 400
    },
    {
      "entropy": 8.451717185974122,
      "epoch": 0.04334100272887795,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000202,
      "loss": 8.1179,
      "mean_token_accuracy": 0.07001868635416031,
      "num_tokens": 876569.0,
      "step": 405
    },
    {
      "entropy": 8.462645149230957,
      "epoch": 0.043876076836641874,
      "grad_norm": 0.921875,
      "learning_rate": 0.00020449999999999998,
      "loss": 8.1782,
      "mean_token_accuracy": 0.05914541408419609,
      "num_tokens": 887580.0,
      "step": 410
    },
    {
      "entropy": 8.390836906433105,
      "epoch": 0.0444111509444058,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000207,
      "loss": 8.1122,
      "mean_token_accuracy": 0.06334667056798934,
      "num_tokens": 897922.0,
      "step": 415
    },
    {
      "entropy": 8.458688259124756,
      "epoch": 0.044946225052169726,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0002095,
      "loss": 8.0675,
      "mean_token_accuracy": 0.06874303668737411,
      "num_tokens": 908522.0,
      "step": 420
    },
    {
      "entropy": 8.417051124572755,
      "epoch": 0.04548129915993365,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000212,
      "loss": 8.0935,
      "mean_token_accuracy": 0.06554699912667275,
      "num_tokens": 918253.0,
      "step": 425
    },
    {
      "entropy": 8.360801124572754,
      "epoch": 0.04601637326769758,
      "grad_norm": 0.73828125,
      "learning_rate": 0.0002145,
      "loss": 8.0775,
      "mean_token_accuracy": 0.07234423086047173,
      "num_tokens": 929150.0,
      "step": 430
    },
    {
      "entropy": 8.457652568817139,
      "epoch": 0.0465514473754615,
      "grad_norm": 0.78125,
      "learning_rate": 0.00021700000000000002,
      "loss": 8.1172,
      "mean_token_accuracy": 0.06926538720726967,
      "num_tokens": 941041.0,
      "step": 435
    },
    {
      "entropy": 8.334060573577881,
      "epoch": 0.047086521483225424,
      "grad_norm": 1.484375,
      "learning_rate": 0.0002195,
      "loss": 8.0663,
      "mean_token_accuracy": 0.07295753248035908,
      "num_tokens": 952714.0,
      "step": 440
    },
    {
      "entropy": 8.448612403869628,
      "epoch": 0.04762159559098935,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000222,
      "loss": 8.1236,
      "mean_token_accuracy": 0.06220442913472653,
      "num_tokens": 964330.0,
      "step": 445
    },
    {
      "entropy": 8.33154468536377,
      "epoch": 0.04815666969875328,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0002245,
      "loss": 8.0483,
      "mean_token_accuracy": 0.06833712309598923,
      "num_tokens": 975257.0,
      "step": 450
    },
    {
      "entropy": 8.419320106506348,
      "epoch": 0.0486917438065172,
      "grad_norm": 0.8203125,
      "learning_rate": 0.00022700000000000002,
      "loss": 8.0769,
      "mean_token_accuracy": 0.06652066558599472,
      "num_tokens": 986335.0,
      "step": 455
    },
    {
      "entropy": 8.301145553588867,
      "epoch": 0.04922681791428113,
      "grad_norm": 0.75,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.9671,
      "mean_token_accuracy": 0.07098020166158676,
      "num_tokens": 997521.0,
      "step": 460
    },
    {
      "entropy": 8.32881212234497,
      "epoch": 0.049761892022045055,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00023200000000000003,
      "loss": 8.0134,
      "mean_token_accuracy": 0.06557687968015671,
      "num_tokens": 1008383.0,
      "step": 465
    },
    {
      "entropy": 8.357664775848388,
      "epoch": 0.05029696612980898,
      "grad_norm": 0.7890625,
      "learning_rate": 0.00023449999999999998,
      "loss": 8.0778,
      "mean_token_accuracy": 0.0682313721626997,
      "num_tokens": 1020552.0,
      "step": 470
    },
    {
      "entropy": 8.34301462173462,
      "epoch": 0.0508320402375729,
      "grad_norm": 0.78125,
      "learning_rate": 0.000237,
      "loss": 8.0512,
      "mean_token_accuracy": 0.07024195492267608,
      "num_tokens": 1031127.0,
      "step": 475
    },
    {
      "entropy": 8.236884784698486,
      "epoch": 0.05136711434533683,
      "grad_norm": 0.74609375,
      "learning_rate": 0.0002395,
      "loss": 7.9657,
      "mean_token_accuracy": 0.07320996709167957,
      "num_tokens": 1043209.0,
      "step": 480
    },
    {
      "entropy": 8.421635818481445,
      "epoch": 0.05190218845310075,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000242,
      "loss": 8.1023,
      "mean_token_accuracy": 0.06352589540183544,
      "num_tokens": 1054032.0,
      "step": 485
    },
    {
      "entropy": 8.233527374267577,
      "epoch": 0.05243726256086468,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0002445,
      "loss": 7.9292,
      "mean_token_accuracy": 0.07238206267356873,
      "num_tokens": 1064565.0,
      "step": 490
    },
    {
      "entropy": 8.403174209594727,
      "epoch": 0.052972336668628606,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000247,
      "loss": 8.0765,
      "mean_token_accuracy": 0.07412495836615562,
      "num_tokens": 1076352.0,
      "step": 495
    },
    {
      "entropy": 8.31511640548706,
      "epoch": 0.05350741077639253,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0002495,
      "loss": 8.0,
      "mean_token_accuracy": 0.0721416525542736,
      "num_tokens": 1087697.0,
      "step": 500
    },
    {
      "entropy": 8.30199728012085,
      "epoch": 0.05404248488415646,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000252,
      "loss": 7.9346,
      "mean_token_accuracy": 0.07342750616371632,
      "num_tokens": 1098056.0,
      "step": 505
    },
    {
      "entropy": 8.233750343322754,
      "epoch": 0.054577558991920384,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0002545,
      "loss": 7.9036,
      "mean_token_accuracy": 0.08062965869903564,
      "num_tokens": 1109219.0,
      "step": 510
    },
    {
      "entropy": 8.323031330108643,
      "epoch": 0.055112633099684304,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000257,
      "loss": 7.987,
      "mean_token_accuracy": 0.06857425346970558,
      "num_tokens": 1120248.0,
      "step": 515
    },
    {
      "entropy": 8.24481725692749,
      "epoch": 0.05564770720744823,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0002595,
      "loss": 8.0086,
      "mean_token_accuracy": 0.07023427784442901,
      "num_tokens": 1131278.0,
      "step": 520
    },
    {
      "entropy": 8.313823032379151,
      "epoch": 0.056182781315212156,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000262,
      "loss": 8.0227,
      "mean_token_accuracy": 0.06274679154157639,
      "num_tokens": 1143147.0,
      "step": 525
    },
    {
      "entropy": 8.287971687316894,
      "epoch": 0.05671785542297608,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.9215,
      "mean_token_accuracy": 0.07099093310534954,
      "num_tokens": 1154316.0,
      "step": 530
    },
    {
      "entropy": 8.246630001068116,
      "epoch": 0.05725292953074001,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.9763,
      "mean_token_accuracy": 0.07050252221524715,
      "num_tokens": 1164696.0,
      "step": 535
    },
    {
      "entropy": 8.212139987945557,
      "epoch": 0.057788003638503935,
      "grad_norm": 0.921875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.9621,
      "mean_token_accuracy": 0.07535928599536419,
      "num_tokens": 1176129.0,
      "step": 540
    },
    {
      "entropy": 8.316623306274414,
      "epoch": 0.05832307774626786,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.9772,
      "mean_token_accuracy": 0.07082564979791642,
      "num_tokens": 1186499.0,
      "step": 545
    },
    {
      "entropy": 8.1876971244812,
      "epoch": 0.05885815185403178,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0002745,
      "loss": 7.9314,
      "mean_token_accuracy": 0.07009159363806247,
      "num_tokens": 1196841.0,
      "step": 550
    },
    {
      "entropy": 8.286906814575195,
      "epoch": 0.059393225961795706,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000277,
      "loss": 7.8784,
      "mean_token_accuracy": 0.07358456589281559,
      "num_tokens": 1207647.0,
      "step": 555
    },
    {
      "entropy": 8.175436544418336,
      "epoch": 0.05992830006955963,
      "grad_norm": 1.0,
      "learning_rate": 0.0002795,
      "loss": 7.8812,
      "mean_token_accuracy": 0.07417330220341682,
      "num_tokens": 1217072.0,
      "step": 560
    },
    {
      "entropy": 8.201017284393311,
      "epoch": 0.06046337417732356,
      "grad_norm": 1.0,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.9853,
      "mean_token_accuracy": 0.07028395347297192,
      "num_tokens": 1228564.0,
      "step": 565
    },
    {
      "entropy": 8.22887988090515,
      "epoch": 0.060998448285087485,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0002845,
      "loss": 7.9124,
      "mean_token_accuracy": 0.072603465244174,
      "num_tokens": 1238755.0,
      "step": 570
    },
    {
      "entropy": 8.181446075439453,
      "epoch": 0.06153352239285141,
      "grad_norm": 0.828125,
      "learning_rate": 0.000287,
      "loss": 7.8745,
      "mean_token_accuracy": 0.07408057898283005,
      "num_tokens": 1248968.0,
      "step": 575
    },
    {
      "entropy": 8.16532106399536,
      "epoch": 0.06206859650061534,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0002895,
      "loss": 7.8461,
      "mean_token_accuracy": 0.07231933623552322,
      "num_tokens": 1259256.0,
      "step": 580
    },
    {
      "entropy": 8.09881591796875,
      "epoch": 0.06260367060837926,
      "grad_norm": 0.984375,
      "learning_rate": 0.000292,
      "loss": 7.8866,
      "mean_token_accuracy": 0.07566695511341096,
      "num_tokens": 1270469.0,
      "step": 585
    },
    {
      "entropy": 8.216452407836915,
      "epoch": 0.06313874471614318,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0002945,
      "loss": 7.8552,
      "mean_token_accuracy": 0.07371087111532688,
      "num_tokens": 1280558.0,
      "step": 590
    },
    {
      "entropy": 8.129891967773437,
      "epoch": 0.06367381882390712,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000297,
      "loss": 7.8999,
      "mean_token_accuracy": 0.06837657615542411,
      "num_tokens": 1291372.0,
      "step": 595
    },
    {
      "entropy": 8.213308429718017,
      "epoch": 0.06420889293167104,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0002995,
      "loss": 7.9753,
      "mean_token_accuracy": 0.06709825992584229,
      "num_tokens": 1302521.0,
      "step": 600
    },
    {
      "entropy": 8.23453779220581,
      "epoch": 0.06474396703943497,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000302,
      "loss": 7.8081,
      "mean_token_accuracy": 0.07309315800666809,
      "num_tokens": 1313632.0,
      "step": 605
    },
    {
      "entropy": 8.05699348449707,
      "epoch": 0.06527904114719889,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003045,
      "loss": 7.8949,
      "mean_token_accuracy": 0.07557445764541626,
      "num_tokens": 1325247.0,
      "step": 610
    },
    {
      "entropy": 8.185565948486328,
      "epoch": 0.06581411525496281,
      "grad_norm": 0.890625,
      "learning_rate": 0.000307,
      "loss": 7.8687,
      "mean_token_accuracy": 0.07852845974266529,
      "num_tokens": 1336026.0,
      "step": 615
    },
    {
      "entropy": 8.095359706878662,
      "epoch": 0.06634918936272674,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0003095,
      "loss": 7.7947,
      "mean_token_accuracy": 0.07415896467864513,
      "num_tokens": 1346920.0,
      "step": 620
    },
    {
      "entropy": 8.091093444824219,
      "epoch": 0.06688426347049066,
      "grad_norm": 1.078125,
      "learning_rate": 0.000312,
      "loss": 7.9316,
      "mean_token_accuracy": 0.0739424929022789,
      "num_tokens": 1359752.0,
      "step": 625
    },
    {
      "entropy": 8.266688632965089,
      "epoch": 0.06741933757825459,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0003145,
      "loss": 7.938,
      "mean_token_accuracy": 0.07053884975612164,
      "num_tokens": 1370874.0,
      "step": 630
    },
    {
      "entropy": 8.154129219055175,
      "epoch": 0.06795441168601851,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000317,
      "loss": 7.8375,
      "mean_token_accuracy": 0.07769538760185242,
      "num_tokens": 1382316.0,
      "step": 635
    },
    {
      "entropy": 8.086437702178955,
      "epoch": 0.06848948579378245,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0003195,
      "loss": 7.8429,
      "mean_token_accuracy": 0.0740839570760727,
      "num_tokens": 1394656.0,
      "step": 640
    },
    {
      "entropy": 8.045439958572388,
      "epoch": 0.06902455990154636,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000322,
      "loss": 7.8192,
      "mean_token_accuracy": 0.07973718792200088,
      "num_tokens": 1405061.0,
      "step": 645
    },
    {
      "entropy": 8.166390562057495,
      "epoch": 0.06955963400931028,
      "grad_norm": 0.87890625,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.7751,
      "mean_token_accuracy": 0.08227450475096702,
      "num_tokens": 1415573.0,
      "step": 650
    },
    {
      "entropy": 8.016403913497925,
      "epoch": 0.07009470811707422,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.8142,
      "mean_token_accuracy": 0.07414669394493104,
      "num_tokens": 1426103.0,
      "step": 655
    },
    {
      "entropy": 8.006370496749877,
      "epoch": 0.07062978222483814,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.8224,
      "mean_token_accuracy": 0.07035662531852722,
      "num_tokens": 1437256.0,
      "step": 660
    },
    {
      "entropy": 8.223912525177003,
      "epoch": 0.07116485633260207,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.7861,
      "mean_token_accuracy": 0.07515646256506443,
      "num_tokens": 1446975.0,
      "step": 665
    },
    {
      "entropy": 8.001112079620361,
      "epoch": 0.07169993044036599,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.8084,
      "mean_token_accuracy": 0.07363461852073669,
      "num_tokens": 1457751.0,
      "step": 670
    },
    {
      "entropy": 8.052605438232423,
      "epoch": 0.07223500454812992,
      "grad_norm": 0.9375,
      "learning_rate": 0.000337,
      "loss": 7.7689,
      "mean_token_accuracy": 0.07661298438906669,
      "num_tokens": 1468186.0,
      "step": 675
    },
    {
      "entropy": 8.094243955612182,
      "epoch": 0.07277007865589384,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0003395,
      "loss": 7.8108,
      "mean_token_accuracy": 0.07768909595906734,
      "num_tokens": 1479780.0,
      "step": 680
    },
    {
      "entropy": 8.057337331771851,
      "epoch": 0.07330515276365776,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000342,
      "loss": 7.7533,
      "mean_token_accuracy": 0.0778307005763054,
      "num_tokens": 1490565.0,
      "step": 685
    },
    {
      "entropy": 8.113437938690186,
      "epoch": 0.0738402268714217,
      "grad_norm": 0.8359375,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.7855,
      "mean_token_accuracy": 0.0767733097076416,
      "num_tokens": 1501552.0,
      "step": 690
    },
    {
      "entropy": 8.02952961921692,
      "epoch": 0.07437530097918561,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000347,
      "loss": 7.7837,
      "mean_token_accuracy": 0.070881437510252,
      "num_tokens": 1513311.0,
      "step": 695
    },
    {
      "entropy": 8.041796684265137,
      "epoch": 0.07491037508694955,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0003495,
      "loss": 7.8081,
      "mean_token_accuracy": 0.07614933587610721,
      "num_tokens": 1524142.0,
      "step": 700
    },
    {
      "entropy": 8.079885721206665,
      "epoch": 0.07544544919471347,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000352,
      "loss": 7.7634,
      "mean_token_accuracy": 0.08070720061659813,
      "num_tokens": 1535571.0,
      "step": 705
    },
    {
      "entropy": 8.080323457717896,
      "epoch": 0.0759805233024774,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0003545,
      "loss": 7.7663,
      "mean_token_accuracy": 0.07677763924002648,
      "num_tokens": 1546528.0,
      "step": 710
    },
    {
      "entropy": 7.932172536849976,
      "epoch": 0.07651559741024132,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000357,
      "loss": 7.7928,
      "mean_token_accuracy": 0.06920730993151665,
      "num_tokens": 1557264.0,
      "step": 715
    },
    {
      "entropy": 8.126523494720459,
      "epoch": 0.07705067151800524,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0003595,
      "loss": 7.7492,
      "mean_token_accuracy": 0.07938187122344971,
      "num_tokens": 1566773.0,
      "step": 720
    },
    {
      "entropy": 7.902943658828735,
      "epoch": 0.07758574562576917,
      "grad_norm": 0.890625,
      "learning_rate": 0.000362,
      "loss": 7.7072,
      "mean_token_accuracy": 0.0833115503191948,
      "num_tokens": 1578173.0,
      "step": 725
    },
    {
      "entropy": 8.02185640335083,
      "epoch": 0.07812081973353309,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0003645,
      "loss": 7.6767,
      "mean_token_accuracy": 0.08211550787091255,
      "num_tokens": 1588537.0,
      "step": 730
    },
    {
      "entropy": 7.918304061889648,
      "epoch": 0.07865589384129702,
      "grad_norm": 0.984375,
      "learning_rate": 0.000367,
      "loss": 7.695,
      "mean_token_accuracy": 0.08645991012454032,
      "num_tokens": 1599159.0,
      "step": 735
    },
    {
      "entropy": 7.982047033309937,
      "epoch": 0.07919096794906094,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0003695,
      "loss": 7.7685,
      "mean_token_accuracy": 0.07947736382484435,
      "num_tokens": 1610378.0,
      "step": 740
    },
    {
      "entropy": 7.9871502876281735,
      "epoch": 0.07972604205682488,
      "grad_norm": 1.03125,
      "learning_rate": 0.000372,
      "loss": 7.7684,
      "mean_token_accuracy": 0.0792300634086132,
      "num_tokens": 1621215.0,
      "step": 745
    },
    {
      "entropy": 7.980308580398559,
      "epoch": 0.0802611161645888,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003745,
      "loss": 7.7133,
      "mean_token_accuracy": 0.07562910728156566,
      "num_tokens": 1631170.0,
      "step": 750
    },
    {
      "entropy": 8.027432918548584,
      "epoch": 0.08079619027235273,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000377,
      "loss": 7.7105,
      "mean_token_accuracy": 0.0776789553463459,
      "num_tokens": 1641897.0,
      "step": 755
    },
    {
      "entropy": 7.974108934402466,
      "epoch": 0.08133126438011665,
      "grad_norm": 0.875,
      "learning_rate": 0.0003795,
      "loss": 7.7762,
      "mean_token_accuracy": 0.07812840193510055,
      "num_tokens": 1653592.0,
      "step": 760
    },
    {
      "entropy": 7.944252681732178,
      "epoch": 0.08186633848788057,
      "grad_norm": 0.921875,
      "learning_rate": 0.000382,
      "loss": 7.7089,
      "mean_token_accuracy": 0.0731413435190916,
      "num_tokens": 1665760.0,
      "step": 765
    },
    {
      "entropy": 7.967342901229858,
      "epoch": 0.0824014125956445,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0003845,
      "loss": 7.7065,
      "mean_token_accuracy": 0.08256332576274872,
      "num_tokens": 1676211.0,
      "step": 770
    },
    {
      "entropy": 7.979905366897583,
      "epoch": 0.08293648670340842,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.7411,
      "mean_token_accuracy": 0.0778872437775135,
      "num_tokens": 1686405.0,
      "step": 775
    },
    {
      "entropy": 8.09218454360962,
      "epoch": 0.08347156081117235,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.7171,
      "mean_token_accuracy": 0.07889893390238285,
      "num_tokens": 1697340.0,
      "step": 780
    },
    {
      "entropy": 7.979829883575439,
      "epoch": 0.08400663491893627,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.6184,
      "mean_token_accuracy": 0.08763319551944733,
      "num_tokens": 1707250.0,
      "step": 785
    },
    {
      "entropy": 7.876405382156372,
      "epoch": 0.0845417090267002,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.5864,
      "mean_token_accuracy": 0.08063115775585175,
      "num_tokens": 1718333.0,
      "step": 790
    },
    {
      "entropy": 7.940855836868286,
      "epoch": 0.08507678313446412,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.6984,
      "mean_token_accuracy": 0.07467846013605595,
      "num_tokens": 1730520.0,
      "step": 795
    },
    {
      "entropy": 7.936916255950928,
      "epoch": 0.08561185724222804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003995,
      "loss": 7.7522,
      "mean_token_accuracy": 0.07317487969994545,
      "num_tokens": 1741898.0,
      "step": 800
    },
    {
      "entropy": 7.9014006614685055,
      "epoch": 0.08614693134999198,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000402,
      "loss": 7.6984,
      "mean_token_accuracy": 0.08368097692728042,
      "num_tokens": 1751671.0,
      "step": 805
    },
    {
      "entropy": 7.829085969924927,
      "epoch": 0.0866820054577559,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004045,
      "loss": 7.7182,
      "mean_token_accuracy": 0.08191005513072014,
      "num_tokens": 1762076.0,
      "step": 810
    },
    {
      "entropy": 8.061158800125122,
      "epoch": 0.08721707956551983,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.685,
      "mean_token_accuracy": 0.07685056664049625,
      "num_tokens": 1772666.0,
      "step": 815
    },
    {
      "entropy": 7.887774229049683,
      "epoch": 0.08775215367328375,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004095,
      "loss": 7.6979,
      "mean_token_accuracy": 0.08384055867791176,
      "num_tokens": 1783621.0,
      "step": 820
    },
    {
      "entropy": 7.9562437534332275,
      "epoch": 0.08828722778104768,
      "grad_norm": 1.390625,
      "learning_rate": 0.000412,
      "loss": 7.7515,
      "mean_token_accuracy": 0.07828780002892018,
      "num_tokens": 1794654.0,
      "step": 825
    },
    {
      "entropy": 7.825522232055664,
      "epoch": 0.0888223018888116,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004145,
      "loss": 7.6368,
      "mean_token_accuracy": 0.08319875784218311,
      "num_tokens": 1805853.0,
      "step": 830
    },
    {
      "entropy": 7.7365336418151855,
      "epoch": 0.08935737599657552,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000417,
      "loss": 7.6227,
      "mean_token_accuracy": 0.0816656306385994,
      "num_tokens": 1817006.0,
      "step": 835
    },
    {
      "entropy": 8.006308937072754,
      "epoch": 0.08989245010433945,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004195,
      "loss": 7.623,
      "mean_token_accuracy": 0.0778016533702612,
      "num_tokens": 1827166.0,
      "step": 840
    },
    {
      "entropy": 8.007753992080689,
      "epoch": 0.09042752421210337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000422,
      "loss": 7.7146,
      "mean_token_accuracy": 0.08085944950580597,
      "num_tokens": 1837182.0,
      "step": 845
    },
    {
      "entropy": 7.7709431648254395,
      "epoch": 0.0909625983198673,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004245,
      "loss": 7.6545,
      "mean_token_accuracy": 0.08275943025946617,
      "num_tokens": 1849034.0,
      "step": 850
    },
    {
      "entropy": 7.842226839065551,
      "epoch": 0.09149767242763122,
      "grad_norm": 0.90625,
      "learning_rate": 0.000427,
      "loss": 7.5383,
      "mean_token_accuracy": 0.08894663602113724,
      "num_tokens": 1859033.0,
      "step": 855
    },
    {
      "entropy": 7.813801717758179,
      "epoch": 0.09203274653539516,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004295,
      "loss": 7.6562,
      "mean_token_accuracy": 0.07978184968233108,
      "num_tokens": 1870274.0,
      "step": 860
    },
    {
      "entropy": 7.786420011520386,
      "epoch": 0.09256782064315908,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000432,
      "loss": 7.5977,
      "mean_token_accuracy": 0.085463797301054,
      "num_tokens": 1881306.0,
      "step": 865
    },
    {
      "entropy": 7.810677576065063,
      "epoch": 0.093102894750923,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004345,
      "loss": 7.5858,
      "mean_token_accuracy": 0.07887605726718902,
      "num_tokens": 1892464.0,
      "step": 870
    },
    {
      "entropy": 7.883417320251465,
      "epoch": 0.09363796885868693,
      "grad_norm": 0.875,
      "learning_rate": 0.000437,
      "loss": 7.6601,
      "mean_token_accuracy": 0.07597660236060619,
      "num_tokens": 1903743.0,
      "step": 875
    },
    {
      "entropy": 7.838293743133545,
      "epoch": 0.09417304296645085,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004395,
      "loss": 7.6112,
      "mean_token_accuracy": 0.08703320100903511,
      "num_tokens": 1915253.0,
      "step": 880
    },
    {
      "entropy": 7.873800992965698,
      "epoch": 0.09470811707421478,
      "grad_norm": 1.15625,
      "learning_rate": 0.000442,
      "loss": 7.5604,
      "mean_token_accuracy": 0.08356410227715969,
      "num_tokens": 1926113.0,
      "step": 885
    },
    {
      "entropy": 7.816686487197876,
      "epoch": 0.0952431911819787,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004445,
      "loss": 7.6579,
      "mean_token_accuracy": 0.08339532688260079,
      "num_tokens": 1937718.0,
      "step": 890
    },
    {
      "entropy": 7.857083654403686,
      "epoch": 0.09577826528974263,
      "grad_norm": 0.875,
      "learning_rate": 0.000447,
      "loss": 7.6419,
      "mean_token_accuracy": 0.07871352508664131,
      "num_tokens": 1948307.0,
      "step": 895
    },
    {
      "entropy": 7.7142290592193605,
      "epoch": 0.09631333939750655,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.4658,
      "mean_token_accuracy": 0.09317997246980667,
      "num_tokens": 1957861.0,
      "step": 900
    },
    {
      "entropy": 7.769102573394775,
      "epoch": 0.09684841350527049,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.5118,
      "mean_token_accuracy": 0.09313379377126693,
      "num_tokens": 1967984.0,
      "step": 905
    },
    {
      "entropy": 7.776102876663208,
      "epoch": 0.0973834876130344,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.531,
      "mean_token_accuracy": 0.08627012819051742,
      "num_tokens": 1978608.0,
      "step": 910
    },
    {
      "entropy": 7.774751234054565,
      "epoch": 0.09791856172079832,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.5571,
      "mean_token_accuracy": 0.08207484185695649,
      "num_tokens": 1989355.0,
      "step": 915
    },
    {
      "entropy": 7.759360122680664,
      "epoch": 0.09845363582856226,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.5519,
      "mean_token_accuracy": 0.09099748730659485,
      "num_tokens": 1999754.0,
      "step": 920
    },
    {
      "entropy": 7.7610972881317135,
      "epoch": 0.09898870993632618,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000462,
      "loss": 7.5582,
      "mean_token_accuracy": 0.08363260142505169,
      "num_tokens": 2010877.0,
      "step": 925
    },
    {
      "entropy": 7.894351291656494,
      "epoch": 0.09952378404409011,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004645,
      "loss": 7.6443,
      "mean_token_accuracy": 0.08178722113370895,
      "num_tokens": 2022301.0,
      "step": 930
    },
    {
      "entropy": 7.720238924026489,
      "epoch": 0.10005885815185403,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000467,
      "loss": 7.5376,
      "mean_token_accuracy": 0.08469678834080696,
      "num_tokens": 2033537.0,
      "step": 935
    },
    {
      "entropy": 7.747068881988525,
      "epoch": 0.10059393225961796,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004695,
      "loss": 7.4595,
      "mean_token_accuracy": 0.08593985140323639,
      "num_tokens": 2044360.0,
      "step": 940
    },
    {
      "entropy": 7.767273759841919,
      "epoch": 0.10112900636738188,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000472,
      "loss": 7.5692,
      "mean_token_accuracy": 0.08553824871778488,
      "num_tokens": 2055000.0,
      "step": 945
    },
    {
      "entropy": 7.675411987304687,
      "epoch": 0.1016640804751458,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004745,
      "loss": 7.5747,
      "mean_token_accuracy": 0.08887513428926468,
      "num_tokens": 2064807.0,
      "step": 950
    },
    {
      "entropy": 7.846766614913941,
      "epoch": 0.10219915458290973,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000477,
      "loss": 7.5338,
      "mean_token_accuracy": 0.08006390556693077,
      "num_tokens": 2075618.0,
      "step": 955
    },
    {
      "entropy": 7.665963554382325,
      "epoch": 0.10273422869067365,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004795,
      "loss": 7.5318,
      "mean_token_accuracy": 0.08242494091391564,
      "num_tokens": 2086226.0,
      "step": 960
    },
    {
      "entropy": 7.758722734451294,
      "epoch": 0.10326930279843759,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000482,
      "loss": 7.5213,
      "mean_token_accuracy": 0.08128740042448043,
      "num_tokens": 2097222.0,
      "step": 965
    },
    {
      "entropy": 7.718167400360107,
      "epoch": 0.1038043769062015,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004845,
      "loss": 7.566,
      "mean_token_accuracy": 0.08480455204844475,
      "num_tokens": 2108919.0,
      "step": 970
    },
    {
      "entropy": 7.67155704498291,
      "epoch": 0.10433945101396544,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000487,
      "loss": 7.5121,
      "mean_token_accuracy": 0.08769758641719819,
      "num_tokens": 2120215.0,
      "step": 975
    },
    {
      "entropy": 7.774061918258667,
      "epoch": 0.10487452512172936,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004895,
      "loss": 7.5377,
      "mean_token_accuracy": 0.08079898655414582,
      "num_tokens": 2131503.0,
      "step": 980
    },
    {
      "entropy": 7.695115470886231,
      "epoch": 0.10540959922949328,
      "grad_norm": 0.921875,
      "learning_rate": 0.000492,
      "loss": 7.4754,
      "mean_token_accuracy": 0.09231993854045868,
      "num_tokens": 2141551.0,
      "step": 985
    },
    {
      "entropy": 7.619196653366089,
      "epoch": 0.10594467333725721,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004945,
      "loss": 7.4805,
      "mean_token_accuracy": 0.0900322899222374,
      "num_tokens": 2151821.0,
      "step": 990
    },
    {
      "entropy": 7.718583822250366,
      "epoch": 0.10647974744502113,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497,
      "loss": 7.5276,
      "mean_token_accuracy": 0.08750699013471604,
      "num_tokens": 2162474.0,
      "step": 995
    },
    {
      "entropy": 7.677374029159546,
      "epoch": 0.10701482155278506,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995,
      "loss": 7.5581,
      "mean_token_accuracy": 0.08569178096950054,
      "num_tokens": 2173985.0,
      "step": 1000
    },
    {
      "entropy": 7.77434720993042,
      "epoch": 0.10754989566054898,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999999979210101,
      "loss": 7.588,
      "mean_token_accuracy": 0.08118958622217179,
      "num_tokens": 2186432.0,
      "step": 1005
    },
    {
      "entropy": 7.720741939544678,
      "epoch": 0.10808496976831292,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999999894751138,
      "loss": 7.4482,
      "mean_token_accuracy": 0.08467482179403304,
      "num_tokens": 2197085.0,
      "step": 1010
    },
    {
      "entropy": 7.661213302612305,
      "epoch": 0.10862004387607684,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999999745323744,
      "loss": 7.4699,
      "mean_token_accuracy": 0.09365503117442131,
      "num_tokens": 2207162.0,
      "step": 1015
    },
    {
      "entropy": 7.673577928543091,
      "epoch": 0.10915511798384077,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999999530927923,
      "loss": 7.5611,
      "mean_token_accuracy": 0.08068385906517506,
      "num_tokens": 2218714.0,
      "step": 1020
    },
    {
      "entropy": 7.6879801750183105,
      "epoch": 0.10969019209160469,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999999251563682,
      "loss": 7.48,
      "mean_token_accuracy": 0.08622352927923202,
      "num_tokens": 2229711.0,
      "step": 1025
    },
    {
      "entropy": 7.641462993621826,
      "epoch": 0.11022526619936861,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999998907231029,
      "loss": 7.4825,
      "mean_token_accuracy": 0.08603335320949554,
      "num_tokens": 2239885.0,
      "step": 1030
    },
    {
      "entropy": 7.723711681365967,
      "epoch": 0.11076034030713254,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999998497929974,
      "loss": 7.4942,
      "mean_token_accuracy": 0.08984629586338996,
      "num_tokens": 2250221.0,
      "step": 1035
    },
    {
      "entropy": 7.636352014541626,
      "epoch": 0.11129541441489646,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999998023660527,
      "loss": 7.5667,
      "mean_token_accuracy": 0.07426251731812954,
      "num_tokens": 2261059.0,
      "step": 1040
    },
    {
      "entropy": 7.627584838867188,
      "epoch": 0.11183048852266039,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004999997484422705,
      "loss": 7.4364,
      "mean_token_accuracy": 0.08743720948696136,
      "num_tokens": 2271174.0,
      "step": 1045
    },
    {
      "entropy": 7.669003200531006,
      "epoch": 0.11236556263042431,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999996880216521,
      "loss": 7.5139,
      "mean_token_accuracy": 0.08555590137839317,
      "num_tokens": 2282431.0,
      "step": 1050
    },
    {
      "entropy": 7.61585578918457,
      "epoch": 0.11290063673818825,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999996211041994,
      "loss": 7.4858,
      "mean_token_accuracy": 0.0835509292781353,
      "num_tokens": 2292891.0,
      "step": 1055
    },
    {
      "entropy": 7.677904653549194,
      "epoch": 0.11343571084595216,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999995476899141,
      "loss": 7.5005,
      "mean_token_accuracy": 0.08647905215620995,
      "num_tokens": 2303764.0,
      "step": 1060
    },
    {
      "entropy": 7.639417934417724,
      "epoch": 0.11397078495371608,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999994677787987,
      "loss": 7.4227,
      "mean_token_accuracy": 0.08917138800024986,
      "num_tokens": 2314172.0,
      "step": 1065
    },
    {
      "entropy": 7.604849863052368,
      "epoch": 0.11450585906148002,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999993813708551,
      "loss": 7.3983,
      "mean_token_accuracy": 0.09371343329548835,
      "num_tokens": 2324848.0,
      "step": 1070
    },
    {
      "entropy": 7.562433481216431,
      "epoch": 0.11504093316924394,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000499999288466086,
      "loss": 7.4262,
      "mean_token_accuracy": 0.0858787901699543,
      "num_tokens": 2336601.0,
      "step": 1075
    },
    {
      "entropy": 7.615593194961548,
      "epoch": 0.11557600727700787,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999991890644941,
      "loss": 7.4123,
      "mean_token_accuracy": 0.08523132391273976,
      "num_tokens": 2347281.0,
      "step": 1080
    },
    {
      "entropy": 7.5736212730407715,
      "epoch": 0.11611108138477179,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999990831660822,
      "loss": 7.4363,
      "mean_token_accuracy": 0.09456784278154373,
      "num_tokens": 2358199.0,
      "step": 1085
    },
    {
      "entropy": 7.702199506759643,
      "epoch": 0.11664615549253572,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999989707708534,
      "loss": 7.4808,
      "mean_token_accuracy": 0.08690382875502109,
      "num_tokens": 2369092.0,
      "step": 1090
    },
    {
      "entropy": 7.544908285140991,
      "epoch": 0.11718122960029964,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999988518788111,
      "loss": 7.4507,
      "mean_token_accuracy": 0.08844347447156906,
      "num_tokens": 2378849.0,
      "step": 1095
    },
    {
      "entropy": 7.622500801086426,
      "epoch": 0.11771630370806356,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999987264899583,
      "loss": 7.4509,
      "mean_token_accuracy": 0.09384607821702957,
      "num_tokens": 2389587.0,
      "step": 1100
    },
    {
      "entropy": 7.658657217025757,
      "epoch": 0.1182513778158275,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000499998594604299,
      "loss": 7.4222,
      "mean_token_accuracy": 0.09075777530670166,
      "num_tokens": 2400442.0,
      "step": 1105
    },
    {
      "entropy": 7.513382863998413,
      "epoch": 0.11878645192359141,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499998456221837,
      "loss": 7.4165,
      "mean_token_accuracy": 0.09110527783632279,
      "num_tokens": 2410771.0,
      "step": 1110
    },
    {
      "entropy": 7.624029874801636,
      "epoch": 0.11932152603135535,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999983113425762,
      "loss": 7.4209,
      "mean_token_accuracy": 0.09688404947519302,
      "num_tokens": 2421112.0,
      "step": 1115
    },
    {
      "entropy": 7.630374145507813,
      "epoch": 0.11985660013911927,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999981599665207,
      "loss": 7.4046,
      "mean_token_accuracy": 0.08740727975964546,
      "num_tokens": 2431891.0,
      "step": 1120
    },
    {
      "entropy": 7.452549457550049,
      "epoch": 0.1203916742468832,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999980020936748,
      "loss": 7.3562,
      "mean_token_accuracy": 0.0897531658411026,
      "num_tokens": 2442987.0,
      "step": 1125
    },
    {
      "entropy": 7.609330129623413,
      "epoch": 0.12092674835464712,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999978377240434,
      "loss": 7.3982,
      "mean_token_accuracy": 0.08765116408467293,
      "num_tokens": 2452703.0,
      "step": 1130
    },
    {
      "entropy": 7.556497859954834,
      "epoch": 0.12146182246241104,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000499997666857631,
      "loss": 7.4509,
      "mean_token_accuracy": 0.09336601346731185,
      "num_tokens": 2462975.0,
      "step": 1135
    },
    {
      "entropy": 7.558685207366944,
      "epoch": 0.12199689657017497,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004999974894944426,
      "loss": 7.3995,
      "mean_token_accuracy": 0.08522969335317612,
      "num_tokens": 2473853.0,
      "step": 1140
    },
    {
      "entropy": 7.575222444534302,
      "epoch": 0.12253197067793889,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999973056344832,
      "loss": 7.3681,
      "mean_token_accuracy": 0.0891667477786541,
      "num_tokens": 2485633.0,
      "step": 1145
    },
    {
      "entropy": 7.465674591064453,
      "epoch": 0.12306704478570282,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999971152777583,
      "loss": 7.3416,
      "mean_token_accuracy": 0.08593353554606438,
      "num_tokens": 2496922.0,
      "step": 1150
    },
    {
      "entropy": 7.701901197433472,
      "epoch": 0.12360211889346674,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004999969184242733,
      "loss": 7.4062,
      "mean_token_accuracy": 0.09027116596698762,
      "num_tokens": 2507890.0,
      "step": 1155
    },
    {
      "entropy": 7.481602525711059,
      "epoch": 0.12413719300123068,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000499996715074034,
      "loss": 7.4012,
      "mean_token_accuracy": 0.0900260142982006,
      "num_tokens": 2519749.0,
      "step": 1160
    },
    {
      "entropy": 7.518426036834716,
      "epoch": 0.1246722671089946,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999965052270461,
      "loss": 7.4146,
      "mean_token_accuracy": 0.08937735855579376,
      "num_tokens": 2530666.0,
      "step": 1165
    },
    {
      "entropy": 7.581199979782104,
      "epoch": 0.12520734121675853,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999962888833157,
      "loss": 7.419,
      "mean_token_accuracy": 0.08899203762412071,
      "num_tokens": 2541594.0,
      "step": 1170
    },
    {
      "entropy": 7.519051456451416,
      "epoch": 0.12574241532452243,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999960660428491,
      "loss": 7.4116,
      "mean_token_accuracy": 0.0924226313829422,
      "num_tokens": 2552330.0,
      "step": 1175
    },
    {
      "entropy": 7.657570695877075,
      "epoch": 0.12627748943228637,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999958367056526,
      "loss": 7.4478,
      "mean_token_accuracy": 0.08449417613446712,
      "num_tokens": 2564601.0,
      "step": 1180
    },
    {
      "entropy": 7.51881742477417,
      "epoch": 0.1268125635400503,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499995600871733,
      "loss": 7.4143,
      "mean_token_accuracy": 0.09234437346458435,
      "num_tokens": 2575539.0,
      "step": 1185
    },
    {
      "entropy": 7.498287916183472,
      "epoch": 0.12734763764781423,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004999953585410971,
      "loss": 7.3524,
      "mean_token_accuracy": 0.08859748020768166,
      "num_tokens": 2586635.0,
      "step": 1190
    },
    {
      "entropy": 7.593882083892822,
      "epoch": 0.12788271175557814,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999951097137518,
      "loss": 7.3818,
      "mean_token_accuracy": 0.09308656603097916,
      "num_tokens": 2596814.0,
      "step": 1195
    },
    {
      "entropy": 7.515226888656616,
      "epoch": 0.12841778586334207,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999948543897044,
      "loss": 7.3996,
      "mean_token_accuracy": 0.08756894245743752,
      "num_tokens": 2608087.0,
      "step": 1200
    },
    {
      "entropy": 7.600381469726562,
      "epoch": 0.128952859971106,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999945925689621,
      "loss": 7.3998,
      "mean_token_accuracy": 0.09394723773002625,
      "num_tokens": 2619377.0,
      "step": 1205
    },
    {
      "entropy": 7.480240154266357,
      "epoch": 0.12948793407886994,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999943242515325,
      "loss": 7.2998,
      "mean_token_accuracy": 0.0932411678135395,
      "num_tokens": 2629089.0,
      "step": 1210
    },
    {
      "entropy": 7.522952890396118,
      "epoch": 0.13002300818663384,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999940494374236,
      "loss": 7.3684,
      "mean_token_accuracy": 0.09049427583813667,
      "num_tokens": 2639743.0,
      "step": 1215
    },
    {
      "entropy": 7.585643148422241,
      "epoch": 0.13055808229439778,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499993768126643,
      "loss": 7.397,
      "mean_token_accuracy": 0.08678129985928536,
      "num_tokens": 2650631.0,
      "step": 1220
    },
    {
      "entropy": 7.525626993179321,
      "epoch": 0.1310931564021617,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499993480319199,
      "loss": 7.366,
      "mean_token_accuracy": 0.09070794358849525,
      "num_tokens": 2660197.0,
      "step": 1225
    },
    {
      "entropy": 7.495192050933838,
      "epoch": 0.13162823050992561,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999931860150999,
      "loss": 7.3461,
      "mean_token_accuracy": 0.08914455473423004,
      "num_tokens": 2670162.0,
      "step": 1230
    },
    {
      "entropy": 7.548467350006104,
      "epoch": 0.13216330461768955,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004999928852143541,
      "loss": 7.3309,
      "mean_token_accuracy": 0.0927185446023941,
      "num_tokens": 2681233.0,
      "step": 1235
    },
    {
      "entropy": 7.398493242263794,
      "epoch": 0.13269837872545348,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999925779169703,
      "loss": 7.2982,
      "mean_token_accuracy": 0.09300850480794906,
      "num_tokens": 2692145.0,
      "step": 1240
    },
    {
      "entropy": 7.526278877258301,
      "epoch": 0.13323345283321741,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999922641229575,
      "loss": 7.3431,
      "mean_token_accuracy": 0.0936188168823719,
      "num_tokens": 2703192.0,
      "step": 1245
    },
    {
      "entropy": 7.520468711853027,
      "epoch": 0.13376852694098132,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999919438323248,
      "loss": 7.3363,
      "mean_token_accuracy": 0.09251101464033126,
      "num_tokens": 2714804.0,
      "step": 1250
    },
    {
      "entropy": 7.4432690143585205,
      "epoch": 0.13430360104874525,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999916170450812,
      "loss": 7.3068,
      "mean_token_accuracy": 0.09502202421426773,
      "num_tokens": 2724340.0,
      "step": 1255
    },
    {
      "entropy": 7.475179243087768,
      "epoch": 0.13483867515650919,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999912837612364,
      "loss": 7.3071,
      "mean_token_accuracy": 0.09392300024628639,
      "num_tokens": 2735767.0,
      "step": 1260
    },
    {
      "entropy": 7.569590282440186,
      "epoch": 0.1353737492642731,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999909439807997,
      "loss": 7.3779,
      "mean_token_accuracy": 0.08663228303194045,
      "num_tokens": 2747658.0,
      "step": 1265
    },
    {
      "entropy": 7.444280576705933,
      "epoch": 0.13590882337203702,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999905977037813,
      "loss": 7.2372,
      "mean_token_accuracy": 0.09776455760002137,
      "num_tokens": 2757901.0,
      "step": 1270
    },
    {
      "entropy": 7.465328884124756,
      "epoch": 0.13644389747980096,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999902449301909,
      "loss": 7.2776,
      "mean_token_accuracy": 0.09301821067929268,
      "num_tokens": 2768945.0,
      "step": 1275
    },
    {
      "entropy": 7.314613389968872,
      "epoch": 0.1369789715875649,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004999898856600387,
      "loss": 7.2582,
      "mean_token_accuracy": 0.1052941657602787,
      "num_tokens": 2780276.0,
      "step": 1280
    },
    {
      "entropy": 7.536267852783203,
      "epoch": 0.1375140456953288,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999895198933354,
      "loss": 7.3323,
      "mean_token_accuracy": 0.09816832616925239,
      "num_tokens": 2791183.0,
      "step": 1285
    },
    {
      "entropy": 7.4270100593566895,
      "epoch": 0.13804911980309273,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999891476300911,
      "loss": 7.2426,
      "mean_token_accuracy": 0.09529703482985497,
      "num_tokens": 2801928.0,
      "step": 1290
    },
    {
      "entropy": 7.473166513442993,
      "epoch": 0.13858419391085666,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499988768870317,
      "loss": 7.3295,
      "mean_token_accuracy": 0.08986801654100418,
      "num_tokens": 2813415.0,
      "step": 1295
    },
    {
      "entropy": 7.48001651763916,
      "epoch": 0.13911926801862057,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999883836140236,
      "loss": 7.3364,
      "mean_token_accuracy": 0.09272250309586524,
      "num_tokens": 2825288.0,
      "step": 1300
    },
    {
      "entropy": 7.421426439285279,
      "epoch": 0.1396543421263845,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999879918612224,
      "loss": 7.2918,
      "mean_token_accuracy": 0.096827931702137,
      "num_tokens": 2836298.0,
      "step": 1305
    },
    {
      "entropy": 7.523803615570069,
      "epoch": 0.14018941623414843,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999875936119245,
      "loss": 7.2778,
      "mean_token_accuracy": 0.0943089857697487,
      "num_tokens": 2846367.0,
      "step": 1310
    },
    {
      "entropy": 7.3315290927886965,
      "epoch": 0.14072449034191237,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999871888661415,
      "loss": 7.2799,
      "mean_token_accuracy": 0.09142087027430534,
      "num_tokens": 2856832.0,
      "step": 1315
    },
    {
      "entropy": 7.507561445236206,
      "epoch": 0.14125956444967627,
      "grad_norm": 0.84375,
      "learning_rate": 0.000499986777623885,
      "loss": 7.3088,
      "mean_token_accuracy": 0.0900144100189209,
      "num_tokens": 2867954.0,
      "step": 1320
    },
    {
      "entropy": 7.509339809417725,
      "epoch": 0.1417946385574402,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999863598851669,
      "loss": 7.3334,
      "mean_token_accuracy": 0.09359355568885804,
      "num_tokens": 2878714.0,
      "step": 1325
    },
    {
      "entropy": 7.325873184204101,
      "epoch": 0.14232971266520414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999859356499992,
      "loss": 7.28,
      "mean_token_accuracy": 0.09667187854647637,
      "num_tokens": 2890697.0,
      "step": 1330
    },
    {
      "entropy": 7.467297649383545,
      "epoch": 0.14286478677296804,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999855049183943,
      "loss": 7.292,
      "mean_token_accuracy": 0.09200986102223396,
      "num_tokens": 2901468.0,
      "step": 1335
    },
    {
      "entropy": 7.420497417449951,
      "epoch": 0.14339986088073198,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999850676903646,
      "loss": 7.256,
      "mean_token_accuracy": 0.09620093181729317,
      "num_tokens": 2913466.0,
      "step": 1340
    },
    {
      "entropy": 7.381121587753296,
      "epoch": 0.1439349349884959,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999846239659226,
      "loss": 7.2224,
      "mean_token_accuracy": 0.10079324766993522,
      "num_tokens": 2923638.0,
      "step": 1345
    },
    {
      "entropy": 7.451461362838745,
      "epoch": 0.14447000909625984,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999841737450812,
      "loss": 7.2969,
      "mean_token_accuracy": 0.1001754067838192,
      "num_tokens": 2934790.0,
      "step": 1350
    },
    {
      "entropy": 7.389082574844361,
      "epoch": 0.14500508320402375,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999837170278535,
      "loss": 7.248,
      "mean_token_accuracy": 0.0918399602174759,
      "num_tokens": 2946320.0,
      "step": 1355
    },
    {
      "entropy": 7.486172676086426,
      "epoch": 0.14554015731178768,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999832538142526,
      "loss": 7.2238,
      "mean_token_accuracy": 0.091259303689003,
      "num_tokens": 2957077.0,
      "step": 1360
    },
    {
      "entropy": 7.342933654785156,
      "epoch": 0.14607523141955162,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999827841042917,
      "loss": 7.2296,
      "mean_token_accuracy": 0.09233908951282502,
      "num_tokens": 2967286.0,
      "step": 1365
    },
    {
      "entropy": 7.4447362422943115,
      "epoch": 0.14661030552731552,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999823078979846,
      "loss": 7.2744,
      "mean_token_accuracy": 0.09223584830760956,
      "num_tokens": 2979516.0,
      "step": 1370
    },
    {
      "entropy": 7.485718393325806,
      "epoch": 0.14714537963507945,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999818251953449,
      "loss": 7.399,
      "mean_token_accuracy": 0.08901449739933014,
      "num_tokens": 2991594.0,
      "step": 1375
    },
    {
      "entropy": 7.445301294326782,
      "epoch": 0.1476804537428434,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999813359963867,
      "loss": 7.286,
      "mean_token_accuracy": 0.09267361760139466,
      "num_tokens": 3002487.0,
      "step": 1380
    },
    {
      "entropy": 7.488003301620483,
      "epoch": 0.14821552785060732,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999808403011241,
      "loss": 7.2851,
      "mean_token_accuracy": 0.09769331067800521,
      "num_tokens": 3013197.0,
      "step": 1385
    },
    {
      "entropy": 7.3263664722442625,
      "epoch": 0.14875060195837123,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999803381095712,
      "loss": 7.3105,
      "mean_token_accuracy": 0.09277946949005127,
      "num_tokens": 3024512.0,
      "step": 1390
    },
    {
      "entropy": 7.529593801498413,
      "epoch": 0.14928567606613516,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999798294217428,
      "loss": 7.2724,
      "mean_token_accuracy": 0.09323014095425605,
      "num_tokens": 3035621.0,
      "step": 1395
    },
    {
      "entropy": 7.383570957183838,
      "epoch": 0.1498207501738991,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999793142376533,
      "loss": 7.2703,
      "mean_token_accuracy": 0.09348074868321418,
      "num_tokens": 3046492.0,
      "step": 1400
    },
    {
      "entropy": 7.38762936592102,
      "epoch": 0.150355824281663,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999787925573177,
      "loss": 7.2831,
      "mean_token_accuracy": 0.09317544624209403,
      "num_tokens": 3057715.0,
      "step": 1405
    },
    {
      "entropy": 7.511090230941773,
      "epoch": 0.15089089838942693,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999782643807513,
      "loss": 7.2558,
      "mean_token_accuracy": 0.0898089624941349,
      "num_tokens": 3068344.0,
      "step": 1410
    },
    {
      "entropy": 7.340856122970581,
      "epoch": 0.15142597249719086,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999777297079689,
      "loss": 7.2191,
      "mean_token_accuracy": 0.0933483250439167,
      "num_tokens": 3079576.0,
      "step": 1415
    },
    {
      "entropy": 7.457035493850708,
      "epoch": 0.1519610466049548,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999771885389863,
      "loss": 7.2075,
      "mean_token_accuracy": 0.09207480028271675,
      "num_tokens": 3090201.0,
      "step": 1420
    },
    {
      "entropy": 7.333426856994629,
      "epoch": 0.1524961207127187,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999766408738189,
      "loss": 7.172,
      "mean_token_accuracy": 0.10389826074242592,
      "num_tokens": 3099824.0,
      "step": 1425
    },
    {
      "entropy": 7.333724737167358,
      "epoch": 0.15303119482048264,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999760867124827,
      "loss": 7.2405,
      "mean_token_accuracy": 0.09329254552721977,
      "num_tokens": 3109948.0,
      "step": 1430
    },
    {
      "entropy": 7.4186999797821045,
      "epoch": 0.15356626892824657,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004999755260549937,
      "loss": 7.2413,
      "mean_token_accuracy": 0.09903448224067687,
      "num_tokens": 3121492.0,
      "step": 1435
    },
    {
      "entropy": 7.313020372390747,
      "epoch": 0.15410134303601047,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004999749589013677,
      "loss": 7.0092,
      "mean_token_accuracy": 0.11508586779236793,
      "num_tokens": 3131990.0,
      "step": 1440
    },
    {
      "entropy": 7.2625288486480715,
      "epoch": 0.1546364171437744,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999743852516217,
      "loss": 7.2038,
      "mean_token_accuracy": 0.09737641140818595,
      "num_tokens": 3142567.0,
      "step": 1445
    },
    {
      "entropy": 7.4858119010925295,
      "epoch": 0.15517149125153834,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999738051057717,
      "loss": 7.2253,
      "mean_token_accuracy": 0.09757704138755799,
      "num_tokens": 3153083.0,
      "step": 1450
    },
    {
      "entropy": 7.336223459243774,
      "epoch": 0.15570656535930227,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999732184638347,
      "loss": 7.2224,
      "mean_token_accuracy": 0.09092248976230621,
      "num_tokens": 3163985.0,
      "step": 1455
    },
    {
      "entropy": 7.436727666854859,
      "epoch": 0.15624163946706618,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999726253258278,
      "loss": 7.2157,
      "mean_token_accuracy": 0.09606200903654098,
      "num_tokens": 3175259.0,
      "step": 1460
    },
    {
      "entropy": 7.293716049194336,
      "epoch": 0.1567767135748301,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999720256917678,
      "loss": 7.1804,
      "mean_token_accuracy": 0.09885367155075073,
      "num_tokens": 3185487.0,
      "step": 1465
    },
    {
      "entropy": 7.31433138847351,
      "epoch": 0.15731178768259405,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999714195616723,
      "loss": 7.1717,
      "mean_token_accuracy": 0.10157929807901382,
      "num_tokens": 3196200.0,
      "step": 1470
    },
    {
      "entropy": 7.441909456253052,
      "epoch": 0.15784686179035795,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999708069355586,
      "loss": 7.2733,
      "mean_token_accuracy": 0.08783909082412719,
      "num_tokens": 3207675.0,
      "step": 1475
    },
    {
      "entropy": 7.31417498588562,
      "epoch": 0.15838193589812188,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999701878134445,
      "loss": 7.1471,
      "mean_token_accuracy": 0.09791682288050652,
      "num_tokens": 3218117.0,
      "step": 1480
    },
    {
      "entropy": 7.3960308074951175,
      "epoch": 0.15891701000588582,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999695621953477,
      "loss": 7.2164,
      "mean_token_accuracy": 0.10362501963973045,
      "num_tokens": 3229208.0,
      "step": 1485
    },
    {
      "entropy": 7.309657001495362,
      "epoch": 0.15945208411364975,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999689300812866,
      "loss": 7.1036,
      "mean_token_accuracy": 0.10788998156785964,
      "num_tokens": 3239586.0,
      "step": 1490
    },
    {
      "entropy": 7.392857885360717,
      "epoch": 0.15998715822141366,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004999682914712791,
      "loss": 7.1983,
      "mean_token_accuracy": 0.09886570796370506,
      "num_tokens": 3250103.0,
      "step": 1495
    },
    {
      "entropy": 7.331405687332153,
      "epoch": 0.1605222323291776,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999676463653439,
      "loss": 7.1785,
      "mean_token_accuracy": 0.10397682413458824,
      "num_tokens": 3261844.0,
      "step": 1500
    },
    {
      "entropy": 7.361588764190674,
      "epoch": 0.16105730643694152,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999669947634996,
      "loss": 7.1436,
      "mean_token_accuracy": 0.0984138660132885,
      "num_tokens": 3272073.0,
      "step": 1505
    },
    {
      "entropy": 7.343551349639893,
      "epoch": 0.16159238054470546,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999663366657647,
      "loss": 7.1261,
      "mean_token_accuracy": 0.1005841538310051,
      "num_tokens": 3281725.0,
      "step": 1510
    },
    {
      "entropy": 7.326076507568359,
      "epoch": 0.16212745465246936,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999656720721586,
      "loss": 7.2726,
      "mean_token_accuracy": 0.0962200105190277,
      "num_tokens": 3292708.0,
      "step": 1515
    },
    {
      "entropy": 7.3782445907592775,
      "epoch": 0.1626625287602333,
      "grad_norm": 0.875,
      "learning_rate": 0.0004999650009827004,
      "loss": 7.2564,
      "mean_token_accuracy": 0.09442284628748894,
      "num_tokens": 3303548.0,
      "step": 1520
    },
    {
      "entropy": 7.326221704483032,
      "epoch": 0.16319760286799723,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999643233974092,
      "loss": 7.117,
      "mean_token_accuracy": 0.10348235741257668,
      "num_tokens": 3314261.0,
      "step": 1525
    },
    {
      "entropy": 7.2751476764678955,
      "epoch": 0.16373267697576113,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999636393163049,
      "loss": 7.1271,
      "mean_token_accuracy": 0.10080837979912757,
      "num_tokens": 3324833.0,
      "step": 1530
    },
    {
      "entropy": 7.2360046863555905,
      "epoch": 0.16426775108352507,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999629487394071,
      "loss": 7.036,
      "mean_token_accuracy": 0.11096755266189576,
      "num_tokens": 3334790.0,
      "step": 1535
    },
    {
      "entropy": 7.339627885818482,
      "epoch": 0.164802825191289,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999622516667358,
      "loss": 7.1602,
      "mean_token_accuracy": 0.09869524762034416,
      "num_tokens": 3345115.0,
      "step": 1540
    },
    {
      "entropy": 7.268951654434204,
      "epoch": 0.16533789929905293,
      "grad_norm": 0.7890625,
      "learning_rate": 0.000499961548098311,
      "loss": 7.0592,
      "mean_token_accuracy": 0.10297729372978211,
      "num_tokens": 3356187.0,
      "step": 1545
    },
    {
      "entropy": 7.235326194763184,
      "epoch": 0.16587297340681684,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004999608380341533,
      "loss": 7.0744,
      "mean_token_accuracy": 0.10658099949359894,
      "num_tokens": 3367218.0,
      "step": 1550
    },
    {
      "entropy": 7.27059588432312,
      "epoch": 0.16640804751458077,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999601214742829,
      "loss": 7.2223,
      "mean_token_accuracy": 0.09261110052466393,
      "num_tokens": 3377557.0,
      "step": 1555
    },
    {
      "entropy": 7.427518892288208,
      "epoch": 0.1669431216223447,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004999593984187206,
      "loss": 7.2095,
      "mean_token_accuracy": 0.10194377526640892,
      "num_tokens": 3387402.0,
      "step": 1560
    },
    {
      "entropy": 7.248392057418823,
      "epoch": 0.1674781957301086,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004999586688674872,
      "loss": 7.0941,
      "mean_token_accuracy": 0.10278604626655578,
      "num_tokens": 3397924.0,
      "step": 1565
    },
    {
      "entropy": 7.3578520774841305,
      "epoch": 0.16801326983787254,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000499957932820604,
      "loss": 7.1743,
      "mean_token_accuracy": 0.09474834129214287,
      "num_tokens": 3408622.0,
      "step": 1570
    },
    {
      "entropy": 7.343464088439942,
      "epoch": 0.16854834394563648,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499957190278092,
      "loss": 7.1367,
      "mean_token_accuracy": 0.10084123015403748,
      "num_tokens": 3419692.0,
      "step": 1575
    },
    {
      "entropy": 7.277517509460449,
      "epoch": 0.1690834180534004,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999564412399728,
      "loss": 7.1842,
      "mean_token_accuracy": 0.09453171491622925,
      "num_tokens": 3430986.0,
      "step": 1580
    },
    {
      "entropy": 7.289031267166138,
      "epoch": 0.1696184921611643,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499955685706268,
      "loss": 7.1056,
      "mean_token_accuracy": 0.10056664869189262,
      "num_tokens": 3442211.0,
      "step": 1585
    },
    {
      "entropy": 7.181107234954834,
      "epoch": 0.17015356626892825,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999549236769993,
      "loss": 7.0973,
      "mean_token_accuracy": 0.10596697479486465,
      "num_tokens": 3453585.0,
      "step": 1590
    },
    {
      "entropy": 7.261494874954224,
      "epoch": 0.17068864037669218,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999541551521888,
      "loss": 7.1281,
      "mean_token_accuracy": 0.10282958298921585,
      "num_tokens": 3464156.0,
      "step": 1595
    },
    {
      "entropy": 7.3783262252807615,
      "epoch": 0.17122371448445609,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999533801318587,
      "loss": 7.1901,
      "mean_token_accuracy": 0.095138331502676,
      "num_tokens": 3474853.0,
      "step": 1600
    },
    {
      "entropy": 7.3372804641723635,
      "epoch": 0.17175878859222002,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999525986160313,
      "loss": 7.1634,
      "mean_token_accuracy": 0.10265101641416549,
      "num_tokens": 3486841.0,
      "step": 1605
    },
    {
      "entropy": 7.3480212688446045,
      "epoch": 0.17229386269998395,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004999518106047293,
      "loss": 7.1702,
      "mean_token_accuracy": 0.0980787456035614,
      "num_tokens": 3498735.0,
      "step": 1610
    },
    {
      "entropy": 7.253286647796631,
      "epoch": 0.17282893680774788,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004999510160979754,
      "loss": 7.2008,
      "mean_token_accuracy": 0.10094587206840515,
      "num_tokens": 3508829.0,
      "step": 1615
    },
    {
      "entropy": 7.2795453548431395,
      "epoch": 0.1733640109155118,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999502150957925,
      "loss": 7.1209,
      "mean_token_accuracy": 0.10228936895728111,
      "num_tokens": 3519465.0,
      "step": 1620
    },
    {
      "entropy": 7.299279642105103,
      "epoch": 0.17389908502327572,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999494075982037,
      "loss": 7.0647,
      "mean_token_accuracy": 0.10867529138922691,
      "num_tokens": 3530401.0,
      "step": 1625
    },
    {
      "entropy": 7.235185194015503,
      "epoch": 0.17443415913103966,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999485936052324,
      "loss": 7.141,
      "mean_token_accuracy": 0.10203565210103989,
      "num_tokens": 3541136.0,
      "step": 1630
    },
    {
      "entropy": 7.373658609390259,
      "epoch": 0.17496923323880356,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499947773116902,
      "loss": 7.1322,
      "mean_token_accuracy": 0.1018851414322853,
      "num_tokens": 3552525.0,
      "step": 1635
    },
    {
      "entropy": 7.22913932800293,
      "epoch": 0.1755043073465675,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999469461332365,
      "loss": 7.0903,
      "mean_token_accuracy": 0.10703189074993133,
      "num_tokens": 3563028.0,
      "step": 1640
    },
    {
      "entropy": 7.259391117095947,
      "epoch": 0.17603938145433143,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999461126542592,
      "loss": 7.0522,
      "mean_token_accuracy": 0.10235123336315155,
      "num_tokens": 3573354.0,
      "step": 1645
    },
    {
      "entropy": 7.158776950836182,
      "epoch": 0.17657445556209536,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004999452726799947,
      "loss": 7.0373,
      "mean_token_accuracy": 0.10366476848721504,
      "num_tokens": 3585576.0,
      "step": 1650
    },
    {
      "entropy": 7.320888566970825,
      "epoch": 0.17710952966985927,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999444262104671,
      "loss": 7.1598,
      "mean_token_accuracy": 0.09593332335352897,
      "num_tokens": 3596478.0,
      "step": 1655
    },
    {
      "entropy": 7.289003372192383,
      "epoch": 0.1776446037776232,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999435732457007,
      "loss": 7.123,
      "mean_token_accuracy": 0.09818345233798027,
      "num_tokens": 3608014.0,
      "step": 1660
    },
    {
      "entropy": 7.291224193572998,
      "epoch": 0.17817967788538713,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999427137857203,
      "loss": 7.1694,
      "mean_token_accuracy": 0.09814337715506553,
      "num_tokens": 3620120.0,
      "step": 1665
    },
    {
      "entropy": 7.274828290939331,
      "epoch": 0.17871475199315104,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999418478305506,
      "loss": 7.0748,
      "mean_token_accuracy": 0.09978235512971878,
      "num_tokens": 3632578.0,
      "step": 1670
    },
    {
      "entropy": 7.185247135162354,
      "epoch": 0.17924982610091497,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999409753802167,
      "loss": 7.0939,
      "mean_token_accuracy": 0.10724005252122878,
      "num_tokens": 3643245.0,
      "step": 1675
    },
    {
      "entropy": 7.27445502281189,
      "epoch": 0.1797849002086789,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999400964347437,
      "loss": 7.0967,
      "mean_token_accuracy": 0.10385240912437439,
      "num_tokens": 3655832.0,
      "step": 1680
    },
    {
      "entropy": 7.2620195865631105,
      "epoch": 0.18031997431644284,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999392109941571,
      "loss": 7.0683,
      "mean_token_accuracy": 0.10449369996786118,
      "num_tokens": 3666566.0,
      "step": 1685
    },
    {
      "entropy": 7.303769207000732,
      "epoch": 0.18085504842420674,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999383190584822,
      "loss": 7.0852,
      "mean_token_accuracy": 0.10249918699264526,
      "num_tokens": 3677572.0,
      "step": 1690
    },
    {
      "entropy": 7.2038905143737795,
      "epoch": 0.18139012253197068,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999374206277451,
      "loss": 7.0113,
      "mean_token_accuracy": 0.10162734165787697,
      "num_tokens": 3687116.0,
      "step": 1695
    },
    {
      "entropy": 7.175737571716309,
      "epoch": 0.1819251966397346,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999365157019717,
      "loss": 7.1189,
      "mean_token_accuracy": 0.10182299092411995,
      "num_tokens": 3697961.0,
      "step": 1700
    },
    {
      "entropy": 7.309969711303711,
      "epoch": 0.18246027074749852,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999356042811878,
      "loss": 7.0444,
      "mean_token_accuracy": 0.10850266963243485,
      "num_tokens": 3708834.0,
      "step": 1705
    },
    {
      "entropy": 7.182546806335449,
      "epoch": 0.18299534485526245,
      "grad_norm": 0.84375,
      "learning_rate": 0.00049993468636542,
      "loss": 7.0991,
      "mean_token_accuracy": 0.10176495090126991,
      "num_tokens": 3719835.0,
      "step": 1710
    },
    {
      "entropy": 7.250284337997437,
      "epoch": 0.18353041896302638,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999337619546948,
      "loss": 7.0009,
      "mean_token_accuracy": 0.10658760368824005,
      "num_tokens": 3730408.0,
      "step": 1715
    },
    {
      "entropy": 7.171570301055908,
      "epoch": 0.18406549307079031,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999328310490387,
      "loss": 7.0695,
      "mean_token_accuracy": 0.10707555189728737,
      "num_tokens": 3741257.0,
      "step": 1720
    },
    {
      "entropy": 7.23467698097229,
      "epoch": 0.18460056717855422,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999318936484789,
      "loss": 7.0647,
      "mean_token_accuracy": 0.10648670792579651,
      "num_tokens": 3752615.0,
      "step": 1725
    },
    {
      "entropy": 7.214528036117554,
      "epoch": 0.18513564128631815,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999309497530422,
      "loss": 7.0949,
      "mean_token_accuracy": 0.09849463924765586,
      "num_tokens": 3763656.0,
      "step": 1730
    },
    {
      "entropy": 7.150198459625244,
      "epoch": 0.1856707153940821,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004999299993627558,
      "loss": 7.0733,
      "mean_token_accuracy": 0.10403963699936866,
      "num_tokens": 3775257.0,
      "step": 1735
    },
    {
      "entropy": 7.201783800125122,
      "epoch": 0.186205789501846,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999290424776475,
      "loss": 7.0384,
      "mean_token_accuracy": 0.11034373492002487,
      "num_tokens": 3786094.0,
      "step": 1740
    },
    {
      "entropy": 7.355918598175049,
      "epoch": 0.18674086360960993,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999280790977445,
      "loss": 7.1286,
      "mean_token_accuracy": 0.09715545624494552,
      "num_tokens": 3796672.0,
      "step": 1745
    },
    {
      "entropy": 7.305685806274414,
      "epoch": 0.18727593771737386,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499927109223075,
      "loss": 7.0958,
      "mean_token_accuracy": 0.10413563027977943,
      "num_tokens": 3807910.0,
      "step": 1750
    },
    {
      "entropy": 7.1276421546936035,
      "epoch": 0.1878110118251378,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004999261328536667,
      "loss": 7.0578,
      "mean_token_accuracy": 0.10628265142440796,
      "num_tokens": 3820792.0,
      "step": 1755
    },
    {
      "entropy": 7.157390975952149,
      "epoch": 0.1883460859329017,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999251499895479,
      "loss": 6.9844,
      "mean_token_accuracy": 0.10621865093708038,
      "num_tokens": 3831126.0,
      "step": 1760
    },
    {
      "entropy": 7.23041090965271,
      "epoch": 0.18888116004066563,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999241606307471,
      "loss": 7.0942,
      "mean_token_accuracy": 0.10165138021111489,
      "num_tokens": 3841923.0,
      "step": 1765
    },
    {
      "entropy": 7.243416452407837,
      "epoch": 0.18941623414842956,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999231647772927,
      "loss": 7.0078,
      "mean_token_accuracy": 0.10644997358322143,
      "num_tokens": 3852504.0,
      "step": 1770
    },
    {
      "entropy": 7.0957677364349365,
      "epoch": 0.1899513082561935,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004999221624292136,
      "loss": 7.0087,
      "mean_token_accuracy": 0.1103954017162323,
      "num_tokens": 3862761.0,
      "step": 1775
    },
    {
      "entropy": 7.252035474777221,
      "epoch": 0.1904863823639574,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004999211535865385,
      "loss": 7.0234,
      "mean_token_accuracy": 0.1052067756652832,
      "num_tokens": 3873662.0,
      "step": 1780
    },
    {
      "entropy": 7.154085254669189,
      "epoch": 0.19102145647172133,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999201382492969,
      "loss": 6.9621,
      "mean_token_accuracy": 0.10425644144415855,
      "num_tokens": 3884835.0,
      "step": 1785
    },
    {
      "entropy": 7.187242269515991,
      "epoch": 0.19155653057948527,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999191164175178,
      "loss": 7.1166,
      "mean_token_accuracy": 0.10480258762836456,
      "num_tokens": 3895073.0,
      "step": 1790
    },
    {
      "entropy": 7.2901567935943605,
      "epoch": 0.19209160468724917,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999180880912309,
      "loss": 7.0556,
      "mean_token_accuracy": 0.10658993124961853,
      "num_tokens": 3905440.0,
      "step": 1795
    },
    {
      "entropy": 7.176698493957519,
      "epoch": 0.1926266787950131,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999170532704657,
      "loss": 7.0712,
      "mean_token_accuracy": 0.10166745483875275,
      "num_tokens": 3917135.0,
      "step": 1800
    },
    {
      "entropy": 7.2208339214324955,
      "epoch": 0.19316175290277704,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004999160119552523,
      "loss": 7.0716,
      "mean_token_accuracy": 0.10915103554725647,
      "num_tokens": 3928247.0,
      "step": 1805
    },
    {
      "entropy": 7.223503684997558,
      "epoch": 0.19369682701054097,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999149641456206,
      "loss": 7.0491,
      "mean_token_accuracy": 0.1053008608520031,
      "num_tokens": 3938609.0,
      "step": 1810
    },
    {
      "entropy": 7.265676212310791,
      "epoch": 0.19423190111830488,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499913909841601,
      "loss": 7.0274,
      "mean_token_accuracy": 0.10771157816052437,
      "num_tokens": 3949221.0,
      "step": 1815
    },
    {
      "entropy": 7.100171852111816,
      "epoch": 0.1947669752260688,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999128490432238,
      "loss": 7.0149,
      "mean_token_accuracy": 0.09949721470475197,
      "num_tokens": 3959475.0,
      "step": 1820
    },
    {
      "entropy": 7.136339998245239,
      "epoch": 0.19530204933383274,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999117817505196,
      "loss": 7.0138,
      "mean_token_accuracy": 0.11632085666060447,
      "num_tokens": 3969638.0,
      "step": 1825
    },
    {
      "entropy": 7.178588724136352,
      "epoch": 0.19583712344159665,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004999107079635194,
      "loss": 7.0644,
      "mean_token_accuracy": 0.10715136379003525,
      "num_tokens": 3980543.0,
      "step": 1830
    },
    {
      "entropy": 7.336812162399292,
      "epoch": 0.19637219754936058,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000499909627682254,
      "loss": 7.1143,
      "mean_token_accuracy": 0.0920241579413414,
      "num_tokens": 3993518.0,
      "step": 1835
    },
    {
      "entropy": 7.216370439529419,
      "epoch": 0.19690727165712452,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999085409067549,
      "loss": 7.0948,
      "mean_token_accuracy": 0.10310237035155297,
      "num_tokens": 4002940.0,
      "step": 1840
    },
    {
      "entropy": 7.141087865829467,
      "epoch": 0.19744234576488845,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004999074476370531,
      "loss": 7.0291,
      "mean_token_accuracy": 0.10522487610578538,
      "num_tokens": 4013982.0,
      "step": 1845
    },
    {
      "entropy": 7.2277580261230465,
      "epoch": 0.19797741987265235,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999063478731805,
      "loss": 6.9933,
      "mean_token_accuracy": 0.11192195788025856,
      "num_tokens": 4025607.0,
      "step": 1850
    },
    {
      "entropy": 7.117806482315063,
      "epoch": 0.1985124939804163,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999052416151687,
      "loss": 7.0568,
      "mean_token_accuracy": 0.10609380379319192,
      "num_tokens": 4037467.0,
      "step": 1855
    },
    {
      "entropy": 7.221655082702637,
      "epoch": 0.19904756808818022,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004999041288630496,
      "loss": 7.065,
      "mean_token_accuracy": 0.10668069496750832,
      "num_tokens": 4049146.0,
      "step": 1860
    },
    {
      "entropy": 7.2340247631073,
      "epoch": 0.19958264219594413,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004999030096168553,
      "loss": 7.1116,
      "mean_token_accuracy": 0.10125580132007599,
      "num_tokens": 4060500.0,
      "step": 1865
    },
    {
      "entropy": 7.248100090026855,
      "epoch": 0.20011771630370806,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999018838766182,
      "loss": 7.0427,
      "mean_token_accuracy": 0.10293570905923843,
      "num_tokens": 4070553.0,
      "step": 1870
    },
    {
      "entropy": 7.2189253807067875,
      "epoch": 0.200652790411472,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499900751642371,
      "loss": 6.988,
      "mean_token_accuracy": 0.1170131392776966,
      "num_tokens": 4079653.0,
      "step": 1875
    },
    {
      "entropy": 7.100331258773804,
      "epoch": 0.20118786451923593,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998996129141461,
      "loss": 7.0654,
      "mean_token_accuracy": 0.10362578108906746,
      "num_tokens": 4089912.0,
      "step": 1880
    },
    {
      "entropy": 7.275002813339233,
      "epoch": 0.20172293862699983,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998984676919764,
      "loss": 7.0342,
      "mean_token_accuracy": 0.10268538147211075,
      "num_tokens": 4100577.0,
      "step": 1885
    },
    {
      "entropy": 7.187543249130249,
      "epoch": 0.20225801273476376,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998973159758952,
      "loss": 7.0658,
      "mean_token_accuracy": 0.10256308168172837,
      "num_tokens": 4111752.0,
      "step": 1890
    },
    {
      "entropy": 7.10661129951477,
      "epoch": 0.2027930868425277,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998961577659355,
      "loss": 7.015,
      "mean_token_accuracy": 0.10711082518100738,
      "num_tokens": 4122616.0,
      "step": 1895
    },
    {
      "entropy": 7.239635181427002,
      "epoch": 0.2033281609502916,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004998949930621309,
      "loss": 7.1135,
      "mean_token_accuracy": 0.10389721244573594,
      "num_tokens": 4134275.0,
      "step": 1900
    },
    {
      "entropy": 7.159951686859131,
      "epoch": 0.20386323505805554,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998938218645149,
      "loss": 7.0093,
      "mean_token_accuracy": 0.10267628878355026,
      "num_tokens": 4146412.0,
      "step": 1905
    },
    {
      "entropy": 7.137516021728516,
      "epoch": 0.20439830916581947,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004998926441731215,
      "loss": 7.0056,
      "mean_token_accuracy": 0.10260871946811675,
      "num_tokens": 4157592.0,
      "step": 1910
    },
    {
      "entropy": 7.170009613037109,
      "epoch": 0.2049333832735834,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998914599879845,
      "loss": 6.9763,
      "mean_token_accuracy": 0.1078723892569542,
      "num_tokens": 4168637.0,
      "step": 1915
    },
    {
      "entropy": 7.15439133644104,
      "epoch": 0.2054684573813473,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998902693091382,
      "loss": 7.0316,
      "mean_token_accuracy": 0.10298030972480773,
      "num_tokens": 4179951.0,
      "step": 1920
    },
    {
      "entropy": 7.146353483200073,
      "epoch": 0.20600353148911124,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998890721366171,
      "loss": 6.9259,
      "mean_token_accuracy": 0.1077031098306179,
      "num_tokens": 4189895.0,
      "step": 1925
    },
    {
      "entropy": 7.1884785175323485,
      "epoch": 0.20653860559687517,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998878684704556,
      "loss": 6.9636,
      "mean_token_accuracy": 0.11009967923164368,
      "num_tokens": 4199796.0,
      "step": 1930
    },
    {
      "entropy": 7.076934337615967,
      "epoch": 0.20707367970463908,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998866583106884,
      "loss": 7.0095,
      "mean_token_accuracy": 0.10127623230218888,
      "num_tokens": 4210428.0,
      "step": 1935
    },
    {
      "entropy": 7.256487798690796,
      "epoch": 0.207608753812403,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998854416573506,
      "loss": 7.0649,
      "mean_token_accuracy": 0.1079666055738926,
      "num_tokens": 4221515.0,
      "step": 1940
    },
    {
      "entropy": 7.144673681259155,
      "epoch": 0.20814382792016695,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998842185104772,
      "loss": 7.0776,
      "mean_token_accuracy": 0.11020539179444314,
      "num_tokens": 4232580.0,
      "step": 1945
    },
    {
      "entropy": 7.1092174530029295,
      "epoch": 0.20867890202793088,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998829888701036,
      "loss": 6.9893,
      "mean_token_accuracy": 0.11271198317408562,
      "num_tokens": 4242871.0,
      "step": 1950
    },
    {
      "entropy": 7.213183069229126,
      "epoch": 0.20921397613569478,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998817527362654,
      "loss": 7.0261,
      "mean_token_accuracy": 0.10301245972514153,
      "num_tokens": 4253419.0,
      "step": 1955
    },
    {
      "entropy": 7.153800678253174,
      "epoch": 0.20974905024345872,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004998805101089982,
      "loss": 6.9999,
      "mean_token_accuracy": 0.10977082848548889,
      "num_tokens": 4264327.0,
      "step": 1960
    },
    {
      "entropy": 7.214887237548828,
      "epoch": 0.21028412435122265,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998792609883378,
      "loss": 6.9766,
      "mean_token_accuracy": 0.10818460211157799,
      "num_tokens": 4274568.0,
      "step": 1965
    },
    {
      "entropy": 7.140443277359009,
      "epoch": 0.21081919845898656,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998780053743203,
      "loss": 6.9731,
      "mean_token_accuracy": 0.11215656697750091,
      "num_tokens": 4284895.0,
      "step": 1970
    },
    {
      "entropy": 7.1276960372924805,
      "epoch": 0.2113542725667505,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998767432669822,
      "loss": 7.0104,
      "mean_token_accuracy": 0.10507805719971657,
      "num_tokens": 4297258.0,
      "step": 1975
    },
    {
      "entropy": 7.149711656570434,
      "epoch": 0.21188934667451442,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998754746663595,
      "loss": 6.9365,
      "mean_token_accuracy": 0.10693479850888252,
      "num_tokens": 4308307.0,
      "step": 1980
    },
    {
      "entropy": 7.124268484115601,
      "epoch": 0.21242442078227836,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004998741995724892,
      "loss": 7.0014,
      "mean_token_accuracy": 0.10538375228643418,
      "num_tokens": 4318801.0,
      "step": 1985
    },
    {
      "entropy": 7.139152193069458,
      "epoch": 0.21295949489004226,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998729179854079,
      "loss": 6.9777,
      "mean_token_accuracy": 0.10666713044047356,
      "num_tokens": 4329531.0,
      "step": 1990
    },
    {
      "entropy": 7.1325037479400635,
      "epoch": 0.2134945689978062,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998716299051527,
      "loss": 6.9611,
      "mean_token_accuracy": 0.11172771900892257,
      "num_tokens": 4341203.0,
      "step": 1995
    },
    {
      "entropy": 7.091319608688354,
      "epoch": 0.21402964310557013,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998703353317609,
      "loss": 7.0092,
      "mean_token_accuracy": 0.10446273684501647,
      "num_tokens": 4352426.0,
      "step": 2000
    },
    {
      "entropy": 7.161164569854736,
      "epoch": 0.21456471721333403,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004998690342652697,
      "loss": 6.9493,
      "mean_token_accuracy": 0.11236951574683189,
      "num_tokens": 4363413.0,
      "step": 2005
    },
    {
      "entropy": 7.104535961151123,
      "epoch": 0.21509979132109797,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998677267057166,
      "loss": 6.8913,
      "mean_token_accuracy": 0.11243644654750824,
      "num_tokens": 4373470.0,
      "step": 2010
    },
    {
      "entropy": 7.114715242385865,
      "epoch": 0.2156348654288619,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004998664126531397,
      "loss": 6.9596,
      "mean_token_accuracy": 0.1114291749894619,
      "num_tokens": 4384438.0,
      "step": 2015
    },
    {
      "entropy": 7.101155853271484,
      "epoch": 0.21616993953662583,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998650921075766,
      "loss": 6.9454,
      "mean_token_accuracy": 0.10823875814676284,
      "num_tokens": 4395100.0,
      "step": 2020
    },
    {
      "entropy": 7.117280292510986,
      "epoch": 0.21670501364438974,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004998637650690656,
      "loss": 7.0544,
      "mean_token_accuracy": 0.10239638835191726,
      "num_tokens": 4407131.0,
      "step": 2025
    },
    {
      "entropy": 7.1200460433959964,
      "epoch": 0.21724008775215367,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499862431537645,
      "loss": 6.8888,
      "mean_token_accuracy": 0.10876853093504905,
      "num_tokens": 4417701.0,
      "step": 2030
    },
    {
      "entropy": 7.097174835205078,
      "epoch": 0.2177751618599176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998610915133533,
      "loss": 6.949,
      "mean_token_accuracy": 0.11059459373354912,
      "num_tokens": 4429309.0,
      "step": 2035
    },
    {
      "entropy": 7.08780870437622,
      "epoch": 0.21831023596768154,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998597449962292,
      "loss": 7.0251,
      "mean_token_accuracy": 0.10617428198456764,
      "num_tokens": 4440850.0,
      "step": 2040
    },
    {
      "entropy": 7.156359386444092,
      "epoch": 0.21884531007544544,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998583919863115,
      "loss": 7.0009,
      "mean_token_accuracy": 0.10219309031963349,
      "num_tokens": 4451995.0,
      "step": 2045
    },
    {
      "entropy": 6.987979745864868,
      "epoch": 0.21938038418320938,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998570324836393,
      "loss": 6.9059,
      "mean_token_accuracy": 0.11296729370951653,
      "num_tokens": 4461916.0,
      "step": 2050
    },
    {
      "entropy": 7.093602132797241,
      "epoch": 0.2199154582909733,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499855666488252,
      "loss": 6.9953,
      "mean_token_accuracy": 0.1073579765856266,
      "num_tokens": 4471603.0,
      "step": 2055
    },
    {
      "entropy": 7.243840980529785,
      "epoch": 0.22045053239873721,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004998542940001889,
      "loss": 7.0352,
      "mean_token_accuracy": 0.10500341430306434,
      "num_tokens": 4483166.0,
      "step": 2060
    },
    {
      "entropy": 7.074850082397461,
      "epoch": 0.22098560650650115,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004998529150194895,
      "loss": 6.9424,
      "mean_token_accuracy": 0.1041251115500927,
      "num_tokens": 4493824.0,
      "step": 2065
    },
    {
      "entropy": 7.116635465621949,
      "epoch": 0.22152068061426508,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998515295461939,
      "loss": 6.837,
      "mean_token_accuracy": 0.1261841081082821,
      "num_tokens": 4505674.0,
      "step": 2070
    },
    {
      "entropy": 7.0095867156982425,
      "epoch": 0.22205575472202901,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499850137580342,
      "loss": 6.9422,
      "mean_token_accuracy": 0.11334002017974854,
      "num_tokens": 4516516.0,
      "step": 2075
    },
    {
      "entropy": 7.119809246063232,
      "epoch": 0.22259082882979292,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998487391219739,
      "loss": 6.9453,
      "mean_token_accuracy": 0.10056448429822921,
      "num_tokens": 4527811.0,
      "step": 2080
    },
    {
      "entropy": 7.050966739654541,
      "epoch": 0.22312590293755685,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004998473341711301,
      "loss": 6.8806,
      "mean_token_accuracy": 0.10502795204520225,
      "num_tokens": 4538988.0,
      "step": 2085
    },
    {
      "entropy": 7.119756889343262,
      "epoch": 0.22366097704532079,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998459227278512,
      "loss": 6.8997,
      "mean_token_accuracy": 0.11361236721277237,
      "num_tokens": 4550049.0,
      "step": 2090
    },
    {
      "entropy": 7.177942180633545,
      "epoch": 0.2241960511530847,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998445047921777,
      "loss": 7.0332,
      "mean_token_accuracy": 0.10892037376761436,
      "num_tokens": 4559784.0,
      "step": 2095
    },
    {
      "entropy": 7.094767713546753,
      "epoch": 0.22473112526084862,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998430803641509,
      "loss": 6.873,
      "mean_token_accuracy": 0.11253026425838471,
      "num_tokens": 4569982.0,
      "step": 2100
    },
    {
      "entropy": 6.99695782661438,
      "epoch": 0.22526619936861256,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998416494438118,
      "loss": 6.8769,
      "mean_token_accuracy": 0.11211913973093032,
      "num_tokens": 4581117.0,
      "step": 2105
    },
    {
      "entropy": 7.084139013290406,
      "epoch": 0.2258012734763765,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998402120312014,
      "loss": 6.9462,
      "mean_token_accuracy": 0.11116094216704368,
      "num_tokens": 4592012.0,
      "step": 2110
    },
    {
      "entropy": 7.116655254364014,
      "epoch": 0.2263363475841404,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998387681263616,
      "loss": 6.9931,
      "mean_token_accuracy": 0.10694418773055077,
      "num_tokens": 4602591.0,
      "step": 2115
    },
    {
      "entropy": 7.098616075515747,
      "epoch": 0.22687142169190433,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000499837317729334,
      "loss": 6.9291,
      "mean_token_accuracy": 0.10939419046044349,
      "num_tokens": 4613418.0,
      "step": 2120
    },
    {
      "entropy": 7.091100645065308,
      "epoch": 0.22740649579966826,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004998358608401603,
      "loss": 6.8499,
      "mean_token_accuracy": 0.11751434430480004,
      "num_tokens": 4625156.0,
      "step": 2125
    },
    {
      "entropy": 7.1228173732757565,
      "epoch": 0.22794156990743217,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499834397458883,
      "loss": 6.8865,
      "mean_token_accuracy": 0.11367959231138229,
      "num_tokens": 4634523.0,
      "step": 2130
    },
    {
      "entropy": 6.990410232543946,
      "epoch": 0.2284766440151961,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998329275855438,
      "loss": 6.87,
      "mean_token_accuracy": 0.11853560954332351,
      "num_tokens": 4646292.0,
      "step": 2135
    },
    {
      "entropy": 7.091177034378052,
      "epoch": 0.22901171812296003,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998314512201855,
      "loss": 6.8548,
      "mean_token_accuracy": 0.10857467800378799,
      "num_tokens": 4656417.0,
      "step": 2140
    },
    {
      "entropy": 7.000166225433349,
      "epoch": 0.22954679223072397,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998299683628507,
      "loss": 6.8661,
      "mean_token_accuracy": 0.11203439608216285,
      "num_tokens": 4666933.0,
      "step": 2145
    },
    {
      "entropy": 7.122287797927856,
      "epoch": 0.23008186633848787,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004998284790135821,
      "loss": 7.0464,
      "mean_token_accuracy": 0.10003090128302575,
      "num_tokens": 4678482.0,
      "step": 2150
    },
    {
      "entropy": 7.111646604537964,
      "epoch": 0.2306169404462518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998269831724228,
      "loss": 6.894,
      "mean_token_accuracy": 0.11034037619829178,
      "num_tokens": 4688454.0,
      "step": 2155
    },
    {
      "entropy": 7.016481971740722,
      "epoch": 0.23115201455401574,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004998254808394159,
      "loss": 6.9201,
      "mean_token_accuracy": 0.11088593229651451,
      "num_tokens": 4699545.0,
      "step": 2160
    },
    {
      "entropy": 7.125952577590942,
      "epoch": 0.23168708866177964,
      "grad_norm": 0.7578125,
      "learning_rate": 0.0004998239720146048,
      "loss": 6.9256,
      "mean_token_accuracy": 0.11005692854523659,
      "num_tokens": 4711149.0,
      "step": 2165
    },
    {
      "entropy": 6.9970673561096195,
      "epoch": 0.23222216276954358,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004998224566980332,
      "loss": 6.9387,
      "mean_token_accuracy": 0.10672764554619789,
      "num_tokens": 4722651.0,
      "step": 2170
    },
    {
      "entropy": 7.063665342330933,
      "epoch": 0.2327572368773075,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004998209348897447,
      "loss": 6.9388,
      "mean_token_accuracy": 0.10191527009010315,
      "num_tokens": 4733883.0,
      "step": 2175
    },
    {
      "entropy": 7.244954586029053,
      "epoch": 0.23329231098507144,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004998194065897833,
      "loss": 6.9158,
      "mean_token_accuracy": 0.10730345025658608,
      "num_tokens": 4744517.0,
      "step": 2180
    },
    {
      "entropy": 7.010083484649658,
      "epoch": 0.23382738509283535,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499817871798193,
      "loss": 6.9234,
      "mean_token_accuracy": 0.1180235929787159,
      "num_tokens": 4755481.0,
      "step": 2185
    },
    {
      "entropy": 7.012146234512329,
      "epoch": 0.23436245920059928,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004998163305150185,
      "loss": 6.8316,
      "mean_token_accuracy": 0.11478903293609619,
      "num_tokens": 4765535.0,
      "step": 2190
    },
    {
      "entropy": 7.069972801208496,
      "epoch": 0.23489753330836322,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004998147827403038,
      "loss": 6.9619,
      "mean_token_accuracy": 0.10954375863075257,
      "num_tokens": 4777045.0,
      "step": 2195
    },
    {
      "entropy": 7.119951629638672,
      "epoch": 0.23543260741612712,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499813228474094,
      "loss": 6.8399,
      "mean_token_accuracy": 0.1167965590953827,
      "num_tokens": 4787958.0,
      "step": 2200
    },
    {
      "entropy": 6.953368234634399,
      "epoch": 0.23596768152389105,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004998116677164338,
      "loss": 6.8911,
      "mean_token_accuracy": 0.10928420349955559,
      "num_tokens": 4798502.0,
      "step": 2205
    },
    {
      "entropy": 7.026491498947143,
      "epoch": 0.236502755631655,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004998101004673682,
      "loss": 6.8906,
      "mean_token_accuracy": 0.11567377597093582,
      "num_tokens": 4809179.0,
      "step": 2210
    },
    {
      "entropy": 7.102079772949219,
      "epoch": 0.23703782973941892,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998085267269426,
      "loss": 6.9476,
      "mean_token_accuracy": 0.10375262200832366,
      "num_tokens": 4819803.0,
      "step": 2215
    },
    {
      "entropy": 7.111145877838135,
      "epoch": 0.23757290384718283,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004998069464952025,
      "loss": 6.9275,
      "mean_token_accuracy": 0.1055069126188755,
      "num_tokens": 4831226.0,
      "step": 2220
    },
    {
      "entropy": 7.114426469802856,
      "epoch": 0.23810797795494676,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998053597721932,
      "loss": 6.955,
      "mean_token_accuracy": 0.11139222607016563,
      "num_tokens": 4842168.0,
      "step": 2225
    },
    {
      "entropy": 7.013771915435791,
      "epoch": 0.2386430520627107,
      "grad_norm": 0.875,
      "learning_rate": 0.0004998037665579608,
      "loss": 6.9151,
      "mean_token_accuracy": 0.11002311557531357,
      "num_tokens": 4852655.0,
      "step": 2230
    },
    {
      "entropy": 6.965934324264526,
      "epoch": 0.2391781261704746,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998021668525513,
      "loss": 6.8002,
      "mean_token_accuracy": 0.11136664599180221,
      "num_tokens": 4862164.0,
      "step": 2235
    },
    {
      "entropy": 7.1173601150512695,
      "epoch": 0.23971320027823853,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998005606560107,
      "loss": 6.9347,
      "mean_token_accuracy": 0.11432894393801689,
      "num_tokens": 4874233.0,
      "step": 2240
    },
    {
      "entropy": 7.07620940208435,
      "epoch": 0.24024827438600246,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004997989479683856,
      "loss": 6.924,
      "mean_token_accuracy": 0.11140381470322609,
      "num_tokens": 4885948.0,
      "step": 2245
    },
    {
      "entropy": 7.045759534835815,
      "epoch": 0.2407833484937664,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004997973287897224,
      "loss": 6.9123,
      "mean_token_accuracy": 0.11342343240976334,
      "num_tokens": 4897128.0,
      "step": 2250
    },
    {
      "entropy": 7.03762469291687,
      "epoch": 0.2413184226015303,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004997957031200681,
      "loss": 6.8835,
      "mean_token_accuracy": 0.11781226769089699,
      "num_tokens": 4907699.0,
      "step": 2255
    },
    {
      "entropy": 6.970374584197998,
      "epoch": 0.24185349670929424,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997940709594692,
      "loss": 6.806,
      "mean_token_accuracy": 0.11417917162179947,
      "num_tokens": 4918772.0,
      "step": 2260
    },
    {
      "entropy": 7.130536031723023,
      "epoch": 0.24238857081705817,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997924323079733,
      "loss": 6.8628,
      "mean_token_accuracy": 0.10981507375836372,
      "num_tokens": 4930046.0,
      "step": 2265
    },
    {
      "entropy": 6.969363880157471,
      "epoch": 0.24292364492482207,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997907871656273,
      "loss": 6.8944,
      "mean_token_accuracy": 0.11376282870769501,
      "num_tokens": 4940609.0,
      "step": 2270
    },
    {
      "entropy": 7.0445019721984865,
      "epoch": 0.243458719032586,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997891355324791,
      "loss": 6.9047,
      "mean_token_accuracy": 0.11144462153315544,
      "num_tokens": 4951339.0,
      "step": 2275
    },
    {
      "entropy": 7.013249683380127,
      "epoch": 0.24399379314034994,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997874774085761,
      "loss": 6.8215,
      "mean_token_accuracy": 0.10943349301815034,
      "num_tokens": 4961083.0,
      "step": 2280
    },
    {
      "entropy": 6.997879314422607,
      "epoch": 0.24452886724811387,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997858127939662,
      "loss": 6.824,
      "mean_token_accuracy": 0.1134270079433918,
      "num_tokens": 4971399.0,
      "step": 2285
    },
    {
      "entropy": 7.0827906131744385,
      "epoch": 0.24506394135587778,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997841416886976,
      "loss": 6.9026,
      "mean_token_accuracy": 0.10839502215385437,
      "num_tokens": 4982429.0,
      "step": 2290
    },
    {
      "entropy": 7.065428256988525,
      "epoch": 0.2455990154636417,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997824640928186,
      "loss": 6.9601,
      "mean_token_accuracy": 0.10927876681089402,
      "num_tokens": 4993706.0,
      "step": 2295
    },
    {
      "entropy": 7.042297983169556,
      "epoch": 0.24613408957140565,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997807800063774,
      "loss": 6.8437,
      "mean_token_accuracy": 0.11474441438913345,
      "num_tokens": 5005411.0,
      "step": 2300
    },
    {
      "entropy": 7.061811208724976,
      "epoch": 0.24666916367916958,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000499779089429423,
      "loss": 6.963,
      "mean_token_accuracy": 0.1087774895131588,
      "num_tokens": 5016011.0,
      "step": 2305
    },
    {
      "entropy": 7.043573570251465,
      "epoch": 0.24720423778693348,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004997773923620037,
      "loss": 6.9238,
      "mean_token_accuracy": 0.10597949847579002,
      "num_tokens": 5027841.0,
      "step": 2310
    },
    {
      "entropy": 7.102751350402832,
      "epoch": 0.24773931189469742,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004997756888041689,
      "loss": 6.9271,
      "mean_token_accuracy": 0.10425131991505623,
      "num_tokens": 5038640.0,
      "step": 2315
    },
    {
      "entropy": 6.972273588180542,
      "epoch": 0.24827438600246135,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997739787559677,
      "loss": 6.8668,
      "mean_token_accuracy": 0.11417348384857177,
      "num_tokens": 5050423.0,
      "step": 2320
    },
    {
      "entropy": 7.035957622528076,
      "epoch": 0.24880946011022526,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997722622174494,
      "loss": 6.8771,
      "mean_token_accuracy": 0.11295794397592544,
      "num_tokens": 5060654.0,
      "step": 2325
    },
    {
      "entropy": 6.993176651000977,
      "epoch": 0.2493445342179892,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997705391886636,
      "loss": 6.799,
      "mean_token_accuracy": 0.11571635380387306,
      "num_tokens": 5070165.0,
      "step": 2330
    },
    {
      "entropy": 6.959088754653931,
      "epoch": 0.24987960832575312,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997688096696601,
      "loss": 6.8572,
      "mean_token_accuracy": 0.1051781103014946,
      "num_tokens": 5080937.0,
      "step": 2335
    },
    {
      "entropy": 6.938587760925293,
      "epoch": 0.25041468243351706,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004997670736604888,
      "loss": 6.808,
      "mean_token_accuracy": 0.11930364593863488,
      "num_tokens": 5090955.0,
      "step": 2340
    },
    {
      "entropy": 7.089204788208008,
      "epoch": 0.250949756541281,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004997653311611998,
      "loss": 6.9384,
      "mean_token_accuracy": 0.10624079629778863,
      "num_tokens": 5102201.0,
      "step": 2345
    },
    {
      "entropy": 7.0677807331085205,
      "epoch": 0.25148483064904487,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004997635821718435,
      "loss": 6.8653,
      "mean_token_accuracy": 0.11593251675367355,
      "num_tokens": 5112235.0,
      "step": 2350
    },
    {
      "entropy": 6.948492336273193,
      "epoch": 0.2520199047568088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997618266924704,
      "loss": 6.7418,
      "mean_token_accuracy": 0.11052383631467819,
      "num_tokens": 5122360.0,
      "step": 2355
    },
    {
      "entropy": 7.005466365814209,
      "epoch": 0.25255497886457273,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499760064723131,
      "loss": 6.9121,
      "mean_token_accuracy": 0.10666457489132881,
      "num_tokens": 5133084.0,
      "step": 2360
    },
    {
      "entropy": 6.970629024505615,
      "epoch": 0.25309005297233667,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004997582962638764,
      "loss": 6.8565,
      "mean_token_accuracy": 0.1083373948931694,
      "num_tokens": 5144495.0,
      "step": 2365
    },
    {
      "entropy": 7.105882930755615,
      "epoch": 0.2536251270801006,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997565213147575,
      "loss": 6.865,
      "mean_token_accuracy": 0.11402732953429222,
      "num_tokens": 5154824.0,
      "step": 2370
    },
    {
      "entropy": 6.978338050842285,
      "epoch": 0.25416020118786453,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997547398758257,
      "loss": 6.8449,
      "mean_token_accuracy": 0.10908161997795104,
      "num_tokens": 5165061.0,
      "step": 2375
    },
    {
      "entropy": 6.9950824737548825,
      "epoch": 0.25469527529562846,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997529519471323,
      "loss": 6.8105,
      "mean_token_accuracy": 0.11739344000816346,
      "num_tokens": 5174964.0,
      "step": 2380
    },
    {
      "entropy": 6.959466028213501,
      "epoch": 0.25523034940339234,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997511575287291,
      "loss": 6.8388,
      "mean_token_accuracy": 0.11174839809536934,
      "num_tokens": 5185458.0,
      "step": 2385
    },
    {
      "entropy": 7.014845609664917,
      "epoch": 0.2557654235111563,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997493566206677,
      "loss": 6.8452,
      "mean_token_accuracy": 0.11382765620946884,
      "num_tokens": 5196193.0,
      "step": 2390
    },
    {
      "entropy": 6.993070125579834,
      "epoch": 0.2563004976189202,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004997475492230003,
      "loss": 6.9517,
      "mean_token_accuracy": 0.10821824818849564,
      "num_tokens": 5208541.0,
      "step": 2395
    },
    {
      "entropy": 7.031092548370362,
      "epoch": 0.25683557172668414,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499745735335779,
      "loss": 6.7908,
      "mean_token_accuracy": 0.11512782797217369,
      "num_tokens": 5218567.0,
      "step": 2400
    },
    {
      "entropy": 6.993920040130615,
      "epoch": 0.2573706458344481,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004997439149590561,
      "loss": 6.8965,
      "mean_token_accuracy": 0.11021987870335578,
      "num_tokens": 5228293.0,
      "step": 2405
    },
    {
      "entropy": 6.987144422531128,
      "epoch": 0.257905719942212,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997420880928843,
      "loss": 6.8368,
      "mean_token_accuracy": 0.11468688249588013,
      "num_tokens": 5238880.0,
      "step": 2410
    },
    {
      "entropy": 7.0282874584198,
      "epoch": 0.25844079404997594,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004997402547373162,
      "loss": 6.9248,
      "mean_token_accuracy": 0.10779067873954773,
      "num_tokens": 5250104.0,
      "step": 2415
    },
    {
      "entropy": 7.077765417098999,
      "epoch": 0.2589758681577399,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004997384148924049,
      "loss": 6.9476,
      "mean_token_accuracy": 0.10701018646359443,
      "num_tokens": 5260355.0,
      "step": 2420
    },
    {
      "entropy": 7.013334131240844,
      "epoch": 0.25951094226550375,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004997365685582035,
      "loss": 6.8839,
      "mean_token_accuracy": 0.11221425458788872,
      "num_tokens": 5272117.0,
      "step": 2425
    },
    {
      "entropy": 7.025537967681885,
      "epoch": 0.2600460163732677,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004997347157347651,
      "loss": 6.9641,
      "mean_token_accuracy": 0.11001334488391876,
      "num_tokens": 5284583.0,
      "step": 2430
    },
    {
      "entropy": 7.070367336273193,
      "epoch": 0.2605810904810316,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004997328564221435,
      "loss": 6.7554,
      "mean_token_accuracy": 0.12274964973330497,
      "num_tokens": 5295846.0,
      "step": 2435
    },
    {
      "entropy": 6.974102544784546,
      "epoch": 0.26111616458879555,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004997309906203922,
      "loss": 6.9064,
      "mean_token_accuracy": 0.10685541406273842,
      "num_tokens": 5307802.0,
      "step": 2440
    },
    {
      "entropy": 7.062164688110352,
      "epoch": 0.2616512386965595,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997291183295652,
      "loss": 6.8217,
      "mean_token_accuracy": 0.11616088747978211,
      "num_tokens": 5318263.0,
      "step": 2445
    },
    {
      "entropy": 7.043830013275146,
      "epoch": 0.2621863128043234,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004997272395497164,
      "loss": 6.8821,
      "mean_token_accuracy": 0.11346295177936554,
      "num_tokens": 5328555.0,
      "step": 2450
    },
    {
      "entropy": 6.941327333450317,
      "epoch": 0.26272138691208735,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004997253542809002,
      "loss": 6.8354,
      "mean_token_accuracy": 0.11198842748999596,
      "num_tokens": 5339954.0,
      "step": 2455
    },
    {
      "entropy": 7.107675504684448,
      "epoch": 0.26325646101985123,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004997234625231709,
      "loss": 6.8836,
      "mean_token_accuracy": 0.11381551921367646,
      "num_tokens": 5350919.0,
      "step": 2460
    },
    {
      "entropy": 6.949951696395874,
      "epoch": 0.26379153512761516,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004997215642765832,
      "loss": 6.8624,
      "mean_token_accuracy": 0.11542123556137085,
      "num_tokens": 5361844.0,
      "step": 2465
    },
    {
      "entropy": 7.018435907363892,
      "epoch": 0.2643266092353791,
      "grad_norm": 0.875,
      "learning_rate": 0.000499719659541192,
      "loss": 6.8726,
      "mean_token_accuracy": 0.10965746343135833,
      "num_tokens": 5372026.0,
      "step": 2470
    },
    {
      "entropy": 7.0428282737731935,
      "epoch": 0.26486168334314303,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004997177483170521,
      "loss": 6.9163,
      "mean_token_accuracy": 0.11546080112457276,
      "num_tokens": 5382965.0,
      "step": 2475
    },
    {
      "entropy": 6.963793182373047,
      "epoch": 0.26539675745090696,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997158306042188,
      "loss": 6.874,
      "mean_token_accuracy": 0.11316567808389663,
      "num_tokens": 5393738.0,
      "step": 2480
    },
    {
      "entropy": 6.974194860458374,
      "epoch": 0.2659318315586709,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004997139064027476,
      "loss": 6.74,
      "mean_token_accuracy": 0.12156007960438728,
      "num_tokens": 5404059.0,
      "step": 2485
    },
    {
      "entropy": 6.964081430435181,
      "epoch": 0.26646690566643483,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997119757126938,
      "loss": 6.7804,
      "mean_token_accuracy": 0.11532729938626289,
      "num_tokens": 5414152.0,
      "step": 2490
    },
    {
      "entropy": 7.028856039047241,
      "epoch": 0.2670019797741987,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004997100385341133,
      "loss": 6.9634,
      "mean_token_accuracy": 0.1126515194773674,
      "num_tokens": 5425227.0,
      "step": 2495
    },
    {
      "entropy": 6.9782177925109865,
      "epoch": 0.26753705388196264,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004997080948670621,
      "loss": 6.789,
      "mean_token_accuracy": 0.11695293188095093,
      "num_tokens": 5435716.0,
      "step": 2500
    },
    {
      "entropy": 6.985808944702148,
      "epoch": 0.26807212798972657,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004997061447115962,
      "loss": 6.8623,
      "mean_token_accuracy": 0.10614107474684716,
      "num_tokens": 5446519.0,
      "step": 2505
    },
    {
      "entropy": 7.017519235610962,
      "epoch": 0.2686072020974905,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004997041880677719,
      "loss": 6.8421,
      "mean_token_accuracy": 0.11282496824860573,
      "num_tokens": 5457496.0,
      "step": 2510
    },
    {
      "entropy": 6.900492763519287,
      "epoch": 0.26914227620525444,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004997022249356457,
      "loss": 6.7495,
      "mean_token_accuracy": 0.11760077178478241,
      "num_tokens": 5469059.0,
      "step": 2515
    },
    {
      "entropy": 7.010392713546753,
      "epoch": 0.26967735031301837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997002553152745,
      "loss": 6.8399,
      "mean_token_accuracy": 0.1156559094786644,
      "num_tokens": 5480324.0,
      "step": 2520
    },
    {
      "entropy": 7.005701923370362,
      "epoch": 0.2702124244207823,
      "grad_norm": 0.875,
      "learning_rate": 0.0004996982792067148,
      "loss": 6.845,
      "mean_token_accuracy": 0.1095408894121647,
      "num_tokens": 5491513.0,
      "step": 2525
    },
    {
      "entropy": 6.884131574630738,
      "epoch": 0.2707474985285462,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000499696296610024,
      "loss": 6.7724,
      "mean_token_accuracy": 0.11539606228470803,
      "num_tokens": 5502414.0,
      "step": 2530
    },
    {
      "entropy": 6.973390340805054,
      "epoch": 0.2712825726363101,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996943075252592,
      "loss": 6.8968,
      "mean_token_accuracy": 0.11106460094451905,
      "num_tokens": 5512962.0,
      "step": 2535
    },
    {
      "entropy": 6.999091625213623,
      "epoch": 0.27181764674407405,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996923119524777,
      "loss": 6.7602,
      "mean_token_accuracy": 0.11408493369817733,
      "num_tokens": 5524648.0,
      "step": 2540
    },
    {
      "entropy": 6.898416757583618,
      "epoch": 0.272352720851838,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004996903098917375,
      "loss": 6.8386,
      "mean_token_accuracy": 0.10790063142776489,
      "num_tokens": 5536158.0,
      "step": 2545
    },
    {
      "entropy": 6.9561662673950195,
      "epoch": 0.2728877949596019,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499688301343096,
      "loss": 6.7565,
      "mean_token_accuracy": 0.11405683234333992,
      "num_tokens": 5547374.0,
      "step": 2550
    },
    {
      "entropy": 7.05834732055664,
      "epoch": 0.27342286906736585,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004996862863066114,
      "loss": 6.8386,
      "mean_token_accuracy": 0.11390230059623718,
      "num_tokens": 5559212.0,
      "step": 2555
    },
    {
      "entropy": 6.979931354522705,
      "epoch": 0.2739579431751298,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004996842647823419,
      "loss": 6.8098,
      "mean_token_accuracy": 0.11664133742451668,
      "num_tokens": 5569302.0,
      "step": 2560
    },
    {
      "entropy": 6.945258140563965,
      "epoch": 0.27449301728289366,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004996822367703458,
      "loss": 6.8112,
      "mean_token_accuracy": 0.10886863917112351,
      "num_tokens": 5581688.0,
      "step": 2565
    },
    {
      "entropy": 7.029229307174683,
      "epoch": 0.2750280913906576,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996802022706817,
      "loss": 6.7402,
      "mean_token_accuracy": 0.11893118396401406,
      "num_tokens": 5591376.0,
      "step": 2570
    },
    {
      "entropy": 6.872675561904908,
      "epoch": 0.2755631654984215,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996781612834083,
      "loss": 6.7905,
      "mean_token_accuracy": 0.11670346409082413,
      "num_tokens": 5601510.0,
      "step": 2575
    },
    {
      "entropy": 7.052143335342407,
      "epoch": 0.27609823960618546,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996761138085846,
      "loss": 6.868,
      "mean_token_accuracy": 0.1093013696372509,
      "num_tokens": 5611436.0,
      "step": 2580
    },
    {
      "entropy": 6.98383355140686,
      "epoch": 0.2766333137139494,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996740598462697,
      "loss": 6.8517,
      "mean_token_accuracy": 0.11311317384243011,
      "num_tokens": 5620470.0,
      "step": 2585
    },
    {
      "entropy": 6.922229051589966,
      "epoch": 0.2771683878217133,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996719993965229,
      "loss": 6.7627,
      "mean_token_accuracy": 0.11485590264201165,
      "num_tokens": 5630740.0,
      "step": 2590
    },
    {
      "entropy": 6.873901319503784,
      "epoch": 0.27770346192947726,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996699324594036,
      "loss": 6.7591,
      "mean_token_accuracy": 0.11999578103423118,
      "num_tokens": 5641138.0,
      "step": 2595
    },
    {
      "entropy": 7.053933334350586,
      "epoch": 0.27823853603724114,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004996678590349717,
      "loss": 6.89,
      "mean_token_accuracy": 0.11600473672151565,
      "num_tokens": 5652474.0,
      "step": 2600
    },
    {
      "entropy": 6.952626085281372,
      "epoch": 0.27877361014500507,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996657791232868,
      "loss": 6.7958,
      "mean_token_accuracy": 0.11903122812509537,
      "num_tokens": 5663178.0,
      "step": 2605
    },
    {
      "entropy": 6.984945917129517,
      "epoch": 0.279308684252769,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996636927244092,
      "loss": 6.8051,
      "mean_token_accuracy": 0.11196058839559556,
      "num_tokens": 5673565.0,
      "step": 2610
    },
    {
      "entropy": 7.001294422149658,
      "epoch": 0.27984375836053293,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000499661599838399,
      "loss": 6.8497,
      "mean_token_accuracy": 0.1122498370707035,
      "num_tokens": 5684651.0,
      "step": 2615
    },
    {
      "entropy": 6.904658508300781,
      "epoch": 0.28037883246829687,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996595004653167,
      "loss": 6.7432,
      "mean_token_accuracy": 0.11533383056521415,
      "num_tokens": 5696081.0,
      "step": 2620
    },
    {
      "entropy": 6.9526012420654295,
      "epoch": 0.2809139065760608,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004996573946052229,
      "loss": 6.7812,
      "mean_token_accuracy": 0.1164487712085247,
      "num_tokens": 5706830.0,
      "step": 2625
    },
    {
      "entropy": 6.8979075908660885,
      "epoch": 0.28144898068382473,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996552822581783,
      "loss": 6.7986,
      "mean_token_accuracy": 0.11754084527492523,
      "num_tokens": 5717261.0,
      "step": 2630
    },
    {
      "entropy": 7.005229568481445,
      "epoch": 0.2819840547915886,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499653163424244,
      "loss": 6.7734,
      "mean_token_accuracy": 0.11206443011760711,
      "num_tokens": 5727280.0,
      "step": 2635
    },
    {
      "entropy": 6.993945360183716,
      "epoch": 0.28251912889935255,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996510381034814,
      "loss": 6.8241,
      "mean_token_accuracy": 0.10772419720888138,
      "num_tokens": 5737485.0,
      "step": 2640
    },
    {
      "entropy": 6.9192140102386475,
      "epoch": 0.2830542030071165,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996489062959515,
      "loss": 6.7957,
      "mean_token_accuracy": 0.11468210741877556,
      "num_tokens": 5747683.0,
      "step": 2645
    },
    {
      "entropy": 6.915011835098267,
      "epoch": 0.2835892771148804,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004996467680017159,
      "loss": 6.862,
      "mean_token_accuracy": 0.10823372080922126,
      "num_tokens": 5758148.0,
      "step": 2650
    },
    {
      "entropy": 6.931305360794068,
      "epoch": 0.28412435122264434,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996446232208365,
      "loss": 6.7896,
      "mean_token_accuracy": 0.11629763916134835,
      "num_tokens": 5769109.0,
      "step": 2655
    },
    {
      "entropy": 6.928414344787598,
      "epoch": 0.2846594253304083,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004996424719533753,
      "loss": 6.8193,
      "mean_token_accuracy": 0.11202871650457383,
      "num_tokens": 5780344.0,
      "step": 2660
    },
    {
      "entropy": 6.936452388763428,
      "epoch": 0.2851944994381722,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996403141993941,
      "loss": 6.627,
      "mean_token_accuracy": 0.12468865811824799,
      "num_tokens": 5790167.0,
      "step": 2665
    },
    {
      "entropy": 6.922191047668457,
      "epoch": 0.2857295735459361,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004996381499589555,
      "loss": 6.791,
      "mean_token_accuracy": 0.11766616478562356,
      "num_tokens": 5800076.0,
      "step": 2670
    },
    {
      "entropy": 7.001429510116577,
      "epoch": 0.2862646476537,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996359792321219,
      "loss": 6.7247,
      "mean_token_accuracy": 0.12332994639873504,
      "num_tokens": 5810547.0,
      "step": 2675
    },
    {
      "entropy": 6.842127323150635,
      "epoch": 0.28679972176146395,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000499633802018956,
      "loss": 6.7651,
      "mean_token_accuracy": 0.11882570460438728,
      "num_tokens": 5821210.0,
      "step": 2680
    },
    {
      "entropy": 6.933741283416748,
      "epoch": 0.2873347958692279,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004996316183195204,
      "loss": 6.8064,
      "mean_token_accuracy": 0.11239668801426887,
      "num_tokens": 5832361.0,
      "step": 2685
    },
    {
      "entropy": 6.997045278549194,
      "epoch": 0.2878698699769918,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004996294281338786,
      "loss": 6.7965,
      "mean_token_accuracy": 0.11312427893280982,
      "num_tokens": 5843242.0,
      "step": 2690
    },
    {
      "entropy": 6.942799377441406,
      "epoch": 0.28840494408475575,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004996272314620936,
      "loss": 6.8534,
      "mean_token_accuracy": 0.11269704699516296,
      "num_tokens": 5855163.0,
      "step": 2695
    },
    {
      "entropy": 6.908647918701172,
      "epoch": 0.2889400181925197,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004996250283042288,
      "loss": 6.7873,
      "mean_token_accuracy": 0.11659785062074661,
      "num_tokens": 5866908.0,
      "step": 2700
    },
    {
      "entropy": 6.983246564865112,
      "epoch": 0.28947509230028357,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499622818660348,
      "loss": 6.8308,
      "mean_token_accuracy": 0.11280549988150597,
      "num_tokens": 5877342.0,
      "step": 2705
    },
    {
      "entropy": 6.902181339263916,
      "epoch": 0.2900101664080475,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996206025305147,
      "loss": 6.7809,
      "mean_token_accuracy": 0.11292893514037132,
      "num_tokens": 5887960.0,
      "step": 2710
    },
    {
      "entropy": 6.913876533508301,
      "epoch": 0.29054524051581143,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004996183799147932,
      "loss": 6.7618,
      "mean_token_accuracy": 0.11444101110100746,
      "num_tokens": 5899887.0,
      "step": 2715
    },
    {
      "entropy": 6.982015180587768,
      "epoch": 0.29108031462357536,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004996161508132475,
      "loss": 6.7616,
      "mean_token_accuracy": 0.10974067524075508,
      "num_tokens": 5911022.0,
      "step": 2720
    },
    {
      "entropy": 6.94197039604187,
      "epoch": 0.2916153887313393,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000499613915225942,
      "loss": 6.8321,
      "mean_token_accuracy": 0.11855507642030716,
      "num_tokens": 5922780.0,
      "step": 2725
    },
    {
      "entropy": 6.964428901672363,
      "epoch": 0.29215046283910323,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004996116731529412,
      "loss": 6.8632,
      "mean_token_accuracy": 0.11426409929990769,
      "num_tokens": 5934112.0,
      "step": 2730
    },
    {
      "entropy": 6.972011709213257,
      "epoch": 0.29268553694686716,
      "grad_norm": 0.953125,
      "learning_rate": 0.00049960942459431,
      "loss": 6.8045,
      "mean_token_accuracy": 0.1121431604027748,
      "num_tokens": 5944386.0,
      "step": 2735
    },
    {
      "entropy": 6.87826681137085,
      "epoch": 0.29322061105463104,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004996071695501132,
      "loss": 6.7394,
      "mean_token_accuracy": 0.1182074099779129,
      "num_tokens": 5955552.0,
      "step": 2740
    },
    {
      "entropy": 6.928379726409912,
      "epoch": 0.293755685162395,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004996049080204159,
      "loss": 6.7346,
      "mean_token_accuracy": 0.11500842422246933,
      "num_tokens": 5966790.0,
      "step": 2745
    },
    {
      "entropy": 6.979641532897949,
      "epoch": 0.2942907592701589,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004996026400052835,
      "loss": 6.7392,
      "mean_token_accuracy": 0.11592580303549767,
      "num_tokens": 5977704.0,
      "step": 2750
    },
    {
      "entropy": 6.8253716945648195,
      "epoch": 0.29482583337792284,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004996003655047814,
      "loss": 6.7484,
      "mean_token_accuracy": 0.11899459958076478,
      "num_tokens": 5989221.0,
      "step": 2755
    },
    {
      "entropy": 6.908860492706299,
      "epoch": 0.2953609074856868,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004995980845189753,
      "loss": 6.8069,
      "mean_token_accuracy": 0.11177898719906806,
      "num_tokens": 6000906.0,
      "step": 2760
    },
    {
      "entropy": 6.967212343215943,
      "epoch": 0.2958959815934507,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004995957970479312,
      "loss": 6.8099,
      "mean_token_accuracy": 0.11111435741186142,
      "num_tokens": 6012039.0,
      "step": 2765
    },
    {
      "entropy": 6.8845288276672365,
      "epoch": 0.29643105570121464,
      "grad_norm": 0.7734375,
      "learning_rate": 0.0004995935030917148,
      "loss": 6.6981,
      "mean_token_accuracy": 0.11647267341613769,
      "num_tokens": 6024383.0,
      "step": 2770
    },
    {
      "entropy": 6.988625144958496,
      "epoch": 0.2969661298089785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004995912026503926,
      "loss": 6.7698,
      "mean_token_accuracy": 0.11584996357560158,
      "num_tokens": 6034904.0,
      "step": 2775
    },
    {
      "entropy": 6.852597904205322,
      "epoch": 0.29750120391674245,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000499588895724031,
      "loss": 6.7743,
      "mean_token_accuracy": 0.111159298568964,
      "num_tokens": 6044449.0,
      "step": 2780
    },
    {
      "entropy": 6.959954643249512,
      "epoch": 0.2980362780245064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995865823126968,
      "loss": 6.7933,
      "mean_token_accuracy": 0.11641821041703224,
      "num_tokens": 6054465.0,
      "step": 2785
    },
    {
      "entropy": 6.91104040145874,
      "epoch": 0.2985713521322703,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995842624164564,
      "loss": 6.8124,
      "mean_token_accuracy": 0.11264995485544205,
      "num_tokens": 6065539.0,
      "step": 2790
    },
    {
      "entropy": 6.946519184112549,
      "epoch": 0.29910642624003425,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000499581936035377,
      "loss": 6.7705,
      "mean_token_accuracy": 0.11285096481442451,
      "num_tokens": 6076135.0,
      "step": 2795
    },
    {
      "entropy": 6.859121942520142,
      "epoch": 0.2996415003477982,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004995796031695257,
      "loss": 6.7652,
      "mean_token_accuracy": 0.11385954692959785,
      "num_tokens": 6086604.0,
      "step": 2800
    },
    {
      "entropy": 6.977310657501221,
      "epoch": 0.3001765744555621,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049957726381897,
      "loss": 6.785,
      "mean_token_accuracy": 0.11666215732693672,
      "num_tokens": 6095672.0,
      "step": 2805
    },
    {
      "entropy": 6.85119218826294,
      "epoch": 0.300711648563326,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995749179837773,
      "loss": 6.6622,
      "mean_token_accuracy": 0.12312485352158546,
      "num_tokens": 6105652.0,
      "step": 2810
    },
    {
      "entropy": 6.840598678588867,
      "epoch": 0.30124672267108993,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995725656640155,
      "loss": 6.7056,
      "mean_token_accuracy": 0.1178930640220642,
      "num_tokens": 6116930.0,
      "step": 2815
    },
    {
      "entropy": 6.875258922576904,
      "epoch": 0.30178179677885386,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004995702068597524,
      "loss": 6.7157,
      "mean_token_accuracy": 0.12197659760713578,
      "num_tokens": 6126659.0,
      "step": 2820
    },
    {
      "entropy": 6.899388074874878,
      "epoch": 0.3023168708866178,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995678415710561,
      "loss": 6.7457,
      "mean_token_accuracy": 0.115099136531353,
      "num_tokens": 6137227.0,
      "step": 2825
    },
    {
      "entropy": 6.978420209884644,
      "epoch": 0.3028519449943817,
      "grad_norm": 0.875,
      "learning_rate": 0.000499565469797995,
      "loss": 6.7977,
      "mean_token_accuracy": 0.1148864470422268,
      "num_tokens": 6147800.0,
      "step": 2830
    },
    {
      "entropy": 6.862562036514282,
      "epoch": 0.30338701910214566,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004995630915406374,
      "loss": 6.6492,
      "mean_token_accuracy": 0.12218020334839821,
      "num_tokens": 6157671.0,
      "step": 2835
    },
    {
      "entropy": 6.896828746795654,
      "epoch": 0.3039220932099096,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004995607067990521,
      "loss": 6.7933,
      "mean_token_accuracy": 0.11226967573165894,
      "num_tokens": 6168651.0,
      "step": 2840
    },
    {
      "entropy": 6.8445981502532955,
      "epoch": 0.30445716731767347,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499558315573308,
      "loss": 6.7653,
      "mean_token_accuracy": 0.11298804879188537,
      "num_tokens": 6179925.0,
      "step": 2845
    },
    {
      "entropy": 6.870256233215332,
      "epoch": 0.3049922414254374,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995559178634741,
      "loss": 6.7605,
      "mean_token_accuracy": 0.11938253119587898,
      "num_tokens": 6190541.0,
      "step": 2850
    },
    {
      "entropy": 7.023727226257324,
      "epoch": 0.30552731553320134,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995535136696196,
      "loss": 6.7961,
      "mean_token_accuracy": 0.11052935272455215,
      "num_tokens": 6201491.0,
      "step": 2855
    },
    {
      "entropy": 6.799321126937866,
      "epoch": 0.30606238964096527,
      "grad_norm": 0.8125,
      "learning_rate": 0.000499551102991814,
      "loss": 6.616,
      "mean_token_accuracy": 0.12260399758815765,
      "num_tokens": 6212443.0,
      "step": 2860
    },
    {
      "entropy": 6.903628635406494,
      "epoch": 0.3065974637487292,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004995486858301268,
      "loss": 6.8178,
      "mean_token_accuracy": 0.108098603785038,
      "num_tokens": 6222623.0,
      "step": 2865
    },
    {
      "entropy": 6.90746922492981,
      "epoch": 0.30713253785649314,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004995462621846279,
      "loss": 6.7064,
      "mean_token_accuracy": 0.11651685237884521,
      "num_tokens": 6233712.0,
      "step": 2870
    },
    {
      "entropy": 6.9290708065032955,
      "epoch": 0.30766761196425707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995438320553871,
      "loss": 6.7715,
      "mean_token_accuracy": 0.11282215490937234,
      "num_tokens": 6244254.0,
      "step": 2875
    },
    {
      "entropy": 6.881653308868408,
      "epoch": 0.30820268607202095,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995413954424748,
      "loss": 6.6712,
      "mean_token_accuracy": 0.12289544939994812,
      "num_tokens": 6254202.0,
      "step": 2880
    },
    {
      "entropy": 6.872118377685547,
      "epoch": 0.3087377601797849,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004995389523459612,
      "loss": 6.7368,
      "mean_token_accuracy": 0.11404762491583824,
      "num_tokens": 6264496.0,
      "step": 2885
    },
    {
      "entropy": 6.84141526222229,
      "epoch": 0.3092728342875488,
      "grad_norm": 0.90625,
      "learning_rate": 0.000499536502765917,
      "loss": 6.7556,
      "mean_token_accuracy": 0.12146841287612915,
      "num_tokens": 6275835.0,
      "step": 2890
    },
    {
      "entropy": 6.885162115097046,
      "epoch": 0.30980790839531275,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995340467024128,
      "loss": 6.6989,
      "mean_token_accuracy": 0.11953729987144471,
      "num_tokens": 6286805.0,
      "step": 2895
    },
    {
      "entropy": 6.878981494903565,
      "epoch": 0.3103429825030767,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995315841555195,
      "loss": 6.7113,
      "mean_token_accuracy": 0.11021909192204475,
      "num_tokens": 6298444.0,
      "step": 2900
    },
    {
      "entropy": 6.813249588012695,
      "epoch": 0.3108780566108406,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004995291151253083,
      "loss": 6.6436,
      "mean_token_accuracy": 0.12175455540418625,
      "num_tokens": 6308259.0,
      "step": 2905
    },
    {
      "entropy": 6.857611513137817,
      "epoch": 0.31141313071860455,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995266396118505,
      "loss": 6.72,
      "mean_token_accuracy": 0.11006433069705963,
      "num_tokens": 6320058.0,
      "step": 2910
    },
    {
      "entropy": 6.8574628829956055,
      "epoch": 0.3119482048263684,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004995241576152174,
      "loss": 6.6938,
      "mean_token_accuracy": 0.11197758540511131,
      "num_tokens": 6331203.0,
      "step": 2915
    },
    {
      "entropy": 6.926673221588135,
      "epoch": 0.31248327893413236,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499521669135481,
      "loss": 6.8007,
      "mean_token_accuracy": 0.11380135864019394,
      "num_tokens": 6342027.0,
      "step": 2920
    },
    {
      "entropy": 6.901452732086182,
      "epoch": 0.3130183530418963,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004995191741727128,
      "loss": 6.6965,
      "mean_token_accuracy": 0.11690429598093033,
      "num_tokens": 6352580.0,
      "step": 2925
    },
    {
      "entropy": 6.897034406661987,
      "epoch": 0.3135534271496602,
      "grad_norm": 0.80078125,
      "learning_rate": 0.000499516672726985,
      "loss": 6.699,
      "mean_token_accuracy": 0.12443051934242248,
      "num_tokens": 6363997.0,
      "step": 2930
    },
    {
      "entropy": 6.855990314483643,
      "epoch": 0.31408850125742416,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004995141647983699,
      "loss": 6.7583,
      "mean_token_accuracy": 0.11855747178196907,
      "num_tokens": 6375284.0,
      "step": 2935
    },
    {
      "entropy": 6.844248676300049,
      "epoch": 0.3146235753651881,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995116503869398,
      "loss": 6.6672,
      "mean_token_accuracy": 0.12428331449627876,
      "num_tokens": 6386698.0,
      "step": 2940
    },
    {
      "entropy": 6.867465877532959,
      "epoch": 0.315158649472952,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004995091294927673,
      "loss": 6.7408,
      "mean_token_accuracy": 0.11660289838910103,
      "num_tokens": 6397237.0,
      "step": 2945
    },
    {
      "entropy": 6.98031268119812,
      "epoch": 0.3156937235807159,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995066021159253,
      "loss": 6.8387,
      "mean_token_accuracy": 0.11507428362965584,
      "num_tokens": 6408266.0,
      "step": 2950
    },
    {
      "entropy": 6.814388084411621,
      "epoch": 0.31622879768847983,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004995040682564867,
      "loss": 6.7375,
      "mean_token_accuracy": 0.12039956524968147,
      "num_tokens": 6419563.0,
      "step": 2955
    },
    {
      "entropy": 6.865634298324585,
      "epoch": 0.31676387179624377,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004995015279145247,
      "loss": 6.6945,
      "mean_token_accuracy": 0.12165094465017319,
      "num_tokens": 6430220.0,
      "step": 2960
    },
    {
      "entropy": 6.879283094406128,
      "epoch": 0.3172989459040077,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004994989810901126,
      "loss": 6.7139,
      "mean_token_accuracy": 0.1167122632265091,
      "num_tokens": 6440157.0,
      "step": 2965
    },
    {
      "entropy": 6.871557569503784,
      "epoch": 0.31783402001177163,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994964277833239,
      "loss": 6.7633,
      "mean_token_accuracy": 0.1225232295691967,
      "num_tokens": 6449511.0,
      "step": 2970
    },
    {
      "entropy": 6.925748014450074,
      "epoch": 0.31836909411953557,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004994938679942324,
      "loss": 6.7496,
      "mean_token_accuracy": 0.11134742796421052,
      "num_tokens": 6460608.0,
      "step": 2975
    },
    {
      "entropy": 6.8563477993011475,
      "epoch": 0.3189041682272995,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994913017229121,
      "loss": 6.7304,
      "mean_token_accuracy": 0.12039977535605431,
      "num_tokens": 6472737.0,
      "step": 2980
    },
    {
      "entropy": 6.803793001174927,
      "epoch": 0.31943924233506343,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994887289694371,
      "loss": 6.6205,
      "mean_token_accuracy": 0.12295867651700973,
      "num_tokens": 6482993.0,
      "step": 2985
    },
    {
      "entropy": 6.9399620532989506,
      "epoch": 0.3199743164428273,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994861497338814,
      "loss": 6.8293,
      "mean_token_accuracy": 0.11786540225148201,
      "num_tokens": 6494530.0,
      "step": 2990
    },
    {
      "entropy": 6.926778125762939,
      "epoch": 0.32050939055059124,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004994835640163198,
      "loss": 6.6588,
      "mean_token_accuracy": 0.11456671357154846,
      "num_tokens": 6504691.0,
      "step": 2995
    },
    {
      "entropy": 6.821675395965576,
      "epoch": 0.3210444646583552,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994809718168267,
      "loss": 6.684,
      "mean_token_accuracy": 0.11950030028820038,
      "num_tokens": 6514924.0,
      "step": 3000
    },
    {
      "epoch": 0.3210444646583552,
      "eval_entropy": 6.548516875521513,
      "eval_loss": 6.794521331787109,
      "eval_mean_token_accuracy": 0.12106629688332907,
      "eval_num_tokens": 6514924.0,
      "eval_runtime": 17.6213,
      "eval_samples_per_second": 1684.272,
      "eval_steps_per_second": 210.541,
      "step": 3000
    },
    {
      "entropy": 6.799951505661011,
      "epoch": 0.3215795387661191,
      "grad_norm": 0.875,
      "learning_rate": 0.0004994783731354771,
      "loss": 6.663,
      "mean_token_accuracy": 0.121905267983675,
      "num_tokens": 6526097.0,
      "step": 3005
    },
    {
      "entropy": 6.887381935119629,
      "epoch": 0.32211461287388304,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004994757679723461,
      "loss": 6.7307,
      "mean_token_accuracy": 0.11466163247823716,
      "num_tokens": 6536384.0,
      "step": 3010
    },
    {
      "entropy": 6.9197609424591064,
      "epoch": 0.322649686981647,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004994731563275088,
      "loss": 6.7868,
      "mean_token_accuracy": 0.11109080761671067,
      "num_tokens": 6547161.0,
      "step": 3015
    },
    {
      "entropy": 6.826877498626709,
      "epoch": 0.3231847610894109,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004994705382010406,
      "loss": 6.6947,
      "mean_token_accuracy": 0.11839734464883804,
      "num_tokens": 6556751.0,
      "step": 3020
    },
    {
      "entropy": 6.831856346130371,
      "epoch": 0.3237198351971748,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004994679135930174,
      "loss": 6.7379,
      "mean_token_accuracy": 0.12035313099622727,
      "num_tokens": 6567655.0,
      "step": 3025
    },
    {
      "entropy": 6.9234333515167235,
      "epoch": 0.3242549093049387,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994652825035144,
      "loss": 6.7726,
      "mean_token_accuracy": 0.11987186148762703,
      "num_tokens": 6579127.0,
      "step": 3030
    },
    {
      "entropy": 6.984951210021973,
      "epoch": 0.32478998341270265,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994626449326081,
      "loss": 6.7544,
      "mean_token_accuracy": 0.11932827979326248,
      "num_tokens": 6590779.0,
      "step": 3035
    },
    {
      "entropy": 6.786751747131348,
      "epoch": 0.3253250575204666,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994600008803745,
      "loss": 6.6644,
      "mean_token_accuracy": 0.12521166279911994,
      "num_tokens": 6600849.0,
      "step": 3040
    },
    {
      "entropy": 6.839286756515503,
      "epoch": 0.3258601316282305,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994573503468899,
      "loss": 6.6739,
      "mean_token_accuracy": 0.12311565428972245,
      "num_tokens": 6610633.0,
      "step": 3045
    },
    {
      "entropy": 6.862971591949463,
      "epoch": 0.32639520573599445,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004994546933322308,
      "loss": 6.7091,
      "mean_token_accuracy": 0.11418486312031746,
      "num_tokens": 6621327.0,
      "step": 3050
    },
    {
      "entropy": 6.812316656112671,
      "epoch": 0.3269302798437584,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499452029836474,
      "loss": 6.6481,
      "mean_token_accuracy": 0.12416826337575912,
      "num_tokens": 6632267.0,
      "step": 3055
    },
    {
      "entropy": 6.841871833801269,
      "epoch": 0.32746535395152226,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004994493598596963,
      "loss": 6.6816,
      "mean_token_accuracy": 0.11459707394242287,
      "num_tokens": 6642714.0,
      "step": 3060
    },
    {
      "entropy": 6.869484233856201,
      "epoch": 0.3280004280592862,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004994466834019749,
      "loss": 6.685,
      "mean_token_accuracy": 0.12255851849913597,
      "num_tokens": 6652353.0,
      "step": 3065
    },
    {
      "entropy": 6.835297298431397,
      "epoch": 0.32853550216705013,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004994440004633871,
      "loss": 6.6661,
      "mean_token_accuracy": 0.11809565648436546,
      "num_tokens": 6663133.0,
      "step": 3070
    },
    {
      "entropy": 6.749080991744995,
      "epoch": 0.32907057627481406,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994413110440103,
      "loss": 6.6284,
      "mean_token_accuracy": 0.1260196879506111,
      "num_tokens": 6673984.0,
      "step": 3075
    },
    {
      "entropy": 6.871647500991822,
      "epoch": 0.329605650382578,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004994386151439222,
      "loss": 6.714,
      "mean_token_accuracy": 0.11568559035658836,
      "num_tokens": 6686171.0,
      "step": 3080
    },
    {
      "entropy": 6.957374811172485,
      "epoch": 0.33014072449034193,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994359127632006,
      "loss": 6.7591,
      "mean_token_accuracy": 0.1136956661939621,
      "num_tokens": 6696609.0,
      "step": 3085
    },
    {
      "entropy": 6.855549001693726,
      "epoch": 0.33067579859810586,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004994332039019235,
      "loss": 6.6983,
      "mean_token_accuracy": 0.11846785247325897,
      "num_tokens": 6706188.0,
      "step": 3090
    },
    {
      "entropy": 6.778521108627319,
      "epoch": 0.33121087270586974,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994304885601694,
      "loss": 6.6773,
      "mean_token_accuracy": 0.11152351349592209,
      "num_tokens": 6716276.0,
      "step": 3095
    },
    {
      "entropy": 6.760422897338867,
      "epoch": 0.3317459468136337,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004994277667380162,
      "loss": 6.6388,
      "mean_token_accuracy": 0.12479276284575462,
      "num_tokens": 6727176.0,
      "step": 3100
    },
    {
      "entropy": 6.854667711257934,
      "epoch": 0.3322810209213976,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004994250384355429,
      "loss": 6.727,
      "mean_token_accuracy": 0.11873743012547493,
      "num_tokens": 6737768.0,
      "step": 3105
    },
    {
      "entropy": 6.849891614913941,
      "epoch": 0.33281609502916154,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004994223036528282,
      "loss": 6.6613,
      "mean_token_accuracy": 0.11627993434667587,
      "num_tokens": 6747962.0,
      "step": 3110
    },
    {
      "entropy": 6.821473026275635,
      "epoch": 0.3333511691369255,
      "grad_norm": 0.9375,
      "learning_rate": 0.000499419562389951,
      "loss": 6.6551,
      "mean_token_accuracy": 0.12350267693400382,
      "num_tokens": 6758400.0,
      "step": 3115
    },
    {
      "entropy": 6.813153457641602,
      "epoch": 0.3338862432446894,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994168146469904,
      "loss": 6.6619,
      "mean_token_accuracy": 0.12469682693481446,
      "num_tokens": 6768855.0,
      "step": 3120
    },
    {
      "entropy": 6.861107110977173,
      "epoch": 0.33442131735245334,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004994140604240259,
      "loss": 6.6474,
      "mean_token_accuracy": 0.11827528327703477,
      "num_tokens": 6778656.0,
      "step": 3125
    },
    {
      "entropy": 6.844491481781006,
      "epoch": 0.3349563914602172,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004994112997211369,
      "loss": 6.7274,
      "mean_token_accuracy": 0.12191558927297592,
      "num_tokens": 6789960.0,
      "step": 3130
    },
    {
      "entropy": 6.770041465759277,
      "epoch": 0.33549146556798115,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994085325384031,
      "loss": 6.7037,
      "mean_token_accuracy": 0.1188412569463253,
      "num_tokens": 6800188.0,
      "step": 3135
    },
    {
      "entropy": 6.833252429962158,
      "epoch": 0.3360265396757451,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004994057588759045,
      "loss": 6.6543,
      "mean_token_accuracy": 0.12379577830433845,
      "num_tokens": 6811032.0,
      "step": 3140
    },
    {
      "entropy": 6.751126480102539,
      "epoch": 0.336561613783509,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004994029787337212,
      "loss": 6.6709,
      "mean_token_accuracy": 0.12116707265377044,
      "num_tokens": 6823060.0,
      "step": 3145
    },
    {
      "entropy": 6.856843137741089,
      "epoch": 0.33709668789127295,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994001921119333,
      "loss": 6.7397,
      "mean_token_accuracy": 0.12123994454741478,
      "num_tokens": 6833559.0,
      "step": 3150
    },
    {
      "entropy": 6.888458251953125,
      "epoch": 0.3376317619990369,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004993973990106215,
      "loss": 6.7015,
      "mean_token_accuracy": 0.12211815863847733,
      "num_tokens": 6843657.0,
      "step": 3155
    },
    {
      "entropy": 6.793592262268066,
      "epoch": 0.3381668361068008,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993945994298662,
      "loss": 6.6387,
      "mean_token_accuracy": 0.11275321766734123,
      "num_tokens": 6853615.0,
      "step": 3160
    },
    {
      "entropy": 6.842955875396728,
      "epoch": 0.3387019102145647,
      "grad_norm": 0.875,
      "learning_rate": 0.0004993917933697484,
      "loss": 6.7319,
      "mean_token_accuracy": 0.11903002113103867,
      "num_tokens": 6865256.0,
      "step": 3165
    },
    {
      "entropy": 7.049095106124878,
      "epoch": 0.3392369843223286,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993889808303493,
      "loss": 6.7589,
      "mean_token_accuracy": 0.1182759940624237,
      "num_tokens": 6875358.0,
      "step": 3170
    },
    {
      "entropy": 6.767399740219116,
      "epoch": 0.33977205843009256,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004993861618117498,
      "loss": 6.6782,
      "mean_token_accuracy": 0.1162776917219162,
      "num_tokens": 6886584.0,
      "step": 3175
    },
    {
      "entropy": 6.765003776550293,
      "epoch": 0.3403071325378565,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993833363140314,
      "loss": 6.6082,
      "mean_token_accuracy": 0.12602533102035524,
      "num_tokens": 6896958.0,
      "step": 3180
    },
    {
      "entropy": 6.886105108261108,
      "epoch": 0.3408422066456204,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993805043372756,
      "loss": 6.6784,
      "mean_token_accuracy": 0.12520490884780883,
      "num_tokens": 6907944.0,
      "step": 3185
    },
    {
      "entropy": 6.877585315704346,
      "epoch": 0.34137728075338436,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004993776658815644,
      "loss": 6.7681,
      "mean_token_accuracy": 0.11567452028393746,
      "num_tokens": 6919486.0,
      "step": 3190
    },
    {
      "entropy": 6.879427862167359,
      "epoch": 0.3419123548611483,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004993748209469796,
      "loss": 6.7403,
      "mean_token_accuracy": 0.12306578159332275,
      "num_tokens": 6930650.0,
      "step": 3195
    },
    {
      "entropy": 6.7968353748321535,
      "epoch": 0.34244742896891217,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993719695336033,
      "loss": 6.6739,
      "mean_token_accuracy": 0.11598326042294502,
      "num_tokens": 6941519.0,
      "step": 3200
    },
    {
      "entropy": 6.843740224838257,
      "epoch": 0.3429825030766761,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993691116415179,
      "loss": 6.7212,
      "mean_token_accuracy": 0.11809414103627205,
      "num_tokens": 6952952.0,
      "step": 3205
    },
    {
      "entropy": 6.8387915134429935,
      "epoch": 0.34351757718444004,
      "grad_norm": 0.859375,
      "learning_rate": 0.000499366247270806,
      "loss": 6.6856,
      "mean_token_accuracy": 0.12287690415978432,
      "num_tokens": 6962867.0,
      "step": 3210
    },
    {
      "entropy": 6.79352445602417,
      "epoch": 0.34405265129220397,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993633764215502,
      "loss": 6.6226,
      "mean_token_accuracy": 0.12643782943487167,
      "num_tokens": 6973434.0,
      "step": 3215
    },
    {
      "entropy": 6.819111680984497,
      "epoch": 0.3445877253999679,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993604990938336,
      "loss": 6.604,
      "mean_token_accuracy": 0.13255805000662804,
      "num_tokens": 6984161.0,
      "step": 3220
    },
    {
      "entropy": 6.868915271759033,
      "epoch": 0.34512279950773184,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004993576152877389,
      "loss": 6.6275,
      "mean_token_accuracy": 0.1198020301759243,
      "num_tokens": 6994700.0,
      "step": 3225
    },
    {
      "entropy": 6.828886413574219,
      "epoch": 0.34565787361549577,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004993547250033496,
      "loss": 6.7612,
      "mean_token_accuracy": 0.11038903146982193,
      "num_tokens": 7006497.0,
      "step": 3230
    },
    {
      "entropy": 6.8887570858001705,
      "epoch": 0.34619294772325965,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993518282407493,
      "loss": 6.6798,
      "mean_token_accuracy": 0.1266617774963379,
      "num_tokens": 7016415.0,
      "step": 3235
    },
    {
      "entropy": 6.814205265045166,
      "epoch": 0.3467280218310236,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004993489250000213,
      "loss": 6.6303,
      "mean_token_accuracy": 0.11938185915350914,
      "num_tokens": 7026610.0,
      "step": 3240
    },
    {
      "entropy": 6.814730072021485,
      "epoch": 0.3472630959387875,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993460152812497,
      "loss": 6.639,
      "mean_token_accuracy": 0.11687793508172035,
      "num_tokens": 7038236.0,
      "step": 3245
    },
    {
      "entropy": 6.799622106552124,
      "epoch": 0.34779817004655145,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993430990845184,
      "loss": 6.6873,
      "mean_token_accuracy": 0.1299070708453655,
      "num_tokens": 7049438.0,
      "step": 3250
    },
    {
      "entropy": 6.822757863998413,
      "epoch": 0.3483332441543154,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004993401764099118,
      "loss": 6.6614,
      "mean_token_accuracy": 0.11520971059799194,
      "num_tokens": 7061001.0,
      "step": 3255
    },
    {
      "entropy": 6.826554918289185,
      "epoch": 0.3488683182620793,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004993372472575139,
      "loss": 6.6494,
      "mean_token_accuracy": 0.11807417050004006,
      "num_tokens": 7071328.0,
      "step": 3260
    },
    {
      "entropy": 6.862842416763305,
      "epoch": 0.34940339236984325,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993343116274096,
      "loss": 6.8306,
      "mean_token_accuracy": 0.11384405344724655,
      "num_tokens": 7082597.0,
      "step": 3265
    },
    {
      "entropy": 6.815863227844238,
      "epoch": 0.3499384664776071,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004993313695196836,
      "loss": 6.6586,
      "mean_token_accuracy": 0.11965244635939598,
      "num_tokens": 7093463.0,
      "step": 3270
    },
    {
      "entropy": 6.786405324935913,
      "epoch": 0.35047354058537106,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004993284209344208,
      "loss": 6.7219,
      "mean_token_accuracy": 0.12500518634915353,
      "num_tokens": 7104350.0,
      "step": 3275
    },
    {
      "entropy": 6.882570743560791,
      "epoch": 0.351008614693135,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993254658717065,
      "loss": 6.661,
      "mean_token_accuracy": 0.1239277258515358,
      "num_tokens": 7114851.0,
      "step": 3280
    },
    {
      "entropy": 6.823478555679321,
      "epoch": 0.3515436888008989,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004993225043316257,
      "loss": 6.7135,
      "mean_token_accuracy": 0.11792605891823768,
      "num_tokens": 7126440.0,
      "step": 3285
    },
    {
      "entropy": 6.880914402008057,
      "epoch": 0.35207876290866286,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004993195363142643,
      "loss": 6.7407,
      "mean_token_accuracy": 0.11478666737675666,
      "num_tokens": 7137884.0,
      "step": 3290
    },
    {
      "entropy": 6.892389440536499,
      "epoch": 0.3526138370164268,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993165618197076,
      "loss": 6.7197,
      "mean_token_accuracy": 0.1155787654221058,
      "num_tokens": 7147930.0,
      "step": 3295
    },
    {
      "entropy": 6.791791582107544,
      "epoch": 0.3531489111241907,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993135808480418,
      "loss": 6.5753,
      "mean_token_accuracy": 0.129153461009264,
      "num_tokens": 7159200.0,
      "step": 3300
    },
    {
      "entropy": 6.859189891815186,
      "epoch": 0.3536839852319546,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993105933993529,
      "loss": 6.6683,
      "mean_token_accuracy": 0.12259154245257378,
      "num_tokens": 7169601.0,
      "step": 3305
    },
    {
      "entropy": 6.754944276809693,
      "epoch": 0.35421905933971853,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004993075994737271,
      "loss": 6.6144,
      "mean_token_accuracy": 0.12793416306376457,
      "num_tokens": 7180465.0,
      "step": 3310
    },
    {
      "entropy": 6.840865564346314,
      "epoch": 0.35475413344748247,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004993045990712508,
      "loss": 6.6155,
      "mean_token_accuracy": 0.12482116669416428,
      "num_tokens": 7190603.0,
      "step": 3315
    },
    {
      "entropy": 6.774190092086792,
      "epoch": 0.3552892075552464,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004993015921920109,
      "loss": 6.5848,
      "mean_token_accuracy": 0.1270811952650547,
      "num_tokens": 7201017.0,
      "step": 3320
    },
    {
      "entropy": 6.705060338973999,
      "epoch": 0.35582428166301033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499298578836094,
      "loss": 6.5837,
      "mean_token_accuracy": 0.12208671867847443,
      "num_tokens": 7210908.0,
      "step": 3325
    },
    {
      "entropy": 6.868721342086792,
      "epoch": 0.35635935577077427,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499295559003587,
      "loss": 6.6211,
      "mean_token_accuracy": 0.12420659512281418,
      "num_tokens": 7221306.0,
      "step": 3330
    },
    {
      "entropy": 6.7805100440979,
      "epoch": 0.3568944298785382,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004992925326945774,
      "loss": 6.7391,
      "mean_token_accuracy": 0.11606214642524719,
      "num_tokens": 7233076.0,
      "step": 3335
    },
    {
      "entropy": 6.79472131729126,
      "epoch": 0.3574295039863021,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992894999091524,
      "loss": 6.7376,
      "mean_token_accuracy": 0.11862620785832405,
      "num_tokens": 7244703.0,
      "step": 3340
    },
    {
      "entropy": 6.858727550506591,
      "epoch": 0.357964578094066,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992864606473996,
      "loss": 6.7249,
      "mean_token_accuracy": 0.11088455319404603,
      "num_tokens": 7255438.0,
      "step": 3345
    },
    {
      "entropy": 6.8269633769989015,
      "epoch": 0.35849965220182994,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992834149094069,
      "loss": 6.6592,
      "mean_token_accuracy": 0.12365302219986915,
      "num_tokens": 7265720.0,
      "step": 3350
    },
    {
      "entropy": 6.849551486968994,
      "epoch": 0.3590347263095939,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992803626952619,
      "loss": 6.7043,
      "mean_token_accuracy": 0.11888771429657936,
      "num_tokens": 7277791.0,
      "step": 3355
    },
    {
      "entropy": 6.7473889827728275,
      "epoch": 0.3595698004173578,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000499277304005053,
      "loss": 6.5728,
      "mean_token_accuracy": 0.1280308946967125,
      "num_tokens": 7288521.0,
      "step": 3360
    },
    {
      "entropy": 6.810279130935669,
      "epoch": 0.36010487452512174,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004992742388388686,
      "loss": 6.7666,
      "mean_token_accuracy": 0.11023237109184265,
      "num_tokens": 7299298.0,
      "step": 3365
    },
    {
      "entropy": 6.792523527145386,
      "epoch": 0.3606399486328857,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992711671967968,
      "loss": 6.595,
      "mean_token_accuracy": 0.11808668822050095,
      "num_tokens": 7310271.0,
      "step": 3370
    },
    {
      "entropy": 6.823890495300293,
      "epoch": 0.36117502274064955,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004992680890789268,
      "loss": 6.6509,
      "mean_token_accuracy": 0.12248839884996414,
      "num_tokens": 7321112.0,
      "step": 3375
    },
    {
      "entropy": 6.706878280639648,
      "epoch": 0.3617100968484135,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004992650044853471,
      "loss": 6.6047,
      "mean_token_accuracy": 0.12075349166989327,
      "num_tokens": 7331976.0,
      "step": 3380
    },
    {
      "entropy": 6.8689617156982425,
      "epoch": 0.3622451709561774,
      "grad_norm": 0.953125,
      "learning_rate": 0.000499261913416147,
      "loss": 6.7263,
      "mean_token_accuracy": 0.11769687235355378,
      "num_tokens": 7342800.0,
      "step": 3385
    },
    {
      "entropy": 6.7807615280151365,
      "epoch": 0.36278024506394135,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004992588158714157,
      "loss": 6.5487,
      "mean_token_accuracy": 0.12072362452745437,
      "num_tokens": 7354213.0,
      "step": 3390
    },
    {
      "entropy": 6.761278200149536,
      "epoch": 0.3633153191717053,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004992557118512424,
      "loss": 6.6253,
      "mean_token_accuracy": 0.12343727201223373,
      "num_tokens": 7363250.0,
      "step": 3395
    },
    {
      "entropy": 6.743581962585449,
      "epoch": 0.3638503932794692,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004992526013557171,
      "loss": 6.6557,
      "mean_token_accuracy": 0.1214616447687149,
      "num_tokens": 7374185.0,
      "step": 3400
    },
    {
      "entropy": 6.7366564750671385,
      "epoch": 0.36438546738723315,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992494843849294,
      "loss": 6.6305,
      "mean_token_accuracy": 0.11691405922174454,
      "num_tokens": 7384579.0,
      "step": 3405
    },
    {
      "entropy": 6.881455659866333,
      "epoch": 0.36492054149499703,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004992463609389694,
      "loss": 6.6406,
      "mean_token_accuracy": 0.11755282357335091,
      "num_tokens": 7395277.0,
      "step": 3410
    },
    {
      "entropy": 6.706863021850586,
      "epoch": 0.36545561560276096,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004992432310179272,
      "loss": 6.6343,
      "mean_token_accuracy": 0.11993120685219764,
      "num_tokens": 7406688.0,
      "step": 3415
    },
    {
      "entropy": 6.736936712265015,
      "epoch": 0.3659906897105249,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004992400946218932,
      "loss": 6.6842,
      "mean_token_accuracy": 0.12183412909507751,
      "num_tokens": 7417378.0,
      "step": 3420
    },
    {
      "entropy": 6.972946405410767,
      "epoch": 0.36652576381828883,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004992369517509581,
      "loss": 6.7056,
      "mean_token_accuracy": 0.11991995200514793,
      "num_tokens": 7427675.0,
      "step": 3425
    },
    {
      "entropy": 6.802153635025024,
      "epoch": 0.36706083792605276,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004992338024052123,
      "loss": 6.7424,
      "mean_token_accuracy": 0.11765128523111343,
      "num_tokens": 7438083.0,
      "step": 3430
    },
    {
      "entropy": 6.767482089996338,
      "epoch": 0.3675959120338167,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004992306465847473,
      "loss": 6.5952,
      "mean_token_accuracy": 0.12695135846734046,
      "num_tokens": 7449051.0,
      "step": 3435
    },
    {
      "entropy": 6.7729590892791744,
      "epoch": 0.36813098614158063,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004992274842896536,
      "loss": 6.6246,
      "mean_token_accuracy": 0.12268055081367493,
      "num_tokens": 7459074.0,
      "step": 3440
    },
    {
      "entropy": 6.784332609176635,
      "epoch": 0.3686660602493445,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004992243155200229,
      "loss": 6.6902,
      "mean_token_accuracy": 0.12212735787034035,
      "num_tokens": 7470472.0,
      "step": 3445
    },
    {
      "entropy": 6.78359055519104,
      "epoch": 0.36920113435710844,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004992211402759466,
      "loss": 6.6568,
      "mean_token_accuracy": 0.12637588679790496,
      "num_tokens": 7481477.0,
      "step": 3450
    },
    {
      "entropy": 6.786507797241211,
      "epoch": 0.3697362084648724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004992179585575164,
      "loss": 6.6085,
      "mean_token_accuracy": 0.12437106668949127,
      "num_tokens": 7491906.0,
      "step": 3455
    },
    {
      "entropy": 6.794345140457153,
      "epoch": 0.3702712825726363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499214770364824,
      "loss": 6.5373,
      "mean_token_accuracy": 0.11893177703022957,
      "num_tokens": 7502447.0,
      "step": 3460
    },
    {
      "entropy": 6.721900892257691,
      "epoch": 0.37080635668040024,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004992115756979617,
      "loss": 6.553,
      "mean_token_accuracy": 0.129140467941761,
      "num_tokens": 7513505.0,
      "step": 3465
    },
    {
      "entropy": 6.754326629638672,
      "epoch": 0.3713414307881642,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004992083745570218,
      "loss": 6.5675,
      "mean_token_accuracy": 0.12708702757954599,
      "num_tokens": 7524884.0,
      "step": 3470
    },
    {
      "entropy": 6.7483885288238525,
      "epoch": 0.3718765048959281,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004992051669420964,
      "loss": 6.6312,
      "mean_token_accuracy": 0.12879489958286286,
      "num_tokens": 7535942.0,
      "step": 3475
    },
    {
      "entropy": 6.7121062755584715,
      "epoch": 0.372411579003692,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004992019528532785,
      "loss": 6.5951,
      "mean_token_accuracy": 0.12708142325282096,
      "num_tokens": 7545506.0,
      "step": 3480
    },
    {
      "entropy": 6.795954084396362,
      "epoch": 0.3729466531114559,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991987322906605,
      "loss": 6.6893,
      "mean_token_accuracy": 0.11757104098796844,
      "num_tokens": 7555328.0,
      "step": 3485
    },
    {
      "entropy": 6.849932861328125,
      "epoch": 0.37348172721921985,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991955052543357,
      "loss": 6.5845,
      "mean_token_accuracy": 0.12351070642471314,
      "num_tokens": 7565762.0,
      "step": 3490
    },
    {
      "entropy": 6.776114749908447,
      "epoch": 0.3740168013269838,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004991922717443972,
      "loss": 6.6028,
      "mean_token_accuracy": 0.12989014089107515,
      "num_tokens": 7575884.0,
      "step": 3495
    },
    {
      "entropy": 6.684484958648682,
      "epoch": 0.3745518754347477,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991890317609384,
      "loss": 6.6374,
      "mean_token_accuracy": 0.12035185471177101,
      "num_tokens": 7586100.0,
      "step": 3500
    },
    {
      "entropy": 6.829225254058838,
      "epoch": 0.37508694954251165,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004991857853040528,
      "loss": 6.6546,
      "mean_token_accuracy": 0.12205979079008103,
      "num_tokens": 7597137.0,
      "step": 3505
    },
    {
      "entropy": 6.783835411071777,
      "epoch": 0.3756220236502756,
      "grad_norm": 0.875,
      "learning_rate": 0.0004991825323738341,
      "loss": 6.676,
      "mean_token_accuracy": 0.12901032269001006,
      "num_tokens": 7607640.0,
      "step": 3510
    },
    {
      "entropy": 6.829906606674195,
      "epoch": 0.3761570977580395,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004991792729703763,
      "loss": 6.7337,
      "mean_token_accuracy": 0.12128165885806083,
      "num_tokens": 7618319.0,
      "step": 3515
    },
    {
      "entropy": 6.812218856811524,
      "epoch": 0.3766921718658034,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004991760070937734,
      "loss": 6.6028,
      "mean_token_accuracy": 0.12587715685367584,
      "num_tokens": 7628350.0,
      "step": 3520
    },
    {
      "entropy": 6.777668380737305,
      "epoch": 0.3772272459735673,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004991727347441199,
      "loss": 6.6396,
      "mean_token_accuracy": 0.11658850014209747,
      "num_tokens": 7637857.0,
      "step": 3525
    },
    {
      "entropy": 6.83516297340393,
      "epoch": 0.37776232008133126,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004991694559215102,
      "loss": 6.6543,
      "mean_token_accuracy": 0.12148249298334121,
      "num_tokens": 7648242.0,
      "step": 3530
    },
    {
      "entropy": 6.782798194885254,
      "epoch": 0.3782973941890952,
      "grad_norm": 0.734375,
      "learning_rate": 0.000499166170626039,
      "loss": 6.6481,
      "mean_token_accuracy": 0.12124886438250541,
      "num_tokens": 7660565.0,
      "step": 3535
    },
    {
      "entropy": 6.772379589080811,
      "epoch": 0.3788324682968591,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499162878857801,
      "loss": 6.6642,
      "mean_token_accuracy": 0.12547138035297395,
      "num_tokens": 7671600.0,
      "step": 3540
    },
    {
      "entropy": 6.786243486404419,
      "epoch": 0.37936754240462306,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991595806168915,
      "loss": 6.6761,
      "mean_token_accuracy": 0.11763999313116073,
      "num_tokens": 7682453.0,
      "step": 3545
    },
    {
      "entropy": 6.761963129043579,
      "epoch": 0.379902616512387,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004991562759034055,
      "loss": 6.6056,
      "mean_token_accuracy": 0.12182423919439316,
      "num_tokens": 7693490.0,
      "step": 3550
    },
    {
      "entropy": 6.8190428733825685,
      "epoch": 0.38043769062015087,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991529647174386,
      "loss": 6.5641,
      "mean_token_accuracy": 0.12036292180418968,
      "num_tokens": 7703478.0,
      "step": 3555
    },
    {
      "entropy": 6.775476408004761,
      "epoch": 0.3809727647279148,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004991496470590864,
      "loss": 6.6654,
      "mean_token_accuracy": 0.1186753623187542,
      "num_tokens": 7714647.0,
      "step": 3560
    },
    {
      "entropy": 6.7765895366668705,
      "epoch": 0.38150783883567874,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991463229284445,
      "loss": 6.6069,
      "mean_token_accuracy": 0.1255822166800499,
      "num_tokens": 7725528.0,
      "step": 3565
    },
    {
      "entropy": 6.741648101806641,
      "epoch": 0.38204291294344267,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004991429923256091,
      "loss": 6.6474,
      "mean_token_accuracy": 0.12289627566933632,
      "num_tokens": 7737708.0,
      "step": 3570
    },
    {
      "entropy": 6.807142543792724,
      "epoch": 0.3825779870512066,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004991396552506762,
      "loss": 6.642,
      "mean_token_accuracy": 0.11627409681677818,
      "num_tokens": 7749093.0,
      "step": 3575
    },
    {
      "entropy": 6.73176646232605,
      "epoch": 0.38311306115897054,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004991363117037423,
      "loss": 6.5812,
      "mean_token_accuracy": 0.12449619993567466,
      "num_tokens": 7760024.0,
      "step": 3580
    },
    {
      "entropy": 6.7203834533691404,
      "epoch": 0.38364813526673447,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004991329616849039,
      "loss": 6.619,
      "mean_token_accuracy": 0.12767036110162736,
      "num_tokens": 7770801.0,
      "step": 3585
    },
    {
      "entropy": 6.82001256942749,
      "epoch": 0.38418320937449835,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004991296051942578,
      "loss": 6.6721,
      "mean_token_accuracy": 0.1217954471707344,
      "num_tokens": 7782257.0,
      "step": 3590
    },
    {
      "entropy": 6.768124771118164,
      "epoch": 0.3847182834822623,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004991262422319008,
      "loss": 6.5965,
      "mean_token_accuracy": 0.12531521171331406,
      "num_tokens": 7793016.0,
      "step": 3595
    },
    {
      "entropy": 6.751244163513183,
      "epoch": 0.3852533575900262,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00049912287279793,
      "loss": 6.5442,
      "mean_token_accuracy": 0.13346222937107086,
      "num_tokens": 7802207.0,
      "step": 3600
    },
    {
      "entropy": 6.784510135650635,
      "epoch": 0.38578843169779015,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991194968924429,
      "loss": 6.656,
      "mean_token_accuracy": 0.12131711095571518,
      "num_tokens": 7813487.0,
      "step": 3605
    },
    {
      "entropy": 6.82340726852417,
      "epoch": 0.3863235058055541,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004991161145155367,
      "loss": 6.6398,
      "mean_token_accuracy": 0.1207100585103035,
      "num_tokens": 7824587.0,
      "step": 3610
    },
    {
      "entropy": 6.838507223129272,
      "epoch": 0.386858579913318,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004991127256673091,
      "loss": 6.6259,
      "mean_token_accuracy": 0.11949132606387139,
      "num_tokens": 7836240.0,
      "step": 3615
    },
    {
      "entropy": 6.724960899353027,
      "epoch": 0.38739365402108195,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004991093303478582,
      "loss": 6.5721,
      "mean_token_accuracy": 0.12099727615714073,
      "num_tokens": 7847014.0,
      "step": 3620
    },
    {
      "entropy": 6.755022096633911,
      "epoch": 0.3879287281288458,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004991059285572818,
      "loss": 6.6905,
      "mean_token_accuracy": 0.12505315169692038,
      "num_tokens": 7857899.0,
      "step": 3625
    },
    {
      "entropy": 6.805648851394653,
      "epoch": 0.38846380223660976,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004991025202956781,
      "loss": 6.5859,
      "mean_token_accuracy": 0.1255096547305584,
      "num_tokens": 7868987.0,
      "step": 3630
    },
    {
      "entropy": 6.84774341583252,
      "epoch": 0.3889988763443737,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004990991055631457,
      "loss": 6.7064,
      "mean_token_accuracy": 0.1195052333176136,
      "num_tokens": 7879547.0,
      "step": 3635
    },
    {
      "entropy": 6.737785577774048,
      "epoch": 0.3895339504521376,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004990956843597832,
      "loss": 6.5782,
      "mean_token_accuracy": 0.12665883377194403,
      "num_tokens": 7889824.0,
      "step": 3640
    },
    {
      "entropy": 6.662966775894165,
      "epoch": 0.39006902455990156,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004990922566856892,
      "loss": 6.5733,
      "mean_token_accuracy": 0.11881136298179626,
      "num_tokens": 7901156.0,
      "step": 3645
    },
    {
      "entropy": 6.7981789112091064,
      "epoch": 0.3906040986676655,
      "grad_norm": 1.0,
      "learning_rate": 0.0004990888225409629,
      "loss": 6.592,
      "mean_token_accuracy": 0.12195635959506035,
      "num_tokens": 7911695.0,
      "step": 3650
    },
    {
      "entropy": 6.686615133285523,
      "epoch": 0.3911391727754294,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004990853819257033,
      "loss": 6.5732,
      "mean_token_accuracy": 0.13316732421517372,
      "num_tokens": 7923559.0,
      "step": 3655
    },
    {
      "entropy": 6.842485618591309,
      "epoch": 0.3916742468831933,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004990819348400097,
      "loss": 6.6183,
      "mean_token_accuracy": 0.12080760896205903,
      "num_tokens": 7935792.0,
      "step": 3660
    },
    {
      "entropy": 6.804695320129395,
      "epoch": 0.39220932099095723,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004990784812839818,
      "loss": 6.5951,
      "mean_token_accuracy": 0.13260335549712182,
      "num_tokens": 7946987.0,
      "step": 3665
    },
    {
      "entropy": 6.670995664596558,
      "epoch": 0.39274439509872117,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004990750212577192,
      "loss": 6.5283,
      "mean_token_accuracy": 0.12538674250245094,
      "num_tokens": 7957160.0,
      "step": 3670
    },
    {
      "entropy": 6.772046279907227,
      "epoch": 0.3932794692064851,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004990715547613219,
      "loss": 6.6304,
      "mean_token_accuracy": 0.11679298281669617,
      "num_tokens": 7968833.0,
      "step": 3675
    },
    {
      "entropy": 6.8003373622894285,
      "epoch": 0.39381454331424903,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004990680817948898,
      "loss": 6.6755,
      "mean_token_accuracy": 0.11913765817880631,
      "num_tokens": 7980661.0,
      "step": 3680
    },
    {
      "entropy": 6.844438648223877,
      "epoch": 0.39434961742201297,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990646023585234,
      "loss": 6.5909,
      "mean_token_accuracy": 0.12314732074737549,
      "num_tokens": 7992677.0,
      "step": 3685
    },
    {
      "entropy": 6.688145542144776,
      "epoch": 0.3948846915297769,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004990611164523231,
      "loss": 6.57,
      "mean_token_accuracy": 0.1285789616405964,
      "num_tokens": 8002796.0,
      "step": 3690
    },
    {
      "entropy": 6.688539314270019,
      "epoch": 0.3954197656375408,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990576240763896,
      "loss": 6.5999,
      "mean_token_accuracy": 0.1252342328429222,
      "num_tokens": 8012779.0,
      "step": 3695
    },
    {
      "entropy": 6.76634202003479,
      "epoch": 0.3959548397453047,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004990541252308237,
      "loss": 6.5824,
      "mean_token_accuracy": 0.12815626487135887,
      "num_tokens": 8023442.0,
      "step": 3700
    },
    {
      "entropy": 6.820317983627319,
      "epoch": 0.39648991385306864,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004990506199157263,
      "loss": 6.6726,
      "mean_token_accuracy": 0.11957266852259636,
      "num_tokens": 8035163.0,
      "step": 3705
    },
    {
      "entropy": 6.777938270568848,
      "epoch": 0.3970249879608326,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004990471081311988,
      "loss": 6.5926,
      "mean_token_accuracy": 0.12503935247659684,
      "num_tokens": 8045330.0,
      "step": 3710
    },
    {
      "entropy": 6.772491693496704,
      "epoch": 0.3975600620685965,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004990435898773426,
      "loss": 6.6203,
      "mean_token_accuracy": 0.12175011858344079,
      "num_tokens": 8056154.0,
      "step": 3715
    },
    {
      "entropy": 6.784107732772827,
      "epoch": 0.39809513617636044,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004990400651542592,
      "loss": 6.5829,
      "mean_token_accuracy": 0.1277529016137123,
      "num_tokens": 8066101.0,
      "step": 3720
    },
    {
      "entropy": 6.751408624649048,
      "epoch": 0.3986302102841244,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004990365339620504,
      "loss": 6.6561,
      "mean_token_accuracy": 0.117050202190876,
      "num_tokens": 8077249.0,
      "step": 3725
    },
    {
      "entropy": 6.759656381607056,
      "epoch": 0.39916528439188825,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499032996300818,
      "loss": 6.5993,
      "mean_token_accuracy": 0.12702321633696556,
      "num_tokens": 8088532.0,
      "step": 3730
    },
    {
      "entropy": 6.800127220153809,
      "epoch": 0.3997003584996522,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990294521706645,
      "loss": 6.6469,
      "mean_token_accuracy": 0.11475934162735939,
      "num_tokens": 8099541.0,
      "step": 3735
    },
    {
      "entropy": 6.765775728225708,
      "epoch": 0.4002354326074161,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000499025901571692,
      "loss": 6.6743,
      "mean_token_accuracy": 0.12347187474370003,
      "num_tokens": 8110532.0,
      "step": 3740
    },
    {
      "entropy": 6.804666519165039,
      "epoch": 0.40077050671518005,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004990223445040031,
      "loss": 6.5906,
      "mean_token_accuracy": 0.12874274030327798,
      "num_tokens": 8120991.0,
      "step": 3745
    },
    {
      "entropy": 6.71906795501709,
      "epoch": 0.401305580822944,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004990187809677004,
      "loss": 6.5306,
      "mean_token_accuracy": 0.12286594063043595,
      "num_tokens": 8130737.0,
      "step": 3750
    },
    {
      "entropy": 6.6623273372650145,
      "epoch": 0.4018406549307079,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004990152109628869,
      "loss": 6.5363,
      "mean_token_accuracy": 0.12400009483098984,
      "num_tokens": 8142315.0,
      "step": 3755
    },
    {
      "entropy": 6.764507102966308,
      "epoch": 0.40237572903847185,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990116344896657,
      "loss": 6.6172,
      "mean_token_accuracy": 0.11884014382958412,
      "num_tokens": 8153618.0,
      "step": 3760
    },
    {
      "entropy": 6.701496410369873,
      "epoch": 0.40291080314623573,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00049900805154814,
      "loss": 6.5736,
      "mean_token_accuracy": 0.13147677332162858,
      "num_tokens": 8164304.0,
      "step": 3765
    },
    {
      "entropy": 6.719698762893676,
      "epoch": 0.40344587725399966,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004990044621384134,
      "loss": 6.5691,
      "mean_token_accuracy": 0.1204621210694313,
      "num_tokens": 8175289.0,
      "step": 3770
    },
    {
      "entropy": 6.799706554412841,
      "epoch": 0.4039809513617636,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004990008662605893,
      "loss": 6.618,
      "mean_token_accuracy": 0.1192642904818058,
      "num_tokens": 8186274.0,
      "step": 3775
    },
    {
      "entropy": 6.766622829437256,
      "epoch": 0.40451602546952753,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989972639147717,
      "loss": 6.5458,
      "mean_token_accuracy": 0.12909928560256959,
      "num_tokens": 8196380.0,
      "step": 3780
    },
    {
      "entropy": 6.709709596633911,
      "epoch": 0.40505109957729146,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004989936551010646,
      "loss": 6.6212,
      "mean_token_accuracy": 0.12540413141250611,
      "num_tokens": 8207204.0,
      "step": 3785
    },
    {
      "entropy": 6.662211322784424,
      "epoch": 0.4055861736850554,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989900398195722,
      "loss": 6.5364,
      "mean_token_accuracy": 0.12927303686738015,
      "num_tokens": 8217745.0,
      "step": 3790
    },
    {
      "entropy": 6.708821868896484,
      "epoch": 0.40612124779281933,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989864180703988,
      "loss": 6.5265,
      "mean_token_accuracy": 0.12706319689750672,
      "num_tokens": 8228171.0,
      "step": 3795
    },
    {
      "entropy": 6.718810129165649,
      "epoch": 0.4066563219005832,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004989827898536491,
      "loss": 6.5805,
      "mean_token_accuracy": 0.11968392059206963,
      "num_tokens": 8239220.0,
      "step": 3800
    },
    {
      "entropy": 6.75914945602417,
      "epoch": 0.40719139600834714,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004989791551694279,
      "loss": 6.6392,
      "mean_token_accuracy": 0.12101104408502579,
      "num_tokens": 8250363.0,
      "step": 3805
    },
    {
      "entropy": 6.7193773746490475,
      "epoch": 0.4077264701161111,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004989755140178398,
      "loss": 6.5934,
      "mean_token_accuracy": 0.12401422187685966,
      "num_tokens": 8261207.0,
      "step": 3810
    },
    {
      "entropy": 6.687061738967896,
      "epoch": 0.408261544223875,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004989718663989905,
      "loss": 6.4898,
      "mean_token_accuracy": 0.12349989488720894,
      "num_tokens": 8272343.0,
      "step": 3815
    },
    {
      "entropy": 6.696186542510986,
      "epoch": 0.40879661833163894,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989682123129849,
      "loss": 6.5487,
      "mean_token_accuracy": 0.12425093948841096,
      "num_tokens": 8282677.0,
      "step": 3820
    },
    {
      "entropy": 6.667926263809204,
      "epoch": 0.40933169243940287,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004989645517599286,
      "loss": 6.5614,
      "mean_token_accuracy": 0.12363334447145462,
      "num_tokens": 8292753.0,
      "step": 3825
    },
    {
      "entropy": 6.730340671539307,
      "epoch": 0.4098667665471668,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004989608847399275,
      "loss": 6.5546,
      "mean_token_accuracy": 0.12061431556940079,
      "num_tokens": 8302747.0,
      "step": 3830
    },
    {
      "entropy": 6.735944366455078,
      "epoch": 0.4104018406549307,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004989572112530871,
      "loss": 6.6175,
      "mean_token_accuracy": 0.11923456266522407,
      "num_tokens": 8314766.0,
      "step": 3835
    },
    {
      "entropy": 6.797013807296753,
      "epoch": 0.4109369147626946,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004989535312995139,
      "loss": 6.655,
      "mean_token_accuracy": 0.12089989259839058,
      "num_tokens": 8325791.0,
      "step": 3840
    },
    {
      "entropy": 6.710721731185913,
      "epoch": 0.41147198887045855,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004989498448793139,
      "loss": 6.5712,
      "mean_token_accuracy": 0.12769034430384635,
      "num_tokens": 8335809.0,
      "step": 3845
    },
    {
      "entropy": 6.67577633857727,
      "epoch": 0.4120070629782225,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004989461519925935,
      "loss": 6.5827,
      "mean_token_accuracy": 0.12097218707203865,
      "num_tokens": 8347505.0,
      "step": 3850
    },
    {
      "entropy": 6.75542802810669,
      "epoch": 0.4125421370859864,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989424526394596,
      "loss": 6.5775,
      "mean_token_accuracy": 0.12156168222427369,
      "num_tokens": 8358363.0,
      "step": 3855
    },
    {
      "entropy": 6.7438743114471436,
      "epoch": 0.41307721119375035,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004989387468200187,
      "loss": 6.5988,
      "mean_token_accuracy": 0.13023255839943887,
      "num_tokens": 8369076.0,
      "step": 3860
    },
    {
      "entropy": 6.752788734436035,
      "epoch": 0.4136122853015143,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000498935034534378,
      "loss": 6.6291,
      "mean_token_accuracy": 0.12350267320871353,
      "num_tokens": 8380071.0,
      "step": 3865
    },
    {
      "entropy": 6.721376562118531,
      "epoch": 0.41414735940927816,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004989313157826446,
      "loss": 6.5938,
      "mean_token_accuracy": 0.12291992232203483,
      "num_tokens": 8390574.0,
      "step": 3870
    },
    {
      "entropy": 6.705049705505371,
      "epoch": 0.4146824335170421,
      "grad_norm": 0.8125,
      "learning_rate": 0.000498927590564926,
      "loss": 6.5749,
      "mean_token_accuracy": 0.12339994534850121,
      "num_tokens": 8401154.0,
      "step": 3875
    },
    {
      "entropy": 6.766706037521362,
      "epoch": 0.415217507624806,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004989238588813296,
      "loss": 6.6672,
      "mean_token_accuracy": 0.11639968454837799,
      "num_tokens": 8412113.0,
      "step": 3880
    },
    {
      "entropy": 6.797072076797486,
      "epoch": 0.41575258173256996,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004989201207319634,
      "loss": 6.6385,
      "mean_token_accuracy": 0.12364522740244865,
      "num_tokens": 8422734.0,
      "step": 3885
    },
    {
      "entropy": 6.736127710342407,
      "epoch": 0.4162876558403339,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004989163761169351,
      "loss": 6.5662,
      "mean_token_accuracy": 0.12166397422552108,
      "num_tokens": 8433647.0,
      "step": 3890
    },
    {
      "entropy": 6.717603731155395,
      "epoch": 0.4168227299480978,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004989126250363529,
      "loss": 6.6043,
      "mean_token_accuracy": 0.12411727085709572,
      "num_tokens": 8444726.0,
      "step": 3895
    },
    {
      "entropy": 6.732759284973144,
      "epoch": 0.41735780405586176,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004989088674903251,
      "loss": 6.6042,
      "mean_token_accuracy": 0.12171476036310196,
      "num_tokens": 8455351.0,
      "step": 3900
    },
    {
      "entropy": 6.69054012298584,
      "epoch": 0.41789287816362564,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004989051034789602,
      "loss": 6.5945,
      "mean_token_accuracy": 0.12778979167342186,
      "num_tokens": 8466423.0,
      "step": 3905
    },
    {
      "entropy": 6.696830415725708,
      "epoch": 0.41842795227138957,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000498901333002367,
      "loss": 6.5204,
      "mean_token_accuracy": 0.12430474385619164,
      "num_tokens": 8475841.0,
      "step": 3910
    },
    {
      "entropy": 6.767647314071655,
      "epoch": 0.4189630263791535,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004988975560606541,
      "loss": 6.6309,
      "mean_token_accuracy": 0.12111587822437286,
      "num_tokens": 8487656.0,
      "step": 3915
    },
    {
      "entropy": 6.737721538543701,
      "epoch": 0.41949810048691744,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004988937726539308,
      "loss": 6.6345,
      "mean_token_accuracy": 0.11631876155734062,
      "num_tokens": 8497959.0,
      "step": 3920
    },
    {
      "entropy": 6.702249479293823,
      "epoch": 0.42003317459468137,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004988899827823062,
      "loss": 6.5598,
      "mean_token_accuracy": 0.12661055102944374,
      "num_tokens": 8508228.0,
      "step": 3925
    },
    {
      "entropy": 6.742639064788818,
      "epoch": 0.4205682487024453,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00049888618644589,
      "loss": 6.5927,
      "mean_token_accuracy": 0.11702611446380615,
      "num_tokens": 8518999.0,
      "step": 3930
    },
    {
      "entropy": 6.690499782562256,
      "epoch": 0.42110332281020924,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004988823836447916,
      "loss": 6.6409,
      "mean_token_accuracy": 0.12285174503922462,
      "num_tokens": 8529583.0,
      "step": 3935
    },
    {
      "entropy": 6.824066638946533,
      "epoch": 0.4216383969179731,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004988785743791207,
      "loss": 6.6921,
      "mean_token_accuracy": 0.11692366898059844,
      "num_tokens": 8540069.0,
      "step": 3940
    },
    {
      "entropy": 6.752025747299195,
      "epoch": 0.42217347102573705,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004988747586489876,
      "loss": 6.5824,
      "mean_token_accuracy": 0.12357909306883812,
      "num_tokens": 8550441.0,
      "step": 3945
    },
    {
      "entropy": 6.681799650192261,
      "epoch": 0.422708545133501,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000498870936454502,
      "loss": 6.543,
      "mean_token_accuracy": 0.13031908869743347,
      "num_tokens": 8560257.0,
      "step": 3950
    },
    {
      "entropy": 6.709078216552735,
      "epoch": 0.4232436192412649,
      "grad_norm": 0.875,
      "learning_rate": 0.0004988671077957749,
      "loss": 6.6124,
      "mean_token_accuracy": 0.12646457999944688,
      "num_tokens": 8570991.0,
      "step": 3955
    },
    {
      "entropy": 6.775239324569702,
      "epoch": 0.42377869334902885,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004988632726729164,
      "loss": 6.6754,
      "mean_token_accuracy": 0.1174784742295742,
      "num_tokens": 8583668.0,
      "step": 3960
    },
    {
      "entropy": 6.746391296386719,
      "epoch": 0.4243137674567928,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004988594310860374,
      "loss": 6.6133,
      "mean_token_accuracy": 0.12330208793282509,
      "num_tokens": 8593727.0,
      "step": 3965
    },
    {
      "entropy": 6.668441295623779,
      "epoch": 0.4248488415645567,
      "grad_norm": 0.76953125,
      "learning_rate": 0.0004988555830352485,
      "loss": 6.5888,
      "mean_token_accuracy": 0.12464616522192955,
      "num_tokens": 8605470.0,
      "step": 3970
    },
    {
      "entropy": 6.752479600906372,
      "epoch": 0.4253839156723206,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004988517285206614,
      "loss": 6.5361,
      "mean_token_accuracy": 0.12684643566608428,
      "num_tokens": 8617176.0,
      "step": 3975
    },
    {
      "entropy": 6.705382680892944,
      "epoch": 0.4259189897800845,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004988478675423871,
      "loss": 6.6269,
      "mean_token_accuracy": 0.12260573580861092,
      "num_tokens": 8627756.0,
      "step": 3980
    },
    {
      "entropy": 6.75893235206604,
      "epoch": 0.42645406388784846,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988440001005369,
      "loss": 6.5524,
      "mean_token_accuracy": 0.12638817578554154,
      "num_tokens": 8637766.0,
      "step": 3985
    },
    {
      "entropy": 6.738568544387817,
      "epoch": 0.4269891379956124,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004988401261952227,
      "loss": 6.5903,
      "mean_token_accuracy": 0.12412031963467599,
      "num_tokens": 8648562.0,
      "step": 3990
    },
    {
      "entropy": 6.768758535385132,
      "epoch": 0.4275242121033763,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004988362458265565,
      "loss": 6.5978,
      "mean_token_accuracy": 0.1206689141690731,
      "num_tokens": 8658393.0,
      "step": 3995
    },
    {
      "entropy": 6.72909688949585,
      "epoch": 0.42805928621114026,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004988323589946499,
      "loss": 6.5505,
      "mean_token_accuracy": 0.12452557906508446,
      "num_tokens": 8668365.0,
      "step": 4000
    },
    {
      "entropy": 6.706186580657959,
      "epoch": 0.4285943603189042,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988284656996156,
      "loss": 6.5742,
      "mean_token_accuracy": 0.12101777195930481,
      "num_tokens": 8679228.0,
      "step": 4005
    },
    {
      "entropy": 6.660093975067139,
      "epoch": 0.42912943442666807,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004988245659415656,
      "loss": 6.4583,
      "mean_token_accuracy": 0.132504703104496,
      "num_tokens": 8688940.0,
      "step": 4010
    },
    {
      "entropy": 6.755603170394897,
      "epoch": 0.429664508534432,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004988206597206129,
      "loss": 6.6261,
      "mean_token_accuracy": 0.11976755112409591,
      "num_tokens": 8699923.0,
      "step": 4015
    },
    {
      "entropy": 6.665601968765259,
      "epoch": 0.43019958264219593,
      "grad_norm": 0.875,
      "learning_rate": 0.00049881674703687,
      "loss": 6.4542,
      "mean_token_accuracy": 0.12824729904532434,
      "num_tokens": 8710445.0,
      "step": 4020
    },
    {
      "entropy": 6.662732791900635,
      "epoch": 0.43073465674995987,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004988128278904499,
      "loss": 6.5787,
      "mean_token_accuracy": 0.12728943005204202,
      "num_tokens": 8720686.0,
      "step": 4025
    },
    {
      "entropy": 6.717475557327271,
      "epoch": 0.4312697308577238,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000498808902281466,
      "loss": 6.5936,
      "mean_token_accuracy": 0.12410746738314629,
      "num_tokens": 8732122.0,
      "step": 4030
    },
    {
      "entropy": 6.69724588394165,
      "epoch": 0.43180480496548773,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004988049702100315,
      "loss": 6.5563,
      "mean_token_accuracy": 0.12644634395837784,
      "num_tokens": 8744122.0,
      "step": 4035
    },
    {
      "entropy": 6.75086350440979,
      "epoch": 0.43233987907325166,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004988010316762598,
      "loss": 6.5638,
      "mean_token_accuracy": 0.1258111245930195,
      "num_tokens": 8754758.0,
      "step": 4040
    },
    {
      "entropy": 6.65916633605957,
      "epoch": 0.4328749531810156,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004987970866802649,
      "loss": 6.5892,
      "mean_token_accuracy": 0.11819620206952094,
      "num_tokens": 8766577.0,
      "step": 4045
    },
    {
      "entropy": 6.749998998641968,
      "epoch": 0.4334100272887795,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987931352221605,
      "loss": 6.5695,
      "mean_token_accuracy": 0.12269530296325684,
      "num_tokens": 8777698.0,
      "step": 4050
    },
    {
      "entropy": 6.717717838287354,
      "epoch": 0.4339451013965434,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004987891773020606,
      "loss": 6.5937,
      "mean_token_accuracy": 0.12456514909863473,
      "num_tokens": 8788637.0,
      "step": 4055
    },
    {
      "entropy": 6.6792045593261715,
      "epoch": 0.43448017550430734,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004987852129200799,
      "loss": 6.5714,
      "mean_token_accuracy": 0.12965806424617768,
      "num_tokens": 8799709.0,
      "step": 4060
    },
    {
      "entropy": 6.719041061401367,
      "epoch": 0.4350152496120713,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987812420763326,
      "loss": 6.5766,
      "mean_token_accuracy": 0.12700210288167,
      "num_tokens": 8810469.0,
      "step": 4065
    },
    {
      "entropy": 6.795799016952515,
      "epoch": 0.4355503237198352,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004987772647709333,
      "loss": 6.5356,
      "mean_token_accuracy": 0.12159947156906128,
      "num_tokens": 8821142.0,
      "step": 4070
    },
    {
      "entropy": 6.621897602081299,
      "epoch": 0.43608539782759914,
      "grad_norm": 0.81640625,
      "learning_rate": 0.000498773281003997,
      "loss": 6.4949,
      "mean_token_accuracy": 0.13529240638017653,
      "num_tokens": 8832951.0,
      "step": 4075
    },
    {
      "entropy": 6.6723052024841305,
      "epoch": 0.4366204719353631,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004987692907756385,
      "loss": 6.6075,
      "mean_token_accuracy": 0.11884131506085396,
      "num_tokens": 8843579.0,
      "step": 4080
    },
    {
      "entropy": 6.653975200653076,
      "epoch": 0.43715554604312695,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004987652940859733,
      "loss": 6.5615,
      "mean_token_accuracy": 0.11667362749576568,
      "num_tokens": 8854612.0,
      "step": 4085
    },
    {
      "entropy": 6.680628633499145,
      "epoch": 0.4376906201508909,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004987612909351167,
      "loss": 6.4945,
      "mean_token_accuracy": 0.12707522585988046,
      "num_tokens": 8865584.0,
      "step": 4090
    },
    {
      "entropy": 6.665642642974854,
      "epoch": 0.4382256942586548,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987572813231842,
      "loss": 6.5216,
      "mean_token_accuracy": 0.13140948340296746,
      "num_tokens": 8876662.0,
      "step": 4095
    },
    {
      "entropy": 6.685274982452393,
      "epoch": 0.43876076836641875,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004987532652502917,
      "loss": 6.5656,
      "mean_token_accuracy": 0.12099597677588463,
      "num_tokens": 8887249.0,
      "step": 4100
    },
    {
      "entropy": 6.698389673233033,
      "epoch": 0.4392958424741827,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987492427165551,
      "loss": 6.3869,
      "mean_token_accuracy": 0.13985745757818221,
      "num_tokens": 8898635.0,
      "step": 4105
    },
    {
      "entropy": 6.655579090118408,
      "epoch": 0.4398309165819466,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004987452137220906,
      "loss": 6.5696,
      "mean_token_accuracy": 0.12459381148219109,
      "num_tokens": 8908794.0,
      "step": 4110
    },
    {
      "entropy": 6.805545902252197,
      "epoch": 0.44036599068971055,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004987411782670144,
      "loss": 6.5815,
      "mean_token_accuracy": 0.12002512514591217,
      "num_tokens": 8919015.0,
      "step": 4115
    },
    {
      "entropy": 6.803833293914795,
      "epoch": 0.44090106479747443,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004987371363514431,
      "loss": 6.6185,
      "mean_token_accuracy": 0.11796738654375076,
      "num_tokens": 8929028.0,
      "step": 4120
    },
    {
      "entropy": 6.633229732513428,
      "epoch": 0.44143613890523836,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004987330879754933,
      "loss": 6.5136,
      "mean_token_accuracy": 0.12121351808309555,
      "num_tokens": 8939789.0,
      "step": 4125
    },
    {
      "entropy": 6.659311389923095,
      "epoch": 0.4419712130130023,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004987290331392823,
      "loss": 6.4902,
      "mean_token_accuracy": 0.13035049065947532,
      "num_tokens": 8950657.0,
      "step": 4130
    },
    {
      "entropy": 6.771459245681763,
      "epoch": 0.44250628712076623,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004987249718429269,
      "loss": 6.5966,
      "mean_token_accuracy": 0.12920275554060937,
      "num_tokens": 8960744.0,
      "step": 4135
    },
    {
      "entropy": 6.6667276382446286,
      "epoch": 0.44304136122853016,
      "grad_norm": 0.875,
      "learning_rate": 0.0004987209040865441,
      "loss": 6.4554,
      "mean_token_accuracy": 0.13071943819522858,
      "num_tokens": 8972411.0,
      "step": 4140
    },
    {
      "entropy": 6.691996479034424,
      "epoch": 0.4435764353362941,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004987168298702519,
      "loss": 6.6141,
      "mean_token_accuracy": 0.12496730759739876,
      "num_tokens": 8982818.0,
      "step": 4145
    },
    {
      "entropy": 6.736303377151489,
      "epoch": 0.44411150944405803,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004987127491941675,
      "loss": 6.5332,
      "mean_token_accuracy": 0.12681084722280503,
      "num_tokens": 8993227.0,
      "step": 4150
    },
    {
      "entropy": 6.658869075775146,
      "epoch": 0.4446465835518219,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004987086620584091,
      "loss": 6.49,
      "mean_token_accuracy": 0.12491258159279824,
      "num_tokens": 9003796.0,
      "step": 4155
    },
    {
      "entropy": 6.673899793624878,
      "epoch": 0.44518165765958584,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004987045684630944,
      "loss": 6.493,
      "mean_token_accuracy": 0.13490606471896172,
      "num_tokens": 9014931.0,
      "step": 4160
    },
    {
      "entropy": 6.723305988311767,
      "epoch": 0.44571673176734977,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004987004684083417,
      "loss": 6.568,
      "mean_token_accuracy": 0.12488022521138191,
      "num_tokens": 9025312.0,
      "step": 4165
    },
    {
      "entropy": 6.641047430038452,
      "epoch": 0.4462518058751137,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004986963618942694,
      "loss": 6.55,
      "mean_token_accuracy": 0.13490004763007163,
      "num_tokens": 9035563.0,
      "step": 4170
    },
    {
      "entropy": 6.764650487899781,
      "epoch": 0.44678687998287764,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004986922489209961,
      "loss": 6.5782,
      "mean_token_accuracy": 0.1268523618578911,
      "num_tokens": 9046065.0,
      "step": 4175
    },
    {
      "entropy": 6.636542081832886,
      "epoch": 0.44732195409064157,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004986881294886406,
      "loss": 6.4771,
      "mean_token_accuracy": 0.13082251697778702,
      "num_tokens": 9056696.0,
      "step": 4180
    },
    {
      "entropy": 6.672894048690796,
      "epoch": 0.4478570281984055,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004986840035973218,
      "loss": 6.5487,
      "mean_token_accuracy": 0.12361067533493042,
      "num_tokens": 9067035.0,
      "step": 4185
    },
    {
      "entropy": 6.693661117553711,
      "epoch": 0.4483921023061694,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004986798712471588,
      "loss": 6.5097,
      "mean_token_accuracy": 0.12609453201293946,
      "num_tokens": 9078253.0,
      "step": 4190
    },
    {
      "entropy": 6.8062968254089355,
      "epoch": 0.4489271764139333,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986757324382709,
      "loss": 6.5838,
      "mean_token_accuracy": 0.11728138923645019,
      "num_tokens": 9088184.0,
      "step": 4195
    },
    {
      "entropy": 6.673310852050781,
      "epoch": 0.44946225052169725,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004986715871707777,
      "loss": 6.6401,
      "mean_token_accuracy": 0.11650597229599953,
      "num_tokens": 9099321.0,
      "step": 4200
    },
    {
      "entropy": 6.752178430557251,
      "epoch": 0.4499973246294612,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004986674354447988,
      "loss": 6.6718,
      "mean_token_accuracy": 0.12286324501037597,
      "num_tokens": 9111410.0,
      "step": 4205
    },
    {
      "entropy": 6.754798460006714,
      "epoch": 0.4505323987372251,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986632772604543,
      "loss": 6.4915,
      "mean_token_accuracy": 0.12803962752223014,
      "num_tokens": 9121932.0,
      "step": 4210
    },
    {
      "entropy": 6.554611539840698,
      "epoch": 0.45106747284498905,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986591126178639,
      "loss": 6.5716,
      "mean_token_accuracy": 0.13324237689375878,
      "num_tokens": 9132686.0,
      "step": 4215
    },
    {
      "entropy": 6.649262714385986,
      "epoch": 0.451602546952753,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986549415171482,
      "loss": 6.501,
      "mean_token_accuracy": 0.12619779780507087,
      "num_tokens": 9143582.0,
      "step": 4220
    },
    {
      "entropy": 6.82189040184021,
      "epoch": 0.45213762106051686,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004986507639584274,
      "loss": 6.5821,
      "mean_token_accuracy": 0.1168325275182724,
      "num_tokens": 9156173.0,
      "step": 4225
    },
    {
      "entropy": 6.66492600440979,
      "epoch": 0.4526726951682808,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004986465799418223,
      "loss": 6.5773,
      "mean_token_accuracy": 0.12092015594244003,
      "num_tokens": 9166882.0,
      "step": 4230
    },
    {
      "entropy": 6.669227886199951,
      "epoch": 0.4532077692760447,
      "grad_norm": 0.77734375,
      "learning_rate": 0.0004986423894674534,
      "loss": 6.5669,
      "mean_token_accuracy": 0.12278837189078332,
      "num_tokens": 9177525.0,
      "step": 4235
    },
    {
      "entropy": 6.783564901351928,
      "epoch": 0.45374284338380866,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004986381925354422,
      "loss": 6.515,
      "mean_token_accuracy": 0.12535930350422858,
      "num_tokens": 9188586.0,
      "step": 4240
    },
    {
      "entropy": 6.634851264953613,
      "epoch": 0.4542779174915726,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004986339891459094,
      "loss": 6.4353,
      "mean_token_accuracy": 0.13272654935717582,
      "num_tokens": 9199920.0,
      "step": 4245
    },
    {
      "entropy": 6.579446649551391,
      "epoch": 0.4548129915993365,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004986297792989768,
      "loss": 6.4489,
      "mean_token_accuracy": 0.13165228366851806,
      "num_tokens": 9210923.0,
      "step": 4250
    },
    {
      "entropy": 6.771414089202881,
      "epoch": 0.45534806570710046,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986255629947655,
      "loss": 6.6181,
      "mean_token_accuracy": 0.1227750226855278,
      "num_tokens": 9221459.0,
      "step": 4255
    },
    {
      "entropy": 6.73143310546875,
      "epoch": 0.45588313981486434,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004986213402333978,
      "loss": 6.5599,
      "mean_token_accuracy": 0.12391374334692955,
      "num_tokens": 9231768.0,
      "step": 4260
    },
    {
      "entropy": 6.687683820724487,
      "epoch": 0.45641821392262827,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004986171110149951,
      "loss": 6.5375,
      "mean_token_accuracy": 0.12756052240729332,
      "num_tokens": 9242230.0,
      "step": 4265
    },
    {
      "entropy": 6.7013014793396,
      "epoch": 0.4569532880303922,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986128753396798,
      "loss": 6.4225,
      "mean_token_accuracy": 0.14026538878679276,
      "num_tokens": 9252485.0,
      "step": 4270
    },
    {
      "entropy": 6.55937352180481,
      "epoch": 0.45748836213815613,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004986086332075742,
      "loss": 6.518,
      "mean_token_accuracy": 0.13188858777284623,
      "num_tokens": 9263014.0,
      "step": 4275
    },
    {
      "entropy": 6.67518835067749,
      "epoch": 0.45802343624592007,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004986043846188006,
      "loss": 6.549,
      "mean_token_accuracy": 0.13570101261138917,
      "num_tokens": 9273777.0,
      "step": 4280
    },
    {
      "entropy": 6.6095668315887455,
      "epoch": 0.458558510353684,
      "grad_norm": 1.0,
      "learning_rate": 0.000498600129573482,
      "loss": 6.4885,
      "mean_token_accuracy": 0.12344526872038841,
      "num_tokens": 9284738.0,
      "step": 4285
    },
    {
      "entropy": 6.706184148788452,
      "epoch": 0.45909358446144793,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000498595868071741,
      "loss": 6.5584,
      "mean_token_accuracy": 0.12425956949591636,
      "num_tokens": 9295921.0,
      "step": 4290
    },
    {
      "entropy": 6.699369287490844,
      "epoch": 0.4596286585692118,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004985916001137006,
      "loss": 6.524,
      "mean_token_accuracy": 0.1261294700205326,
      "num_tokens": 9306568.0,
      "step": 4295
    },
    {
      "entropy": 6.693387126922607,
      "epoch": 0.46016373267697575,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004985873256994843,
      "loss": 6.4816,
      "mean_token_accuracy": 0.12667348012328147,
      "num_tokens": 9316468.0,
      "step": 4300
    },
    {
      "entropy": 6.6449432373046875,
      "epoch": 0.4606988067847397,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985830448292154,
      "loss": 6.5646,
      "mean_token_accuracy": 0.12484502121806144,
      "num_tokens": 9328185.0,
      "step": 4305
    },
    {
      "entropy": 6.655733108520508,
      "epoch": 0.4612338808925036,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004985787575030175,
      "loss": 6.5055,
      "mean_token_accuracy": 0.12664854377508164,
      "num_tokens": 9338591.0,
      "step": 4310
    },
    {
      "entropy": 6.71276707649231,
      "epoch": 0.46176895500026754,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004985744637210144,
      "loss": 6.5447,
      "mean_token_accuracy": 0.12165323868393899,
      "num_tokens": 9349286.0,
      "step": 4315
    },
    {
      "entropy": 6.670126962661743,
      "epoch": 0.4623040291080315,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00049857016348333,
      "loss": 6.5149,
      "mean_token_accuracy": 0.1324791319668293,
      "num_tokens": 9359752.0,
      "step": 4320
    },
    {
      "entropy": 6.737763357162476,
      "epoch": 0.4628391032157954,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004985658567900886,
      "loss": 6.5894,
      "mean_token_accuracy": 0.11995674669742584,
      "num_tokens": 9371823.0,
      "step": 4325
    },
    {
      "entropy": 6.588179349899292,
      "epoch": 0.4633741773235593,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004985615436414145,
      "loss": 6.4261,
      "mean_token_accuracy": 0.12827276140451432,
      "num_tokens": 9383127.0,
      "step": 4330
    },
    {
      "entropy": 6.624917984008789,
      "epoch": 0.4639092514313232,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985572240374321,
      "loss": 6.5481,
      "mean_token_accuracy": 0.12790464386343955,
      "num_tokens": 9394296.0,
      "step": 4335
    },
    {
      "entropy": 6.71473331451416,
      "epoch": 0.46444432553908715,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004985528979782665,
      "loss": 6.5219,
      "mean_token_accuracy": 0.12147142961621285,
      "num_tokens": 9405988.0,
      "step": 4340
    },
    {
      "entropy": 6.728419351577759,
      "epoch": 0.4649793996468511,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004985485654640423,
      "loss": 6.5307,
      "mean_token_accuracy": 0.1294780731201172,
      "num_tokens": 9417479.0,
      "step": 4345
    },
    {
      "entropy": 6.677046489715576,
      "epoch": 0.465514473754615,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004985442264948847,
      "loss": 6.4781,
      "mean_token_accuracy": 0.13104128241539,
      "num_tokens": 9428050.0,
      "step": 4350
    },
    {
      "entropy": 6.602668333053589,
      "epoch": 0.46604954786237895,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004985398810709189,
      "loss": 6.4289,
      "mean_token_accuracy": 0.13291953578591348,
      "num_tokens": 9438229.0,
      "step": 4355
    },
    {
      "entropy": 6.621841144561768,
      "epoch": 0.4665846219701429,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004985355291922704,
      "loss": 6.5209,
      "mean_token_accuracy": 0.12576833069324495,
      "num_tokens": 9449611.0,
      "step": 4360
    },
    {
      "entropy": 6.626019430160523,
      "epoch": 0.46711969607790677,
      "grad_norm": 0.984375,
      "learning_rate": 0.000498531170859065,
      "loss": 6.4803,
      "mean_token_accuracy": 0.12464792206883431,
      "num_tokens": 9460234.0,
      "step": 4365
    },
    {
      "entropy": 6.688359355926513,
      "epoch": 0.4676547701856707,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004985268060714285,
      "loss": 6.5359,
      "mean_token_accuracy": 0.128062354773283,
      "num_tokens": 9472136.0,
      "step": 4370
    },
    {
      "entropy": 6.6489707946777346,
      "epoch": 0.46818984429343463,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004985224348294868,
      "loss": 6.4977,
      "mean_token_accuracy": 0.13154730200767517,
      "num_tokens": 9482602.0,
      "step": 4375
    },
    {
      "entropy": 6.75223126411438,
      "epoch": 0.46872491840119856,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004985180571333663,
      "loss": 6.4894,
      "mean_token_accuracy": 0.13144660815596582,
      "num_tokens": 9493782.0,
      "step": 4380
    },
    {
      "entropy": 6.5902422904968265,
      "epoch": 0.4692599925089625,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004985136729831932,
      "loss": 6.4712,
      "mean_token_accuracy": 0.12834268659353257,
      "num_tokens": 9504389.0,
      "step": 4385
    },
    {
      "entropy": 6.615573263168335,
      "epoch": 0.46979506661672643,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004985092823790942,
      "loss": 6.5912,
      "mean_token_accuracy": 0.12514225915074348,
      "num_tokens": 9515166.0,
      "step": 4390
    },
    {
      "entropy": 6.733069658279419,
      "epoch": 0.47033014072449036,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004985048853211961,
      "loss": 6.5085,
      "mean_token_accuracy": 0.12728274688124658,
      "num_tokens": 9525566.0,
      "step": 4395
    },
    {
      "entropy": 6.700274705886841,
      "epoch": 0.47086521483225424,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498500481809626,
      "loss": 6.4194,
      "mean_token_accuracy": 0.1366426721215248,
      "num_tokens": 9536021.0,
      "step": 4400
    },
    {
      "entropy": 6.616627550125122,
      "epoch": 0.4714002889400182,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004984960718445107,
      "loss": 6.5091,
      "mean_token_accuracy": 0.12142780721187592,
      "num_tokens": 9546827.0,
      "step": 4405
    },
    {
      "entropy": 6.756778526306152,
      "epoch": 0.4719353630477821,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004984916554259778,
      "loss": 6.6272,
      "mean_token_accuracy": 0.11766278594732285,
      "num_tokens": 9558104.0,
      "step": 4410
    },
    {
      "entropy": 6.780975437164306,
      "epoch": 0.47247043715554604,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004984872325541547,
      "loss": 6.5192,
      "mean_token_accuracy": 0.1305229753255844,
      "num_tokens": 9568487.0,
      "step": 4415
    },
    {
      "entropy": 6.638660430908203,
      "epoch": 0.47300551126331,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984828032291692,
      "loss": 6.4634,
      "mean_token_accuracy": 0.13527418449521064,
      "num_tokens": 9579509.0,
      "step": 4420
    },
    {
      "entropy": 6.663227844238281,
      "epoch": 0.4735405853710739,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984783674511492,
      "loss": 6.5034,
      "mean_token_accuracy": 0.12841146439313889,
      "num_tokens": 9589931.0,
      "step": 4425
    },
    {
      "entropy": 6.6811174869537355,
      "epoch": 0.47407565947883784,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984739252202227,
      "loss": 6.5353,
      "mean_token_accuracy": 0.12742502838373185,
      "num_tokens": 9599764.0,
      "step": 4430
    },
    {
      "entropy": 6.678421831130981,
      "epoch": 0.4746107335866017,
      "grad_norm": 0.90625,
      "learning_rate": 0.000498469476536518,
      "loss": 6.5053,
      "mean_token_accuracy": 0.13000835999846458,
      "num_tokens": 9611145.0,
      "step": 4435
    },
    {
      "entropy": 6.644073152542115,
      "epoch": 0.47514580769436565,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004984650214001636,
      "loss": 6.502,
      "mean_token_accuracy": 0.1319687359035015,
      "num_tokens": 9621441.0,
      "step": 4440
    },
    {
      "entropy": 6.634088039398193,
      "epoch": 0.4756808818021296,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004984605598112881,
      "loss": 6.5429,
      "mean_token_accuracy": 0.13188807293772697,
      "num_tokens": 9631874.0,
      "step": 4445
    },
    {
      "entropy": 6.6144147396087645,
      "epoch": 0.4762159559098935,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004984560917700204,
      "loss": 6.522,
      "mean_token_accuracy": 0.11959985718131065,
      "num_tokens": 9643006.0,
      "step": 4450
    },
    {
      "entropy": 6.652404069900513,
      "epoch": 0.47675103001765745,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004984516172764894,
      "loss": 6.4059,
      "mean_token_accuracy": 0.1327241726219654,
      "num_tokens": 9653897.0,
      "step": 4455
    },
    {
      "entropy": 6.608202123641968,
      "epoch": 0.4772861041254214,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004984471363308243,
      "loss": 6.4968,
      "mean_token_accuracy": 0.12554212734103204,
      "num_tokens": 9664460.0,
      "step": 4460
    },
    {
      "entropy": 6.681749153137207,
      "epoch": 0.4778211782331853,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004984426489331546,
      "loss": 6.494,
      "mean_token_accuracy": 0.1332928016781807,
      "num_tokens": 9674534.0,
      "step": 4465
    },
    {
      "entropy": 6.706970596313477,
      "epoch": 0.4783562523409492,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004984381550836097,
      "loss": 6.5747,
      "mean_token_accuracy": 0.12295873761177063,
      "num_tokens": 9684828.0,
      "step": 4470
    },
    {
      "entropy": 6.735209989547729,
      "epoch": 0.47889132644871313,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004984336547823196,
      "loss": 6.538,
      "mean_token_accuracy": 0.13103948533535004,
      "num_tokens": 9694536.0,
      "step": 4475
    },
    {
      "entropy": 6.634707307815551,
      "epoch": 0.47942640055647706,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000498429148029414,
      "loss": 6.4102,
      "mean_token_accuracy": 0.13732634782791137,
      "num_tokens": 9705715.0,
      "step": 4480
    },
    {
      "entropy": 6.622171640396118,
      "epoch": 0.479961474664241,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004984246348250233,
      "loss": 6.4753,
      "mean_token_accuracy": 0.13292167633771895,
      "num_tokens": 9717505.0,
      "step": 4485
    },
    {
      "entropy": 6.633249378204345,
      "epoch": 0.4804965487720049,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004984201151692775,
      "loss": 6.3711,
      "mean_token_accuracy": 0.1316321723163128,
      "num_tokens": 9727910.0,
      "step": 4490
    },
    {
      "entropy": 6.62043080329895,
      "epoch": 0.48103162287976886,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004984155890623074,
      "loss": 6.4698,
      "mean_token_accuracy": 0.13449255973100663,
      "num_tokens": 9738528.0,
      "step": 4495
    },
    {
      "entropy": 6.671240425109863,
      "epoch": 0.4815666969875328,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004984110565042435,
      "loss": 6.4875,
      "mean_token_accuracy": 0.12762050181627274,
      "num_tokens": 9749744.0,
      "step": 4500
    },
    {
      "entropy": 6.687379980087281,
      "epoch": 0.48210177109529667,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004984065174952168,
      "loss": 6.5035,
      "mean_token_accuracy": 0.12369845062494278,
      "num_tokens": 9759968.0,
      "step": 4505
    },
    {
      "entropy": 6.60925521850586,
      "epoch": 0.4826368452030606,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004984019720353582,
      "loss": 6.5227,
      "mean_token_accuracy": 0.12860732153058052,
      "num_tokens": 9769791.0,
      "step": 4510
    },
    {
      "entropy": 6.633432817459107,
      "epoch": 0.48317191931082454,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983974201247991,
      "loss": 6.4889,
      "mean_token_accuracy": 0.1387922890484333,
      "num_tokens": 9780369.0,
      "step": 4515
    },
    {
      "entropy": 6.638261842727661,
      "epoch": 0.48370699341858847,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004983928617636709,
      "loss": 6.5307,
      "mean_token_accuracy": 0.12319683730602264,
      "num_tokens": 9791125.0,
      "step": 4520
    },
    {
      "entropy": 6.675742292404175,
      "epoch": 0.4842420675263524,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004983882969521052,
      "loss": 6.526,
      "mean_token_accuracy": 0.12952028438448906,
      "num_tokens": 9801697.0,
      "step": 4525
    },
    {
      "entropy": 6.754634857177734,
      "epoch": 0.48477714163411634,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004983837256902338,
      "loss": 6.5621,
      "mean_token_accuracy": 0.1297781951725483,
      "num_tokens": 9812171.0,
      "step": 4530
    },
    {
      "entropy": 6.68020339012146,
      "epoch": 0.48531221574188027,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004983791479781887,
      "loss": 6.4054,
      "mean_token_accuracy": 0.1366664409637451,
      "num_tokens": 9821596.0,
      "step": 4535
    },
    {
      "entropy": 6.679528045654297,
      "epoch": 0.48584728984964415,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004983745638161021,
      "loss": 6.5296,
      "mean_token_accuracy": 0.1205269493162632,
      "num_tokens": 9831634.0,
      "step": 4540
    },
    {
      "entropy": 6.666993045806885,
      "epoch": 0.4863823639574081,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004983699732041064,
      "loss": 6.4882,
      "mean_token_accuracy": 0.13205414190888404,
      "num_tokens": 9841662.0,
      "step": 4545
    },
    {
      "entropy": 6.670028972625732,
      "epoch": 0.486917438065172,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000498365376142334,
      "loss": 6.4637,
      "mean_token_accuracy": 0.1234777070581913,
      "num_tokens": 9853797.0,
      "step": 4550
    },
    {
      "entropy": 6.681587982177734,
      "epoch": 0.48745251217293595,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004983607726309179,
      "loss": 6.5233,
      "mean_token_accuracy": 0.12715924233198167,
      "num_tokens": 9865245.0,
      "step": 4555
    },
    {
      "entropy": 6.6820274829864506,
      "epoch": 0.4879875862806999,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983561626699908,
      "loss": 6.6009,
      "mean_token_accuracy": 0.11727365702390671,
      "num_tokens": 9876207.0,
      "step": 4560
    },
    {
      "entropy": 6.639146041870117,
      "epoch": 0.4885226603884638,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004983515462596859,
      "loss": 6.4204,
      "mean_token_accuracy": 0.13496807143092154,
      "num_tokens": 9887210.0,
      "step": 4565
    },
    {
      "entropy": 6.691262245178223,
      "epoch": 0.48905773449622775,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983469234001365,
      "loss": 6.5983,
      "mean_token_accuracy": 0.12019298523664475,
      "num_tokens": 9899607.0,
      "step": 4570
    },
    {
      "entropy": 6.595501470565796,
      "epoch": 0.4895928086039917,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000498342294091476,
      "loss": 6.3773,
      "mean_token_accuracy": 0.13430055379867553,
      "num_tokens": 9909910.0,
      "step": 4575
    },
    {
      "entropy": 6.615827751159668,
      "epoch": 0.49012788271175556,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004983376583338382,
      "loss": 6.5079,
      "mean_token_accuracy": 0.1254624530673027,
      "num_tokens": 9922137.0,
      "step": 4580
    },
    {
      "entropy": 6.638407421112061,
      "epoch": 0.4906629568195195,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004983330161273569,
      "loss": 6.5069,
      "mean_token_accuracy": 0.13186536133289337,
      "num_tokens": 9932514.0,
      "step": 4585
    },
    {
      "entropy": 6.654112815856934,
      "epoch": 0.4911980309272834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983283674721662,
      "loss": 6.5293,
      "mean_token_accuracy": 0.129039965569973,
      "num_tokens": 9943054.0,
      "step": 4590
    },
    {
      "entropy": 6.740719270706177,
      "epoch": 0.49173310503504736,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004983237123684002,
      "loss": 6.5094,
      "mean_token_accuracy": 0.12170372903347015,
      "num_tokens": 9953791.0,
      "step": 4595
    },
    {
      "entropy": 6.587079620361328,
      "epoch": 0.4922681791428113,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004983190508161934,
      "loss": 6.4627,
      "mean_token_accuracy": 0.1294872209429741,
      "num_tokens": 9964782.0,
      "step": 4600
    },
    {
      "entropy": 6.6498559474945065,
      "epoch": 0.4928032532505752,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004983143828156804,
      "loss": 6.5266,
      "mean_token_accuracy": 0.12838228419423103,
      "num_tokens": 9975843.0,
      "step": 4605
    },
    {
      "entropy": 6.704116678237915,
      "epoch": 0.49333832735833916,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004983097083669959,
      "loss": 6.5007,
      "mean_token_accuracy": 0.13298282846808435,
      "num_tokens": 9987060.0,
      "step": 4610
    },
    {
      "entropy": 6.646341371536255,
      "epoch": 0.49387340146610303,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004983050274702751,
      "loss": 6.465,
      "mean_token_accuracy": 0.12663726061582564,
      "num_tokens": 9997380.0,
      "step": 4615
    },
    {
      "entropy": 6.6445447444915775,
      "epoch": 0.49440847557386697,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004983003401256529,
      "loss": 6.5172,
      "mean_token_accuracy": 0.12875118479132652,
      "num_tokens": 10007955.0,
      "step": 4620
    },
    {
      "entropy": 6.78430962562561,
      "epoch": 0.4949435496816309,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004982956463332647,
      "loss": 6.585,
      "mean_token_accuracy": 0.12407600358128548,
      "num_tokens": 10017645.0,
      "step": 4625
    },
    {
      "entropy": 6.629924869537353,
      "epoch": 0.49547862378939483,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004982909460932463,
      "loss": 6.5323,
      "mean_token_accuracy": 0.1285700298845768,
      "num_tokens": 10027619.0,
      "step": 4630
    },
    {
      "entropy": 6.580510091781616,
      "epoch": 0.49601369789715877,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000498286239405733,
      "loss": 6.3937,
      "mean_token_accuracy": 0.13874556571245195,
      "num_tokens": 10037551.0,
      "step": 4635
    },
    {
      "entropy": 6.605573320388794,
      "epoch": 0.4965487720049227,
      "grad_norm": 0.86328125,
      "learning_rate": 0.000498281526270861,
      "loss": 6.4215,
      "mean_token_accuracy": 0.13800892382860183,
      "num_tokens": 10047771.0,
      "step": 4640
    },
    {
      "entropy": 6.736871671676636,
      "epoch": 0.49708384611268663,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982768066887663,
      "loss": 6.5738,
      "mean_token_accuracy": 0.1221847340464592,
      "num_tokens": 10058771.0,
      "step": 4645
    },
    {
      "entropy": 6.605147457122802,
      "epoch": 0.4976189202204505,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004982720806595851,
      "loss": 6.4377,
      "mean_token_accuracy": 0.13613661155104637,
      "num_tokens": 10070318.0,
      "step": 4650
    },
    {
      "entropy": 6.566076612472534,
      "epoch": 0.49815399432821444,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004982673481834541,
      "loss": 6.4794,
      "mean_token_accuracy": 0.12480002790689468,
      "num_tokens": 10081810.0,
      "step": 4655
    },
    {
      "entropy": 6.641260051727295,
      "epoch": 0.4986890684359784,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004982626092605097,
      "loss": 6.4814,
      "mean_token_accuracy": 0.1259951874613762,
      "num_tokens": 10091724.0,
      "step": 4660
    },
    {
      "entropy": 6.687929010391235,
      "epoch": 0.4992241425437423,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004982578638908888,
      "loss": 6.4605,
      "mean_token_accuracy": 0.13417887166142464,
      "num_tokens": 10102510.0,
      "step": 4665
    },
    {
      "entropy": 6.617040157318115,
      "epoch": 0.49975921665150624,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004982531120747286,
      "loss": 6.5499,
      "mean_token_accuracy": 0.1230244629085064,
      "num_tokens": 10112636.0,
      "step": 4670
    },
    {
      "entropy": 6.66232738494873,
      "epoch": 0.5002942907592701,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000498248353812166,
      "loss": 6.5287,
      "mean_token_accuracy": 0.131372894346714,
      "num_tokens": 10123562.0,
      "step": 4675
    },
    {
      "entropy": 6.659600734710693,
      "epoch": 0.5008293648670341,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004982435891033387,
      "loss": 6.4954,
      "mean_token_accuracy": 0.12019536346197128,
      "num_tokens": 10134731.0,
      "step": 4680
    },
    {
      "entropy": 6.658867216110229,
      "epoch": 0.501364438974798,
      "grad_norm": 0.82421875,
      "learning_rate": 0.000498238817948384,
      "loss": 6.4867,
      "mean_token_accuracy": 0.12581022083759308,
      "num_tokens": 10145232.0,
      "step": 4685
    },
    {
      "entropy": 6.658832597732544,
      "epoch": 0.501899513082562,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982340403474398,
      "loss": 6.4544,
      "mean_token_accuracy": 0.12711327224969865,
      "num_tokens": 10155857.0,
      "step": 4690
    },
    {
      "entropy": 6.601750755310059,
      "epoch": 0.5024345871903259,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004982292563006441,
      "loss": 6.4533,
      "mean_token_accuracy": 0.13474077358841896,
      "num_tokens": 10166145.0,
      "step": 4695
    },
    {
      "entropy": 6.6569538593292235,
      "epoch": 0.5029696612980897,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498224465808135,
      "loss": 6.5942,
      "mean_token_accuracy": 0.12159125730395318,
      "num_tokens": 10177086.0,
      "step": 4700
    },
    {
      "entropy": 6.703319120407104,
      "epoch": 0.5035047354058537,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004982196688700509,
      "loss": 6.4545,
      "mean_token_accuracy": 0.12969210147857665,
      "num_tokens": 10187120.0,
      "step": 4705
    },
    {
      "entropy": 6.615395879745483,
      "epoch": 0.5040398095136176,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004982148654865302,
      "loss": 6.5139,
      "mean_token_accuracy": 0.1280078411102295,
      "num_tokens": 10197784.0,
      "step": 4710
    },
    {
      "entropy": 6.6520003318786625,
      "epoch": 0.5045748836213816,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004982100556577115,
      "loss": 6.3837,
      "mean_token_accuracy": 0.13710010796785355,
      "num_tokens": 10209373.0,
      "step": 4715
    },
    {
      "entropy": 6.653470230102539,
      "epoch": 0.5051099577291455,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004982052393837338,
      "loss": 6.5106,
      "mean_token_accuracy": 0.13445726856589318,
      "num_tokens": 10220316.0,
      "step": 4720
    },
    {
      "entropy": 6.573184156417847,
      "epoch": 0.5056450318369095,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004982004166647363,
      "loss": 6.4179,
      "mean_token_accuracy": 0.13162229359149932,
      "num_tokens": 10230478.0,
      "step": 4725
    },
    {
      "entropy": 6.572769546508789,
      "epoch": 0.5061801059446733,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004981955875008582,
      "loss": 6.5068,
      "mean_token_accuracy": 0.13034035861492158,
      "num_tokens": 10240641.0,
      "step": 4730
    },
    {
      "entropy": 6.663148355484009,
      "epoch": 0.5067151800524372,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004981907518922387,
      "loss": 6.473,
      "mean_token_accuracy": 0.13470017835497855,
      "num_tokens": 10250676.0,
      "step": 4735
    },
    {
      "entropy": 6.7320473194122314,
      "epoch": 0.5072502541602012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981859098390177,
      "loss": 6.5316,
      "mean_token_accuracy": 0.1317993402481079,
      "num_tokens": 10262016.0,
      "step": 4740
    },
    {
      "entropy": 6.629784202575683,
      "epoch": 0.5077853282679651,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004981810613413348,
      "loss": 6.4807,
      "mean_token_accuracy": 0.1275458127260208,
      "num_tokens": 10271985.0,
      "step": 4745
    },
    {
      "entropy": 6.645000886917114,
      "epoch": 0.5083204023757291,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004981762063993302,
      "loss": 6.4468,
      "mean_token_accuracy": 0.13643463030457498,
      "num_tokens": 10282156.0,
      "step": 4750
    },
    {
      "entropy": 6.608729267120362,
      "epoch": 0.5088554764834929,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000498171345013144,
      "loss": 6.476,
      "mean_token_accuracy": 0.12745949104428292,
      "num_tokens": 10292576.0,
      "step": 4755
    },
    {
      "entropy": 6.594528150558472,
      "epoch": 0.5093905505912569,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981664771829165,
      "loss": 6.4777,
      "mean_token_accuracy": 0.1305478774011135,
      "num_tokens": 10303648.0,
      "step": 4760
    },
    {
      "entropy": 6.6102478981018065,
      "epoch": 0.5099256246990208,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004981616029087884,
      "loss": 6.354,
      "mean_token_accuracy": 0.1412102237343788,
      "num_tokens": 10314275.0,
      "step": 4765
    },
    {
      "entropy": 6.6168890476226805,
      "epoch": 0.5104606988067847,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004981567221909004,
      "loss": 6.48,
      "mean_token_accuracy": 0.12254874557256698,
      "num_tokens": 10325386.0,
      "step": 4770
    },
    {
      "entropy": 6.62398509979248,
      "epoch": 0.5109957729145487,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004981518350293935,
      "loss": 6.4652,
      "mean_token_accuracy": 0.12788971215486528,
      "num_tokens": 10335041.0,
      "step": 4775
    },
    {
      "entropy": 6.60414490699768,
      "epoch": 0.5115308470223126,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004981469414244086,
      "loss": 6.4153,
      "mean_token_accuracy": 0.13254307731986045,
      "num_tokens": 10345746.0,
      "step": 4780
    },
    {
      "entropy": 6.59508867263794,
      "epoch": 0.5120659211300765,
      "grad_norm": 0.890625,
      "learning_rate": 0.000498142041376087,
      "loss": 6.5051,
      "mean_token_accuracy": 0.12980739027261734,
      "num_tokens": 10356995.0,
      "step": 4785
    },
    {
      "entropy": 6.638574075698853,
      "epoch": 0.5126009952378404,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004981371348845705,
      "loss": 6.4396,
      "mean_token_accuracy": 0.13524362444877625,
      "num_tokens": 10368445.0,
      "step": 4790
    },
    {
      "entropy": 6.630716753005982,
      "epoch": 0.5131360693456044,
      "grad_norm": 0.875,
      "learning_rate": 0.0004981322219500006,
      "loss": 6.3772,
      "mean_token_accuracy": 0.1370594806969166,
      "num_tokens": 10378256.0,
      "step": 4795
    },
    {
      "entropy": 6.631394529342652,
      "epoch": 0.5136711434533683,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000498127302572519,
      "loss": 6.4759,
      "mean_token_accuracy": 0.12833066135644913,
      "num_tokens": 10388698.0,
      "step": 4800
    },
    {
      "entropy": 6.581023120880127,
      "epoch": 0.5142062175611322,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004981223767522679,
      "loss": 6.4706,
      "mean_token_accuracy": 0.1341756135225296,
      "num_tokens": 10399975.0,
      "step": 4805
    },
    {
      "entropy": 6.611735057830811,
      "epoch": 0.5147412916688962,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004981174444893896,
      "loss": 6.4831,
      "mean_token_accuracy": 0.12776545137166978,
      "num_tokens": 10410480.0,
      "step": 4810
    },
    {
      "entropy": 6.629581546783447,
      "epoch": 0.51527636577666,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004981125057840264,
      "loss": 6.4404,
      "mean_token_accuracy": 0.13301406651735306,
      "num_tokens": 10421023.0,
      "step": 4815
    },
    {
      "entropy": 6.635268592834473,
      "epoch": 0.515811439884424,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000498107560636321,
      "loss": 6.4832,
      "mean_token_accuracy": 0.13234892711043358,
      "num_tokens": 10431152.0,
      "step": 4820
    },
    {
      "entropy": 6.609623670578003,
      "epoch": 0.5163465139921879,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004981026090464161,
      "loss": 6.3907,
      "mean_token_accuracy": 0.12726799473166467,
      "num_tokens": 10441986.0,
      "step": 4825
    },
    {
      "entropy": 6.665307092666626,
      "epoch": 0.5168815880999519,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004980976510144548,
      "loss": 6.4697,
      "mean_token_accuracy": 0.1312413029372692,
      "num_tokens": 10452932.0,
      "step": 4830
    },
    {
      "entropy": 6.657898330688477,
      "epoch": 0.5174166622077158,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980926865405801,
      "loss": 6.5334,
      "mean_token_accuracy": 0.13006979301571847,
      "num_tokens": 10463411.0,
      "step": 4835
    },
    {
      "entropy": 6.612054872512817,
      "epoch": 0.5179517363154797,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004980877156249354,
      "loss": 6.4312,
      "mean_token_accuracy": 0.12908995747566224,
      "num_tokens": 10474820.0,
      "step": 4840
    },
    {
      "entropy": 6.605563735961914,
      "epoch": 0.5184868104232436,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004980827382676643,
      "loss": 6.4336,
      "mean_token_accuracy": 0.13449087291955947,
      "num_tokens": 10485701.0,
      "step": 4845
    },
    {
      "entropy": 6.629001617431641,
      "epoch": 0.5190218845310075,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004980777544689105,
      "loss": 6.4242,
      "mean_token_accuracy": 0.13291734606027603,
      "num_tokens": 10495825.0,
      "step": 4850
    },
    {
      "entropy": 6.63130373954773,
      "epoch": 0.5195569586387715,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004980727642288178,
      "loss": 6.4235,
      "mean_token_accuracy": 0.1373104438185692,
      "num_tokens": 10505847.0,
      "step": 4855
    },
    {
      "entropy": 6.620811843872071,
      "epoch": 0.5200920327465354,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004980677675475305,
      "loss": 6.4745,
      "mean_token_accuracy": 0.12906499356031417,
      "num_tokens": 10516596.0,
      "step": 4860
    },
    {
      "entropy": 6.615943861007691,
      "epoch": 0.5206271068542994,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004980627644251926,
      "loss": 6.3881,
      "mean_token_accuracy": 0.13836821541190147,
      "num_tokens": 10528175.0,
      "step": 4865
    },
    {
      "entropy": 6.57639708518982,
      "epoch": 0.5211621809620632,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004980577548619489,
      "loss": 6.4849,
      "mean_token_accuracy": 0.12840275838971138,
      "num_tokens": 10540076.0,
      "step": 4870
    },
    {
      "entropy": 6.621571588516235,
      "epoch": 0.5216972550698272,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004980527388579437,
      "loss": 6.5063,
      "mean_token_accuracy": 0.12991923615336418,
      "num_tokens": 10552025.0,
      "step": 4875
    },
    {
      "entropy": 6.643404912948609,
      "epoch": 0.5222323291775911,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004980477164133221,
      "loss": 6.4985,
      "mean_token_accuracy": 0.12801033034920692,
      "num_tokens": 10562535.0,
      "step": 4880
    },
    {
      "entropy": 6.591488313674927,
      "epoch": 0.522767403285355,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000498042687528229,
      "loss": 6.433,
      "mean_token_accuracy": 0.1368302322924137,
      "num_tokens": 10572989.0,
      "step": 4885
    },
    {
      "entropy": 6.578029823303223,
      "epoch": 0.523302477393119,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004980376522028098,
      "loss": 6.4311,
      "mean_token_accuracy": 0.13608148843050002,
      "num_tokens": 10583033.0,
      "step": 4890
    },
    {
      "entropy": 6.67541823387146,
      "epoch": 0.5238375515008828,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004980326104372095,
      "loss": 6.5134,
      "mean_token_accuracy": 0.13113198056817055,
      "num_tokens": 10595205.0,
      "step": 4895
    },
    {
      "entropy": 6.633107995986938,
      "epoch": 0.5243726256086468,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000498027562231574,
      "loss": 6.3799,
      "mean_token_accuracy": 0.1369300030171871,
      "num_tokens": 10605244.0,
      "step": 4900
    },
    {
      "entropy": 6.548080635070801,
      "epoch": 0.5249076997164107,
      "grad_norm": 0.83203125,
      "learning_rate": 0.000498022507586049,
      "loss": 6.4421,
      "mean_token_accuracy": 0.13095175921916963,
      "num_tokens": 10615817.0,
      "step": 4905
    },
    {
      "entropy": 6.63610873222351,
      "epoch": 0.5254427738241747,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004980174465007804,
      "loss": 6.4653,
      "mean_token_accuracy": 0.1272575758397579,
      "num_tokens": 10627160.0,
      "step": 4910
    },
    {
      "entropy": 6.613020944595337,
      "epoch": 0.5259778479319386,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004980123789759145,
      "loss": 6.4621,
      "mean_token_accuracy": 0.13695163801312446,
      "num_tokens": 10637552.0,
      "step": 4915
    },
    {
      "entropy": 6.633289575576782,
      "epoch": 0.5265129220397025,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004980073050115973,
      "loss": 6.4837,
      "mean_token_accuracy": 0.13580890074372293,
      "num_tokens": 10648773.0,
      "step": 4920
    },
    {
      "entropy": 6.6644628047943115,
      "epoch": 0.5270479961474664,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004980022246079755,
      "loss": 6.4893,
      "mean_token_accuracy": 0.12900413498282431,
      "num_tokens": 10659537.0,
      "step": 4925
    },
    {
      "entropy": 6.678333568572998,
      "epoch": 0.5275830702552303,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000497997137765196,
      "loss": 6.5279,
      "mean_token_accuracy": 0.1248202569782734,
      "num_tokens": 10671390.0,
      "step": 4930
    },
    {
      "entropy": 6.620291471481323,
      "epoch": 0.5281181443629943,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004979920444834054,
      "loss": 6.3809,
      "mean_token_accuracy": 0.14244376122951508,
      "num_tokens": 10682389.0,
      "step": 4935
    },
    {
      "entropy": 6.521419858932495,
      "epoch": 0.5286532184707582,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004979869447627508,
      "loss": 6.4436,
      "mean_token_accuracy": 0.13305451348423958,
      "num_tokens": 10692400.0,
      "step": 4940
    },
    {
      "entropy": 6.563894271850586,
      "epoch": 0.5291882925785222,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004979818386033795,
      "loss": 6.3639,
      "mean_token_accuracy": 0.13524384126067163,
      "num_tokens": 10702396.0,
      "step": 4945
    },
    {
      "entropy": 6.578342390060425,
      "epoch": 0.5297233666862861,
      "grad_norm": 0.83984375,
      "learning_rate": 0.000497976726005439,
      "loss": 6.4823,
      "mean_token_accuracy": 0.1315547928214073,
      "num_tokens": 10712863.0,
      "step": 4950
    },
    {
      "entropy": 6.675341939926147,
      "epoch": 0.5302584407940499,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004979716069690768,
      "loss": 6.4897,
      "mean_token_accuracy": 0.12950288206338884,
      "num_tokens": 10724234.0,
      "step": 4955
    },
    {
      "entropy": 6.723803901672364,
      "epoch": 0.5307935149018139,
      "grad_norm": 1.25,
      "learning_rate": 0.0004979664814944409,
      "loss": 6.5076,
      "mean_token_accuracy": 0.12615261673927308,
      "num_tokens": 10734405.0,
      "step": 4960
    },
    {
      "entropy": 6.647532176971436,
      "epoch": 0.5313285890095778,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004979613495816791,
      "loss": 6.4501,
      "mean_token_accuracy": 0.13496298864483833,
      "num_tokens": 10744573.0,
      "step": 4965
    },
    {
      "entropy": 6.642793846130371,
      "epoch": 0.5318636631173418,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004979562112309397,
      "loss": 6.5223,
      "mean_token_accuracy": 0.13261223658919336,
      "num_tokens": 10755555.0,
      "step": 4970
    },
    {
      "entropy": 6.597110176086426,
      "epoch": 0.5323987372251057,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000497951066442371,
      "loss": 6.53,
      "mean_token_accuracy": 0.123849056661129,
      "num_tokens": 10766493.0,
      "step": 4975
    },
    {
      "entropy": 6.682430124282837,
      "epoch": 0.5329338113328697,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004979459152161215,
      "loss": 6.4762,
      "mean_token_accuracy": 0.12446538880467414,
      "num_tokens": 10776735.0,
      "step": 4980
    },
    {
      "entropy": 6.585795831680298,
      "epoch": 0.5334688854406335,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004979407575523402,
      "loss": 6.4788,
      "mean_token_accuracy": 0.13038726896047592,
      "num_tokens": 10787500.0,
      "step": 4985
    },
    {
      "entropy": 6.5987447738647464,
      "epoch": 0.5340039595483974,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004979355934511757,
      "loss": 6.4956,
      "mean_token_accuracy": 0.13150699660182,
      "num_tokens": 10798002.0,
      "step": 4990
    },
    {
      "entropy": 6.60825834274292,
      "epoch": 0.5345390336561614,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004979304229127773,
      "loss": 6.44,
      "mean_token_accuracy": 0.13437903374433519,
      "num_tokens": 10808982.0,
      "step": 4995
    },
    {
      "entropy": 6.610709095001221,
      "epoch": 0.5350741077639253,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004979252459372944,
      "loss": 6.3928,
      "mean_token_accuracy": 0.13411291986703872,
      "num_tokens": 10821619.0,
      "step": 5000
    },
    {
      "entropy": 6.623658800125122,
      "epoch": 0.5356091818716893,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004979200625248761,
      "loss": 6.4286,
      "mean_token_accuracy": 0.13552514240145683,
      "num_tokens": 10832344.0,
      "step": 5005
    },
    {
      "entropy": 6.642703866958618,
      "epoch": 0.5361442559794531,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004979148726756725,
      "loss": 6.3927,
      "mean_token_accuracy": 0.13105069994926452,
      "num_tokens": 10843149.0,
      "step": 5010
    },
    {
      "entropy": 6.594963884353637,
      "epoch": 0.5366793300872171,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004979096763898333,
      "loss": 6.4409,
      "mean_token_accuracy": 0.13659456893801689,
      "num_tokens": 10854590.0,
      "step": 5015
    },
    {
      "entropy": 6.635641431808471,
      "epoch": 0.537214404194981,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004979044736675084,
      "loss": 6.4088,
      "mean_token_accuracy": 0.13173776045441626,
      "num_tokens": 10865500.0,
      "step": 5020
    },
    {
      "entropy": 6.591180086135864,
      "epoch": 0.5377494783027449,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004978992645088483,
      "loss": 6.4565,
      "mean_token_accuracy": 0.13780646920204162,
      "num_tokens": 10876866.0,
      "step": 5025
    },
    {
      "entropy": 6.486580514907837,
      "epoch": 0.5382845524105089,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000497894048914003,
      "loss": 6.4116,
      "mean_token_accuracy": 0.13192799538373948,
      "num_tokens": 10888627.0,
      "step": 5030
    },
    {
      "entropy": 6.687264013290405,
      "epoch": 0.5388196265182728,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004978888268831236,
      "loss": 6.5199,
      "mean_token_accuracy": 0.12521881535649299,
      "num_tokens": 10900992.0,
      "step": 5035
    },
    {
      "entropy": 6.708090019226074,
      "epoch": 0.5393547006260367,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978835984163606,
      "loss": 6.5639,
      "mean_token_accuracy": 0.1180639810860157,
      "num_tokens": 10912323.0,
      "step": 5040
    },
    {
      "entropy": 6.620733070373535,
      "epoch": 0.5398897747338006,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004978783635138649,
      "loss": 6.4856,
      "mean_token_accuracy": 0.1326413333415985,
      "num_tokens": 10923991.0,
      "step": 5045
    },
    {
      "entropy": 6.610896444320678,
      "epoch": 0.5404248488415646,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978731221757878,
      "loss": 6.3531,
      "mean_token_accuracy": 0.13417525365948677,
      "num_tokens": 10934081.0,
      "step": 5050
    },
    {
      "entropy": 6.5825488567352295,
      "epoch": 0.5409599229493285,
      "grad_norm": 1.0,
      "learning_rate": 0.0004978678744022808,
      "loss": 6.4329,
      "mean_token_accuracy": 0.135209359228611,
      "num_tokens": 10944729.0,
      "step": 5055
    },
    {
      "entropy": 6.665228748321534,
      "epoch": 0.5414949970570924,
      "grad_norm": 0.80859375,
      "learning_rate": 0.000497862620193495,
      "loss": 6.4855,
      "mean_token_accuracy": 0.1289301633834839,
      "num_tokens": 10955708.0,
      "step": 5060
    },
    {
      "entropy": 6.603465986251831,
      "epoch": 0.5420300711648564,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004978573595495826,
      "loss": 6.4197,
      "mean_token_accuracy": 0.12746732607483863,
      "num_tokens": 10966427.0,
      "step": 5065
    },
    {
      "entropy": 6.606443548202515,
      "epoch": 0.5425651452726202,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000497852092470695,
      "loss": 6.4768,
      "mean_token_accuracy": 0.13206957876682282,
      "num_tokens": 10977741.0,
      "step": 5070
    },
    {
      "entropy": 6.642637538909912,
      "epoch": 0.5431002193803842,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004978468189569847,
      "loss": 6.4354,
      "mean_token_accuracy": 0.1358490638434887,
      "num_tokens": 10989127.0,
      "step": 5075
    },
    {
      "entropy": 6.555223178863526,
      "epoch": 0.5436352934881481,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004978415390086038,
      "loss": 6.4441,
      "mean_token_accuracy": 0.12885510921478271,
      "num_tokens": 10999791.0,
      "step": 5080
    },
    {
      "entropy": 6.667173433303833,
      "epoch": 0.5441703675959121,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004978362526257047,
      "loss": 6.5002,
      "mean_token_accuracy": 0.12577469870448113,
      "num_tokens": 11010045.0,
      "step": 5085
    },
    {
      "entropy": 6.640448665618896,
      "epoch": 0.544705441703676,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004978309598084403,
      "loss": 6.4332,
      "mean_token_accuracy": 0.13017582893371582,
      "num_tokens": 11020729.0,
      "step": 5090
    },
    {
      "entropy": 6.631437110900879,
      "epoch": 0.5452405158114398,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004978256605569631,
      "loss": 6.4522,
      "mean_token_accuracy": 0.1322314776480198,
      "num_tokens": 11030641.0,
      "step": 5095
    },
    {
      "entropy": 6.630730199813843,
      "epoch": 0.5457755899192038,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004978203548714262,
      "loss": 6.4586,
      "mean_token_accuracy": 0.13038678020238875,
      "num_tokens": 11041533.0,
      "step": 5100
    },
    {
      "entropy": 6.595696926116943,
      "epoch": 0.5463106640269677,
      "grad_norm": 0.79296875,
      "learning_rate": 0.000497815042751983,
      "loss": 6.4069,
      "mean_token_accuracy": 0.12977249100804328,
      "num_tokens": 11052170.0,
      "step": 5105
    },
    {
      "entropy": 6.540053987503052,
      "epoch": 0.5468457381347317,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004978097241987868,
      "loss": 6.3767,
      "mean_token_accuracy": 0.1344059519469738,
      "num_tokens": 11062872.0,
      "step": 5110
    },
    {
      "entropy": 6.614701986312866,
      "epoch": 0.5473808122424956,
      "grad_norm": 0.890625,
      "learning_rate": 0.000497804399211991,
      "loss": 6.4883,
      "mean_token_accuracy": 0.1307993248105049,
      "num_tokens": 11074199.0,
      "step": 5115
    },
    {
      "entropy": 6.635615634918213,
      "epoch": 0.5479158863502596,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977990677917495,
      "loss": 6.4407,
      "mean_token_accuracy": 0.13429288640618325,
      "num_tokens": 11083881.0,
      "step": 5120
    },
    {
      "entropy": 6.56947956085205,
      "epoch": 0.5484509604580234,
      "grad_norm": 0.875,
      "learning_rate": 0.0004977937299382162,
      "loss": 6.4897,
      "mean_token_accuracy": 0.12704017013311386,
      "num_tokens": 11094374.0,
      "step": 5125
    },
    {
      "entropy": 6.591516447067261,
      "epoch": 0.5489860345657873,
      "grad_norm": 1.25,
      "learning_rate": 0.0004977883856515453,
      "loss": 6.3252,
      "mean_token_accuracy": 0.14053784012794496,
      "num_tokens": 11105174.0,
      "step": 5130
    },
    {
      "entropy": 6.547975683212281,
      "epoch": 0.5495211086735513,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000497783034931891,
      "loss": 6.498,
      "mean_token_accuracy": 0.13169048056006433,
      "num_tokens": 11117223.0,
      "step": 5135
    },
    {
      "entropy": 6.671089124679566,
      "epoch": 0.5500561827813152,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000497777677779408,
      "loss": 6.4192,
      "mean_token_accuracy": 0.12979228496551515,
      "num_tokens": 11127240.0,
      "step": 5140
    },
    {
      "entropy": 6.596736049652099,
      "epoch": 0.5505912568890792,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004977723141942509,
      "loss": 6.3724,
      "mean_token_accuracy": 0.1445842854678631,
      "num_tokens": 11137287.0,
      "step": 5145
    },
    {
      "entropy": 6.581566190719604,
      "epoch": 0.551126330996843,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004977669441765743,
      "loss": 6.5663,
      "mean_token_accuracy": 0.12571654692292214,
      "num_tokens": 11148988.0,
      "step": 5150
    },
    {
      "entropy": 6.66612982749939,
      "epoch": 0.551661405104607,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977615677265336,
      "loss": 6.4666,
      "mean_token_accuracy": 0.13236356154084206,
      "num_tokens": 11159874.0,
      "step": 5155
    },
    {
      "entropy": 6.656392335891724,
      "epoch": 0.5521964792123709,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000497756184844284,
      "loss": 6.4194,
      "mean_token_accuracy": 0.1348666474223137,
      "num_tokens": 11170703.0,
      "step": 5160
    },
    {
      "entropy": 6.514084815979004,
      "epoch": 0.5527315533201348,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004977507955299809,
      "loss": 6.303,
      "mean_token_accuracy": 0.14367334991693498,
      "num_tokens": 11182230.0,
      "step": 5165
    },
    {
      "entropy": 6.5705219268798825,
      "epoch": 0.5532666274278988,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977453997837797,
      "loss": 6.4047,
      "mean_token_accuracy": 0.12478601410984994,
      "num_tokens": 11192897.0,
      "step": 5170
    },
    {
      "entropy": 6.637944078445434,
      "epoch": 0.5538017015356627,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004977399976058366,
      "loss": 6.4917,
      "mean_token_accuracy": 0.12794631570577622,
      "num_tokens": 11203942.0,
      "step": 5175
    },
    {
      "entropy": 6.657525253295899,
      "epoch": 0.5543367756434266,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977345889963072,
      "loss": 6.5191,
      "mean_token_accuracy": 0.12642529159784316,
      "num_tokens": 11214361.0,
      "step": 5180
    },
    {
      "entropy": 6.554194641113281,
      "epoch": 0.5548718497511905,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004977291739553479,
      "loss": 6.4321,
      "mean_token_accuracy": 0.13422846421599388,
      "num_tokens": 11225928.0,
      "step": 5185
    },
    {
      "entropy": 6.594499778747559,
      "epoch": 0.5554069238589545,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004977237524831149,
      "loss": 6.3949,
      "mean_token_accuracy": 0.12422026544809342,
      "num_tokens": 11236660.0,
      "step": 5190
    },
    {
      "entropy": 6.540923309326172,
      "epoch": 0.5559419979667184,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004977183245797649,
      "loss": 6.4001,
      "mean_token_accuracy": 0.14284231811761855,
      "num_tokens": 11247272.0,
      "step": 5195
    },
    {
      "entropy": 6.628225469589234,
      "epoch": 0.5564770720744823,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004977128902454547,
      "loss": 6.5004,
      "mean_token_accuracy": 0.12502486854791642,
      "num_tokens": 11258472.0,
      "step": 5200
    },
    {
      "entropy": 6.670383024215698,
      "epoch": 0.5570121461822463,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004977074494803409,
      "loss": 6.392,
      "mean_token_accuracy": 0.14113382995128632,
      "num_tokens": 11268344.0,
      "step": 5205
    },
    {
      "entropy": 6.5701521873474125,
      "epoch": 0.5575472202900101,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004977020022845809,
      "loss": 6.4905,
      "mean_token_accuracy": 0.12506455108523368,
      "num_tokens": 11279931.0,
      "step": 5210
    },
    {
      "entropy": 6.671869421005249,
      "epoch": 0.5580822943977741,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976965486583318,
      "loss": 6.4459,
      "mean_token_accuracy": 0.1281243644654751,
      "num_tokens": 11290595.0,
      "step": 5215
    },
    {
      "entropy": 6.58588662147522,
      "epoch": 0.558617368505538,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004976910886017513,
      "loss": 6.4646,
      "mean_token_accuracy": 0.13403649777173995,
      "num_tokens": 11300774.0,
      "step": 5220
    },
    {
      "entropy": 6.6014903545379635,
      "epoch": 0.559152442613302,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004976856221149969,
      "loss": 6.3569,
      "mean_token_accuracy": 0.14249046072363852,
      "num_tokens": 11310288.0,
      "step": 5225
    },
    {
      "entropy": 6.588343715667724,
      "epoch": 0.5596875167210659,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976801491982263,
      "loss": 6.4484,
      "mean_token_accuracy": 0.1302250012755394,
      "num_tokens": 11321599.0,
      "step": 5230
    },
    {
      "entropy": 6.632518863677978,
      "epoch": 0.5602225908288297,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004976746698515977,
      "loss": 6.4876,
      "mean_token_accuracy": 0.13621862158179282,
      "num_tokens": 11331383.0,
      "step": 5235
    },
    {
      "entropy": 6.606530237197876,
      "epoch": 0.5607576649365937,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004976691840752694,
      "loss": 6.452,
      "mean_token_accuracy": 0.1273598402738571,
      "num_tokens": 11342085.0,
      "step": 5240
    },
    {
      "entropy": 6.596298360824585,
      "epoch": 0.5612927390443576,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004976636918693996,
      "loss": 6.4856,
      "mean_token_accuracy": 0.12637143433094025,
      "num_tokens": 11354061.0,
      "step": 5245
    },
    {
      "entropy": 6.6884543895721436,
      "epoch": 0.5618278131521216,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004976581932341471,
      "loss": 6.4222,
      "mean_token_accuracy": 0.13169787973165512,
      "num_tokens": 11364448.0,
      "step": 5250
    },
    {
      "entropy": 6.689448833465576,
      "epoch": 0.5623628872598855,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976526881696706,
      "loss": 6.3971,
      "mean_token_accuracy": 0.13254007026553155,
      "num_tokens": 11374838.0,
      "step": 5255
    },
    {
      "entropy": 6.508886289596558,
      "epoch": 0.5628979613676495,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004976471766761288,
      "loss": 6.4524,
      "mean_token_accuracy": 0.13301875442266464,
      "num_tokens": 11385310.0,
      "step": 5260
    },
    {
      "entropy": 6.587552404403686,
      "epoch": 0.5634330354754133,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004976416587536812,
      "loss": 6.4347,
      "mean_token_accuracy": 0.12732182890176774,
      "num_tokens": 11396590.0,
      "step": 5265
    },
    {
      "entropy": 6.618044662475586,
      "epoch": 0.5639681095831772,
      "grad_norm": 0.875,
      "learning_rate": 0.0004976361344024871,
      "loss": 6.3971,
      "mean_token_accuracy": 0.13341034501791,
      "num_tokens": 11407218.0,
      "step": 5270
    },
    {
      "entropy": 6.498771047592163,
      "epoch": 0.5645031836909412,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004976306036227056,
      "loss": 6.4152,
      "mean_token_accuracy": 0.1334387518465519,
      "num_tokens": 11417716.0,
      "step": 5275
    },
    {
      "entropy": 6.6630606174469,
      "epoch": 0.5650382577987051,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000497625066414497,
      "loss": 6.4962,
      "mean_token_accuracy": 0.12818157002329827,
      "num_tokens": 11428295.0,
      "step": 5280
    },
    {
      "entropy": 6.655110549926758,
      "epoch": 0.5655733319064691,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004976195227780207,
      "loss": 6.432,
      "mean_token_accuracy": 0.12855115234851838,
      "num_tokens": 11439044.0,
      "step": 5285
    },
    {
      "entropy": 6.605824327468872,
      "epoch": 0.566108406014233,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004976139727134371,
      "loss": 6.3837,
      "mean_token_accuracy": 0.1281607411801815,
      "num_tokens": 11450029.0,
      "step": 5290
    },
    {
      "entropy": 6.5651922702789305,
      "epoch": 0.566643480121997,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004976084162209062,
      "loss": 6.5481,
      "mean_token_accuracy": 0.12115254476666451,
      "num_tokens": 11461190.0,
      "step": 5295
    },
    {
      "entropy": 6.587199306488037,
      "epoch": 0.5671785542297608,
      "grad_norm": 0.79296875,
      "learning_rate": 0.0004976028533005886,
      "loss": 6.4074,
      "mean_token_accuracy": 0.13242645785212517,
      "num_tokens": 11472143.0,
      "step": 5300
    },
    {
      "entropy": 6.648195028305054,
      "epoch": 0.5677136283375247,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004975972839526449,
      "loss": 6.5453,
      "mean_token_accuracy": 0.12057378143072128,
      "num_tokens": 11483745.0,
      "step": 5305
    },
    {
      "entropy": 6.601594543457031,
      "epoch": 0.5682487024452887,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975917081772358,
      "loss": 6.3929,
      "mean_token_accuracy": 0.13485434129834176,
      "num_tokens": 11495695.0,
      "step": 5310
    },
    {
      "entropy": 6.634777784347534,
      "epoch": 0.5687837765530526,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004975861259745225,
      "loss": 6.4454,
      "mean_token_accuracy": 0.13447388708591462,
      "num_tokens": 11505576.0,
      "step": 5315
    },
    {
      "entropy": 6.576272439956665,
      "epoch": 0.5693188506608166,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004975805373446662,
      "loss": 6.2933,
      "mean_token_accuracy": 0.13931943476200104,
      "num_tokens": 11516614.0,
      "step": 5320
    },
    {
      "entropy": 6.575680208206177,
      "epoch": 0.5698539247685804,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000497574942287828,
      "loss": 6.4265,
      "mean_token_accuracy": 0.1257178418338299,
      "num_tokens": 11528032.0,
      "step": 5325
    },
    {
      "entropy": 6.551600313186645,
      "epoch": 0.5703889988763444,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975693408041697,
      "loss": 6.4361,
      "mean_token_accuracy": 0.12763728350400924,
      "num_tokens": 11538822.0,
      "step": 5330
    },
    {
      "entropy": 6.6312580585479735,
      "epoch": 0.5709240729841083,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004975637328938529,
      "loss": 6.4506,
      "mean_token_accuracy": 0.12766908183693887,
      "num_tokens": 11548895.0,
      "step": 5335
    },
    {
      "entropy": 6.579705810546875,
      "epoch": 0.5714591470918722,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004975581185570398,
      "loss": 6.3573,
      "mean_token_accuracy": 0.128676937520504,
      "num_tokens": 11559613.0,
      "step": 5340
    },
    {
      "entropy": 6.55549578666687,
      "epoch": 0.5719942211996362,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004975524977938921,
      "loss": 6.3779,
      "mean_token_accuracy": 0.1374692440032959,
      "num_tokens": 11571092.0,
      "step": 5345
    },
    {
      "entropy": 6.607112216949463,
      "epoch": 0.5725292953074,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004975468706045722,
      "loss": 6.4942,
      "mean_token_accuracy": 0.1302151046693325,
      "num_tokens": 11581448.0,
      "step": 5350
    },
    {
      "entropy": 6.614511585235595,
      "epoch": 0.573064369415164,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004975412369892429,
      "loss": 6.4029,
      "mean_token_accuracy": 0.13683656677603723,
      "num_tokens": 11592273.0,
      "step": 5355
    },
    {
      "entropy": 6.586702966690064,
      "epoch": 0.5735994435229279,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004975355969480665,
      "loss": 6.3903,
      "mean_token_accuracy": 0.1379762977361679,
      "num_tokens": 11602759.0,
      "step": 5360
    },
    {
      "entropy": 6.574518823623658,
      "epoch": 0.5741345176306919,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004975299504812061,
      "loss": 6.4258,
      "mean_token_accuracy": 0.12733170837163926,
      "num_tokens": 11613977.0,
      "step": 5365
    },
    {
      "entropy": 6.657155513763428,
      "epoch": 0.5746695917384558,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004975242975888246,
      "loss": 6.4517,
      "mean_token_accuracy": 0.13247386440634729,
      "num_tokens": 11625425.0,
      "step": 5370
    },
    {
      "entropy": 6.603577184677124,
      "epoch": 0.5752046658462197,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004975186382710854,
      "loss": 6.4175,
      "mean_token_accuracy": 0.13235969319939614,
      "num_tokens": 11636530.0,
      "step": 5375
    },
    {
      "entropy": 6.6505959033966064,
      "epoch": 0.5757397399539836,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004975129725281516,
      "loss": 6.418,
      "mean_token_accuracy": 0.1344207614660263,
      "num_tokens": 11646516.0,
      "step": 5380
    },
    {
      "entropy": 6.549999094009399,
      "epoch": 0.5762748140617475,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004975073003601871,
      "loss": 6.3904,
      "mean_token_accuracy": 0.13115372210741044,
      "num_tokens": 11657424.0,
      "step": 5385
    },
    {
      "entropy": 6.580683898925781,
      "epoch": 0.5768098881695115,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004975016217673556,
      "loss": 6.4263,
      "mean_token_accuracy": 0.1307824395596981,
      "num_tokens": 11669447.0,
      "step": 5390
    },
    {
      "entropy": 6.633586883544922,
      "epoch": 0.5773449622772754,
      "grad_norm": 0.90625,
      "learning_rate": 0.000497495936749821,
      "loss": 6.3755,
      "mean_token_accuracy": 0.14045795276761056,
      "num_tokens": 11679612.0,
      "step": 5395
    },
    {
      "entropy": 6.564531373977661,
      "epoch": 0.5778800363850394,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004974902453077474,
      "loss": 6.3665,
      "mean_token_accuracy": 0.140810589492321,
      "num_tokens": 11689527.0,
      "step": 5400
    },
    {
      "entropy": 6.607308626174927,
      "epoch": 0.5784151104928033,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004974845474412993,
      "loss": 6.4125,
      "mean_token_accuracy": 0.13287024796009064,
      "num_tokens": 11700452.0,
      "step": 5405
    },
    {
      "entropy": 6.539466714859008,
      "epoch": 0.5789501846005671,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004974788431506412,
      "loss": 6.4179,
      "mean_token_accuracy": 0.13111766055226326,
      "num_tokens": 11710902.0,
      "step": 5410
    },
    {
      "entropy": 6.580818510055542,
      "epoch": 0.5794852587083311,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004974731324359378,
      "loss": 6.4516,
      "mean_token_accuracy": 0.12524218559265138,
      "num_tokens": 11721901.0,
      "step": 5415
    },
    {
      "entropy": 6.574299621582031,
      "epoch": 0.580020332816095,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004974674152973538,
      "loss": 6.436,
      "mean_token_accuracy": 0.12758637517690657,
      "num_tokens": 11733622.0,
      "step": 5420
    },
    {
      "entropy": 6.645131921768188,
      "epoch": 0.580555406923859,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004974616917350545,
      "loss": 6.4384,
      "mean_token_accuracy": 0.13184043169021606,
      "num_tokens": 11745116.0,
      "step": 5425
    },
    {
      "entropy": 6.514034986495972,
      "epoch": 0.5810904810316229,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004974559617492052,
      "loss": 6.3601,
      "mean_token_accuracy": 0.14398131519556046,
      "num_tokens": 11756203.0,
      "step": 5430
    },
    {
      "entropy": 6.586912155151367,
      "epoch": 0.5816255551393869,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004974502253399712,
      "loss": 6.4155,
      "mean_token_accuracy": 0.13222005367279052,
      "num_tokens": 11766797.0,
      "step": 5435
    },
    {
      "entropy": 6.65447473526001,
      "epoch": 0.5821606292471507,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004974444825075182,
      "loss": 6.5038,
      "mean_token_accuracy": 0.12940247878432273,
      "num_tokens": 11778110.0,
      "step": 5440
    },
    {
      "entropy": 6.526040124893188,
      "epoch": 0.5826957033549146,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497438733252012,
      "loss": 6.3593,
      "mean_token_accuracy": 0.13486856520175933,
      "num_tokens": 11788396.0,
      "step": 5445
    },
    {
      "entropy": 6.630442190170288,
      "epoch": 0.5832307774626786,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004974329775736188,
      "loss": 6.4293,
      "mean_token_accuracy": 0.1297837421298027,
      "num_tokens": 11798691.0,
      "step": 5450
    },
    {
      "entropy": 6.666377258300781,
      "epoch": 0.5837658515704425,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004974272154725044,
      "loss": 6.496,
      "mean_token_accuracy": 0.1329609051346779,
      "num_tokens": 11808965.0,
      "step": 5455
    },
    {
      "entropy": 6.572327184677124,
      "epoch": 0.5843009256782065,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004974214469488355,
      "loss": 6.3343,
      "mean_token_accuracy": 0.13848894461989403,
      "num_tokens": 11819932.0,
      "step": 5460
    },
    {
      "entropy": 6.626104116439819,
      "epoch": 0.5848359997859703,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004974156720027786,
      "loss": 6.426,
      "mean_token_accuracy": 0.1278026707470417,
      "num_tokens": 11831332.0,
      "step": 5465
    },
    {
      "entropy": 6.579942083358764,
      "epoch": 0.5853710738937343,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004974098906345004,
      "loss": 6.4187,
      "mean_token_accuracy": 0.1270743988454342,
      "num_tokens": 11840947.0,
      "step": 5470
    },
    {
      "entropy": 6.554082345962525,
      "epoch": 0.5859061480014982,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004974041028441679,
      "loss": 6.3534,
      "mean_token_accuracy": 0.13192323818802834,
      "num_tokens": 11851640.0,
      "step": 5475
    },
    {
      "entropy": 6.650488948822021,
      "epoch": 0.5864412221092621,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004973983086319481,
      "loss": 6.4799,
      "mean_token_accuracy": 0.12282219007611275,
      "num_tokens": 11863608.0,
      "step": 5480
    },
    {
      "entropy": 6.599178647994995,
      "epoch": 0.5869762962170261,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004973925079980085,
      "loss": 6.3844,
      "mean_token_accuracy": 0.1334873415529728,
      "num_tokens": 11874707.0,
      "step": 5485
    },
    {
      "entropy": 6.564403486251831,
      "epoch": 0.58751137032479,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973867009425164,
      "loss": 6.4576,
      "mean_token_accuracy": 0.13435769751667975,
      "num_tokens": 11885118.0,
      "step": 5490
    },
    {
      "entropy": 6.636895275115966,
      "epoch": 0.5880464444325539,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004973808874656396,
      "loss": 6.433,
      "mean_token_accuracy": 0.1338953197002411,
      "num_tokens": 11895859.0,
      "step": 5495
    },
    {
      "entropy": 6.667998456954956,
      "epoch": 0.5885815185403178,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004973750675675458,
      "loss": 6.4449,
      "mean_token_accuracy": 0.1335778370499611,
      "num_tokens": 11907019.0,
      "step": 5500
    },
    {
      "entropy": 6.66331148147583,
      "epoch": 0.5891165926480818,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004973692412484034,
      "loss": 6.3889,
      "mean_token_accuracy": 0.13348127380013466,
      "num_tokens": 11918257.0,
      "step": 5505
    },
    {
      "entropy": 6.5581518650054935,
      "epoch": 0.5896516667558457,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004973634085083802,
      "loss": 6.3977,
      "mean_token_accuracy": 0.14219407886266708,
      "num_tokens": 11929003.0,
      "step": 5510
    },
    {
      "entropy": 6.622295188903808,
      "epoch": 0.5901867408636096,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000497357569347645,
      "loss": 6.5309,
      "mean_token_accuracy": 0.1261256143450737,
      "num_tokens": 11939786.0,
      "step": 5515
    },
    {
      "entropy": 6.6540138721466064,
      "epoch": 0.5907218149713735,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004973517237663661,
      "loss": 6.4607,
      "mean_token_accuracy": 0.1294330932199955,
      "num_tokens": 11951042.0,
      "step": 5520
    },
    {
      "entropy": 6.569264698028564,
      "epoch": 0.5912568890791374,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004973458717647124,
      "loss": 6.3549,
      "mean_token_accuracy": 0.13981595113873482,
      "num_tokens": 11961982.0,
      "step": 5525
    },
    {
      "entropy": 6.532705926895142,
      "epoch": 0.5917919631869014,
      "grad_norm": 0.90234375,
      "learning_rate": 0.000497340013342853,
      "loss": 6.4091,
      "mean_token_accuracy": 0.13178442418575287,
      "num_tokens": 11972078.0,
      "step": 5530
    },
    {
      "entropy": 6.568721675872803,
      "epoch": 0.5923270372946653,
      "grad_norm": 0.921875,
      "learning_rate": 0.000497334148500957,
      "loss": 6.4584,
      "mean_token_accuracy": 0.13278395012021066,
      "num_tokens": 11984037.0,
      "step": 5535
    },
    {
      "entropy": 6.6697516441345215,
      "epoch": 0.5928621114024293,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004973282772391936,
      "loss": 6.394,
      "mean_token_accuracy": 0.13035947531461717,
      "num_tokens": 11993688.0,
      "step": 5540
    },
    {
      "entropy": 6.565344905853271,
      "epoch": 0.5933971855101932,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004973223995577324,
      "loss": 6.3777,
      "mean_token_accuracy": 0.13061075434088706,
      "num_tokens": 12003759.0,
      "step": 5545
    },
    {
      "entropy": 6.531543827056884,
      "epoch": 0.593932259617957,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004973165154567432,
      "loss": 6.3875,
      "mean_token_accuracy": 0.13072411194443703,
      "num_tokens": 12015030.0,
      "step": 5550
    },
    {
      "entropy": 6.590817785263061,
      "epoch": 0.594467333725721,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004973106249363959,
      "loss": 6.4772,
      "mean_token_accuracy": 0.13105777204036712,
      "num_tokens": 12026181.0,
      "step": 5555
    },
    {
      "entropy": 6.541166162490844,
      "epoch": 0.5950024078334849,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004973047279968606,
      "loss": 6.2499,
      "mean_token_accuracy": 0.1466530330479145,
      "num_tokens": 12037336.0,
      "step": 5560
    },
    {
      "entropy": 6.567782735824585,
      "epoch": 0.5955374819412489,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004972988246383074,
      "loss": 6.3037,
      "mean_token_accuracy": 0.13920291215181352,
      "num_tokens": 12048425.0,
      "step": 5565
    },
    {
      "entropy": 6.436773157119751,
      "epoch": 0.5960725560490128,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972929148609068,
      "loss": 6.2698,
      "mean_token_accuracy": 0.15105066001415252,
      "num_tokens": 12059814.0,
      "step": 5570
    },
    {
      "entropy": 6.525291872024536,
      "epoch": 0.5966076301567768,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004972869986648296,
      "loss": 6.3441,
      "mean_token_accuracy": 0.138674733042717,
      "num_tokens": 12069377.0,
      "step": 5575
    },
    {
      "entropy": 6.570910167694092,
      "epoch": 0.5971427042645406,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004972810760502467,
      "loss": 6.4875,
      "mean_token_accuracy": 0.12600740566849708,
      "num_tokens": 12080277.0,
      "step": 5580
    },
    {
      "entropy": 6.599872684478759,
      "epoch": 0.5976777783723045,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972751470173287,
      "loss": 6.3497,
      "mean_token_accuracy": 0.13464435711503028,
      "num_tokens": 12091375.0,
      "step": 5585
    },
    {
      "entropy": 6.550978755950927,
      "epoch": 0.5982128524800685,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004972692115662473,
      "loss": 6.4211,
      "mean_token_accuracy": 0.13656509220600127,
      "num_tokens": 12103110.0,
      "step": 5590
    },
    {
      "entropy": 6.599859666824341,
      "epoch": 0.5987479265878324,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004972632696971735,
      "loss": 6.4603,
      "mean_token_accuracy": 0.13045730218291282,
      "num_tokens": 12112454.0,
      "step": 5595
    },
    {
      "entropy": 6.55696702003479,
      "epoch": 0.5992830006955964,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972573214102791,
      "loss": 6.3395,
      "mean_token_accuracy": 0.1349827118217945,
      "num_tokens": 12125140.0,
      "step": 5600
    },
    {
      "entropy": 6.5705962657928465,
      "epoch": 0.5998180748033602,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004972513667057358,
      "loss": 6.3656,
      "mean_token_accuracy": 0.13562314063310624,
      "num_tokens": 12136061.0,
      "step": 5605
    },
    {
      "entropy": 6.650278472900391,
      "epoch": 0.6003531489111242,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004972454055837155,
      "loss": 6.404,
      "mean_token_accuracy": 0.13498041927814483,
      "num_tokens": 12146654.0,
      "step": 5610
    },
    {
      "entropy": 6.614070034027099,
      "epoch": 0.6008882230188881,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004972394380443903,
      "loss": 6.4097,
      "mean_token_accuracy": 0.1308725282549858,
      "num_tokens": 12157881.0,
      "step": 5615
    },
    {
      "entropy": 6.578127241134643,
      "epoch": 0.601423297126652,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004972334640879325,
      "loss": 6.4287,
      "mean_token_accuracy": 0.13418816700577735,
      "num_tokens": 12168534.0,
      "step": 5620
    },
    {
      "entropy": 6.554200172424316,
      "epoch": 0.601958371234416,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004972274837145147,
      "loss": 6.3578,
      "mean_token_accuracy": 0.13448846638202666,
      "num_tokens": 12178773.0,
      "step": 5625
    },
    {
      "entropy": 6.577337265014648,
      "epoch": 0.6024934453421799,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004972214969243096,
      "loss": 6.3879,
      "mean_token_accuracy": 0.13618157282471657,
      "num_tokens": 12188819.0,
      "step": 5630
    },
    {
      "entropy": 6.515968942642212,
      "epoch": 0.6030285194499438,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00049721550371749,
      "loss": 6.375,
      "mean_token_accuracy": 0.138833886384964,
      "num_tokens": 12199198.0,
      "step": 5635
    },
    {
      "entropy": 6.569277572631836,
      "epoch": 0.6035635935577077,
      "grad_norm": 0.7890625,
      "learning_rate": 0.0004972095040942289,
      "loss": 6.4012,
      "mean_token_accuracy": 0.1318491853773594,
      "num_tokens": 12210765.0,
      "step": 5640
    },
    {
      "entropy": 6.580835676193237,
      "epoch": 0.6040986676654717,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004972034980546996,
      "loss": 6.4207,
      "mean_token_accuracy": 0.128849096596241,
      "num_tokens": 12221067.0,
      "step": 5645
    },
    {
      "entropy": 6.530771636962891,
      "epoch": 0.6046337417732356,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004971974855990755,
      "loss": 6.4017,
      "mean_token_accuracy": 0.13618824183940886,
      "num_tokens": 12232994.0,
      "step": 5650
    },
    {
      "entropy": 6.593167781829834,
      "epoch": 0.6051688158809995,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004971914667275302,
      "loss": 6.3796,
      "mean_token_accuracy": 0.13560357168316842,
      "num_tokens": 12243327.0,
      "step": 5655
    },
    {
      "entropy": 6.473536396026612,
      "epoch": 0.6057038899887635,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004971854414402377,
      "loss": 6.3126,
      "mean_token_accuracy": 0.13819175511598586,
      "num_tokens": 12254249.0,
      "step": 5660
    },
    {
      "entropy": 6.540711259841919,
      "epoch": 0.6062389640965273,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004971794097373716,
      "loss": 6.3412,
      "mean_token_accuracy": 0.13511233404278755,
      "num_tokens": 12264487.0,
      "step": 5665
    },
    {
      "entropy": 6.494623517990112,
      "epoch": 0.6067740382042913,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004971733716191063,
      "loss": 6.347,
      "mean_token_accuracy": 0.1392446607351303,
      "num_tokens": 12274801.0,
      "step": 5670
    },
    {
      "entropy": 6.561354398727417,
      "epoch": 0.6073091123120552,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004971673270856162,
      "loss": 6.3022,
      "mean_token_accuracy": 0.13149403035640717,
      "num_tokens": 12284647.0,
      "step": 5675
    },
    {
      "entropy": 6.544518184661865,
      "epoch": 0.6078441864198192,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004971612761370756,
      "loss": 6.4221,
      "mean_token_accuracy": 0.13287723660469056,
      "num_tokens": 12295215.0,
      "step": 5680
    },
    {
      "entropy": 6.517327117919922,
      "epoch": 0.6083792605275831,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004971552187736596,
      "loss": 6.339,
      "mean_token_accuracy": 0.1339027963578701,
      "num_tokens": 12305324.0,
      "step": 5685
    },
    {
      "entropy": 6.5198009490966795,
      "epoch": 0.6089143346353469,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004971491549955427,
      "loss": 6.3515,
      "mean_token_accuracy": 0.1382587730884552,
      "num_tokens": 12316399.0,
      "step": 5690
    },
    {
      "entropy": 6.480385828018188,
      "epoch": 0.6094494087431109,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004971430848029002,
      "loss": 6.3333,
      "mean_token_accuracy": 0.1392750360071659,
      "num_tokens": 12328033.0,
      "step": 5695
    },
    {
      "entropy": 6.518033885955811,
      "epoch": 0.6099844828508748,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004971370081959073,
      "loss": 6.3701,
      "mean_token_accuracy": 0.1361820012331009,
      "num_tokens": 12338962.0,
      "step": 5700
    },
    {
      "entropy": 6.588574457168579,
      "epoch": 0.6105195569586388,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004971309251747396,
      "loss": 6.3557,
      "mean_token_accuracy": 0.1340286947786808,
      "num_tokens": 12350216.0,
      "step": 5705
    },
    {
      "entropy": 6.601768064498901,
      "epoch": 0.6110546310664027,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004971248357395726,
      "loss": 6.3923,
      "mean_token_accuracy": 0.13332446068525314,
      "num_tokens": 12361277.0,
      "step": 5710
    },
    {
      "entropy": 6.487619066238404,
      "epoch": 0.6115897051741667,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004971187398905821,
      "loss": 6.3086,
      "mean_token_accuracy": 0.13468469381332399,
      "num_tokens": 12373308.0,
      "step": 5715
    },
    {
      "entropy": 6.604460716247559,
      "epoch": 0.6121247792819305,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004971126376279443,
      "loss": 6.5461,
      "mean_token_accuracy": 0.12380608022212983,
      "num_tokens": 12383943.0,
      "step": 5720
    },
    {
      "entropy": 6.578613996505737,
      "epoch": 0.6126598533896944,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004971065289518354,
      "loss": 6.3125,
      "mean_token_accuracy": 0.13367188200354577,
      "num_tokens": 12394575.0,
      "step": 5725
    },
    {
      "entropy": 6.603823232650757,
      "epoch": 0.6131949274974584,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004971004138624315,
      "loss": 6.3981,
      "mean_token_accuracy": 0.13316768556833267,
      "num_tokens": 12405536.0,
      "step": 5730
    },
    {
      "entropy": 6.497972536087036,
      "epoch": 0.6137300016052223,
      "grad_norm": 0.8125,
      "learning_rate": 0.0004970942923599094,
      "loss": 6.3005,
      "mean_token_accuracy": 0.14006469771265984,
      "num_tokens": 12416148.0,
      "step": 5735
    },
    {
      "entropy": 6.541106081008911,
      "epoch": 0.6142650757129863,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004970881644444459,
      "loss": 6.3076,
      "mean_token_accuracy": 0.13991214632987975,
      "num_tokens": 12426182.0,
      "step": 5740
    },
    {
      "entropy": 6.590402269363404,
      "epoch": 0.6148001498207502,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004970820301162178,
      "loss": 6.4833,
      "mean_token_accuracy": 0.1292591094970703,
      "num_tokens": 12437548.0,
      "step": 5745
    },
    {
      "entropy": 6.592732715606689,
      "epoch": 0.6153352239285141,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004970758893754022,
      "loss": 6.4092,
      "mean_token_accuracy": 0.12948114275932313,
      "num_tokens": 12449720.0,
      "step": 5750
    },
    {
      "entropy": 6.630194807052613,
      "epoch": 0.615870298036278,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004970697422221767,
      "loss": 6.3971,
      "mean_token_accuracy": 0.13294799551367759,
      "num_tokens": 12460666.0,
      "step": 5755
    },
    {
      "entropy": 6.558688449859619,
      "epoch": 0.6164053721440419,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004970635886567185,
      "loss": 6.3956,
      "mean_token_accuracy": 0.1334315963089466,
      "num_tokens": 12471472.0,
      "step": 5760
    },
    {
      "entropy": 6.496923494338989,
      "epoch": 0.6169404462518059,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004970574286792054,
      "loss": 6.3846,
      "mean_token_accuracy": 0.13120403438806533,
      "num_tokens": 12482127.0,
      "step": 5765
    },
    {
      "entropy": 6.568459987640381,
      "epoch": 0.6174755203595698,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004970512622898152,
      "loss": 6.3757,
      "mean_token_accuracy": 0.13721705228090286,
      "num_tokens": 12493680.0,
      "step": 5770
    },
    {
      "entropy": 6.528800058364868,
      "epoch": 0.6180105944673338,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000497045089488726,
      "loss": 6.3354,
      "mean_token_accuracy": 0.13322472646832467,
      "num_tokens": 12505838.0,
      "step": 5775
    },
    {
      "entropy": 6.586766910552979,
      "epoch": 0.6185456685750976,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004970389102761161,
      "loss": 6.3525,
      "mean_token_accuracy": 0.13949506655335425,
      "num_tokens": 12516334.0,
      "step": 5780
    },
    {
      "entropy": 6.524970054626465,
      "epoch": 0.6190807426828616,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970327246521638,
      "loss": 6.3963,
      "mean_token_accuracy": 0.1295120485126972,
      "num_tokens": 12526440.0,
      "step": 5785
    },
    {
      "entropy": 6.550500583648682,
      "epoch": 0.6196158167906255,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004970265326170478,
      "loss": 6.3384,
      "mean_token_accuracy": 0.12922370806336403,
      "num_tokens": 12538127.0,
      "step": 5790
    },
    {
      "entropy": 6.614361572265625,
      "epoch": 0.6201508908983894,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004970203341709469,
      "loss": 6.4204,
      "mean_token_accuracy": 0.13372067511081695,
      "num_tokens": 12548459.0,
      "step": 5795
    },
    {
      "entropy": 6.504186201095581,
      "epoch": 0.6206859650061534,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00049701412931404,
      "loss": 6.3551,
      "mean_token_accuracy": 0.13731214702129363,
      "num_tokens": 12558989.0,
      "step": 5800
    },
    {
      "entropy": 6.5102873802185055,
      "epoch": 0.6212210391139172,
      "grad_norm": 0.875,
      "learning_rate": 0.0004970079180465064,
      "loss": 6.2834,
      "mean_token_accuracy": 0.14180205017328262,
      "num_tokens": 12569449.0,
      "step": 5805
    },
    {
      "entropy": 6.547976398468018,
      "epoch": 0.6217561132216812,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004970017003685253,
      "loss": 6.3543,
      "mean_token_accuracy": 0.1336100295186043,
      "num_tokens": 12580733.0,
      "step": 5810
    },
    {
      "entropy": 6.552909994125367,
      "epoch": 0.6222911873294451,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004969954762802763,
      "loss": 6.3611,
      "mean_token_accuracy": 0.13334628865122794,
      "num_tokens": 12590869.0,
      "step": 5815
    },
    {
      "entropy": 6.600517606735229,
      "epoch": 0.6228262614372091,
      "grad_norm": 0.9765625,
      "learning_rate": 0.000496989245781939,
      "loss": 6.4181,
      "mean_token_accuracy": 0.13195589184761047,
      "num_tokens": 12602109.0,
      "step": 5820
    },
    {
      "entropy": 6.524480295181275,
      "epoch": 0.623361335544973,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004969830088736937,
      "loss": 6.3577,
      "mean_token_accuracy": 0.13163684085011482,
      "num_tokens": 12612234.0,
      "step": 5825
    },
    {
      "entropy": 6.640629863739013,
      "epoch": 0.6238964096527368,
      "grad_norm": 0.890625,
      "learning_rate": 0.00049697676555572,
      "loss": 6.5171,
      "mean_token_accuracy": 0.13004111051559447,
      "num_tokens": 12623252.0,
      "step": 5830
    },
    {
      "entropy": 6.637633466720581,
      "epoch": 0.6244314837605008,
      "grad_norm": 0.875,
      "learning_rate": 0.0004969705158281985,
      "loss": 6.2591,
      "mean_token_accuracy": 0.1430523671209812,
      "num_tokens": 12632796.0,
      "step": 5835
    },
    {
      "entropy": 6.52786078453064,
      "epoch": 0.6249665578682647,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969642596913095,
      "loss": 6.3471,
      "mean_token_accuracy": 0.1345833159983158,
      "num_tokens": 12643278.0,
      "step": 5840
    },
    {
      "entropy": 6.538140487670899,
      "epoch": 0.6255016319760287,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004969579971452337,
      "loss": 6.4262,
      "mean_token_accuracy": 0.13390445560216904,
      "num_tokens": 12654446.0,
      "step": 5845
    },
    {
      "entropy": 6.564716768264771,
      "epoch": 0.6260367060837926,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004969517281901519,
      "loss": 6.4057,
      "mean_token_accuracy": 0.13656433820724487,
      "num_tokens": 12664281.0,
      "step": 5850
    },
    {
      "entropy": 6.584988307952881,
      "epoch": 0.6265717801915566,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004969454528262451,
      "loss": 6.3973,
      "mean_token_accuracy": 0.13601098954677582,
      "num_tokens": 12675252.0,
      "step": 5855
    },
    {
      "entropy": 6.534916353225708,
      "epoch": 0.6271068542993204,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004969391710536946,
      "loss": 6.3701,
      "mean_token_accuracy": 0.13320676386356353,
      "num_tokens": 12685764.0,
      "step": 5860
    },
    {
      "entropy": 6.454722547531128,
      "epoch": 0.6276419284070843,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004969328828726817,
      "loss": 6.2897,
      "mean_token_accuracy": 0.13955534845590592,
      "num_tokens": 12696974.0,
      "step": 5865
    },
    {
      "entropy": 6.6120130062103275,
      "epoch": 0.6281770025148483,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004969265882833879,
      "loss": 6.452,
      "mean_token_accuracy": 0.12457979917526245,
      "num_tokens": 12708217.0,
      "step": 5870
    },
    {
      "entropy": 6.594373893737793,
      "epoch": 0.6287120766226122,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004969202872859952,
      "loss": 6.3037,
      "mean_token_accuracy": 0.13776595368981362,
      "num_tokens": 12717840.0,
      "step": 5875
    },
    {
      "entropy": 6.483970737457275,
      "epoch": 0.6292471507303762,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004969139798806854,
      "loss": 6.3973,
      "mean_token_accuracy": 0.1298674687743187,
      "num_tokens": 12728390.0,
      "step": 5880
    },
    {
      "entropy": 6.540892505645752,
      "epoch": 0.6297822248381401,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004969076660676405,
      "loss": 6.4003,
      "mean_token_accuracy": 0.13340216055512427,
      "num_tokens": 12738650.0,
      "step": 5885
    },
    {
      "entropy": 6.595161819458008,
      "epoch": 0.630317298945904,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000496901345847043,
      "loss": 6.4032,
      "mean_token_accuracy": 0.13205326199531556,
      "num_tokens": 12748989.0,
      "step": 5890
    },
    {
      "entropy": 6.5864780902862545,
      "epoch": 0.6308523730536679,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004968950192190752,
      "loss": 6.3649,
      "mean_token_accuracy": 0.1320453979074955,
      "num_tokens": 12761132.0,
      "step": 5895
    },
    {
      "entropy": 6.517518615722656,
      "epoch": 0.6313874471614318,
      "grad_norm": 0.890625,
      "learning_rate": 0.00049688868618392,
      "loss": 6.3788,
      "mean_token_accuracy": 0.13421476930379866,
      "num_tokens": 12771478.0,
      "step": 5900
    },
    {
      "entropy": 6.581002044677734,
      "epoch": 0.6319225212691958,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004968823467417602,
      "loss": 6.4344,
      "mean_token_accuracy": 0.13052433654665946,
      "num_tokens": 12782096.0,
      "step": 5905
    },
    {
      "entropy": 6.563936185836792,
      "epoch": 0.6324575953769597,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004968760008927788,
      "loss": 6.3382,
      "mean_token_accuracy": 0.13559918701648713,
      "num_tokens": 12792467.0,
      "step": 5910
    },
    {
      "entropy": 6.576694536209106,
      "epoch": 0.6329926694847237,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496869648637159,
      "loss": 6.3267,
      "mean_token_accuracy": 0.14201204627752304,
      "num_tokens": 12803248.0,
      "step": 5915
    },
    {
      "entropy": 6.562338829040527,
      "epoch": 0.6335277435924875,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004968632899750844,
      "loss": 6.354,
      "mean_token_accuracy": 0.13680187240242958,
      "num_tokens": 12813701.0,
      "step": 5920
    },
    {
      "entropy": 6.6017234325408936,
      "epoch": 0.6340628177002515,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004968569249067383,
      "loss": 6.4335,
      "mean_token_accuracy": 0.12925372123718262,
      "num_tokens": 12824945.0,
      "step": 5925
    },
    {
      "entropy": 6.566445016860962,
      "epoch": 0.6345978918080154,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004968505534323047,
      "loss": 6.3768,
      "mean_token_accuracy": 0.13098957762122154,
      "num_tokens": 12836699.0,
      "step": 5930
    },
    {
      "entropy": 6.5367591381073,
      "epoch": 0.6351329659157794,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004968441755519677,
      "loss": 6.3637,
      "mean_token_accuracy": 0.131689964979887,
      "num_tokens": 12846905.0,
      "step": 5935
    },
    {
      "entropy": 6.582563877105713,
      "epoch": 0.6356680400235433,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004968377912659112,
      "loss": 6.2806,
      "mean_token_accuracy": 0.1423302873969078,
      "num_tokens": 12856919.0,
      "step": 5940
    },
    {
      "entropy": 6.4202080249786375,
      "epoch": 0.6362031141313071,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004968314005743196,
      "loss": 6.272,
      "mean_token_accuracy": 0.14061389490962029,
      "num_tokens": 12867863.0,
      "step": 5945
    },
    {
      "entropy": 6.480769872665405,
      "epoch": 0.6367381882390711,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004968250034773776,
      "loss": 6.346,
      "mean_token_accuracy": 0.1376643568277359,
      "num_tokens": 12877877.0,
      "step": 5950
    },
    {
      "entropy": 6.656388235092163,
      "epoch": 0.637273262346835,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004968185999752697,
      "loss": 6.4354,
      "mean_token_accuracy": 0.12367877438664436,
      "num_tokens": 12887954.0,
      "step": 5955
    },
    {
      "entropy": 6.554588699340821,
      "epoch": 0.637808336454599,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004968121900681809,
      "loss": 6.361,
      "mean_token_accuracy": 0.13090814650058746,
      "num_tokens": 12897744.0,
      "step": 5960
    },
    {
      "entropy": 6.55085768699646,
      "epoch": 0.6383434105623629,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004968057737562964,
      "loss": 6.4384,
      "mean_token_accuracy": 0.13585126921534538,
      "num_tokens": 12909193.0,
      "step": 5965
    },
    {
      "entropy": 6.55703182220459,
      "epoch": 0.6388784846701269,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967993510398012,
      "loss": 6.3074,
      "mean_token_accuracy": 0.13894035816192626,
      "num_tokens": 12919879.0,
      "step": 5970
    },
    {
      "entropy": 6.53627986907959,
      "epoch": 0.6394135587778907,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967929219188809,
      "loss": 6.367,
      "mean_token_accuracy": 0.13364173769950866,
      "num_tokens": 12930297.0,
      "step": 5975
    },
    {
      "entropy": 6.533883190155029,
      "epoch": 0.6399486328856546,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004967864863937212,
      "loss": 6.3185,
      "mean_token_accuracy": 0.1434766449034214,
      "num_tokens": 12940467.0,
      "step": 5980
    },
    {
      "entropy": 6.5478757381439205,
      "epoch": 0.6404837069934186,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967800444645079,
      "loss": 6.3365,
      "mean_token_accuracy": 0.13498707339167595,
      "num_tokens": 12950417.0,
      "step": 5985
    },
    {
      "entropy": 6.495828294754029,
      "epoch": 0.6410187811011825,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004967735961314269,
      "loss": 6.29,
      "mean_token_accuracy": 0.14207748472690582,
      "num_tokens": 12960957.0,
      "step": 5990
    },
    {
      "entropy": 6.482624292373657,
      "epoch": 0.6415538552089465,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004967671413946646,
      "loss": 6.3971,
      "mean_token_accuracy": 0.13469642549753189,
      "num_tokens": 12971194.0,
      "step": 5995
    },
    {
      "entropy": 6.558772277832031,
      "epoch": 0.6420889293167104,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004967606802544071,
      "loss": 6.3775,
      "mean_token_accuracy": 0.13109284415841102,
      "num_tokens": 12982479.0,
      "step": 6000
    },
    {
      "epoch": 0.6420889293167104,
      "eval_entropy": 6.383020276329267,
      "eval_loss": 6.415825366973877,
      "eval_mean_token_accuracy": 0.13898081381747504,
      "eval_num_tokens": 12982479.0,
      "eval_runtime": 17.5694,
      "eval_samples_per_second": 1689.24,
      "eval_steps_per_second": 211.162,
      "step": 6000
    },
    {
      "entropy": 6.591076946258545,
      "epoch": 0.6426240034244743,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004967542127108412,
      "loss": 6.4288,
      "mean_token_accuracy": 0.12930821031332015,
      "num_tokens": 12994119.0,
      "step": 6005
    },
    {
      "entropy": 6.578906488418579,
      "epoch": 0.6431590775322382,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004967477387641537,
      "loss": 6.3498,
      "mean_token_accuracy": 0.1371295042335987,
      "num_tokens": 13005437.0,
      "step": 6010
    },
    {
      "entropy": 6.5153087139129635,
      "epoch": 0.6436941516400021,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004967412584145312,
      "loss": 6.3602,
      "mean_token_accuracy": 0.1351504795253277,
      "num_tokens": 13017278.0,
      "step": 6015
    },
    {
      "entropy": 6.5156495571136475,
      "epoch": 0.6442292257477661,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004967347716621611,
      "loss": 6.3372,
      "mean_token_accuracy": 0.13614037558436393,
      "num_tokens": 13028010.0,
      "step": 6020
    },
    {
      "entropy": 6.577405118942261,
      "epoch": 0.64476429985553,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004967282785072306,
      "loss": 6.3242,
      "mean_token_accuracy": 0.1378284588456154,
      "num_tokens": 13038740.0,
      "step": 6025
    },
    {
      "entropy": 6.50105562210083,
      "epoch": 0.645299373963294,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004967217789499272,
      "loss": 6.3512,
      "mean_token_accuracy": 0.13472929447889329,
      "num_tokens": 13049921.0,
      "step": 6030
    },
    {
      "entropy": 6.4590959548950195,
      "epoch": 0.6458344480710578,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004967152729904386,
      "loss": 6.2649,
      "mean_token_accuracy": 0.14219320937991142,
      "num_tokens": 13060907.0,
      "step": 6035
    },
    {
      "entropy": 6.536409521102906,
      "epoch": 0.6463695221788218,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004967087606289527,
      "loss": 6.3143,
      "mean_token_accuracy": 0.14407092854380607,
      "num_tokens": 13071720.0,
      "step": 6040
    },
    {
      "entropy": 6.536934518814087,
      "epoch": 0.6469045962865857,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004967022418656575,
      "loss": 6.3551,
      "mean_token_accuracy": 0.1360303618013859,
      "num_tokens": 13082474.0,
      "step": 6045
    },
    {
      "entropy": 6.517414903640747,
      "epoch": 0.6474396703943496,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004966957167007412,
      "loss": 6.2695,
      "mean_token_accuracy": 0.13758466467261316,
      "num_tokens": 13092777.0,
      "step": 6050
    },
    {
      "entropy": 6.464146471023559,
      "epoch": 0.6479747445021136,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966891851343922,
      "loss": 6.2711,
      "mean_token_accuracy": 0.14155081808567047,
      "num_tokens": 13104019.0,
      "step": 6055
    },
    {
      "entropy": 6.608423948287964,
      "epoch": 0.6485098186098774,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004966826471667992,
      "loss": 6.2927,
      "mean_token_accuracy": 0.1424348384141922,
      "num_tokens": 13115741.0,
      "step": 6060
    },
    {
      "entropy": 6.54501576423645,
      "epoch": 0.6490448927176414,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004966761027981509,
      "loss": 6.3272,
      "mean_token_accuracy": 0.13803260400891304,
      "num_tokens": 13126838.0,
      "step": 6065
    },
    {
      "entropy": 6.464467906951905,
      "epoch": 0.6495799668254053,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004966695520286362,
      "loss": 6.416,
      "mean_token_accuracy": 0.13495837524533272,
      "num_tokens": 13137485.0,
      "step": 6070
    },
    {
      "entropy": 6.549513626098633,
      "epoch": 0.6501150409331693,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004966629948584444,
      "loss": 6.298,
      "mean_token_accuracy": 0.138609080016613,
      "num_tokens": 13147631.0,
      "step": 6075
    },
    {
      "entropy": 6.469826984405517,
      "epoch": 0.6506501150409332,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966564312877647,
      "loss": 6.3508,
      "mean_token_accuracy": 0.13750356957316398,
      "num_tokens": 13158468.0,
      "step": 6080
    },
    {
      "entropy": 6.581138849258423,
      "epoch": 0.651185189148697,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966498613167868,
      "loss": 6.3767,
      "mean_token_accuracy": 0.13204531446099282,
      "num_tokens": 13169263.0,
      "step": 6085
    },
    {
      "entropy": 6.567144441604614,
      "epoch": 0.651720263256461,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004966432849457002,
      "loss": 6.3019,
      "mean_token_accuracy": 0.14222824424505234,
      "num_tokens": 13178647.0,
      "step": 6090
    },
    {
      "entropy": 6.483666181564331,
      "epoch": 0.6522553373642249,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000496636702174695,
      "loss": 6.2931,
      "mean_token_accuracy": 0.13921134546399117,
      "num_tokens": 13189883.0,
      "step": 6095
    },
    {
      "entropy": 6.533740234375,
      "epoch": 0.6527904114719889,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004966301130039611,
      "loss": 6.4142,
      "mean_token_accuracy": 0.13619857504963875,
      "num_tokens": 13200920.0,
      "step": 6100
    },
    {
      "entropy": 6.497662544250488,
      "epoch": 0.6533254855797528,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004966235174336889,
      "loss": 6.3993,
      "mean_token_accuracy": 0.12920630872249603,
      "num_tokens": 13211765.0,
      "step": 6105
    },
    {
      "entropy": 6.5689723014831545,
      "epoch": 0.6538605596875168,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004966169154640687,
      "loss": 6.3311,
      "mean_token_accuracy": 0.13576391562819481,
      "num_tokens": 13221768.0,
      "step": 6110
    },
    {
      "entropy": 6.542498636245727,
      "epoch": 0.6543956337952807,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004966103070952912,
      "loss": 6.3216,
      "mean_token_accuracy": 0.14183156117796897,
      "num_tokens": 13233545.0,
      "step": 6115
    },
    {
      "entropy": 6.4608306884765625,
      "epoch": 0.6549307079030445,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004966036923275472,
      "loss": 6.3757,
      "mean_token_accuracy": 0.14127599596977233,
      "num_tokens": 13245018.0,
      "step": 6120
    },
    {
      "entropy": 6.5395012378692625,
      "epoch": 0.6554657820108085,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004965970711610278,
      "loss": 6.2876,
      "mean_token_accuracy": 0.13406386077404023,
      "num_tokens": 13255772.0,
      "step": 6125
    },
    {
      "entropy": 6.547164535522461,
      "epoch": 0.6560008561185724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965904435959241,
      "loss": 6.3557,
      "mean_token_accuracy": 0.13506652861833573,
      "num_tokens": 13267271.0,
      "step": 6130
    },
    {
      "entropy": 6.556967639923096,
      "epoch": 0.6565359302263364,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004965838096324274,
      "loss": 6.3603,
      "mean_token_accuracy": 0.13804064318537712,
      "num_tokens": 13278467.0,
      "step": 6135
    },
    {
      "entropy": 6.55072660446167,
      "epoch": 0.6570710043341003,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004965771692707294,
      "loss": 6.4318,
      "mean_token_accuracy": 0.13093264251947404,
      "num_tokens": 13288522.0,
      "step": 6140
    },
    {
      "entropy": 6.602171182632446,
      "epoch": 0.6576060784418643,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004965705225110218,
      "loss": 6.4291,
      "mean_token_accuracy": 0.12572090178728104,
      "num_tokens": 13299355.0,
      "step": 6145
    },
    {
      "entropy": 6.55772123336792,
      "epoch": 0.6581411525496281,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004965638693534964,
      "loss": 6.3431,
      "mean_token_accuracy": 0.14018580093979835,
      "num_tokens": 13310334.0,
      "step": 6150
    },
    {
      "entropy": 6.550020837783814,
      "epoch": 0.658676226657392,
      "grad_norm": 0.875,
      "learning_rate": 0.0004965572097983455,
      "loss": 6.3695,
      "mean_token_accuracy": 0.13387906774878502,
      "num_tokens": 13320917.0,
      "step": 6155
    },
    {
      "entropy": 6.569891929626465,
      "epoch": 0.659211300765156,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004965505438457613,
      "loss": 6.2847,
      "mean_token_accuracy": 0.13883678689599038,
      "num_tokens": 13331707.0,
      "step": 6160
    },
    {
      "entropy": 6.5394927024841305,
      "epoch": 0.6597463748729199,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004965438714959363,
      "loss": 6.429,
      "mean_token_accuracy": 0.12999134734272957,
      "num_tokens": 13341784.0,
      "step": 6165
    },
    {
      "entropy": 6.608139896392823,
      "epoch": 0.6602814489806839,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000496537192749063,
      "loss": 6.3849,
      "mean_token_accuracy": 0.13332228735089302,
      "num_tokens": 13351633.0,
      "step": 6170
    },
    {
      "entropy": 6.569412803649902,
      "epoch": 0.6608165230884477,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004965305076053345,
      "loss": 6.2882,
      "mean_token_accuracy": 0.14427342563867568,
      "num_tokens": 13362444.0,
      "step": 6175
    },
    {
      "entropy": 6.502989387512207,
      "epoch": 0.6613515971962117,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004965238160649437,
      "loss": 6.4167,
      "mean_token_accuracy": 0.13150877356529236,
      "num_tokens": 13374322.0,
      "step": 6180
    },
    {
      "entropy": 6.573364210128784,
      "epoch": 0.6618866713039756,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004965171181280838,
      "loss": 6.3844,
      "mean_token_accuracy": 0.1348087079823017,
      "num_tokens": 13384188.0,
      "step": 6185
    },
    {
      "entropy": 6.585802316665649,
      "epoch": 0.6624217454117395,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004965104137949483,
      "loss": 6.383,
      "mean_token_accuracy": 0.12980616390705108,
      "num_tokens": 13395648.0,
      "step": 6190
    },
    {
      "entropy": 6.606305265426636,
      "epoch": 0.6629568195195035,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004965037030657308,
      "loss": 6.3957,
      "mean_token_accuracy": 0.13427068889141083,
      "num_tokens": 13406769.0,
      "step": 6195
    },
    {
      "entropy": 6.60039119720459,
      "epoch": 0.6634918936272673,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004964969859406248,
      "loss": 6.3627,
      "mean_token_accuracy": 0.12875289767980574,
      "num_tokens": 13417509.0,
      "step": 6200
    },
    {
      "entropy": 6.547433280944825,
      "epoch": 0.6640269677350313,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004964902624198246,
      "loss": 6.2895,
      "mean_token_accuracy": 0.13346439301967622,
      "num_tokens": 13429441.0,
      "step": 6205
    },
    {
      "entropy": 6.470551347732544,
      "epoch": 0.6645620418427952,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004964835325035243,
      "loss": 6.3223,
      "mean_token_accuracy": 0.1374746859073639,
      "num_tokens": 13439332.0,
      "step": 6210
    },
    {
      "entropy": 6.480902814865113,
      "epoch": 0.6650971159505592,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004964767961919179,
      "loss": 6.3016,
      "mean_token_accuracy": 0.13590085953474046,
      "num_tokens": 13449400.0,
      "step": 6215
    },
    {
      "entropy": 6.597186613082886,
      "epoch": 0.6656321900583231,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004964700534852003,
      "loss": 6.4121,
      "mean_token_accuracy": 0.13142415061593055,
      "num_tokens": 13460134.0,
      "step": 6220
    },
    {
      "entropy": 6.536060047149658,
      "epoch": 0.666167264166087,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004964633043835658,
      "loss": 6.3634,
      "mean_token_accuracy": 0.13759579584002496,
      "num_tokens": 13469954.0,
      "step": 6225
    },
    {
      "entropy": 6.550062990188598,
      "epoch": 0.666702338273851,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004964565488872097,
      "loss": 6.3406,
      "mean_token_accuracy": 0.1375626415014267,
      "num_tokens": 13479866.0,
      "step": 6230
    },
    {
      "entropy": 6.457085180282593,
      "epoch": 0.6672374123816148,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004964497869963268,
      "loss": 6.2851,
      "mean_token_accuracy": 0.1391478732228279,
      "num_tokens": 13490088.0,
      "step": 6235
    },
    {
      "entropy": 6.536445426940918,
      "epoch": 0.6677724864893788,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004964430187111125,
      "loss": 6.3925,
      "mean_token_accuracy": 0.13275325670838356,
      "num_tokens": 13500898.0,
      "step": 6240
    },
    {
      "entropy": 6.529649829864502,
      "epoch": 0.6683075605971427,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004964362440317621,
      "loss": 6.4025,
      "mean_token_accuracy": 0.13160213604569435,
      "num_tokens": 13511537.0,
      "step": 6245
    },
    {
      "entropy": 6.5402873992919925,
      "epoch": 0.6688426347049067,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004964294629584712,
      "loss": 6.3436,
      "mean_token_accuracy": 0.13360676541924477,
      "num_tokens": 13522916.0,
      "step": 6250
    },
    {
      "entropy": 6.50937123298645,
      "epoch": 0.6693777088126706,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004964226754914357,
      "loss": 6.2894,
      "mean_token_accuracy": 0.13634295910596847,
      "num_tokens": 13533817.0,
      "step": 6255
    },
    {
      "entropy": 6.536599111557007,
      "epoch": 0.6699127829204344,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004964158816308517,
      "loss": 6.3785,
      "mean_token_accuracy": 0.13008598610758781,
      "num_tokens": 13543723.0,
      "step": 6260
    },
    {
      "entropy": 6.50706934928894,
      "epoch": 0.6704478570281984,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004964090813769151,
      "loss": 6.2569,
      "mean_token_accuracy": 0.14148564413189887,
      "num_tokens": 13555269.0,
      "step": 6265
    },
    {
      "entropy": 6.566424083709717,
      "epoch": 0.6709829311359623,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004964022747298224,
      "loss": 6.4121,
      "mean_token_accuracy": 0.13113898783922195,
      "num_tokens": 13567465.0,
      "step": 6270
    },
    {
      "entropy": 6.65675458908081,
      "epoch": 0.6715180052437263,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004963954616897701,
      "loss": 6.3532,
      "mean_token_accuracy": 0.13427343294024469,
      "num_tokens": 13578637.0,
      "step": 6275
    },
    {
      "entropy": 6.5185935497283936,
      "epoch": 0.6720530793514902,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000496388642256955,
      "loss": 6.3624,
      "mean_token_accuracy": 0.13522644713521004,
      "num_tokens": 13589074.0,
      "step": 6280
    },
    {
      "entropy": 6.453289651870728,
      "epoch": 0.6725881534592542,
      "grad_norm": 0.890625,
      "learning_rate": 0.000496381816431574,
      "loss": 6.3123,
      "mean_token_accuracy": 0.13581889048218726,
      "num_tokens": 13599624.0,
      "step": 6285
    },
    {
      "entropy": 6.583613777160645,
      "epoch": 0.673123227567018,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004963749842138241,
      "loss": 6.3408,
      "mean_token_accuracy": 0.13643057271838188,
      "num_tokens": 13610293.0,
      "step": 6290
    },
    {
      "entropy": 6.5441490650177006,
      "epoch": 0.6736583016747819,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004963681456039027,
      "loss": 6.3573,
      "mean_token_accuracy": 0.1366145357489586,
      "num_tokens": 13621641.0,
      "step": 6295
    },
    {
      "entropy": 6.528734970092773,
      "epoch": 0.6741933757825459,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004963613006020072,
      "loss": 6.3254,
      "mean_token_accuracy": 0.12962572723627092,
      "num_tokens": 13632083.0,
      "step": 6300
    },
    {
      "entropy": 6.487830448150635,
      "epoch": 0.6747284498903098,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004963544492083352,
      "loss": 6.1881,
      "mean_token_accuracy": 0.1562342181801796,
      "num_tokens": 13644084.0,
      "step": 6305
    },
    {
      "entropy": 6.463047790527344,
      "epoch": 0.6752635239980738,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004963475914230845,
      "loss": 6.3522,
      "mean_token_accuracy": 0.13347619250416756,
      "num_tokens": 13656231.0,
      "step": 6310
    },
    {
      "entropy": 6.610400629043579,
      "epoch": 0.6757985981058376,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004963407272464533,
      "loss": 6.4062,
      "mean_token_accuracy": 0.13457160592079162,
      "num_tokens": 13667334.0,
      "step": 6315
    },
    {
      "entropy": 6.599489688873291,
      "epoch": 0.6763336722136016,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004963338566786398,
      "loss": 6.3742,
      "mean_token_accuracy": 0.1362266592681408,
      "num_tokens": 13678615.0,
      "step": 6320
    },
    {
      "entropy": 6.542865371704101,
      "epoch": 0.6768687463213655,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004963269797198422,
      "loss": 6.3715,
      "mean_token_accuracy": 0.13775787129998207,
      "num_tokens": 13689890.0,
      "step": 6325
    },
    {
      "entropy": 6.4303265571594235,
      "epoch": 0.6774038204291294,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004963200963702592,
      "loss": 6.2377,
      "mean_token_accuracy": 0.14377759397029877,
      "num_tokens": 13699120.0,
      "step": 6330
    },
    {
      "entropy": 6.489156198501587,
      "epoch": 0.6779388945368934,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004963132066300894,
      "loss": 6.3484,
      "mean_token_accuracy": 0.1312595598399639,
      "num_tokens": 13710016.0,
      "step": 6335
    },
    {
      "entropy": 6.591976928710937,
      "epoch": 0.6784739686446573,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004963063104995319,
      "loss": 6.4208,
      "mean_token_accuracy": 0.123978491127491,
      "num_tokens": 13721576.0,
      "step": 6340
    },
    {
      "entropy": 6.604969596862793,
      "epoch": 0.6790090427524212,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004962994079787859,
      "loss": 6.4036,
      "mean_token_accuracy": 0.13739463537931443,
      "num_tokens": 13732568.0,
      "step": 6345
    },
    {
      "entropy": 6.506201934814453,
      "epoch": 0.6795441168601851,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962924990680504,
      "loss": 6.28,
      "mean_token_accuracy": 0.1411899022758007,
      "num_tokens": 13742569.0,
      "step": 6350
    },
    {
      "entropy": 6.52852578163147,
      "epoch": 0.6800791909679491,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004962855837675252,
      "loss": 6.4066,
      "mean_token_accuracy": 0.1306900292634964,
      "num_tokens": 13752689.0,
      "step": 6355
    },
    {
      "entropy": 6.58823971748352,
      "epoch": 0.680614265075713,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004962786620774099,
      "loss": 6.3729,
      "mean_token_accuracy": 0.1385651059448719,
      "num_tokens": 13763709.0,
      "step": 6360
    },
    {
      "entropy": 6.528443193435669,
      "epoch": 0.6811493391834769,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004962717339979043,
      "loss": 6.2834,
      "mean_token_accuracy": 0.14149898886680604,
      "num_tokens": 13773726.0,
      "step": 6365
    },
    {
      "entropy": 6.517940521240234,
      "epoch": 0.6816844132912409,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0004962647995292084,
      "loss": 6.3242,
      "mean_token_accuracy": 0.13827024698257445,
      "num_tokens": 13785435.0,
      "step": 6370
    },
    {
      "entropy": 6.535262107849121,
      "epoch": 0.6822194873990047,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004962578586715226,
      "loss": 6.3226,
      "mean_token_accuracy": 0.1321915939450264,
      "num_tokens": 13797104.0,
      "step": 6375
    },
    {
      "entropy": 6.607646894454956,
      "epoch": 0.6827545615067687,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004962509114250471,
      "loss": 6.4155,
      "mean_token_accuracy": 0.13438887000083924,
      "num_tokens": 13807756.0,
      "step": 6380
    },
    {
      "entropy": 6.371178197860718,
      "epoch": 0.6832896356145326,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004962439577899828,
      "loss": 6.1084,
      "mean_token_accuracy": 0.15405212193727494,
      "num_tokens": 13818490.0,
      "step": 6385
    },
    {
      "entropy": 6.540214729309082,
      "epoch": 0.6838247097222966,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004962369977665301,
      "loss": 6.3995,
      "mean_token_accuracy": 0.1326890356838703,
      "num_tokens": 13829465.0,
      "step": 6390
    },
    {
      "entropy": 6.588983011245728,
      "epoch": 0.6843597838300605,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004962300313548903,
      "loss": 6.3343,
      "mean_token_accuracy": 0.1356295645236969,
      "num_tokens": 13840863.0,
      "step": 6395
    },
    {
      "entropy": 6.582163715362549,
      "epoch": 0.6848948579378243,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004962230585552645,
      "loss": 6.356,
      "mean_token_accuracy": 0.1350361503660679,
      "num_tokens": 13851711.0,
      "step": 6400
    },
    {
      "entropy": 6.51849479675293,
      "epoch": 0.6854299320455883,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004962160793678539,
      "loss": 6.3847,
      "mean_token_accuracy": 0.13450609743595124,
      "num_tokens": 13864576.0,
      "step": 6405
    },
    {
      "entropy": 6.497533178329467,
      "epoch": 0.6859650061533522,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049620909379286,
      "loss": 6.2858,
      "mean_token_accuracy": 0.14179206341505052,
      "num_tokens": 13875150.0,
      "step": 6410
    },
    {
      "entropy": 6.5381104946136475,
      "epoch": 0.6865000802611162,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004962021018304847,
      "loss": 6.3236,
      "mean_token_accuracy": 0.1370361328125,
      "num_tokens": 13886070.0,
      "step": 6415
    },
    {
      "entropy": 6.475835657119751,
      "epoch": 0.6870351543688801,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004961951034809298,
      "loss": 6.2492,
      "mean_token_accuracy": 0.13783403784036635,
      "num_tokens": 13897898.0,
      "step": 6420
    },
    {
      "entropy": 6.595751667022705,
      "epoch": 0.6875702284766441,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961880987443974,
      "loss": 6.3956,
      "mean_token_accuracy": 0.13372014537453653,
      "num_tokens": 13909578.0,
      "step": 6425
    },
    {
      "entropy": 6.412465286254883,
      "epoch": 0.6881053025844079,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004961810876210897,
      "loss": 6.2446,
      "mean_token_accuracy": 0.14092109501361846,
      "num_tokens": 13920537.0,
      "step": 6430
    },
    {
      "entropy": 6.472372817993164,
      "epoch": 0.6886403766921718,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004961740701112091,
      "loss": 6.2984,
      "mean_token_accuracy": 0.14048361852765084,
      "num_tokens": 13931015.0,
      "step": 6435
    },
    {
      "entropy": 6.5291259765625,
      "epoch": 0.6891754507999358,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004961670462149583,
      "loss": 6.339,
      "mean_token_accuracy": 0.1330929972231388,
      "num_tokens": 13942691.0,
      "step": 6440
    },
    {
      "entropy": 6.630201196670532,
      "epoch": 0.6897105249076997,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004961600159325403,
      "loss": 6.4517,
      "mean_token_accuracy": 0.12625204175710678,
      "num_tokens": 13953945.0,
      "step": 6445
    },
    {
      "entropy": 6.603420209884644,
      "epoch": 0.6902455990154637,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961529792641577,
      "loss": 6.4248,
      "mean_token_accuracy": 0.13203159719705582,
      "num_tokens": 13965658.0,
      "step": 6450
    },
    {
      "entropy": 6.566073894500732,
      "epoch": 0.6907806731232276,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004961459362100141,
      "loss": 6.3969,
      "mean_token_accuracy": 0.13572529330849648,
      "num_tokens": 13976184.0,
      "step": 6455
    },
    {
      "entropy": 6.502432203292846,
      "epoch": 0.6913157472309915,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004961388867703125,
      "loss": 6.3307,
      "mean_token_accuracy": 0.12880197241902352,
      "num_tokens": 13986243.0,
      "step": 6460
    },
    {
      "entropy": 6.538561773300171,
      "epoch": 0.6918508213387554,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004961318309452566,
      "loss": 6.309,
      "mean_token_accuracy": 0.13768778294324874,
      "num_tokens": 13995904.0,
      "step": 6465
    },
    {
      "entropy": 6.54346489906311,
      "epoch": 0.6923858954465193,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004961247687350504,
      "loss": 6.3409,
      "mean_token_accuracy": 0.13864050060510635,
      "num_tokens": 14006711.0,
      "step": 6470
    },
    {
      "entropy": 6.533595180511474,
      "epoch": 0.6929209695542833,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004961177001398974,
      "loss": 6.3398,
      "mean_token_accuracy": 0.13462782576680182,
      "num_tokens": 14018177.0,
      "step": 6475
    },
    {
      "entropy": 6.522426795959473,
      "epoch": 0.6934560436620472,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004961106251600018,
      "loss": 6.2816,
      "mean_token_accuracy": 0.13624507933855057,
      "num_tokens": 14028725.0,
      "step": 6480
    },
    {
      "entropy": 6.472539377212525,
      "epoch": 0.6939911177698111,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004961035437955681,
      "loss": 6.3279,
      "mean_token_accuracy": 0.14601708576083183,
      "num_tokens": 14040448.0,
      "step": 6485
    },
    {
      "entropy": 6.4963243961334225,
      "epoch": 0.694526191877575,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004960964560468005,
      "loss": 6.3587,
      "mean_token_accuracy": 0.13147530034184457,
      "num_tokens": 14051731.0,
      "step": 6490
    },
    {
      "entropy": 6.584133195877075,
      "epoch": 0.695061265985339,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004960893619139039,
      "loss": 6.3762,
      "mean_token_accuracy": 0.1344805397093296,
      "num_tokens": 14062330.0,
      "step": 6495
    },
    {
      "entropy": 6.511853218078613,
      "epoch": 0.6955963400931029,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004960822613970831,
      "loss": 6.2513,
      "mean_token_accuracy": 0.13797787055373192,
      "num_tokens": 14073991.0,
      "step": 6500
    },
    {
      "entropy": 6.546637916564942,
      "epoch": 0.6961314142008668,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004960751544965431,
      "loss": 6.3092,
      "mean_token_accuracy": 0.13491785749793053,
      "num_tokens": 14085132.0,
      "step": 6505
    },
    {
      "entropy": 6.482801246643066,
      "epoch": 0.6966664883086308,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004960680412124889,
      "loss": 6.2574,
      "mean_token_accuracy": 0.14044845551252366,
      "num_tokens": 14095371.0,
      "step": 6510
    },
    {
      "entropy": 6.499139022827149,
      "epoch": 0.6972015624163946,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004960609215451261,
      "loss": 6.3275,
      "mean_token_accuracy": 0.13540011495351792,
      "num_tokens": 14106117.0,
      "step": 6515
    },
    {
      "entropy": 6.5513159275054935,
      "epoch": 0.6977366365241586,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004960537954946604,
      "loss": 6.3081,
      "mean_token_accuracy": 0.13781920000910758,
      "num_tokens": 14117112.0,
      "step": 6520
    },
    {
      "entropy": 6.448483991622925,
      "epoch": 0.6982717106319225,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004960466630612974,
      "loss": 6.2546,
      "mean_token_accuracy": 0.14362177699804307,
      "num_tokens": 14128037.0,
      "step": 6525
    },
    {
      "entropy": 6.441561079025268,
      "epoch": 0.6988067847396865,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004960395242452429,
      "loss": 6.2604,
      "mean_token_accuracy": 0.14045727550983428,
      "num_tokens": 14140183.0,
      "step": 6530
    },
    {
      "entropy": 6.5423798084259035,
      "epoch": 0.6993418588474504,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004960323790467033,
      "loss": 6.3996,
      "mean_token_accuracy": 0.13265951126813888,
      "num_tokens": 14151635.0,
      "step": 6535
    },
    {
      "entropy": 6.566456127166748,
      "epoch": 0.6998769329552142,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004960252274658849,
      "loss": 6.3558,
      "mean_token_accuracy": 0.13347001150250434,
      "num_tokens": 14162660.0,
      "step": 6540
    },
    {
      "entropy": 6.609951591491699,
      "epoch": 0.7004120070629782,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004960180695029939,
      "loss": 6.3121,
      "mean_token_accuracy": 0.14251735731959342,
      "num_tokens": 14173496.0,
      "step": 6545
    },
    {
      "entropy": 6.492941761016846,
      "epoch": 0.7009470811707421,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004960109051582374,
      "loss": 6.2757,
      "mean_token_accuracy": 0.13342020139098168,
      "num_tokens": 14185479.0,
      "step": 6550
    },
    {
      "entropy": 6.486592578887939,
      "epoch": 0.7014821552785061,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004960037344318221,
      "loss": 6.253,
      "mean_token_accuracy": 0.1364479474723339,
      "num_tokens": 14197441.0,
      "step": 6555
    },
    {
      "entropy": 6.492273426055908,
      "epoch": 0.70201722938627,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004959965573239548,
      "loss": 6.2036,
      "mean_token_accuracy": 0.14330939874053,
      "num_tokens": 14207832.0,
      "step": 6560
    },
    {
      "entropy": 6.451603555679322,
      "epoch": 0.702552303494034,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004959893738348432,
      "loss": 6.285,
      "mean_token_accuracy": 0.1442940965294838,
      "num_tokens": 14217976.0,
      "step": 6565
    },
    {
      "entropy": 6.565099763870239,
      "epoch": 0.7030873776017978,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004959821839646943,
      "loss": 6.2514,
      "mean_token_accuracy": 0.14138402938842773,
      "num_tokens": 14227557.0,
      "step": 6570
    },
    {
      "entropy": 6.510221767425537,
      "epoch": 0.7036224517095617,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000495974987713716,
      "loss": 6.3514,
      "mean_token_accuracy": 0.13358439654111862,
      "num_tokens": 14238912.0,
      "step": 6575
    },
    {
      "entropy": 6.507627773284912,
      "epoch": 0.7041575258173257,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004959677850821159,
      "loss": 6.291,
      "mean_token_accuracy": 0.135826575756073,
      "num_tokens": 14249968.0,
      "step": 6580
    },
    {
      "entropy": 6.553734540939331,
      "epoch": 0.7046925999250896,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004959605760701021,
      "loss": 6.3686,
      "mean_token_accuracy": 0.13807943612337112,
      "num_tokens": 14260809.0,
      "step": 6585
    },
    {
      "entropy": 6.527280855178833,
      "epoch": 0.7052276740328536,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004959533606778827,
      "loss": 6.3241,
      "mean_token_accuracy": 0.13484576269984244,
      "num_tokens": 14271971.0,
      "step": 6590
    },
    {
      "entropy": 6.548121643066406,
      "epoch": 0.7057627481406175,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004959461389056661,
      "loss": 6.3927,
      "mean_token_accuracy": 0.14100139141082763,
      "num_tokens": 14283550.0,
      "step": 6595
    },
    {
      "entropy": 6.486141300201416,
      "epoch": 0.7062978222483814,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004959389107536608,
      "loss": 6.23,
      "mean_token_accuracy": 0.1378169223666191,
      "num_tokens": 14293769.0,
      "step": 6600
    },
    {
      "entropy": 6.50356159210205,
      "epoch": 0.7068328963561453,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004959316762220754,
      "loss": 6.3178,
      "mean_token_accuracy": 0.13625159189105035,
      "num_tokens": 14304668.0,
      "step": 6605
    },
    {
      "entropy": 6.549765253067017,
      "epoch": 0.7073679704639092,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000495924435311119,
      "loss": 6.388,
      "mean_token_accuracy": 0.13758754581212998,
      "num_tokens": 14315912.0,
      "step": 6610
    },
    {
      "entropy": 6.493173265457154,
      "epoch": 0.7079030445716732,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004959171880210005,
      "loss": 6.342,
      "mean_token_accuracy": 0.13686094656586648,
      "num_tokens": 14327386.0,
      "step": 6615
    },
    {
      "entropy": 6.528710269927979,
      "epoch": 0.7084381186794371,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004959099343519294,
      "loss": 6.3504,
      "mean_token_accuracy": 0.13063807785511017,
      "num_tokens": 14338629.0,
      "step": 6620
    },
    {
      "entropy": 6.5862815380096436,
      "epoch": 0.7089731927872011,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004959026743041148,
      "loss": 6.3635,
      "mean_token_accuracy": 0.1376403383910656,
      "num_tokens": 14349195.0,
      "step": 6625
    },
    {
      "entropy": 6.5694060802459715,
      "epoch": 0.7095082668949649,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004958954078777665,
      "loss": 6.3743,
      "mean_token_accuracy": 0.1331467352807522,
      "num_tokens": 14359449.0,
      "step": 6630
    },
    {
      "entropy": 6.4865234375,
      "epoch": 0.7100433410027289,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004958881350730944,
      "loss": 6.2833,
      "mean_token_accuracy": 0.14074085056781768,
      "num_tokens": 14369904.0,
      "step": 6635
    },
    {
      "entropy": 6.574102067947388,
      "epoch": 0.7105784151104928,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004958808558903085,
      "loss": 6.3086,
      "mean_token_accuracy": 0.13960615694522857,
      "num_tokens": 14382635.0,
      "step": 6640
    },
    {
      "entropy": 6.518001365661621,
      "epoch": 0.7111134892182567,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958735703296189,
      "loss": 6.2667,
      "mean_token_accuracy": 0.1383349306881428,
      "num_tokens": 14393238.0,
      "step": 6645
    },
    {
      "entropy": 6.499480247497559,
      "epoch": 0.7116485633260207,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000495866278391236,
      "loss": 6.314,
      "mean_token_accuracy": 0.1402275539934635,
      "num_tokens": 14403734.0,
      "step": 6650
    },
    {
      "entropy": 6.535943222045899,
      "epoch": 0.7121836374337845,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958589800753703,
      "loss": 6.282,
      "mean_token_accuracy": 0.1409186951816082,
      "num_tokens": 14414201.0,
      "step": 6655
    },
    {
      "entropy": 6.521246767044067,
      "epoch": 0.7127187115415485,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004958516753822326,
      "loss": 6.312,
      "mean_token_accuracy": 0.14062219932675363,
      "num_tokens": 14425153.0,
      "step": 6660
    },
    {
      "entropy": 6.527857637405395,
      "epoch": 0.7132537856493124,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004958443643120338,
      "loss": 6.2973,
      "mean_token_accuracy": 0.143423992395401,
      "num_tokens": 14435942.0,
      "step": 6665
    },
    {
      "entropy": 6.532345724105835,
      "epoch": 0.7137888597570764,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000495837046864985,
      "loss": 6.2887,
      "mean_token_accuracy": 0.13781173378229142,
      "num_tokens": 14446371.0,
      "step": 6670
    },
    {
      "entropy": 6.526313304901123,
      "epoch": 0.7143239338648403,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004958297230412976,
      "loss": 6.2101,
      "mean_token_accuracy": 0.14608053863048553,
      "num_tokens": 14456598.0,
      "step": 6675
    },
    {
      "entropy": 6.465871715545655,
      "epoch": 0.7148590079726042,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004958223928411829,
      "loss": 6.3143,
      "mean_token_accuracy": 0.13014661967754365,
      "num_tokens": 14466775.0,
      "step": 6680
    },
    {
      "entropy": 6.443863487243652,
      "epoch": 0.7153940820803681,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004958150562648526,
      "loss": 6.2738,
      "mean_token_accuracy": 0.1357213944196701,
      "num_tokens": 14476876.0,
      "step": 6685
    },
    {
      "entropy": 6.475451421737671,
      "epoch": 0.715929156188132,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004958077133125187,
      "loss": 6.2752,
      "mean_token_accuracy": 0.14165417924523355,
      "num_tokens": 14487698.0,
      "step": 6690
    },
    {
      "entropy": 6.581821632385254,
      "epoch": 0.716464230295896,
      "grad_norm": 0.8203125,
      "learning_rate": 0.000495800363984393,
      "loss": 6.3035,
      "mean_token_accuracy": 0.13958917260169984,
      "num_tokens": 14498572.0,
      "step": 6695
    },
    {
      "entropy": 6.469868612289429,
      "epoch": 0.7169993044036599,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004957930082806878,
      "loss": 6.339,
      "mean_token_accuracy": 0.1359809160232544,
      "num_tokens": 14509673.0,
      "step": 6700
    },
    {
      "entropy": 6.560486268997193,
      "epoch": 0.7175343785114239,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957856462016155,
      "loss": 6.4157,
      "mean_token_accuracy": 0.12751279547810554,
      "num_tokens": 14519974.0,
      "step": 6705
    },
    {
      "entropy": 6.577579021453857,
      "epoch": 0.7180694526191878,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004957782777473888,
      "loss": 6.3455,
      "mean_token_accuracy": 0.12905636057257652,
      "num_tokens": 14531194.0,
      "step": 6710
    },
    {
      "entropy": 6.547890329360962,
      "epoch": 0.7186045267269516,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004957709029182203,
      "loss": 6.2847,
      "mean_token_accuracy": 0.13680249452590942,
      "num_tokens": 14542393.0,
      "step": 6715
    },
    {
      "entropy": 6.480940198898315,
      "epoch": 0.7191396008347156,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0004957635217143231,
      "loss": 6.2978,
      "mean_token_accuracy": 0.14570009410381318,
      "num_tokens": 14552934.0,
      "step": 6720
    },
    {
      "entropy": 6.516141891479492,
      "epoch": 0.7196746749424795,
      "grad_norm": 0.83984375,
      "learning_rate": 0.00049575613413591,
      "loss": 6.3065,
      "mean_token_accuracy": 0.13669617399573325,
      "num_tokens": 14563643.0,
      "step": 6725
    },
    {
      "entropy": 6.5035957336425785,
      "epoch": 0.7202097490502435,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004957487401831947,
      "loss": 6.2305,
      "mean_token_accuracy": 0.14224187284708023,
      "num_tokens": 14574422.0,
      "step": 6730
    },
    {
      "entropy": 6.463456058502198,
      "epoch": 0.7207448231580074,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004957413398563906,
      "loss": 6.2275,
      "mean_token_accuracy": 0.1363912045955658,
      "num_tokens": 14585002.0,
      "step": 6735
    },
    {
      "entropy": 6.5203992366790775,
      "epoch": 0.7212798972657714,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004957339331557113,
      "loss": 6.2579,
      "mean_token_accuracy": 0.13723741695284844,
      "num_tokens": 14595172.0,
      "step": 6740
    },
    {
      "entropy": 6.494306182861328,
      "epoch": 0.7218149713735352,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004957265200813706,
      "loss": 6.377,
      "mean_token_accuracy": 0.13376473039388656,
      "num_tokens": 14606316.0,
      "step": 6745
    },
    {
      "entropy": 6.549130630493164,
      "epoch": 0.7223500454812991,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004957191006335827,
      "loss": 6.2721,
      "mean_token_accuracy": 0.1368239663541317,
      "num_tokens": 14616394.0,
      "step": 6750
    },
    {
      "entropy": 6.474317359924316,
      "epoch": 0.7228851195890631,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004957116748125618,
      "loss": 6.2913,
      "mean_token_accuracy": 0.13458259999752045,
      "num_tokens": 14627474.0,
      "step": 6755
    },
    {
      "entropy": 6.475166845321655,
      "epoch": 0.723420193696827,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004957042426185223,
      "loss": 6.2185,
      "mean_token_accuracy": 0.14519866183400154,
      "num_tokens": 14638343.0,
      "step": 6760
    },
    {
      "entropy": 6.481177186965942,
      "epoch": 0.723955267804591,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004956968040516789,
      "loss": 6.2988,
      "mean_token_accuracy": 0.13848926946520806,
      "num_tokens": 14649115.0,
      "step": 6765
    },
    {
      "entropy": 6.493714237213135,
      "epoch": 0.7244903419123548,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004956893591122461,
      "loss": 6.3486,
      "mean_token_accuracy": 0.13865093067288398,
      "num_tokens": 14659714.0,
      "step": 6770
    },
    {
      "entropy": 6.630873727798462,
      "epoch": 0.7250254160201188,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004956819078004392,
      "loss": 6.4034,
      "mean_token_accuracy": 0.13012188225984572,
      "num_tokens": 14670481.0,
      "step": 6775
    },
    {
      "entropy": 6.4983642578125,
      "epoch": 0.7255604901278827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004956744501164732,
      "loss": 6.2132,
      "mean_token_accuracy": 0.14504143223166466,
      "num_tokens": 14680816.0,
      "step": 6780
    },
    {
      "entropy": 6.473732042312622,
      "epoch": 0.7260955642356466,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004956669860605634,
      "loss": 6.3109,
      "mean_token_accuracy": 0.13706372082233428,
      "num_tokens": 14692723.0,
      "step": 6785
    },
    {
      "entropy": 6.536991310119629,
      "epoch": 0.7266306383434106,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004956595156329254,
      "loss": 6.3442,
      "mean_token_accuracy": 0.13278391361236572,
      "num_tokens": 14703824.0,
      "step": 6790
    },
    {
      "entropy": 6.548037528991699,
      "epoch": 0.7271657124511745,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000495652038833775,
      "loss": 6.3036,
      "mean_token_accuracy": 0.14187096431851387,
      "num_tokens": 14715934.0,
      "step": 6795
    },
    {
      "entropy": 6.484328985214233,
      "epoch": 0.7277007865589384,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004956445556633279,
      "loss": 6.3026,
      "mean_token_accuracy": 0.1362330436706543,
      "num_tokens": 14725617.0,
      "step": 6800
    },
    {
      "entropy": 6.548808860778808,
      "epoch": 0.7282358606667023,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004956370661218003,
      "loss": 6.2984,
      "mean_token_accuracy": 0.1377137780189514,
      "num_tokens": 14736430.0,
      "step": 6805
    },
    {
      "entropy": 6.4918732166290285,
      "epoch": 0.7287709347744663,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004956295702094084,
      "loss": 6.3021,
      "mean_token_accuracy": 0.14272141754627227,
      "num_tokens": 14746740.0,
      "step": 6810
    },
    {
      "entropy": 6.435430526733398,
      "epoch": 0.7293060088822302,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004956220679263687,
      "loss": 6.2465,
      "mean_token_accuracy": 0.14273511692881585,
      "num_tokens": 14756884.0,
      "step": 6815
    },
    {
      "entropy": 6.491763591766357,
      "epoch": 0.7298410829899941,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004956145592728976,
      "loss": 6.2684,
      "mean_token_accuracy": 0.1443004846572876,
      "num_tokens": 14766974.0,
      "step": 6820
    },
    {
      "entropy": 6.512017059326172,
      "epoch": 0.730376157097758,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004956070442492123,
      "loss": 6.3979,
      "mean_token_accuracy": 0.13255006521940232,
      "num_tokens": 14778449.0,
      "step": 6825
    },
    {
      "entropy": 6.544721984863282,
      "epoch": 0.7309112312055219,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004955995228555296,
      "loss": 6.3548,
      "mean_token_accuracy": 0.1371677577495575,
      "num_tokens": 14788989.0,
      "step": 6830
    },
    {
      "entropy": 6.485946607589722,
      "epoch": 0.7314463053132859,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955919950920666,
      "loss": 6.3078,
      "mean_token_accuracy": 0.14651093482971192,
      "num_tokens": 14798454.0,
      "step": 6835
    },
    {
      "entropy": 6.537364959716797,
      "epoch": 0.7319813794210498,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955844609590408,
      "loss": 6.2658,
      "mean_token_accuracy": 0.1357029877603054,
      "num_tokens": 14808920.0,
      "step": 6840
    },
    {
      "entropy": 6.457472896575927,
      "epoch": 0.7325164535288138,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004955769204566696,
      "loss": 6.2406,
      "mean_token_accuracy": 0.14403316751122475,
      "num_tokens": 14819250.0,
      "step": 6845
    },
    {
      "entropy": 6.523740148544311,
      "epoch": 0.7330515276365777,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004955693735851709,
      "loss": 6.3091,
      "mean_token_accuracy": 0.13773782774806023,
      "num_tokens": 14829789.0,
      "step": 6850
    },
    {
      "entropy": 6.4774055004119875,
      "epoch": 0.7335866017443415,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004955618203447625,
      "loss": 6.3008,
      "mean_token_accuracy": 0.1299881212413311,
      "num_tokens": 14840527.0,
      "step": 6855
    },
    {
      "entropy": 6.492670202255249,
      "epoch": 0.7341216758521055,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004955542607356625,
      "loss": 6.3558,
      "mean_token_accuracy": 0.13976651281118393,
      "num_tokens": 14850922.0,
      "step": 6860
    },
    {
      "entropy": 6.528111457824707,
      "epoch": 0.7346567499598694,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004955466947580893,
      "loss": 6.2987,
      "mean_token_accuracy": 0.13667705580592154,
      "num_tokens": 14862226.0,
      "step": 6865
    },
    {
      "entropy": 6.494947338104248,
      "epoch": 0.7351918240676334,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004955391224122611,
      "loss": 6.2078,
      "mean_token_accuracy": 0.14821575954556465,
      "num_tokens": 14872461.0,
      "step": 6870
    },
    {
      "entropy": 6.470346212387085,
      "epoch": 0.7357268981753973,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000495531543698397,
      "loss": 6.3264,
      "mean_token_accuracy": 0.1390345349907875,
      "num_tokens": 14883088.0,
      "step": 6875
    },
    {
      "entropy": 6.571719551086426,
      "epoch": 0.7362619722831613,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004955239586167153,
      "loss": 6.3397,
      "mean_token_accuracy": 0.1373453587293625,
      "num_tokens": 14894055.0,
      "step": 6880
    },
    {
      "entropy": 6.488383483886719,
      "epoch": 0.7367970463909251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004955163671674354,
      "loss": 6.3135,
      "mean_token_accuracy": 0.1341870181262493,
      "num_tokens": 14904760.0,
      "step": 6885
    },
    {
      "entropy": 6.571750116348267,
      "epoch": 0.737332120498689,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004955087693507764,
      "loss": 6.4294,
      "mean_token_accuracy": 0.13335092067718507,
      "num_tokens": 14915983.0,
      "step": 6890
    },
    {
      "entropy": 6.5187903881073,
      "epoch": 0.737867194606453,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004955011651669577,
      "loss": 6.3275,
      "mean_token_accuracy": 0.13620130643248557,
      "num_tokens": 14925959.0,
      "step": 6895
    },
    {
      "entropy": 6.510292673110962,
      "epoch": 0.7384022687142169,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004954935546161988,
      "loss": 6.267,
      "mean_token_accuracy": 0.1400763660669327,
      "num_tokens": 14936138.0,
      "step": 6900
    },
    {
      "entropy": 6.588149738311768,
      "epoch": 0.7389373428219809,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004954859376987195,
      "loss": 6.2944,
      "mean_token_accuracy": 0.14030833616852761,
      "num_tokens": 14946979.0,
      "step": 6905
    },
    {
      "entropy": 6.4428019523620605,
      "epoch": 0.7394724169297447,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004954783144147397,
      "loss": 6.3625,
      "mean_token_accuracy": 0.13160200417041779,
      "num_tokens": 14958324.0,
      "step": 6910
    },
    {
      "entropy": 6.572534799575806,
      "epoch": 0.7400074910375087,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004954706847644796,
      "loss": 6.3568,
      "mean_token_accuracy": 0.1344327799975872,
      "num_tokens": 14969048.0,
      "step": 6915
    },
    {
      "entropy": 6.569218111038208,
      "epoch": 0.7405425651452726,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954630487481594,
      "loss": 6.3444,
      "mean_token_accuracy": 0.13555286452174187,
      "num_tokens": 14979825.0,
      "step": 6920
    },
    {
      "entropy": 6.464891624450684,
      "epoch": 0.7410776392530365,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004954554063659998,
      "loss": 6.2767,
      "mean_token_accuracy": 0.13890406489372253,
      "num_tokens": 14990138.0,
      "step": 6925
    },
    {
      "entropy": 6.475687170028687,
      "epoch": 0.7416127133608005,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004954477576182212,
      "loss": 6.2233,
      "mean_token_accuracy": 0.14126780778169631,
      "num_tokens": 15002276.0,
      "step": 6930
    },
    {
      "entropy": 6.53695969581604,
      "epoch": 0.7421477874685644,
      "grad_norm": 0.875,
      "learning_rate": 0.0004954401025050445,
      "loss": 6.3723,
      "mean_token_accuracy": 0.1353396236896515,
      "num_tokens": 15013622.0,
      "step": 6935
    },
    {
      "entropy": 6.512791585922241,
      "epoch": 0.7426828615763283,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954324410266909,
      "loss": 6.3342,
      "mean_token_accuracy": 0.1394040659070015,
      "num_tokens": 15024067.0,
      "step": 6940
    },
    {
      "entropy": 6.527772712707519,
      "epoch": 0.7432179356840922,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004954247731833815,
      "loss": 6.212,
      "mean_token_accuracy": 0.14988886192440987,
      "num_tokens": 15033646.0,
      "step": 6945
    },
    {
      "entropy": 6.422814083099365,
      "epoch": 0.7437530097918562,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004954170989753377,
      "loss": 6.2725,
      "mean_token_accuracy": 0.13537086322903633,
      "num_tokens": 15044655.0,
      "step": 6950
    },
    {
      "entropy": 6.562718105316162,
      "epoch": 0.7442880838996201,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004954094184027813,
      "loss": 6.3271,
      "mean_token_accuracy": 0.13692593872547149,
      "num_tokens": 15055492.0,
      "step": 6955
    },
    {
      "entropy": 6.535667085647583,
      "epoch": 0.744823158007384,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004954017314659339,
      "loss": 6.3194,
      "mean_token_accuracy": 0.1304670549929142,
      "num_tokens": 15066472.0,
      "step": 6960
    },
    {
      "entropy": 6.524055433273316,
      "epoch": 0.745358232115148,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004953940381650174,
      "loss": 6.2694,
      "mean_token_accuracy": 0.14626826345920563,
      "num_tokens": 15076851.0,
      "step": 6965
    },
    {
      "entropy": 6.380783891677856,
      "epoch": 0.7458933062229118,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004953863385002541,
      "loss": 6.2059,
      "mean_token_accuracy": 0.14161457493901253,
      "num_tokens": 15086997.0,
      "step": 6970
    },
    {
      "entropy": 6.487274694442749,
      "epoch": 0.7464283803306758,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004953786324718661,
      "loss": 6.3764,
      "mean_token_accuracy": 0.13502368032932283,
      "num_tokens": 15098625.0,
      "step": 6975
    },
    {
      "entropy": 6.480925273895264,
      "epoch": 0.7469634544384397,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953709200800761,
      "loss": 6.192,
      "mean_token_accuracy": 0.14741757214069368,
      "num_tokens": 15108265.0,
      "step": 6980
    },
    {
      "entropy": 6.521458864212036,
      "epoch": 0.7474985285462037,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004953632013251068,
      "loss": 6.2952,
      "mean_token_accuracy": 0.13556876108050347,
      "num_tokens": 15118794.0,
      "step": 6985
    },
    {
      "entropy": 6.51433048248291,
      "epoch": 0.7480336026539676,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004953554762071811,
      "loss": 6.3227,
      "mean_token_accuracy": 0.13437229096889497,
      "num_tokens": 15130084.0,
      "step": 6990
    },
    {
      "entropy": 6.459391403198242,
      "epoch": 0.7485686767617316,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004953477447265218,
      "loss": 6.313,
      "mean_token_accuracy": 0.1361713796854019,
      "num_tokens": 15142193.0,
      "step": 6995
    },
    {
      "entropy": 6.561881446838379,
      "epoch": 0.7491037508694954,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004953400068833524,
      "loss": 6.4312,
      "mean_token_accuracy": 0.13202383518218994,
      "num_tokens": 15152817.0,
      "step": 7000
    },
    {
      "entropy": 6.437725496292114,
      "epoch": 0.7496388249772593,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004953322626778964,
      "loss": 6.0928,
      "mean_token_accuracy": 0.1460914634168148,
      "num_tokens": 15162352.0,
      "step": 7005
    },
    {
      "entropy": 6.522571277618408,
      "epoch": 0.7501738990850233,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004953245121103771,
      "loss": 6.2814,
      "mean_token_accuracy": 0.1433742970228195,
      "num_tokens": 15173091.0,
      "step": 7010
    },
    {
      "entropy": 6.577470827102661,
      "epoch": 0.7507089731927872,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004953167551810185,
      "loss": 6.4482,
      "mean_token_accuracy": 0.12787895053625106,
      "num_tokens": 15184422.0,
      "step": 7015
    },
    {
      "entropy": 6.445052719116211,
      "epoch": 0.7512440473005512,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004953089918900447,
      "loss": 6.1741,
      "mean_token_accuracy": 0.148541921377182,
      "num_tokens": 15194765.0,
      "step": 7020
    },
    {
      "entropy": 6.526875019073486,
      "epoch": 0.751779121408315,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004953012222376795,
      "loss": 6.3365,
      "mean_token_accuracy": 0.13461314439773558,
      "num_tokens": 15206007.0,
      "step": 7025
    },
    {
      "entropy": 6.518378210067749,
      "epoch": 0.752314195516079,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004952934462241476,
      "loss": 6.2863,
      "mean_token_accuracy": 0.13811305165290833,
      "num_tokens": 15216634.0,
      "step": 7030
    },
    {
      "entropy": 6.502454376220703,
      "epoch": 0.7528492696238429,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004952856638496734,
      "loss": 6.346,
      "mean_token_accuracy": 0.1339038170874119,
      "num_tokens": 15226931.0,
      "step": 7035
    },
    {
      "entropy": 6.485996055603027,
      "epoch": 0.7533843437316068,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004952778751144817,
      "loss": 6.3191,
      "mean_token_accuracy": 0.13655798137187958,
      "num_tokens": 15239034.0,
      "step": 7040
    },
    {
      "entropy": 6.558846187591553,
      "epoch": 0.7539194178393708,
      "grad_norm": 0.875,
      "learning_rate": 0.0004952700800187971,
      "loss": 6.3061,
      "mean_token_accuracy": 0.13760880902409553,
      "num_tokens": 15250088.0,
      "step": 7045
    },
    {
      "entropy": 6.536484813690185,
      "epoch": 0.7544544919471347,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000495262278562845,
      "loss": 6.2541,
      "mean_token_accuracy": 0.1410465881228447,
      "num_tokens": 15261150.0,
      "step": 7050
    },
    {
      "entropy": 6.504215478897095,
      "epoch": 0.7549895660548986,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004952544707468506,
      "loss": 6.3425,
      "mean_token_accuracy": 0.13632960692048074,
      "num_tokens": 15270865.0,
      "step": 7055
    },
    {
      "entropy": 6.520873260498047,
      "epoch": 0.7555246401626625,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952466565710391,
      "loss": 6.3659,
      "mean_token_accuracy": 0.13299903199076651,
      "num_tokens": 15282809.0,
      "step": 7060
    },
    {
      "entropy": 6.5031005859375,
      "epoch": 0.7560597142704265,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004952388360356366,
      "loss": 6.309,
      "mean_token_accuracy": 0.14305069372057916,
      "num_tokens": 15293166.0,
      "step": 7065
    },
    {
      "entropy": 6.478386497497558,
      "epoch": 0.7565947883781904,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004952310091408685,
      "loss": 6.2773,
      "mean_token_accuracy": 0.14135212823748589,
      "num_tokens": 15302985.0,
      "step": 7070
    },
    {
      "entropy": 6.526683950424195,
      "epoch": 0.7571298624859543,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004952231758869611,
      "loss": 6.3361,
      "mean_token_accuracy": 0.14280917197465898,
      "num_tokens": 15314737.0,
      "step": 7075
    },
    {
      "entropy": 6.488175630569458,
      "epoch": 0.7576649365937183,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004952153362741403,
      "loss": 6.2683,
      "mean_token_accuracy": 0.14548469185829163,
      "num_tokens": 15325441.0,
      "step": 7080
    },
    {
      "entropy": 6.470507526397705,
      "epoch": 0.7582000107014821,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004952074903026327,
      "loss": 6.1998,
      "mean_token_accuracy": 0.14595614969730378,
      "num_tokens": 15334939.0,
      "step": 7085
    },
    {
      "entropy": 6.497086811065674,
      "epoch": 0.7587350848092461,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004951996379726647,
      "loss": 6.3673,
      "mean_token_accuracy": 0.13187039270997047,
      "num_tokens": 15345655.0,
      "step": 7090
    },
    {
      "entropy": 6.433161306381225,
      "epoch": 0.75927015891701,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004951917792844631,
      "loss": 6.1866,
      "mean_token_accuracy": 0.1500830665230751,
      "num_tokens": 15356035.0,
      "step": 7095
    },
    {
      "entropy": 6.509328460693359,
      "epoch": 0.759805233024774,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004951839142382548,
      "loss": 6.2964,
      "mean_token_accuracy": 0.13920380026102067,
      "num_tokens": 15368509.0,
      "step": 7100
    },
    {
      "entropy": 6.496813440322876,
      "epoch": 0.7603403071325379,
      "grad_norm": 0.8515625,
      "learning_rate": 0.000495176042834267,
      "loss": 6.2823,
      "mean_token_accuracy": 0.1411573439836502,
      "num_tokens": 15379662.0,
      "step": 7105
    },
    {
      "entropy": 6.466855239868164,
      "epoch": 0.7608753812403017,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004951681650727268,
      "loss": 6.2743,
      "mean_token_accuracy": 0.13163355365395546,
      "num_tokens": 15389698.0,
      "step": 7110
    },
    {
      "entropy": 6.547094535827637,
      "epoch": 0.7614104553480657,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004951602809538619,
      "loss": 6.3742,
      "mean_token_accuracy": 0.13446253389120102,
      "num_tokens": 15400724.0,
      "step": 7115
    },
    {
      "entropy": 6.632762432098389,
      "epoch": 0.7619455294558296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004951523904778997,
      "loss": 6.3676,
      "mean_token_accuracy": 0.13552323952317238,
      "num_tokens": 15410732.0,
      "step": 7120
    },
    {
      "entropy": 6.4792533874511715,
      "epoch": 0.7624806035635936,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004951444936450682,
      "loss": 6.2567,
      "mean_token_accuracy": 0.1431451417505741,
      "num_tokens": 15420185.0,
      "step": 7125
    },
    {
      "entropy": 6.458725023269653,
      "epoch": 0.7630156776713575,
      "grad_norm": 0.80859375,
      "learning_rate": 0.0004951365904555954,
      "loss": 6.2023,
      "mean_token_accuracy": 0.1460730217397213,
      "num_tokens": 15431904.0,
      "step": 7130
    },
    {
      "entropy": 6.484750652313233,
      "epoch": 0.7635507517791215,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004951286809097094,
      "loss": 6.3222,
      "mean_token_accuracy": 0.13236253559589387,
      "num_tokens": 15442650.0,
      "step": 7135
    },
    {
      "entropy": 6.478479433059692,
      "epoch": 0.7640858258868853,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004951207650076388,
      "loss": 6.2951,
      "mean_token_accuracy": 0.13835720494389533,
      "num_tokens": 15454273.0,
      "step": 7140
    },
    {
      "entropy": 6.5286139965057375,
      "epoch": 0.7646208999946492,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000495112842749612,
      "loss": 6.2692,
      "mean_token_accuracy": 0.14157717898488045,
      "num_tokens": 15465751.0,
      "step": 7145
    },
    {
      "entropy": 6.504359912872315,
      "epoch": 0.7651559741024132,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004951049141358578,
      "loss": 6.3795,
      "mean_token_accuracy": 0.13437190428376197,
      "num_tokens": 15475340.0,
      "step": 7150
    },
    {
      "entropy": 6.570265865325927,
      "epoch": 0.7656910482101771,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004950969791666051,
      "loss": 6.3583,
      "mean_token_accuracy": 0.13302911072969437,
      "num_tokens": 15486246.0,
      "step": 7155
    },
    {
      "entropy": 6.566818809509277,
      "epoch": 0.7662261223179411,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004950890378420831,
      "loss": 6.3382,
      "mean_token_accuracy": 0.1381997674703598,
      "num_tokens": 15497331.0,
      "step": 7160
    },
    {
      "entropy": 6.5047430992126465,
      "epoch": 0.766761196425705,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000495081090162521,
      "loss": 6.2881,
      "mean_token_accuracy": 0.1369309015572071,
      "num_tokens": 15508036.0,
      "step": 7165
    },
    {
      "entropy": 6.557389926910401,
      "epoch": 0.7672962705334689,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004950731361281483,
      "loss": 6.2913,
      "mean_token_accuracy": 0.13860526978969573,
      "num_tokens": 15518381.0,
      "step": 7170
    },
    {
      "entropy": 6.527024841308593,
      "epoch": 0.7678313446412328,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004950651757391948,
      "loss": 6.2942,
      "mean_token_accuracy": 0.13651483356952668,
      "num_tokens": 15528939.0,
      "step": 7175
    },
    {
      "entropy": 6.493138170242309,
      "epoch": 0.7683664187489967,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004950572089958904,
      "loss": 6.2547,
      "mean_token_accuracy": 0.1347493626177311,
      "num_tokens": 15539718.0,
      "step": 7180
    },
    {
      "entropy": 6.5450633525848385,
      "epoch": 0.7689014928567607,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004950492358984648,
      "loss": 6.4417,
      "mean_token_accuracy": 0.12941004782915116,
      "num_tokens": 15550970.0,
      "step": 7185
    },
    {
      "entropy": 6.564019250869751,
      "epoch": 0.7694365669645246,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004950412564471486,
      "loss": 6.2162,
      "mean_token_accuracy": 0.1421010449528694,
      "num_tokens": 15560490.0,
      "step": 7190
    },
    {
      "entropy": 6.430169534683228,
      "epoch": 0.7699716410722885,
      "grad_norm": 0.921875,
      "learning_rate": 0.000495033270642172,
      "loss": 6.2196,
      "mean_token_accuracy": 0.1391077883541584,
      "num_tokens": 15570188.0,
      "step": 7195
    },
    {
      "entropy": 6.453687524795532,
      "epoch": 0.7705067151800524,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004950252784837655,
      "loss": 6.2574,
      "mean_token_accuracy": 0.13299167156219482,
      "num_tokens": 15581446.0,
      "step": 7200
    },
    {
      "entropy": 6.561170434951782,
      "epoch": 0.7710417892878164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004950172799721601,
      "loss": 6.3462,
      "mean_token_accuracy": 0.13516487926244736,
      "num_tokens": 15594362.0,
      "step": 7205
    },
    {
      "entropy": 6.5357495784759525,
      "epoch": 0.7715768633955803,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004950092751075866,
      "loss": 6.2202,
      "mean_token_accuracy": 0.14535379260778428,
      "num_tokens": 15605531.0,
      "step": 7210
    },
    {
      "entropy": 6.469838094711304,
      "epoch": 0.7721119375033442,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004950012638902763,
      "loss": 6.269,
      "mean_token_accuracy": 0.14220933765172958,
      "num_tokens": 15614962.0,
      "step": 7215
    },
    {
      "entropy": 6.4516149997711185,
      "epoch": 0.7726470116111082,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004949932463204603,
      "loss": 6.2946,
      "mean_token_accuracy": 0.13934430181980134,
      "num_tokens": 15625475.0,
      "step": 7220
    },
    {
      "entropy": 6.499975156784058,
      "epoch": 0.773182085718872,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004949852223983703,
      "loss": 6.2849,
      "mean_token_accuracy": 0.13335466906428337,
      "num_tokens": 15637359.0,
      "step": 7225
    },
    {
      "entropy": 6.643149709701538,
      "epoch": 0.773717159826636,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004949771921242379,
      "loss": 6.3373,
      "mean_token_accuracy": 0.1320371888577938,
      "num_tokens": 15648368.0,
      "step": 7230
    },
    {
      "entropy": 6.58695502281189,
      "epoch": 0.7742522339343999,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004949691554982951,
      "loss": 6.2177,
      "mean_token_accuracy": 0.1455841079354286,
      "num_tokens": 15659295.0,
      "step": 7235
    },
    {
      "entropy": 6.4842101573944095,
      "epoch": 0.7747873080421639,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004949611125207737,
      "loss": 6.205,
      "mean_token_accuracy": 0.14328436851501464,
      "num_tokens": 15669975.0,
      "step": 7240
    },
    {
      "entropy": 6.410018348693848,
      "epoch": 0.7753223821499278,
      "grad_norm": 0.953125,
      "learning_rate": 0.000494953063191906,
      "loss": 6.2806,
      "mean_token_accuracy": 0.1361421063542366,
      "num_tokens": 15680343.0,
      "step": 7245
    },
    {
      "entropy": 6.517525291442871,
      "epoch": 0.7758574562576916,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004949450075119247,
      "loss": 6.3599,
      "mean_token_accuracy": 0.1313583545386791,
      "num_tokens": 15692246.0,
      "step": 7250
    },
    {
      "entropy": 6.567987680435181,
      "epoch": 0.7763925303654556,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004949369454810621,
      "loss": 6.2859,
      "mean_token_accuracy": 0.14094367995858192,
      "num_tokens": 15704262.0,
      "step": 7255
    },
    {
      "entropy": 6.46034517288208,
      "epoch": 0.7769276044732195,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004949288770995512,
      "loss": 6.2987,
      "mean_token_accuracy": 0.1368984915316105,
      "num_tokens": 15715078.0,
      "step": 7260
    },
    {
      "entropy": 6.491894006729126,
      "epoch": 0.7774626785809835,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004949208023676249,
      "loss": 6.3364,
      "mean_token_accuracy": 0.13559149354696273,
      "num_tokens": 15726503.0,
      "step": 7265
    },
    {
      "entropy": 6.455145072937012,
      "epoch": 0.7779977526887474,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004949127212855161,
      "loss": 6.1977,
      "mean_token_accuracy": 0.14185231328010559,
      "num_tokens": 15738854.0,
      "step": 7270
    },
    {
      "entropy": 6.532480049133301,
      "epoch": 0.7785328267965114,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004949046338534587,
      "loss": 6.3709,
      "mean_token_accuracy": 0.13298935443162918,
      "num_tokens": 15750836.0,
      "step": 7275
    },
    {
      "entropy": 6.631442880630493,
      "epoch": 0.7790679009042752,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004948965400716857,
      "loss": 6.3906,
      "mean_token_accuracy": 0.13619265481829643,
      "num_tokens": 15762355.0,
      "step": 7280
    },
    {
      "entropy": 6.5851335525512695,
      "epoch": 0.7796029750120391,
      "grad_norm": 0.890625,
      "learning_rate": 0.000494888439940431,
      "loss": 6.3474,
      "mean_token_accuracy": 0.13436082154512405,
      "num_tokens": 15772952.0,
      "step": 7285
    },
    {
      "entropy": 6.4406662940979,
      "epoch": 0.7801380491198031,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004948803334599286,
      "loss": 6.1803,
      "mean_token_accuracy": 0.14825765639543534,
      "num_tokens": 15782997.0,
      "step": 7290
    },
    {
      "entropy": 6.4774942874908445,
      "epoch": 0.780673123227567,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004948722206304123,
      "loss": 6.2942,
      "mean_token_accuracy": 0.13698788061738015,
      "num_tokens": 15794060.0,
      "step": 7295
    },
    {
      "entropy": 6.551709508895874,
      "epoch": 0.781208197335331,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004948641014521167,
      "loss": 6.3483,
      "mean_token_accuracy": 0.13397666439414024,
      "num_tokens": 15804149.0,
      "step": 7300
    },
    {
      "entropy": 6.533293533325195,
      "epoch": 0.7817432714430949,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000494855975925276,
      "loss": 6.3033,
      "mean_token_accuracy": 0.1421580545604229,
      "num_tokens": 15814643.0,
      "step": 7305
    },
    {
      "entropy": 6.514427042007446,
      "epoch": 0.7822783455508588,
      "grad_norm": 1.0,
      "learning_rate": 0.0004948478440501249,
      "loss": 6.2427,
      "mean_token_accuracy": 0.14081592559814454,
      "num_tokens": 15824640.0,
      "step": 7310
    },
    {
      "entropy": 6.438489103317261,
      "epoch": 0.7828134196586227,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004948397058268982,
      "loss": 6.2631,
      "mean_token_accuracy": 0.14093610346317292,
      "num_tokens": 15836160.0,
      "step": 7315
    },
    {
      "entropy": 6.4757349491119385,
      "epoch": 0.7833484937663866,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004948315612558308,
      "loss": 6.3136,
      "mean_token_accuracy": 0.13938710689544678,
      "num_tokens": 15846886.0,
      "step": 7320
    },
    {
      "entropy": 6.5189488410949705,
      "epoch": 0.7838835678741506,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000494823410337158,
      "loss": 6.2338,
      "mean_token_accuracy": 0.14782554805278778,
      "num_tokens": 15857309.0,
      "step": 7325
    },
    {
      "entropy": 6.511603593826294,
      "epoch": 0.7844186419819145,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004948152530711153,
      "loss": 6.3667,
      "mean_token_accuracy": 0.13546707183122636,
      "num_tokens": 15869620.0,
      "step": 7330
    },
    {
      "entropy": 6.480473375320434,
      "epoch": 0.7849537160896785,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004948070894579378,
      "loss": 6.2377,
      "mean_token_accuracy": 0.1438289165496826,
      "num_tokens": 15880915.0,
      "step": 7335
    },
    {
      "entropy": 6.483896780014038,
      "epoch": 0.7854887901974423,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004947989194978616,
      "loss": 6.2679,
      "mean_token_accuracy": 0.14402638673782348,
      "num_tokens": 15891150.0,
      "step": 7340
    },
    {
      "entropy": 6.506072473526001,
      "epoch": 0.7860238643052063,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004947907431911225,
      "loss": 6.2821,
      "mean_token_accuracy": 0.14244200363755227,
      "num_tokens": 15902526.0,
      "step": 7345
    },
    {
      "entropy": 6.445666742324829,
      "epoch": 0.7865589384129702,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004947825605379566,
      "loss": 6.2245,
      "mean_token_accuracy": 0.14417237937450408,
      "num_tokens": 15913425.0,
      "step": 7350
    },
    {
      "entropy": 6.506277179718017,
      "epoch": 0.7870940125207341,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004947743715386,
      "loss": 6.2047,
      "mean_token_accuracy": 0.1404588595032692,
      "num_tokens": 15924324.0,
      "step": 7355
    },
    {
      "entropy": 6.448118591308594,
      "epoch": 0.7876290866284981,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004947661761932894,
      "loss": 6.2429,
      "mean_token_accuracy": 0.1403199277818203,
      "num_tokens": 15935533.0,
      "step": 7360
    },
    {
      "entropy": 6.474745893478394,
      "epoch": 0.7881641607362619,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004947579745022613,
      "loss": 6.2019,
      "mean_token_accuracy": 0.13793543130159377,
      "num_tokens": 15947007.0,
      "step": 7365
    },
    {
      "entropy": 6.482897329330444,
      "epoch": 0.7886992348440259,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004947497664657527,
      "loss": 6.2921,
      "mean_token_accuracy": 0.13485845997929574,
      "num_tokens": 15957405.0,
      "step": 7370
    },
    {
      "entropy": 6.466152811050415,
      "epoch": 0.7892343089517898,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004947415520840003,
      "loss": 6.2382,
      "mean_token_accuracy": 0.14553275927901269,
      "num_tokens": 15969895.0,
      "step": 7375
    },
    {
      "entropy": 6.459381437301635,
      "epoch": 0.7897693830595538,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004947333313572416,
      "loss": 6.1793,
      "mean_token_accuracy": 0.15283698141574859,
      "num_tokens": 15979854.0,
      "step": 7380
    },
    {
      "entropy": 6.411194515228272,
      "epoch": 0.7903044571673177,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004947251042857137,
      "loss": 6.1885,
      "mean_token_accuracy": 0.14348533600568772,
      "num_tokens": 15991341.0,
      "step": 7385
    },
    {
      "entropy": 6.450728845596314,
      "epoch": 0.7908395312750816,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004947168708696544,
      "loss": 6.2289,
      "mean_token_accuracy": 0.14246075004339218,
      "num_tokens": 16001944.0,
      "step": 7390
    },
    {
      "entropy": 6.454413890838623,
      "epoch": 0.7913746053828455,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004947086311093013,
      "loss": 6.2178,
      "mean_token_accuracy": 0.14334433525800705,
      "num_tokens": 16011771.0,
      "step": 7395
    },
    {
      "entropy": 6.520086002349854,
      "epoch": 0.7919096794906094,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004947003850048924,
      "loss": 6.3877,
      "mean_token_accuracy": 0.12969294264912606,
      "num_tokens": 16023901.0,
      "step": 7400
    },
    {
      "entropy": 6.558329677581787,
      "epoch": 0.7924447535983734,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004946921325566656,
      "loss": 6.3008,
      "mean_token_accuracy": 0.13578956350684165,
      "num_tokens": 16035014.0,
      "step": 7405
    },
    {
      "entropy": 6.559424161911011,
      "epoch": 0.7929798277061373,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004946838737648595,
      "loss": 6.3806,
      "mean_token_accuracy": 0.13097170516848564,
      "num_tokens": 16046927.0,
      "step": 7410
    },
    {
      "entropy": 6.52015266418457,
      "epoch": 0.7935149018139013,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004946756086297124,
      "loss": 6.2912,
      "mean_token_accuracy": 0.14482976719737053,
      "num_tokens": 16057341.0,
      "step": 7415
    },
    {
      "entropy": 6.5183099746704105,
      "epoch": 0.7940499759216652,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004946673371514628,
      "loss": 6.2748,
      "mean_token_accuracy": 0.13880268409848212,
      "num_tokens": 16068408.0,
      "step": 7420
    },
    {
      "entropy": 6.438851547241211,
      "epoch": 0.794585050029429,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004946590593303499,
      "loss": 6.2351,
      "mean_token_accuracy": 0.14763287976384162,
      "num_tokens": 16078693.0,
      "step": 7425
    },
    {
      "entropy": 6.442134380340576,
      "epoch": 0.795120124137193,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004946507751666124,
      "loss": 6.2892,
      "mean_token_accuracy": 0.14073902294039725,
      "num_tokens": 16089127.0,
      "step": 7430
    },
    {
      "entropy": 6.42581148147583,
      "epoch": 0.7956551982449569,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004946424846604897,
      "loss": 6.1725,
      "mean_token_accuracy": 0.15482478812336922,
      "num_tokens": 16099726.0,
      "step": 7435
    },
    {
      "entropy": 6.476835060119629,
      "epoch": 0.7961902723527209,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004946341878122212,
      "loss": 6.2498,
      "mean_token_accuracy": 0.14192514270544052,
      "num_tokens": 16111017.0,
      "step": 7440
    },
    {
      "entropy": 6.483833980560303,
      "epoch": 0.7967253464604848,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004946258846220462,
      "loss": 6.3086,
      "mean_token_accuracy": 0.13704771995544435,
      "num_tokens": 16121406.0,
      "step": 7445
    },
    {
      "entropy": 6.468227291107178,
      "epoch": 0.7972604205682488,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004946175750902049,
      "loss": 6.2649,
      "mean_token_accuracy": 0.14069657698273658,
      "num_tokens": 16131587.0,
      "step": 7450
    },
    {
      "entropy": 6.4237853527069095,
      "epoch": 0.7977954946760126,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004946092592169368,
      "loss": 6.2036,
      "mean_token_accuracy": 0.1471470907330513,
      "num_tokens": 16142578.0,
      "step": 7455
    },
    {
      "entropy": 6.50896520614624,
      "epoch": 0.7983305687837765,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004946009370024822,
      "loss": 6.208,
      "mean_token_accuracy": 0.14368785321712493,
      "num_tokens": 16153501.0,
      "step": 7460
    },
    {
      "entropy": 6.466834688186646,
      "epoch": 0.7988656428915405,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004945926084470814,
      "loss": 6.2857,
      "mean_token_accuracy": 0.13894426599144935,
      "num_tokens": 16164273.0,
      "step": 7465
    },
    {
      "entropy": 6.542187261581421,
      "epoch": 0.7994007169993044,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004945842735509749,
      "loss": 6.2918,
      "mean_token_accuracy": 0.1417921483516693,
      "num_tokens": 16175071.0,
      "step": 7470
    },
    {
      "entropy": 6.563093519210815,
      "epoch": 0.7999357911070684,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004945759323144034,
      "loss": 6.3163,
      "mean_token_accuracy": 0.13780677914619446,
      "num_tokens": 16185404.0,
      "step": 7475
    },
    {
      "entropy": 6.476992416381836,
      "epoch": 0.8004708652148322,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004945675847376077,
      "loss": 6.276,
      "mean_token_accuracy": 0.13502865731716157,
      "num_tokens": 16196881.0,
      "step": 7480
    },
    {
      "entropy": 6.546551465988159,
      "epoch": 0.8010059393225962,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004945592308208288,
      "loss": 6.3374,
      "mean_token_accuracy": 0.13504851162433623,
      "num_tokens": 16207534.0,
      "step": 7485
    },
    {
      "entropy": 6.499393558502197,
      "epoch": 0.8015410134303601,
      "grad_norm": 0.953125,
      "learning_rate": 0.000494550870564308,
      "loss": 6.204,
      "mean_token_accuracy": 0.1507472276687622,
      "num_tokens": 16217371.0,
      "step": 7490
    },
    {
      "entropy": 6.49650092124939,
      "epoch": 0.802076087538124,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004945425039682866,
      "loss": 6.2668,
      "mean_token_accuracy": 0.1396873563528061,
      "num_tokens": 16227898.0,
      "step": 7495
    },
    {
      "entropy": 6.4722301959991455,
      "epoch": 0.802611161645888,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004945341310330064,
      "loss": 6.2443,
      "mean_token_accuracy": 0.14260734915733336,
      "num_tokens": 16238966.0,
      "step": 7500
    },
    {
      "entropy": 6.465616369247437,
      "epoch": 0.8031462357536518,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004945257517587089,
      "loss": 6.261,
      "mean_token_accuracy": 0.15234632641077042,
      "num_tokens": 16250129.0,
      "step": 7505
    },
    {
      "entropy": 6.531611108779908,
      "epoch": 0.8036813098614158,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004945173661456361,
      "loss": 6.2977,
      "mean_token_accuracy": 0.1357301101088524,
      "num_tokens": 16260877.0,
      "step": 7510
    },
    {
      "entropy": 6.510767412185669,
      "epoch": 0.8042163839691797,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004945089741940303,
      "loss": 6.2692,
      "mean_token_accuracy": 0.14449671357870103,
      "num_tokens": 16270548.0,
      "step": 7515
    },
    {
      "entropy": 6.4516441345214846,
      "epoch": 0.8047514580769437,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004945005759041338,
      "loss": 6.2777,
      "mean_token_accuracy": 0.1392606034874916,
      "num_tokens": 16281014.0,
      "step": 7520
    },
    {
      "entropy": 6.521145486831665,
      "epoch": 0.8052865321847076,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004944921712761889,
      "loss": 6.3218,
      "mean_token_accuracy": 0.14316845685243607,
      "num_tokens": 16291653.0,
      "step": 7525
    },
    {
      "entropy": 6.515569591522217,
      "epoch": 0.8058216062924715,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004944837603104383,
      "loss": 6.2009,
      "mean_token_accuracy": 0.14263664186000824,
      "num_tokens": 16301703.0,
      "step": 7530
    },
    {
      "entropy": 6.403749990463257,
      "epoch": 0.8063566804002354,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004944753430071252,
      "loss": 6.2495,
      "mean_token_accuracy": 0.1420724242925644,
      "num_tokens": 16312252.0,
      "step": 7535
    },
    {
      "entropy": 6.496534729003907,
      "epoch": 0.8068917545079993,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004944669193664923,
      "loss": 6.208,
      "mean_token_accuracy": 0.13835022747516632,
      "num_tokens": 16324073.0,
      "step": 7540
    },
    {
      "entropy": 6.540986967086792,
      "epoch": 0.8074268286157633,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004944584893887829,
      "loss": 6.2375,
      "mean_token_accuracy": 0.14323000907897948,
      "num_tokens": 16334637.0,
      "step": 7545
    },
    {
      "entropy": 6.4887460231781,
      "epoch": 0.8079619027235272,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004944500530742404,
      "loss": 6.3489,
      "mean_token_accuracy": 0.13313830867409707,
      "num_tokens": 16345581.0,
      "step": 7550
    },
    {
      "entropy": 6.544404220581055,
      "epoch": 0.8084969768312912,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004944416104231086,
      "loss": 6.3414,
      "mean_token_accuracy": 0.13447642773389817,
      "num_tokens": 16356995.0,
      "step": 7555
    },
    {
      "entropy": 6.5167601108551025,
      "epoch": 0.8090320509390551,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004944331614356311,
      "loss": 6.2975,
      "mean_token_accuracy": 0.1369791805744171,
      "num_tokens": 16368841.0,
      "step": 7560
    },
    {
      "entropy": 6.459184885025024,
      "epoch": 0.8095671250468189,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004944247061120519,
      "loss": 6.2208,
      "mean_token_accuracy": 0.13835411071777343,
      "num_tokens": 16379295.0,
      "step": 7565
    },
    {
      "entropy": 6.506670951843262,
      "epoch": 0.8101021991545829,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004944162444526151,
      "loss": 6.2571,
      "mean_token_accuracy": 0.1426844835281372,
      "num_tokens": 16389512.0,
      "step": 7570
    },
    {
      "entropy": 6.395688676834107,
      "epoch": 0.8106372732623468,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004944077764575651,
      "loss": 6.1798,
      "mean_token_accuracy": 0.14822551906108855,
      "num_tokens": 16400720.0,
      "step": 7575
    },
    {
      "entropy": 6.448258018493652,
      "epoch": 0.8111723473701108,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004943993021271463,
      "loss": 6.2379,
      "mean_token_accuracy": 0.14293486773967742,
      "num_tokens": 16411278.0,
      "step": 7580
    },
    {
      "entropy": 6.4964300155639645,
      "epoch": 0.8117074214778747,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004943908214616035,
      "loss": 6.2955,
      "mean_token_accuracy": 0.1392256498336792,
      "num_tokens": 16421665.0,
      "step": 7585
    },
    {
      "entropy": 6.513621187210083,
      "epoch": 0.8122424955856387,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004943823344611818,
      "loss": 6.3402,
      "mean_token_accuracy": 0.13683698028326036,
      "num_tokens": 16432943.0,
      "step": 7590
    },
    {
      "entropy": 6.549855375289917,
      "epoch": 0.8127775696934025,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004943738411261258,
      "loss": 6.2327,
      "mean_token_accuracy": 0.14013942778110505,
      "num_tokens": 16444506.0,
      "step": 7595
    },
    {
      "entropy": 6.475243091583252,
      "epoch": 0.8133126438011664,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004943653414566809,
      "loss": 6.2576,
      "mean_token_accuracy": 0.13932344764471055,
      "num_tokens": 16455403.0,
      "step": 7600
    },
    {
      "entropy": 6.567423343658447,
      "epoch": 0.8138477179089304,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004943568354530927,
      "loss": 6.3487,
      "mean_token_accuracy": 0.13028792813420295,
      "num_tokens": 16465616.0,
      "step": 7605
    },
    {
      "entropy": 6.561347150802613,
      "epoch": 0.8143827920166943,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004943483231156068,
      "loss": 6.3302,
      "mean_token_accuracy": 0.13280778825283052,
      "num_tokens": 16476850.0,
      "step": 7610
    },
    {
      "entropy": 6.5337916851043705,
      "epoch": 0.8149178661244583,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004943398044444687,
      "loss": 6.276,
      "mean_token_accuracy": 0.14516093358397483,
      "num_tokens": 16489118.0,
      "step": 7615
    },
    {
      "entropy": 6.516808271408081,
      "epoch": 0.8154529402322221,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004943312794399246,
      "loss": 6.2344,
      "mean_token_accuracy": 0.14010219648480415,
      "num_tokens": 16499336.0,
      "step": 7620
    },
    {
      "entropy": 6.481103897094727,
      "epoch": 0.8159880143399861,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004943227481022207,
      "loss": 6.2849,
      "mean_token_accuracy": 0.13623519912362098,
      "num_tokens": 16510158.0,
      "step": 7625
    },
    {
      "entropy": 6.429189777374267,
      "epoch": 0.81652308844775,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004943142104316033,
      "loss": 6.2389,
      "mean_token_accuracy": 0.13902947455644607,
      "num_tokens": 16521080.0,
      "step": 7630
    },
    {
      "entropy": 6.469725894927978,
      "epoch": 0.8170581625555139,
      "grad_norm": 1.0,
      "learning_rate": 0.0004943056664283189,
      "loss": 6.256,
      "mean_token_accuracy": 0.14541354924440383,
      "num_tokens": 16531072.0,
      "step": 7635
    },
    {
      "entropy": 6.523196601867676,
      "epoch": 0.8175932366632779,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000494297116092614,
      "loss": 6.3556,
      "mean_token_accuracy": 0.13215273022651672,
      "num_tokens": 16542588.0,
      "step": 7640
    },
    {
      "entropy": 6.533563232421875,
      "epoch": 0.8181283107710418,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942885594247359,
      "loss": 6.2192,
      "mean_token_accuracy": 0.1388680048286915,
      "num_tokens": 16553466.0,
      "step": 7645
    },
    {
      "entropy": 6.462963390350342,
      "epoch": 0.8186633848788057,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004942799964249314,
      "loss": 6.2498,
      "mean_token_accuracy": 0.13584778681397439,
      "num_tokens": 16564084.0,
      "step": 7650
    },
    {
      "entropy": 6.413071632385254,
      "epoch": 0.8191984589865696,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004942714270934479,
      "loss": 6.2168,
      "mean_token_accuracy": 0.15287627652287483,
      "num_tokens": 16575665.0,
      "step": 7655
    },
    {
      "entropy": 6.4788531303405765,
      "epoch": 0.8197335330943336,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004942628514305326,
      "loss": 6.2394,
      "mean_token_accuracy": 0.135659921169281,
      "num_tokens": 16586518.0,
      "step": 7660
    },
    {
      "entropy": 6.41478214263916,
      "epoch": 0.8202686072020975,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004942542694364333,
      "loss": 6.2567,
      "mean_token_accuracy": 0.14273435920476912,
      "num_tokens": 16597345.0,
      "step": 7665
    },
    {
      "entropy": 6.482707834243774,
      "epoch": 0.8208036813098614,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004942456811113978,
      "loss": 6.2869,
      "mean_token_accuracy": 0.14190693125128745,
      "num_tokens": 16609586.0,
      "step": 7670
    },
    {
      "entropy": 6.546649169921875,
      "epoch": 0.8213387554176254,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004942370864556741,
      "loss": 6.1482,
      "mean_token_accuracy": 0.14869638085365294,
      "num_tokens": 16620165.0,
      "step": 7675
    },
    {
      "entropy": 6.420868015289306,
      "epoch": 0.8218738295253892,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004942284854695104,
      "loss": 6.254,
      "mean_token_accuracy": 0.14137813225388526,
      "num_tokens": 16631200.0,
      "step": 7680
    },
    {
      "entropy": 6.4423301219940186,
      "epoch": 0.8224089036331532,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004942198781531549,
      "loss": 6.2996,
      "mean_token_accuracy": 0.13697852045297623,
      "num_tokens": 16642553.0,
      "step": 7685
    },
    {
      "entropy": 6.513464164733887,
      "epoch": 0.8229439777409171,
      "grad_norm": 0.875,
      "learning_rate": 0.0004942112645068561,
      "loss": 6.311,
      "mean_token_accuracy": 0.13813328519463539,
      "num_tokens": 16653431.0,
      "step": 7690
    },
    {
      "entropy": 6.487181377410889,
      "epoch": 0.8234790518486811,
      "grad_norm": 0.8359375,
      "learning_rate": 0.000494202644530863,
      "loss": 6.2405,
      "mean_token_accuracy": 0.14050233140587806,
      "num_tokens": 16664880.0,
      "step": 7695
    },
    {
      "entropy": 6.423042058944702,
      "epoch": 0.824014125956445,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004941940182254244,
      "loss": 6.2495,
      "mean_token_accuracy": 0.14572507813572882,
      "num_tokens": 16676224.0,
      "step": 7700
    },
    {
      "entropy": 6.482410526275634,
      "epoch": 0.8245492000642088,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004941853855907892,
      "loss": 6.2969,
      "mean_token_accuracy": 0.13951638638973235,
      "num_tokens": 16687732.0,
      "step": 7705
    },
    {
      "entropy": 6.510223007202148,
      "epoch": 0.8250842741719728,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004941767466272068,
      "loss": 6.2403,
      "mean_token_accuracy": 0.14558646380901336,
      "num_tokens": 16698257.0,
      "step": 7710
    },
    {
      "entropy": 6.490880250930786,
      "epoch": 0.8256193482797367,
      "grad_norm": 0.875,
      "learning_rate": 0.0004941681013349267,
      "loss": 6.2086,
      "mean_token_accuracy": 0.14403536319732665,
      "num_tokens": 16708860.0,
      "step": 7715
    },
    {
      "entropy": 6.5099608421325685,
      "epoch": 0.8261544223875007,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004941594497141985,
      "loss": 6.2156,
      "mean_token_accuracy": 0.14348786622285842,
      "num_tokens": 16719987.0,
      "step": 7720
    },
    {
      "entropy": 6.505326843261718,
      "epoch": 0.8266894964952646,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004941507917652718,
      "loss": 6.3102,
      "mean_token_accuracy": 0.1376115381717682,
      "num_tokens": 16730861.0,
      "step": 7725
    },
    {
      "entropy": 6.42150502204895,
      "epoch": 0.8272245706030286,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004941421274883969,
      "loss": 6.1608,
      "mean_token_accuracy": 0.14996344447135926,
      "num_tokens": 16740968.0,
      "step": 7730
    },
    {
      "entropy": 6.461458396911621,
      "epoch": 0.8277596447107924,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000494133456883824,
      "loss": 6.2661,
      "mean_token_accuracy": 0.14037820547819138,
      "num_tokens": 16750700.0,
      "step": 7735
    },
    {
      "entropy": 6.4964001178741455,
      "epoch": 0.8282947188185563,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004941247799518031,
      "loss": 6.2483,
      "mean_token_accuracy": 0.13828410655260087,
      "num_tokens": 16760719.0,
      "step": 7740
    },
    {
      "entropy": 6.481279420852661,
      "epoch": 0.8288297929263203,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004941160966925851,
      "loss": 6.1388,
      "mean_token_accuracy": 0.14534148275852204,
      "num_tokens": 16771331.0,
      "step": 7745
    },
    {
      "entropy": 6.364374017715454,
      "epoch": 0.8293648670340842,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004941074071064205,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14480406567454338,
      "num_tokens": 16781306.0,
      "step": 7750
    },
    {
      "entropy": 6.483348798751831,
      "epoch": 0.8298999411418482,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004940987111935605,
      "loss": 6.2633,
      "mean_token_accuracy": 0.14158105328679085,
      "num_tokens": 16793013.0,
      "step": 7755
    },
    {
      "entropy": 6.473084163665772,
      "epoch": 0.830435015249612,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940900089542558,
      "loss": 6.251,
      "mean_token_accuracy": 0.14182406961917876,
      "num_tokens": 16803987.0,
      "step": 7760
    },
    {
      "entropy": 6.475079774856567,
      "epoch": 0.830970089357376,
      "grad_norm": 0.890625,
      "learning_rate": 0.000494081300388758,
      "loss": 6.2544,
      "mean_token_accuracy": 0.13191908672451974,
      "num_tokens": 16815031.0,
      "step": 7765
    },
    {
      "entropy": 6.544781827926636,
      "epoch": 0.8315051634651399,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004940725854973184,
      "loss": 6.3045,
      "mean_token_accuracy": 0.1363958887755871,
      "num_tokens": 16826313.0,
      "step": 7770
    },
    {
      "entropy": 6.531337690353394,
      "epoch": 0.8320402375729038,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004940638642801886,
      "loss": 6.2939,
      "mean_token_accuracy": 0.13946133702993393,
      "num_tokens": 16838470.0,
      "step": 7775
    },
    {
      "entropy": 6.44612717628479,
      "epoch": 0.8325753116806678,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004940551367376205,
      "loss": 6.1915,
      "mean_token_accuracy": 0.1401160478591919,
      "num_tokens": 16849266.0,
      "step": 7780
    },
    {
      "entropy": 6.450345659255982,
      "epoch": 0.8331103857884317,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004940464028698662,
      "loss": 6.2943,
      "mean_token_accuracy": 0.1405651241540909,
      "num_tokens": 16860720.0,
      "step": 7785
    },
    {
      "entropy": 6.456397151947021,
      "epoch": 0.8336454598961957,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004940376626771779,
      "loss": 6.2642,
      "mean_token_accuracy": 0.1413808658719063,
      "num_tokens": 16872004.0,
      "step": 7790
    },
    {
      "entropy": 6.518138599395752,
      "epoch": 0.8341805340039595,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004940289161598076,
      "loss": 6.2874,
      "mean_token_accuracy": 0.1319346934556961,
      "num_tokens": 16882898.0,
      "step": 7795
    },
    {
      "entropy": 6.523390483856201,
      "epoch": 0.8347156081117235,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004940201633180084,
      "loss": 6.2289,
      "mean_token_accuracy": 0.12887770161032677,
      "num_tokens": 16894331.0,
      "step": 7800
    },
    {
      "entropy": 6.580140399932861,
      "epoch": 0.8352506822194874,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004940114041520326,
      "loss": 6.3364,
      "mean_token_accuracy": 0.13916995525360107,
      "num_tokens": 16903656.0,
      "step": 7805
    },
    {
      "entropy": 6.45752329826355,
      "epoch": 0.8357857563272513,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004940026386621333,
      "loss": 6.2934,
      "mean_token_accuracy": 0.1385620042681694,
      "num_tokens": 16915878.0,
      "step": 7810
    },
    {
      "entropy": 6.562590312957764,
      "epoch": 0.8363208304350153,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004939938668485636,
      "loss": 6.3424,
      "mean_token_accuracy": 0.13664351105690004,
      "num_tokens": 16926577.0,
      "step": 7815
    },
    {
      "entropy": 6.488303852081299,
      "epoch": 0.8368559045427791,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004939850887115768,
      "loss": 6.1861,
      "mean_token_accuracy": 0.1507130466401577,
      "num_tokens": 16938012.0,
      "step": 7820
    },
    {
      "entropy": 6.470408582687378,
      "epoch": 0.8373909786505431,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004939763042514263,
      "loss": 6.1905,
      "mean_token_accuracy": 0.1373836576938629,
      "num_tokens": 16949007.0,
      "step": 7825
    },
    {
      "entropy": 6.3479162693023685,
      "epoch": 0.837926052758307,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004939675134683658,
      "loss": 6.161,
      "mean_token_accuracy": 0.15042434334754945,
      "num_tokens": 16959425.0,
      "step": 7830
    },
    {
      "entropy": 6.506083917617798,
      "epoch": 0.838461126866071,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004939587163626491,
      "loss": 6.2723,
      "mean_token_accuracy": 0.13412620574235917,
      "num_tokens": 16970002.0,
      "step": 7835
    },
    {
      "entropy": 6.482392454147339,
      "epoch": 0.8389962009738349,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004939499129345302,
      "loss": 6.2986,
      "mean_token_accuracy": 0.137315384298563,
      "num_tokens": 16980739.0,
      "step": 7840
    },
    {
      "entropy": 6.4975001335144045,
      "epoch": 0.8395312750815987,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004939411031842633,
      "loss": 6.2563,
      "mean_token_accuracy": 0.14315761104226113,
      "num_tokens": 16991643.0,
      "step": 7845
    },
    {
      "entropy": 6.358150863647461,
      "epoch": 0.8400663491893627,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000493932287112103,
      "loss": 6.1031,
      "mean_token_accuracy": 0.15902615636587142,
      "num_tokens": 17001975.0,
      "step": 7850
    },
    {
      "entropy": 6.524006414413452,
      "epoch": 0.8406014232971266,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004939234647183035,
      "loss": 6.236,
      "mean_token_accuracy": 0.14151140302419662,
      "num_tokens": 17012658.0,
      "step": 7855
    },
    {
      "entropy": 6.500154781341553,
      "epoch": 0.8411364974048906,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004939146360031197,
      "loss": 6.266,
      "mean_token_accuracy": 0.13744446337223054,
      "num_tokens": 17023196.0,
      "step": 7860
    },
    {
      "entropy": 6.466902256011963,
      "epoch": 0.8416715715126545,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004939058009668067,
      "loss": 6.2894,
      "mean_token_accuracy": 0.13646864593029023,
      "num_tokens": 17034589.0,
      "step": 7865
    },
    {
      "entropy": 6.53296594619751,
      "epoch": 0.8422066456204185,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004938969596096194,
      "loss": 6.2518,
      "mean_token_accuracy": 0.14518792033195496,
      "num_tokens": 17046139.0,
      "step": 7870
    },
    {
      "entropy": 6.460026597976684,
      "epoch": 0.8427417197281823,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000493888111931813,
      "loss": 6.2687,
      "mean_token_accuracy": 0.1366637207567692,
      "num_tokens": 17056926.0,
      "step": 7875
    },
    {
      "entropy": 6.364657974243164,
      "epoch": 0.8432767938359462,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004938792579336433,
      "loss": 6.212,
      "mean_token_accuracy": 0.14265245869755744,
      "num_tokens": 17067339.0,
      "step": 7880
    },
    {
      "entropy": 6.538064432144165,
      "epoch": 0.8438118679437102,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004938703976153657,
      "loss": 6.232,
      "mean_token_accuracy": 0.13975303694605828,
      "num_tokens": 17078037.0,
      "step": 7885
    },
    {
      "entropy": 6.5085241317749025,
      "epoch": 0.8443469420514741,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004938615309772362,
      "loss": 6.2712,
      "mean_token_accuracy": 0.1315981440246105,
      "num_tokens": 17089069.0,
      "step": 7890
    },
    {
      "entropy": 6.457151651382446,
      "epoch": 0.8448820161592381,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004938526580195107,
      "loss": 6.2399,
      "mean_token_accuracy": 0.14801932424306868,
      "num_tokens": 17100006.0,
      "step": 7895
    },
    {
      "entropy": 6.516610717773437,
      "epoch": 0.845417090267002,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004938437787424454,
      "loss": 6.252,
      "mean_token_accuracy": 0.1433919169008732,
      "num_tokens": 17111896.0,
      "step": 7900
    },
    {
      "entropy": 6.469233131408691,
      "epoch": 0.845952164374766,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004938348931462969,
      "loss": 6.2477,
      "mean_token_accuracy": 0.1522398389875889,
      "num_tokens": 17123202.0,
      "step": 7905
    },
    {
      "entropy": 6.4610546112060545,
      "epoch": 0.8464872384825298,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004938260012313215,
      "loss": 6.2122,
      "mean_token_accuracy": 0.14675726890563964,
      "num_tokens": 17134013.0,
      "step": 7910
    },
    {
      "entropy": 6.401585674285888,
      "epoch": 0.8470223125902937,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004938171029977761,
      "loss": 6.2183,
      "mean_token_accuracy": 0.14334620982408525,
      "num_tokens": 17144041.0,
      "step": 7915
    },
    {
      "entropy": 6.4813121318817135,
      "epoch": 0.8475573866980577,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004938081984459176,
      "loss": 6.2195,
      "mean_token_accuracy": 0.14281805008649825,
      "num_tokens": 17155334.0,
      "step": 7920
    },
    {
      "entropy": 6.465215826034546,
      "epoch": 0.8480924608058216,
      "grad_norm": 0.953125,
      "learning_rate": 0.000493799287576003,
      "loss": 6.1704,
      "mean_token_accuracy": 0.14565350413322448,
      "num_tokens": 17166243.0,
      "step": 7925
    },
    {
      "entropy": 6.44447431564331,
      "epoch": 0.8486275349135856,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004937903703882898,
      "loss": 6.2449,
      "mean_token_accuracy": 0.13910187482833863,
      "num_tokens": 17177709.0,
      "step": 7930
    },
    {
      "entropy": 6.437526512145996,
      "epoch": 0.8491626090213494,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937814468830353,
      "loss": 6.2173,
      "mean_token_accuracy": 0.13859330490231514,
      "num_tokens": 17188275.0,
      "step": 7935
    },
    {
      "entropy": 6.508519554138184,
      "epoch": 0.8496976831291134,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004937725170604975,
      "loss": 6.336,
      "mean_token_accuracy": 0.1381486490368843,
      "num_tokens": 17199527.0,
      "step": 7940
    },
    {
      "entropy": 6.510644769668579,
      "epoch": 0.8502327572368773,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004937635809209339,
      "loss": 6.2432,
      "mean_token_accuracy": 0.1443048372864723,
      "num_tokens": 17209500.0,
      "step": 7945
    },
    {
      "entropy": 6.451434659957886,
      "epoch": 0.8507678313446412,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004937546384646027,
      "loss": 6.2004,
      "mean_token_accuracy": 0.1390951469540596,
      "num_tokens": 17219589.0,
      "step": 7950
    },
    {
      "entropy": 6.546543407440185,
      "epoch": 0.8513029054524052,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004937456896917619,
      "loss": 6.2371,
      "mean_token_accuracy": 0.13986699059605598,
      "num_tokens": 17230508.0,
      "step": 7955
    },
    {
      "entropy": 6.432592868804932,
      "epoch": 0.851837979560169,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004937367346026702,
      "loss": 6.2493,
      "mean_token_accuracy": 0.14182270914316178,
      "num_tokens": 17240886.0,
      "step": 7960
    },
    {
      "entropy": 6.409507751464844,
      "epoch": 0.852373053667933,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004937277731975858,
      "loss": 6.2577,
      "mean_token_accuracy": 0.14627861008048057,
      "num_tokens": 17252247.0,
      "step": 7965
    },
    {
      "entropy": 6.531063508987427,
      "epoch": 0.8529081277756969,
      "grad_norm": 0.84765625,
      "learning_rate": 0.0004937188054767679,
      "loss": 6.2349,
      "mean_token_accuracy": 0.1466575562953949,
      "num_tokens": 17262845.0,
      "step": 7970
    },
    {
      "entropy": 6.495129871368408,
      "epoch": 0.8534432018834609,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004937098314404751,
      "loss": 6.2361,
      "mean_token_accuracy": 0.14176514744758606,
      "num_tokens": 17274247.0,
      "step": 7975
    },
    {
      "entropy": 6.48998498916626,
      "epoch": 0.8539782759912248,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004937008510889668,
      "loss": 6.2478,
      "mean_token_accuracy": 0.14324330165982246,
      "num_tokens": 17285471.0,
      "step": 7980
    },
    {
      "entropy": 6.500499153137207,
      "epoch": 0.8545133500989887,
      "grad_norm": 1.0,
      "learning_rate": 0.0004936918644225021,
      "loss": 6.2902,
      "mean_token_accuracy": 0.14687259569764138,
      "num_tokens": 17296243.0,
      "step": 7985
    },
    {
      "entropy": 6.475070428848267,
      "epoch": 0.8550484242067526,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004936828714413405,
      "loss": 6.2357,
      "mean_token_accuracy": 0.1498035214841366,
      "num_tokens": 17307301.0,
      "step": 7990
    },
    {
      "entropy": 6.383671283721924,
      "epoch": 0.8555834983145165,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004936738721457417,
      "loss": 6.1702,
      "mean_token_accuracy": 0.1501430407166481,
      "num_tokens": 17318550.0,
      "step": 7995
    },
    {
      "entropy": 6.492444467544556,
      "epoch": 0.8561185724222805,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004936648665359657,
      "loss": 6.2398,
      "mean_token_accuracy": 0.1399112895131111,
      "num_tokens": 17328577.0,
      "step": 8000
    },
    {
      "entropy": 6.535345792770386,
      "epoch": 0.8566536465300444,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004936558546122723,
      "loss": 6.2893,
      "mean_token_accuracy": 0.13243073150515555,
      "num_tokens": 17341013.0,
      "step": 8005
    },
    {
      "entropy": 6.432257556915284,
      "epoch": 0.8571887206378084,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004936468363749217,
      "loss": 6.2318,
      "mean_token_accuracy": 0.1384417898952961,
      "num_tokens": 17351956.0,
      "step": 8010
    },
    {
      "entropy": 6.519441795349121,
      "epoch": 0.8577237947455723,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004936378118241746,
      "loss": 6.2193,
      "mean_token_accuracy": 0.13666695952415467,
      "num_tokens": 17363465.0,
      "step": 8015
    },
    {
      "entropy": 6.458926010131836,
      "epoch": 0.8582588688533361,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004936287809602913,
      "loss": 6.2675,
      "mean_token_accuracy": 0.14383048564195633,
      "num_tokens": 17374458.0,
      "step": 8020
    },
    {
      "entropy": 6.470918083190918,
      "epoch": 0.8587939429611001,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004936197437835329,
      "loss": 6.278,
      "mean_token_accuracy": 0.13847905173897743,
      "num_tokens": 17385492.0,
      "step": 8025
    },
    {
      "entropy": 6.547710371017456,
      "epoch": 0.859329017068864,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004936107002941599,
      "loss": 6.215,
      "mean_token_accuracy": 0.1442974954843521,
      "num_tokens": 17395640.0,
      "step": 8030
    },
    {
      "entropy": 6.489324522018433,
      "epoch": 0.859864091176628,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004936016504924337,
      "loss": 6.2492,
      "mean_token_accuracy": 0.13880942687392234,
      "num_tokens": 17405211.0,
      "step": 8035
    },
    {
      "entropy": 6.460426187515258,
      "epoch": 0.8603991652843919,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004935925943786155,
      "loss": 6.2162,
      "mean_token_accuracy": 0.14056197851896285,
      "num_tokens": 17415421.0,
      "step": 8040
    },
    {
      "entropy": 6.484164619445801,
      "epoch": 0.8609342393921559,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004935835319529669,
      "loss": 6.2758,
      "mean_token_accuracy": 0.14110616594552994,
      "num_tokens": 17426040.0,
      "step": 8045
    },
    {
      "entropy": 6.4609459400177,
      "epoch": 0.8614693134999197,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004935744632157496,
      "loss": 6.2578,
      "mean_token_accuracy": 0.14142901971936225,
      "num_tokens": 17436305.0,
      "step": 8050
    },
    {
      "entropy": 6.494168329238891,
      "epoch": 0.8620043876076836,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004935653881672253,
      "loss": 6.3011,
      "mean_token_accuracy": 0.13791701942682266,
      "num_tokens": 17446786.0,
      "step": 8055
    },
    {
      "entropy": 6.4791186332702635,
      "epoch": 0.8625394617154476,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004935563068076561,
      "loss": 6.1918,
      "mean_token_accuracy": 0.147944313287735,
      "num_tokens": 17456613.0,
      "step": 8060
    },
    {
      "entropy": 6.38516206741333,
      "epoch": 0.8630745358232115,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004935472191373044,
      "loss": 6.1426,
      "mean_token_accuracy": 0.14823999702930452,
      "num_tokens": 17467281.0,
      "step": 8065
    },
    {
      "entropy": 6.463254356384278,
      "epoch": 0.8636096099309755,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004935381251564324,
      "loss": 6.2615,
      "mean_token_accuracy": 0.13736518770456313,
      "num_tokens": 17478271.0,
      "step": 8070
    },
    {
      "entropy": 6.48794641494751,
      "epoch": 0.8641446840387393,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004935290248653027,
      "loss": 6.2459,
      "mean_token_accuracy": 0.14025778323411942,
      "num_tokens": 17489519.0,
      "step": 8075
    },
    {
      "entropy": 6.524880504608154,
      "epoch": 0.8646797581465033,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004935199182641781,
      "loss": 6.2158,
      "mean_token_accuracy": 0.14082201719284057,
      "num_tokens": 17500076.0,
      "step": 8080
    },
    {
      "entropy": 6.5165893077850345,
      "epoch": 0.8652148322542672,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004935108053533216,
      "loss": 6.2167,
      "mean_token_accuracy": 0.13490459620952605,
      "num_tokens": 17509898.0,
      "step": 8085
    },
    {
      "entropy": 6.435736274719238,
      "epoch": 0.8657499063620312,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004935016861329964,
      "loss": 6.1613,
      "mean_token_accuracy": 0.1448570892214775,
      "num_tokens": 17519890.0,
      "step": 8090
    },
    {
      "entropy": 6.4099431991577145,
      "epoch": 0.8662849804697951,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004934925606034656,
      "loss": 6.2016,
      "mean_token_accuracy": 0.13980754762887954,
      "num_tokens": 17530927.0,
      "step": 8095
    },
    {
      "entropy": 6.510073041915893,
      "epoch": 0.866820054577559,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004934834287649928,
      "loss": 6.2963,
      "mean_token_accuracy": 0.13630046993494033,
      "num_tokens": 17541174.0,
      "step": 8100
    },
    {
      "entropy": 6.504948520660401,
      "epoch": 0.8673551286853229,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004934742906178417,
      "loss": 6.2603,
      "mean_token_accuracy": 0.13654820397496223,
      "num_tokens": 17552460.0,
      "step": 8105
    },
    {
      "entropy": 6.531632614135742,
      "epoch": 0.8678902027930868,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004934651461622762,
      "loss": 6.2525,
      "mean_token_accuracy": 0.14166218042373657,
      "num_tokens": 17563908.0,
      "step": 8110
    },
    {
      "entropy": 6.466775894165039,
      "epoch": 0.8684252769008508,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004934559953985603,
      "loss": 6.2573,
      "mean_token_accuracy": 0.14609386771917343,
      "num_tokens": 17574314.0,
      "step": 8115
    },
    {
      "entropy": 6.459406852722168,
      "epoch": 0.8689603510086147,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004934468383269582,
      "loss": 6.2311,
      "mean_token_accuracy": 0.14299913197755815,
      "num_tokens": 17583719.0,
      "step": 8120
    },
    {
      "entropy": 6.401128053665161,
      "epoch": 0.8694954251163787,
      "grad_norm": 0.875,
      "learning_rate": 0.0004934376749477344,
      "loss": 6.1965,
      "mean_token_accuracy": 0.1476762294769287,
      "num_tokens": 17594522.0,
      "step": 8125
    },
    {
      "entropy": 6.4326173782348635,
      "epoch": 0.8700304992241426,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004934285052611533,
      "loss": 6.1711,
      "mean_token_accuracy": 0.14994932785630227,
      "num_tokens": 17604632.0,
      "step": 8130
    },
    {
      "entropy": 6.506157541275025,
      "epoch": 0.8705655733319064,
      "grad_norm": 0.9375,
      "learning_rate": 0.00049341932926748,
      "loss": 6.3052,
      "mean_token_accuracy": 0.1354134902358055,
      "num_tokens": 17615459.0,
      "step": 8135
    },
    {
      "entropy": 6.474421977996826,
      "epoch": 0.8711006474396704,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004934101469669791,
      "loss": 6.2506,
      "mean_token_accuracy": 0.1445701651275158,
      "num_tokens": 17625800.0,
      "step": 8140
    },
    {
      "entropy": 6.369798183441162,
      "epoch": 0.8716357215474343,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004934009583599159,
      "loss": 6.0743,
      "mean_token_accuracy": 0.1552235320210457,
      "num_tokens": 17637264.0,
      "step": 8145
    },
    {
      "entropy": 6.448842477798462,
      "epoch": 0.8721707956551983,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004933917634465558,
      "loss": 6.1273,
      "mean_token_accuracy": 0.1460499256849289,
      "num_tokens": 17648256.0,
      "step": 8150
    },
    {
      "entropy": 6.388854551315307,
      "epoch": 0.8727058697629622,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004933825622271642,
      "loss": 6.2548,
      "mean_token_accuracy": 0.13975195586681366,
      "num_tokens": 17659096.0,
      "step": 8155
    },
    {
      "entropy": 6.4904468059539795,
      "epoch": 0.8732409438707261,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004933733547020069,
      "loss": 6.2317,
      "mean_token_accuracy": 0.14684420675039292,
      "num_tokens": 17669808.0,
      "step": 8160
    },
    {
      "entropy": 6.506088829040527,
      "epoch": 0.87377601797849,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004933641408713495,
      "loss": 6.1948,
      "mean_token_accuracy": 0.14477132260799408,
      "num_tokens": 17680375.0,
      "step": 8165
    },
    {
      "entropy": 6.355422782897949,
      "epoch": 0.8743110920862539,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004933549207354582,
      "loss": 6.164,
      "mean_token_accuracy": 0.14456060230731965,
      "num_tokens": 17690780.0,
      "step": 8170
    },
    {
      "entropy": 6.455756998062133,
      "epoch": 0.8748461661940179,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004933456942945994,
      "loss": 6.2794,
      "mean_token_accuracy": 0.14372563734650612,
      "num_tokens": 17702170.0,
      "step": 8175
    },
    {
      "entropy": 6.447136497497558,
      "epoch": 0.8753812403017818,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004933364615490393,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14216097965836524,
      "num_tokens": 17712631.0,
      "step": 8180
    },
    {
      "entropy": 6.510778856277466,
      "epoch": 0.8759163144095458,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933272224990445,
      "loss": 6.3019,
      "mean_token_accuracy": 0.13427601531147956,
      "num_tokens": 17723692.0,
      "step": 8185
    },
    {
      "entropy": 6.537478256225586,
      "epoch": 0.8764513885173096,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004933179771448819,
      "loss": 6.1829,
      "mean_token_accuracy": 0.15159363076090812,
      "num_tokens": 17735264.0,
      "step": 8190
    },
    {
      "entropy": 6.390658330917359,
      "epoch": 0.8769864626250736,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004933087254868183,
      "loss": 6.0927,
      "mean_token_accuracy": 0.1521390289068222,
      "num_tokens": 17747060.0,
      "step": 8195
    },
    {
      "entropy": 6.394728803634644,
      "epoch": 0.8775215367328375,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000493299467525121,
      "loss": 6.228,
      "mean_token_accuracy": 0.138492451608181,
      "num_tokens": 17758159.0,
      "step": 8200
    },
    {
      "entropy": 6.475798273086548,
      "epoch": 0.8780566108406014,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004932902032600573,
      "loss": 6.2271,
      "mean_token_accuracy": 0.14088183417916297,
      "num_tokens": 17768444.0,
      "step": 8205
    },
    {
      "entropy": 6.456412696838379,
      "epoch": 0.8785916849483654,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004932809326918945,
      "loss": 6.1086,
      "mean_token_accuracy": 0.1525212988257408,
      "num_tokens": 17779303.0,
      "step": 8210
    },
    {
      "entropy": 6.44280481338501,
      "epoch": 0.8791267590561292,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004932716558209005,
      "loss": 6.3172,
      "mean_token_accuracy": 0.13491011261940003,
      "num_tokens": 17789853.0,
      "step": 8215
    },
    {
      "entropy": 6.517859029769897,
      "epoch": 0.8796618331638932,
      "grad_norm": 1.0,
      "learning_rate": 0.0004932623726473432,
      "loss": 6.1848,
      "mean_token_accuracy": 0.1506144881248474,
      "num_tokens": 17799232.0,
      "step": 8220
    },
    {
      "entropy": 6.394598865509034,
      "epoch": 0.8801969072716571,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004932530831714904,
      "loss": 6.1785,
      "mean_token_accuracy": 0.14645391702651978,
      "num_tokens": 17808870.0,
      "step": 8225
    },
    {
      "entropy": 6.391938924789429,
      "epoch": 0.8807319813794211,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004932437873936107,
      "loss": 6.2461,
      "mean_token_accuracy": 0.1392425887286663,
      "num_tokens": 17819394.0,
      "step": 8230
    },
    {
      "entropy": 6.5266252040863035,
      "epoch": 0.881267055487185,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004932344853139721,
      "loss": 6.2443,
      "mean_token_accuracy": 0.14133281037211418,
      "num_tokens": 17830220.0,
      "step": 8235
    },
    {
      "entropy": 6.428589296340943,
      "epoch": 0.8818021295949489,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004932251769328436,
      "loss": 6.1881,
      "mean_token_accuracy": 0.1395357370376587,
      "num_tokens": 17841971.0,
      "step": 8240
    },
    {
      "entropy": 6.449368190765381,
      "epoch": 0.8823372037027128,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004932158622504937,
      "loss": 6.1969,
      "mean_token_accuracy": 0.1442675620317459,
      "num_tokens": 17851890.0,
      "step": 8245
    },
    {
      "entropy": 6.4090735912323,
      "epoch": 0.8828722778104767,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004932065412671915,
      "loss": 6.1875,
      "mean_token_accuracy": 0.1493501305580139,
      "num_tokens": 17862855.0,
      "step": 8250
    },
    {
      "entropy": 6.498145198822021,
      "epoch": 0.8834073519182407,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004931972139832058,
      "loss": 6.1969,
      "mean_token_accuracy": 0.14035707712173462,
      "num_tokens": 17873550.0,
      "step": 8255
    },
    {
      "entropy": 6.4655333995819095,
      "epoch": 0.8839424260260046,
      "grad_norm": 0.875,
      "learning_rate": 0.0004931878803988066,
      "loss": 6.2113,
      "mean_token_accuracy": 0.14411091208457946,
      "num_tokens": 17884067.0,
      "step": 8260
    },
    {
      "entropy": 6.44084734916687,
      "epoch": 0.8844775001337686,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004931785405142627,
      "loss": 6.2264,
      "mean_token_accuracy": 0.14473579600453376,
      "num_tokens": 17894587.0,
      "step": 8265
    },
    {
      "entropy": 6.494640493392945,
      "epoch": 0.8850125742415325,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004931691943298443,
      "loss": 6.2298,
      "mean_token_accuracy": 0.138222074508667,
      "num_tokens": 17905297.0,
      "step": 8270
    },
    {
      "entropy": 6.427201509475708,
      "epoch": 0.8855476483492963,
      "grad_norm": 0.8671875,
      "learning_rate": 0.000493159841845821,
      "loss": 6.1419,
      "mean_token_accuracy": 0.144347370415926,
      "num_tokens": 17916225.0,
      "step": 8275
    },
    {
      "entropy": 6.504813957214355,
      "epoch": 0.8860827224570603,
      "grad_norm": 1.0,
      "learning_rate": 0.0004931504830624629,
      "loss": 6.1912,
      "mean_token_accuracy": 0.142536723613739,
      "num_tokens": 17925795.0,
      "step": 8280
    },
    {
      "entropy": 6.44462628364563,
      "epoch": 0.8866177965648242,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004931411179800403,
      "loss": 6.1903,
      "mean_token_accuracy": 0.14010872691869736,
      "num_tokens": 17936134.0,
      "step": 8285
    },
    {
      "entropy": 6.471136140823364,
      "epoch": 0.8871528706725882,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004931317465988235,
      "loss": 6.2576,
      "mean_token_accuracy": 0.1403445340692997,
      "num_tokens": 17946393.0,
      "step": 8290
    },
    {
      "entropy": 6.517022895812988,
      "epoch": 0.8876879447803521,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004931223689190832,
      "loss": 6.2823,
      "mean_token_accuracy": 0.13318857923150063,
      "num_tokens": 17957527.0,
      "step": 8295
    },
    {
      "entropy": 6.520452547073364,
      "epoch": 0.8882230188881161,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004931129849410903,
      "loss": 6.262,
      "mean_token_accuracy": 0.13815884441137313,
      "num_tokens": 17968390.0,
      "step": 8300
    },
    {
      "entropy": 6.518455266952515,
      "epoch": 0.8887580929958799,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004931035946651154,
      "loss": 6.1898,
      "mean_token_accuracy": 0.1582094445824623,
      "num_tokens": 17978667.0,
      "step": 8305
    },
    {
      "entropy": 6.480077505111694,
      "epoch": 0.8892931671036438,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930941980914299,
      "loss": 6.2282,
      "mean_token_accuracy": 0.1417049378156662,
      "num_tokens": 17989008.0,
      "step": 8310
    },
    {
      "entropy": 6.405767440795898,
      "epoch": 0.8898282412114078,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004930847952203052,
      "loss": 6.2001,
      "mean_token_accuracy": 0.13765018358826636,
      "num_tokens": 18000199.0,
      "step": 8315
    },
    {
      "entropy": 6.522706699371338,
      "epoch": 0.8903633153191717,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004930753860520126,
      "loss": 6.1332,
      "mean_token_accuracy": 0.1412038892507553,
      "num_tokens": 18010941.0,
      "step": 8320
    },
    {
      "entropy": 6.549570560455322,
      "epoch": 0.8908983894269357,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004930659705868237,
      "loss": 6.2465,
      "mean_token_accuracy": 0.14199950248003007,
      "num_tokens": 18021519.0,
      "step": 8325
    },
    {
      "entropy": 6.491166019439698,
      "epoch": 0.8914334635346995,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004930565488250108,
      "loss": 6.2108,
      "mean_token_accuracy": 0.14039984792470933,
      "num_tokens": 18032773.0,
      "step": 8330
    },
    {
      "entropy": 6.4073690414428714,
      "epoch": 0.8919685376424635,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004930471207668456,
      "loss": 6.1647,
      "mean_token_accuracy": 0.1511008135974407,
      "num_tokens": 18043250.0,
      "step": 8335
    },
    {
      "entropy": 6.425286436080933,
      "epoch": 0.8925036117502274,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004930376864126005,
      "loss": 6.1657,
      "mean_token_accuracy": 0.15050228536128998,
      "num_tokens": 18055165.0,
      "step": 8340
    },
    {
      "entropy": 6.4237892627716064,
      "epoch": 0.8930386858579913,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004930282457625476,
      "loss": 6.126,
      "mean_token_accuracy": 0.15399046689271928,
      "num_tokens": 18065296.0,
      "step": 8345
    },
    {
      "entropy": 6.419537687301636,
      "epoch": 0.8935737599657553,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00049301879881696,
      "loss": 6.1576,
      "mean_token_accuracy": 0.1441187486052513,
      "num_tokens": 18075536.0,
      "step": 8350
    },
    {
      "entropy": 6.477183246612549,
      "epoch": 0.8941088340735192,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004930093455761101,
      "loss": 6.2611,
      "mean_token_accuracy": 0.14573011174798012,
      "num_tokens": 18086253.0,
      "step": 8355
    },
    {
      "entropy": 6.452179193496704,
      "epoch": 0.8946439081812831,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000492999886040271,
      "loss": 6.2447,
      "mean_token_accuracy": 0.14390550181269646,
      "num_tokens": 18097335.0,
      "step": 8360
    },
    {
      "entropy": 6.4713541030883786,
      "epoch": 0.895178982289047,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004929904202097159,
      "loss": 6.1639,
      "mean_token_accuracy": 0.1419937528669834,
      "num_tokens": 18107293.0,
      "step": 8365
    },
    {
      "entropy": 6.413883543014526,
      "epoch": 0.895714056396811,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004929809480847179,
      "loss": 6.3171,
      "mean_token_accuracy": 0.13175713121891022,
      "num_tokens": 18117705.0,
      "step": 8370
    },
    {
      "entropy": 6.511688280105591,
      "epoch": 0.8962491305045749,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004929714696655507,
      "loss": 6.1998,
      "mean_token_accuracy": 0.138552325963974,
      "num_tokens": 18129003.0,
      "step": 8375
    },
    {
      "entropy": 6.435767269134521,
      "epoch": 0.8967842046123388,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000492961984952488,
      "loss": 6.155,
      "mean_token_accuracy": 0.15483056306838988,
      "num_tokens": 18140843.0,
      "step": 8380
    },
    {
      "entropy": 6.433523893356323,
      "epoch": 0.8973192787201028,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004929524939458037,
      "loss": 6.25,
      "mean_token_accuracy": 0.13802943378686905,
      "num_tokens": 18151476.0,
      "step": 8385
    },
    {
      "entropy": 6.464963054656982,
      "epoch": 0.8978543528278666,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004929429966457716,
      "loss": 6.269,
      "mean_token_accuracy": 0.13858315348625183,
      "num_tokens": 18161304.0,
      "step": 8390
    },
    {
      "entropy": 6.500939226150512,
      "epoch": 0.8983894269356306,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004929334930526662,
      "loss": 6.207,
      "mean_token_accuracy": 0.15046523809432982,
      "num_tokens": 18173696.0,
      "step": 8395
    },
    {
      "entropy": 6.452808618545532,
      "epoch": 0.8989245010433945,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004929239831667617,
      "loss": 6.225,
      "mean_token_accuracy": 0.13648192286491395,
      "num_tokens": 18186024.0,
      "step": 8400
    },
    {
      "entropy": 6.4754187107086185,
      "epoch": 0.8994595751511585,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004929144669883329,
      "loss": 6.234,
      "mean_token_accuracy": 0.14507345035672187,
      "num_tokens": 18196976.0,
      "step": 8405
    },
    {
      "entropy": 6.4965128898620605,
      "epoch": 0.8999946492589224,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004929049445176545,
      "loss": 6.2435,
      "mean_token_accuracy": 0.13944084644317628,
      "num_tokens": 18208086.0,
      "step": 8410
    },
    {
      "entropy": 6.443697834014893,
      "epoch": 0.9005297233666862,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004928954157550013,
      "loss": 6.1727,
      "mean_token_accuracy": 0.15264458656311036,
      "num_tokens": 18218468.0,
      "step": 8415
    },
    {
      "entropy": 6.506827449798584,
      "epoch": 0.9010647974744502,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004928858807006488,
      "loss": 6.2814,
      "mean_token_accuracy": 0.146379953622818,
      "num_tokens": 18230489.0,
      "step": 8420
    },
    {
      "entropy": 6.446054220199585,
      "epoch": 0.9015998715822141,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004928763393548721,
      "loss": 6.2164,
      "mean_token_accuracy": 0.14563113823533058,
      "num_tokens": 18239968.0,
      "step": 8425
    },
    {
      "entropy": 6.470791387557983,
      "epoch": 0.9021349456899781,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004928667917179465,
      "loss": 6.2636,
      "mean_token_accuracy": 0.13968126103281975,
      "num_tokens": 18250902.0,
      "step": 8430
    },
    {
      "entropy": 6.570216989517212,
      "epoch": 0.902670019797742,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004928572377901481,
      "loss": 6.2619,
      "mean_token_accuracy": 0.14227017015218735,
      "num_tokens": 18262094.0,
      "step": 8435
    },
    {
      "entropy": 6.512625312805175,
      "epoch": 0.903205093905506,
      "grad_norm": 0.875,
      "learning_rate": 0.0004928476775717524,
      "loss": 6.2391,
      "mean_token_accuracy": 0.14174272492527962,
      "num_tokens": 18272765.0,
      "step": 8440
    },
    {
      "entropy": 6.415294218063354,
      "epoch": 0.9037401680132698,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004928381110630358,
      "loss": 6.1184,
      "mean_token_accuracy": 0.15091901198029517,
      "num_tokens": 18284570.0,
      "step": 8445
    },
    {
      "entropy": 6.5518545627594,
      "epoch": 0.9042752421210337,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004928285382642742,
      "loss": 6.2444,
      "mean_token_accuracy": 0.13692313879728318,
      "num_tokens": 18295170.0,
      "step": 8450
    },
    {
      "entropy": 6.426909875869751,
      "epoch": 0.9048103162287977,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004928189591757443,
      "loss": 6.1804,
      "mean_token_accuracy": 0.13876531049609184,
      "num_tokens": 18306388.0,
      "step": 8455
    },
    {
      "entropy": 6.438180589675904,
      "epoch": 0.9053453903365616,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004928093737977225,
      "loss": 6.1954,
      "mean_token_accuracy": 0.14388242959976197,
      "num_tokens": 18316363.0,
      "step": 8460
    },
    {
      "entropy": 6.367106246948242,
      "epoch": 0.9058804644443256,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004927997821304856,
      "loss": 6.1385,
      "mean_token_accuracy": 0.14865853562951087,
      "num_tokens": 18327397.0,
      "step": 8465
    },
    {
      "entropy": 6.493155765533447,
      "epoch": 0.9064155385520895,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004927901841743107,
      "loss": 6.2671,
      "mean_token_accuracy": 0.13924047946929932,
      "num_tokens": 18338909.0,
      "step": 8470
    },
    {
      "entropy": 6.51797890663147,
      "epoch": 0.9069506126598534,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004927805799294747,
      "loss": 6.2214,
      "mean_token_accuracy": 0.1413904421031475,
      "num_tokens": 18350110.0,
      "step": 8475
    },
    {
      "entropy": 6.50021653175354,
      "epoch": 0.9074856867676173,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004927709693962551,
      "loss": 6.275,
      "mean_token_accuracy": 0.1353155218064785,
      "num_tokens": 18361590.0,
      "step": 8480
    },
    {
      "entropy": 6.43279881477356,
      "epoch": 0.9080207608753812,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004927613525749295,
      "loss": 6.2202,
      "mean_token_accuracy": 0.1394275188446045,
      "num_tokens": 18372602.0,
      "step": 8485
    },
    {
      "entropy": 6.522397661209107,
      "epoch": 0.9085558349831452,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004927517294657753,
      "loss": 6.2601,
      "mean_token_accuracy": 0.14095053374767302,
      "num_tokens": 18384653.0,
      "step": 8490
    },
    {
      "entropy": 6.572888708114624,
      "epoch": 0.9090909090909091,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004927421000690705,
      "loss": 6.2316,
      "mean_token_accuracy": 0.13683453053236008,
      "num_tokens": 18395379.0,
      "step": 8495
    },
    {
      "entropy": 6.454066371917724,
      "epoch": 0.909625983198673,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004927324643850933,
      "loss": 6.2005,
      "mean_token_accuracy": 0.1451219365000725,
      "num_tokens": 18405915.0,
      "step": 8500
    },
    {
      "entropy": 6.350123691558838,
      "epoch": 0.9101610573064369,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004927228224141217,
      "loss": 6.0896,
      "mean_token_accuracy": 0.1604235753417015,
      "num_tokens": 18416656.0,
      "step": 8505
    },
    {
      "entropy": 6.436915111541748,
      "epoch": 0.9106961314142009,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004927131741564342,
      "loss": 6.2332,
      "mean_token_accuracy": 0.13724781796336175,
      "num_tokens": 18427141.0,
      "step": 8510
    },
    {
      "entropy": 6.4622210502624515,
      "epoch": 0.9112312055219648,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004927035196123094,
      "loss": 6.245,
      "mean_token_accuracy": 0.14686613231897355,
      "num_tokens": 18437135.0,
      "step": 8515
    },
    {
      "entropy": 6.589114761352539,
      "epoch": 0.9117662796297287,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000492693858782026,
      "loss": 6.3184,
      "mean_token_accuracy": 0.13272974640130997,
      "num_tokens": 18448567.0,
      "step": 8520
    },
    {
      "entropy": 6.575060272216797,
      "epoch": 0.9123013537374927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000492684191665863,
      "loss": 6.2913,
      "mean_token_accuracy": 0.1354794941842556,
      "num_tokens": 18459385.0,
      "step": 8525
    },
    {
      "entropy": 6.4733765602111815,
      "epoch": 0.9128364278452565,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004926745182640996,
      "loss": 6.233,
      "mean_token_accuracy": 0.14534093737602233,
      "num_tokens": 18471376.0,
      "step": 8530
    },
    {
      "entropy": 6.50493049621582,
      "epoch": 0.9133715019530205,
      "grad_norm": 1.0,
      "learning_rate": 0.0004926648385770152,
      "loss": 6.3131,
      "mean_token_accuracy": 0.13382241129875183,
      "num_tokens": 18481950.0,
      "step": 8535
    },
    {
      "entropy": 6.538687610626221,
      "epoch": 0.9139065760607844,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004926551526048891,
      "loss": 6.2224,
      "mean_token_accuracy": 0.14380355328321456,
      "num_tokens": 18492436.0,
      "step": 8540
    },
    {
      "entropy": 6.475391149520874,
      "epoch": 0.9144416501685484,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004926454603480009,
      "loss": 6.063,
      "mean_token_accuracy": 0.1486901819705963,
      "num_tokens": 18504279.0,
      "step": 8545
    },
    {
      "entropy": 6.49855489730835,
      "epoch": 0.9149767242763123,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004926357618066307,
      "loss": 6.2083,
      "mean_token_accuracy": 0.1432658165693283,
      "num_tokens": 18515262.0,
      "step": 8550
    },
    {
      "entropy": 6.475007104873657,
      "epoch": 0.9155117983840761,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004926260569810586,
      "loss": 6.2505,
      "mean_token_accuracy": 0.1455526813864708,
      "num_tokens": 18525176.0,
      "step": 8555
    },
    {
      "entropy": 6.429649925231933,
      "epoch": 0.9160468724918401,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004926163458715645,
      "loss": 6.2405,
      "mean_token_accuracy": 0.13883231729269027,
      "num_tokens": 18536674.0,
      "step": 8560
    },
    {
      "entropy": 6.489083862304687,
      "epoch": 0.916581946599604,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004926066284784292,
      "loss": 6.2105,
      "mean_token_accuracy": 0.14140415489673613,
      "num_tokens": 18547398.0,
      "step": 8565
    },
    {
      "entropy": 6.499289560317993,
      "epoch": 0.917117020707368,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004925969048019329,
      "loss": 6.233,
      "mean_token_accuracy": 0.14758445024490358,
      "num_tokens": 18558838.0,
      "step": 8570
    },
    {
      "entropy": 6.472250127792359,
      "epoch": 0.9176520948151319,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004925871748423566,
      "loss": 6.191,
      "mean_token_accuracy": 0.1454857811331749,
      "num_tokens": 18569881.0,
      "step": 8575
    },
    {
      "entropy": 6.422400283813476,
      "epoch": 0.9181871689228959,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004925774385999812,
      "loss": 6.0968,
      "mean_token_accuracy": 0.15187532901763917,
      "num_tokens": 18579556.0,
      "step": 8580
    },
    {
      "entropy": 6.341235733032226,
      "epoch": 0.9187222430306597,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004925676960750878,
      "loss": 6.1269,
      "mean_token_accuracy": 0.15357157960534096,
      "num_tokens": 18590580.0,
      "step": 8585
    },
    {
      "entropy": 6.512661075592041,
      "epoch": 0.9192573171384236,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004925579472679578,
      "loss": 6.2386,
      "mean_token_accuracy": 0.14388099759817125,
      "num_tokens": 18600813.0,
      "step": 8590
    },
    {
      "entropy": 6.445880270004272,
      "epoch": 0.9197923912461876,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004925481921788726,
      "loss": 6.1593,
      "mean_token_accuracy": 0.14935692250728608,
      "num_tokens": 18610807.0,
      "step": 8595
    },
    {
      "entropy": 6.3805183410644535,
      "epoch": 0.9203274653539515,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000492538430808114,
      "loss": 6.1273,
      "mean_token_accuracy": 0.14852006882429122,
      "num_tokens": 18622108.0,
      "step": 8600
    },
    {
      "entropy": 6.461879444122315,
      "epoch": 0.9208625394617155,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004925286631559636,
      "loss": 6.2957,
      "mean_token_accuracy": 0.14091389775276184,
      "num_tokens": 18633098.0,
      "step": 8605
    },
    {
      "entropy": 6.484151840209961,
      "epoch": 0.9213976135694794,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004925188892227038,
      "loss": 6.1845,
      "mean_token_accuracy": 0.14727117121219635,
      "num_tokens": 18643603.0,
      "step": 8610
    },
    {
      "entropy": 6.45102071762085,
      "epoch": 0.9219326876772433,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004925091090086166,
      "loss": 6.1988,
      "mean_token_accuracy": 0.14656163156032562,
      "num_tokens": 18654354.0,
      "step": 8615
    },
    {
      "entropy": 6.527435445785523,
      "epoch": 0.9224677617850072,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004924993225139845,
      "loss": 6.2663,
      "mean_token_accuracy": 0.14755606800317764,
      "num_tokens": 18665349.0,
      "step": 8620
    },
    {
      "entropy": 6.503271293640137,
      "epoch": 0.9230028358927711,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004924895297390899,
      "loss": 6.2307,
      "mean_token_accuracy": 0.14059074074029923,
      "num_tokens": 18676575.0,
      "step": 8625
    },
    {
      "entropy": 6.476508283615113,
      "epoch": 0.9235379100005351,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004924797306842157,
      "loss": 6.1679,
      "mean_token_accuracy": 0.14684823229908944,
      "num_tokens": 18686501.0,
      "step": 8630
    },
    {
      "entropy": 6.4832337379455565,
      "epoch": 0.924072984108299,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004924699253496449,
      "loss": 6.2238,
      "mean_token_accuracy": 0.1421502150595188,
      "num_tokens": 18697246.0,
      "step": 8635
    },
    {
      "entropy": 6.374241018295288,
      "epoch": 0.924608058216063,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924601137356605,
      "loss": 6.1548,
      "mean_token_accuracy": 0.15231939256191254,
      "num_tokens": 18708177.0,
      "step": 8640
    },
    {
      "entropy": 6.458330821990967,
      "epoch": 0.9251431323238268,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000492450295842546,
      "loss": 6.2111,
      "mean_token_accuracy": 0.1425161123275757,
      "num_tokens": 18719334.0,
      "step": 8645
    },
    {
      "entropy": 6.508578205108643,
      "epoch": 0.9256782064315908,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004924404716705847,
      "loss": 6.1882,
      "mean_token_accuracy": 0.14905410408973693,
      "num_tokens": 18729174.0,
      "step": 8650
    },
    {
      "entropy": 6.470645904541016,
      "epoch": 0.9262132805393547,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004924306412200603,
      "loss": 6.2955,
      "mean_token_accuracy": 0.1374170996248722,
      "num_tokens": 18740277.0,
      "step": 8655
    },
    {
      "entropy": 6.488519620895386,
      "epoch": 0.9267483546471186,
      "grad_norm": 0.875,
      "learning_rate": 0.0004924208044912567,
      "loss": 6.2312,
      "mean_token_accuracy": 0.14602856040000917,
      "num_tokens": 18751864.0,
      "step": 8660
    },
    {
      "entropy": 6.390778303146362,
      "epoch": 0.9272834287548826,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004924109614844579,
      "loss": 6.1468,
      "mean_token_accuracy": 0.15138288885354995,
      "num_tokens": 18762786.0,
      "step": 8665
    },
    {
      "entropy": 6.440236520767212,
      "epoch": 0.9278185028626464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004924011121999482,
      "loss": 6.1989,
      "mean_token_accuracy": 0.14837858080863953,
      "num_tokens": 18772913.0,
      "step": 8670
    },
    {
      "entropy": 6.466088914871216,
      "epoch": 0.9283535769704104,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004923912566380118,
      "loss": 6.1633,
      "mean_token_accuracy": 0.14807765036821366,
      "num_tokens": 18784204.0,
      "step": 8675
    },
    {
      "entropy": 6.371707487106323,
      "epoch": 0.9288886510781743,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004923813947989336,
      "loss": 6.2646,
      "mean_token_accuracy": 0.1406092531979084,
      "num_tokens": 18795425.0,
      "step": 8680
    },
    {
      "entropy": 6.511398839950561,
      "epoch": 0.9294237251859383,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004923715266829981,
      "loss": 6.1844,
      "mean_token_accuracy": 0.14496176391839982,
      "num_tokens": 18806850.0,
      "step": 8685
    },
    {
      "entropy": 6.440308332443237,
      "epoch": 0.9299587992937022,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004923616522904903,
      "loss": 6.1365,
      "mean_token_accuracy": 0.14974386394023895,
      "num_tokens": 18817272.0,
      "step": 8690
    },
    {
      "entropy": 6.476079654693604,
      "epoch": 0.930493873401466,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004923517716216953,
      "loss": 6.2815,
      "mean_token_accuracy": 0.1370019644498825,
      "num_tokens": 18827075.0,
      "step": 8695
    },
    {
      "entropy": 6.515619707107544,
      "epoch": 0.93102894750923,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004923418846768985,
      "loss": 6.1977,
      "mean_token_accuracy": 0.14271625205874444,
      "num_tokens": 18837635.0,
      "step": 8700
    },
    {
      "entropy": 6.484317922592163,
      "epoch": 0.9315640216169939,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004923319914563852,
      "loss": 6.1917,
      "mean_token_accuracy": 0.15020837187767028,
      "num_tokens": 18848559.0,
      "step": 8705
    },
    {
      "entropy": 6.4977153778076175,
      "epoch": 0.9320990957247579,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004923220919604413,
      "loss": 6.2153,
      "mean_token_accuracy": 0.139048457890749,
      "num_tokens": 18859327.0,
      "step": 8710
    },
    {
      "entropy": 6.413622426986694,
      "epoch": 0.9326341698325218,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004923121861893525,
      "loss": 6.168,
      "mean_token_accuracy": 0.14036728888750077,
      "num_tokens": 18869403.0,
      "step": 8715
    },
    {
      "entropy": 6.488254070281982,
      "epoch": 0.9331692439402858,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004923022741434049,
      "loss": 6.2026,
      "mean_token_accuracy": 0.13896684646606444,
      "num_tokens": 18879890.0,
      "step": 8720
    },
    {
      "entropy": 6.50945143699646,
      "epoch": 0.9337043180480497,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004922923558228846,
      "loss": 6.2224,
      "mean_token_accuracy": 0.13936150074005127,
      "num_tokens": 18890361.0,
      "step": 8725
    },
    {
      "entropy": 6.441441202163697,
      "epoch": 0.9342393921558135,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004922824312280781,
      "loss": 6.1417,
      "mean_token_accuracy": 0.14964256584644317,
      "num_tokens": 18902344.0,
      "step": 8730
    },
    {
      "entropy": 6.401036024093628,
      "epoch": 0.9347744662635775,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000492272500359272,
      "loss": 6.1394,
      "mean_token_accuracy": 0.149148790538311,
      "num_tokens": 18911364.0,
      "step": 8735
    },
    {
      "entropy": 6.446852159500122,
      "epoch": 0.9353095403713414,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004922625632167529,
      "loss": 6.2362,
      "mean_token_accuracy": 0.13709417283535003,
      "num_tokens": 18922254.0,
      "step": 8740
    },
    {
      "entropy": 6.521650123596191,
      "epoch": 0.9358446144791054,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004922526198008078,
      "loss": 6.2111,
      "mean_token_accuracy": 0.1378822520375252,
      "num_tokens": 18932943.0,
      "step": 8745
    },
    {
      "entropy": 6.495054626464844,
      "epoch": 0.9363796885868693,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004922426701117239,
      "loss": 6.2902,
      "mean_token_accuracy": 0.13874457031488419,
      "num_tokens": 18943523.0,
      "step": 8750
    },
    {
      "entropy": 6.535897827148437,
      "epoch": 0.9369147626946333,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004922327141497884,
      "loss": 6.2695,
      "mean_token_accuracy": 0.13618727624416352,
      "num_tokens": 18955220.0,
      "step": 8755
    },
    {
      "entropy": 6.555761384963989,
      "epoch": 0.9374498368023971,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004922227519152888,
      "loss": 6.2969,
      "mean_token_accuracy": 0.13864321932196616,
      "num_tokens": 18966714.0,
      "step": 8760
    },
    {
      "entropy": 6.439997386932373,
      "epoch": 0.937984910910161,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004922127834085128,
      "loss": 6.1404,
      "mean_token_accuracy": 0.1468714326620102,
      "num_tokens": 18977998.0,
      "step": 8765
    },
    {
      "entropy": 6.443836832046509,
      "epoch": 0.938519985017925,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004922028086297481,
      "loss": 6.1427,
      "mean_token_accuracy": 0.14381246864795685,
      "num_tokens": 18990210.0,
      "step": 8770
    },
    {
      "entropy": 6.5187712669372555,
      "epoch": 0.9390550591256889,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004921928275792828,
      "loss": 6.2347,
      "mean_token_accuracy": 0.14691857248544693,
      "num_tokens": 19000529.0,
      "step": 8775
    },
    {
      "entropy": 6.440783977508545,
      "epoch": 0.9395901332334529,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004921828402574052,
      "loss": 6.1755,
      "mean_token_accuracy": 0.14634492248296738,
      "num_tokens": 19011386.0,
      "step": 8780
    },
    {
      "entropy": 6.440403699874878,
      "epoch": 0.9401252073412167,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004921728466644037,
      "loss": 6.1758,
      "mean_token_accuracy": 0.1455728828907013,
      "num_tokens": 19022196.0,
      "step": 8785
    },
    {
      "entropy": 6.453129720687866,
      "epoch": 0.9406602814489807,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004921628468005666,
      "loss": 6.1924,
      "mean_token_accuracy": 0.14931313693523407,
      "num_tokens": 19033221.0,
      "step": 8790
    },
    {
      "entropy": 6.437537908554077,
      "epoch": 0.9411953555567446,
      "grad_norm": 0.859375,
      "learning_rate": 0.000492152840666183,
      "loss": 6.233,
      "mean_token_accuracy": 0.14072544127702713,
      "num_tokens": 19044953.0,
      "step": 8795
    },
    {
      "entropy": 6.534979820251465,
      "epoch": 0.9417304296645085,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004921428282615415,
      "loss": 6.2606,
      "mean_token_accuracy": 0.13626675456762313,
      "num_tokens": 19055512.0,
      "step": 8800
    },
    {
      "entropy": 6.504602432250977,
      "epoch": 0.9422655037722725,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004921328095869313,
      "loss": 6.2723,
      "mean_token_accuracy": 0.13825024142861367,
      "num_tokens": 19065659.0,
      "step": 8805
    },
    {
      "entropy": 6.445130443572998,
      "epoch": 0.9428005778800364,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004921227846426418,
      "loss": 6.1581,
      "mean_token_accuracy": 0.15683560371398925,
      "num_tokens": 19076955.0,
      "step": 8810
    },
    {
      "entropy": 6.480764627456665,
      "epoch": 0.9433356519878003,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004921127534289624,
      "loss": 6.256,
      "mean_token_accuracy": 0.1354398809373379,
      "num_tokens": 19088305.0,
      "step": 8815
    },
    {
      "entropy": 6.466386985778809,
      "epoch": 0.9438707260955642,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004921027159461829,
      "loss": 6.2487,
      "mean_token_accuracy": 0.14003829956054686,
      "num_tokens": 19098046.0,
      "step": 8820
    },
    {
      "entropy": 6.453748321533203,
      "epoch": 0.9444058002033282,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004920926721945927,
      "loss": 6.1619,
      "mean_token_accuracy": 0.1483352944254875,
      "num_tokens": 19109019.0,
      "step": 8825
    },
    {
      "entropy": 6.5257744789123535,
      "epoch": 0.9449408743110921,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004920826221744823,
      "loss": 6.1889,
      "mean_token_accuracy": 0.14775989204645157,
      "num_tokens": 19119624.0,
      "step": 8830
    },
    {
      "entropy": 6.474110651016235,
      "epoch": 0.945475948418856,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004920725658861417,
      "loss": 6.2369,
      "mean_token_accuracy": 0.13665593937039375,
      "num_tokens": 19130577.0,
      "step": 8835
    },
    {
      "entropy": 6.482955360412598,
      "epoch": 0.94601102252662,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004920625033298611,
      "loss": 6.1342,
      "mean_token_accuracy": 0.14972887337207794,
      "num_tokens": 19140254.0,
      "step": 8840
    },
    {
      "entropy": 6.428343439102173,
      "epoch": 0.9465460966343838,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004920524345059314,
      "loss": 6.1564,
      "mean_token_accuracy": 0.1475004658102989,
      "num_tokens": 19151357.0,
      "step": 8845
    },
    {
      "entropy": 6.507237672805786,
      "epoch": 0.9470811707421478,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004920423594146431,
      "loss": 6.1864,
      "mean_token_accuracy": 0.1468237280845642,
      "num_tokens": 19162069.0,
      "step": 8850
    },
    {
      "entropy": 6.380544519424438,
      "epoch": 0.9476162448499117,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004920322780562872,
      "loss": 6.0752,
      "mean_token_accuracy": 0.1535169079899788,
      "num_tokens": 19172515.0,
      "step": 8855
    },
    {
      "entropy": 6.3271417140960695,
      "epoch": 0.9481513189576757,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004920221904311546,
      "loss": 6.0952,
      "mean_token_accuracy": 0.1441791236400604,
      "num_tokens": 19182450.0,
      "step": 8860
    },
    {
      "entropy": 6.520216560363769,
      "epoch": 0.9486863930654396,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004920120965395369,
      "loss": 6.2842,
      "mean_token_accuracy": 0.14113576114177703,
      "num_tokens": 19193833.0,
      "step": 8865
    },
    {
      "entropy": 6.541619491577149,
      "epoch": 0.9492214671732034,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004920019963817253,
      "loss": 6.2349,
      "mean_token_accuracy": 0.13758261725306511,
      "num_tokens": 19204970.0,
      "step": 8870
    },
    {
      "entropy": 6.479512119293213,
      "epoch": 0.9497565412809674,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004919918899580117,
      "loss": 6.1924,
      "mean_token_accuracy": 0.14385030567646026,
      "num_tokens": 19215795.0,
      "step": 8875
    },
    {
      "entropy": 6.46100172996521,
      "epoch": 0.9502916153887313,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004919817772686876,
      "loss": 6.2304,
      "mean_token_accuracy": 0.136734214425087,
      "num_tokens": 19226375.0,
      "step": 8880
    },
    {
      "entropy": 6.495023155212403,
      "epoch": 0.9508266894964953,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004919716583140453,
      "loss": 6.2109,
      "mean_token_accuracy": 0.1478329211473465,
      "num_tokens": 19236066.0,
      "step": 8885
    },
    {
      "entropy": 6.430351591110229,
      "epoch": 0.9513617636042592,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004919615330943769,
      "loss": 6.1003,
      "mean_token_accuracy": 0.1485234946012497,
      "num_tokens": 19246757.0,
      "step": 8890
    },
    {
      "entropy": 6.510757064819336,
      "epoch": 0.9518968377120232,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004919514016099746,
      "loss": 6.2589,
      "mean_token_accuracy": 0.14149574860930442,
      "num_tokens": 19257875.0,
      "step": 8895
    },
    {
      "entropy": 6.511518335342407,
      "epoch": 0.952431911819787,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000491941263861131,
      "loss": 6.2017,
      "mean_token_accuracy": 0.14484582766890525,
      "num_tokens": 19270236.0,
      "step": 8900
    },
    {
      "entropy": 6.438380193710327,
      "epoch": 0.9529669859275509,
      "grad_norm": 0.953125,
      "learning_rate": 0.000491931119848139,
      "loss": 6.1947,
      "mean_token_accuracy": 0.14166287556290627,
      "num_tokens": 19280877.0,
      "step": 8905
    },
    {
      "entropy": 6.3953382015228275,
      "epoch": 0.9535020600353149,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004919209695712915,
      "loss": 6.1011,
      "mean_token_accuracy": 0.15066876411437988,
      "num_tokens": 19292852.0,
      "step": 8910
    },
    {
      "entropy": 6.5155603885650635,
      "epoch": 0.9540371341430788,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004919108130308814,
      "loss": 6.3096,
      "mean_token_accuracy": 0.13334777876734732,
      "num_tokens": 19305516.0,
      "step": 8915
    },
    {
      "entropy": 6.442837429046631,
      "epoch": 0.9545722082508428,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004919006502272019,
      "loss": 6.1258,
      "mean_token_accuracy": 0.15061808079481126,
      "num_tokens": 19315436.0,
      "step": 8920
    },
    {
      "entropy": 6.428034067153931,
      "epoch": 0.9551072823586066,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004918904811605468,
      "loss": 6.1724,
      "mean_token_accuracy": 0.14929206520318986,
      "num_tokens": 19325393.0,
      "step": 8925
    },
    {
      "entropy": 6.3922606945037845,
      "epoch": 0.9556423564663706,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004918803058312094,
      "loss": 6.1811,
      "mean_token_accuracy": 0.14905931875109674,
      "num_tokens": 19336722.0,
      "step": 8930
    },
    {
      "entropy": 6.537478923797607,
      "epoch": 0.9561774305741345,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004918701242394837,
      "loss": 6.3348,
      "mean_token_accuracy": 0.13431016877293586,
      "num_tokens": 19348406.0,
      "step": 8935
    },
    {
      "entropy": 6.4235334396362305,
      "epoch": 0.9567125046818984,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004918599363856637,
      "loss": 6.1407,
      "mean_token_accuracy": 0.1554851695895195,
      "num_tokens": 19359714.0,
      "step": 8940
    },
    {
      "entropy": 6.455921173095703,
      "epoch": 0.9572475787896624,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004918497422700434,
      "loss": 6.1837,
      "mean_token_accuracy": 0.14434646666049958,
      "num_tokens": 19370833.0,
      "step": 8945
    },
    {
      "entropy": 6.479765129089356,
      "epoch": 0.9577826528974263,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004918395418929174,
      "loss": 6.2673,
      "mean_token_accuracy": 0.13628800809383393,
      "num_tokens": 19382098.0,
      "step": 8950
    },
    {
      "entropy": 6.48378963470459,
      "epoch": 0.9583177270051902,
      "grad_norm": 1.0,
      "learning_rate": 0.00049182933525458,
      "loss": 6.2217,
      "mean_token_accuracy": 0.1419902265071869,
      "num_tokens": 19393595.0,
      "step": 8955
    },
    {
      "entropy": 6.4629234790802,
      "epoch": 0.9588528011129541,
      "grad_norm": 0.875,
      "learning_rate": 0.000491819122355326,
      "loss": 6.1891,
      "mean_token_accuracy": 0.1453189082443714,
      "num_tokens": 19405296.0,
      "step": 8960
    },
    {
      "entropy": 6.477556037902832,
      "epoch": 0.9593878752207181,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004918089031954503,
      "loss": 6.1989,
      "mean_token_accuracy": 0.14241567626595497,
      "num_tokens": 19415693.0,
      "step": 8965
    },
    {
      "entropy": 6.462906360626221,
      "epoch": 0.959922949328482,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000491798677775248,
      "loss": 6.2413,
      "mean_token_accuracy": 0.1438746988773346,
      "num_tokens": 19426140.0,
      "step": 8970
    },
    {
      "entropy": 6.52581205368042,
      "epoch": 0.9604580234362459,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004917884460950144,
      "loss": 6.2969,
      "mean_token_accuracy": 0.13814738094806672,
      "num_tokens": 19437543.0,
      "step": 8975
    },
    {
      "entropy": 6.449632263183593,
      "epoch": 0.9609930975440099,
      "grad_norm": 0.89453125,
      "learning_rate": 0.000491778208155045,
      "loss": 6.2024,
      "mean_token_accuracy": 0.1431194320321083,
      "num_tokens": 19447788.0,
      "step": 8980
    },
    {
      "entropy": 6.5050982475280765,
      "epoch": 0.9615281716517737,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004917679639556351,
      "loss": 6.2044,
      "mean_token_accuracy": 0.13954243138432504,
      "num_tokens": 19457587.0,
      "step": 8985
    },
    {
      "entropy": 6.463346862792969,
      "epoch": 0.9620632457595377,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917577134970808,
      "loss": 6.1178,
      "mean_token_accuracy": 0.14681217819452286,
      "num_tokens": 19467422.0,
      "step": 8990
    },
    {
      "entropy": 6.4243536472320555,
      "epoch": 0.9625983198673016,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004917474567796778,
      "loss": 6.1887,
      "mean_token_accuracy": 0.14375904724001884,
      "num_tokens": 19479037.0,
      "step": 8995
    },
    {
      "entropy": 6.403169393539429,
      "epoch": 0.9631333939750656,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004917371938037226,
      "loss": 6.1154,
      "mean_token_accuracy": 0.16109271273016929,
      "num_tokens": 19489639.0,
      "step": 9000
    },
    {
      "epoch": 0.9631333939750656,
      "eval_entropy": 6.2245820548013855,
      "eval_loss": 6.231032371520996,
      "eval_mean_token_accuracy": 0.14841360884500202,
      "eval_num_tokens": 19489639.0,
      "eval_runtime": 17.4846,
      "eval_samples_per_second": 1697.436,
      "eval_steps_per_second": 212.187,
      "step": 9000
    },
    {
      "entropy": 6.456880140304565,
      "epoch": 0.9636684680828295,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004917269245695114,
      "loss": 6.2426,
      "mean_token_accuracy": 0.14219753742218016,
      "num_tokens": 19500592.0,
      "step": 9005
    },
    {
      "entropy": 6.46939902305603,
      "epoch": 0.9642035421905933,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004917166490773406,
      "loss": 6.2435,
      "mean_token_accuracy": 0.13975200057029724,
      "num_tokens": 19510818.0,
      "step": 9010
    },
    {
      "entropy": 6.489015245437622,
      "epoch": 0.9647386162983573,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004917063673275071,
      "loss": 6.2637,
      "mean_token_accuracy": 0.14110172167420387,
      "num_tokens": 19521283.0,
      "step": 9015
    },
    {
      "entropy": 6.492032527923584,
      "epoch": 0.9652736904061212,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004916960793203077,
      "loss": 6.1112,
      "mean_token_accuracy": 0.14708787500858306,
      "num_tokens": 19531799.0,
      "step": 9020
    },
    {
      "entropy": 6.399127197265625,
      "epoch": 0.9658087645138852,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004916857850560394,
      "loss": 6.0772,
      "mean_token_accuracy": 0.1587416410446167,
      "num_tokens": 19542377.0,
      "step": 9025
    },
    {
      "entropy": 6.407155561447143,
      "epoch": 0.9663438386216491,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004916754845349996,
      "loss": 6.0939,
      "mean_token_accuracy": 0.15201009660959244,
      "num_tokens": 19552654.0,
      "step": 9030
    },
    {
      "entropy": 6.428509712219238,
      "epoch": 0.9668789127294131,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004916651777574856,
      "loss": 6.1679,
      "mean_token_accuracy": 0.14325408041477203,
      "num_tokens": 19563439.0,
      "step": 9035
    },
    {
      "entropy": 6.504122638702393,
      "epoch": 0.9674139868371769,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000491654864723795,
      "loss": 6.1981,
      "mean_token_accuracy": 0.1442462347447872,
      "num_tokens": 19574264.0,
      "step": 9040
    },
    {
      "entropy": 6.3582710266113285,
      "epoch": 0.9679490609449408,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004916445454342256,
      "loss": 6.0701,
      "mean_token_accuracy": 0.1570579081773758,
      "num_tokens": 19585293.0,
      "step": 9045
    },
    {
      "entropy": 6.437489986419678,
      "epoch": 0.9684841350527048,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004916342198890754,
      "loss": 6.1163,
      "mean_token_accuracy": 0.1407800741493702,
      "num_tokens": 19594384.0,
      "step": 9050
    },
    {
      "entropy": 6.396086120605469,
      "epoch": 0.9690192091604687,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004916238880886426,
      "loss": 6.2122,
      "mean_token_accuracy": 0.14744801074266434,
      "num_tokens": 19605144.0,
      "step": 9055
    },
    {
      "entropy": 6.401884508132935,
      "epoch": 0.9695542832682327,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004916135500332255,
      "loss": 6.1257,
      "mean_token_accuracy": 0.1512026757001877,
      "num_tokens": 19615885.0,
      "step": 9060
    },
    {
      "entropy": 6.4324079036712645,
      "epoch": 0.9700893573759966,
      "grad_norm": 1.0,
      "learning_rate": 0.0004916032057231225,
      "loss": 6.1336,
      "mean_token_accuracy": 0.14521500766277312,
      "num_tokens": 19625749.0,
      "step": 9065
    },
    {
      "entropy": 6.470341300964355,
      "epoch": 0.9706244314837605,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004915928551586324,
      "loss": 6.2287,
      "mean_token_accuracy": 0.1395932786166668,
      "num_tokens": 19638285.0,
      "step": 9070
    },
    {
      "entropy": 6.533346557617188,
      "epoch": 0.9711595055915244,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004915824983400541,
      "loss": 6.231,
      "mean_token_accuracy": 0.14019075855612756,
      "num_tokens": 19649697.0,
      "step": 9075
    },
    {
      "entropy": 6.501502466201782,
      "epoch": 0.9716945796992883,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004915721352676864,
      "loss": 6.2484,
      "mean_token_accuracy": 0.14228077232837677,
      "num_tokens": 19659668.0,
      "step": 9080
    },
    {
      "entropy": 6.437466859817505,
      "epoch": 0.9722296538070523,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004915617659418289,
      "loss": 6.1152,
      "mean_token_accuracy": 0.14758161455392838,
      "num_tokens": 19670301.0,
      "step": 9085
    },
    {
      "entropy": 6.433247423171997,
      "epoch": 0.9727647279148162,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004915513903627807,
      "loss": 6.1608,
      "mean_token_accuracy": 0.14314992651343345,
      "num_tokens": 19680926.0,
      "step": 9090
    },
    {
      "entropy": 6.424163961410523,
      "epoch": 0.9732998020225802,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004915410085308417,
      "loss": 6.151,
      "mean_token_accuracy": 0.15092322528362273,
      "num_tokens": 19691467.0,
      "step": 9095
    },
    {
      "entropy": 6.482801675796509,
      "epoch": 0.973834876130344,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004915306204463115,
      "loss": 6.161,
      "mean_token_accuracy": 0.146022729575634,
      "num_tokens": 19701139.0,
      "step": 9100
    },
    {
      "entropy": 6.432535314559937,
      "epoch": 0.974369950238108,
      "grad_norm": 0.984375,
      "learning_rate": 0.00049152022610949,
      "loss": 6.156,
      "mean_token_accuracy": 0.1427958242595196,
      "num_tokens": 19710628.0,
      "step": 9105
    },
    {
      "entropy": 6.450985336303711,
      "epoch": 0.9749050243458719,
      "grad_norm": 0.82421875,
      "learning_rate": 0.0004915098255206773,
      "loss": 6.1535,
      "mean_token_accuracy": 0.14337446838617324,
      "num_tokens": 19721824.0,
      "step": 9110
    },
    {
      "entropy": 6.422918701171875,
      "epoch": 0.9754400984536358,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000491499418680174,
      "loss": 6.1813,
      "mean_token_accuracy": 0.14035857543349267,
      "num_tokens": 19732557.0,
      "step": 9115
    },
    {
      "entropy": 6.451064109802246,
      "epoch": 0.9759751725613998,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004914890055882804,
      "loss": 6.1706,
      "mean_token_accuracy": 0.1465473085641861,
      "num_tokens": 19743436.0,
      "step": 9120
    },
    {
      "entropy": 6.443043327331543,
      "epoch": 0.9765102466691636,
      "grad_norm": 0.85546875,
      "learning_rate": 0.000491478586245297,
      "loss": 6.0791,
      "mean_token_accuracy": 0.15131891667842864,
      "num_tokens": 19755305.0,
      "step": 9125
    },
    {
      "entropy": 6.472864103317261,
      "epoch": 0.9770453207769276,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000491468160651525,
      "loss": 6.089,
      "mean_token_accuracy": 0.15562343299388887,
      "num_tokens": 19766223.0,
      "step": 9130
    },
    {
      "entropy": 6.435733985900879,
      "epoch": 0.9775803948846915,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004914577288072652,
      "loss": 6.1992,
      "mean_token_accuracy": 0.1429711565375328,
      "num_tokens": 19777268.0,
      "step": 9135
    },
    {
      "entropy": 6.475059175491333,
      "epoch": 0.9781154689924555,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004914472907128189,
      "loss": 6.2691,
      "mean_token_accuracy": 0.14372571781277657,
      "num_tokens": 19787875.0,
      "step": 9140
    },
    {
      "entropy": 6.494171762466431,
      "epoch": 0.9786505431002194,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004914368463684876,
      "loss": 6.1505,
      "mean_token_accuracy": 0.14933271259069442,
      "num_tokens": 19798919.0,
      "step": 9145
    },
    {
      "entropy": 6.470934581756592,
      "epoch": 0.9791856172079834,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004914263957745726,
      "loss": 6.136,
      "mean_token_accuracy": 0.14965204894542694,
      "num_tokens": 19809320.0,
      "step": 9150
    },
    {
      "entropy": 6.505920267105102,
      "epoch": 0.9797206913157472,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004914159389313759,
      "loss": 6.207,
      "mean_token_accuracy": 0.14603523388504983,
      "num_tokens": 19819511.0,
      "step": 9155
    },
    {
      "entropy": 6.499664258956909,
      "epoch": 0.9802557654235111,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004914054758391994,
      "loss": 6.2488,
      "mean_token_accuracy": 0.1425292134284973,
      "num_tokens": 19830381.0,
      "step": 9160
    },
    {
      "entropy": 6.363729572296142,
      "epoch": 0.9807908395312751,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004913950064983452,
      "loss": 6.1127,
      "mean_token_accuracy": 0.14317781180143357,
      "num_tokens": 19840806.0,
      "step": 9165
    },
    {
      "entropy": 6.396924781799316,
      "epoch": 0.981325913639039,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004913845309091156,
      "loss": 6.1143,
      "mean_token_accuracy": 0.1530517816543579,
      "num_tokens": 19851095.0,
      "step": 9170
    },
    {
      "entropy": 6.435604286193848,
      "epoch": 0.981860987746803,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004913740490718131,
      "loss": 6.079,
      "mean_token_accuracy": 0.15113799571990966,
      "num_tokens": 19860235.0,
      "step": 9175
    },
    {
      "entropy": 6.397189712524414,
      "epoch": 0.9823960618545668,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004913635609867403,
      "loss": 6.1705,
      "mean_token_accuracy": 0.13944099545478822,
      "num_tokens": 19872264.0,
      "step": 9180
    },
    {
      "entropy": 6.483842229843139,
      "epoch": 0.9829311359623308,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004913530666542001,
      "loss": 6.262,
      "mean_token_accuracy": 0.13980590477585791,
      "num_tokens": 19883628.0,
      "step": 9185
    },
    {
      "entropy": 6.503321313858033,
      "epoch": 0.9834662100700947,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004913425660744955,
      "loss": 6.1548,
      "mean_token_accuracy": 0.1449754886329174,
      "num_tokens": 19894624.0,
      "step": 9190
    },
    {
      "entropy": 6.501814842224121,
      "epoch": 0.9840012841778586,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004913320592479297,
      "loss": 6.1622,
      "mean_token_accuracy": 0.1453083112835884,
      "num_tokens": 19904882.0,
      "step": 9195
    },
    {
      "entropy": 6.426853370666504,
      "epoch": 0.9845363582856226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004913215461748061,
      "loss": 6.1122,
      "mean_token_accuracy": 0.15256331562995912,
      "num_tokens": 19915344.0,
      "step": 9200
    },
    {
      "entropy": 6.419813108444214,
      "epoch": 0.9850714323933865,
      "grad_norm": 0.8046875,
      "learning_rate": 0.0004913110268554282,
      "loss": 6.2431,
      "mean_token_accuracy": 0.13632122725248336,
      "num_tokens": 19927527.0,
      "step": 9205
    },
    {
      "entropy": 6.43550968170166,
      "epoch": 0.9856065065011504,
      "grad_norm": 0.875,
      "learning_rate": 0.0004913005012900998,
      "loss": 6.1881,
      "mean_token_accuracy": 0.1410766676068306,
      "num_tokens": 19938893.0,
      "step": 9210
    },
    {
      "entropy": 6.4580189228057865,
      "epoch": 0.9861415806089143,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004912899694791248,
      "loss": 6.1423,
      "mean_token_accuracy": 0.14898031800985337,
      "num_tokens": 19949297.0,
      "step": 9215
    },
    {
      "entropy": 6.426213169097901,
      "epoch": 0.9866766547166783,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004912794314228074,
      "loss": 6.1286,
      "mean_token_accuracy": 0.1483857363462448,
      "num_tokens": 19959194.0,
      "step": 9220
    },
    {
      "entropy": 6.472116231918335,
      "epoch": 0.9872117288244422,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004912688871214518,
      "loss": 6.1711,
      "mean_token_accuracy": 0.14535824656486512,
      "num_tokens": 19969896.0,
      "step": 9225
    },
    {
      "entropy": 6.479356098175049,
      "epoch": 0.9877468029322061,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004912583365753625,
      "loss": 6.1511,
      "mean_token_accuracy": 0.14686021953821182,
      "num_tokens": 19981505.0,
      "step": 9230
    },
    {
      "entropy": 6.457101821899414,
      "epoch": 0.9882818770399701,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004912477797848441,
      "loss": 6.1263,
      "mean_token_accuracy": 0.15228600800037384,
      "num_tokens": 19992770.0,
      "step": 9235
    },
    {
      "entropy": 6.4596668720245365,
      "epoch": 0.9888169511477339,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004912372167502014,
      "loss": 6.1378,
      "mean_token_accuracy": 0.14391888231039046,
      "num_tokens": 20003572.0,
      "step": 9240
    },
    {
      "entropy": 6.404411506652832,
      "epoch": 0.9893520252554979,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004912266474717394,
      "loss": 6.1672,
      "mean_token_accuracy": 0.14781463146209717,
      "num_tokens": 20014500.0,
      "step": 9245
    },
    {
      "entropy": 6.466900205612182,
      "epoch": 0.9898870993632618,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004912160719497635,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1432821810245514,
      "num_tokens": 20025745.0,
      "step": 9250
    },
    {
      "entropy": 6.4329530715942385,
      "epoch": 0.9904221734710258,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004912054901845788,
      "loss": 6.0478,
      "mean_token_accuracy": 0.14896151423454285,
      "num_tokens": 20036346.0,
      "step": 9255
    },
    {
      "entropy": 6.379064083099365,
      "epoch": 0.9909572475787897,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000491194902176491,
      "loss": 6.1982,
      "mean_token_accuracy": 0.14473242461681365,
      "num_tokens": 20048162.0,
      "step": 9260
    },
    {
      "entropy": 6.446938228607178,
      "epoch": 0.9914923216865535,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004911843079258057,
      "loss": 6.2876,
      "mean_token_accuracy": 0.13625169321894645,
      "num_tokens": 20058947.0,
      "step": 9265
    },
    {
      "entropy": 6.563446998596191,
      "epoch": 0.9920273957943175,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004911737074328289,
      "loss": 6.2142,
      "mean_token_accuracy": 0.14483553171157837,
      "num_tokens": 20069282.0,
      "step": 9270
    },
    {
      "entropy": 6.430431079864502,
      "epoch": 0.9925624699020814,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004911631006978667,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1452360913157463,
      "num_tokens": 20078347.0,
      "step": 9275
    },
    {
      "entropy": 6.431726169586182,
      "epoch": 0.9930975440098454,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004911524877212254,
      "loss": 6.2,
      "mean_token_accuracy": 0.14119307920336724,
      "num_tokens": 20089129.0,
      "step": 9280
    },
    {
      "entropy": 6.492161798477173,
      "epoch": 0.9936326181176093,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004911418685032114,
      "loss": 6.1975,
      "mean_token_accuracy": 0.14033432081341743,
      "num_tokens": 20101008.0,
      "step": 9285
    },
    {
      "entropy": 6.388562440872192,
      "epoch": 0.9941676922253733,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004911312430441312,
      "loss": 6.1917,
      "mean_token_accuracy": 0.1472950503230095,
      "num_tokens": 20112272.0,
      "step": 9290
    },
    {
      "entropy": 6.461758422851562,
      "epoch": 0.9947027663331371,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004911206113442918,
      "loss": 6.0834,
      "mean_token_accuracy": 0.15697775781154633,
      "num_tokens": 20122814.0,
      "step": 9295
    },
    {
      "entropy": 6.401407289505005,
      "epoch": 0.995237840440901,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004911099734040001,
      "loss": 6.0537,
      "mean_token_accuracy": 0.14913687855005264,
      "num_tokens": 20133169.0,
      "step": 9300
    },
    {
      "entropy": 6.461355257034302,
      "epoch": 0.995772914548665,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004910993292235634,
      "loss": 6.193,
      "mean_token_accuracy": 0.13840402215719222,
      "num_tokens": 20144355.0,
      "step": 9305
    },
    {
      "entropy": 6.4919617652893065,
      "epoch": 0.9963079886564289,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004910886788032887,
      "loss": 6.1984,
      "mean_token_accuracy": 0.13854726180434226,
      "num_tokens": 20154486.0,
      "step": 9310
    },
    {
      "entropy": 6.449913454055786,
      "epoch": 0.9968430627641929,
      "grad_norm": 0.81640625,
      "learning_rate": 0.0004910780221434839,
      "loss": 6.1459,
      "mean_token_accuracy": 0.1506755456328392,
      "num_tokens": 20166359.0,
      "step": 9315
    },
    {
      "entropy": 6.434816360473633,
      "epoch": 0.9973781368719568,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004910673592444566,
      "loss": 6.1642,
      "mean_token_accuracy": 0.1463884525001049,
      "num_tokens": 20176469.0,
      "step": 9320
    },
    {
      "entropy": 6.43997974395752,
      "epoch": 0.9979132109797207,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004910566901065147,
      "loss": 6.1962,
      "mean_token_accuracy": 0.1458169311285019,
      "num_tokens": 20188334.0,
      "step": 9325
    },
    {
      "entropy": 6.4144402027130125,
      "epoch": 0.9984482850874846,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004910460147299661,
      "loss": 6.101,
      "mean_token_accuracy": 0.15269924700260162,
      "num_tokens": 20198594.0,
      "step": 9330
    },
    {
      "entropy": 6.422747039794922,
      "epoch": 0.9989833591952485,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004910353331151193,
      "loss": 6.2026,
      "mean_token_accuracy": 0.14397644847631455,
      "num_tokens": 20209456.0,
      "step": 9335
    },
    {
      "entropy": 6.396505403518677,
      "epoch": 0.9995184333030125,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004910246452622825,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14899104088544846,
      "num_tokens": 20219835.0,
      "step": 9340
    },
    {
      "entropy": 6.544154008229573,
      "epoch": 1.0,
      "grad_norm": 1.8671875,
      "learning_rate": 0.0004910139511717646,
      "loss": 6.2128,
      "mean_token_accuracy": 0.14114661680327523,
      "num_tokens": 20228258.0,
      "step": 9345
    },
    {
      "entropy": 6.45707426071167,
      "epoch": 1.0005350741077639,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004910032508438739,
      "loss": 6.0695,
      "mean_token_accuracy": 0.1519175037741661,
      "num_tokens": 20238875.0,
      "step": 9350
    },
    {
      "entropy": 6.488657760620117,
      "epoch": 1.0010701482155278,
      "grad_norm": 0.875,
      "learning_rate": 0.00049099254427892,
      "loss": 6.1027,
      "mean_token_accuracy": 0.13913752511143684,
      "num_tokens": 20249750.0,
      "step": 9355
    },
    {
      "entropy": 6.55254979133606,
      "epoch": 1.0016052223232919,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004909818314772116,
      "loss": 6.13,
      "mean_token_accuracy": 0.13821920976042748,
      "num_tokens": 20259772.0,
      "step": 9360
    },
    {
      "entropy": 6.397948837280273,
      "epoch": 1.0021402964310557,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909711124390582,
      "loss": 6.0467,
      "mean_token_accuracy": 0.15047429502010345,
      "num_tokens": 20270230.0,
      "step": 9365
    },
    {
      "entropy": 6.337224769592285,
      "epoch": 1.0026753705388196,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004909603871647692,
      "loss": 5.8949,
      "mean_token_accuracy": 0.16912632584571838,
      "num_tokens": 20280317.0,
      "step": 9370
    },
    {
      "entropy": 6.389275026321411,
      "epoch": 1.0032104446465835,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004909496556546544,
      "loss": 5.988,
      "mean_token_accuracy": 0.15474062114953996,
      "num_tokens": 20291147.0,
      "step": 9375
    },
    {
      "entropy": 6.4034810066223145,
      "epoch": 1.0037455187543476,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004909389179090238,
      "loss": 6.0441,
      "mean_token_accuracy": 0.15618184059858323,
      "num_tokens": 20301095.0,
      "step": 9380
    },
    {
      "entropy": 6.422483968734741,
      "epoch": 1.0042805928621115,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004909281739281871,
      "loss": 6.0625,
      "mean_token_accuracy": 0.1534925326704979,
      "num_tokens": 20311216.0,
      "step": 9385
    },
    {
      "entropy": 6.495018529891968,
      "epoch": 1.0048156669698753,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004909174237124548,
      "loss": 6.1216,
      "mean_token_accuracy": 0.14018214344978333,
      "num_tokens": 20321719.0,
      "step": 9390
    },
    {
      "entropy": 6.4697959423065186,
      "epoch": 1.0053507410776392,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004909066672621372,
      "loss": 6.1141,
      "mean_token_accuracy": 0.1479663632810116,
      "num_tokens": 20331997.0,
      "step": 9395
    },
    {
      "entropy": 6.416800117492675,
      "epoch": 1.005885815185403,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004908959045775449,
      "loss": 6.158,
      "mean_token_accuracy": 0.14297408759593963,
      "num_tokens": 20342973.0,
      "step": 9400
    },
    {
      "entropy": 6.424854278564453,
      "epoch": 1.0064208892931672,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004908851356589887,
      "loss": 6.0485,
      "mean_token_accuracy": 0.15194151401519776,
      "num_tokens": 20352652.0,
      "step": 9405
    },
    {
      "entropy": 6.466174459457397,
      "epoch": 1.006955963400931,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0004908743605067796,
      "loss": 6.1054,
      "mean_token_accuracy": 0.14904245138168334,
      "num_tokens": 20364577.0,
      "step": 9410
    },
    {
      "entropy": 6.453578186035156,
      "epoch": 1.007491037508695,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004908635791212287,
      "loss": 6.1632,
      "mean_token_accuracy": 0.14845566898584367,
      "num_tokens": 20375451.0,
      "step": 9415
    },
    {
      "entropy": 6.412635469436646,
      "epoch": 1.0080261116164588,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004908527915026472,
      "loss": 6.0725,
      "mean_token_accuracy": 0.14759569838643075,
      "num_tokens": 20385174.0,
      "step": 9420
    },
    {
      "entropy": 6.440878963470459,
      "epoch": 1.0085611857242227,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004908419976513467,
      "loss": 6.005,
      "mean_token_accuracy": 0.1531747967004776,
      "num_tokens": 20395710.0,
      "step": 9425
    },
    {
      "entropy": 6.400674343109131,
      "epoch": 1.0090962598319868,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004908311975676388,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15267980545759202,
      "num_tokens": 20406572.0,
      "step": 9430
    },
    {
      "entropy": 6.371784019470215,
      "epoch": 1.0096313339397507,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004908203912518355,
      "loss": 6.0186,
      "mean_token_accuracy": 0.15572094172239304,
      "num_tokens": 20417786.0,
      "step": 9435
    },
    {
      "entropy": 6.462334108352661,
      "epoch": 1.0101664080475146,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004908095787042487,
      "loss": 6.0396,
      "mean_token_accuracy": 0.15110276639461517,
      "num_tokens": 20427680.0,
      "step": 9440
    },
    {
      "entropy": 6.384920072555542,
      "epoch": 1.0107014821552784,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004907987599251907,
      "loss": 6.0417,
      "mean_token_accuracy": 0.15148375183343887,
      "num_tokens": 20438640.0,
      "step": 9445
    },
    {
      "entropy": 6.421800231933593,
      "epoch": 1.0112365562630425,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004907879349149737,
      "loss": 6.0384,
      "mean_token_accuracy": 0.15655026361346244,
      "num_tokens": 20449939.0,
      "step": 9450
    },
    {
      "entropy": 6.382502889633178,
      "epoch": 1.0117716303708064,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004907771036739105,
      "loss": 5.9766,
      "mean_token_accuracy": 0.1538733087480068,
      "num_tokens": 20460845.0,
      "step": 9455
    },
    {
      "entropy": 6.4499505996704105,
      "epoch": 1.0123067044785703,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004907662662023137,
      "loss": 6.1259,
      "mean_token_accuracy": 0.14487328231334687,
      "num_tokens": 20472407.0,
      "step": 9460
    },
    {
      "entropy": 6.467405939102173,
      "epoch": 1.0128417785863342,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004907554225004962,
      "loss": 6.122,
      "mean_token_accuracy": 0.14684601947665216,
      "num_tokens": 20483192.0,
      "step": 9465
    },
    {
      "entropy": 6.473992156982422,
      "epoch": 1.013376852694098,
      "grad_norm": 0.875,
      "learning_rate": 0.0004907445725687714,
      "loss": 6.1922,
      "mean_token_accuracy": 0.1419432170689106,
      "num_tokens": 20493498.0,
      "step": 9470
    },
    {
      "entropy": 6.419417190551758,
      "epoch": 1.0139119268018622,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004907337164074523,
      "loss": 6.0895,
      "mean_token_accuracy": 0.15110560208559037,
      "num_tokens": 20503781.0,
      "step": 9475
    },
    {
      "entropy": 6.3973390579223635,
      "epoch": 1.014447000909626,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004907228540168523,
      "loss": 6.145,
      "mean_token_accuracy": 0.14667920246720315,
      "num_tokens": 20515142.0,
      "step": 9480
    },
    {
      "entropy": 6.460160827636718,
      "epoch": 1.01498207501739,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004907119853972854,
      "loss": 6.1105,
      "mean_token_accuracy": 0.1487506926059723,
      "num_tokens": 20527568.0,
      "step": 9485
    },
    {
      "entropy": 6.370030069351197,
      "epoch": 1.0155171491251538,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004907011105490651,
      "loss": 6.0245,
      "mean_token_accuracy": 0.15469107478857042,
      "num_tokens": 20538198.0,
      "step": 9490
    },
    {
      "entropy": 6.445659017562866,
      "epoch": 1.0160522232329177,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004906902294725056,
      "loss": 5.9996,
      "mean_token_accuracy": 0.15493054389953614,
      "num_tokens": 20549355.0,
      "step": 9495
    },
    {
      "entropy": 6.407544088363648,
      "epoch": 1.0165872973406818,
      "grad_norm": 1.0,
      "learning_rate": 0.000490679342167921,
      "loss": 6.1166,
      "mean_token_accuracy": 0.14722240716218948,
      "num_tokens": 20558984.0,
      "step": 9500
    },
    {
      "entropy": 6.4405745506286625,
      "epoch": 1.0171223714484456,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004906684486356258,
      "loss": 6.1206,
      "mean_token_accuracy": 0.1429874800145626,
      "num_tokens": 20570143.0,
      "step": 9505
    },
    {
      "entropy": 6.440524482727051,
      "epoch": 1.0176574455562095,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004906575488759343,
      "loss": 6.13,
      "mean_token_accuracy": 0.14288547039031982,
      "num_tokens": 20580973.0,
      "step": 9510
    },
    {
      "entropy": 6.479393720626831,
      "epoch": 1.0181925196639734,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004906466428891616,
      "loss": 6.1014,
      "mean_token_accuracy": 0.14325934648513794,
      "num_tokens": 20591473.0,
      "step": 9515
    },
    {
      "entropy": 6.4329283237457275,
      "epoch": 1.0187275937717375,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004906357306756222,
      "loss": 6.0451,
      "mean_token_accuracy": 0.14811489060521127,
      "num_tokens": 20602347.0,
      "step": 9520
    },
    {
      "entropy": 6.465449523925781,
      "epoch": 1.0192626678795014,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004906248122356315,
      "loss": 6.1187,
      "mean_token_accuracy": 0.15244316533207894,
      "num_tokens": 20613168.0,
      "step": 9525
    },
    {
      "entropy": 6.390532112121582,
      "epoch": 1.0197977419872652,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004906138875695045,
      "loss": 6.0004,
      "mean_token_accuracy": 0.1514841079711914,
      "num_tokens": 20623523.0,
      "step": 9530
    },
    {
      "entropy": 6.336051368713379,
      "epoch": 1.0203328160950291,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004906029566775569,
      "loss": 6.0354,
      "mean_token_accuracy": 0.15277280956506728,
      "num_tokens": 20634085.0,
      "step": 9535
    },
    {
      "entropy": 6.456353998184204,
      "epoch": 1.020867890202793,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004905920195601042,
      "loss": 6.1026,
      "mean_token_accuracy": 0.14824522584676741,
      "num_tokens": 20644747.0,
      "step": 9540
    },
    {
      "entropy": 6.445295906066894,
      "epoch": 1.021402964310557,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004905810762174622,
      "loss": 6.0449,
      "mean_token_accuracy": 0.14302776530385017,
      "num_tokens": 20655704.0,
      "step": 9545
    },
    {
      "entropy": 6.453642654418945,
      "epoch": 1.021938038418321,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004905701266499469,
      "loss": 6.0483,
      "mean_token_accuracy": 0.156907856464386,
      "num_tokens": 20665844.0,
      "step": 9550
    },
    {
      "entropy": 6.45090389251709,
      "epoch": 1.0224731125260849,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004905591708578746,
      "loss": 6.1766,
      "mean_token_accuracy": 0.1401354804635048,
      "num_tokens": 20676921.0,
      "step": 9555
    },
    {
      "entropy": 6.469284963607788,
      "epoch": 1.0230081866338487,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004905482088415615,
      "loss": 6.1143,
      "mean_token_accuracy": 0.14324075654149054,
      "num_tokens": 20688476.0,
      "step": 9560
    },
    {
      "entropy": 6.459663772583008,
      "epoch": 1.0235432607416126,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004905372406013241,
      "loss": 6.1118,
      "mean_token_accuracy": 0.1481221318244934,
      "num_tokens": 20698588.0,
      "step": 9565
    },
    {
      "entropy": 6.414258289337158,
      "epoch": 1.0240783348493767,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004905262661374792,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14821909964084626,
      "num_tokens": 20709181.0,
      "step": 9570
    },
    {
      "entropy": 6.410443258285523,
      "epoch": 1.0246134089571406,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004905152854503436,
      "loss": 6.0265,
      "mean_token_accuracy": 0.15928781181573867,
      "num_tokens": 20719724.0,
      "step": 9575
    },
    {
      "entropy": 6.4836194038391115,
      "epoch": 1.0251484830649045,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004905042985402345,
      "loss": 6.1244,
      "mean_token_accuracy": 0.1476878434419632,
      "num_tokens": 20730548.0,
      "step": 9580
    },
    {
      "entropy": 6.4444591999053955,
      "epoch": 1.0256835571726683,
      "grad_norm": 0.8828125,
      "learning_rate": 0.000490493305407469,
      "loss": 6.1096,
      "mean_token_accuracy": 0.14357411563396455,
      "num_tokens": 20742124.0,
      "step": 9585
    },
    {
      "entropy": 6.498137474060059,
      "epoch": 1.0262186312804324,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004904823060523646,
      "loss": 6.1732,
      "mean_token_accuracy": 0.14311810359358787,
      "num_tokens": 20753313.0,
      "step": 9590
    },
    {
      "entropy": 6.492930173873901,
      "epoch": 1.0267537053881963,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004904713004752389,
      "loss": 6.1291,
      "mean_token_accuracy": 0.15013999938964845,
      "num_tokens": 20763539.0,
      "step": 9595
    },
    {
      "entropy": 6.434520149230957,
      "epoch": 1.0272887794959602,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004904602886764097,
      "loss": 6.21,
      "mean_token_accuracy": 0.14225629568099976,
      "num_tokens": 20775849.0,
      "step": 9600
    },
    {
      "entropy": 6.417896127700805,
      "epoch": 1.027823853603724,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004904492706561948,
      "loss": 6.0924,
      "mean_token_accuracy": 0.1449787475168705,
      "num_tokens": 20787460.0,
      "step": 9605
    },
    {
      "entropy": 6.53659200668335,
      "epoch": 1.028358927711488,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004904382464149127,
      "loss": 6.0114,
      "mean_token_accuracy": 0.15352628827095033,
      "num_tokens": 20798478.0,
      "step": 9610
    },
    {
      "entropy": 6.506520175933838,
      "epoch": 1.028894001819252,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004904272159528814,
      "loss": 6.106,
      "mean_token_accuracy": 0.1513581931591034,
      "num_tokens": 20809730.0,
      "step": 9615
    },
    {
      "entropy": 6.368850517272949,
      "epoch": 1.029429075927016,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004904161792704195,
      "loss": 6.109,
      "mean_token_accuracy": 0.1596040368080139,
      "num_tokens": 20821438.0,
      "step": 9620
    },
    {
      "entropy": 6.427107143402099,
      "epoch": 1.0299641500347798,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004904051363678457,
      "loss": 6.08,
      "mean_token_accuracy": 0.1430250570178032,
      "num_tokens": 20832328.0,
      "step": 9625
    },
    {
      "entropy": 6.392775964736939,
      "epoch": 1.0304992241425437,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004903940872454789,
      "loss": 6.0616,
      "mean_token_accuracy": 0.14926150143146516,
      "num_tokens": 20843956.0,
      "step": 9630
    },
    {
      "entropy": 6.488288164138794,
      "epoch": 1.0310342982503076,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004903830319036381,
      "loss": 6.0901,
      "mean_token_accuracy": 0.15633318722248077,
      "num_tokens": 20854578.0,
      "step": 9635
    },
    {
      "entropy": 6.402007007598877,
      "epoch": 1.0315693723580717,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004903719703426425,
      "loss": 6.0171,
      "mean_token_accuracy": 0.14938449412584304,
      "num_tokens": 20865190.0,
      "step": 9640
    },
    {
      "entropy": 6.443171644210816,
      "epoch": 1.0321044464658355,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004903609025628115,
      "loss": 6.0682,
      "mean_token_accuracy": 0.1459045395255089,
      "num_tokens": 20875733.0,
      "step": 9645
    },
    {
      "entropy": 6.381960773468018,
      "epoch": 1.0326395205735994,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004903498285644647,
      "loss": 6.0726,
      "mean_token_accuracy": 0.14591895639896393,
      "num_tokens": 20886900.0,
      "step": 9650
    },
    {
      "entropy": 6.45991883277893,
      "epoch": 1.0331745946813633,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004903387483479219,
      "loss": 6.1333,
      "mean_token_accuracy": 0.145784579962492,
      "num_tokens": 20898125.0,
      "step": 9655
    },
    {
      "entropy": 6.5207103252410885,
      "epoch": 1.0337096687891274,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004903276619135029,
      "loss": 6.1563,
      "mean_token_accuracy": 0.14228807538747787,
      "num_tokens": 20909788.0,
      "step": 9660
    },
    {
      "entropy": 6.45972318649292,
      "epoch": 1.0342447428968913,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000490316569261528,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14990933537483214,
      "num_tokens": 20921853.0,
      "step": 9665
    },
    {
      "entropy": 6.455929327011108,
      "epoch": 1.0347798170046552,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004903054703923175,
      "loss": 6.1131,
      "mean_token_accuracy": 0.13926180005073546,
      "num_tokens": 20932574.0,
      "step": 9670
    },
    {
      "entropy": 6.416140031814575,
      "epoch": 1.035314891112419,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004902943653061916,
      "loss": 6.0817,
      "mean_token_accuracy": 0.14819479882717132,
      "num_tokens": 20943458.0,
      "step": 9675
    },
    {
      "entropy": 6.456738138198853,
      "epoch": 1.035849965220183,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004902832540034712,
      "loss": 6.0442,
      "mean_token_accuracy": 0.15593548864126205,
      "num_tokens": 20953213.0,
      "step": 9680
    },
    {
      "entropy": 6.447682666778564,
      "epoch": 1.036385039327947,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004902721364844773,
      "loss": 6.0844,
      "mean_token_accuracy": 0.1495116487145424,
      "num_tokens": 20963885.0,
      "step": 9685
    },
    {
      "entropy": 6.3958375453948975,
      "epoch": 1.0369201134357109,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004902610127495305,
      "loss": 6.0906,
      "mean_token_accuracy": 0.14940711706876755,
      "num_tokens": 20975529.0,
      "step": 9690
    },
    {
      "entropy": 6.407362413406372,
      "epoch": 1.0374551875434748,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004902498827989523,
      "loss": 6.0379,
      "mean_token_accuracy": 0.15741287767887116,
      "num_tokens": 20985279.0,
      "step": 9695
    },
    {
      "entropy": 6.491542053222656,
      "epoch": 1.0379902616512386,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004902387466330639,
      "loss": 6.1023,
      "mean_token_accuracy": 0.14826459288597107,
      "num_tokens": 20995818.0,
      "step": 9700
    },
    {
      "entropy": 6.428172922134399,
      "epoch": 1.0385253357590025,
      "grad_norm": 0.921875,
      "learning_rate": 0.000490227604252187,
      "loss": 6.0285,
      "mean_token_accuracy": 0.14653588086366653,
      "num_tokens": 21006786.0,
      "step": 9705
    },
    {
      "entropy": 6.4387706279754635,
      "epoch": 1.0390604098667666,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004902164556566433,
      "loss": 6.0824,
      "mean_token_accuracy": 0.15597511380910872,
      "num_tokens": 21017441.0,
      "step": 9710
    },
    {
      "entropy": 6.307831192016602,
      "epoch": 1.0395954839745305,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004902053008467546,
      "loss": 5.9689,
      "mean_token_accuracy": 0.15480395182967185,
      "num_tokens": 21027812.0,
      "step": 9715
    },
    {
      "entropy": 6.4775238037109375,
      "epoch": 1.0401305580822944,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004901941398228431,
      "loss": 6.1265,
      "mean_token_accuracy": 0.1492430344223976,
      "num_tokens": 21038960.0,
      "step": 9720
    },
    {
      "entropy": 6.484867095947266,
      "epoch": 1.0406656321900583,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004901829725852311,
      "loss": 6.0881,
      "mean_token_accuracy": 0.15482718497514725,
      "num_tokens": 21050467.0,
      "step": 9725
    },
    {
      "entropy": 6.385688924789429,
      "epoch": 1.0412007062978224,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000490171799134241,
      "loss": 6.0058,
      "mean_token_accuracy": 0.148017618060112,
      "num_tokens": 21061208.0,
      "step": 9730
    },
    {
      "entropy": 6.417812156677246,
      "epoch": 1.0417357804055862,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004901606194701954,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1493859887123108,
      "num_tokens": 21072872.0,
      "step": 9735
    },
    {
      "entropy": 6.49041428565979,
      "epoch": 1.04227085451335,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004901494335934172,
      "loss": 6.0684,
      "mean_token_accuracy": 0.15393005311489105,
      "num_tokens": 21084878.0,
      "step": 9740
    },
    {
      "entropy": 6.501065063476562,
      "epoch": 1.042805928621114,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004901382415042293,
      "loss": 6.1239,
      "mean_token_accuracy": 0.14754006862640381,
      "num_tokens": 21095289.0,
      "step": 9745
    },
    {
      "entropy": 6.461180543899536,
      "epoch": 1.0433410027288779,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004901270432029549,
      "loss": 6.0764,
      "mean_token_accuracy": 0.14892422109842302,
      "num_tokens": 21106077.0,
      "step": 9750
    },
    {
      "entropy": 6.337989759445191,
      "epoch": 1.043876076836642,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004901158386899174,
      "loss": 5.9935,
      "mean_token_accuracy": 0.15678749233484268,
      "num_tokens": 21116548.0,
      "step": 9755
    },
    {
      "entropy": 6.388410520553589,
      "epoch": 1.0444111509444058,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004901046279654402,
      "loss": 6.0207,
      "mean_token_accuracy": 0.16142363250255584,
      "num_tokens": 21126163.0,
      "step": 9760
    },
    {
      "entropy": 6.4837563037872314,
      "epoch": 1.0449462250521697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004900934110298471,
      "loss": 6.1429,
      "mean_token_accuracy": 0.14407059773802758,
      "num_tokens": 21136830.0,
      "step": 9765
    },
    {
      "entropy": 6.482270765304565,
      "epoch": 1.0454812991599336,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004900821878834621,
      "loss": 6.0542,
      "mean_token_accuracy": 0.15313467383384705,
      "num_tokens": 21146793.0,
      "step": 9770
    },
    {
      "entropy": 6.424727058410644,
      "epoch": 1.0460163732676975,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000490070958526609,
      "loss": 6.126,
      "mean_token_accuracy": 0.1466248482465744,
      "num_tokens": 21156799.0,
      "step": 9775
    },
    {
      "entropy": 6.39983811378479,
      "epoch": 1.0465514473754616,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004900597229596124,
      "loss": 6.1172,
      "mean_token_accuracy": 0.14409211352467538,
      "num_tokens": 21167757.0,
      "step": 9780
    },
    {
      "entropy": 6.495046710968017,
      "epoch": 1.0470865214832255,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004900484811827963,
      "loss": 6.1084,
      "mean_token_accuracy": 0.1545184776186943,
      "num_tokens": 21178832.0,
      "step": 9785
    },
    {
      "entropy": 6.407345342636108,
      "epoch": 1.0476215955909893,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004900372331964856,
      "loss": 6.0139,
      "mean_token_accuracy": 0.1554933547973633,
      "num_tokens": 21189637.0,
      "step": 9790
    },
    {
      "entropy": 6.372892141342163,
      "epoch": 1.0481566696987532,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000490025979001005,
      "loss": 6.0427,
      "mean_token_accuracy": 0.15000387877225876,
      "num_tokens": 21200194.0,
      "step": 9795
    },
    {
      "entropy": 6.3619026184082035,
      "epoch": 1.0486917438065173,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004900147185966795,
      "loss": 6.0009,
      "mean_token_accuracy": 0.15102414786815643,
      "num_tokens": 21211405.0,
      "step": 9800
    },
    {
      "entropy": 6.437906265258789,
      "epoch": 1.0492268179142812,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004900034519838342,
      "loss": 6.0514,
      "mean_token_accuracy": 0.14703138917684555,
      "num_tokens": 21222722.0,
      "step": 9805
    },
    {
      "entropy": 6.393271970748901,
      "epoch": 1.049761892022045,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004899921791627945,
      "loss": 6.008,
      "mean_token_accuracy": 0.1607884645462036,
      "num_tokens": 21233510.0,
      "step": 9810
    },
    {
      "entropy": 6.452353286743164,
      "epoch": 1.050296966129809,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004899809001338857,
      "loss": 6.1611,
      "mean_token_accuracy": 0.13560899049043657,
      "num_tokens": 21243235.0,
      "step": 9815
    },
    {
      "entropy": 6.439401531219483,
      "epoch": 1.0508320402375728,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004899696148974338,
      "loss": 6.0679,
      "mean_token_accuracy": 0.15340595170855523,
      "num_tokens": 21254680.0,
      "step": 9820
    },
    {
      "entropy": 6.427839517593384,
      "epoch": 1.051367114345337,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004899583234537644,
      "loss": 6.0567,
      "mean_token_accuracy": 0.15005040615797044,
      "num_tokens": 21265116.0,
      "step": 9825
    },
    {
      "entropy": 6.423818683624267,
      "epoch": 1.0519021884531008,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004899470258032034,
      "loss": 6.0282,
      "mean_token_accuracy": 0.15112557560205458,
      "num_tokens": 21277066.0,
      "step": 9830
    },
    {
      "entropy": 6.444361448287964,
      "epoch": 1.0524372625608647,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004899357219460775,
      "loss": 6.0969,
      "mean_token_accuracy": 0.14964328855276107,
      "num_tokens": 21287332.0,
      "step": 9835
    },
    {
      "entropy": 6.417037725448608,
      "epoch": 1.0529723366686286,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004899244118827128,
      "loss": 6.083,
      "mean_token_accuracy": 0.14961449801921844,
      "num_tokens": 21298377.0,
      "step": 9840
    },
    {
      "entropy": 6.472538852691651,
      "epoch": 1.0535074107763924,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004899130956134358,
      "loss": 6.0702,
      "mean_token_accuracy": 0.1478628024458885,
      "num_tokens": 21309966.0,
      "step": 9845
    },
    {
      "entropy": 6.37747106552124,
      "epoch": 1.0540424848841565,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004899017731385735,
      "loss": 5.9809,
      "mean_token_accuracy": 0.16066834628582,
      "num_tokens": 21320399.0,
      "step": 9850
    },
    {
      "entropy": 6.415247440338135,
      "epoch": 1.0545775589919204,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004898904444584527,
      "loss": 6.094,
      "mean_token_accuracy": 0.14864722043275833,
      "num_tokens": 21331716.0,
      "step": 9855
    },
    {
      "entropy": 6.475224733352661,
      "epoch": 1.0551126330996843,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004898791095734004,
      "loss": 6.1132,
      "mean_token_accuracy": 0.14484637528657912,
      "num_tokens": 21341664.0,
      "step": 9860
    },
    {
      "entropy": 6.441167688369751,
      "epoch": 1.0556477072074482,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000489867768483744,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14595111310482026,
      "num_tokens": 21352220.0,
      "step": 9865
    },
    {
      "entropy": 6.472434234619141,
      "epoch": 1.0561827813152123,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004898564211898111,
      "loss": 6.1762,
      "mean_token_accuracy": 0.1460395909845829,
      "num_tokens": 21363947.0,
      "step": 9870
    },
    {
      "entropy": 6.50465030670166,
      "epoch": 1.0567178554229761,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004898450676919291,
      "loss": 6.0466,
      "mean_token_accuracy": 0.15059653520584107,
      "num_tokens": 21374011.0,
      "step": 9875
    },
    {
      "entropy": 6.432668781280517,
      "epoch": 1.05725292953074,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000489833707990426,
      "loss": 6.1043,
      "mean_token_accuracy": 0.14760383665561677,
      "num_tokens": 21384232.0,
      "step": 9880
    },
    {
      "entropy": 6.411658668518067,
      "epoch": 1.057788003638504,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004898223420856298,
      "loss": 6.0748,
      "mean_token_accuracy": 0.1460980847477913,
      "num_tokens": 21395928.0,
      "step": 9885
    },
    {
      "entropy": 6.411401748657227,
      "epoch": 1.0583230777462678,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004898109699778686,
      "loss": 6.0739,
      "mean_token_accuracy": 0.15150696486234666,
      "num_tokens": 21406391.0,
      "step": 9890
    },
    {
      "entropy": 6.4182164669036865,
      "epoch": 1.0588581518540319,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000489799591667471,
      "loss": 6.1218,
      "mean_token_accuracy": 0.14150176271796228,
      "num_tokens": 21417859.0,
      "step": 9895
    },
    {
      "entropy": 6.422651481628418,
      "epoch": 1.0593932259617957,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004897882071547651,
      "loss": 5.9764,
      "mean_token_accuracy": 0.1521952971816063,
      "num_tokens": 21428284.0,
      "step": 9900
    },
    {
      "entropy": 6.4054419040679935,
      "epoch": 1.0599283000695596,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004897768164400801,
      "loss": 6.1287,
      "mean_token_accuracy": 0.1415218196809292,
      "num_tokens": 21439610.0,
      "step": 9905
    },
    {
      "entropy": 6.463895034790039,
      "epoch": 1.0604633741773235,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004897654195237446,
      "loss": 6.1507,
      "mean_token_accuracy": 0.14277879372239113,
      "num_tokens": 21449319.0,
      "step": 9910
    },
    {
      "entropy": 6.455616092681884,
      "epoch": 1.0609984482850874,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004897540164060879,
      "loss": 6.1315,
      "mean_token_accuracy": 0.15181272178888322,
      "num_tokens": 21459851.0,
      "step": 9915
    },
    {
      "entropy": 6.370935535430908,
      "epoch": 1.0615335223928515,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004897426070874392,
      "loss": 6.0755,
      "mean_token_accuracy": 0.15603728890419005,
      "num_tokens": 21469717.0,
      "step": 9920
    },
    {
      "entropy": 6.490514135360717,
      "epoch": 1.0620685965006154,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004897311915681278,
      "loss": 6.1789,
      "mean_token_accuracy": 0.14462916627526284,
      "num_tokens": 21480815.0,
      "step": 9925
    },
    {
      "entropy": 6.501011514663697,
      "epoch": 1.0626036706083792,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004897197698484834,
      "loss": 6.0743,
      "mean_token_accuracy": 0.15044210404157637,
      "num_tokens": 21492000.0,
      "step": 9930
    },
    {
      "entropy": 6.428673505783081,
      "epoch": 1.0631387447161431,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004897083419288358,
      "loss": 6.0922,
      "mean_token_accuracy": 0.14192008078098298,
      "num_tokens": 21502006.0,
      "step": 9935
    },
    {
      "entropy": 6.378690624237061,
      "epoch": 1.0636738188239072,
      "grad_norm": 1.046875,
      "learning_rate": 0.000489696907809515,
      "loss": 6.1074,
      "mean_token_accuracy": 0.1524524912238121,
      "num_tokens": 21513066.0,
      "step": 9940
    },
    {
      "entropy": 6.402763700485229,
      "epoch": 1.064208892931671,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004896854674908512,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14280223995447158,
      "num_tokens": 21524853.0,
      "step": 9945
    },
    {
      "entropy": 6.431840705871582,
      "epoch": 1.064743967039435,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004896740209731746,
      "loss": 6.0093,
      "mean_token_accuracy": 0.14984188228845596,
      "num_tokens": 21536497.0,
      "step": 9950
    },
    {
      "entropy": 6.418710803985595,
      "epoch": 1.0652790411471988,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004896625682568159,
      "loss": 6.0514,
      "mean_token_accuracy": 0.15341312289237977,
      "num_tokens": 21546213.0,
      "step": 9955
    },
    {
      "entropy": 6.3661713123321535,
      "epoch": 1.0658141152549627,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004896511093421057,
      "loss": 6.0194,
      "mean_token_accuracy": 0.1580762155354023,
      "num_tokens": 21556990.0,
      "step": 9960
    },
    {
      "entropy": 6.384196376800537,
      "epoch": 1.0663491893627268,
      "grad_norm": 1.0,
      "learning_rate": 0.0004896396442293749,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15247415006160736,
      "num_tokens": 21567567.0,
      "step": 9965
    },
    {
      "entropy": 6.442054796218872,
      "epoch": 1.0668842634704907,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004896281729189544,
      "loss": 6.0543,
      "mean_token_accuracy": 0.14573955312371253,
      "num_tokens": 21577457.0,
      "step": 9970
    },
    {
      "entropy": 6.386103725433349,
      "epoch": 1.0674193375782546,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004896166954111758,
      "loss": 6.0526,
      "mean_token_accuracy": 0.15782974362373353,
      "num_tokens": 21588173.0,
      "step": 9975
    },
    {
      "entropy": 6.4095924377441404,
      "epoch": 1.0679544116860185,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048960521170637,
      "loss": 6.0912,
      "mean_token_accuracy": 0.14669934809207916,
      "num_tokens": 21599154.0,
      "step": 9980
    },
    {
      "entropy": 6.4254683494567875,
      "epoch": 1.0684894857937826,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895937218048692,
      "loss": 6.1226,
      "mean_token_accuracy": 0.1450969859957695,
      "num_tokens": 21610134.0,
      "step": 9985
    },
    {
      "entropy": 6.390247631072998,
      "epoch": 1.0690245599015464,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895822257070046,
      "loss": 6.0105,
      "mean_token_accuracy": 0.15353728979825973,
      "num_tokens": 21619989.0,
      "step": 9990
    },
    {
      "entropy": 6.397879266738892,
      "epoch": 1.0695596340093103,
      "grad_norm": 0.796875,
      "learning_rate": 0.0004895707234131084,
      "loss": 6.0663,
      "mean_token_accuracy": 0.15498469173908233,
      "num_tokens": 21632707.0,
      "step": 9995
    },
    {
      "entropy": 6.447797775268555,
      "epoch": 1.0700947081170742,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004895592149235128,
      "loss": 6.0317,
      "mean_token_accuracy": 0.15208378732204436,
      "num_tokens": 21643001.0,
      "step": 10000
    },
    {
      "entropy": 6.4747251033782955,
      "epoch": 1.070629782224838,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048954770023855,
      "loss": 6.1152,
      "mean_token_accuracy": 0.14609223753213882,
      "num_tokens": 21653425.0,
      "step": 10005
    },
    {
      "entropy": 6.459932565689087,
      "epoch": 1.0711648563326022,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004895361793585525,
      "loss": 6.127,
      "mean_token_accuracy": 0.1472277447581291,
      "num_tokens": 21664212.0,
      "step": 10010
    },
    {
      "entropy": 6.4673868179321286,
      "epoch": 1.071699930440366,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004895246522838529,
      "loss": 6.1235,
      "mean_token_accuracy": 0.14691547006368638,
      "num_tokens": 21675564.0,
      "step": 10015
    },
    {
      "entropy": 6.406800746917725,
      "epoch": 1.07223500454813,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004895131190147842,
      "loss": 6.1095,
      "mean_token_accuracy": 0.15103839933872223,
      "num_tokens": 21686225.0,
      "step": 10020
    },
    {
      "entropy": 6.386848735809326,
      "epoch": 1.0727700786558938,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004895015795516793,
      "loss": 6.023,
      "mean_token_accuracy": 0.15253046602010728,
      "num_tokens": 21697479.0,
      "step": 10025
    },
    {
      "entropy": 6.480775499343872,
      "epoch": 1.0733051527636577,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004894900338948714,
      "loss": 6.1523,
      "mean_token_accuracy": 0.14475396350026132,
      "num_tokens": 21708496.0,
      "step": 10030
    },
    {
      "entropy": 6.421282005310059,
      "epoch": 1.0738402268714218,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004894784820446939,
      "loss": 6.0074,
      "mean_token_accuracy": 0.15607749968767165,
      "num_tokens": 21718290.0,
      "step": 10035
    },
    {
      "entropy": 6.361752891540528,
      "epoch": 1.0743753009791857,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004894669240014804,
      "loss": 5.9993,
      "mean_token_accuracy": 0.1518443137407303,
      "num_tokens": 21729185.0,
      "step": 10040
    },
    {
      "entropy": 6.4178660869598385,
      "epoch": 1.0749103750869495,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004894553597655646,
      "loss": 6.0837,
      "mean_token_accuracy": 0.14546559676527976,
      "num_tokens": 21740787.0,
      "step": 10045
    },
    {
      "entropy": 6.486740016937256,
      "epoch": 1.0754454491947134,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004894437893372804,
      "loss": 6.1284,
      "mean_token_accuracy": 0.14527418315410615,
      "num_tokens": 21751804.0,
      "step": 10050
    },
    {
      "entropy": 6.479933214187622,
      "epoch": 1.0759805233024773,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000489432212716962,
      "loss": 6.0605,
      "mean_token_accuracy": 0.1507222905755043,
      "num_tokens": 21762730.0,
      "step": 10055
    },
    {
      "entropy": 6.391507148742676,
      "epoch": 1.0765155974102414,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004894206299049436,
      "loss": 6.0869,
      "mean_token_accuracy": 0.14692523330450058,
      "num_tokens": 21773470.0,
      "step": 10060
    },
    {
      "entropy": 6.440608644485474,
      "epoch": 1.0770506715180053,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004894090409015595,
      "loss": 6.0895,
      "mean_token_accuracy": 0.146237114071846,
      "num_tokens": 21783385.0,
      "step": 10065
    },
    {
      "entropy": 6.364365720748902,
      "epoch": 1.0775857456257691,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004893974457071445,
      "loss": 6.1004,
      "mean_token_accuracy": 0.14771676436066628,
      "num_tokens": 21794857.0,
      "step": 10070
    },
    {
      "entropy": 6.463506364822388,
      "epoch": 1.078120819733533,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004893858443220335,
      "loss": 6.0968,
      "mean_token_accuracy": 0.1443526454269886,
      "num_tokens": 21804296.0,
      "step": 10075
    },
    {
      "entropy": 6.484564781188965,
      "epoch": 1.0786558938412971,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004893742367465613,
      "loss": 6.0937,
      "mean_token_accuracy": 0.14892311319708823,
      "num_tokens": 21815899.0,
      "step": 10080
    },
    {
      "entropy": 6.397023391723633,
      "epoch": 1.079190967949061,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004893626229810631,
      "loss": 6.0655,
      "mean_token_accuracy": 0.1513351768255234,
      "num_tokens": 21825964.0,
      "step": 10085
    },
    {
      "entropy": 6.454033517837525,
      "epoch": 1.0797260420568249,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0004893510030258743,
      "loss": 6.1646,
      "mean_token_accuracy": 0.1430799700319767,
      "num_tokens": 21836656.0,
      "step": 10090
    },
    {
      "entropy": 6.44245924949646,
      "epoch": 1.0802611161645888,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004893393768813305,
      "loss": 6.1613,
      "mean_token_accuracy": 0.1456031396985054,
      "num_tokens": 21847658.0,
      "step": 10095
    },
    {
      "entropy": 6.51317343711853,
      "epoch": 1.0807961902723526,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004893277445477673,
      "loss": 6.125,
      "mean_token_accuracy": 0.13948384076356887,
      "num_tokens": 21858338.0,
      "step": 10100
    },
    {
      "entropy": 6.406071662902832,
      "epoch": 1.0813312643801167,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004893161060255206,
      "loss": 5.9892,
      "mean_token_accuracy": 0.1562623143196106,
      "num_tokens": 21870656.0,
      "step": 10105
    },
    {
      "entropy": 6.394718933105469,
      "epoch": 1.0818663384878806,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004893044613149263,
      "loss": 6.0988,
      "mean_token_accuracy": 0.15282596796751022,
      "num_tokens": 21881806.0,
      "step": 10110
    },
    {
      "entropy": 6.343643856048584,
      "epoch": 1.0824014125956445,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004892928104163209,
      "loss": 6.0196,
      "mean_token_accuracy": 0.14847707897424697,
      "num_tokens": 21892277.0,
      "step": 10115
    },
    {
      "entropy": 6.375891828536988,
      "epoch": 1.0829364867034084,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004892811533300407,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15485417544841767,
      "num_tokens": 21903509.0,
      "step": 10120
    },
    {
      "entropy": 6.47607159614563,
      "epoch": 1.0834715608111725,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004892694900564223,
      "loss": 6.127,
      "mean_token_accuracy": 0.14488525539636612,
      "num_tokens": 21914308.0,
      "step": 10125
    },
    {
      "entropy": 6.376104402542114,
      "epoch": 1.0840066349189363,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004892578205958025,
      "loss": 6.0255,
      "mean_token_accuracy": 0.15569847971200942,
      "num_tokens": 21924302.0,
      "step": 10130
    },
    {
      "entropy": 6.447469568252563,
      "epoch": 1.0845417090267002,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004892461449485182,
      "loss": 6.0979,
      "mean_token_accuracy": 0.14397566318511962,
      "num_tokens": 21935066.0,
      "step": 10135
    },
    {
      "entropy": 6.424494314193725,
      "epoch": 1.085076783134464,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004892344631149066,
      "loss": 6.0024,
      "mean_token_accuracy": 0.156691525131464,
      "num_tokens": 21945798.0,
      "step": 10140
    },
    {
      "entropy": 6.371891069412231,
      "epoch": 1.085611857242228,
      "grad_norm": 0.96875,
      "learning_rate": 0.000489222775095305,
      "loss": 6.0936,
      "mean_token_accuracy": 0.14902716130018234,
      "num_tokens": 21955667.0,
      "step": 10145
    },
    {
      "entropy": 6.391481828689575,
      "epoch": 1.086146931349992,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004892110808900509,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14894235879182816,
      "num_tokens": 21966107.0,
      "step": 10150
    },
    {
      "entropy": 6.4563398361206055,
      "epoch": 1.086682005457756,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004891993804994818,
      "loss": 6.065,
      "mean_token_accuracy": 0.15464599430561066,
      "num_tokens": 21976408.0,
      "step": 10155
    },
    {
      "entropy": 6.532507419586182,
      "epoch": 1.0872170795655198,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004891876739239358,
      "loss": 6.0548,
      "mean_token_accuracy": 0.1521727569401264,
      "num_tokens": 21986481.0,
      "step": 10160
    },
    {
      "entropy": 6.417515563964844,
      "epoch": 1.0877521536732837,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004891759611637508,
      "loss": 6.0943,
      "mean_token_accuracy": 0.14886776804924012,
      "num_tokens": 21997554.0,
      "step": 10165
    },
    {
      "entropy": 6.49372968673706,
      "epoch": 1.0882872277810476,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000489164242219265,
      "loss": 6.1459,
      "mean_token_accuracy": 0.14391707479953766,
      "num_tokens": 22009048.0,
      "step": 10170
    },
    {
      "entropy": 6.520137166976928,
      "epoch": 1.0888223018888117,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004891525170908169,
      "loss": 6.1609,
      "mean_token_accuracy": 0.14751541018486022,
      "num_tokens": 22020625.0,
      "step": 10175
    },
    {
      "entropy": 6.480677938461303,
      "epoch": 1.0893573759965756,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004891407857787448,
      "loss": 6.0437,
      "mean_token_accuracy": 0.1444891132414341,
      "num_tokens": 22032298.0,
      "step": 10180
    },
    {
      "entropy": 6.392166805267334,
      "epoch": 1.0898924501043394,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004891290482833876,
      "loss": 6.0356,
      "mean_token_accuracy": 0.1518296092748642,
      "num_tokens": 22043693.0,
      "step": 10185
    },
    {
      "entropy": 6.409099960327149,
      "epoch": 1.0904275242121033,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004891173046050844,
      "loss": 6.1008,
      "mean_token_accuracy": 0.1553488165140152,
      "num_tokens": 22054373.0,
      "step": 10190
    },
    {
      "entropy": 6.437811851501465,
      "epoch": 1.0909625983198672,
      "grad_norm": 1.03125,
      "learning_rate": 0.000489105554744174,
      "loss": 6.0729,
      "mean_token_accuracy": 0.150793220102787,
      "num_tokens": 22065039.0,
      "step": 10195
    },
    {
      "entropy": 6.50524754524231,
      "epoch": 1.0914976724276313,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004890937987009958,
      "loss": 6.0908,
      "mean_token_accuracy": 0.151939994096756,
      "num_tokens": 22077003.0,
      "step": 10200
    },
    {
      "entropy": 6.487708902359008,
      "epoch": 1.0920327465353952,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004890820364758892,
      "loss": 6.0878,
      "mean_token_accuracy": 0.15026550590991974,
      "num_tokens": 22088558.0,
      "step": 10205
    },
    {
      "entropy": 6.479548788070678,
      "epoch": 1.092567820643159,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000489070268069194,
      "loss": 6.0623,
      "mean_token_accuracy": 0.1517270565032959,
      "num_tokens": 22097777.0,
      "step": 10210
    },
    {
      "entropy": 6.419154262542724,
      "epoch": 1.093102894750923,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004890584934812498,
      "loss": 5.9928,
      "mean_token_accuracy": 0.15226405560970308,
      "num_tokens": 22107803.0,
      "step": 10215
    },
    {
      "entropy": 6.403665971755982,
      "epoch": 1.093637968858687,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004890467127123967,
      "loss": 6.003,
      "mean_token_accuracy": 0.15619039833545684,
      "num_tokens": 22117965.0,
      "step": 10220
    },
    {
      "entropy": 6.473029899597168,
      "epoch": 1.094173042966451,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004890349257629749,
      "loss": 6.0483,
      "mean_token_accuracy": 0.1545405551791191,
      "num_tokens": 22127759.0,
      "step": 10225
    },
    {
      "entropy": 6.424864292144775,
      "epoch": 1.0947081170742148,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004890231326333246,
      "loss": 6.1063,
      "mean_token_accuracy": 0.14265400022268296,
      "num_tokens": 22138477.0,
      "step": 10230
    },
    {
      "entropy": 6.430114078521728,
      "epoch": 1.0952431911819787,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004890113333237865,
      "loss": 6.0726,
      "mean_token_accuracy": 0.14977218061685563,
      "num_tokens": 22149568.0,
      "step": 10235
    },
    {
      "entropy": 6.419962358474732,
      "epoch": 1.0957782652897425,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004889995278347012,
      "loss": 6.0571,
      "mean_token_accuracy": 0.14955754801630974,
      "num_tokens": 22159949.0,
      "step": 10240
    },
    {
      "entropy": 6.383775806427002,
      "epoch": 1.0963133393975066,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004889877161664096,
      "loss": 6.0624,
      "mean_token_accuracy": 0.14545081555843353,
      "num_tokens": 22170115.0,
      "step": 10245
    },
    {
      "entropy": 6.41920108795166,
      "epoch": 1.0968484135052705,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004889758983192528,
      "loss": 6.0901,
      "mean_token_accuracy": 0.14915428459644317,
      "num_tokens": 22181323.0,
      "step": 10250
    },
    {
      "entropy": 6.4047595977783205,
      "epoch": 1.0973834876130344,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004889640742935719,
      "loss": 6.0299,
      "mean_token_accuracy": 0.1494324542582035,
      "num_tokens": 22192078.0,
      "step": 10255
    },
    {
      "entropy": 6.485225772857666,
      "epoch": 1.0979185617207983,
      "grad_norm": 1.0,
      "learning_rate": 0.0004889522440897085,
      "loss": 6.1086,
      "mean_token_accuracy": 0.14816955849528313,
      "num_tokens": 22203211.0,
      "step": 10260
    },
    {
      "entropy": 6.451343822479248,
      "epoch": 1.0984536358285624,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004889404077080041,
      "loss": 6.0448,
      "mean_token_accuracy": 0.14499879628419876,
      "num_tokens": 22213487.0,
      "step": 10265
    },
    {
      "entropy": 6.426690006256104,
      "epoch": 1.0989887099363262,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004889285651488005,
      "loss": 6.0715,
      "mean_token_accuracy": 0.1457860678434372,
      "num_tokens": 22223717.0,
      "step": 10270
    },
    {
      "entropy": 6.391878318786621,
      "epoch": 1.0995237840440901,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004889167164124396,
      "loss": 6.0901,
      "mean_token_accuracy": 0.15345896035432816,
      "num_tokens": 22235141.0,
      "step": 10275
    },
    {
      "entropy": 6.480268955230713,
      "epoch": 1.100058858151854,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004889048614992636,
      "loss": 6.1523,
      "mean_token_accuracy": 0.1405470959842205,
      "num_tokens": 22246601.0,
      "step": 10280
    },
    {
      "entropy": 6.421423244476318,
      "epoch": 1.1005939322596179,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004888930004096148,
      "loss": 6.0746,
      "mean_token_accuracy": 0.1453216128051281,
      "num_tokens": 22257206.0,
      "step": 10285
    },
    {
      "entropy": 6.49395112991333,
      "epoch": 1.101129006367382,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004888811331438356,
      "loss": 6.14,
      "mean_token_accuracy": 0.1408381849527359,
      "num_tokens": 22267799.0,
      "step": 10290
    },
    {
      "entropy": 6.4962990283966064,
      "epoch": 1.1016640804751459,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004888692597022689,
      "loss": 6.185,
      "mean_token_accuracy": 0.14611935019493102,
      "num_tokens": 22279368.0,
      "step": 10295
    },
    {
      "entropy": 6.411830472946167,
      "epoch": 1.1021991545829097,
      "grad_norm": 0.84375,
      "learning_rate": 0.0004888573800852572,
      "loss": 5.9155,
      "mean_token_accuracy": 0.1589374363422394,
      "num_tokens": 22289998.0,
      "step": 10300
    },
    {
      "entropy": 6.334289979934693,
      "epoch": 1.1027342286906736,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004888454942931438,
      "loss": 6.0336,
      "mean_token_accuracy": 0.16020656377077103,
      "num_tokens": 22300058.0,
      "step": 10305
    },
    {
      "entropy": 6.443489122390747,
      "epoch": 1.1032693027984375,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004888336023262718,
      "loss": 6.1598,
      "mean_token_accuracy": 0.1443057283759117,
      "num_tokens": 22310962.0,
      "step": 10310
    },
    {
      "entropy": 6.468604373931885,
      "epoch": 1.1038043769062016,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004888217041849846,
      "loss": 6.0147,
      "mean_token_accuracy": 0.1587409719824791,
      "num_tokens": 22321526.0,
      "step": 10315
    },
    {
      "entropy": 6.45356707572937,
      "epoch": 1.1043394510139655,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004888097998696256,
      "loss": 6.1147,
      "mean_token_accuracy": 0.14810292199254035,
      "num_tokens": 22332558.0,
      "step": 10320
    },
    {
      "entropy": 6.42706069946289,
      "epoch": 1.1048745251217293,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004887978893805387,
      "loss": 6.0606,
      "mean_token_accuracy": 0.14955779910087585,
      "num_tokens": 22342392.0,
      "step": 10325
    },
    {
      "entropy": 6.425696277618409,
      "epoch": 1.1054095992294932,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004887859727180679,
      "loss": 6.1102,
      "mean_token_accuracy": 0.14016611725091935,
      "num_tokens": 22353089.0,
      "step": 10330
    },
    {
      "entropy": 6.3832090377807615,
      "epoch": 1.105944673337257,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004887740498825572,
      "loss": 6.0284,
      "mean_token_accuracy": 0.1551051214337349,
      "num_tokens": 22364314.0,
      "step": 10335
    },
    {
      "entropy": 6.3845062255859375,
      "epoch": 1.1064797474450212,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004887621208743507,
      "loss": 6.0454,
      "mean_token_accuracy": 0.15220018178224565,
      "num_tokens": 22374985.0,
      "step": 10340
    },
    {
      "entropy": 6.418020820617675,
      "epoch": 1.107014821552785,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004887501856937932,
      "loss": 5.9543,
      "mean_token_accuracy": 0.15764311850070953,
      "num_tokens": 22385618.0,
      "step": 10345
    },
    {
      "entropy": 6.470552587509156,
      "epoch": 1.107549895660549,
      "grad_norm": 0.953125,
      "learning_rate": 0.000488738244341229,
      "loss": 6.0695,
      "mean_token_accuracy": 0.15393458679318428,
      "num_tokens": 22395871.0,
      "step": 10350
    },
    {
      "entropy": 6.433292579650879,
      "epoch": 1.1080849697683128,
      "grad_norm": 0.84765625,
      "learning_rate": 0.000488726296817003,
      "loss": 6.102,
      "mean_token_accuracy": 0.1510417215526104,
      "num_tokens": 22406707.0,
      "step": 10355
    },
    {
      "entropy": 6.445325803756714,
      "epoch": 1.108620043876077,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004887143431214602,
      "loss": 6.107,
      "mean_token_accuracy": 0.14697629138827323,
      "num_tokens": 22417543.0,
      "step": 10360
    },
    {
      "entropy": 6.37202787399292,
      "epoch": 1.1091551179838408,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004887023832549459,
      "loss": 6.0356,
      "mean_token_accuracy": 0.15357186049222946,
      "num_tokens": 22428283.0,
      "step": 10365
    },
    {
      "entropy": 6.3737739562988285,
      "epoch": 1.1096901920916047,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004886904172178052,
      "loss": 6.0245,
      "mean_token_accuracy": 0.1518484517931938,
      "num_tokens": 22439860.0,
      "step": 10370
    },
    {
      "entropy": 6.4564087867736815,
      "epoch": 1.1102252661993686,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004886784450103838,
      "loss": 6.0878,
      "mean_token_accuracy": 0.1529605954885483,
      "num_tokens": 22451250.0,
      "step": 10375
    },
    {
      "entropy": 6.386963558197022,
      "epoch": 1.1107603403071324,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004886664666330273,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1488528370857239,
      "num_tokens": 22461112.0,
      "step": 10380
    },
    {
      "entropy": 6.459249639511109,
      "epoch": 1.1112954144148965,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004886544820860816,
      "loss": 6.0548,
      "mean_token_accuracy": 0.15672071278095245,
      "num_tokens": 22471368.0,
      "step": 10385
    },
    {
      "entropy": 6.439004373550415,
      "epoch": 1.1118304885226604,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004886424913698927,
      "loss": 6.0558,
      "mean_token_accuracy": 0.15102000236511232,
      "num_tokens": 22482011.0,
      "step": 10390
    },
    {
      "entropy": 6.394613552093506,
      "epoch": 1.1123655626304243,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004886304944848069,
      "loss": 5.9945,
      "mean_token_accuracy": 0.16399833112955092,
      "num_tokens": 22493648.0,
      "step": 10395
    },
    {
      "entropy": 6.472637605667114,
      "epoch": 1.1129006367381882,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004886184914311706,
      "loss": 6.1355,
      "mean_token_accuracy": 0.1481831580400467,
      "num_tokens": 22505577.0,
      "step": 10400
    },
    {
      "entropy": 6.457976293563843,
      "epoch": 1.1134357108459523,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004886064822093305,
      "loss": 6.0942,
      "mean_token_accuracy": 0.14395636469125747,
      "num_tokens": 22515546.0,
      "step": 10405
    },
    {
      "entropy": 6.395815181732178,
      "epoch": 1.1139707849537162,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004885944668196332,
      "loss": 5.9838,
      "mean_token_accuracy": 0.16027534753084183,
      "num_tokens": 22526710.0,
      "step": 10410
    },
    {
      "entropy": 6.456317710876465,
      "epoch": 1.11450585906148,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004885824452624254,
      "loss": 6.1173,
      "mean_token_accuracy": 0.1452877067029476,
      "num_tokens": 22537809.0,
      "step": 10415
    },
    {
      "entropy": 6.453176689147949,
      "epoch": 1.115040933169244,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004885704175380548,
      "loss": 6.0965,
      "mean_token_accuracy": 0.1449737437069416,
      "num_tokens": 22549301.0,
      "step": 10420
    },
    {
      "entropy": 6.458534908294678,
      "epoch": 1.1155760072770078,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004885583836468683,
      "loss": 6.0099,
      "mean_token_accuracy": 0.15261961966753007,
      "num_tokens": 22560483.0,
      "step": 10425
    },
    {
      "entropy": 6.396278381347656,
      "epoch": 1.1161110813847719,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004885463435892136,
      "loss": 6.1155,
      "mean_token_accuracy": 0.144923784583807,
      "num_tokens": 22570524.0,
      "step": 10430
    },
    {
      "entropy": 6.422711277008057,
      "epoch": 1.1166461554925358,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000488534297365438,
      "loss": 6.1168,
      "mean_token_accuracy": 0.14799824953079224,
      "num_tokens": 22581120.0,
      "step": 10435
    },
    {
      "entropy": 6.4675966739654545,
      "epoch": 1.1171812296002996,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004885222449758899,
      "loss": 6.0934,
      "mean_token_accuracy": 0.1485187105834484,
      "num_tokens": 22592658.0,
      "step": 10440
    },
    {
      "entropy": 6.462940645217896,
      "epoch": 1.1177163037080635,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004885101864209167,
      "loss": 6.0817,
      "mean_token_accuracy": 0.1439016804099083,
      "num_tokens": 22603497.0,
      "step": 10445
    },
    {
      "entropy": 6.413467311859131,
      "epoch": 1.1182513778158274,
      "grad_norm": 0.96875,
      "learning_rate": 0.000488498121700867,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15722592622041703,
      "num_tokens": 22612916.0,
      "step": 10450
    },
    {
      "entropy": 6.366930341720581,
      "epoch": 1.1187864519235915,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004884860508160891,
      "loss": 6.0723,
      "mean_token_accuracy": 0.15238417387008668,
      "num_tokens": 22625345.0,
      "step": 10455
    },
    {
      "entropy": 6.421725797653198,
      "epoch": 1.1193215260313554,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004884739737669314,
      "loss": 5.9876,
      "mean_token_accuracy": 0.15892833918333055,
      "num_tokens": 22637000.0,
      "step": 10460
    },
    {
      "entropy": 6.416720533370972,
      "epoch": 1.1198566001391193,
      "grad_norm": 1.0,
      "learning_rate": 0.0004884618905537427,
      "loss": 6.0153,
      "mean_token_accuracy": 0.15657200962305068,
      "num_tokens": 22647025.0,
      "step": 10465
    },
    {
      "entropy": 6.3941301822662355,
      "epoch": 1.1203916742468831,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004884498011768719,
      "loss": 5.9879,
      "mean_token_accuracy": 0.15711487084627151,
      "num_tokens": 22657253.0,
      "step": 10470
    },
    {
      "entropy": 6.40429720878601,
      "epoch": 1.120926748354647,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000488437705636668,
      "loss": 6.0399,
      "mean_token_accuracy": 0.1520557001233101,
      "num_tokens": 22668432.0,
      "step": 10475
    },
    {
      "entropy": 6.41401424407959,
      "epoch": 1.121461822462411,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004884256039334805,
      "loss": 6.0915,
      "mean_token_accuracy": 0.14589745476841925,
      "num_tokens": 22678538.0,
      "step": 10480
    },
    {
      "entropy": 6.414116954803466,
      "epoch": 1.121996896570175,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004884134960676586,
      "loss": 6.0485,
      "mean_token_accuracy": 0.14792819917201996,
      "num_tokens": 22688994.0,
      "step": 10485
    },
    {
      "entropy": 6.43791618347168,
      "epoch": 1.1225319706779389,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000488401382039552,
      "loss": 6.0791,
      "mean_token_accuracy": 0.1497262015938759,
      "num_tokens": 22699584.0,
      "step": 10490
    },
    {
      "entropy": 6.422362899780273,
      "epoch": 1.1230670447857027,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004883892618495104,
      "loss": 5.9752,
      "mean_token_accuracy": 0.15234384685754776,
      "num_tokens": 22710093.0,
      "step": 10495
    },
    {
      "entropy": 6.4493261814117435,
      "epoch": 1.1236021188934668,
      "grad_norm": 0.921875,
      "learning_rate": 0.000488377135497884,
      "loss": 6.0803,
      "mean_token_accuracy": 0.14836570024490356,
      "num_tokens": 22720533.0,
      "step": 10500
    },
    {
      "entropy": 6.3820295333862305,
      "epoch": 1.1241371930012307,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004883650029850226,
      "loss": 5.9731,
      "mean_token_accuracy": 0.15201365649700166,
      "num_tokens": 22731010.0,
      "step": 10505
    },
    {
      "entropy": 6.37717604637146,
      "epoch": 1.1246722671089946,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004883528643112769,
      "loss": 6.0614,
      "mean_token_accuracy": 0.1477554202079773,
      "num_tokens": 22743045.0,
      "step": 10510
    },
    {
      "entropy": 6.408057165145874,
      "epoch": 1.1252073412167585,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004883407194769972,
      "loss": 6.0455,
      "mean_token_accuracy": 0.15190024971961974,
      "num_tokens": 22753746.0,
      "step": 10515
    },
    {
      "entropy": 6.460110425949097,
      "epoch": 1.1257424153245223,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004883285684825342,
      "loss": 6.0673,
      "mean_token_accuracy": 0.15546210557222367,
      "num_tokens": 22765264.0,
      "step": 10520
    },
    {
      "entropy": 6.468258857727051,
      "epoch": 1.1262774894322864,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004883164113282386,
      "loss": 6.0587,
      "mean_token_accuracy": 0.15260644555091857,
      "num_tokens": 22775727.0,
      "step": 10525
    },
    {
      "entropy": 6.440642976760865,
      "epoch": 1.1268125635400503,
      "grad_norm": 0.8203125,
      "learning_rate": 0.0004883042480144618,
      "loss": 6.055,
      "mean_token_accuracy": 0.1503731533885002,
      "num_tokens": 22787228.0,
      "step": 10530
    },
    {
      "entropy": 6.494921445846558,
      "epoch": 1.1273476376478142,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004882920785415547,
      "loss": 6.125,
      "mean_token_accuracy": 0.14749099910259247,
      "num_tokens": 22797467.0,
      "step": 10535
    },
    {
      "entropy": 6.4185178756713865,
      "epoch": 1.127882711755578,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004882799029098689,
      "loss": 6.0907,
      "mean_token_accuracy": 0.14464054852724076,
      "num_tokens": 22808281.0,
      "step": 10540
    },
    {
      "entropy": 6.448378658294677,
      "epoch": 1.1284177858633422,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048826772111975583,
      "loss": 6.1357,
      "mean_token_accuracy": 0.15108129233121873,
      "num_tokens": 22819387.0,
      "step": 10545
    },
    {
      "entropy": 6.423340892791748,
      "epoch": 1.128952859971106,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048825553317156717,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14872741252183913,
      "num_tokens": 22829649.0,
      "step": 10550
    },
    {
      "entropy": 6.383133172988892,
      "epoch": 1.12948793407887,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004882433390656551,
      "loss": 6.084,
      "mean_token_accuracy": 0.1525307387113571,
      "num_tokens": 22840667.0,
      "step": 10555
    },
    {
      "entropy": 6.4322563171386715,
      "epoch": 1.1300230081866338,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004882311388023715,
      "loss": 6.0314,
      "mean_token_accuracy": 0.15103217586874962,
      "num_tokens": 22852461.0,
      "step": 10560
    },
    {
      "entropy": 6.501431894302368,
      "epoch": 1.1305580822943977,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004882189323820688,
      "loss": 6.1142,
      "mean_token_accuracy": 0.1415798097848892,
      "num_tokens": 22862781.0,
      "step": 10565
    },
    {
      "entropy": 6.504143524169922,
      "epoch": 1.1310931564021618,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048820671980509935,
      "loss": 6.1136,
      "mean_token_accuracy": 0.14972802996635437,
      "num_tokens": 22873393.0,
      "step": 10570
    },
    {
      "entropy": 6.489892244338989,
      "epoch": 1.1316282305099257,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048819450107181583,
      "loss": 6.1076,
      "mean_token_accuracy": 0.14779544323682786,
      "num_tokens": 22883931.0,
      "step": 10575
    },
    {
      "entropy": 6.423511743545532,
      "epoch": 1.1321633046176895,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004881822761825711,
      "loss": 6.0435,
      "mean_token_accuracy": 0.14574084728956221,
      "num_tokens": 22894829.0,
      "step": 10580
    },
    {
      "entropy": 6.411485481262207,
      "epoch": 1.1326983787254534,
      "grad_norm": 0.875,
      "learning_rate": 0.000488170045137718,
      "loss": 6.078,
      "mean_token_accuracy": 0.14932154715061188,
      "num_tokens": 22906059.0,
      "step": 10585
    },
    {
      "entropy": 6.457902097702027,
      "epoch": 1.1332334528332173,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004881578079376099,
      "loss": 6.0841,
      "mean_token_accuracy": 0.14441045373678207,
      "num_tokens": 22917834.0,
      "step": 10590
    },
    {
      "entropy": 6.421571207046509,
      "epoch": 1.1337685269409814,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048814556458259996,
      "loss": 5.9901,
      "mean_token_accuracy": 0.16167470067739487,
      "num_tokens": 22929157.0,
      "step": 10595
    },
    {
      "entropy": 6.490074348449707,
      "epoch": 1.1343036010487453,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004881333150730419,
      "loss": 6.1165,
      "mean_token_accuracy": 0.14964948296546937,
      "num_tokens": 22939396.0,
      "step": 10600
    },
    {
      "entropy": 6.438498306274414,
      "epoch": 1.1348386751565092,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048812105940928917,
      "loss": 6.0823,
      "mean_token_accuracy": 0.14990349262952804,
      "num_tokens": 22950253.0,
      "step": 10605
    },
    {
      "entropy": 6.455872678756714,
      "epoch": 1.135373749264273,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048810879759169593,
      "loss": 6.0999,
      "mean_token_accuracy": 0.15298274010419846,
      "num_tokens": 22960417.0,
      "step": 10610
    },
    {
      "entropy": 6.4434606552124025,
      "epoch": 1.135908823372037,
      "grad_norm": 1.1875,
      "learning_rate": 0.000488096529620616,
      "loss": 6.0578,
      "mean_token_accuracy": 0.14702137559652328,
      "num_tokens": 22971763.0,
      "step": 10615
    },
    {
      "entropy": 6.505296325683593,
      "epoch": 1.136443897479801,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048808425549640383,
      "loss": 6.0842,
      "mean_token_accuracy": 0.14849616065621377,
      "num_tokens": 22981571.0,
      "step": 10620
    },
    {
      "entropy": 6.426681184768677,
      "epoch": 1.136978971587565,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048807197521941366,
      "loss": 6.0539,
      "mean_token_accuracy": 0.15373365134000777,
      "num_tokens": 22991644.0,
      "step": 10625
    },
    {
      "entropy": 6.362898826599121,
      "epoch": 1.1375140456953288,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004880596887900002,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15561307817697526,
      "num_tokens": 23001557.0,
      "step": 10630
    },
    {
      "entropy": 6.400293016433716,
      "epoch": 1.1380491198030926,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048804739620851806,
      "loss": 6.0948,
      "mean_token_accuracy": 0.14370866864919662,
      "num_tokens": 23012715.0,
      "step": 10635
    },
    {
      "entropy": 6.431537914276123,
      "epoch": 1.1385841939108567,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048803509747532235,
      "loss": 6.0855,
      "mean_token_accuracy": 0.14991748481988906,
      "num_tokens": 23023016.0,
      "step": 10640
    },
    {
      "entropy": 6.4581732749938965,
      "epoch": 1.1391192680186206,
      "grad_norm": 0.8359375,
      "learning_rate": 0.0004880227925907682,
      "loss": 6.0812,
      "mean_token_accuracy": 0.1482681669294834,
      "num_tokens": 23033511.0,
      "step": 10645
    },
    {
      "entropy": 6.463703298568726,
      "epoch": 1.1396543421263845,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004880104815552108,
      "loss": 6.0827,
      "mean_token_accuracy": 0.1531242087483406,
      "num_tokens": 23045149.0,
      "step": 10650
    },
    {
      "entropy": 6.3737654209136965,
      "epoch": 1.1401894162341484,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004879981643690056,
      "loss": 6.0441,
      "mean_token_accuracy": 0.1592191517353058,
      "num_tokens": 23054489.0,
      "step": 10655
    },
    {
      "entropy": 6.449917793273926,
      "epoch": 1.1407244903419125,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048798584103250847,
      "loss": 6.1902,
      "mean_token_accuracy": 0.1439359351992607,
      "num_tokens": 23064725.0,
      "step": 10660
    },
    {
      "entropy": 6.453944969177246,
      "epoch": 1.1412595644496764,
      "grad_norm": 0.88671875,
      "learning_rate": 0.000487973511546075,
      "loss": 6.1371,
      "mean_token_accuracy": 0.14356259480118752,
      "num_tokens": 23076473.0,
      "step": 10665
    },
    {
      "entropy": 6.474194765090942,
      "epoch": 1.1417946385574402,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004879611759100614,
      "loss": 6.0447,
      "mean_token_accuracy": 0.15839172154664993,
      "num_tokens": 23087449.0,
      "step": 10670
    },
    {
      "entropy": 6.465250492095947,
      "epoch": 1.1423297126652041,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004879488341248237,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15290169715881347,
      "num_tokens": 23098284.0,
      "step": 10675
    },
    {
      "entropy": 6.3778749942779545,
      "epoch": 1.142864786772968,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048793648619071834,
      "loss": 6.0355,
      "mean_token_accuracy": 0.15722760632634164,
      "num_tokens": 23108545.0,
      "step": 10680
    },
    {
      "entropy": 6.441332674026489,
      "epoch": 1.143399860880732,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004879241321081019,
      "loss": 5.9952,
      "mean_token_accuracy": 0.1549189269542694,
      "num_tokens": 23118568.0,
      "step": 10685
    },
    {
      "entropy": 6.470385646820068,
      "epoch": 1.143934934988496,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048791177187733104,
      "loss": 6.0969,
      "mean_token_accuracy": 0.14864624291658401,
      "num_tokens": 23129189.0,
      "step": 10690
    },
    {
      "entropy": 6.455295276641846,
      "epoch": 1.1444700090962598,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004878994054987627,
      "loss": 6.1191,
      "mean_token_accuracy": 0.14697857052087784,
      "num_tokens": 23140238.0,
      "step": 10695
    },
    {
      "entropy": 6.407076358795166,
      "epoch": 1.1450050832040237,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004878870329727539,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15791965872049332,
      "num_tokens": 23149567.0,
      "step": 10700
    },
    {
      "entropy": 6.448425197601319,
      "epoch": 1.1455401573117876,
      "grad_norm": 0.984375,
      "learning_rate": 0.000487874654299662,
      "loss": 6.0039,
      "mean_token_accuracy": 0.1536222830414772,
      "num_tokens": 23159208.0,
      "step": 10705
    },
    {
      "entropy": 6.361423063278198,
      "epoch": 1.1460752314195517,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048786226947984435,
      "loss": 6.051,
      "mean_token_accuracy": 0.1475197084248066,
      "num_tokens": 23169092.0,
      "step": 10710
    },
    {
      "entropy": 6.4091840267181395,
      "epoch": 1.1466103055273156,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004878498785136586,
      "loss": 6.0642,
      "mean_token_accuracy": 0.15199013203382492,
      "num_tokens": 23180385.0,
      "step": 10715
    },
    {
      "entropy": 6.467318820953369,
      "epoch": 1.1471453796350795,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048783748140146245,
      "loss": 6.0578,
      "mean_token_accuracy": 0.1475761592388153,
      "num_tokens": 23191180.0,
      "step": 10720
    },
    {
      "entropy": 6.4394677639007565,
      "epoch": 1.1476804537428433,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000487825078143614,
      "loss": 6.0301,
      "mean_token_accuracy": 0.15278345644474028,
      "num_tokens": 23200920.0,
      "step": 10725
    },
    {
      "entropy": 6.3841925144195555,
      "epoch": 1.1482155278506072,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004878126687404713,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15247055590152742,
      "num_tokens": 23211140.0,
      "step": 10730
    },
    {
      "entropy": 6.441495704650879,
      "epoch": 1.1487506019583713,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004878002531923927,
      "loss": 6.0748,
      "mean_token_accuracy": 0.14650596231222152,
      "num_tokens": 23222039.0,
      "step": 10735
    },
    {
      "entropy": 6.519817543029785,
      "epoch": 1.1492856760661352,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048778783149973674,
      "loss": 6.1204,
      "mean_token_accuracy": 0.14570426791906357,
      "num_tokens": 23232502.0,
      "step": 10740
    },
    {
      "entropy": 6.466353082656861,
      "epoch": 1.149820750173899,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048777540366286195,
      "loss": 6.0287,
      "mean_token_accuracy": 0.15317765027284622,
      "num_tokens": 23243172.0,
      "step": 10745
    },
    {
      "entropy": 6.4662481307983395,
      "epoch": 1.150355824281663,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004877629696821273,
      "loss": 6.1176,
      "mean_token_accuracy": 0.150849911570549,
      "num_tokens": 23254558.0,
      "step": 10750
    },
    {
      "entropy": 6.497811889648437,
      "epoch": 1.1508908983894268,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048775052955789185,
      "loss": 6.1396,
      "mean_token_accuracy": 0.14723906069993972,
      "num_tokens": 23266283.0,
      "step": 10755
    },
    {
      "entropy": 6.3984321594238285,
      "epoch": 1.151425972497191,
      "grad_norm": 0.83203125,
      "learning_rate": 0.0004877380832905147,
      "loss": 6.0021,
      "mean_token_accuracy": 0.15511785745620726,
      "num_tokens": 23277823.0,
      "step": 10760
    },
    {
      "entropy": 6.390422391891479,
      "epoch": 1.1519610466049548,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048772563088035533,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15505800396203995,
      "num_tokens": 23288159.0,
      "step": 10765
    },
    {
      "entropy": 6.3896759986877445,
      "epoch": 1.1524961207127187,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004877131723277732,
      "loss": 6.0365,
      "mean_token_accuracy": 0.14486374631524085,
      "num_tokens": 23299379.0,
      "step": 10770
    },
    {
      "entropy": 6.436257076263428,
      "epoch": 1.1530311948204826,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004877007076331282,
      "loss": 6.0734,
      "mean_token_accuracy": 0.1472310595214367,
      "num_tokens": 23310572.0,
      "step": 10775
    },
    {
      "entropy": 6.497855949401855,
      "epoch": 1.1535662689282467,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004876882367967801,
      "loss": 6.049,
      "mean_token_accuracy": 0.15402799546718599,
      "num_tokens": 23320782.0,
      "step": 10780
    },
    {
      "entropy": 6.503010988235474,
      "epoch": 1.1541013430360105,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048767575981908907,
      "loss": 6.1285,
      "mean_token_accuracy": 0.1449247770011425,
      "num_tokens": 23331427.0,
      "step": 10785
    },
    {
      "entropy": 6.463255071640015,
      "epoch": 1.1546364171437744,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048766327670041534,
      "loss": 6.2422,
      "mean_token_accuracy": 0.1385998956859112,
      "num_tokens": 23344250.0,
      "step": 10790
    },
    {
      "entropy": 6.453471755981445,
      "epoch": 1.1551714912515383,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004876507874411194,
      "loss": 6.0793,
      "mean_token_accuracy": 0.1520368903875351,
      "num_tokens": 23354404.0,
      "step": 10795
    },
    {
      "entropy": 6.4722288131713865,
      "epoch": 1.1557065653593024,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048763829204156187,
      "loss": 6.1042,
      "mean_token_accuracy": 0.1406518802046776,
      "num_tokens": 23364634.0,
      "step": 10800
    },
    {
      "entropy": 6.372135829925537,
      "epoch": 1.1562416394670663,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048762579050210344,
      "loss": 5.963,
      "mean_token_accuracy": 0.15279010832309722,
      "num_tokens": 23375816.0,
      "step": 10805
    },
    {
      "entropy": 6.4805442810058596,
      "epoch": 1.1567767135748301,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048761328282310534,
      "loss": 6.1059,
      "mean_token_accuracy": 0.14840065985918044,
      "num_tokens": 23386583.0,
      "step": 10810
    },
    {
      "entropy": 6.394677257537841,
      "epoch": 1.157311787682594,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048760076900492846,
      "loss": 6.0292,
      "mean_token_accuracy": 0.15753632560372352,
      "num_tokens": 23396826.0,
      "step": 10815
    },
    {
      "entropy": 6.449412679672241,
      "epoch": 1.157846861790358,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004875882490479343,
      "loss": 6.1105,
      "mean_token_accuracy": 0.15231072157621384,
      "num_tokens": 23408661.0,
      "step": 10820
    },
    {
      "entropy": 6.422153186798096,
      "epoch": 1.158381935898122,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048757572295248425,
      "loss": 6.0275,
      "mean_token_accuracy": 0.14887704849243164,
      "num_tokens": 23419308.0,
      "step": 10825
    },
    {
      "entropy": 6.3802698135375975,
      "epoch": 1.1589170100058859,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004875631907189402,
      "loss": 6.0182,
      "mean_token_accuracy": 0.16104642450809478,
      "num_tokens": 23430043.0,
      "step": 10830
    },
    {
      "entropy": 6.36746654510498,
      "epoch": 1.1594520841136498,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004875506523476638,
      "loss": 5.9477,
      "mean_token_accuracy": 0.15474530458450317,
      "num_tokens": 23441246.0,
      "step": 10835
    },
    {
      "entropy": 6.485027217864991,
      "epoch": 1.1599871582214136,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048753810783901716,
      "loss": 6.1044,
      "mean_token_accuracy": 0.14729998707771302,
      "num_tokens": 23452250.0,
      "step": 10840
    },
    {
      "entropy": 6.453603839874267,
      "epoch": 1.1605222323291775,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048752555719336257,
      "loss": 6.0772,
      "mean_token_accuracy": 0.15116316229104995,
      "num_tokens": 23464174.0,
      "step": 10845
    },
    {
      "entropy": 6.366837453842163,
      "epoch": 1.1610573064369416,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004875130004110623,
      "loss": 5.9586,
      "mean_token_accuracy": 0.15381108969449997,
      "num_tokens": 23475159.0,
      "step": 10850
    },
    {
      "entropy": 6.3404120922088625,
      "epoch": 1.1615923805447055,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048750043749247907,
      "loss": 6.0178,
      "mean_token_accuracy": 0.1559457913041115,
      "num_tokens": 23486210.0,
      "step": 10855
    },
    {
      "entropy": 6.384507942199707,
      "epoch": 1.1621274546524694,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004874878684379756,
      "loss": 6.1167,
      "mean_token_accuracy": 0.15019274055957793,
      "num_tokens": 23497122.0,
      "step": 10860
    },
    {
      "entropy": 6.420570230484008,
      "epoch": 1.1626625287602332,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004874752932479148,
      "loss": 5.9687,
      "mean_token_accuracy": 0.15829487442970275,
      "num_tokens": 23507765.0,
      "step": 10865
    },
    {
      "entropy": 6.48632779121399,
      "epoch": 1.1631976028679971,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048746271192265974,
      "loss": 6.0518,
      "mean_token_accuracy": 0.14846138283610344,
      "num_tokens": 23517686.0,
      "step": 10870
    },
    {
      "entropy": 6.373788785934448,
      "epoch": 1.1637326769757612,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004874501244625737,
      "loss": 6.0202,
      "mean_token_accuracy": 0.15190632790327072,
      "num_tokens": 23528325.0,
      "step": 10875
    },
    {
      "entropy": 6.399011945724487,
      "epoch": 1.164267751083525,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004874375308680202,
      "loss": 6.0953,
      "mean_token_accuracy": 0.14939468652009963,
      "num_tokens": 23540006.0,
      "step": 10880
    },
    {
      "entropy": 6.496952390670776,
      "epoch": 1.164802825191289,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048742493113936274,
      "loss": 6.0054,
      "mean_token_accuracy": 0.14794512689113617,
      "num_tokens": 23550419.0,
      "step": 10885
    },
    {
      "entropy": 6.504558753967285,
      "epoch": 1.1653378992990528,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004874123252769653,
      "loss": 6.1161,
      "mean_token_accuracy": 0.14421150386333464,
      "num_tokens": 23561227.0,
      "step": 10890
    },
    {
      "entropy": 6.418510055541992,
      "epoch": 1.1658729734068167,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048739971328119194,
      "loss": 6.0484,
      "mean_token_accuracy": 0.15052332282066344,
      "num_tokens": 23571357.0,
      "step": 10895
    },
    {
      "entropy": 6.446261215209961,
      "epoch": 1.1664080475145808,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004873870951524066,
      "loss": 6.0785,
      "mean_token_accuracy": 0.14799331724643708,
      "num_tokens": 23581875.0,
      "step": 10900
    },
    {
      "entropy": 6.465222358703613,
      "epoch": 1.1669431216223447,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048737447089097373,
      "loss": 6.1338,
      "mean_token_accuracy": 0.14500605091452598,
      "num_tokens": 23594251.0,
      "step": 10905
    },
    {
      "entropy": 6.465814876556396,
      "epoch": 1.1674781957301086,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004873618404972579,
      "loss": 6.0921,
      "mean_token_accuracy": 0.14288632869720458,
      "num_tokens": 23605159.0,
      "step": 10910
    },
    {
      "entropy": 6.488463592529297,
      "epoch": 1.1680132698378725,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048734920397162376,
      "loss": 6.0064,
      "mean_token_accuracy": 0.15894824713468553,
      "num_tokens": 23615737.0,
      "step": 10915
    },
    {
      "entropy": 6.4052753925323485,
      "epoch": 1.1685483439456366,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004873365613144361,
      "loss": 6.0443,
      "mean_token_accuracy": 0.15159929990768434,
      "num_tokens": 23626207.0,
      "step": 10920
    },
    {
      "entropy": 6.465179443359375,
      "epoch": 1.1690834180534004,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004873239125260602,
      "loss": 6.0928,
      "mean_token_accuracy": 0.15343543291091918,
      "num_tokens": 23637900.0,
      "step": 10925
    },
    {
      "entropy": 6.4907252311706545,
      "epoch": 1.1696184921611643,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048731125760686106,
      "loss": 5.9701,
      "mean_token_accuracy": 0.15678429305553437,
      "num_tokens": 23647867.0,
      "step": 10930
    },
    {
      "entropy": 6.461256313323974,
      "epoch": 1.1701535662689282,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048729859655720424,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15132727473974228,
      "num_tokens": 23658055.0,
      "step": 10935
    },
    {
      "entropy": 6.337489748001099,
      "epoch": 1.1706886403766923,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004872859293774553,
      "loss": 6.0465,
      "mean_token_accuracy": 0.15058431327342986,
      "num_tokens": 23669458.0,
      "step": 10940
    },
    {
      "entropy": 6.388458490371704,
      "epoch": 1.1712237144844562,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048727325606797995,
      "loss": 6.0464,
      "mean_token_accuracy": 0.14643949568271636,
      "num_tokens": 23679702.0,
      "step": 10945
    },
    {
      "entropy": 6.480168628692627,
      "epoch": 1.17175878859222,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004872605766291441,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1538378491997719,
      "num_tokens": 23690844.0,
      "step": 10950
    },
    {
      "entropy": 6.372534894943238,
      "epoch": 1.172293862699984,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000487247891061314,
      "loss": 6.0334,
      "mean_token_accuracy": 0.15365362986922265,
      "num_tokens": 23701484.0,
      "step": 10955
    },
    {
      "entropy": 6.447096824645996,
      "epoch": 1.1728289368077478,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004872351993648559,
      "loss": 6.0684,
      "mean_token_accuracy": 0.14949404299259186,
      "num_tokens": 23711452.0,
      "step": 10960
    },
    {
      "entropy": 6.44826865196228,
      "epoch": 1.173364010915512,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048722250154013613,
      "loss": 6.0131,
      "mean_token_accuracy": 0.14880230724811555,
      "num_tokens": 23723514.0,
      "step": 10965
    },
    {
      "entropy": 6.450251531600952,
      "epoch": 1.1738990850232758,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004872097975875216,
      "loss": 6.0941,
      "mean_token_accuracy": 0.15424433797597886,
      "num_tokens": 23733722.0,
      "step": 10970
    },
    {
      "entropy": 6.48226318359375,
      "epoch": 1.1744341591310397,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004871970875073789,
      "loss": 6.0579,
      "mean_token_accuracy": 0.14862788021564483,
      "num_tokens": 23743403.0,
      "step": 10975
    },
    {
      "entropy": 6.426197052001953,
      "epoch": 1.1749692332388035,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004871843713000751,
      "loss": 6.0156,
      "mean_token_accuracy": 0.15023545622825624,
      "num_tokens": 23754558.0,
      "step": 10980
    },
    {
      "entropy": 6.471014785766601,
      "epoch": 1.1755043073465674,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048717164896597737,
      "loss": 6.0834,
      "mean_token_accuracy": 0.15474623143672944,
      "num_tokens": 23766346.0,
      "step": 10985
    },
    {
      "entropy": 6.477969932556152,
      "epoch": 1.1760393814543315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004871589205054532,
      "loss": 6.076,
      "mean_token_accuracy": 0.15379417836666107,
      "num_tokens": 23776851.0,
      "step": 10990
    },
    {
      "entropy": 6.439497947692871,
      "epoch": 1.1765744555620954,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048714618591886997,
      "loss": 6.1216,
      "mean_token_accuracy": 0.14844486713409424,
      "num_tokens": 23787350.0,
      "step": 10995
    },
    {
      "entropy": 6.362220573425293,
      "epoch": 1.1771095296698593,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004871334452065954,
      "loss": 5.9855,
      "mean_token_accuracy": 0.15159978866577148,
      "num_tokens": 23797738.0,
      "step": 11000
    },
    {
      "entropy": 6.424336576461792,
      "epoch": 1.1776446037776231,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004871206983689974,
      "loss": 5.971,
      "mean_token_accuracy": 0.1643219843506813,
      "num_tokens": 23809516.0,
      "step": 11005
    },
    {
      "entropy": 6.431756162643433,
      "epoch": 1.178179677885387,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000487107945406444,
      "loss": 6.0541,
      "mean_token_accuracy": 0.15629515945911407,
      "num_tokens": 23819850.0,
      "step": 11010
    },
    {
      "entropy": 6.393584156036377,
      "epoch": 1.1787147519931511,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004870951863193036,
      "loss": 6.0464,
      "mean_token_accuracy": 0.1514424443244934,
      "num_tokens": 23829891.0,
      "step": 11015
    },
    {
      "entropy": 6.397791385650635,
      "epoch": 1.179249826100915,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004870824211079444,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1531070053577423,
      "num_tokens": 23840816.0,
      "step": 11020
    },
    {
      "entropy": 6.524810886383056,
      "epoch": 1.1797849002086789,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048706964977273515,
      "loss": 6.09,
      "mean_token_accuracy": 0.1509687826037407,
      "num_tokens": 23852595.0,
      "step": 11025
    },
    {
      "entropy": 6.493875980377197,
      "epoch": 1.1803199743164428,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048705687231404455,
      "loss": 6.0734,
      "mean_token_accuracy": 0.14644408375024795,
      "num_tokens": 23863048.0,
      "step": 11030
    },
    {
      "entropy": 6.463421964645386,
      "epoch": 1.1808550484242066,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048704408873224156,
      "loss": 6.0352,
      "mean_token_accuracy": 0.1540014013648033,
      "num_tokens": 23873499.0,
      "step": 11035
    },
    {
      "entropy": 6.3715019702911375,
      "epoch": 1.1813901225319707,
      "grad_norm": 1.0,
      "learning_rate": 0.00048703129902769526,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15743489116430281,
      "num_tokens": 23885129.0,
      "step": 11040
    },
    {
      "entropy": 6.4595049858093265,
      "epoch": 1.1819251966397346,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004870185032007751,
      "loss": 6.0927,
      "mean_token_accuracy": 0.14818481355905533,
      "num_tokens": 23895538.0,
      "step": 11045
    },
    {
      "entropy": 6.464940881729126,
      "epoch": 1.1824602707474985,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004870057012518504,
      "loss": 6.1261,
      "mean_token_accuracy": 0.1414801925420761,
      "num_tokens": 23907341.0,
      "step": 11050
    },
    {
      "entropy": 6.4046837329864506,
      "epoch": 1.1829953448552624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048699289318129087,
      "loss": 5.9599,
      "mean_token_accuracy": 0.15888493061065673,
      "num_tokens": 23918501.0,
      "step": 11055
    },
    {
      "entropy": 6.456871604919433,
      "epoch": 1.1835304189630265,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004869800789894663,
      "loss": 6.0441,
      "mean_token_accuracy": 0.15033092945814133,
      "num_tokens": 23929204.0,
      "step": 11060
    },
    {
      "entropy": 6.427889442443847,
      "epoch": 1.1840654930707903,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004869672586767468,
      "loss": 6.0818,
      "mean_token_accuracy": 0.15578549206256867,
      "num_tokens": 23939855.0,
      "step": 11065
    },
    {
      "entropy": 6.49022479057312,
      "epoch": 1.1846005671785542,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004869544322435024,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15362068712711335,
      "num_tokens": 23949871.0,
      "step": 11070
    },
    {
      "entropy": 6.465183353424072,
      "epoch": 1.185135641286318,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004869415996901036,
      "loss": 6.1399,
      "mean_token_accuracy": 0.1459863841533661,
      "num_tokens": 23961650.0,
      "step": 11075
    },
    {
      "entropy": 6.439829683303833,
      "epoch": 1.1856707153940822,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004869287610169209,
      "loss": 6.1244,
      "mean_token_accuracy": 0.14908740743994714,
      "num_tokens": 23973103.0,
      "step": 11080
    },
    {
      "entropy": 6.440314865112304,
      "epoch": 1.186205789501846,
      "grad_norm": 0.9296875,
      "learning_rate": 0.000486915916224325,
      "loss": 6.0521,
      "mean_token_accuracy": 0.1492658242583275,
      "num_tokens": 23982972.0,
      "step": 11085
    },
    {
      "entropy": 6.359928703308105,
      "epoch": 1.18674086360961,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004869030653126868,
      "loss": 5.9413,
      "mean_token_accuracy": 0.1565922185778618,
      "num_tokens": 23993044.0,
      "step": 11090
    },
    {
      "entropy": 6.457144451141358,
      "epoch": 1.1872759377173738,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004868902082823774,
      "loss": 6.0044,
      "mean_token_accuracy": 0.15196296125650405,
      "num_tokens": 24003871.0,
      "step": 11095
    },
    {
      "entropy": 6.458668184280396,
      "epoch": 1.1878110118251377,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004868773451337679,
      "loss": 6.0639,
      "mean_token_accuracy": 0.14637100100517272,
      "num_tokens": 24014613.0,
      "step": 11100
    },
    {
      "entropy": 6.412311792373657,
      "epoch": 1.1883460859329018,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048686447586722995,
      "loss": 6.0757,
      "mean_token_accuracy": 0.15174557641148567,
      "num_tokens": 24024172.0,
      "step": 11105
    },
    {
      "entropy": 6.48825101852417,
      "epoch": 1.1888811600406657,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048685160048313506,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1455383963882923,
      "num_tokens": 24035301.0,
      "step": 11110
    },
    {
      "entropy": 6.494930219650269,
      "epoch": 1.1894162341484296,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048683871898185486,
      "loss": 6.0516,
      "mean_token_accuracy": 0.15569785237312317,
      "num_tokens": 24045661.0,
      "step": 11115
    },
    {
      "entropy": 6.438175868988037,
      "epoch": 1.1899513082561934,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048682583136376145,
      "loss": 6.0423,
      "mean_token_accuracy": 0.1516675606369972,
      "num_tokens": 24057027.0,
      "step": 11120
    },
    {
      "entropy": 6.4651689529418945,
      "epoch": 1.1904863823639573,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004868129376292269,
      "loss": 6.168,
      "mean_token_accuracy": 0.14580827355384826,
      "num_tokens": 24068322.0,
      "step": 11125
    },
    {
      "entropy": 6.468797302246093,
      "epoch": 1.1910214564717214,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048680003777862356,
      "loss": 5.9974,
      "mean_token_accuracy": 0.15292215943336487,
      "num_tokens": 24078506.0,
      "step": 11130
    },
    {
      "entropy": 6.438296890258789,
      "epoch": 1.1915565305794853,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048678713181232394,
      "loss": 6.0343,
      "mean_token_accuracy": 0.15208195596933366,
      "num_tokens": 24090296.0,
      "step": 11135
    },
    {
      "entropy": 6.374386644363403,
      "epoch": 1.1920916046872492,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048677421973070063,
      "loss": 6.0443,
      "mean_token_accuracy": 0.14469144716858864,
      "num_tokens": 24101478.0,
      "step": 11140
    },
    {
      "entropy": 6.453811120986939,
      "epoch": 1.192626678795013,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048676130153412644,
      "loss": 6.0681,
      "mean_token_accuracy": 0.15139658153057098,
      "num_tokens": 24112552.0,
      "step": 11145
    },
    {
      "entropy": 6.356844568252564,
      "epoch": 1.193161752902777,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004867483772229745,
      "loss": 5.9399,
      "mean_token_accuracy": 0.1556419163942337,
      "num_tokens": 24124210.0,
      "step": 11150
    },
    {
      "entropy": 6.447923374176026,
      "epoch": 1.193696827010541,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004867354467976178,
      "loss": 6.0453,
      "mean_token_accuracy": 0.15147082954645158,
      "num_tokens": 24133525.0,
      "step": 11155
    },
    {
      "entropy": 6.467784261703491,
      "epoch": 1.194231901118305,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048672251025842994,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14874400049448014,
      "num_tokens": 24145400.0,
      "step": 11160
    },
    {
      "entropy": 6.4576493263244625,
      "epoch": 1.1947669752260688,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004867095676057843,
      "loss": 6.0812,
      "mean_token_accuracy": 0.1450783997774124,
      "num_tokens": 24155596.0,
      "step": 11165
    },
    {
      "entropy": 6.474843740463257,
      "epoch": 1.1953020493338327,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048669661884005467,
      "loss": 6.095,
      "mean_token_accuracy": 0.1418712928891182,
      "num_tokens": 24166683.0,
      "step": 11170
    },
    {
      "entropy": 6.479637908935547,
      "epoch": 1.1958371234415965,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048668366396161503,
      "loss": 6.0112,
      "mean_token_accuracy": 0.15480864942073821,
      "num_tokens": 24177649.0,
      "step": 11175
    },
    {
      "entropy": 6.443119621276855,
      "epoch": 1.1963721975493606,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004866707029708392,
      "loss": 6.0614,
      "mean_token_accuracy": 0.16101958900690078,
      "num_tokens": 24188522.0,
      "step": 11180
    },
    {
      "entropy": 6.424028301239014,
      "epoch": 1.1969072716571245,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004866577358681015,
      "loss": 6.0257,
      "mean_token_accuracy": 0.15157176703214645,
      "num_tokens": 24199285.0,
      "step": 11185
    },
    {
      "entropy": 6.390493392944336,
      "epoch": 1.1974423457648884,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004866447626537765,
      "loss": 5.987,
      "mean_token_accuracy": 0.1504751116037369,
      "num_tokens": 24210929.0,
      "step": 11190
    },
    {
      "entropy": 6.420280265808105,
      "epoch": 1.1979774198726523,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004866317833282387,
      "loss": 6.0742,
      "mean_token_accuracy": 0.15018317624926567,
      "num_tokens": 24221394.0,
      "step": 11195
    },
    {
      "entropy": 6.435041618347168,
      "epoch": 1.1985124939804164,
      "grad_norm": 1.140625,
      "learning_rate": 0.00048661879789186295,
      "loss": 6.0244,
      "mean_token_accuracy": 0.15483270287513734,
      "num_tokens": 24230297.0,
      "step": 11200
    },
    {
      "entropy": 6.420574760437011,
      "epoch": 1.1990475680881802,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048660580634502415,
      "loss": 6.167,
      "mean_token_accuracy": 0.14766187146306037,
      "num_tokens": 24241505.0,
      "step": 11205
    },
    {
      "entropy": 6.408963394165039,
      "epoch": 1.1995826421959441,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048659280868809744,
      "loss": 6.0165,
      "mean_token_accuracy": 0.15162643641233445,
      "num_tokens": 24252638.0,
      "step": 11210
    },
    {
      "entropy": 6.404865312576294,
      "epoch": 1.200117716303708,
      "grad_norm": 1.0,
      "learning_rate": 0.000486579804921458,
      "loss": 6.061,
      "mean_token_accuracy": 0.15557684600353242,
      "num_tokens": 24263156.0,
      "step": 11215
    },
    {
      "entropy": 6.377303171157837,
      "epoch": 1.200652790411472,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048656679504548145,
      "loss": 5.9287,
      "mean_token_accuracy": 0.16459331214427947,
      "num_tokens": 24274384.0,
      "step": 11220
    },
    {
      "entropy": 6.513161277770996,
      "epoch": 1.201187864519236,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004865537790605435,
      "loss": 6.1531,
      "mean_token_accuracy": 0.143333275616169,
      "num_tokens": 24285159.0,
      "step": 11225
    },
    {
      "entropy": 6.476498651504516,
      "epoch": 1.2017229386269999,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004865407569670198,
      "loss": 6.0909,
      "mean_token_accuracy": 0.1453853964805603,
      "num_tokens": 24296202.0,
      "step": 11230
    },
    {
      "entropy": 6.448228645324707,
      "epoch": 1.2022580127347637,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004865277287652866,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15065984055399895,
      "num_tokens": 24306691.0,
      "step": 11235
    },
    {
      "entropy": 6.411701011657715,
      "epoch": 1.2027930868425276,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004865146944557199,
      "loss": 6.0538,
      "mean_token_accuracy": 0.15151648819446564,
      "num_tokens": 24317499.0,
      "step": 11240
    },
    {
      "entropy": 6.447541999816894,
      "epoch": 1.2033281609502917,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004865016540386961,
      "loss": 6.0272,
      "mean_token_accuracy": 0.14429401308298112,
      "num_tokens": 24328096.0,
      "step": 11245
    },
    {
      "entropy": 6.444887781143189,
      "epoch": 1.2038632350580556,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048648860751459173,
      "loss": 6.0458,
      "mean_token_accuracy": 0.1462721660733223,
      "num_tokens": 24338188.0,
      "step": 11250
    },
    {
      "entropy": 6.436428880691528,
      "epoch": 1.2043983091658195,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048647555488378355,
      "loss": 6.0734,
      "mean_token_accuracy": 0.1602361723780632,
      "num_tokens": 24348578.0,
      "step": 11255
    },
    {
      "entropy": 6.408139324188232,
      "epoch": 1.2049333832735833,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004864624961466485,
      "loss": 6.0522,
      "mean_token_accuracy": 0.1554632969200611,
      "num_tokens": 24358807.0,
      "step": 11260
    },
    {
      "entropy": 6.4370527267456055,
      "epoch": 1.2054684573813472,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004864494313035635,
      "loss": 6.0646,
      "mean_token_accuracy": 0.145206705480814,
      "num_tokens": 24369074.0,
      "step": 11265
    },
    {
      "entropy": 6.448667812347412,
      "epoch": 1.2060035314891113,
      "grad_norm": 0.90625,
      "learning_rate": 0.000486436360354906,
      "loss": 6.0289,
      "mean_token_accuracy": 0.15937023758888244,
      "num_tokens": 24380655.0,
      "step": 11270
    },
    {
      "entropy": 6.405341863632202,
      "epoch": 1.2065386055968752,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048642328330105323,
      "loss": 6.0084,
      "mean_token_accuracy": 0.15117370784282685,
      "num_tokens": 24391701.0,
      "step": 11275
    },
    {
      "entropy": 6.350640439987183,
      "epoch": 1.207073679704639,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048641020014238286,
      "loss": 6.1055,
      "mean_token_accuracy": 0.14622513055801392,
      "num_tokens": 24403690.0,
      "step": 11280
    },
    {
      "entropy": 6.5028821468353275,
      "epoch": 1.207608753812403,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004863971108792727,
      "loss": 6.0782,
      "mean_token_accuracy": 0.14830457121133805,
      "num_tokens": 24414670.0,
      "step": 11285
    },
    {
      "entropy": 6.505238008499146,
      "epoch": 1.2081438279201668,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048638401551210063,
      "loss": 6.1272,
      "mean_token_accuracy": 0.14568008184432985,
      "num_tokens": 24425237.0,
      "step": 11290
    },
    {
      "entropy": 6.457878446578979,
      "epoch": 1.208678902027931,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048637091404124484,
      "loss": 6.0794,
      "mean_token_accuracy": 0.15169409960508345,
      "num_tokens": 24437142.0,
      "step": 11295
    },
    {
      "entropy": 6.41042160987854,
      "epoch": 1.2092139761356948,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004863578064670837,
      "loss": 5.9887,
      "mean_token_accuracy": 0.15622535347938538,
      "num_tokens": 24448819.0,
      "step": 11300
    },
    {
      "entropy": 6.372686815261841,
      "epoch": 1.2097490502434587,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00048634469278999545,
      "loss": 5.9385,
      "mean_token_accuracy": 0.16393271535634996,
      "num_tokens": 24459392.0,
      "step": 11305
    },
    {
      "entropy": 6.453218460083008,
      "epoch": 1.2102841243512226,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048633157301035895,
      "loss": 6.0454,
      "mean_token_accuracy": 0.14742689952254295,
      "num_tokens": 24470298.0,
      "step": 11310
    },
    {
      "entropy": 6.4517792701721195,
      "epoch": 1.2108191984589864,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000486318447128553,
      "loss": 6.0206,
      "mean_token_accuracy": 0.15606458187103273,
      "num_tokens": 24480477.0,
      "step": 11315
    },
    {
      "entropy": 6.426034069061279,
      "epoch": 1.2113542725667505,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004863053151449566,
      "loss": 6.0997,
      "mean_token_accuracy": 0.1392226055264473,
      "num_tokens": 24491890.0,
      "step": 11320
    },
    {
      "entropy": 6.482469272613526,
      "epoch": 1.2118893466745144,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048629217705994883,
      "loss": 6.036,
      "mean_token_accuracy": 0.15356713458895682,
      "num_tokens": 24502583.0,
      "step": 11325
    },
    {
      "entropy": 6.46385931968689,
      "epoch": 1.2124244207822783,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004862790328739091,
      "loss": 6.1352,
      "mean_token_accuracy": 0.14395268484950066,
      "num_tokens": 24514491.0,
      "step": 11330
    },
    {
      "entropy": 6.384811925888061,
      "epoch": 1.2129594948900422,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048626588258721717,
      "loss": 5.9564,
      "mean_token_accuracy": 0.1573146991431713,
      "num_tokens": 24524788.0,
      "step": 11335
    },
    {
      "entropy": 6.431461572647095,
      "epoch": 1.2134945689978063,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004862527262002524,
      "loss": 5.9998,
      "mean_token_accuracy": 0.14600226208567618,
      "num_tokens": 24535699.0,
      "step": 11340
    },
    {
      "entropy": 6.407611513137818,
      "epoch": 1.2140296431055702,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048623956371339494,
      "loss": 6.0184,
      "mean_token_accuracy": 0.1541600078344345,
      "num_tokens": 24547339.0,
      "step": 11345
    },
    {
      "entropy": 6.429941463470459,
      "epoch": 1.214564717213334,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048622639512702477,
      "loss": 6.1282,
      "mean_token_accuracy": 0.14357974976301194,
      "num_tokens": 24558778.0,
      "step": 11350
    },
    {
      "entropy": 6.451292991638184,
      "epoch": 1.215099791321098,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048621322044152207,
      "loss": 6.0434,
      "mean_token_accuracy": 0.1508875846862793,
      "num_tokens": 24569032.0,
      "step": 11355
    },
    {
      "entropy": 6.451923894882202,
      "epoch": 1.215634865428862,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048620003965726727,
      "loss": 6.1425,
      "mean_token_accuracy": 0.13858878761529922,
      "num_tokens": 24581356.0,
      "step": 11360
    },
    {
      "entropy": 6.409787034988403,
      "epoch": 1.2161699395366259,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048618685277464116,
      "loss": 6.0182,
      "mean_token_accuracy": 0.15386898517608644,
      "num_tokens": 24591809.0,
      "step": 11365
    },
    {
      "entropy": 6.354791307449341,
      "epoch": 1.2167050136443898,
      "grad_norm": 0.875,
      "learning_rate": 0.00048617365979402417,
      "loss": 5.9184,
      "mean_token_accuracy": 0.15894464403390884,
      "num_tokens": 24603492.0,
      "step": 11370
    },
    {
      "entropy": 6.436268472671509,
      "epoch": 1.2172400877521536,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004861604607157975,
      "loss": 6.0817,
      "mean_token_accuracy": 0.14820944964885713,
      "num_tokens": 24614837.0,
      "step": 11375
    },
    {
      "entropy": 6.4668169021606445,
      "epoch": 1.2177751618599175,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004861472555403423,
      "loss": 6.0766,
      "mean_token_accuracy": 0.14768256843090058,
      "num_tokens": 24624714.0,
      "step": 11380
    },
    {
      "entropy": 6.473559999465943,
      "epoch": 1.2183102359676816,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048613404426803964,
      "loss": 6.1114,
      "mean_token_accuracy": 0.14593157917261124,
      "num_tokens": 24635758.0,
      "step": 11385
    },
    {
      "entropy": 6.507129096984864,
      "epoch": 1.2188453100754455,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048612082689927116,
      "loss": 6.089,
      "mean_token_accuracy": 0.1490107871592045,
      "num_tokens": 24645362.0,
      "step": 11390
    },
    {
      "entropy": 6.451828718185425,
      "epoch": 1.2193803841832094,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004861076034344185,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14778612405061722,
      "num_tokens": 24656898.0,
      "step": 11395
    },
    {
      "entropy": 6.386983633041382,
      "epoch": 1.2199154582909733,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004860943738738634,
      "loss": 6.015,
      "mean_token_accuracy": 0.15682487487792968,
      "num_tokens": 24668191.0,
      "step": 11400
    },
    {
      "entropy": 6.4507026195526125,
      "epoch": 1.2204505323987371,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00048608113821798786,
      "loss": 6.0587,
      "mean_token_accuracy": 0.14989481419324874,
      "num_tokens": 24678034.0,
      "step": 11405
    },
    {
      "entropy": 6.465731048583985,
      "epoch": 1.2209856065065012,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004860678964671743,
      "loss": 6.0832,
      "mean_token_accuracy": 0.15081679373979567,
      "num_tokens": 24689062.0,
      "step": 11410
    },
    {
      "entropy": 6.509672069549561,
      "epoch": 1.221520680614265,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048605464862180474,
      "loss": 6.047,
      "mean_token_accuracy": 0.14699062705039978,
      "num_tokens": 24699787.0,
      "step": 11415
    },
    {
      "entropy": 6.438560056686401,
      "epoch": 1.222055754722029,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004860413946822619,
      "loss": 6.0049,
      "mean_token_accuracy": 0.14929670542478563,
      "num_tokens": 24711233.0,
      "step": 11420
    },
    {
      "entropy": 6.395307207107544,
      "epoch": 1.2225908288297929,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004860281346489284,
      "loss": 6.0495,
      "mean_token_accuracy": 0.1426271140575409,
      "num_tokens": 24722414.0,
      "step": 11425
    },
    {
      "entropy": 6.474143981933594,
      "epoch": 1.2231259029375567,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004860148685221873,
      "loss": 6.063,
      "mean_token_accuracy": 0.14952090233564377,
      "num_tokens": 24733093.0,
      "step": 11430
    },
    {
      "entropy": 6.400910997390747,
      "epoch": 1.2236609770453208,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048600159630242135,
      "loss": 6.0075,
      "mean_token_accuracy": 0.15858413130044938,
      "num_tokens": 24743195.0,
      "step": 11435
    },
    {
      "entropy": 6.463865900039673,
      "epoch": 1.2241960511530847,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048598831799001406,
      "loss": 6.0239,
      "mean_token_accuracy": 0.15156161040067673,
      "num_tokens": 24753680.0,
      "step": 11440
    },
    {
      "entropy": 6.439968919754028,
      "epoch": 1.2247311252608486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004859750335853488,
      "loss": 6.0075,
      "mean_token_accuracy": 0.15582909286022187,
      "num_tokens": 24764135.0,
      "step": 11445
    },
    {
      "entropy": 6.3277181625366214,
      "epoch": 1.2252661993686125,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000485961743088809,
      "loss": 5.9506,
      "mean_token_accuracy": 0.1694600835442543,
      "num_tokens": 24775740.0,
      "step": 11450
    },
    {
      "entropy": 6.44970006942749,
      "epoch": 1.2258012734763766,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004859484465007784,
      "loss": 6.1045,
      "mean_token_accuracy": 0.14759573638439177,
      "num_tokens": 24786169.0,
      "step": 11455
    },
    {
      "entropy": 6.434850025177002,
      "epoch": 1.2263363475841405,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048593514382164123,
      "loss": 5.9837,
      "mean_token_accuracy": 0.15727620720863342,
      "num_tokens": 24796218.0,
      "step": 11460
    },
    {
      "entropy": 6.492648553848267,
      "epoch": 1.2268714216919043,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004859218350517814,
      "loss": 6.1233,
      "mean_token_accuracy": 0.14665577784180642,
      "num_tokens": 24806957.0,
      "step": 11465
    },
    {
      "entropy": 6.509902763366699,
      "epoch": 1.2274064957996682,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004859085201915831,
      "loss": 6.0625,
      "mean_token_accuracy": 0.1473672591149807,
      "num_tokens": 24818761.0,
      "step": 11470
    },
    {
      "entropy": 6.433614778518677,
      "epoch": 1.227941569907432,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004858951992414311,
      "loss": 5.9685,
      "mean_token_accuracy": 0.1604271337389946,
      "num_tokens": 24829163.0,
      "step": 11475
    },
    {
      "entropy": 6.391783905029297,
      "epoch": 1.2284766440151962,
      "grad_norm": 1.0,
      "learning_rate": 0.0004858818722017097,
      "loss": 5.9982,
      "mean_token_accuracy": 0.1541475310921669,
      "num_tokens": 24839334.0,
      "step": 11480
    },
    {
      "entropy": 6.356706190109253,
      "epoch": 1.22901171812296,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048586853907280395,
      "loss": 6.0313,
      "mean_token_accuracy": 0.15621337294578552,
      "num_tokens": 24850312.0,
      "step": 11485
    },
    {
      "entropy": 6.4042762279510494,
      "epoch": 1.229546792230724,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004858551998550987,
      "loss": 5.9679,
      "mean_token_accuracy": 0.1612516313791275,
      "num_tokens": 24861835.0,
      "step": 11490
    },
    {
      "entropy": 6.460929346084595,
      "epoch": 1.2300818663384878,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048584185454897925,
      "loss": 6.0696,
      "mean_token_accuracy": 0.1503744199872017,
      "num_tokens": 24872489.0,
      "step": 11495
    },
    {
      "entropy": 6.43942346572876,
      "epoch": 1.230616940446252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004858285031548309,
      "loss": 6.0648,
      "mean_token_accuracy": 0.15106016993522645,
      "num_tokens": 24882553.0,
      "step": 11500
    },
    {
      "entropy": 6.473426771163941,
      "epoch": 1.2311520145540158,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004858151456730391,
      "loss": 6.0545,
      "mean_token_accuracy": 0.1527333989739418,
      "num_tokens": 24894386.0,
      "step": 11505
    },
    {
      "entropy": 6.422532558441162,
      "epoch": 1.2316870886617797,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048580178210398954,
      "loss": 6.0569,
      "mean_token_accuracy": 0.15017782598733903,
      "num_tokens": 24905049.0,
      "step": 11510
    },
    {
      "entropy": 6.3755429744720455,
      "epoch": 1.2322221627695435,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048578841244806823,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15683013796806336,
      "num_tokens": 24914831.0,
      "step": 11515
    },
    {
      "entropy": 6.450905418395996,
      "epoch": 1.2327572368773074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048577503670566116,
      "loss": 6.0697,
      "mean_token_accuracy": 0.1543118640780449,
      "num_tokens": 24925668.0,
      "step": 11520
    },
    {
      "entropy": 6.427761268615723,
      "epoch": 1.2332923109850715,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048576165487715443,
      "loss": 6.0523,
      "mean_token_accuracy": 0.1527203783392906,
      "num_tokens": 24935983.0,
      "step": 11525
    },
    {
      "entropy": 6.3934684753417965,
      "epoch": 1.2338273850928354,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048574826696293455,
      "loss": 6.0105,
      "mean_token_accuracy": 0.1605793297290802,
      "num_tokens": 24946673.0,
      "step": 11530
    },
    {
      "entropy": 6.410521984100342,
      "epoch": 1.2343624592005993,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048573487296338814,
      "loss": 5.9274,
      "mean_token_accuracy": 0.1642535611987114,
      "num_tokens": 24957075.0,
      "step": 11535
    },
    {
      "entropy": 6.415241861343384,
      "epoch": 1.2348975333083632,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048572147287890183,
      "loss": 6.0926,
      "mean_token_accuracy": 0.1484502986073494,
      "num_tokens": 24967261.0,
      "step": 11540
    },
    {
      "entropy": 6.4251689434051515,
      "epoch": 1.235432607416127,
      "grad_norm": 0.84765625,
      "learning_rate": 0.00048570806670986263,
      "loss": 6.0322,
      "mean_token_accuracy": 0.15339961946010588,
      "num_tokens": 24978097.0,
      "step": 11545
    },
    {
      "entropy": 6.445659732818603,
      "epoch": 1.2359676815238911,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048569465445665754,
      "loss": 6.0493,
      "mean_token_accuracy": 0.14872749894857407,
      "num_tokens": 24987932.0,
      "step": 11550
    },
    {
      "entropy": 6.442898464202881,
      "epoch": 1.236502755631655,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048568123611967396,
      "loss": 6.0303,
      "mean_token_accuracy": 0.15056683868169785,
      "num_tokens": 24998919.0,
      "step": 11555
    },
    {
      "entropy": 6.358511638641358,
      "epoch": 1.237037829739419,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004856678116992993,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15270962566137314,
      "num_tokens": 25010545.0,
      "step": 11560
    },
    {
      "entropy": 6.369730043411255,
      "epoch": 1.2375729038471828,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004856543811959212,
      "loss": 5.968,
      "mean_token_accuracy": 0.1635975420475006,
      "num_tokens": 25021133.0,
      "step": 11565
    },
    {
      "entropy": 6.416121292114258,
      "epoch": 1.2381079779549466,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004856409446099274,
      "loss": 6.0097,
      "mean_token_accuracy": 0.1515389308333397,
      "num_tokens": 25031838.0,
      "step": 11570
    },
    {
      "entropy": 6.4113880634307865,
      "epoch": 1.2386430520627107,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048562750194170595,
      "loss": 6.0423,
      "mean_token_accuracy": 0.14863985180854797,
      "num_tokens": 25042737.0,
      "step": 11575
    },
    {
      "entropy": 6.460899686813354,
      "epoch": 1.2391781261704746,
      "grad_norm": 0.921875,
      "learning_rate": 0.000485614053191645,
      "loss": 6.119,
      "mean_token_accuracy": 0.15056845545768738,
      "num_tokens": 25054442.0,
      "step": 11580
    },
    {
      "entropy": 6.369869089126587,
      "epoch": 1.2397132002782385,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004856005983601328,
      "loss": 5.8733,
      "mean_token_accuracy": 0.1647731527686119,
      "num_tokens": 25065509.0,
      "step": 11585
    },
    {
      "entropy": 6.5033973217010494,
      "epoch": 1.2402482743860024,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048558713744755794,
      "loss": 6.1347,
      "mean_token_accuracy": 0.14068517908453942,
      "num_tokens": 25077306.0,
      "step": 11590
    },
    {
      "entropy": 6.456674671173095,
      "epoch": 1.2407833484937665,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004855736704543091,
      "loss": 6.0511,
      "mean_token_accuracy": 0.1524140328168869,
      "num_tokens": 25088230.0,
      "step": 11595
    },
    {
      "entropy": 6.5009246349334715,
      "epoch": 1.2413184226015304,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004855601973807751,
      "loss": 6.124,
      "mean_token_accuracy": 0.14436076506972312,
      "num_tokens": 25099658.0,
      "step": 11600
    },
    {
      "entropy": 6.414297199249267,
      "epoch": 1.2418534967092942,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00048554671822734495,
      "loss": 6.0132,
      "mean_token_accuracy": 0.15196641236543657,
      "num_tokens": 25110512.0,
      "step": 11605
    },
    {
      "entropy": 6.4297562599182125,
      "epoch": 1.2423885708170581,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048553323299440807,
      "loss": 6.0413,
      "mean_token_accuracy": 0.1562209665775299,
      "num_tokens": 25120193.0,
      "step": 11610
    },
    {
      "entropy": 6.422174119949341,
      "epoch": 1.242923644924822,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048551974168235346,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15415681302547454,
      "num_tokens": 25131752.0,
      "step": 11615
    },
    {
      "entropy": 6.447862243652343,
      "epoch": 1.243458719032586,
      "grad_norm": 0.984375,
      "learning_rate": 0.000485506244291571,
      "loss": 6.0643,
      "mean_token_accuracy": 0.14912200719118118,
      "num_tokens": 25141541.0,
      "step": 11620
    },
    {
      "entropy": 6.356384563446045,
      "epoch": 1.24399379314035,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004854927408224503,
      "loss": 6.0487,
      "mean_token_accuracy": 0.1481182225048542,
      "num_tokens": 25152219.0,
      "step": 11625
    },
    {
      "entropy": 6.516604280471801,
      "epoch": 1.2445288672481138,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048547923127538126,
      "loss": 6.0576,
      "mean_token_accuracy": 0.14661484807729722,
      "num_tokens": 25163030.0,
      "step": 11630
    },
    {
      "entropy": 6.423681020736694,
      "epoch": 1.2450639413558777,
      "grad_norm": 0.859375,
      "learning_rate": 0.00048546571565075396,
      "loss": 5.9348,
      "mean_token_accuracy": 0.15883410722017288,
      "num_tokens": 25174637.0,
      "step": 11635
    },
    {
      "entropy": 6.36015100479126,
      "epoch": 1.2455990154636418,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004854521939489587,
      "loss": 5.9384,
      "mean_token_accuracy": 0.16270433068275453,
      "num_tokens": 25185067.0,
      "step": 11640
    },
    {
      "entropy": 6.389402389526367,
      "epoch": 1.2461340895714057,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048543866617038604,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1570732831954956,
      "num_tokens": 25195046.0,
      "step": 11645
    },
    {
      "entropy": 6.395199346542358,
      "epoch": 1.2466691636791696,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048542513231542635,
      "loss": 6.0534,
      "mean_token_accuracy": 0.1497385948896408,
      "num_tokens": 25206265.0,
      "step": 11650
    },
    {
      "entropy": 6.513322162628174,
      "epoch": 1.2472042377869335,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004854115923844705,
      "loss": 6.065,
      "mean_token_accuracy": 0.1459661528468132,
      "num_tokens": 25217690.0,
      "step": 11655
    },
    {
      "entropy": 6.516798257827759,
      "epoch": 1.2477393118946973,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004853980463779096,
      "loss": 6.0496,
      "mean_token_accuracy": 0.1488141432404518,
      "num_tokens": 25228229.0,
      "step": 11660
    },
    {
      "entropy": 6.3864569664001465,
      "epoch": 1.2482743860024614,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004853844942961346,
      "loss": 5.9913,
      "mean_token_accuracy": 0.1539249375462532,
      "num_tokens": 25240073.0,
      "step": 11665
    },
    {
      "entropy": 6.387154531478882,
      "epoch": 1.2488094601102253,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000485370936139537,
      "loss": 6.051,
      "mean_token_accuracy": 0.15638132840394975,
      "num_tokens": 25251915.0,
      "step": 11670
    },
    {
      "entropy": 6.394068431854248,
      "epoch": 1.2493445342179892,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000485357371908508,
      "loss": 6.0313,
      "mean_token_accuracy": 0.15474005043506622,
      "num_tokens": 25262932.0,
      "step": 11675
    },
    {
      "entropy": 6.416157341003418,
      "epoch": 1.249879608325753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004853438016034396,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14561654552817344,
      "num_tokens": 25274693.0,
      "step": 11680
    },
    {
      "entropy": 6.4382599830627445,
      "epoch": 1.250414682433517,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048533022522472344,
      "loss": 5.9723,
      "mean_token_accuracy": 0.152760349214077,
      "num_tokens": 25285359.0,
      "step": 11685
    },
    {
      "entropy": 6.390935039520263,
      "epoch": 1.250949756541281,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048531664277275154,
      "loss": 6.0125,
      "mean_token_accuracy": 0.15364968180656433,
      "num_tokens": 25296730.0,
      "step": 11690
    },
    {
      "entropy": 6.403318452835083,
      "epoch": 1.251484830649045,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004853030542479162,
      "loss": 6.054,
      "mean_token_accuracy": 0.15899120271205902,
      "num_tokens": 25306422.0,
      "step": 11695
    },
    {
      "entropy": 6.411199760437012,
      "epoch": 1.2520199047568088,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004852894596506096,
      "loss": 6.0722,
      "mean_token_accuracy": 0.15196021944284438,
      "num_tokens": 25317980.0,
      "step": 11700
    },
    {
      "entropy": 6.398972988128662,
      "epoch": 1.2525549788645727,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004852758589812245,
      "loss": 6.0871,
      "mean_token_accuracy": 0.15879200249910355,
      "num_tokens": 25329905.0,
      "step": 11705
    },
    {
      "entropy": 6.503306484222412,
      "epoch": 1.2530900529723366,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048526225224015355,
      "loss": 6.0584,
      "mean_token_accuracy": 0.14375267922878265,
      "num_tokens": 25340759.0,
      "step": 11710
    },
    {
      "entropy": 6.454129076004028,
      "epoch": 1.2536251270801007,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004852486394277896,
      "loss": 6.0849,
      "mean_token_accuracy": 0.1449347823858261,
      "num_tokens": 25351124.0,
      "step": 11715
    },
    {
      "entropy": 6.450143051147461,
      "epoch": 1.2541602011878645,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048523502054452565,
      "loss": 6.0318,
      "mean_token_accuracy": 0.15178735107183455,
      "num_tokens": 25361337.0,
      "step": 11720
    },
    {
      "entropy": 6.482757234573365,
      "epoch": 1.2546952752956284,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048522139559075507,
      "loss": 6.0434,
      "mean_token_accuracy": 0.15659603625535964,
      "num_tokens": 25371893.0,
      "step": 11725
    },
    {
      "entropy": 6.347899389266968,
      "epoch": 1.2552303494033923,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004852077645668712,
      "loss": 5.9397,
      "mean_token_accuracy": 0.16070793718099594,
      "num_tokens": 25383365.0,
      "step": 11730
    },
    {
      "entropy": 6.4331028938293455,
      "epoch": 1.2557654235111562,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004851941274732678,
      "loss": 6.0349,
      "mean_token_accuracy": 0.15329697728157043,
      "num_tokens": 25393211.0,
      "step": 11735
    },
    {
      "entropy": 6.3608129024505615,
      "epoch": 1.2563004976189203,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004851804843103384,
      "loss": 5.9134,
      "mean_token_accuracy": 0.1540319427847862,
      "num_tokens": 25404828.0,
      "step": 11740
    },
    {
      "entropy": 6.458667135238647,
      "epoch": 1.2568355717266841,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048516683507847705,
      "loss": 6.0485,
      "mean_token_accuracy": 0.15592851638793945,
      "num_tokens": 25416016.0,
      "step": 11745
    },
    {
      "entropy": 6.406210088729859,
      "epoch": 1.257370645834448,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004851531797780779,
      "loss": 5.9902,
      "mean_token_accuracy": 0.15909887552261354,
      "num_tokens": 25426173.0,
      "step": 11750
    },
    {
      "entropy": 6.384048175811768,
      "epoch": 1.2579057199422121,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004851395184095352,
      "loss": 5.9686,
      "mean_token_accuracy": 0.15794549584388734,
      "num_tokens": 25437507.0,
      "step": 11755
    },
    {
      "entropy": 6.4623936176300045,
      "epoch": 1.258440794049976,
      "grad_norm": 1.0,
      "learning_rate": 0.0004851258509732434,
      "loss": 6.067,
      "mean_token_accuracy": 0.14506628960371018,
      "num_tokens": 25448201.0,
      "step": 11760
    },
    {
      "entropy": 6.4682936668396,
      "epoch": 1.2589758681577399,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048511217746959723,
      "loss": 6.0675,
      "mean_token_accuracy": 0.149390310049057,
      "num_tokens": 25458640.0,
      "step": 11765
    },
    {
      "entropy": 6.466487216949463,
      "epoch": 1.2595109422655038,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048509849789899146,
      "loss": 6.0617,
      "mean_token_accuracy": 0.1591961294412613,
      "num_tokens": 25469474.0,
      "step": 11770
    },
    {
      "entropy": 6.464839506149292,
      "epoch": 1.2600460163732676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000485084812261821,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14502721130847931,
      "num_tokens": 25479565.0,
      "step": 11775
    },
    {
      "entropy": 6.459881591796875,
      "epoch": 1.2605810904810317,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004850711205584812,
      "loss": 6.0147,
      "mean_token_accuracy": 0.15503039658069612,
      "num_tokens": 25490142.0,
      "step": 11780
    },
    {
      "entropy": 6.433891248703003,
      "epoch": 1.2611161645887956,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048505742278936726,
      "loss": 6.1127,
      "mean_token_accuracy": 0.1525495395064354,
      "num_tokens": 25500824.0,
      "step": 11785
    },
    {
      "entropy": 6.452680253982544,
      "epoch": 1.2616512386965595,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048504371895487474,
      "loss": 6.0525,
      "mean_token_accuracy": 0.15128812193870544,
      "num_tokens": 25511251.0,
      "step": 11790
    },
    {
      "entropy": 6.4330840587615965,
      "epoch": 1.2621863128043234,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048503000905539945,
      "loss": 6.0238,
      "mean_token_accuracy": 0.14805231541395186,
      "num_tokens": 25521368.0,
      "step": 11795
    },
    {
      "entropy": 6.410334300994873,
      "epoch": 1.2627213869120872,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048501629309133707,
      "loss": 6.0315,
      "mean_token_accuracy": 0.15120382755994796,
      "num_tokens": 25532486.0,
      "step": 11800
    },
    {
      "entropy": 6.4322896003723145,
      "epoch": 1.2632564610198513,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004850025710630838,
      "loss": 6.0165,
      "mean_token_accuracy": 0.15547343716025352,
      "num_tokens": 25542867.0,
      "step": 11805
    },
    {
      "entropy": 6.421828842163086,
      "epoch": 1.2637915351276152,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048498884297103575,
      "loss": 6.0049,
      "mean_token_accuracy": 0.15357961356639863,
      "num_tokens": 25553564.0,
      "step": 11810
    },
    {
      "entropy": 6.4211489200592045,
      "epoch": 1.264326609235379,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004849751088155894,
      "loss": 5.9812,
      "mean_token_accuracy": 0.14905180633068085,
      "num_tokens": 25564017.0,
      "step": 11815
    },
    {
      "entropy": 6.388357353210449,
      "epoch": 1.264861683343143,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004849613685971413,
      "loss": 5.9394,
      "mean_token_accuracy": 0.16021719723939895,
      "num_tokens": 25575125.0,
      "step": 11820
    },
    {
      "entropy": 6.448769617080688,
      "epoch": 1.2653967574509069,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004849476223160882,
      "loss": 6.0909,
      "mean_token_accuracy": 0.15273791253566743,
      "num_tokens": 25586405.0,
      "step": 11825
    },
    {
      "entropy": 6.450822591781616,
      "epoch": 1.265931831558671,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004849338699728269,
      "loss": 6.0363,
      "mean_token_accuracy": 0.14729649648070336,
      "num_tokens": 25597494.0,
      "step": 11830
    },
    {
      "entropy": 6.386255884170533,
      "epoch": 1.2664669056664348,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048492011156775467,
      "loss": 6.0067,
      "mean_token_accuracy": 0.15772907435894012,
      "num_tokens": 25608425.0,
      "step": 11835
    },
    {
      "entropy": 6.452467632293701,
      "epoch": 1.2670019797741987,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004849063471012688,
      "loss": 6.0187,
      "mean_token_accuracy": 0.1434104934334755,
      "num_tokens": 25619411.0,
      "step": 11840
    },
    {
      "entropy": 6.4729187965393065,
      "epoch": 1.2675370538819626,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004848925765737665,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14374926313757896,
      "num_tokens": 25630477.0,
      "step": 11845
    },
    {
      "entropy": 6.441867733001709,
      "epoch": 1.2680721279897265,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048487879998564565,
      "loss": 5.9919,
      "mean_token_accuracy": 0.14997110217809678,
      "num_tokens": 25640129.0,
      "step": 11850
    },
    {
      "entropy": 6.415208387374878,
      "epoch": 1.2686072020974906,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048486501733730386,
      "loss": 6.0753,
      "mean_token_accuracy": 0.15035825669765474,
      "num_tokens": 25650592.0,
      "step": 11855
    },
    {
      "entropy": 6.3990668773651125,
      "epoch": 1.2691422762052544,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048485122862913934,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15741816014051438,
      "num_tokens": 25660408.0,
      "step": 11860
    },
    {
      "entropy": 6.398413181304932,
      "epoch": 1.2696773503130183,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048483743386154994,
      "loss": 6.0729,
      "mean_token_accuracy": 0.15436518490314483,
      "num_tokens": 25671563.0,
      "step": 11865
    },
    {
      "entropy": 6.472416830062866,
      "epoch": 1.2702124244207824,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048482363303493415,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15327411144971848,
      "num_tokens": 25682474.0,
      "step": 11870
    },
    {
      "entropy": 6.50192699432373,
      "epoch": 1.270747498528546,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004848098261496905,
      "loss": 6.0225,
      "mean_token_accuracy": 0.14838091433048248,
      "num_tokens": 25692594.0,
      "step": 11875
    },
    {
      "entropy": 6.30156979560852,
      "epoch": 1.2712825726363102,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004847960132062175,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16908981651067734,
      "num_tokens": 25703724.0,
      "step": 11880
    },
    {
      "entropy": 6.459439325332641,
      "epoch": 1.271817646744074,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004847821942049142,
      "loss": 6.1642,
      "mean_token_accuracy": 0.13672943115234376,
      "num_tokens": 25714911.0,
      "step": 11885
    },
    {
      "entropy": 6.508573770523071,
      "epoch": 1.272352720851838,
      "grad_norm": 1.109375,
      "learning_rate": 0.00048476836914617945,
      "loss": 6.0328,
      "mean_token_accuracy": 0.1507526382803917,
      "num_tokens": 25725098.0,
      "step": 11890
    },
    {
      "entropy": 6.454191493988037,
      "epoch": 1.272887794959602,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048475453803041254,
      "loss": 5.9966,
      "mean_token_accuracy": 0.14595401734113694,
      "num_tokens": 25735099.0,
      "step": 11895
    },
    {
      "entropy": 6.415300607681274,
      "epoch": 1.273422869067366,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004847407008580128,
      "loss": 6.0041,
      "mean_token_accuracy": 0.1523135185241699,
      "num_tokens": 25746852.0,
      "step": 11900
    },
    {
      "entropy": 6.4201719760894775,
      "epoch": 1.2739579431751298,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048472685762937983,
      "loss": 6.0684,
      "mean_token_accuracy": 0.15304538011550903,
      "num_tokens": 25758579.0,
      "step": 11905
    },
    {
      "entropy": 6.387421560287476,
      "epoch": 1.2744930172828937,
      "grad_norm": 0.86328125,
      "learning_rate": 0.00048471300834491325,
      "loss": 6.0515,
      "mean_token_accuracy": 0.1467440605163574,
      "num_tokens": 25769965.0,
      "step": 11910
    },
    {
      "entropy": 6.429838371276856,
      "epoch": 1.2750280913906575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004846991530050132,
      "loss": 6.0303,
      "mean_token_accuracy": 0.1531519263982773,
      "num_tokens": 25780264.0,
      "step": 11915
    },
    {
      "entropy": 6.477902126312256,
      "epoch": 1.2755631654984216,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004846852916100794,
      "loss": 5.9798,
      "mean_token_accuracy": 0.1537883996963501,
      "num_tokens": 25791228.0,
      "step": 11920
    },
    {
      "entropy": 6.478234767913818,
      "epoch": 1.2760982396061855,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004846714241605123,
      "loss": 5.9742,
      "mean_token_accuracy": 0.15666651129722595,
      "num_tokens": 25802428.0,
      "step": 11925
    },
    {
      "entropy": 6.451391696929932,
      "epoch": 1.2766333137139494,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048465755065671233,
      "loss": 6.0465,
      "mean_token_accuracy": 0.15374656468629838,
      "num_tokens": 25814220.0,
      "step": 11930
    },
    {
      "entropy": 6.446829986572266,
      "epoch": 1.2771683878217133,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048464367109908005,
      "loss": 6.0244,
      "mean_token_accuracy": 0.1516493707895279,
      "num_tokens": 25824382.0,
      "step": 11935
    },
    {
      "entropy": 6.366900110244751,
      "epoch": 1.2777034619294771,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004846297854880162,
      "loss": 5.9872,
      "mean_token_accuracy": 0.15378809869289398,
      "num_tokens": 25834896.0,
      "step": 11940
    },
    {
      "entropy": 6.4640082836151125,
      "epoch": 1.2782385360372412,
      "grad_norm": 0.82421875,
      "learning_rate": 0.00048461589382392176,
      "loss": 6.1005,
      "mean_token_accuracy": 0.147010001540184,
      "num_tokens": 25847300.0,
      "step": 11945
    },
    {
      "entropy": 6.461084651947021,
      "epoch": 1.2787736101450051,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004846019961071978,
      "loss": 6.0374,
      "mean_token_accuracy": 0.15376857966184615,
      "num_tokens": 25858599.0,
      "step": 11950
    },
    {
      "entropy": 6.434991693496704,
      "epoch": 1.279308684252769,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004845880923382457,
      "loss": 5.9773,
      "mean_token_accuracy": 0.15862847864627838,
      "num_tokens": 25869135.0,
      "step": 11955
    },
    {
      "entropy": 6.439998912811279,
      "epoch": 1.2798437583605329,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048457418251746695,
      "loss": 5.9704,
      "mean_token_accuracy": 0.1583736091852188,
      "num_tokens": 25879234.0,
      "step": 11960
    },
    {
      "entropy": 6.425737905502319,
      "epoch": 1.2803788324682968,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004845602666452629,
      "loss": 6.0154,
      "mean_token_accuracy": 0.1578753650188446,
      "num_tokens": 25888230.0,
      "step": 11965
    },
    {
      "entropy": 6.42799711227417,
      "epoch": 1.2809139065760609,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004845463447220358,
      "loss": 6.0453,
      "mean_token_accuracy": 0.16157087236642836,
      "num_tokens": 25899429.0,
      "step": 11970
    },
    {
      "entropy": 6.398941230773926,
      "epoch": 1.2814489806838247,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004845324167481874,
      "loss": 5.9156,
      "mean_token_accuracy": 0.1674194172024727,
      "num_tokens": 25911042.0,
      "step": 11975
    },
    {
      "entropy": 6.4437507629394535,
      "epoch": 1.2819840547915886,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004845184827241199,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15630557835102082,
      "num_tokens": 25921466.0,
      "step": 11980
    },
    {
      "entropy": 6.435121631622314,
      "epoch": 1.2825191288993525,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004845045426502357,
      "loss": 6.0997,
      "mean_token_accuracy": 0.14636427015066147,
      "num_tokens": 25931059.0,
      "step": 11985
    },
    {
      "entropy": 6.460503435134887,
      "epoch": 1.2830542030071164,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048449059652693724,
      "loss": 6.0242,
      "mean_token_accuracy": 0.15136034339666365,
      "num_tokens": 25941669.0,
      "step": 11990
    },
    {
      "entropy": 6.41490740776062,
      "epoch": 1.2835892771148805,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004844766443546272,
      "loss": 6.0514,
      "mean_token_accuracy": 0.15163252204656602,
      "num_tokens": 25952066.0,
      "step": 11995
    },
    {
      "entropy": 6.4431534767150875,
      "epoch": 1.2841243512226443,
      "grad_norm": 1.0,
      "learning_rate": 0.0004844626861337085,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15751319229602814,
      "num_tokens": 25962388.0,
      "step": 12000
    },
    {
      "epoch": 1.2841243512226443,
      "eval_entropy": 6.175162490166101,
      "eval_loss": 6.133983612060547,
      "eval_mean_token_accuracy": 0.15550359565779728,
      "eval_num_tokens": 25962388.0,
      "eval_runtime": 17.6404,
      "eval_samples_per_second": 1682.442,
      "eval_steps_per_second": 210.312,
      "step": 12000
    }
  ],
  "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": 5646475164764160.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}