heb-hebr-10mb-100mb_seed3407/checkpoint-33000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.531328589009578,
  "eval_steps": 3000,
  "global_step": 33000,
  "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
    },
    {
      "entropy": 6.333190059661865,
      "epoch": 1.2846594253304082,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004844487218645843,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15236114710569382,
      "num_tokens": 25973865.0,
      "step": 12005
    },
    {
      "entropy": 6.378346300125122,
      "epoch": 1.2851944994381723,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004844347515476575,
      "loss": 5.989,
      "mean_token_accuracy": 0.15576977729797364,
      "num_tokens": 25985113.0,
      "step": 12010
    },
    {
      "entropy": 6.415798568725586,
      "epoch": 1.285729573545936,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048442077518333175,
      "loss": 5.9161,
      "mean_token_accuracy": 0.16710020750761032,
      "num_tokens": 25995980.0,
      "step": 12015
    },
    {
      "entropy": 6.392321634292602,
      "epoch": 1.2862646476537,
      "grad_norm": 1.0,
      "learning_rate": 0.00048440679277201063,
      "loss": 6.0031,
      "mean_token_accuracy": 0.1561504051089287,
      "num_tokens": 26006362.0,
      "step": 12020
    },
    {
      "entropy": 6.455795860290527,
      "epoch": 1.286799721761464,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048439280431409774,
      "loss": 6.0966,
      "mean_token_accuracy": 0.1500068873167038,
      "num_tokens": 26016623.0,
      "step": 12025
    },
    {
      "entropy": 6.357543182373047,
      "epoch": 1.2873347958692278,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048437880980999694,
      "loss": 5.9105,
      "mean_token_accuracy": 0.16127020865678787,
      "num_tokens": 26027202.0,
      "step": 12030
    },
    {
      "entropy": 6.42196626663208,
      "epoch": 1.287869869976992,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004843648092601125,
      "loss": 5.9975,
      "mean_token_accuracy": 0.1481295645236969,
      "num_tokens": 26037330.0,
      "step": 12035
    },
    {
      "entropy": 6.471438837051392,
      "epoch": 1.2884049440847558,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004843508026648487,
      "loss": 6.0191,
      "mean_token_accuracy": 0.15306735336780547,
      "num_tokens": 26047661.0,
      "step": 12040
    },
    {
      "entropy": 6.443175220489502,
      "epoch": 1.2889400181925197,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004843367900246098,
      "loss": 6.0663,
      "mean_token_accuracy": 0.1495279110968113,
      "num_tokens": 26058582.0,
      "step": 12045
    },
    {
      "entropy": 6.3907555103302,
      "epoch": 1.2894750923002836,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048432277133980055,
      "loss": 6.0724,
      "mean_token_accuracy": 0.14319442361593246,
      "num_tokens": 26069117.0,
      "step": 12050
    },
    {
      "entropy": 6.4476766109466555,
      "epoch": 1.2900101664080474,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00048430874661082565,
      "loss": 6.0367,
      "mean_token_accuracy": 0.15534177124500276,
      "num_tokens": 26079327.0,
      "step": 12055
    },
    {
      "entropy": 6.526661777496338,
      "epoch": 1.2905452405158115,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048429471583809016,
      "loss": 6.0591,
      "mean_token_accuracy": 0.14655549973249435,
      "num_tokens": 26090709.0,
      "step": 12060
    },
    {
      "entropy": 6.4874342918396,
      "epoch": 1.2910803146235754,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004842806790219991,
      "loss": 6.0464,
      "mean_token_accuracy": 0.14699550569057465,
      "num_tokens": 26101570.0,
      "step": 12065
    },
    {
      "entropy": 6.36808590888977,
      "epoch": 1.2916153887313393,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048426663616295783,
      "loss": 5.996,
      "mean_token_accuracy": 0.15453232526779176,
      "num_tokens": 26112827.0,
      "step": 12070
    },
    {
      "entropy": 6.438121938705445,
      "epoch": 1.2921504628391032,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048425258726137187,
      "loss": 6.0714,
      "mean_token_accuracy": 0.14548972621560097,
      "num_tokens": 26124141.0,
      "step": 12075
    },
    {
      "entropy": 6.445812797546386,
      "epoch": 1.292685536946867,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048423853231764684,
      "loss": 6.0189,
      "mean_token_accuracy": 0.15440742522478104,
      "num_tokens": 26135372.0,
      "step": 12080
    },
    {
      "entropy": 6.405870962142944,
      "epoch": 1.2932206110546312,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048422447133218867,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15272736847400664,
      "num_tokens": 26145933.0,
      "step": 12085
    },
    {
      "entropy": 6.459271335601807,
      "epoch": 1.293755685162395,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048421040430540323,
      "loss": 6.0716,
      "mean_token_accuracy": 0.1470033660531044,
      "num_tokens": 26155948.0,
      "step": 12090
    },
    {
      "entropy": 6.425253915786743,
      "epoch": 1.294290759270159,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004841963312376967,
      "loss": 5.9656,
      "mean_token_accuracy": 0.1603505089879036,
      "num_tokens": 26166410.0,
      "step": 12095
    },
    {
      "entropy": 6.378489351272583,
      "epoch": 1.2948258333779228,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048418225212947566,
      "loss": 6.0071,
      "mean_token_accuracy": 0.15442349910736083,
      "num_tokens": 26177334.0,
      "step": 12100
    },
    {
      "entropy": 6.506193733215332,
      "epoch": 1.2953609074856867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004841681669811464,
      "loss": 6.086,
      "mean_token_accuracy": 0.1429160974919796,
      "num_tokens": 26187987.0,
      "step": 12105
    },
    {
      "entropy": 6.457412767410278,
      "epoch": 1.2958959815934508,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004841540757931157,
      "loss": 6.0171,
      "mean_token_accuracy": 0.1484729066491127,
      "num_tokens": 26198600.0,
      "step": 12110
    },
    {
      "entropy": 6.440911054611206,
      "epoch": 1.2964310557012146,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004841399785657904,
      "loss": 6.0629,
      "mean_token_accuracy": 0.1492348551750183,
      "num_tokens": 26210977.0,
      "step": 12115
    },
    {
      "entropy": 6.437525844573974,
      "epoch": 1.2969661298089785,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048412587529957774,
      "loss": 6.012,
      "mean_token_accuracy": 0.15377116352319717,
      "num_tokens": 26220813.0,
      "step": 12120
    },
    {
      "entropy": 6.434960651397705,
      "epoch": 1.2975012039167424,
      "grad_norm": 0.859375,
      "learning_rate": 0.0004841117659948848,
      "loss": 6.0446,
      "mean_token_accuracy": 0.1535068929195404,
      "num_tokens": 26232042.0,
      "step": 12125
    },
    {
      "entropy": 6.46825270652771,
      "epoch": 1.2980362780245063,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048409765065211884,
      "loss": 6.0511,
      "mean_token_accuracy": 0.15129367187619208,
      "num_tokens": 26242968.0,
      "step": 12130
    },
    {
      "entropy": 6.486259698867798,
      "epoch": 1.2985713521322704,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048408352927168774,
      "loss": 6.0432,
      "mean_token_accuracy": 0.15426448583602906,
      "num_tokens": 26252751.0,
      "step": 12135
    },
    {
      "entropy": 6.4638776779174805,
      "epoch": 1.2991064262400343,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004840694018539991,
      "loss": 6.1052,
      "mean_token_accuracy": 0.14575298503041267,
      "num_tokens": 26263270.0,
      "step": 12140
    },
    {
      "entropy": 6.463758897781372,
      "epoch": 1.2996415003477981,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048405526839946086,
      "loss": 6.1122,
      "mean_token_accuracy": 0.14519880339503288,
      "num_tokens": 26274279.0,
      "step": 12145
    },
    {
      "entropy": 6.423062467575074,
      "epoch": 1.3001765744555622,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048404112890848105,
      "loss": 6.1165,
      "mean_token_accuracy": 0.15198323279619216,
      "num_tokens": 26286020.0,
      "step": 12150
    },
    {
      "entropy": 6.492597818374634,
      "epoch": 1.3007116485633259,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048402698338146803,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1561700791120529,
      "num_tokens": 26296798.0,
      "step": 12155
    },
    {
      "entropy": 6.431667852401733,
      "epoch": 1.30124672267109,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004840128318188303,
      "loss": 6.0109,
      "mean_token_accuracy": 0.15001762583851813,
      "num_tokens": 26306952.0,
      "step": 12160
    },
    {
      "entropy": 6.475614261627197,
      "epoch": 1.3017817967788539,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048399867422097644,
      "loss": 6.0907,
      "mean_token_accuracy": 0.15140620172023772,
      "num_tokens": 26317686.0,
      "step": 12165
    },
    {
      "entropy": 6.452743816375732,
      "epoch": 1.3023168708866177,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048398451058831514,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15510403513908386,
      "num_tokens": 26327997.0,
      "step": 12170
    },
    {
      "entropy": 6.4132490158081055,
      "epoch": 1.3028519449943818,
      "grad_norm": 0.90625,
      "learning_rate": 0.00048397034092125554,
      "loss": 6.0361,
      "mean_token_accuracy": 0.15372907370328903,
      "num_tokens": 26340227.0,
      "step": 12175
    },
    {
      "entropy": 6.548251295089722,
      "epoch": 1.3033870191021457,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048395616522020667,
      "loss": 6.1317,
      "mean_token_accuracy": 0.15375821515917779,
      "num_tokens": 26350549.0,
      "step": 12180
    },
    {
      "entropy": 6.457336473464966,
      "epoch": 1.3039220932099096,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004839419834855779,
      "loss": 6.0515,
      "mean_token_accuracy": 0.14976763129234313,
      "num_tokens": 26361822.0,
      "step": 12185
    },
    {
      "entropy": 6.347854137420654,
      "epoch": 1.3044571673176735,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004839277957177787,
      "loss": 6.0693,
      "mean_token_accuracy": 0.15542332977056503,
      "num_tokens": 26372720.0,
      "step": 12190
    },
    {
      "entropy": 6.403561353683472,
      "epoch": 1.3049922414254373,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004839136019172188,
      "loss": 6.0143,
      "mean_token_accuracy": 0.1563357502222061,
      "num_tokens": 26383092.0,
      "step": 12195
    },
    {
      "entropy": 6.409371471405029,
      "epoch": 1.3055273155332014,
      "grad_norm": 0.984375,
      "learning_rate": 0.000483899402084308,
      "loss": 5.9604,
      "mean_token_accuracy": 0.16210002303123475,
      "num_tokens": 26395188.0,
      "step": 12200
    },
    {
      "entropy": 6.405869722366333,
      "epoch": 1.3060623896409653,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004838851962194563,
      "loss": 6.1106,
      "mean_token_accuracy": 0.15249233990907668,
      "num_tokens": 26406338.0,
      "step": 12205
    },
    {
      "entropy": 6.429240465164185,
      "epoch": 1.3065974637487292,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004838709843230739,
      "loss": 6.0308,
      "mean_token_accuracy": 0.15293858274817468,
      "num_tokens": 26416983.0,
      "step": 12210
    },
    {
      "entropy": 6.464216423034668,
      "epoch": 1.307132537856493,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004838567663955712,
      "loss": 5.9882,
      "mean_token_accuracy": 0.16073604822158813,
      "num_tokens": 26426966.0,
      "step": 12215
    },
    {
      "entropy": 6.418710374832154,
      "epoch": 1.307667611964257,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004838425424373587,
      "loss": 5.9538,
      "mean_token_accuracy": 0.15561669766902925,
      "num_tokens": 26437711.0,
      "step": 12220
    },
    {
      "entropy": 6.433170127868652,
      "epoch": 1.308202686072021,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004838283124488472,
      "loss": 6.0857,
      "mean_token_accuracy": 0.15388013273477555,
      "num_tokens": 26448657.0,
      "step": 12225
    },
    {
      "entropy": 6.461983919143677,
      "epoch": 1.308737760179785,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048381407643044737,
      "loss": 5.9786,
      "mean_token_accuracy": 0.15819147676229478,
      "num_tokens": 26459363.0,
      "step": 12230
    },
    {
      "entropy": 6.407735157012939,
      "epoch": 1.3092728342875488,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004837998343825705,
      "loss": 6.008,
      "mean_token_accuracy": 0.15663945227861403,
      "num_tokens": 26470149.0,
      "step": 12235
    },
    {
      "entropy": 6.410347604751587,
      "epoch": 1.3098079083953127,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048378558630562783,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15960146933794023,
      "num_tokens": 26480424.0,
      "step": 12240
    },
    {
      "entropy": 6.443244791030883,
      "epoch": 1.3103429825030766,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004837713322000306,
      "loss": 6.0002,
      "mean_token_accuracy": 0.15054864510893823,
      "num_tokens": 26490312.0,
      "step": 12245
    },
    {
      "entropy": 6.3710370540618895,
      "epoch": 1.3108780566108407,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004837570720661905,
      "loss": 5.9959,
      "mean_token_accuracy": 0.16077183485031127,
      "num_tokens": 26501119.0,
      "step": 12250
    },
    {
      "entropy": 6.449281978607178,
      "epoch": 1.3114131307186045,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048374280590451934,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15321921557188034,
      "num_tokens": 26512438.0,
      "step": 12255
    },
    {
      "entropy": 6.400368165969849,
      "epoch": 1.3119482048263684,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00048372853371542895,
      "loss": 6.0303,
      "mean_token_accuracy": 0.14798753410577775,
      "num_tokens": 26523930.0,
      "step": 12260
    },
    {
      "entropy": 6.441335821151734,
      "epoch": 1.3124832789341323,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004837142554993315,
      "loss": 6.0881,
      "mean_token_accuracy": 0.15518404245376588,
      "num_tokens": 26534937.0,
      "step": 12265
    },
    {
      "entropy": 6.353445386886596,
      "epoch": 1.3130183530418962,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004836999712566392,
      "loss": 5.8891,
      "mean_token_accuracy": 0.16114138141274453,
      "num_tokens": 26544555.0,
      "step": 12270
    },
    {
      "entropy": 6.443650627136231,
      "epoch": 1.3135534271496603,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048368568098776463,
      "loss": 6.0327,
      "mean_token_accuracy": 0.1635463148355484,
      "num_tokens": 26556144.0,
      "step": 12275
    },
    {
      "entropy": 6.508738708496094,
      "epoch": 1.3140885012574242,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004836713846931203,
      "loss": 6.0984,
      "mean_token_accuracy": 0.14364687874913215,
      "num_tokens": 26567350.0,
      "step": 12280
    },
    {
      "entropy": 6.399505472183227,
      "epoch": 1.314623575365188,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048365708237311904,
      "loss": 6.0061,
      "mean_token_accuracy": 0.1546790763735771,
      "num_tokens": 26579486.0,
      "step": 12285
    },
    {
      "entropy": 6.4119771957397464,
      "epoch": 1.3151586494729521,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004836427740281739,
      "loss": 5.9969,
      "mean_token_accuracy": 0.1514202192425728,
      "num_tokens": 26589665.0,
      "step": 12290
    },
    {
      "entropy": 6.446229124069214,
      "epoch": 1.3156937235807158,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00048362845965869793,
      "loss": 6.0244,
      "mean_token_accuracy": 0.15393809825181962,
      "num_tokens": 26600187.0,
      "step": 12295
    },
    {
      "entropy": 6.429892492294312,
      "epoch": 1.31622879768848,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004836141392651046,
      "loss": 5.986,
      "mean_token_accuracy": 0.1569882944226265,
      "num_tokens": 26612155.0,
      "step": 12300
    },
    {
      "entropy": 6.450652694702148,
      "epoch": 1.3167638717962438,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048359981284780727,
      "loss": 6.0281,
      "mean_token_accuracy": 0.14950771778821945,
      "num_tokens": 26623077.0,
      "step": 12305
    },
    {
      "entropy": 6.399432706832886,
      "epoch": 1.3172989459040076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004835854804072196,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15725656747817993,
      "num_tokens": 26632979.0,
      "step": 12310
    },
    {
      "entropy": 6.4788518905639645,
      "epoch": 1.3178340200117717,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048357114194375555,
      "loss": 6.0809,
      "mean_token_accuracy": 0.1508488178253174,
      "num_tokens": 26644377.0,
      "step": 12315
    },
    {
      "entropy": 6.444859313964844,
      "epoch": 1.3183690941195356,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000483556797457829,
      "loss": 6.0471,
      "mean_token_accuracy": 0.14296522736549377,
      "num_tokens": 26655538.0,
      "step": 12320
    },
    {
      "entropy": 6.465205526351928,
      "epoch": 1.3189041682272995,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048354244694985435,
      "loss": 5.9911,
      "mean_token_accuracy": 0.15424570590257644,
      "num_tokens": 26668011.0,
      "step": 12325
    },
    {
      "entropy": 6.434666919708252,
      "epoch": 1.3194392423350634,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004835280904202457,
      "loss": 6.0281,
      "mean_token_accuracy": 0.14812832027673722,
      "num_tokens": 26678996.0,
      "step": 12330
    },
    {
      "entropy": 6.444580459594727,
      "epoch": 1.3199743164428273,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004835137278694178,
      "loss": 5.9908,
      "mean_token_accuracy": 0.15598600804805757,
      "num_tokens": 26689041.0,
      "step": 12335
    },
    {
      "entropy": 6.448755931854248,
      "epoch": 1.3205093905505914,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004834993592977854,
      "loss": 5.9895,
      "mean_token_accuracy": 0.1587250277400017,
      "num_tokens": 26699459.0,
      "step": 12340
    },
    {
      "entropy": 6.421284246444702,
      "epoch": 1.3210444646583552,
      "grad_norm": 0.828125,
      "learning_rate": 0.0004834849847057632,
      "loss": 6.1168,
      "mean_token_accuracy": 0.1508992299437523,
      "num_tokens": 26711663.0,
      "step": 12345
    },
    {
      "entropy": 6.48749942779541,
      "epoch": 1.321579538766119,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048347060409376644,
      "loss": 6.0618,
      "mean_token_accuracy": 0.14786320328712463,
      "num_tokens": 26722960.0,
      "step": 12350
    },
    {
      "entropy": 6.434947299957275,
      "epoch": 1.322114612873883,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048345621746221024,
      "loss": 5.9953,
      "mean_token_accuracy": 0.15536460503935814,
      "num_tokens": 26733274.0,
      "step": 12355
    },
    {
      "entropy": 6.431057453155518,
      "epoch": 1.3226496869816469,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048344182481151003,
      "loss": 6.053,
      "mean_token_accuracy": 0.14910613149404525,
      "num_tokens": 26743906.0,
      "step": 12360
    },
    {
      "entropy": 6.401789379119873,
      "epoch": 1.323184761089411,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048342742614208145,
      "loss": 6.0422,
      "mean_token_accuracy": 0.14974934756755828,
      "num_tokens": 26755093.0,
      "step": 12365
    },
    {
      "entropy": 6.425383853912353,
      "epoch": 1.3237198351971748,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004834130214543402,
      "loss": 6.0112,
      "mean_token_accuracy": 0.1535753443837166,
      "num_tokens": 26765614.0,
      "step": 12370
    },
    {
      "entropy": 6.407527446746826,
      "epoch": 1.3242549093049387,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00048339861074870227,
      "loss": 6.0177,
      "mean_token_accuracy": 0.15520277470350266,
      "num_tokens": 26778078.0,
      "step": 12375
    },
    {
      "entropy": 6.456593036651611,
      "epoch": 1.3247899834127026,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048338419402558375,
      "loss": 6.0508,
      "mean_token_accuracy": 0.14981769770383835,
      "num_tokens": 26788707.0,
      "step": 12380
    },
    {
      "entropy": 6.477190923690796,
      "epoch": 1.3253250575204665,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004833697712854009,
      "loss": 5.9966,
      "mean_token_accuracy": 0.15447067022323607,
      "num_tokens": 26798650.0,
      "step": 12385
    },
    {
      "entropy": 6.40105938911438,
      "epoch": 1.3258601316282306,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048335534252857016,
      "loss": 6.0495,
      "mean_token_accuracy": 0.1529637470841408,
      "num_tokens": 26809822.0,
      "step": 12390
    },
    {
      "entropy": 6.3931135654449465,
      "epoch": 1.3263952057359945,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004833409077555082,
      "loss": 6.0227,
      "mean_token_accuracy": 0.14945009499788284,
      "num_tokens": 26821061.0,
      "step": 12395
    },
    {
      "entropy": 6.477404832839966,
      "epoch": 1.3269302798437583,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048332646696663177,
      "loss": 6.0688,
      "mean_token_accuracy": 0.15109701007604598,
      "num_tokens": 26832020.0,
      "step": 12400
    },
    {
      "entropy": 6.49910979270935,
      "epoch": 1.3274653539515222,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004833120201623579,
      "loss": 6.041,
      "mean_token_accuracy": 0.1525729402899742,
      "num_tokens": 26844287.0,
      "step": 12405
    },
    {
      "entropy": 6.35153169631958,
      "epoch": 1.328000428059286,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048329756734310375,
      "loss": 5.9172,
      "mean_token_accuracy": 0.16074969619512558,
      "num_tokens": 26855790.0,
      "step": 12410
    },
    {
      "entropy": 6.410444974899292,
      "epoch": 1.3285355021670502,
      "grad_norm": 0.87109375,
      "learning_rate": 0.00048328310850928656,
      "loss": 6.0446,
      "mean_token_accuracy": 0.15391015037894248,
      "num_tokens": 26867391.0,
      "step": 12415
    },
    {
      "entropy": 6.4366028785705565,
      "epoch": 1.329070576274814,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048326864366132397,
      "loss": 6.0368,
      "mean_token_accuracy": 0.1570956066250801,
      "num_tokens": 26879173.0,
      "step": 12420
    },
    {
      "entropy": 6.4614588737487795,
      "epoch": 1.329605650382578,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004832541727996335,
      "loss": 5.9573,
      "mean_token_accuracy": 0.16303762793540955,
      "num_tokens": 26889839.0,
      "step": 12425
    },
    {
      "entropy": 6.360187435150147,
      "epoch": 1.330140724490342,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000483239695924633,
      "loss": 6.0186,
      "mean_token_accuracy": 0.1571821913123131,
      "num_tokens": 26900217.0,
      "step": 12430
    },
    {
      "entropy": 6.393980073928833,
      "epoch": 1.330675798598106,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048322521303674064,
      "loss": 6.0107,
      "mean_token_accuracy": 0.1587739109992981,
      "num_tokens": 26911366.0,
      "step": 12435
    },
    {
      "entropy": 6.486601877212524,
      "epoch": 1.3312108727058698,
      "grad_norm": 1.0,
      "learning_rate": 0.00048321072413637454,
      "loss": 6.0584,
      "mean_token_accuracy": 0.15189179480075837,
      "num_tokens": 26922135.0,
      "step": 12440
    },
    {
      "entropy": 6.430215358734131,
      "epoch": 1.3317459468136337,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004831962292239529,
      "loss": 6.0023,
      "mean_token_accuracy": 0.15735085755586625,
      "num_tokens": 26932502.0,
      "step": 12445
    },
    {
      "entropy": 6.408875370025635,
      "epoch": 1.3322810209213976,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048318172829989447,
      "loss": 6.0377,
      "mean_token_accuracy": 0.15644336491823196,
      "num_tokens": 26944607.0,
      "step": 12450
    },
    {
      "entropy": 6.434620094299317,
      "epoch": 1.3328160950291617,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004831672213646179,
      "loss": 6.0571,
      "mean_token_accuracy": 0.15193878412246703,
      "num_tokens": 26955205.0,
      "step": 12455
    },
    {
      "entropy": 6.411489677429199,
      "epoch": 1.3333511691369255,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004831527084185421,
      "loss": 5.9333,
      "mean_token_accuracy": 0.16186214238405228,
      "num_tokens": 26965558.0,
      "step": 12460
    },
    {
      "entropy": 6.401693534851074,
      "epoch": 1.3338862432446894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000483138189462086,
      "loss": 6.0118,
      "mean_token_accuracy": 0.14909350275993347,
      "num_tokens": 26976176.0,
      "step": 12465
    },
    {
      "entropy": 6.452073049545288,
      "epoch": 1.3344213173524533,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048312366449566895,
      "loss": 6.0559,
      "mean_token_accuracy": 0.15355814546346663,
      "num_tokens": 26987099.0,
      "step": 12470
    },
    {
      "entropy": 6.424422931671143,
      "epoch": 1.3349563914602172,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004831091335197104,
      "loss": 6.017,
      "mean_token_accuracy": 0.15274921655654908,
      "num_tokens": 26998122.0,
      "step": 12475
    },
    {
      "entropy": 6.375023269653321,
      "epoch": 1.3354914655679813,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004830945965346298,
      "loss": 5.9691,
      "mean_token_accuracy": 0.1556036040186882,
      "num_tokens": 27009427.0,
      "step": 12480
    },
    {
      "entropy": 6.481918144226074,
      "epoch": 1.3360265396757451,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000483080053540847,
      "loss": 6.1107,
      "mean_token_accuracy": 0.14653872326016426,
      "num_tokens": 27020701.0,
      "step": 12485
    },
    {
      "entropy": 6.493611860275268,
      "epoch": 1.336561613783509,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004830655045387818,
      "loss": 6.0673,
      "mean_token_accuracy": 0.1468011662364006,
      "num_tokens": 27033215.0,
      "step": 12490
    },
    {
      "entropy": 6.429815053939819,
      "epoch": 1.337096687891273,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048305094952885453,
      "loss": 6.0518,
      "mean_token_accuracy": 0.1474909372627735,
      "num_tokens": 27043846.0,
      "step": 12495
    },
    {
      "entropy": 6.408537673950195,
      "epoch": 1.3376317619990368,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004830363885114853,
      "loss": 6.0165,
      "mean_token_accuracy": 0.15108912587165832,
      "num_tokens": 27054783.0,
      "step": 12500
    },
    {
      "entropy": 6.438246393203736,
      "epoch": 1.3381668361068009,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004830218214870946,
      "loss": 6.0437,
      "mean_token_accuracy": 0.15150342136621475,
      "num_tokens": 27066051.0,
      "step": 12505
    },
    {
      "entropy": 6.486574459075928,
      "epoch": 1.3387019102145647,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004830072484561029,
      "loss": 6.0588,
      "mean_token_accuracy": 0.15050753355026245,
      "num_tokens": 27078210.0,
      "step": 12510
    },
    {
      "entropy": 6.469337797164917,
      "epoch": 1.3392369843223286,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004829926694189312,
      "loss": 6.0751,
      "mean_token_accuracy": 0.14790621846914292,
      "num_tokens": 27089693.0,
      "step": 12515
    },
    {
      "entropy": 6.444461011886597,
      "epoch": 1.3397720584300925,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004829780843760005,
      "loss": 6.0522,
      "mean_token_accuracy": 0.14840853810310364,
      "num_tokens": 27099856.0,
      "step": 12520
    },
    {
      "entropy": 6.448481225967408,
      "epoch": 1.3403071325378564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004829634933277317,
      "loss": 6.0614,
      "mean_token_accuracy": 0.1480243980884552,
      "num_tokens": 27110218.0,
      "step": 12525
    },
    {
      "entropy": 6.435736656188965,
      "epoch": 1.3408422066456205,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048294889627454636,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15498586148023605,
      "num_tokens": 27120911.0,
      "step": 12530
    },
    {
      "entropy": 6.401727390289307,
      "epoch": 1.3413772807533844,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048293429321686585,
      "loss": 5.9878,
      "mean_token_accuracy": 0.15649719834327697,
      "num_tokens": 27130964.0,
      "step": 12535
    },
    {
      "entropy": 6.435080575942993,
      "epoch": 1.3419123548611482,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048291968415511174,
      "loss": 6.0161,
      "mean_token_accuracy": 0.15310217142105104,
      "num_tokens": 27141641.0,
      "step": 12540
    },
    {
      "entropy": 6.448747396469116,
      "epoch": 1.3424474289689121,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00048290506908970607,
      "loss": 6.0854,
      "mean_token_accuracy": 0.14812047630548478,
      "num_tokens": 27152591.0,
      "step": 12545
    },
    {
      "entropy": 6.478668451309204,
      "epoch": 1.342982503076676,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004828904480210707,
      "loss": 6.1625,
      "mean_token_accuracy": 0.14315567463636397,
      "num_tokens": 27163839.0,
      "step": 12550
    },
    {
      "entropy": 6.49043231010437,
      "epoch": 1.34351757718444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004828758209496278,
      "loss": 6.026,
      "mean_token_accuracy": 0.14880292117595673,
      "num_tokens": 27174480.0,
      "step": 12555
    },
    {
      "entropy": 6.382992458343506,
      "epoch": 1.344052651292204,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00048286118787579983,
      "loss": 5.9943,
      "mean_token_accuracy": 0.15013302788138388,
      "num_tokens": 27184546.0,
      "step": 12560
    },
    {
      "entropy": 6.371896266937256,
      "epoch": 1.3445877253999678,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048284654880000933,
      "loss": 5.8925,
      "mean_token_accuracy": 0.16564469188451766,
      "num_tokens": 27195093.0,
      "step": 12565
    },
    {
      "entropy": 6.447955560684204,
      "epoch": 1.345122799507732,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048283190372267886,
      "loss": 6.0652,
      "mean_token_accuracy": 0.150056292116642,
      "num_tokens": 27206416.0,
      "step": 12570
    },
    {
      "entropy": 6.470468187332154,
      "epoch": 1.3456578736154958,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004828172526442314,
      "loss": 5.9909,
      "mean_token_accuracy": 0.156391042470932,
      "num_tokens": 27215533.0,
      "step": 12575
    },
    {
      "entropy": 6.45537543296814,
      "epoch": 1.3461929477232597,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048280259556509,
      "loss": 6.0423,
      "mean_token_accuracy": 0.15147729963064194,
      "num_tokens": 27225985.0,
      "step": 12580
    },
    {
      "entropy": 6.406804895401001,
      "epoch": 1.3467280218310236,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004827879324856778,
      "loss": 5.9979,
      "mean_token_accuracy": 0.1577044978737831,
      "num_tokens": 27235967.0,
      "step": 12585
    },
    {
      "entropy": 6.434601449966431,
      "epoch": 1.3472630959387875,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048277326340641835,
      "loss": 5.955,
      "mean_token_accuracy": 0.15415469855070113,
      "num_tokens": 27246854.0,
      "step": 12590
    },
    {
      "entropy": 6.43442735671997,
      "epoch": 1.3477981700465516,
      "grad_norm": 1.015625,
      "learning_rate": 0.000482758588327735,
      "loss": 6.0778,
      "mean_token_accuracy": 0.15348837226629258,
      "num_tokens": 27256640.0,
      "step": 12595
    },
    {
      "entropy": 6.438212728500366,
      "epoch": 1.3483332441543154,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004827439072500516,
      "loss": 6.0308,
      "mean_token_accuracy": 0.15508270114660264,
      "num_tokens": 27266907.0,
      "step": 12600
    },
    {
      "entropy": 6.45332145690918,
      "epoch": 1.3488683182620793,
      "grad_norm": 0.875,
      "learning_rate": 0.00048272922017379217,
      "loss": 6.0268,
      "mean_token_accuracy": 0.1500389665365219,
      "num_tokens": 27277089.0,
      "step": 12605
    },
    {
      "entropy": 6.418215751647949,
      "epoch": 1.3494033923698432,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004827145270993807,
      "loss": 5.9844,
      "mean_token_accuracy": 0.15779301822185515,
      "num_tokens": 27286781.0,
      "step": 12610
    },
    {
      "entropy": 6.438915252685547,
      "epoch": 1.349938466477607,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004826998280272414,
      "loss": 6.0178,
      "mean_token_accuracy": 0.15893317759037018,
      "num_tokens": 27298011.0,
      "step": 12615
    },
    {
      "entropy": 6.450037860870362,
      "epoch": 1.3504735405853712,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048268512295779873,
      "loss": 6.0454,
      "mean_token_accuracy": 0.1544107422232628,
      "num_tokens": 27309052.0,
      "step": 12620
    },
    {
      "entropy": 6.3779655456542965,
      "epoch": 1.351008614693135,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004826704118914773,
      "loss": 5.9795,
      "mean_token_accuracy": 0.1526784658432007,
      "num_tokens": 27319945.0,
      "step": 12625
    },
    {
      "entropy": 6.42637071609497,
      "epoch": 1.351543688800899,
      "grad_norm": 1.046875,
      "learning_rate": 0.000482655694828702,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1613290399312973,
      "num_tokens": 27330543.0,
      "step": 12630
    },
    {
      "entropy": 6.4632445812225345,
      "epoch": 1.3520787629086628,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004826409717698976,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15123631358146666,
      "num_tokens": 27341957.0,
      "step": 12635
    },
    {
      "entropy": 6.442296123504638,
      "epoch": 1.3526138370164267,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048262624271548936,
      "loss": 5.997,
      "mean_token_accuracy": 0.15276422053575517,
      "num_tokens": 27351506.0,
      "step": 12640
    },
    {
      "entropy": 6.418410062789917,
      "epoch": 1.3531489111241908,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048261150766590257,
      "loss": 5.9667,
      "mean_token_accuracy": 0.15033604055643082,
      "num_tokens": 27362208.0,
      "step": 12645
    },
    {
      "entropy": 6.426398134231567,
      "epoch": 1.3536839852319547,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048259676662156263,
      "loss": 5.9967,
      "mean_token_accuracy": 0.1552975133061409,
      "num_tokens": 27372890.0,
      "step": 12650
    },
    {
      "entropy": 6.381421041488648,
      "epoch": 1.3542190593397185,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004825820195828952,
      "loss": 5.9391,
      "mean_token_accuracy": 0.1610363468527794,
      "num_tokens": 27382879.0,
      "step": 12655
    },
    {
      "entropy": 6.441811370849609,
      "epoch": 1.3547541334474824,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004825672665503262,
      "loss": 5.9697,
      "mean_token_accuracy": 0.159789077937603,
      "num_tokens": 27393878.0,
      "step": 12660
    },
    {
      "entropy": 6.437974166870117,
      "epoch": 1.3552892075552463,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004825525075242815,
      "loss": 6.07,
      "mean_token_accuracy": 0.1516162060201168,
      "num_tokens": 27404928.0,
      "step": 12665
    },
    {
      "entropy": 6.46609787940979,
      "epoch": 1.3558242816630104,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004825377425051873,
      "loss": 6.0061,
      "mean_token_accuracy": 0.150531093031168,
      "num_tokens": 27417473.0,
      "step": 12670
    },
    {
      "entropy": 6.455836439132691,
      "epoch": 1.3563593557707743,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00048252297149346997,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14940343871712686,
      "num_tokens": 27428378.0,
      "step": 12675
    },
    {
      "entropy": 6.319449996948242,
      "epoch": 1.3568944298785381,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048250819448955595,
      "loss": 6.0371,
      "mean_token_accuracy": 0.15481272637844085,
      "num_tokens": 27438815.0,
      "step": 12680
    },
    {
      "entropy": 6.511168193817139,
      "epoch": 1.357429503986302,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000482493411493872,
      "loss": 6.0768,
      "mean_token_accuracy": 0.151198048889637,
      "num_tokens": 27449295.0,
      "step": 12685
    },
    {
      "entropy": 6.437607145309448,
      "epoch": 1.357964578094066,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000482478622506845,
      "loss": 5.9351,
      "mean_token_accuracy": 0.16188137978315353,
      "num_tokens": 27459357.0,
      "step": 12690
    },
    {
      "entropy": 6.43757815361023,
      "epoch": 1.35849965220183,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004824638275289019,
      "loss": 5.9815,
      "mean_token_accuracy": 0.15351207554340363,
      "num_tokens": 27469405.0,
      "step": 12695
    },
    {
      "entropy": 6.413189649581909,
      "epoch": 1.3590347263095939,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004824490265604699,
      "loss": 5.9939,
      "mean_token_accuracy": 0.15322716683149337,
      "num_tokens": 27478818.0,
      "step": 12700
    },
    {
      "entropy": 6.404108285903931,
      "epoch": 1.3595698004173578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004824342196019764,
      "loss": 5.9436,
      "mean_token_accuracy": 0.16296510994434357,
      "num_tokens": 27488631.0,
      "step": 12705
    },
    {
      "entropy": 6.44634690284729,
      "epoch": 1.3601048745251219,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048241940665384904,
      "loss": 6.0283,
      "mean_token_accuracy": 0.15203226655721663,
      "num_tokens": 27500895.0,
      "step": 12710
    },
    {
      "entropy": 6.417416906356811,
      "epoch": 1.3606399486328857,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004824045877165154,
      "loss": 6.065,
      "mean_token_accuracy": 0.14959678798913956,
      "num_tokens": 27512578.0,
      "step": 12715
    },
    {
      "entropy": 6.41545262336731,
      "epoch": 1.3611750227406496,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048238976279040344,
      "loss": 6.008,
      "mean_token_accuracy": 0.15087068900465966,
      "num_tokens": 27523448.0,
      "step": 12720
    },
    {
      "entropy": 6.417819833755493,
      "epoch": 1.3617100968484135,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004823749318759413,
      "loss": 6.089,
      "mean_token_accuracy": 0.14826277643442154,
      "num_tokens": 27534919.0,
      "step": 12725
    },
    {
      "entropy": 6.380204153060913,
      "epoch": 1.3622451709561774,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000482360094973557,
      "loss": 5.9102,
      "mean_token_accuracy": 0.15930013507604598,
      "num_tokens": 27545603.0,
      "step": 12730
    },
    {
      "entropy": 6.476020431518554,
      "epoch": 1.3627802450639415,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004823452520836792,
      "loss": 6.0287,
      "mean_token_accuracy": 0.1466519758105278,
      "num_tokens": 27556868.0,
      "step": 12735
    },
    {
      "entropy": 6.523520374298096,
      "epoch": 1.3633153191717053,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00048233040320673634,
      "loss": 6.1662,
      "mean_token_accuracy": 0.14357537180185317,
      "num_tokens": 27567200.0,
      "step": 12740
    },
    {
      "entropy": 6.396815824508667,
      "epoch": 1.3638503932794692,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048231554834315716,
      "loss": 5.9635,
      "mean_token_accuracy": 0.15427367091178895,
      "num_tokens": 27577759.0,
      "step": 12745
    },
    {
      "entropy": 6.427941608428955,
      "epoch": 1.364385467387233,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004823006874933708,
      "loss": 6.0707,
      "mean_token_accuracy": 0.14745811074972154,
      "num_tokens": 27588935.0,
      "step": 12750
    },
    {
      "entropy": 6.42252106666565,
      "epoch": 1.364920541494997,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004822858206578061,
      "loss": 6.0128,
      "mean_token_accuracy": 0.14758319407701492,
      "num_tokens": 27600306.0,
      "step": 12755
    },
    {
      "entropy": 6.425823831558228,
      "epoch": 1.365455615602761,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048227094783689244,
      "loss": 6.069,
      "mean_token_accuracy": 0.15159383565187454,
      "num_tokens": 27612322.0,
      "step": 12760
    },
    {
      "entropy": 6.446111249923706,
      "epoch": 1.365990689710525,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004822560690310594,
      "loss": 5.9969,
      "mean_token_accuracy": 0.15962644517421723,
      "num_tokens": 27623812.0,
      "step": 12765
    },
    {
      "entropy": 6.396254634857177,
      "epoch": 1.3665257638182888,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048224118424073644,
      "loss": 5.8876,
      "mean_token_accuracy": 0.15767722576856613,
      "num_tokens": 27633848.0,
      "step": 12770
    },
    {
      "entropy": 6.377055215835571,
      "epoch": 1.3670608379260527,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048222629346635336,
      "loss": 6.0117,
      "mean_token_accuracy": 0.1598612055182457,
      "num_tokens": 27644641.0,
      "step": 12775
    },
    {
      "entropy": 6.408574771881104,
      "epoch": 1.3675959120338166,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048221139670834025,
      "loss": 6.0071,
      "mean_token_accuracy": 0.15948888510465623,
      "num_tokens": 27655552.0,
      "step": 12780
    },
    {
      "entropy": 6.4131247997283936,
      "epoch": 1.3681309861415807,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004821964939671272,
      "loss": 6.0093,
      "mean_token_accuracy": 0.15965822488069534,
      "num_tokens": 27666396.0,
      "step": 12785
    },
    {
      "entropy": 6.4880167007446286,
      "epoch": 1.3686660602493446,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004821815852431444,
      "loss": 5.9463,
      "mean_token_accuracy": 0.16070968508720399,
      "num_tokens": 27676522.0,
      "step": 12790
    },
    {
      "entropy": 6.400334739685059,
      "epoch": 1.3692011343571084,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048216667053682243,
      "loss": 5.9724,
      "mean_token_accuracy": 0.1559053987264633,
      "num_tokens": 27687088.0,
      "step": 12795
    },
    {
      "entropy": 6.462497854232788,
      "epoch": 1.3697362084648723,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048215174984859207,
      "loss": 5.9912,
      "mean_token_accuracy": 0.16018935889005662,
      "num_tokens": 27697587.0,
      "step": 12800
    },
    {
      "entropy": 6.384485292434692,
      "epoch": 1.3702712825726362,
      "grad_norm": 0.9140625,
      "learning_rate": 0.000482136823178884,
      "loss": 5.9277,
      "mean_token_accuracy": 0.16198198348283768,
      "num_tokens": 27708345.0,
      "step": 12805
    },
    {
      "entropy": 6.489076042175293,
      "epoch": 1.3708063566804003,
      "grad_norm": 1.0,
      "learning_rate": 0.0004821218905281292,
      "loss": 6.1347,
      "mean_token_accuracy": 0.14937125220894815,
      "num_tokens": 27719592.0,
      "step": 12810
    },
    {
      "entropy": 6.467997217178345,
      "epoch": 1.3713414307881642,
      "grad_norm": 0.94140625,
      "learning_rate": 0.000482106951896759,
      "loss": 6.0115,
      "mean_token_accuracy": 0.1557191178202629,
      "num_tokens": 27730071.0,
      "step": 12815
    },
    {
      "entropy": 6.446466970443725,
      "epoch": 1.371876504895928,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048209200728520466,
      "loss": 6.0366,
      "mean_token_accuracy": 0.14917937219142913,
      "num_tokens": 27740514.0,
      "step": 12820
    },
    {
      "entropy": 6.4713990688323975,
      "epoch": 1.372411579003692,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048207705669389765,
      "loss": 6.0444,
      "mean_token_accuracy": 0.14687745720148088,
      "num_tokens": 27752088.0,
      "step": 12825
    },
    {
      "entropy": 6.521247386932373,
      "epoch": 1.3729466531114558,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048206210012326976,
      "loss": 6.0524,
      "mean_token_accuracy": 0.15126116424798966,
      "num_tokens": 27763401.0,
      "step": 12830
    },
    {
      "entropy": 6.432258462905883,
      "epoch": 1.37348172721922,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00048204713757375283,
      "loss": 5.9872,
      "mean_token_accuracy": 0.16090262979269027,
      "num_tokens": 27773665.0,
      "step": 12835
    },
    {
      "entropy": 6.409731006622314,
      "epoch": 1.3740168013269838,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004820321690457789,
      "loss": 5.9972,
      "mean_token_accuracy": 0.15933276414871217,
      "num_tokens": 27785474.0,
      "step": 12840
    },
    {
      "entropy": 6.37489800453186,
      "epoch": 1.3745518754347477,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048201719453978023,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15256000310182571,
      "num_tokens": 27796520.0,
      "step": 12845
    },
    {
      "entropy": 6.488599252700806,
      "epoch": 1.3750869495425118,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004820022140561891,
      "loss": 6.0788,
      "mean_token_accuracy": 0.14809737354516983,
      "num_tokens": 27807559.0,
      "step": 12850
    },
    {
      "entropy": 6.457851886749268,
      "epoch": 1.3756220236502756,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004819872275954381,
      "loss": 6.0379,
      "mean_token_accuracy": 0.15214517563581467,
      "num_tokens": 27818208.0,
      "step": 12855
    },
    {
      "entropy": 6.450079488754272,
      "epoch": 1.3761570977580395,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048197223515796,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15912457406520844,
      "num_tokens": 27828238.0,
      "step": 12860
    },
    {
      "entropy": 6.435713100433349,
      "epoch": 1.3766921718658034,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004819572367441877,
      "loss": 6.0334,
      "mean_token_accuracy": 0.16275415122509002,
      "num_tokens": 27837885.0,
      "step": 12865
    },
    {
      "entropy": 6.353283405303955,
      "epoch": 1.3772272459735673,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004819422323545543,
      "loss": 5.9351,
      "mean_token_accuracy": 0.1646234706044197,
      "num_tokens": 27847425.0,
      "step": 12870
    },
    {
      "entropy": 6.325208854675293,
      "epoch": 1.3777623200813314,
      "grad_norm": 0.87109375,
      "learning_rate": 0.000481927221989493,
      "loss": 5.9787,
      "mean_token_accuracy": 0.15377440303564072,
      "num_tokens": 27859606.0,
      "step": 12875
    },
    {
      "entropy": 6.499888134002686,
      "epoch": 1.3782973941890952,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004819122056494373,
      "loss": 6.0688,
      "mean_token_accuracy": 0.15369155406951904,
      "num_tokens": 27871651.0,
      "step": 12880
    },
    {
      "entropy": 6.487256002426148,
      "epoch": 1.3788324682968591,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048189718333482064,
      "loss": 5.9805,
      "mean_token_accuracy": 0.1562919318675995,
      "num_tokens": 27883138.0,
      "step": 12885
    },
    {
      "entropy": 6.5226898193359375,
      "epoch": 1.379367542404623,
      "grad_norm": 1.03125,
      "learning_rate": 0.000481882155046077,
      "loss": 6.054,
      "mean_token_accuracy": 0.15528296157717705,
      "num_tokens": 27893597.0,
      "step": 12890
    },
    {
      "entropy": 6.428260421752929,
      "epoch": 1.3799026165123869,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048186712078364006,
      "loss": 5.9623,
      "mean_token_accuracy": 0.16272054463624955,
      "num_tokens": 27904249.0,
      "step": 12895
    },
    {
      "entropy": 6.401567506790161,
      "epoch": 1.380437690620151,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048185208054794414,
      "loss": 6.0064,
      "mean_token_accuracy": 0.151760071516037,
      "num_tokens": 27914713.0,
      "step": 12900
    },
    {
      "entropy": 6.3788416385650635,
      "epoch": 1.3809727647279149,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004818370343394235,
      "loss": 6.0211,
      "mean_token_accuracy": 0.15688443183898926,
      "num_tokens": 27926257.0,
      "step": 12905
    },
    {
      "entropy": 6.43711633682251,
      "epoch": 1.3815078388356787,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004818219821585125,
      "loss": 5.9917,
      "mean_token_accuracy": 0.15891422033309938,
      "num_tokens": 27936839.0,
      "step": 12910
    },
    {
      "entropy": 6.517184543609619,
      "epoch": 1.3820429129434426,
      "grad_norm": 1.015625,
      "learning_rate": 0.00048180692400564586,
      "loss": 6.1005,
      "mean_token_accuracy": 0.1468397483229637,
      "num_tokens": 27946318.0,
      "step": 12915
    },
    {
      "entropy": 6.49223952293396,
      "epoch": 1.3825779870512065,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00048179185988125834,
      "loss": 6.1093,
      "mean_token_accuracy": 0.14687327891588212,
      "num_tokens": 27956990.0,
      "step": 12920
    },
    {
      "entropy": 6.467932176589966,
      "epoch": 1.3831130611589706,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000481776789785785,
      "loss": 6.0345,
      "mean_token_accuracy": 0.14623228833079338,
      "num_tokens": 27968187.0,
      "step": 12925
    },
    {
      "entropy": 6.5313972473144535,
      "epoch": 1.3836481352667345,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048176171371966083,
      "loss": 6.0844,
      "mean_token_accuracy": 0.15385643690824508,
      "num_tokens": 27978337.0,
      "step": 12930
    },
    {
      "entropy": 6.441867971420288,
      "epoch": 1.3841832093744983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004817466316833212,
      "loss": 6.0493,
      "mean_token_accuracy": 0.14992309361696243,
      "num_tokens": 27988681.0,
      "step": 12935
    },
    {
      "entropy": 6.3984767436981205,
      "epoch": 1.3847182834822622,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004817315436772017,
      "loss": 5.9232,
      "mean_token_accuracy": 0.16402436047792435,
      "num_tokens": 27999233.0,
      "step": 12940
    },
    {
      "entropy": 6.447150754928589,
      "epoch": 1.385253357590026,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004817164497017379,
      "loss": 6.0592,
      "mean_token_accuracy": 0.15348842069506646,
      "num_tokens": 28010242.0,
      "step": 12945
    },
    {
      "entropy": 6.417877101898194,
      "epoch": 1.3857884316977902,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004817013497573656,
      "loss": 6.0555,
      "mean_token_accuracy": 0.14896124973893166,
      "num_tokens": 28020424.0,
      "step": 12950
    },
    {
      "entropy": 6.46284589767456,
      "epoch": 1.386323505805554,
      "grad_norm": 0.984375,
      "learning_rate": 0.000481686243844521,
      "loss": 6.018,
      "mean_token_accuracy": 0.15482072681188583,
      "num_tokens": 28030287.0,
      "step": 12955
    },
    {
      "entropy": 6.464932823181153,
      "epoch": 1.386858579913318,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00048167113196364,
      "loss": 5.9746,
      "mean_token_accuracy": 0.1603196531534195,
      "num_tokens": 28039997.0,
      "step": 12960
    },
    {
      "entropy": 6.432046937942505,
      "epoch": 1.387393654021082,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004816560141151593,
      "loss": 6.0032,
      "mean_token_accuracy": 0.1493307515978813,
      "num_tokens": 28051381.0,
      "step": 12965
    },
    {
      "entropy": 6.39271821975708,
      "epoch": 1.3879287281288457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004816408902995151,
      "loss": 5.9746,
      "mean_token_accuracy": 0.15370654463768005,
      "num_tokens": 28063603.0,
      "step": 12970
    },
    {
      "entropy": 6.416665506362915,
      "epoch": 1.3884638022366098,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004816257605171442,
      "loss": 5.979,
      "mean_token_accuracy": 0.15920588076114656,
      "num_tokens": 28073942.0,
      "step": 12975
    },
    {
      "entropy": 6.409616470336914,
      "epoch": 1.3889988763443737,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048161062476848363,
      "loss": 6.0087,
      "mean_token_accuracy": 0.1566917985677719,
      "num_tokens": 28083837.0,
      "step": 12980
    },
    {
      "entropy": 6.373976755142212,
      "epoch": 1.3895339504521376,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004815954830539702,
      "loss": 5.9356,
      "mean_token_accuracy": 0.16198163628578185,
      "num_tokens": 28094150.0,
      "step": 12985
    },
    {
      "entropy": 6.405348110198974,
      "epoch": 1.3900690245599017,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048158033537404126,
      "loss": 6.0176,
      "mean_token_accuracy": 0.15602971911430358,
      "num_tokens": 28106345.0,
      "step": 12990
    },
    {
      "entropy": 6.4165972709655765,
      "epoch": 1.3906040986676655,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048156518172913417,
      "loss": 5.931,
      "mean_token_accuracy": 0.15412394255399703,
      "num_tokens": 28116858.0,
      "step": 12995
    },
    {
      "entropy": 6.430170249938965,
      "epoch": 1.3911391727754294,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004815500221196865,
      "loss": 6.1149,
      "mean_token_accuracy": 0.14792469888925552,
      "num_tokens": 28128390.0,
      "step": 13000
    },
    {
      "entropy": 6.402363967895508,
      "epoch": 1.3916742468831933,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048153485654613587,
      "loss": 5.9481,
      "mean_token_accuracy": 0.1604602798819542,
      "num_tokens": 28139421.0,
      "step": 13005
    },
    {
      "entropy": 6.466789484024048,
      "epoch": 1.3922093209909572,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00048151968500892043,
      "loss": 6.0539,
      "mean_token_accuracy": 0.15307622402906418,
      "num_tokens": 28151076.0,
      "step": 13010
    },
    {
      "entropy": 6.495349407196045,
      "epoch": 1.3927443950987213,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048150450750847795,
      "loss": 6.1418,
      "mean_token_accuracy": 0.14950639456510545,
      "num_tokens": 28162079.0,
      "step": 13015
    },
    {
      "entropy": 6.449616622924805,
      "epoch": 1.3932794692064852,
      "grad_norm": 0.87890625,
      "learning_rate": 0.000481489324045247,
      "loss": 6.0635,
      "mean_token_accuracy": 0.14990775436162948,
      "num_tokens": 28174442.0,
      "step": 13020
    },
    {
      "entropy": 6.442434406280517,
      "epoch": 1.393814543314249,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048147413461966564,
      "loss": 5.9401,
      "mean_token_accuracy": 0.16076561957597732,
      "num_tokens": 28184643.0,
      "step": 13025
    },
    {
      "entropy": 6.358885478973389,
      "epoch": 1.394349617422013,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048145893923217276,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1577655032277107,
      "num_tokens": 28194021.0,
      "step": 13030
    },
    {
      "entropy": 6.463983583450317,
      "epoch": 1.3948846915297768,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004814437378832071,
      "loss": 6.0316,
      "mean_token_accuracy": 0.14799418672919273,
      "num_tokens": 28204750.0,
      "step": 13035
    },
    {
      "entropy": 6.512043094635009,
      "epoch": 1.3954197656375409,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004814285305732074,
      "loss": 6.0765,
      "mean_token_accuracy": 0.1507783204317093,
      "num_tokens": 28216499.0,
      "step": 13040
    },
    {
      "entropy": 6.5124509811401365,
      "epoch": 1.3959548397453048,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048141331730261293,
      "loss": 6.0215,
      "mean_token_accuracy": 0.15749555081129074,
      "num_tokens": 28226573.0,
      "step": 13045
    },
    {
      "entropy": 6.401469039916992,
      "epoch": 1.3964899138530686,
      "grad_norm": 1.09375,
      "learning_rate": 0.000481398098071863,
      "loss": 5.984,
      "mean_token_accuracy": 0.15759838968515397,
      "num_tokens": 28236688.0,
      "step": 13050
    },
    {
      "entropy": 6.405260801315308,
      "epoch": 1.3970249879608325,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00048138287288139686,
      "loss": 6.0505,
      "mean_token_accuracy": 0.15214207395911217,
      "num_tokens": 28247477.0,
      "step": 13055
    },
    {
      "entropy": 6.411235809326172,
      "epoch": 1.3975600620685964,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004813676417316543,
      "loss": 6.035,
      "mean_token_accuracy": 0.15534068644046783,
      "num_tokens": 28258286.0,
      "step": 13060
    },
    {
      "entropy": 6.522611951828003,
      "epoch": 1.3980951361763605,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004813524046230751,
      "loss": 6.1833,
      "mean_token_accuracy": 0.14365952610969543,
      "num_tokens": 28268587.0,
      "step": 13065
    },
    {
      "entropy": 6.5240504264831545,
      "epoch": 1.3986302102841244,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048133716155609926,
      "loss": 6.0632,
      "mean_token_accuracy": 0.14694000482559205,
      "num_tokens": 28279352.0,
      "step": 13070
    },
    {
      "entropy": 6.4104633808135985,
      "epoch": 1.3991652843918883,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004813219125311668,
      "loss": 5.9957,
      "mean_token_accuracy": 0.15012027472257614,
      "num_tokens": 28290114.0,
      "step": 13075
    },
    {
      "entropy": 6.426528739929199,
      "epoch": 1.3997003584996521,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048130665754871814,
      "loss": 6.072,
      "mean_token_accuracy": 0.15256094485521315,
      "num_tokens": 28301845.0,
      "step": 13080
    },
    {
      "entropy": 6.536268520355224,
      "epoch": 1.400235432607416,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048129139660919374,
      "loss": 6.0533,
      "mean_token_accuracy": 0.1526922568678856,
      "num_tokens": 28313162.0,
      "step": 13085
    },
    {
      "entropy": 6.450088310241699,
      "epoch": 1.40077050671518,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048127612971303425,
      "loss": 5.9966,
      "mean_token_accuracy": 0.148436439037323,
      "num_tokens": 28323888.0,
      "step": 13090
    },
    {
      "entropy": 6.440907049179077,
      "epoch": 1.401305580822944,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004812608568606805,
      "loss": 6.0277,
      "mean_token_accuracy": 0.14747866094112397,
      "num_tokens": 28333735.0,
      "step": 13095
    },
    {
      "entropy": 6.500397348403931,
      "epoch": 1.4018406549307079,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004812455780525735,
      "loss": 6.0413,
      "mean_token_accuracy": 0.14899859428405762,
      "num_tokens": 28343947.0,
      "step": 13100
    },
    {
      "entropy": 6.491208744049072,
      "epoch": 1.402375729038472,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004812302932891544,
      "loss": 6.0942,
      "mean_token_accuracy": 0.1461116909980774,
      "num_tokens": 28354635.0,
      "step": 13105
    },
    {
      "entropy": 6.450083065032959,
      "epoch": 1.4029108031462356,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004812150025708646,
      "loss": 6.0227,
      "mean_token_accuracy": 0.14991040155291557,
      "num_tokens": 28365392.0,
      "step": 13110
    },
    {
      "entropy": 6.49298734664917,
      "epoch": 1.4034458772539997,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00048119970589814557,
      "loss": 6.0137,
      "mean_token_accuracy": 0.1489061161875725,
      "num_tokens": 28376527.0,
      "step": 13115
    },
    {
      "entropy": 6.442909860610962,
      "epoch": 1.4039809513617636,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004811844032714389,
      "loss": 6.0249,
      "mean_token_accuracy": 0.15619166791439057,
      "num_tokens": 28387178.0,
      "step": 13120
    },
    {
      "entropy": 6.375845956802368,
      "epoch": 1.4045160254695275,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048116909469118663,
      "loss": 5.9244,
      "mean_token_accuracy": 0.1544420450925827,
      "num_tokens": 28397578.0,
      "step": 13125
    },
    {
      "entropy": 6.408326292037964,
      "epoch": 1.4050510995772916,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004811537801578308,
      "loss": 6.0604,
      "mean_token_accuracy": 0.1458878591656685,
      "num_tokens": 28408933.0,
      "step": 13130
    },
    {
      "entropy": 6.419409513473511,
      "epoch": 1.4055861736850555,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00048113845967181346,
      "loss": 6.0437,
      "mean_token_accuracy": 0.14672365486621858,
      "num_tokens": 28421906.0,
      "step": 13135
    },
    {
      "entropy": 6.471211099624634,
      "epoch": 1.4061212477928193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000481123133233577,
      "loss": 5.988,
      "mean_token_accuracy": 0.15552569627761842,
      "num_tokens": 28432677.0,
      "step": 13140
    },
    {
      "entropy": 6.435432100296021,
      "epoch": 1.4066563219005832,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048110780084356414,
      "loss": 6.0879,
      "mean_token_accuracy": 0.15051815211772918,
      "num_tokens": 28443795.0,
      "step": 13145
    },
    {
      "entropy": 6.452101373672486,
      "epoch": 1.407191396008347,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004810924625022174,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15602184683084488,
      "num_tokens": 28455727.0,
      "step": 13150
    },
    {
      "entropy": 6.50575270652771,
      "epoch": 1.4077264701161112,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004810771182099799,
      "loss": 6.0706,
      "mean_token_accuracy": 0.14557401835918427,
      "num_tokens": 28467554.0,
      "step": 13155
    },
    {
      "entropy": 6.51044111251831,
      "epoch": 1.408261544223875,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004810617679672945,
      "loss": 6.0814,
      "mean_token_accuracy": 0.15813643634319305,
      "num_tokens": 28479268.0,
      "step": 13160
    },
    {
      "entropy": 6.480827951431275,
      "epoch": 1.408796618331639,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048104641177460443,
      "loss": 6.0596,
      "mean_token_accuracy": 0.15060470327734948,
      "num_tokens": 28489887.0,
      "step": 13165
    },
    {
      "entropy": 6.378494024276733,
      "epoch": 1.4093316924394028,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004810310496323533,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15918494611978531,
      "num_tokens": 28500736.0,
      "step": 13170
    },
    {
      "entropy": 6.3602564334869385,
      "epoch": 1.4098667665471667,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048101568154098446,
      "loss": 5.9306,
      "mean_token_accuracy": 0.16413117498159407,
      "num_tokens": 28511458.0,
      "step": 13175
    },
    {
      "entropy": 6.494165372848511,
      "epoch": 1.4104018406549308,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048100030750094185,
      "loss": 6.0457,
      "mean_token_accuracy": 0.14674134701490402,
      "num_tokens": 28522354.0,
      "step": 13180
    },
    {
      "entropy": 6.407954835891724,
      "epoch": 1.4109369147626947,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004809849275126693,
      "loss": 5.9012,
      "mean_token_accuracy": 0.16964951157569885,
      "num_tokens": 28533061.0,
      "step": 13185
    },
    {
      "entropy": 6.43375973701477,
      "epoch": 1.4114719888704585,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00048096954157661085,
      "loss": 6.0442,
      "mean_token_accuracy": 0.1490870751440525,
      "num_tokens": 28544230.0,
      "step": 13190
    },
    {
      "entropy": 6.464255905151367,
      "epoch": 1.4120070629782224,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048095414969321086,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1566719025373459,
      "num_tokens": 28555461.0,
      "step": 13195
    },
    {
      "entropy": 6.436396980285645,
      "epoch": 1.4125421370859863,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00048093875186291376,
      "loss": 5.9561,
      "mean_token_accuracy": 0.1570437878370285,
      "num_tokens": 28565992.0,
      "step": 13200
    },
    {
      "entropy": 6.420431995391846,
      "epoch": 1.4130772111937504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048092334808616413,
      "loss": 5.9898,
      "mean_token_accuracy": 0.15589201897382737,
      "num_tokens": 28575802.0,
      "step": 13205
    },
    {
      "entropy": 6.391258096694946,
      "epoch": 1.4136122853015143,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004809079383634067,
      "loss": 5.9597,
      "mean_token_accuracy": 0.16196082532405853,
      "num_tokens": 28586982.0,
      "step": 13210
    },
    {
      "entropy": 6.461773347854614,
      "epoch": 1.4141473594092782,
      "grad_norm": 0.875,
      "learning_rate": 0.00048089252269508656,
      "loss": 6.0539,
      "mean_token_accuracy": 0.15034496039152145,
      "num_tokens": 28598267.0,
      "step": 13215
    },
    {
      "entropy": 6.449415063858032,
      "epoch": 1.414682433517042,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048087710108164874,
      "loss": 6.0412,
      "mean_token_accuracy": 0.1543801337480545,
      "num_tokens": 28608213.0,
      "step": 13220
    },
    {
      "entropy": 6.409115505218506,
      "epoch": 1.415217507624806,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004808616735235385,
      "loss": 6.0228,
      "mean_token_accuracy": 0.14734771996736526,
      "num_tokens": 28619674.0,
      "step": 13225
    },
    {
      "entropy": 6.419378709793091,
      "epoch": 1.41575258173257,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004808462400212014,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15777531862258912,
      "num_tokens": 28630824.0,
      "step": 13230
    },
    {
      "entropy": 6.428067588806153,
      "epoch": 1.416287655840334,
      "grad_norm": 0.921875,
      "learning_rate": 0.00048083080057508313,
      "loss": 6.0609,
      "mean_token_accuracy": 0.14990590065717696,
      "num_tokens": 28642270.0,
      "step": 13235
    },
    {
      "entropy": 6.459448528289795,
      "epoch": 1.4168227299480978,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00048081535518562933,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14875992387533188,
      "num_tokens": 28653649.0,
      "step": 13240
    },
    {
      "entropy": 6.420959186553955,
      "epoch": 1.4173578040558619,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000480799903853286,
      "loss": 5.9322,
      "mean_token_accuracy": 0.15323849022388458,
      "num_tokens": 28664044.0,
      "step": 13245
    },
    {
      "entropy": 6.42501220703125,
      "epoch": 1.4178928781636255,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004807844465784995,
      "loss": 6.0026,
      "mean_token_accuracy": 0.15736944377422332,
      "num_tokens": 28675394.0,
      "step": 13250
    },
    {
      "entropy": 6.37998104095459,
      "epoch": 1.4184279522713896,
      "grad_norm": 1.0,
      "learning_rate": 0.00048076898336171596,
      "loss": 5.9756,
      "mean_token_accuracy": 0.16297362595796586,
      "num_tokens": 28686292.0,
      "step": 13255
    },
    {
      "entropy": 6.435836601257324,
      "epoch": 1.4189630263791535,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048075351420338196,
      "loss": 6.033,
      "mean_token_accuracy": 0.15263700634241104,
      "num_tokens": 28696928.0,
      "step": 13260
    },
    {
      "entropy": 6.420453691482544,
      "epoch": 1.4194981004869174,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004807380391039441,
      "loss": 6.0347,
      "mean_token_accuracy": 0.15490647107362748,
      "num_tokens": 28707986.0,
      "step": 13265
    },
    {
      "entropy": 6.488261604309082,
      "epoch": 1.4200331745946815,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00048072255806384936,
      "loss": 6.0516,
      "mean_token_accuracy": 0.15507817268371582,
      "num_tokens": 28719296.0,
      "step": 13270
    },
    {
      "entropy": 6.418520593643189,
      "epoch": 1.4205682487024454,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004807070710835447,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16128877103328704,
      "num_tokens": 28729523.0,
      "step": 13275
    },
    {
      "entropy": 6.47039761543274,
      "epoch": 1.4211033228102092,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004806915781634771,
      "loss": 6.0353,
      "mean_token_accuracy": 0.1538715183734894,
      "num_tokens": 28739026.0,
      "step": 13280
    },
    {
      "entropy": 6.483090829849243,
      "epoch": 1.4216383969179731,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004806760793040942,
      "loss": 6.1054,
      "mean_token_accuracy": 0.1519848808646202,
      "num_tokens": 28749513.0,
      "step": 13285
    },
    {
      "entropy": 6.504253101348877,
      "epoch": 1.422173471025737,
      "grad_norm": 0.890625,
      "learning_rate": 0.00048066057450584334,
      "loss": 6.003,
      "mean_token_accuracy": 0.1540033996105194,
      "num_tokens": 28760722.0,
      "step": 13290
    },
    {
      "entropy": 6.437024927139282,
      "epoch": 1.422708545133501,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004806450637691723,
      "loss": 5.9442,
      "mean_token_accuracy": 0.16350374221801758,
      "num_tokens": 28771235.0,
      "step": 13295
    },
    {
      "entropy": 6.385517406463623,
      "epoch": 1.423243619241265,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048062954709452907,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15432808101177214,
      "num_tokens": 28782642.0,
      "step": 13300
    },
    {
      "entropy": 6.470485210418701,
      "epoch": 1.4237786933490288,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004806140244823615,
      "loss": 5.9938,
      "mean_token_accuracy": 0.1512120172381401,
      "num_tokens": 28793628.0,
      "step": 13305
    },
    {
      "entropy": 6.526476335525513,
      "epoch": 1.4243137674567927,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048059849593311776,
      "loss": 6.0541,
      "mean_token_accuracy": 0.15319129526615144,
      "num_tokens": 28804034.0,
      "step": 13310
    },
    {
      "entropy": 6.471170854568482,
      "epoch": 1.4248488415645566,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004805829614472464,
      "loss": 5.9965,
      "mean_token_accuracy": 0.15557298213243484,
      "num_tokens": 28814679.0,
      "step": 13315
    },
    {
      "entropy": 6.413617420196533,
      "epoch": 1.4253839156723207,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00048056742102519593,
      "loss": 6.0185,
      "mean_token_accuracy": 0.15456014275550842,
      "num_tokens": 28825130.0,
      "step": 13320
    },
    {
      "entropy": 6.451051092147827,
      "epoch": 1.4259189897800846,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00048055187466741506,
      "loss": 6.0418,
      "mean_token_accuracy": 0.15200690031051636,
      "num_tokens": 28835500.0,
      "step": 13325
    },
    {
      "entropy": 6.438564252853394,
      "epoch": 1.4264540638878485,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004805363223743527,
      "loss": 5.9875,
      "mean_token_accuracy": 0.15405036211013795,
      "num_tokens": 28845794.0,
      "step": 13330
    },
    {
      "entropy": 6.457977056503296,
      "epoch": 1.4269891379956123,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004805207641464579,
      "loss": 6.0036,
      "mean_token_accuracy": 0.1515161857008934,
      "num_tokens": 28856326.0,
      "step": 13335
    },
    {
      "entropy": 6.377807998657227,
      "epoch": 1.4275242121033762,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004805051999841799,
      "loss": 5.9685,
      "mean_token_accuracy": 0.15481216460466385,
      "num_tokens": 28867592.0,
      "step": 13340
    },
    {
      "entropy": 6.5410112857818605,
      "epoch": 1.4280592862111403,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004804896298879682,
      "loss": 6.1362,
      "mean_token_accuracy": 0.1461635410785675,
      "num_tokens": 28880352.0,
      "step": 13345
    },
    {
      "entropy": 6.483788013458252,
      "epoch": 1.4285943603189042,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004804740538582723,
      "loss": 5.9966,
      "mean_token_accuracy": 0.1561988338828087,
      "num_tokens": 28892012.0,
      "step": 13350
    },
    {
      "entropy": 6.4211190700531,
      "epoch": 1.429129434426668,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004804584718955419,
      "loss": 5.9691,
      "mean_token_accuracy": 0.15964429453015327,
      "num_tokens": 28902247.0,
      "step": 13355
    },
    {
      "entropy": 6.420061016082764,
      "epoch": 1.429664508534432,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000480442884000227,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14670104086399077,
      "num_tokens": 28913222.0,
      "step": 13360
    },
    {
      "entropy": 6.483000135421753,
      "epoch": 1.4301995826421958,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004804272901727778,
      "loss": 6.0031,
      "mean_token_accuracy": 0.15413570106029512,
      "num_tokens": 28924038.0,
      "step": 13365
    },
    {
      "entropy": 6.447080278396607,
      "epoch": 1.43073465674996,
      "grad_norm": 1.078125,
      "learning_rate": 0.00048041169041364437,
      "loss": 6.029,
      "mean_token_accuracy": 0.15813076347112656,
      "num_tokens": 28934448.0,
      "step": 13370
    },
    {
      "entropy": 6.447243022918701,
      "epoch": 1.4312697308577238,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048039608472327724,
      "loss": 5.9812,
      "mean_token_accuracy": 0.15394294559955596,
      "num_tokens": 28945175.0,
      "step": 13375
    },
    {
      "entropy": 6.432204723358154,
      "epoch": 1.4318048049654877,
      "grad_norm": 1.0625,
      "learning_rate": 0.00048038047310212706,
      "loss": 5.9526,
      "mean_token_accuracy": 0.16240905076265336,
      "num_tokens": 28955116.0,
      "step": 13380
    },
    {
      "entropy": 6.500591564178467,
      "epoch": 1.4323398790732518,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00048036485555064454,
      "loss": 6.0372,
      "mean_token_accuracy": 0.1503146469593048,
      "num_tokens": 28966999.0,
      "step": 13385
    },
    {
      "entropy": 6.429563426971436,
      "epoch": 1.4328749531810157,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00048034923206928065,
      "loss": 6.0274,
      "mean_token_accuracy": 0.14817124456167222,
      "num_tokens": 28977388.0,
      "step": 13390
    },
    {
      "entropy": 6.4212727546691895,
      "epoch": 1.4334100272887795,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048033360265848664,
      "loss": 5.9077,
      "mean_token_accuracy": 0.16248583197593688,
      "num_tokens": 28987792.0,
      "step": 13395
    },
    {
      "entropy": 6.517446422576905,
      "epoch": 1.4339451013965434,
      "grad_norm": 0.96875,
      "learning_rate": 0.00048031796731871364,
      "loss": 6.0643,
      "mean_token_accuracy": 0.15267074555158616,
      "num_tokens": 28998319.0,
      "step": 13400
    },
    {
      "entropy": 6.4156323909759525,
      "epoch": 1.4344801755043073,
      "grad_norm": 0.984375,
      "learning_rate": 0.00048030232605041323,
      "loss": 6.0442,
      "mean_token_accuracy": 0.149434694647789,
      "num_tokens": 29007942.0,
      "step": 13405
    },
    {
      "entropy": 6.456836318969726,
      "epoch": 1.4350152496120714,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004802866788540369,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15850310474634172,
      "num_tokens": 29018217.0,
      "step": 13410
    },
    {
      "entropy": 6.438801908493042,
      "epoch": 1.4355503237198353,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004802710257300367,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15498670041561127,
      "num_tokens": 29028472.0,
      "step": 13415
    },
    {
      "entropy": 6.444475984573364,
      "epoch": 1.4360853978275991,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004802553666788644,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15801013261079788,
      "num_tokens": 29040553.0,
      "step": 13420
    },
    {
      "entropy": 6.502419281005859,
      "epoch": 1.436620471935363,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00048023970170097226,
      "loss": 6.0311,
      "mean_token_accuracy": 0.14904619604349137,
      "num_tokens": 29050142.0,
      "step": 13425
    },
    {
      "entropy": 6.42969970703125,
      "epoch": 1.437155546043127,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00048022403079681254,
      "loss": 5.9682,
      "mean_token_accuracy": 0.15215924382209778,
      "num_tokens": 29061762.0,
      "step": 13430
    },
    {
      "entropy": 6.436580085754395,
      "epoch": 1.437690620150891,
      "grad_norm": 0.953125,
      "learning_rate": 0.00048020835396683775,
      "loss": 6.0387,
      "mean_token_accuracy": 0.15467469394207,
      "num_tokens": 29072095.0,
      "step": 13435
    },
    {
      "entropy": 6.4877783298492435,
      "epoch": 1.4382256942586549,
      "grad_norm": 1.03125,
      "learning_rate": 0.00048019267121150066,
      "loss": 6.0487,
      "mean_token_accuracy": 0.1525133952498436,
      "num_tokens": 29082561.0,
      "step": 13440
    },
    {
      "entropy": 6.499413871765137,
      "epoch": 1.4387607683664188,
      "grad_norm": 0.9375,
      "learning_rate": 0.00048017698253125397,
      "loss": 5.9992,
      "mean_token_accuracy": 0.15325118899345397,
      "num_tokens": 29093813.0,
      "step": 13445
    },
    {
      "entropy": 6.442175674438476,
      "epoch": 1.4392958424741826,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004801612879265507,
      "loss": 6.0021,
      "mean_token_accuracy": 0.15744624584913253,
      "num_tokens": 29104064.0,
      "step": 13450
    },
    {
      "entropy": 6.412386512756347,
      "epoch": 1.4398309165819465,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004801455873978442,
      "loss": 5.9966,
      "mean_token_accuracy": 0.15732175409793853,
      "num_tokens": 29115086.0,
      "step": 13455
    },
    {
      "entropy": 6.427259111404419,
      "epoch": 1.4403659906897106,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004801298809455876,
      "loss": 5.981,
      "mean_token_accuracy": 0.16022816002368928,
      "num_tokens": 29125536.0,
      "step": 13460
    },
    {
      "entropy": 6.479255342483521,
      "epoch": 1.4409010647974745,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004801141685702345,
      "loss": 6.0601,
      "mean_token_accuracy": 0.15428247302770615,
      "num_tokens": 29137007.0,
      "step": 13465
    },
    {
      "entropy": 6.433736753463745,
      "epoch": 1.4414361389052384,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00048009845027223864,
      "loss": 6.0448,
      "mean_token_accuracy": 0.14426361992955208,
      "num_tokens": 29147072.0,
      "step": 13470
    },
    {
      "entropy": 6.41332426071167,
      "epoch": 1.4419712130130022,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00048008272605205377,
      "loss": 6.0068,
      "mean_token_accuracy": 0.1537831097841263,
      "num_tokens": 29157803.0,
      "step": 13475
    },
    {
      "entropy": 6.483035373687744,
      "epoch": 1.4425062871207661,
      "grad_norm": 1.046875,
      "learning_rate": 0.00048006699591013417,
      "loss": 5.9728,
      "mean_token_accuracy": 0.1570359244942665,
      "num_tokens": 29168173.0,
      "step": 13480
    },
    {
      "entropy": 6.504761266708374,
      "epoch": 1.4430413612285302,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004800512598469337,
      "loss": 6.0409,
      "mean_token_accuracy": 0.1504758045077324,
      "num_tokens": 29179377.0,
      "step": 13485
    },
    {
      "entropy": 6.447898864746094,
      "epoch": 1.443576435336294,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000480035517862907,
      "loss": 6.0517,
      "mean_token_accuracy": 0.15051351338624955,
      "num_tokens": 29190909.0,
      "step": 13490
    },
    {
      "entropy": 6.409933423995971,
      "epoch": 1.444111509444058,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00048001976995850856,
      "loss": 5.9648,
      "mean_token_accuracy": 0.16024267673492432,
      "num_tokens": 29200977.0,
      "step": 13495
    },
    {
      "entropy": 6.457130432128906,
      "epoch": 1.4446465835518219,
      "grad_norm": 1.0,
      "learning_rate": 0.00048000401613419296,
      "loss": 5.9975,
      "mean_token_accuracy": 0.15116413235664367,
      "num_tokens": 29210702.0,
      "step": 13500
    },
    {
      "entropy": 6.46932692527771,
      "epoch": 1.4451816576595857,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00047998825639041534,
      "loss": 6.1069,
      "mean_token_accuracy": 0.14859481751918793,
      "num_tokens": 29221338.0,
      "step": 13505
    },
    {
      "entropy": 6.472613525390625,
      "epoch": 1.4457167317673498,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004799724907276306,
      "loss": 5.946,
      "mean_token_accuracy": 0.16109993010759355,
      "num_tokens": 29232091.0,
      "step": 13510
    },
    {
      "entropy": 6.364707040786743,
      "epoch": 1.4462518058751137,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004799567191462939,
      "loss": 5.9597,
      "mean_token_accuracy": 0.15927157253026963,
      "num_tokens": 29241903.0,
      "step": 13515
    },
    {
      "entropy": 6.416997718811035,
      "epoch": 1.4467868799828776,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047994094164686074,
      "loss": 6.0538,
      "mean_token_accuracy": 0.1580089658498764,
      "num_tokens": 29253414.0,
      "step": 13520
    },
    {
      "entropy": 6.4926714420318605,
      "epoch": 1.4473219540906417,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004799251582297868,
      "loss": 5.9419,
      "mean_token_accuracy": 0.16425145417451859,
      "num_tokens": 29264066.0,
      "step": 13525
    },
    {
      "entropy": 6.471937274932861,
      "epoch": 1.4478570281984056,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047990936889552767,
      "loss": 6.009,
      "mean_token_accuracy": 0.1541047215461731,
      "num_tokens": 29274915.0,
      "step": 13530
    },
    {
      "entropy": 6.428443861007691,
      "epoch": 1.4483921023061694,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004798935736445392,
      "loss": 5.9946,
      "mean_token_accuracy": 0.1537620648741722,
      "num_tokens": 29287345.0,
      "step": 13535
    },
    {
      "entropy": 6.4170855522155765,
      "epoch": 1.4489271764139333,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004798777724772777,
      "loss": 5.9728,
      "mean_token_accuracy": 0.16247347593307496,
      "num_tokens": 29297432.0,
      "step": 13540
    },
    {
      "entropy": 6.430151891708374,
      "epoch": 1.4494622505216972,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047986196539419936,
      "loss": 5.9096,
      "mean_token_accuracy": 0.16250572055578233,
      "num_tokens": 29307534.0,
      "step": 13545
    },
    {
      "entropy": 6.440432834625244,
      "epoch": 1.4499973246294613,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004798461523957604,
      "loss": 5.9661,
      "mean_token_accuracy": 0.1589341029524803,
      "num_tokens": 29319412.0,
      "step": 13550
    },
    {
      "entropy": 6.3658668994903564,
      "epoch": 1.4505323987372252,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047983033348241767,
      "loss": 5.955,
      "mean_token_accuracy": 0.16490764170885086,
      "num_tokens": 29330477.0,
      "step": 13555
    },
    {
      "entropy": 6.375678586959839,
      "epoch": 1.451067472844989,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004798145086546278,
      "loss": 5.9982,
      "mean_token_accuracy": 0.15576742440462113,
      "num_tokens": 29341481.0,
      "step": 13560
    },
    {
      "entropy": 6.433413457870484,
      "epoch": 1.451602546952753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004797986779128478,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15724315047264098,
      "num_tokens": 29353273.0,
      "step": 13565
    },
    {
      "entropy": 6.467847871780395,
      "epoch": 1.4521376210605168,
      "grad_norm": 1.0,
      "learning_rate": 0.0004797828412575348,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15608400851488113,
      "num_tokens": 29364108.0,
      "step": 13570
    },
    {
      "entropy": 6.377606010437011,
      "epoch": 1.452672695168281,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047976699868914594,
      "loss": 5.9603,
      "mean_token_accuracy": 0.16145821958780288,
      "num_tokens": 29375272.0,
      "step": 13575
    },
    {
      "entropy": 6.416382551193237,
      "epoch": 1.4532077692760448,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004797511502081388,
      "loss": 5.996,
      "mean_token_accuracy": 0.15781972855329512,
      "num_tokens": 29385749.0,
      "step": 13580
    },
    {
      "entropy": 6.3964580535888675,
      "epoch": 1.4537428433838087,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000479735295814971,
      "loss": 5.94,
      "mean_token_accuracy": 0.16068863570690156,
      "num_tokens": 29396430.0,
      "step": 13585
    },
    {
      "entropy": 6.468331289291382,
      "epoch": 1.4542779174915725,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047971943551010035,
      "loss": 5.9895,
      "mean_token_accuracy": 0.15232473164796828,
      "num_tokens": 29407020.0,
      "step": 13590
    },
    {
      "entropy": 6.441915845870971,
      "epoch": 1.4548129915993364,
      "grad_norm": 1.0,
      "learning_rate": 0.00047970356929398467,
      "loss": 5.9805,
      "mean_token_accuracy": 0.1538350597023964,
      "num_tokens": 29418428.0,
      "step": 13595
    },
    {
      "entropy": 6.479972982406617,
      "epoch": 1.4553480657071005,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004796876971670822,
      "loss": 6.091,
      "mean_token_accuracy": 0.1516618862748146,
      "num_tokens": 29430611.0,
      "step": 13600
    },
    {
      "entropy": 6.433150434494019,
      "epoch": 1.4558831398148644,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004796718191298512,
      "loss": 5.9909,
      "mean_token_accuracy": 0.14898306429386138,
      "num_tokens": 29441545.0,
      "step": 13605
    },
    {
      "entropy": 6.481541919708252,
      "epoch": 1.4564182139226283,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004796559351827503,
      "loss": 6.0781,
      "mean_token_accuracy": 0.15533004850149154,
      "num_tokens": 29452529.0,
      "step": 13610
    },
    {
      "entropy": 6.406052303314209,
      "epoch": 1.4569532880303921,
      "grad_norm": 0.91015625,
      "learning_rate": 0.000479640045326238,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1580472007393837,
      "num_tokens": 29463860.0,
      "step": 13615
    },
    {
      "entropy": 6.48933367729187,
      "epoch": 1.457488362138156,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004796241495607731,
      "loss": 6.0355,
      "mean_token_accuracy": 0.15531430542469024,
      "num_tokens": 29475004.0,
      "step": 13620
    },
    {
      "entropy": 6.490535163879395,
      "epoch": 1.4580234362459201,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004796082478868146,
      "loss": 6.1165,
      "mean_token_accuracy": 0.14709247276186943,
      "num_tokens": 29485921.0,
      "step": 13625
    },
    {
      "entropy": 6.4906196117401125,
      "epoch": 1.458558510353684,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047959234030482185,
      "loss": 5.9658,
      "mean_token_accuracy": 0.16365956962108613,
      "num_tokens": 29495729.0,
      "step": 13630
    },
    {
      "entropy": 6.456753206253052,
      "epoch": 1.4590935844614479,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004795764268152538,
      "loss": 6.0405,
      "mean_token_accuracy": 0.15199915915727616,
      "num_tokens": 29506051.0,
      "step": 13635
    },
    {
      "entropy": 6.407734298706055,
      "epoch": 1.4596286585692118,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004795605074185704,
      "loss": 6.0004,
      "mean_token_accuracy": 0.15606719702482225,
      "num_tokens": 29516234.0,
      "step": 13640
    },
    {
      "entropy": 6.465189456939697,
      "epoch": 1.4601637326769756,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047954458211523095,
      "loss": 6.0025,
      "mean_token_accuracy": 0.15261295586824417,
      "num_tokens": 29527649.0,
      "step": 13645
    },
    {
      "entropy": 6.479425668716431,
      "epoch": 1.4606988067847397,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047952865090569545,
      "loss": 6.0578,
      "mean_token_accuracy": 0.14844823703169824,
      "num_tokens": 29539150.0,
      "step": 13650
    },
    {
      "entropy": 6.409944534301758,
      "epoch": 1.4612338808925036,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047951271379042393,
      "loss": 5.9345,
      "mean_token_accuracy": 0.1595616802573204,
      "num_tokens": 29550058.0,
      "step": 13655
    },
    {
      "entropy": 6.3570703029632565,
      "epoch": 1.4617689550002675,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004794967707698765,
      "loss": 5.9365,
      "mean_token_accuracy": 0.16504984349012375,
      "num_tokens": 29560377.0,
      "step": 13660
    },
    {
      "entropy": 6.384401130676269,
      "epoch": 1.4623040291080316,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004794808218445136,
      "loss": 5.8754,
      "mean_token_accuracy": 0.16511950194835662,
      "num_tokens": 29571378.0,
      "step": 13665
    },
    {
      "entropy": 6.482756471633911,
      "epoch": 1.4628391032157955,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047946486701479576,
      "loss": 6.0302,
      "mean_token_accuracy": 0.15018941834568977,
      "num_tokens": 29582187.0,
      "step": 13670
    },
    {
      "entropy": 6.3954747200012205,
      "epoch": 1.4633741773235593,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004794489062811835,
      "loss": 6.0233,
      "mean_token_accuracy": 0.14970337450504304,
      "num_tokens": 29592982.0,
      "step": 13675
    },
    {
      "entropy": 6.494070911407471,
      "epoch": 1.4639092514313232,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004794329396441378,
      "loss": 5.9979,
      "mean_token_accuracy": 0.15004080086946486,
      "num_tokens": 29603639.0,
      "step": 13680
    },
    {
      "entropy": 6.466432189941406,
      "epoch": 1.464444325539087,
      "grad_norm": 1.0,
      "learning_rate": 0.00047941696710411975,
      "loss": 5.9281,
      "mean_token_accuracy": 0.16575905084609985,
      "num_tokens": 29615103.0,
      "step": 13685
    },
    {
      "entropy": 6.471026468276977,
      "epoch": 1.4649793996468512,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004794009886615904,
      "loss": 6.0125,
      "mean_token_accuracy": 0.1531608931720257,
      "num_tokens": 29626184.0,
      "step": 13690
    },
    {
      "entropy": 6.447490262985229,
      "epoch": 1.465514473754615,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047938500431701135,
      "loss": 5.9906,
      "mean_token_accuracy": 0.16663542985916138,
      "num_tokens": 29637615.0,
      "step": 13695
    },
    {
      "entropy": 6.4909755229949955,
      "epoch": 1.466049547862379,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000479369014070844,
      "loss": 6.0502,
      "mean_token_accuracy": 0.15159168168902398,
      "num_tokens": 29648648.0,
      "step": 13700
    },
    {
      "entropy": 6.481311368942261,
      "epoch": 1.4665846219701428,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047935301792355003,
      "loss": 5.9869,
      "mean_token_accuracy": 0.16029854863882065,
      "num_tokens": 29660542.0,
      "step": 13705
    },
    {
      "entropy": 6.459662914276123,
      "epoch": 1.4671196960779067,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004793370158755914,
      "loss": 6.0267,
      "mean_token_accuracy": 0.15619974881410598,
      "num_tokens": 29670342.0,
      "step": 13710
    },
    {
      "entropy": 6.426430034637451,
      "epoch": 1.4676547701856708,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00047932100792743014,
      "loss": 5.9889,
      "mean_token_accuracy": 0.16116821020841599,
      "num_tokens": 29681191.0,
      "step": 13715
    },
    {
      "entropy": 6.4639264106750485,
      "epoch": 1.4681898442934347,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047930499407952855,
      "loss": 5.9252,
      "mean_token_accuracy": 0.16342314779758454,
      "num_tokens": 29691299.0,
      "step": 13720
    },
    {
      "entropy": 6.388714218139649,
      "epoch": 1.4687249184011986,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047928897433234893,
      "loss": 5.9883,
      "mean_token_accuracy": 0.15282868593931198,
      "num_tokens": 29702125.0,
      "step": 13725
    },
    {
      "entropy": 6.434034204483032,
      "epoch": 1.4692599925089624,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004792729486863539,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15368908494710923,
      "num_tokens": 29712679.0,
      "step": 13730
    },
    {
      "entropy": 6.4496959209442135,
      "epoch": 1.4697950666167263,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004792569171420061,
      "loss": 5.9402,
      "mean_token_accuracy": 0.16062451153993607,
      "num_tokens": 29723833.0,
      "step": 13735
    },
    {
      "entropy": 6.490697526931763,
      "epoch": 1.4703301407244904,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004792408796997687,
      "loss": 6.0508,
      "mean_token_accuracy": 0.15469037741422653,
      "num_tokens": 29734646.0,
      "step": 13740
    },
    {
      "entropy": 6.468696212768554,
      "epoch": 1.4708652148322543,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004792248363601044,
      "loss": 6.002,
      "mean_token_accuracy": 0.15406298786401748,
      "num_tokens": 29745129.0,
      "step": 13745
    },
    {
      "entropy": 6.486354684829712,
      "epoch": 1.4714002889400182,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004792087871234767,
      "loss": 6.0099,
      "mean_token_accuracy": 0.15309463888406755,
      "num_tokens": 29756308.0,
      "step": 13750
    },
    {
      "entropy": 6.44014835357666,
      "epoch": 1.471935363047782,
      "grad_norm": 1.015625,
      "learning_rate": 0.000479192731990349,
      "loss": 6.0019,
      "mean_token_accuracy": 0.1532890632748604,
      "num_tokens": 29767835.0,
      "step": 13755
    },
    {
      "entropy": 6.422342443466187,
      "epoch": 1.472470437155546,
      "grad_norm": 1.0,
      "learning_rate": 0.0004791766709611849,
      "loss": 6.0346,
      "mean_token_accuracy": 0.15643304735422134,
      "num_tokens": 29778776.0,
      "step": 13760
    },
    {
      "entropy": 6.514595460891724,
      "epoch": 1.47300551126331,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000479160604036448,
      "loss": 6.0599,
      "mean_token_accuracy": 0.14576017409563063,
      "num_tokens": 29788800.0,
      "step": 13765
    },
    {
      "entropy": 6.429922437667846,
      "epoch": 1.473540585371074,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004791445312166025,
      "loss": 5.9619,
      "mean_token_accuracy": 0.16213579177856446,
      "num_tokens": 29799375.0,
      "step": 13770
    },
    {
      "entropy": 6.494619655609131,
      "epoch": 1.4740756594788378,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0004791284525021122,
      "loss": 6.0758,
      "mean_token_accuracy": 0.14598800390958785,
      "num_tokens": 29811267.0,
      "step": 13775
    },
    {
      "entropy": 6.468775987625122,
      "epoch": 1.4746107335866017,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004791123678934416,
      "loss": 6.0226,
      "mean_token_accuracy": 0.15053389221429825,
      "num_tokens": 29821616.0,
      "step": 13780
    },
    {
      "entropy": 6.441968727111816,
      "epoch": 1.4751458076943655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004790962773910551,
      "loss": 6.0073,
      "mean_token_accuracy": 0.15939292162656785,
      "num_tokens": 29832233.0,
      "step": 13785
    },
    {
      "entropy": 6.446916103363037,
      "epoch": 1.4756808818021296,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047908018099541717,
      "loss": 5.977,
      "mean_token_accuracy": 0.155632121860981,
      "num_tokens": 29843465.0,
      "step": 13790
    },
    {
      "entropy": 6.437223052978515,
      "epoch": 1.4762159559098935,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004790640787069927,
      "loss": 5.8603,
      "mean_token_accuracy": 0.1654409348964691,
      "num_tokens": 29854123.0,
      "step": 13795
    },
    {
      "entropy": 6.468881750106812,
      "epoch": 1.4767510300176574,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004790479705262467,
      "loss": 6.0735,
      "mean_token_accuracy": 0.14499761611223222,
      "num_tokens": 29864831.0,
      "step": 13800
    },
    {
      "entropy": 6.385055828094482,
      "epoch": 1.4772861041254215,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004790318564536441,
      "loss": 5.9633,
      "mean_token_accuracy": 0.15197936296463013,
      "num_tokens": 29876273.0,
      "step": 13805
    },
    {
      "entropy": 6.42480731010437,
      "epoch": 1.4778211782331854,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047901573648965047,
      "loss": 5.9518,
      "mean_token_accuracy": 0.15442708283662795,
      "num_tokens": 29886560.0,
      "step": 13810
    },
    {
      "entropy": 6.425076627731324,
      "epoch": 1.4783562523409493,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000478999610634731,
      "loss": 5.9689,
      "mean_token_accuracy": 0.15788633674383162,
      "num_tokens": 29896916.0,
      "step": 13815
    },
    {
      "entropy": 6.422893810272217,
      "epoch": 1.4788913264487131,
      "grad_norm": 1.0,
      "learning_rate": 0.0004789834788893515,
      "loss": 6.0369,
      "mean_token_accuracy": 0.15061721354722976,
      "num_tokens": 29907288.0,
      "step": 13820
    },
    {
      "entropy": 6.408447027206421,
      "epoch": 1.479426400556477,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004789673412539777,
      "loss": 5.957,
      "mean_token_accuracy": 0.15547766983509065,
      "num_tokens": 29918194.0,
      "step": 13825
    },
    {
      "entropy": 6.442624711990357,
      "epoch": 1.479961474664241,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047895119772907566,
      "loss": 5.9865,
      "mean_token_accuracy": 0.15572214871644974,
      "num_tokens": 29928521.0,
      "step": 13830
    },
    {
      "entropy": 6.3985248565673825,
      "epoch": 1.480496548772005,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004789350483151113,
      "loss": 5.9536,
      "mean_token_accuracy": 0.1536981388926506,
      "num_tokens": 29938885.0,
      "step": 13835
    },
    {
      "entropy": 6.4583570003509525,
      "epoch": 1.4810316228797689,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004789188930125512,
      "loss": 6.0046,
      "mean_token_accuracy": 0.16502514928579332,
      "num_tokens": 29948921.0,
      "step": 13840
    },
    {
      "entropy": 6.427733993530273,
      "epoch": 1.4815666969875327,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004789027318218616,
      "loss": 5.9426,
      "mean_token_accuracy": 0.1591596856713295,
      "num_tokens": 29959633.0,
      "step": 13845
    },
    {
      "entropy": 6.386487531661987,
      "epoch": 1.4821017710952966,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004788865647435093,
      "loss": 5.9703,
      "mean_token_accuracy": 0.1516585133969784,
      "num_tokens": 29970639.0,
      "step": 13850
    },
    {
      "entropy": 6.42505111694336,
      "epoch": 1.4826368452030607,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004788703917779612,
      "loss": 6.0666,
      "mean_token_accuracy": 0.15420404821634293,
      "num_tokens": 29981752.0,
      "step": 13855
    },
    {
      "entropy": 6.4827244758605955,
      "epoch": 1.4831719193108246,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004788542129256841,
      "loss": 5.9163,
      "mean_token_accuracy": 0.1600962460041046,
      "num_tokens": 29991037.0,
      "step": 13860
    },
    {
      "entropy": 6.4501340866088865,
      "epoch": 1.4837069934185885,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047883802818714526,
      "loss": 6.0204,
      "mean_token_accuracy": 0.14502203986048698,
      "num_tokens": 30001980.0,
      "step": 13865
    },
    {
      "entropy": 6.436656284332275,
      "epoch": 1.4842420675263523,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000478821837562812,
      "loss": 6.0049,
      "mean_token_accuracy": 0.15540240406990052,
      "num_tokens": 30013377.0,
      "step": 13870
    },
    {
      "entropy": 6.458589363098144,
      "epoch": 1.4847771416341162,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004788056410531518,
      "loss": 6.0724,
      "mean_token_accuracy": 0.15364046692848204,
      "num_tokens": 30023909.0,
      "step": 13875
    },
    {
      "entropy": 6.393020725250244,
      "epoch": 1.4853122157418803,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004787894386586324,
      "loss": 5.9391,
      "mean_token_accuracy": 0.15834247469902038,
      "num_tokens": 30034069.0,
      "step": 13880
    },
    {
      "entropy": 6.388837623596191,
      "epoch": 1.4858472898496442,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047877323037972153,
      "loss": 5.9494,
      "mean_token_accuracy": 0.1644696205854416,
      "num_tokens": 30043940.0,
      "step": 13885
    },
    {
      "entropy": 6.35999493598938,
      "epoch": 1.486382363957408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004787570162168874,
      "loss": 5.9193,
      "mean_token_accuracy": 0.16718800216913224,
      "num_tokens": 30054985.0,
      "step": 13890
    },
    {
      "entropy": 6.442635536193848,
      "epoch": 1.486917438065172,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047874079617059783,
      "loss": 5.9689,
      "mean_token_accuracy": 0.15539864748716353,
      "num_tokens": 30065675.0,
      "step": 13895
    },
    {
      "entropy": 6.468201351165772,
      "epoch": 1.4874525121729358,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004787245702413216,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15834919065237046,
      "num_tokens": 30075305.0,
      "step": 13900
    },
    {
      "entropy": 6.438325071334839,
      "epoch": 1.4879875862807,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047870833842952697,
      "loss": 6.014,
      "mean_token_accuracy": 0.15502126067876815,
      "num_tokens": 30087309.0,
      "step": 13905
    },
    {
      "entropy": 6.388030624389648,
      "epoch": 1.4885226603884638,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004786921007356827,
      "loss": 5.9316,
      "mean_token_accuracy": 0.15282392650842666,
      "num_tokens": 30097262.0,
      "step": 13910
    },
    {
      "entropy": 6.446554899215698,
      "epoch": 1.4890577344962277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047867585716025767,
      "loss": 6.0317,
      "mean_token_accuracy": 0.15957216620445253,
      "num_tokens": 30107067.0,
      "step": 13915
    },
    {
      "entropy": 6.461298131942749,
      "epoch": 1.4895928086039918,
      "grad_norm": 1.0,
      "learning_rate": 0.0004786596077037209,
      "loss": 6.0376,
      "mean_token_accuracy": 0.1490042671561241,
      "num_tokens": 30117115.0,
      "step": 13920
    },
    {
      "entropy": 6.4253089904785154,
      "epoch": 1.4901278827117554,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047864335236654144,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15946621894836427,
      "num_tokens": 30127202.0,
      "step": 13925
    },
    {
      "entropy": 6.346633338928223,
      "epoch": 1.4906629568195195,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004786270911491889,
      "loss": 5.9577,
      "mean_token_accuracy": 0.15520149618387222,
      "num_tokens": 30138608.0,
      "step": 13930
    },
    {
      "entropy": 6.514653110504151,
      "epoch": 1.4911980309272834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004786108240521327,
      "loss": 6.0182,
      "mean_token_accuracy": 0.15999771431088447,
      "num_tokens": 30148500.0,
      "step": 13935
    },
    {
      "entropy": 6.420861768722534,
      "epoch": 1.4917331050350473,
      "grad_norm": 1.0,
      "learning_rate": 0.00047859455107584256,
      "loss": 5.9435,
      "mean_token_accuracy": 0.1613678514957428,
      "num_tokens": 30159062.0,
      "step": 13940
    },
    {
      "entropy": 6.373835659027099,
      "epoch": 1.4922681791428114,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047857827222078835,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15471021980047225,
      "num_tokens": 30170468.0,
      "step": 13945
    },
    {
      "entropy": 6.3641805171966555,
      "epoch": 1.4928032532505753,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004785619874874402,
      "loss": 5.889,
      "mean_token_accuracy": 0.16263294070959092,
      "num_tokens": 30180944.0,
      "step": 13950
    },
    {
      "entropy": 6.378198575973511,
      "epoch": 1.4933383273583392,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004785456968762682,
      "loss": 5.8975,
      "mean_token_accuracy": 0.16140799969434738,
      "num_tokens": 30191614.0,
      "step": 13955
    },
    {
      "entropy": 6.380344867706299,
      "epoch": 1.493873401466103,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047852940038774276,
      "loss": 5.9163,
      "mean_token_accuracy": 0.1616358920931816,
      "num_tokens": 30202035.0,
      "step": 13960
    },
    {
      "entropy": 6.386508750915527,
      "epoch": 1.494408475573867,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047851309802233457,
      "loss": 6.0137,
      "mean_token_accuracy": 0.15903844982385634,
      "num_tokens": 30213205.0,
      "step": 13965
    },
    {
      "entropy": 6.475919485092163,
      "epoch": 1.494943549681631,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004784967897805143,
      "loss": 6.0151,
      "mean_token_accuracy": 0.15711776167154312,
      "num_tokens": 30223971.0,
      "step": 13970
    },
    {
      "entropy": 6.43127293586731,
      "epoch": 1.495478623789395,
      "grad_norm": 0.8671875,
      "learning_rate": 0.0004784804756627528,
      "loss": 6.0237,
      "mean_token_accuracy": 0.15176371559500695,
      "num_tokens": 30235346.0,
      "step": 13975
    },
    {
      "entropy": 6.405437374114991,
      "epoch": 1.4960136978971588,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047846415566952106,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15892144292593002,
      "num_tokens": 30246033.0,
      "step": 13980
    },
    {
      "entropy": 6.425986862182617,
      "epoch": 1.4965487720049226,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047844782980129057,
      "loss": 6.0147,
      "mean_token_accuracy": 0.15101836621761322,
      "num_tokens": 30258208.0,
      "step": 13985
    },
    {
      "entropy": 6.450596857070923,
      "epoch": 1.4970838461126865,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047843149805853254,
      "loss": 6.0218,
      "mean_token_accuracy": 0.15674971044063568,
      "num_tokens": 30268824.0,
      "step": 13990
    },
    {
      "entropy": 6.4718118667602536,
      "epoch": 1.4976189202204506,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004784151604417185,
      "loss": 6.043,
      "mean_token_accuracy": 0.14687362760305406,
      "num_tokens": 30280078.0,
      "step": 13995
    },
    {
      "entropy": 6.392143440246582,
      "epoch": 1.4981539943282145,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047839881695132034,
      "loss": 5.9479,
      "mean_token_accuracy": 0.16477046459913253,
      "num_tokens": 30290939.0,
      "step": 14000
    },
    {
      "entropy": 6.444397258758545,
      "epoch": 1.4986890684359784,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047838246758780996,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15928044766187668,
      "num_tokens": 30301198.0,
      "step": 14005
    },
    {
      "entropy": 6.412237596511841,
      "epoch": 1.4992241425437423,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004783661123516593,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1586026057600975,
      "num_tokens": 30312021.0,
      "step": 14010
    },
    {
      "entropy": 6.456020164489746,
      "epoch": 1.4997592166515061,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004783497512433408,
      "loss": 5.965,
      "mean_token_accuracy": 0.1588841274380684,
      "num_tokens": 30323335.0,
      "step": 14015
    },
    {
      "entropy": 6.4273943424224855,
      "epoch": 1.5002942907592702,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004783333842633268,
      "loss": 5.9807,
      "mean_token_accuracy": 0.1590695708990097,
      "num_tokens": 30334389.0,
      "step": 14020
    },
    {
      "entropy": 6.3824134349823,
      "epoch": 1.500829364867034,
      "grad_norm": 1.0,
      "learning_rate": 0.00047831701141208994,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14907802939414977,
      "num_tokens": 30345407.0,
      "step": 14025
    },
    {
      "entropy": 6.439800357818603,
      "epoch": 1.501364438974798,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047830063269010286,
      "loss": 5.9803,
      "mean_token_accuracy": 0.15205998718738556,
      "num_tokens": 30356718.0,
      "step": 14030
    },
    {
      "entropy": 6.408035945892334,
      "epoch": 1.501899513082562,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047828424809783865,
      "loss": 5.9169,
      "mean_token_accuracy": 0.15723048001527787,
      "num_tokens": 30367617.0,
      "step": 14035
    },
    {
      "entropy": 6.428130149841309,
      "epoch": 1.5024345871903257,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004782678576357703,
      "loss": 5.991,
      "mean_token_accuracy": 0.1570350393652916,
      "num_tokens": 30377923.0,
      "step": 14040
    },
    {
      "entropy": 6.433382511138916,
      "epoch": 1.5029696612980898,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004782514613043711,
      "loss": 5.9706,
      "mean_token_accuracy": 0.16139114052057266,
      "num_tokens": 30388719.0,
      "step": 14045
    },
    {
      "entropy": 6.415397882461548,
      "epoch": 1.5035047354058537,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047823505910411465,
      "loss": 5.9194,
      "mean_token_accuracy": 0.1591455489397049,
      "num_tokens": 30400191.0,
      "step": 14050
    },
    {
      "entropy": 6.423733806610107,
      "epoch": 1.5040398095136176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047821865103547437,
      "loss": 5.9953,
      "mean_token_accuracy": 0.1548720195889473,
      "num_tokens": 30410267.0,
      "step": 14055
    },
    {
      "entropy": 6.431436395645141,
      "epoch": 1.5045748836213817,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000478202237098924,
      "loss": 6.0142,
      "mean_token_accuracy": 0.1523920089006424,
      "num_tokens": 30420852.0,
      "step": 14060
    },
    {
      "entropy": 6.455730676651001,
      "epoch": 1.5051099577291454,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004781858172949376,
      "loss": 6.0464,
      "mean_token_accuracy": 0.15612104162573814,
      "num_tokens": 30432214.0,
      "step": 14065
    },
    {
      "entropy": 6.508595752716064,
      "epoch": 1.5056450318369095,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004781693916239894,
      "loss": 6.06,
      "mean_token_accuracy": 0.14718416035175325,
      "num_tokens": 30443919.0,
      "step": 14070
    },
    {
      "entropy": 6.502174711227417,
      "epoch": 1.5061801059446733,
      "grad_norm": 1.0,
      "learning_rate": 0.00047815296008655347,
      "loss": 6.1191,
      "mean_token_accuracy": 0.1441225826740265,
      "num_tokens": 30453930.0,
      "step": 14075
    },
    {
      "entropy": 6.4613954544067385,
      "epoch": 1.5067151800524372,
      "grad_norm": 0.8828125,
      "learning_rate": 0.00047813652268310436,
      "loss": 6.0061,
      "mean_token_accuracy": 0.15176257640123367,
      "num_tokens": 30464612.0,
      "step": 14080
    },
    {
      "entropy": 6.434748744964599,
      "epoch": 1.5072502541602013,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047812007941411673,
      "loss": 5.9487,
      "mean_token_accuracy": 0.15387781262397765,
      "num_tokens": 30474819.0,
      "step": 14085
    },
    {
      "entropy": 6.402099514007569,
      "epoch": 1.507785328267965,
      "grad_norm": 0.8671875,
      "learning_rate": 0.00047810363028006535,
      "loss": 5.946,
      "mean_token_accuracy": 0.16535178422927857,
      "num_tokens": 30485481.0,
      "step": 14090
    },
    {
      "entropy": 6.387093353271484,
      "epoch": 1.508320402375729,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004780871752814252,
      "loss": 5.9657,
      "mean_token_accuracy": 0.1639707699418068,
      "num_tokens": 30496321.0,
      "step": 14095
    },
    {
      "entropy": 6.4116963863372805,
      "epoch": 1.508855476483493,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004780707144186714,
      "loss": 6.0795,
      "mean_token_accuracy": 0.14904560521245003,
      "num_tokens": 30507578.0,
      "step": 14100
    },
    {
      "entropy": 6.485542249679566,
      "epoch": 1.5093905505912568,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047805424769227923,
      "loss": 6.0045,
      "mean_token_accuracy": 0.1545480564236641,
      "num_tokens": 30518328.0,
      "step": 14105
    },
    {
      "entropy": 6.4175670623779295,
      "epoch": 1.509925624699021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047803777510272425,
      "loss": 5.9533,
      "mean_token_accuracy": 0.15541483759880065,
      "num_tokens": 30528555.0,
      "step": 14110
    },
    {
      "entropy": 6.364946508407593,
      "epoch": 1.5104606988067846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000478021296650482,
      "loss": 5.8973,
      "mean_token_accuracy": 0.1622355341911316,
      "num_tokens": 30540217.0,
      "step": 14115
    },
    {
      "entropy": 6.492854070663452,
      "epoch": 1.5109957729145487,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004780048123360283,
      "loss": 6.0338,
      "mean_token_accuracy": 0.1586015999317169,
      "num_tokens": 30550937.0,
      "step": 14120
    },
    {
      "entropy": 6.533609771728516,
      "epoch": 1.5115308470223126,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047798832215983916,
      "loss": 6.0211,
      "mean_token_accuracy": 0.15706411749124527,
      "num_tokens": 30562248.0,
      "step": 14125
    },
    {
      "entropy": 6.493888998031617,
      "epoch": 1.5120659211300764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004779718261223908,
      "loss": 6.0077,
      "mean_token_accuracy": 0.16029556095600128,
      "num_tokens": 30573362.0,
      "step": 14130
    },
    {
      "entropy": 6.447023344039917,
      "epoch": 1.5126009952378405,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047795532422415947,
      "loss": 5.9961,
      "mean_token_accuracy": 0.15161878913640975,
      "num_tokens": 30584052.0,
      "step": 14135
    },
    {
      "entropy": 6.438470792770386,
      "epoch": 1.5131360693456044,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004779388164656217,
      "loss": 6.0273,
      "mean_token_accuracy": 0.15476970076560975,
      "num_tokens": 30594630.0,
      "step": 14140
    },
    {
      "entropy": 6.408714246749878,
      "epoch": 1.5136711434533683,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047792230284725415,
      "loss": 5.9962,
      "mean_token_accuracy": 0.15219680964946747,
      "num_tokens": 30605625.0,
      "step": 14145
    },
    {
      "entropy": 6.437816572189331,
      "epoch": 1.5142062175611322,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047790578336953356,
      "loss": 5.964,
      "mean_token_accuracy": 0.15795817077159882,
      "num_tokens": 30616520.0,
      "step": 14150
    },
    {
      "entropy": 6.430609941482544,
      "epoch": 1.514741291668896,
      "grad_norm": 0.93359375,
      "learning_rate": 0.000477889258032937,
      "loss": 5.9467,
      "mean_token_accuracy": 0.1636806271970272,
      "num_tokens": 30627372.0,
      "step": 14155
    },
    {
      "entropy": 6.4336189270019535,
      "epoch": 1.5152763657766601,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047787272683794157,
      "loss": 5.9434,
      "mean_token_accuracy": 0.16420188844203948,
      "num_tokens": 30636974.0,
      "step": 14160
    },
    {
      "entropy": 6.425327968597412,
      "epoch": 1.515811439884424,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004778561897850247,
      "loss": 5.9934,
      "mean_token_accuracy": 0.1616658478975296,
      "num_tokens": 30648202.0,
      "step": 14165
    },
    {
      "entropy": 6.421125411987305,
      "epoch": 1.516346513992188,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047783964687466384,
      "loss": 6.0448,
      "mean_token_accuracy": 0.15415506660938263,
      "num_tokens": 30657981.0,
      "step": 14170
    },
    {
      "entropy": 6.45186128616333,
      "epoch": 1.516881588099952,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047782309810733676,
      "loss": 6.0123,
      "mean_token_accuracy": 0.14989591166377067,
      "num_tokens": 30668496.0,
      "step": 14175
    },
    {
      "entropy": 6.3835649490356445,
      "epoch": 1.5174166622077156,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047780654348352115,
      "loss": 5.9059,
      "mean_token_accuracy": 0.16182158291339874,
      "num_tokens": 30678868.0,
      "step": 14180
    },
    {
      "entropy": 6.4336316108703615,
      "epoch": 1.5179517363154797,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047778998300369516,
      "loss": 6.0449,
      "mean_token_accuracy": 0.15890242010354996,
      "num_tokens": 30690066.0,
      "step": 14185
    },
    {
      "entropy": 6.4090416431427,
      "epoch": 1.5184868104232436,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004777734166683368,
      "loss": 5.9347,
      "mean_token_accuracy": 0.15853980630636216,
      "num_tokens": 30700624.0,
      "step": 14190
    },
    {
      "entropy": 6.391382360458374,
      "epoch": 1.5190218845310075,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047775684447792457,
      "loss": 5.9469,
      "mean_token_accuracy": 0.16479996144771575,
      "num_tokens": 30711183.0,
      "step": 14195
    },
    {
      "entropy": 6.416874361038208,
      "epoch": 1.5195569586387716,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000477740266432937,
      "loss": 5.972,
      "mean_token_accuracy": 0.16482598185539246,
      "num_tokens": 30721446.0,
      "step": 14200
    },
    {
      "entropy": 6.483606958389283,
      "epoch": 1.5200920327465353,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047772368253385267,
      "loss": 6.0123,
      "mean_token_accuracy": 0.14917905628681183,
      "num_tokens": 30731901.0,
      "step": 14205
    },
    {
      "entropy": 6.47951192855835,
      "epoch": 1.5206271068542994,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004777070927811505,
      "loss": 6.0027,
      "mean_token_accuracy": 0.15324360877275467,
      "num_tokens": 30741701.0,
      "step": 14210
    },
    {
      "entropy": 6.4225176811218265,
      "epoch": 1.5211621809620632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047769049717530956,
      "loss": 5.9892,
      "mean_token_accuracy": 0.15065563470125198,
      "num_tokens": 30751594.0,
      "step": 14215
    },
    {
      "entropy": 6.4378081321716305,
      "epoch": 1.5216972550698271,
      "grad_norm": 0.90625,
      "learning_rate": 0.00047767389571680896,
      "loss": 6.0517,
      "mean_token_accuracy": 0.14456998705863952,
      "num_tokens": 30764418.0,
      "step": 14220
    },
    {
      "entropy": 6.489158010482788,
      "epoch": 1.5222323291775912,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004776572884061281,
      "loss": 5.9787,
      "mean_token_accuracy": 0.15952815860509872,
      "num_tokens": 30773976.0,
      "step": 14225
    },
    {
      "entropy": 6.461334657669068,
      "epoch": 1.5227674032853549,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004776406752437465,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15898415297269822,
      "num_tokens": 30784210.0,
      "step": 14230
    },
    {
      "entropy": 6.433219623565674,
      "epoch": 1.523302477393119,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004776240562301438,
      "loss": 6.0584,
      "mean_token_accuracy": 0.14807326942682267,
      "num_tokens": 30794322.0,
      "step": 14235
    },
    {
      "entropy": 6.471520280838012,
      "epoch": 1.5238375515008828,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004776074313658001,
      "loss": 6.0008,
      "mean_token_accuracy": 0.15867502838373185,
      "num_tokens": 30805264.0,
      "step": 14240
    },
    {
      "entropy": 6.388085842132568,
      "epoch": 1.5243726256086467,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004775908006511952,
      "loss": 5.9816,
      "mean_token_accuracy": 0.1524159222841263,
      "num_tokens": 30815909.0,
      "step": 14245
    },
    {
      "entropy": 6.482439851760864,
      "epoch": 1.5249076997164108,
      "grad_norm": 1.0,
      "learning_rate": 0.0004775741640868095,
      "loss": 5.9831,
      "mean_token_accuracy": 0.16120698153972626,
      "num_tokens": 30826867.0,
      "step": 14250
    },
    {
      "entropy": 6.501117706298828,
      "epoch": 1.5254427738241747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047755752167312314,
      "loss": 5.9372,
      "mean_token_accuracy": 0.15576794147491455,
      "num_tokens": 30837696.0,
      "step": 14255
    },
    {
      "entropy": 6.403955459594727,
      "epoch": 1.5259778479319386,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004775408734106169,
      "loss": 5.974,
      "mean_token_accuracy": 0.15726611465215684,
      "num_tokens": 30848155.0,
      "step": 14260
    },
    {
      "entropy": 6.377830696105957,
      "epoch": 1.5265129220397025,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047752421929977143,
      "loss": 5.8692,
      "mean_token_accuracy": 0.16495280116796493,
      "num_tokens": 30857804.0,
      "step": 14265
    },
    {
      "entropy": 6.460524702072144,
      "epoch": 1.5270479961474663,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004775075593410675,
      "loss": 6.016,
      "mean_token_accuracy": 0.15396036505699157,
      "num_tokens": 30867290.0,
      "step": 14270
    },
    {
      "entropy": 6.454971218109131,
      "epoch": 1.5275830702552304,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047749089353498625,
      "loss": 6.0783,
      "mean_token_accuracy": 0.15616631954908372,
      "num_tokens": 30877634.0,
      "step": 14275
    },
    {
      "entropy": 6.456519031524659,
      "epoch": 1.5281181443629943,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000477474221882009,
      "loss": 6.0429,
      "mean_token_accuracy": 0.15429300293326378,
      "num_tokens": 30889176.0,
      "step": 14280
    },
    {
      "entropy": 6.445395851135254,
      "epoch": 1.5286532184707582,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004774575443826169,
      "loss": 5.9486,
      "mean_token_accuracy": 0.15877715200185777,
      "num_tokens": 30899516.0,
      "step": 14285
    },
    {
      "entropy": 6.460542964935303,
      "epoch": 1.5291882925785223,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004774408610372918,
      "loss": 6.0255,
      "mean_token_accuracy": 0.1542531818151474,
      "num_tokens": 30910155.0,
      "step": 14290
    },
    {
      "entropy": 6.490932846069336,
      "epoch": 1.529723366686286,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004774241718465152,
      "loss": 6.0542,
      "mean_token_accuracy": 0.15218367129564286,
      "num_tokens": 30921486.0,
      "step": 14295
    },
    {
      "entropy": 6.462147617340088,
      "epoch": 1.53025844079405,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004774074768107691,
      "loss": 5.9744,
      "mean_token_accuracy": 0.15398589372634888,
      "num_tokens": 30932361.0,
      "step": 14300
    },
    {
      "entropy": 6.4190033912658695,
      "epoch": 1.530793514901814,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047739077593053554,
      "loss": 5.9546,
      "mean_token_accuracy": 0.15889570564031602,
      "num_tokens": 30943033.0,
      "step": 14305
    },
    {
      "entropy": 6.440966463088989,
      "epoch": 1.5313285890095778,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047737406920629675,
      "loss": 5.9768,
      "mean_token_accuracy": 0.16125437319278718,
      "num_tokens": 30953094.0,
      "step": 14310
    },
    {
      "entropy": 6.468457126617432,
      "epoch": 1.531863663117342,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047735735663853514,
      "loss": 5.9881,
      "mean_token_accuracy": 0.15418486446142196,
      "num_tokens": 30964604.0,
      "step": 14315
    },
    {
      "entropy": 6.438661670684814,
      "epoch": 1.5323987372251056,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004773406382277334,
      "loss": 6.0698,
      "mean_token_accuracy": 0.15072232633829116,
      "num_tokens": 30975681.0,
      "step": 14320
    },
    {
      "entropy": 6.498667287826538,
      "epoch": 1.5329338113328697,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004773239139743741,
      "loss": 6.0215,
      "mean_token_accuracy": 0.15297286212444305,
      "num_tokens": 30986319.0,
      "step": 14325
    },
    {
      "entropy": 6.482737493515015,
      "epoch": 1.5334688854406335,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004773071838789402,
      "loss": 5.9833,
      "mean_token_accuracy": 0.15202575027942658,
      "num_tokens": 30997077.0,
      "step": 14330
    },
    {
      "entropy": 6.380411624908447,
      "epoch": 1.5340039595483974,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004772904479419148,
      "loss": 5.8987,
      "mean_token_accuracy": 0.16325486004352568,
      "num_tokens": 31007485.0,
      "step": 14335
    },
    {
      "entropy": 6.420392751693726,
      "epoch": 1.5345390336561615,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004772737061637811,
      "loss": 5.9751,
      "mean_token_accuracy": 0.16343308240175247,
      "num_tokens": 31017827.0,
      "step": 14340
    },
    {
      "entropy": 6.459635210037232,
      "epoch": 1.5350741077639252,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047725695854502267,
      "loss": 6.0238,
      "mean_token_accuracy": 0.15414443165063857,
      "num_tokens": 31028167.0,
      "step": 14345
    },
    {
      "entropy": 6.412749767303467,
      "epoch": 1.5356091818716893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047724020508612287,
      "loss": 5.9648,
      "mean_token_accuracy": 0.15934168994426728,
      "num_tokens": 31038941.0,
      "step": 14350
    },
    {
      "entropy": 6.420382928848267,
      "epoch": 1.5361442559794531,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047722344578756564,
      "loss": 5.9667,
      "mean_token_accuracy": 0.15343968272209169,
      "num_tokens": 31049949.0,
      "step": 14355
    },
    {
      "entropy": 6.478651857376098,
      "epoch": 1.536679330087217,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047720668064983476,
      "loss": 5.968,
      "mean_token_accuracy": 0.15733518451452255,
      "num_tokens": 31059299.0,
      "step": 14360
    },
    {
      "entropy": 6.432805013656616,
      "epoch": 1.5372144041949811,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004771899096734145,
      "loss": 6.0795,
      "mean_token_accuracy": 0.14781002700328827,
      "num_tokens": 31070508.0,
      "step": 14365
    },
    {
      "entropy": 6.371538496017456,
      "epoch": 1.5377494783027448,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047717313285878886,
      "loss": 5.954,
      "mean_token_accuracy": 0.15986112356185914,
      "num_tokens": 31082354.0,
      "step": 14370
    },
    {
      "entropy": 6.633033609390258,
      "epoch": 1.5382845524105089,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047715635020644247,
      "loss": 6.0862,
      "mean_token_accuracy": 0.1477938786149025,
      "num_tokens": 31093496.0,
      "step": 14375
    },
    {
      "entropy": 6.46688666343689,
      "epoch": 1.5388196265182728,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047713956171685994,
      "loss": 5.9539,
      "mean_token_accuracy": 0.1581593170762062,
      "num_tokens": 31104289.0,
      "step": 14380
    },
    {
      "entropy": 6.345303869247436,
      "epoch": 1.5393547006260366,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000477122767390526,
      "loss": 5.916,
      "mean_token_accuracy": 0.1693064123392105,
      "num_tokens": 31114998.0,
      "step": 14385
    },
    {
      "entropy": 6.425399589538574,
      "epoch": 1.5398897747338007,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047710596722792543,
      "loss": 5.952,
      "mean_token_accuracy": 0.15377293303608894,
      "num_tokens": 31127393.0,
      "step": 14390
    },
    {
      "entropy": 6.364899206161499,
      "epoch": 1.5404248488415646,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770891612295435,
      "loss": 5.8076,
      "mean_token_accuracy": 0.17347481995820999,
      "num_tokens": 31137484.0,
      "step": 14395
    },
    {
      "entropy": 6.444191026687622,
      "epoch": 1.5409599229493285,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004770723493958654,
      "loss": 6.056,
      "mean_token_accuracy": 0.15321120917797088,
      "num_tokens": 31147863.0,
      "step": 14400
    },
    {
      "entropy": 6.412865924835205,
      "epoch": 1.5414949970570924,
      "grad_norm": 1.0,
      "learning_rate": 0.0004770555317273767,
      "loss": 5.9138,
      "mean_token_accuracy": 0.16136762350797654,
      "num_tokens": 31159018.0,
      "step": 14405
    },
    {
      "entropy": 6.455521106719971,
      "epoch": 1.5420300711648562,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047703870822456287,
      "loss": 5.9282,
      "mean_token_accuracy": 0.1573185384273529,
      "num_tokens": 31170673.0,
      "step": 14410
    },
    {
      "entropy": 6.319633340835571,
      "epoch": 1.5425651452726203,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004770218788879097,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16714126020669937,
      "num_tokens": 31181165.0,
      "step": 14415
    },
    {
      "entropy": 6.376730489730835,
      "epoch": 1.5431002193803842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004770050437179033,
      "loss": 5.9521,
      "mean_token_accuracy": 0.16174726486206054,
      "num_tokens": 31192976.0,
      "step": 14420
    },
    {
      "entropy": 6.523973751068115,
      "epoch": 1.543635293488148,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004769882027150295,
      "loss": 6.0303,
      "mean_token_accuracy": 0.1576248213648796,
      "num_tokens": 31203979.0,
      "step": 14425
    },
    {
      "entropy": 6.450204610824585,
      "epoch": 1.5441703675959122,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047697135587977473,
      "loss": 5.8973,
      "mean_token_accuracy": 0.16452319025993348,
      "num_tokens": 31214972.0,
      "step": 14430
    },
    {
      "entropy": 6.496889257431031,
      "epoch": 1.5447054417036759,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047695450321262546,
      "loss": 6.0497,
      "mean_token_accuracy": 0.154098679125309,
      "num_tokens": 31224726.0,
      "step": 14435
    },
    {
      "entropy": 6.442439031600952,
      "epoch": 1.54524051581144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004769376447140683,
      "loss": 5.9116,
      "mean_token_accuracy": 0.16731070578098298,
      "num_tokens": 31235761.0,
      "step": 14440
    },
    {
      "entropy": 6.430090713500976,
      "epoch": 1.5457755899192038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047692078038459015,
      "loss": 6.0148,
      "mean_token_accuracy": 0.16013336479663848,
      "num_tokens": 31246221.0,
      "step": 14445
    },
    {
      "entropy": 6.454028606414795,
      "epoch": 1.5463106640269677,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004769039102246777,
      "loss": 5.9788,
      "mean_token_accuracy": 0.1530252903699875,
      "num_tokens": 31257364.0,
      "step": 14450
    },
    {
      "entropy": 6.453850269317627,
      "epoch": 1.5468457381347318,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047688703423481827,
      "loss": 5.9856,
      "mean_token_accuracy": 0.15455833077430725,
      "num_tokens": 31268343.0,
      "step": 14455
    },
    {
      "entropy": 6.460255146026611,
      "epoch": 1.5473808122424955,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047687015241549896,
      "loss": 6.0184,
      "mean_token_accuracy": 0.15228765457868576,
      "num_tokens": 31278335.0,
      "step": 14460
    },
    {
      "entropy": 6.435940217971802,
      "epoch": 1.5479158863502596,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004768532647672075,
      "loss": 5.9447,
      "mean_token_accuracy": 0.15146819949150087,
      "num_tokens": 31288537.0,
      "step": 14465
    },
    {
      "entropy": 6.413521766662598,
      "epoch": 1.5484509604580234,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004768363712904314,
      "loss": 5.9616,
      "mean_token_accuracy": 0.15913720801472664,
      "num_tokens": 31299353.0,
      "step": 14470
    },
    {
      "entropy": 6.5064427852630615,
      "epoch": 1.5489860345657873,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004768194719856584,
      "loss": 5.9737,
      "mean_token_accuracy": 0.15467377603054047,
      "num_tokens": 31309655.0,
      "step": 14475
    },
    {
      "entropy": 6.442363166809082,
      "epoch": 1.5495211086735514,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004768025668533765,
      "loss": 6.0261,
      "mean_token_accuracy": 0.16192231923341752,
      "num_tokens": 31321008.0,
      "step": 14480
    },
    {
      "entropy": 6.397524309158325,
      "epoch": 1.550056182781315,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00047678565589407383,
      "loss": 5.9026,
      "mean_token_accuracy": 0.16072767674922944,
      "num_tokens": 31331978.0,
      "step": 14485
    },
    {
      "entropy": 6.4599395275115965,
      "epoch": 1.5505912568890792,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004767687391082387,
      "loss": 5.97,
      "mean_token_accuracy": 0.16145619004964828,
      "num_tokens": 31341703.0,
      "step": 14490
    },
    {
      "entropy": 6.453878021240234,
      "epoch": 1.551126330996843,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047675181649635965,
      "loss": 5.9955,
      "mean_token_accuracy": 0.16121636778116227,
      "num_tokens": 31352787.0,
      "step": 14495
    },
    {
      "entropy": 6.400994873046875,
      "epoch": 1.551661405104607,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047673488805892515,
      "loss": 5.9753,
      "mean_token_accuracy": 0.15595806688070296,
      "num_tokens": 31363702.0,
      "step": 14500
    },
    {
      "entropy": 6.445324659347534,
      "epoch": 1.552196479212371,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047671795379642415,
      "loss": 5.9475,
      "mean_token_accuracy": 0.14955118000507356,
      "num_tokens": 31374719.0,
      "step": 14505
    },
    {
      "entropy": 6.529526948928833,
      "epoch": 1.5527315533201347,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004767010137093456,
      "loss": 6.0703,
      "mean_token_accuracy": 0.14686402678489685,
      "num_tokens": 31386289.0,
      "step": 14510
    },
    {
      "entropy": 6.443260669708252,
      "epoch": 1.5532666274278988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047668406779817864,
      "loss": 5.9096,
      "mean_token_accuracy": 0.15902149379253389,
      "num_tokens": 31397103.0,
      "step": 14515
    },
    {
      "entropy": 6.4135784149169925,
      "epoch": 1.5538017015356627,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004766671160634125,
      "loss": 6.0152,
      "mean_token_accuracy": 0.15964665859937668,
      "num_tokens": 31408715.0,
      "step": 14520
    },
    {
      "entropy": 6.525426244735717,
      "epoch": 1.5543367756434265,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004766501585055367,
      "loss": 6.0287,
      "mean_token_accuracy": 0.1492007330060005,
      "num_tokens": 31420365.0,
      "step": 14525
    },
    {
      "entropy": 6.422760438919068,
      "epoch": 1.5548718497511906,
      "grad_norm": 1.0,
      "learning_rate": 0.000476633195125041,
      "loss": 5.9416,
      "mean_token_accuracy": 0.1561449483036995,
      "num_tokens": 31431635.0,
      "step": 14530
    },
    {
      "entropy": 6.463779020309448,
      "epoch": 1.5554069238589545,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047661622592241507,
      "loss": 5.9528,
      "mean_token_accuracy": 0.1607808694243431,
      "num_tokens": 31442513.0,
      "step": 14535
    },
    {
      "entropy": 6.449018335342407,
      "epoch": 1.5559419979667184,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004765992508981489,
      "loss": 5.9874,
      "mean_token_accuracy": 0.14925160109996796,
      "num_tokens": 31453726.0,
      "step": 14540
    },
    {
      "entropy": 6.432442760467529,
      "epoch": 1.5564770720744823,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004765822700527328,
      "loss": 5.9197,
      "mean_token_accuracy": 0.15924156457185745,
      "num_tokens": 31463980.0,
      "step": 14545
    },
    {
      "entropy": 6.4585768699646,
      "epoch": 1.5570121461822461,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004765652833866569,
      "loss": 6.0755,
      "mean_token_accuracy": 0.1500492051243782,
      "num_tokens": 31474615.0,
      "step": 14550
    },
    {
      "entropy": 6.386797142028809,
      "epoch": 1.5575472202900102,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004765482909004118,
      "loss": 5.9203,
      "mean_token_accuracy": 0.16462710052728652,
      "num_tokens": 31485283.0,
      "step": 14555
    },
    {
      "entropy": 6.460398626327515,
      "epoch": 1.5580822943977741,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004765312925944881,
      "loss": 5.9674,
      "mean_token_accuracy": 0.15228799879550933,
      "num_tokens": 31496997.0,
      "step": 14560
    },
    {
      "entropy": 6.515994262695313,
      "epoch": 1.558617368505538,
      "grad_norm": 0.9375,
      "learning_rate": 0.00047651428846937673,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15816565304994584,
      "num_tokens": 31507585.0,
      "step": 14565
    },
    {
      "entropy": 6.450551891326905,
      "epoch": 1.559152442613302,
      "grad_norm": 1.0,
      "learning_rate": 0.00047649727852556854,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15922337621450425,
      "num_tokens": 31517078.0,
      "step": 14570
    },
    {
      "entropy": 6.364662408828735,
      "epoch": 1.5596875167210658,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004764802627635547,
      "loss": 5.902,
      "mean_token_accuracy": 0.16972979605197908,
      "num_tokens": 31528576.0,
      "step": 14575
    },
    {
      "entropy": 6.499505615234375,
      "epoch": 1.5602225908288299,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004764632411838266,
      "loss": 6.0178,
      "mean_token_accuracy": 0.15273881778120996,
      "num_tokens": 31540095.0,
      "step": 14580
    },
    {
      "entropy": 6.427064418792725,
      "epoch": 1.5607576649365937,
      "grad_norm": 1.0,
      "learning_rate": 0.00047644621378687573,
      "loss": 5.9204,
      "mean_token_accuracy": 0.15550186559557916,
      "num_tokens": 31550980.0,
      "step": 14585
    },
    {
      "entropy": 6.281198263168335,
      "epoch": 1.5612927390443576,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047642918057319377,
      "loss": 5.7224,
      "mean_token_accuracy": 0.18287506997585296,
      "num_tokens": 31561842.0,
      "step": 14590
    },
    {
      "entropy": 6.438292932510376,
      "epoch": 1.5618278131521217,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004764121415432725,
      "loss": 6.0548,
      "mean_token_accuracy": 0.15207030922174453,
      "num_tokens": 31572798.0,
      "step": 14595
    },
    {
      "entropy": 6.4317432880401615,
      "epoch": 1.5623628872598854,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004763950966976039,
      "loss": 6.0408,
      "mean_token_accuracy": 0.15332448482513428,
      "num_tokens": 31584774.0,
      "step": 14600
    },
    {
      "entropy": 6.445364618301392,
      "epoch": 1.5628979613676495,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004763780460366803,
      "loss": 5.9617,
      "mean_token_accuracy": 0.1550541788339615,
      "num_tokens": 31596197.0,
      "step": 14605
    },
    {
      "entropy": 6.480567789077758,
      "epoch": 1.5634330354754133,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004763609895609939,
      "loss": 5.9846,
      "mean_token_accuracy": 0.15567363798618317,
      "num_tokens": 31607455.0,
      "step": 14610
    },
    {
      "entropy": 6.455243062973023,
      "epoch": 1.5639681095831772,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047634392727103714,
      "loss": 6.0086,
      "mean_token_accuracy": 0.1544455111026764,
      "num_tokens": 31618995.0,
      "step": 14615
    },
    {
      "entropy": 6.348547458648682,
      "epoch": 1.5645031836909413,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004763268591673028,
      "loss": 5.8761,
      "mean_token_accuracy": 0.15905967801809312,
      "num_tokens": 31629984.0,
      "step": 14620
    },
    {
      "entropy": 6.4232855319976805,
      "epoch": 1.565038257798705,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004763097852502837,
      "loss": 5.9078,
      "mean_token_accuracy": 0.16182886362075805,
      "num_tokens": 31641362.0,
      "step": 14625
    },
    {
      "entropy": 6.500755548477173,
      "epoch": 1.565573331906469,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047629270552047284,
      "loss": 6.0881,
      "mean_token_accuracy": 0.1471730351448059,
      "num_tokens": 31652650.0,
      "step": 14630
    },
    {
      "entropy": 6.457883691787719,
      "epoch": 1.566108406014233,
      "grad_norm": 1.0,
      "learning_rate": 0.00047627561997836337,
      "loss": 5.9966,
      "mean_token_accuracy": 0.15006966888904572,
      "num_tokens": 31664201.0,
      "step": 14635
    },
    {
      "entropy": 6.463002252578735,
      "epoch": 1.5666434801219968,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004762585286244487,
      "loss": 5.9496,
      "mean_token_accuracy": 0.15809212625026703,
      "num_tokens": 31674480.0,
      "step": 14640
    },
    {
      "entropy": 6.4485033512115475,
      "epoch": 1.567178554229761,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047624143145922227,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15839379280805588,
      "num_tokens": 31685377.0,
      "step": 14645
    },
    {
      "entropy": 6.46656174659729,
      "epoch": 1.5677136283375246,
      "grad_norm": 1.0,
      "learning_rate": 0.00047622432848317775,
      "loss": 5.9477,
      "mean_token_accuracy": 0.1618150517344475,
      "num_tokens": 31695199.0,
      "step": 14650
    },
    {
      "entropy": 6.42689094543457,
      "epoch": 1.5682487024452887,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004762072196968091,
      "loss": 5.9886,
      "mean_token_accuracy": 0.15703034847974778,
      "num_tokens": 31705528.0,
      "step": 14655
    },
    {
      "entropy": 6.317888784408569,
      "epoch": 1.5687837765530526,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004761901051006102,
      "loss": 5.903,
      "mean_token_accuracy": 0.1605489045381546,
      "num_tokens": 31718156.0,
      "step": 14660
    },
    {
      "entropy": 6.489415168762207,
      "epoch": 1.5693188506608164,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047617298469507534,
      "loss": 5.9506,
      "mean_token_accuracy": 0.15866044759750367,
      "num_tokens": 31729783.0,
      "step": 14665
    },
    {
      "entropy": 6.489036273956299,
      "epoch": 1.5698539247685805,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004761558584806988,
      "loss": 5.9296,
      "mean_token_accuracy": 0.1568288043141365,
      "num_tokens": 31740796.0,
      "step": 14670
    },
    {
      "entropy": 6.4686554908752445,
      "epoch": 1.5703889988763444,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004761387264579751,
      "loss": 6.0173,
      "mean_token_accuracy": 0.15667476654052734,
      "num_tokens": 31750559.0,
      "step": 14675
    },
    {
      "entropy": 6.395568418502807,
      "epoch": 1.5709240729841083,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004761215886273989,
      "loss": 5.9307,
      "mean_token_accuracy": 0.15865208208560944,
      "num_tokens": 31761559.0,
      "step": 14680
    },
    {
      "entropy": 6.407035064697266,
      "epoch": 1.5714591470918722,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047610444498946507,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16182800084352494,
      "num_tokens": 31771677.0,
      "step": 14685
    },
    {
      "entropy": 6.4499688148498535,
      "epoch": 1.571994221199636,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004760872955446688,
      "loss": 6.0233,
      "mean_token_accuracy": 0.1523835375905037,
      "num_tokens": 31783028.0,
      "step": 14690
    },
    {
      "entropy": 6.444378280639649,
      "epoch": 1.5725292953074002,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047607014029350504,
      "loss": 5.9964,
      "mean_token_accuracy": 0.16041022092103957,
      "num_tokens": 31793110.0,
      "step": 14695
    },
    {
      "entropy": 6.383259534835815,
      "epoch": 1.573064369415164,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047605297923646923,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15845825970172883,
      "num_tokens": 31804193.0,
      "step": 14700
    },
    {
      "entropy": 6.4257618427276615,
      "epoch": 1.573599443522928,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047603581237405684,
      "loss": 5.9779,
      "mean_token_accuracy": 0.1595204859972,
      "num_tokens": 31815752.0,
      "step": 14705
    },
    {
      "entropy": 6.461740636825562,
      "epoch": 1.574134517630692,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047601863970676367,
      "loss": 6.0404,
      "mean_token_accuracy": 0.15752119868993758,
      "num_tokens": 31825936.0,
      "step": 14710
    },
    {
      "entropy": 6.445704317092895,
      "epoch": 1.5746695917384557,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047600146123508554,
      "loss": 5.9515,
      "mean_token_accuracy": 0.15699632167816163,
      "num_tokens": 31836601.0,
      "step": 14715
    },
    {
      "entropy": 6.329979801177979,
      "epoch": 1.5752046658462198,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047598427695951846,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1728365808725357,
      "num_tokens": 31846476.0,
      "step": 14720
    },
    {
      "entropy": 6.374997806549072,
      "epoch": 1.5757397399539836,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047596708688055857,
      "loss": 5.8181,
      "mean_token_accuracy": 0.16548366695642472,
      "num_tokens": 31856151.0,
      "step": 14725
    },
    {
      "entropy": 6.4829412460327145,
      "epoch": 1.5762748140617475,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004759498909987023,
      "loss": 6.0269,
      "mean_token_accuracy": 0.14958452582359313,
      "num_tokens": 31867419.0,
      "step": 14730
    },
    {
      "entropy": 6.446636438369751,
      "epoch": 1.5768098881695116,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004759326893144462,
      "loss": 6.0496,
      "mean_token_accuracy": 0.15669282525777817,
      "num_tokens": 31878036.0,
      "step": 14735
    },
    {
      "entropy": 6.439410209655762,
      "epoch": 1.5773449622772753,
      "grad_norm": 0.953125,
      "learning_rate": 0.000475915481828287,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15094943419098855,
      "num_tokens": 31887919.0,
      "step": 14740
    },
    {
      "entropy": 6.399717998504639,
      "epoch": 1.5778800363850394,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047589826854072133,
      "loss": 5.9767,
      "mean_token_accuracy": 0.1534537985920906,
      "num_tokens": 31898962.0,
      "step": 14745
    },
    {
      "entropy": 6.46125226020813,
      "epoch": 1.5784151104928033,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00047588104945224654,
      "loss": 6.0759,
      "mean_token_accuracy": 0.14978457540273665,
      "num_tokens": 31911027.0,
      "step": 14750
    },
    {
      "entropy": 6.499993705749512,
      "epoch": 1.5789501846005671,
      "grad_norm": 1.0,
      "learning_rate": 0.0004758638245633595,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15864563435316087,
      "num_tokens": 31921226.0,
      "step": 14755
    },
    {
      "entropy": 6.3904369354248045,
      "epoch": 1.5794852587083312,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004758465938745579,
      "loss": 5.9488,
      "mean_token_accuracy": 0.16000099629163742,
      "num_tokens": 31932848.0,
      "step": 14760
    },
    {
      "entropy": 6.370209169387818,
      "epoch": 1.5800203328160949,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047582935738633915,
      "loss": 5.8585,
      "mean_token_accuracy": 0.15453814268112182,
      "num_tokens": 31943757.0,
      "step": 14765
    },
    {
      "entropy": 6.45740156173706,
      "epoch": 1.580555406923859,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004758121150992009,
      "loss": 6.0151,
      "mean_token_accuracy": 0.14922859519720078,
      "num_tokens": 31954391.0,
      "step": 14770
    },
    {
      "entropy": 6.451048707962036,
      "epoch": 1.5810904810316229,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047579486701364093,
      "loss": 6.013,
      "mean_token_accuracy": 0.15529725104570388,
      "num_tokens": 31964497.0,
      "step": 14775
    },
    {
      "entropy": 6.498961925506592,
      "epoch": 1.5816255551393867,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004757776131301575,
      "loss": 6.0102,
      "mean_token_accuracy": 0.1508101060986519,
      "num_tokens": 31974466.0,
      "step": 14780
    },
    {
      "entropy": 6.511969375610351,
      "epoch": 1.5821606292471508,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004757603534492487,
      "loss": 6.0252,
      "mean_token_accuracy": 0.154202464222908,
      "num_tokens": 31985890.0,
      "step": 14785
    },
    {
      "entropy": 6.425185918807983,
      "epoch": 1.5826957033549145,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004757430879714128,
      "loss": 6.0683,
      "mean_token_accuracy": 0.15507468581199646,
      "num_tokens": 31996865.0,
      "step": 14790
    },
    {
      "entropy": 6.31184401512146,
      "epoch": 1.5832307774626786,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004757258166971485,
      "loss": 5.845,
      "mean_token_accuracy": 0.16904890537261963,
      "num_tokens": 32008121.0,
      "step": 14795
    },
    {
      "entropy": 6.494567394256592,
      "epoch": 1.5837658515704425,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004757085396269545,
      "loss": 6.0607,
      "mean_token_accuracy": 0.15636117234826089,
      "num_tokens": 32020552.0,
      "step": 14800
    },
    {
      "entropy": 6.493414354324341,
      "epoch": 1.5843009256782064,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047569125676132963,
      "loss": 6.0115,
      "mean_token_accuracy": 0.15258451104164122,
      "num_tokens": 32032361.0,
      "step": 14805
    },
    {
      "entropy": 6.519532155990601,
      "epoch": 1.5848359997859705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047567396810077286,
      "loss": 5.9971,
      "mean_token_accuracy": 0.14865338206291198,
      "num_tokens": 32043233.0,
      "step": 14810
    },
    {
      "entropy": 6.471884298324585,
      "epoch": 1.5853710738937343,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047565667364578344,
      "loss": 5.9512,
      "mean_token_accuracy": 0.16173680424690245,
      "num_tokens": 32054273.0,
      "step": 14815
    },
    {
      "entropy": 6.396596765518188,
      "epoch": 1.5859061480014982,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004756393733968608,
      "loss": 5.9506,
      "mean_token_accuracy": 0.16215929239988328,
      "num_tokens": 32064200.0,
      "step": 14820
    },
    {
      "entropy": 6.421055603027344,
      "epoch": 1.586441222109262,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004756220673545045,
      "loss": 5.9312,
      "mean_token_accuracy": 0.1541180282831192,
      "num_tokens": 32074823.0,
      "step": 14825
    },
    {
      "entropy": 6.486859369277954,
      "epoch": 1.586976296217026,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047560475551921416,
      "loss": 6.02,
      "mean_token_accuracy": 0.15864337682724,
      "num_tokens": 32085333.0,
      "step": 14830
    },
    {
      "entropy": 6.473203802108765,
      "epoch": 1.58751137032479,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004755874378914897,
      "loss": 6.014,
      "mean_token_accuracy": 0.14848120808601378,
      "num_tokens": 32095827.0,
      "step": 14835
    },
    {
      "entropy": 6.4260095119476315,
      "epoch": 1.588046444432554,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004755701144718311,
      "loss": 5.9977,
      "mean_token_accuracy": 0.15794107913970948,
      "num_tokens": 32106949.0,
      "step": 14840
    },
    {
      "entropy": 6.429665374755859,
      "epoch": 1.5885815185403178,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047555278526073867,
      "loss": 5.95,
      "mean_token_accuracy": 0.16066475808620453,
      "num_tokens": 32118269.0,
      "step": 14845
    },
    {
      "entropy": 6.429636430740357,
      "epoch": 1.589116592648082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004755354502587127,
      "loss": 5.9715,
      "mean_token_accuracy": 0.15288643091917037,
      "num_tokens": 32128406.0,
      "step": 14850
    },
    {
      "entropy": 6.403693675994873,
      "epoch": 1.5896516667558456,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047551810946625385,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15519902855157852,
      "num_tokens": 32140112.0,
      "step": 14855
    },
    {
      "entropy": 6.460103225708008,
      "epoch": 1.5901867408636097,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047550076288386276,
      "loss": 5.916,
      "mean_token_accuracy": 0.16386743038892745,
      "num_tokens": 32150010.0,
      "step": 14860
    },
    {
      "entropy": 6.444653749465942,
      "epoch": 1.5907218149713735,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004754834105120403,
      "loss": 6.0259,
      "mean_token_accuracy": 0.1547081857919693,
      "num_tokens": 32160694.0,
      "step": 14865
    },
    {
      "entropy": 6.422895431518555,
      "epoch": 1.5912568890791374,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047546605235128755,
      "loss": 6.0484,
      "mean_token_accuracy": 0.1528032213449478,
      "num_tokens": 32171947.0,
      "step": 14870
    },
    {
      "entropy": 6.479681873321534,
      "epoch": 1.5917919631869015,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004754486884021057,
      "loss": 6.004,
      "mean_token_accuracy": 0.15590064227581024,
      "num_tokens": 32182849.0,
      "step": 14875
    },
    {
      "entropy": 6.474952268600464,
      "epoch": 1.5923270372946652,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004754313186649961,
      "loss": 6.0161,
      "mean_token_accuracy": 0.15239461362361909,
      "num_tokens": 32192871.0,
      "step": 14880
    },
    {
      "entropy": 6.468064069747925,
      "epoch": 1.5928621114024293,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047541394314046037,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15061285197734833,
      "num_tokens": 32203289.0,
      "step": 14885
    },
    {
      "entropy": 6.465286016464233,
      "epoch": 1.5933971855101932,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004753965618290002,
      "loss": 5.9927,
      "mean_token_accuracy": 0.1517588183283806,
      "num_tokens": 32213238.0,
      "step": 14890
    },
    {
      "entropy": 6.395171165466309,
      "epoch": 1.593932259617957,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004753791747311175,
      "loss": 5.8934,
      "mean_token_accuracy": 0.1645382285118103,
      "num_tokens": 32223384.0,
      "step": 14895
    },
    {
      "entropy": 6.44575686454773,
      "epoch": 1.5944673337257211,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047536178184731426,
      "loss": 5.9907,
      "mean_token_accuracy": 0.1572466164827347,
      "num_tokens": 32235358.0,
      "step": 14900
    },
    {
      "entropy": 6.553261756896973,
      "epoch": 1.5950024078334848,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004753443831780927,
      "loss": 6.0463,
      "mean_token_accuracy": 0.1530042201280594,
      "num_tokens": 32246459.0,
      "step": 14905
    },
    {
      "entropy": 6.424081754684448,
      "epoch": 1.595537481941249,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004753269787239552,
      "loss": 5.9231,
      "mean_token_accuracy": 0.15765887796878814,
      "num_tokens": 32257798.0,
      "step": 14910
    },
    {
      "entropy": 6.443861389160157,
      "epoch": 1.5960725560490128,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047530956848540435,
      "loss": 5.9725,
      "mean_token_accuracy": 0.15628000497817993,
      "num_tokens": 32268118.0,
      "step": 14915
    },
    {
      "entropy": 6.399180889129639,
      "epoch": 1.5966076301567766,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004752921524629429,
      "loss": 5.8838,
      "mean_token_accuracy": 0.1606060743331909,
      "num_tokens": 32279676.0,
      "step": 14920
    },
    {
      "entropy": 6.450523471832275,
      "epoch": 1.5971427042645407,
      "grad_norm": 1.0,
      "learning_rate": 0.0004752747306570737,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15819958597421646,
      "num_tokens": 32290613.0,
      "step": 14925
    },
    {
      "entropy": 6.475983428955078,
      "epoch": 1.5976777783723044,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047525730306829977,
      "loss": 5.9771,
      "mean_token_accuracy": 0.15590822249650954,
      "num_tokens": 32300873.0,
      "step": 14930
    },
    {
      "entropy": 6.457882213592529,
      "epoch": 1.5982128524800685,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004752398696971244,
      "loss": 5.9886,
      "mean_token_accuracy": 0.1568312995135784,
      "num_tokens": 32312326.0,
      "step": 14935
    },
    {
      "entropy": 6.479448032379151,
      "epoch": 1.5987479265878324,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047522243054405085,
      "loss": 5.9697,
      "mean_token_accuracy": 0.15000547468662262,
      "num_tokens": 32324322.0,
      "step": 14940
    },
    {
      "entropy": 6.474199485778809,
      "epoch": 1.5992830006955963,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004752049856095828,
      "loss": 5.9561,
      "mean_token_accuracy": 0.15689892768859864,
      "num_tokens": 32334397.0,
      "step": 14945
    },
    {
      "entropy": 6.4489668846130375,
      "epoch": 1.5998180748033604,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047518753489422393,
      "loss": 5.9492,
      "mean_token_accuracy": 0.16277362555265426,
      "num_tokens": 32345131.0,
      "step": 14950
    },
    {
      "entropy": 6.383879995346069,
      "epoch": 1.6003531489111242,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047517007839847816,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15686536729335784,
      "num_tokens": 32355520.0,
      "step": 14955
    },
    {
      "entropy": 6.4584451675415036,
      "epoch": 1.6008882230188881,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004751526161228494,
      "loss": 6.0172,
      "mean_token_accuracy": 0.15973505824804307,
      "num_tokens": 32366695.0,
      "step": 14960
    },
    {
      "entropy": 6.381473970413208,
      "epoch": 1.601423297126652,
      "grad_norm": 1.03125,
      "learning_rate": 0.000475135148067842,
      "loss": 5.9725,
      "mean_token_accuracy": 0.1534843325614929,
      "num_tokens": 32377467.0,
      "step": 14965
    },
    {
      "entropy": 6.503041934967041,
      "epoch": 1.6019583712344159,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047511767423396033,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15630608648061753,
      "num_tokens": 32388021.0,
      "step": 14970
    },
    {
      "entropy": 6.469640636444092,
      "epoch": 1.60249344534218,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000475100194621709,
      "loss": 5.933,
      "mean_token_accuracy": 0.1623366042971611,
      "num_tokens": 32398105.0,
      "step": 14975
    },
    {
      "entropy": 6.433776330947876,
      "epoch": 1.6030285194499438,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004750827092315926,
      "loss": 6.0698,
      "mean_token_accuracy": 0.15262497290968896,
      "num_tokens": 32409876.0,
      "step": 14980
    },
    {
      "entropy": 6.462302160263062,
      "epoch": 1.6035635935577077,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047506521806411613,
      "loss": 5.9395,
      "mean_token_accuracy": 0.16792636513710021,
      "num_tokens": 32420283.0,
      "step": 14985
    },
    {
      "entropy": 6.466500377655029,
      "epoch": 1.6040986676654718,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047504772111978463,
      "loss": 6.0482,
      "mean_token_accuracy": 0.145388600230217,
      "num_tokens": 32431164.0,
      "step": 14990
    },
    {
      "entropy": 6.471164703369141,
      "epoch": 1.6046337417732355,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004750302183991032,
      "loss": 5.9542,
      "mean_token_accuracy": 0.15546939596533776,
      "num_tokens": 32442253.0,
      "step": 14995
    },
    {
      "entropy": 6.463207006454468,
      "epoch": 1.6051688158809996,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004750127099025774,
      "loss": 6.0489,
      "mean_token_accuracy": 0.15478157997131348,
      "num_tokens": 32452800.0,
      "step": 15000
    },
    {
      "epoch": 1.6051688158809996,
      "eval_entropy": 6.269833217733954,
      "eval_loss": 6.059521198272705,
      "eval_mean_token_accuracy": 0.1604631654920704,
      "eval_num_tokens": 32452800.0,
      "eval_runtime": 17.627,
      "eval_samples_per_second": 1683.724,
      "eval_steps_per_second": 210.473,
      "step": 15000
    },
    {
      "entropy": 6.483892440795898,
      "epoch": 1.6057038899887635,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004749951956307126,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15313233733177184,
      "num_tokens": 32462890.0,
      "step": 15005
    },
    {
      "entropy": 6.419841670989991,
      "epoch": 1.6062389640965273,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004749776755840148,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1624774754047394,
      "num_tokens": 32473422.0,
      "step": 15010
    },
    {
      "entropy": 6.4280518054962155,
      "epoch": 1.6067740382042914,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047496014976298957,
      "loss": 5.9913,
      "mean_token_accuracy": 0.15928715616464614,
      "num_tokens": 32483980.0,
      "step": 15015
    },
    {
      "entropy": 6.534251260757446,
      "epoch": 1.607309112312055,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004749426181681432,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15095625817775726,
      "num_tokens": 32494905.0,
      "step": 15020
    },
    {
      "entropy": 6.523290157318115,
      "epoch": 1.6078441864198192,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047492508079998173,
      "loss": 6.0402,
      "mean_token_accuracy": 0.15197864025831223,
      "num_tokens": 32505799.0,
      "step": 15025
    },
    {
      "entropy": 6.484270286560059,
      "epoch": 1.608379260527583,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004749075376590117,
      "loss": 6.0392,
      "mean_token_accuracy": 0.15244773775339127,
      "num_tokens": 32516328.0,
      "step": 15030
    },
    {
      "entropy": 6.476931428909301,
      "epoch": 1.608914334635347,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004748899887457396,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15822982788085938,
      "num_tokens": 32526335.0,
      "step": 15035
    },
    {
      "entropy": 6.4626367568969725,
      "epoch": 1.609449408743111,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047487243406067227,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15801745355129243,
      "num_tokens": 32536896.0,
      "step": 15040
    },
    {
      "entropy": 6.40813570022583,
      "epoch": 1.6099844828508747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004748548736043164,
      "loss": 5.8736,
      "mean_token_accuracy": 0.1608843594789505,
      "num_tokens": 32547604.0,
      "step": 15045
    },
    {
      "entropy": 6.495395469665527,
      "epoch": 1.6105195569586388,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047483730737717914,
      "loss": 6.0158,
      "mean_token_accuracy": 0.15088473558425902,
      "num_tokens": 32558539.0,
      "step": 15050
    },
    {
      "entropy": 6.485409069061279,
      "epoch": 1.6110546310664027,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004748197353797677,
      "loss": 6.0918,
      "mean_token_accuracy": 0.14981062859296798,
      "num_tokens": 32569787.0,
      "step": 15055
    },
    {
      "entropy": 6.432846784591675,
      "epoch": 1.6115897051741666,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004748021576125895,
      "loss": 6.0097,
      "mean_token_accuracy": 0.1574966460466385,
      "num_tokens": 32580146.0,
      "step": 15060
    },
    {
      "entropy": 6.433947563171387,
      "epoch": 1.6121247792819307,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004747845740761521,
      "loss": 5.9731,
      "mean_token_accuracy": 0.1553043581545353,
      "num_tokens": 32591243.0,
      "step": 15065
    },
    {
      "entropy": 6.509514856338501,
      "epoch": 1.6126598533896943,
      "grad_norm": 1.125,
      "learning_rate": 0.0004747669847709631,
      "loss": 6.0237,
      "mean_token_accuracy": 0.15529226064682006,
      "num_tokens": 32602424.0,
      "step": 15070
    },
    {
      "entropy": 6.5432251453399655,
      "epoch": 1.6131949274974584,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004747493896975306,
      "loss": 6.0883,
      "mean_token_accuracy": 0.15103659629821778,
      "num_tokens": 32613703.0,
      "step": 15075
    },
    {
      "entropy": 6.472999000549317,
      "epoch": 1.6137300016052223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004747317888563625,
      "loss": 6.0343,
      "mean_token_accuracy": 0.15049496293067932,
      "num_tokens": 32624371.0,
      "step": 15080
    },
    {
      "entropy": 6.471946573257446,
      "epoch": 1.6142650757129862,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047471418224796704,
      "loss": 5.989,
      "mean_token_accuracy": 0.15650333166122438,
      "num_tokens": 32634039.0,
      "step": 15085
    },
    {
      "entropy": 6.433880949020386,
      "epoch": 1.6148001498207503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004746965698728526,
      "loss": 5.9475,
      "mean_token_accuracy": 0.1586874783039093,
      "num_tokens": 32644644.0,
      "step": 15090
    },
    {
      "entropy": 6.4981037139892575,
      "epoch": 1.6153352239285141,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00047467895173152787,
      "loss": 6.0074,
      "mean_token_accuracy": 0.1604248508810997,
      "num_tokens": 32655961.0,
      "step": 15095
    },
    {
      "entropy": 6.423077154159546,
      "epoch": 1.615870298036278,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047466132782450146,
      "loss": 5.91,
      "mean_token_accuracy": 0.16534264683723449,
      "num_tokens": 32665506.0,
      "step": 15100
    },
    {
      "entropy": 6.388642072677612,
      "epoch": 1.616405372144042,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047464369815228215,
      "loss": 5.9295,
      "mean_token_accuracy": 0.15987130701541902,
      "num_tokens": 32676278.0,
      "step": 15105
    },
    {
      "entropy": 6.450550222396851,
      "epoch": 1.6169404462518058,
      "grad_norm": 1.0,
      "learning_rate": 0.0004746260627153792,
      "loss": 5.9246,
      "mean_token_accuracy": 0.1558288276195526,
      "num_tokens": 32686835.0,
      "step": 15110
    },
    {
      "entropy": 6.500835180282593,
      "epoch": 1.6174755203595699,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047460842151430166,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14826445132493973,
      "num_tokens": 32696856.0,
      "step": 15115
    },
    {
      "entropy": 6.47217960357666,
      "epoch": 1.6180105944673338,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047459077454955914,
      "loss": 5.9804,
      "mean_token_accuracy": 0.15692849159240724,
      "num_tokens": 32707514.0,
      "step": 15120
    },
    {
      "entropy": 6.517728662490844,
      "epoch": 1.6185456685750976,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047457312182166094,
      "loss": 6.069,
      "mean_token_accuracy": 0.14918784722685813,
      "num_tokens": 32718934.0,
      "step": 15125
    },
    {
      "entropy": 6.4754081726074215,
      "epoch": 1.6190807426828617,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004745554633311169,
      "loss": 5.9522,
      "mean_token_accuracy": 0.1574022263288498,
      "num_tokens": 32729249.0,
      "step": 15130
    },
    {
      "entropy": 6.485927152633667,
      "epoch": 1.6196158167906254,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047453779907843686,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15268929600715636,
      "num_tokens": 32739948.0,
      "step": 15135
    },
    {
      "entropy": 6.479439067840576,
      "epoch": 1.6201508908983895,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047452012906413105,
      "loss": 6.0235,
      "mean_token_accuracy": 0.15684626996517181,
      "num_tokens": 32751482.0,
      "step": 15140
    },
    {
      "entropy": 6.457862806320191,
      "epoch": 1.6206859650061534,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047450245328870943,
      "loss": 5.8499,
      "mean_token_accuracy": 0.17029210925102234,
      "num_tokens": 32761260.0,
      "step": 15145
    },
    {
      "entropy": 6.490730857849121,
      "epoch": 1.6212210391139172,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047448477175268255,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1561057075858116,
      "num_tokens": 32772684.0,
      "step": 15150
    },
    {
      "entropy": 6.3258082389831545,
      "epoch": 1.6217561132216813,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047446708445656095,
      "loss": 5.8128,
      "mean_token_accuracy": 0.17214541137218475,
      "num_tokens": 32784022.0,
      "step": 15155
    },
    {
      "entropy": 6.421033716201782,
      "epoch": 1.622291187329445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004744493914008552,
      "loss": 5.9324,
      "mean_token_accuracy": 0.15665444433689119,
      "num_tokens": 32794104.0,
      "step": 15160
    },
    {
      "entropy": 6.482336854934692,
      "epoch": 1.622826261437209,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004744316925860764,
      "loss": 5.9226,
      "mean_token_accuracy": 0.1626945748925209,
      "num_tokens": 32804038.0,
      "step": 15165
    },
    {
      "entropy": 6.422618722915649,
      "epoch": 1.623361335544973,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004744139880127355,
      "loss": 5.9958,
      "mean_token_accuracy": 0.15421227216720582,
      "num_tokens": 32815620.0,
      "step": 15170
    },
    {
      "entropy": 6.445086479187012,
      "epoch": 1.6238964096527368,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00047439627768134366,
      "loss": 5.8984,
      "mean_token_accuracy": 0.16761928647756577,
      "num_tokens": 32826353.0,
      "step": 15175
    },
    {
      "entropy": 6.465128374099732,
      "epoch": 1.624431483760501,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004743785615924124,
      "loss": 6.0324,
      "mean_token_accuracy": 0.15319160521030425,
      "num_tokens": 32836582.0,
      "step": 15180
    },
    {
      "entropy": 6.422350597381592,
      "epoch": 1.6249665578682646,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004743608397464531,
      "loss": 5.9751,
      "mean_token_accuracy": 0.15573810636997223,
      "num_tokens": 32848073.0,
      "step": 15185
    },
    {
      "entropy": 6.469088840484619,
      "epoch": 1.6255016319760287,
      "grad_norm": 1.125,
      "learning_rate": 0.00047434311214397765,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15321725308895112,
      "num_tokens": 32857839.0,
      "step": 15190
    },
    {
      "entropy": 6.440592241287232,
      "epoch": 1.6260367060837926,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00047432537878549786,
      "loss": 5.9062,
      "mean_token_accuracy": 0.17497546672821046,
      "num_tokens": 32868087.0,
      "step": 15195
    },
    {
      "entropy": 6.487994956970215,
      "epoch": 1.6265717801915565,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047430763967152564,
      "loss": 5.9884,
      "mean_token_accuracy": 0.15571620911359788,
      "num_tokens": 32880622.0,
      "step": 15200
    },
    {
      "entropy": 6.496773958206177,
      "epoch": 1.6271068542993206,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047428989480257337,
      "loss": 5.9781,
      "mean_token_accuracy": 0.15535206645727156,
      "num_tokens": 32891405.0,
      "step": 15205
    },
    {
      "entropy": 6.429591131210327,
      "epoch": 1.6276419284070842,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004742721441791535,
      "loss": 5.9287,
      "mean_token_accuracy": 0.15862076133489608,
      "num_tokens": 32900487.0,
      "step": 15210
    },
    {
      "entropy": 6.395563793182373,
      "epoch": 1.6281770025148483,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004742543878017784,
      "loss": 5.8903,
      "mean_token_accuracy": 0.16640800535678862,
      "num_tokens": 32912139.0,
      "step": 15215
    },
    {
      "entropy": 6.479650640487671,
      "epoch": 1.6287120766226122,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047423662567096085,
      "loss": 5.9815,
      "mean_token_accuracy": 0.1562500551342964,
      "num_tokens": 32922972.0,
      "step": 15220
    },
    {
      "entropy": 6.46083550453186,
      "epoch": 1.629247150730376,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047421885778721376,
      "loss": 5.9299,
      "mean_token_accuracy": 0.1612228736281395,
      "num_tokens": 32932609.0,
      "step": 15225
    },
    {
      "entropy": 6.376732349395752,
      "epoch": 1.6297822248381402,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047420108415105006,
      "loss": 5.942,
      "mean_token_accuracy": 0.16554344892501832,
      "num_tokens": 32942706.0,
      "step": 15230
    },
    {
      "entropy": 6.393526315689087,
      "epoch": 1.630317298945904,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047418330476298317,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15983299314975738,
      "num_tokens": 32952240.0,
      "step": 15235
    },
    {
      "entropy": 6.470578145980835,
      "epoch": 1.630852373053668,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004741655196235262,
      "loss": 6.0053,
      "mean_token_accuracy": 0.1538558319211006,
      "num_tokens": 32963249.0,
      "step": 15240
    },
    {
      "entropy": 6.495720624923706,
      "epoch": 1.6313874471614318,
      "grad_norm": 1.0,
      "learning_rate": 0.00047414772873319294,
      "loss": 5.9904,
      "mean_token_accuracy": 0.16168263405561448,
      "num_tokens": 32974004.0,
      "step": 15245
    },
    {
      "entropy": 6.502746677398681,
      "epoch": 1.6319225212691957,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000474129932092497,
      "loss": 5.9633,
      "mean_token_accuracy": 0.15608260482549668,
      "num_tokens": 32985236.0,
      "step": 15250
    },
    {
      "entropy": 6.488498210906982,
      "epoch": 1.6324575953769598,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004741121297019522,
      "loss": 6.0166,
      "mean_token_accuracy": 0.15756164416670798,
      "num_tokens": 32995155.0,
      "step": 15255
    },
    {
      "entropy": 6.491855907440185,
      "epoch": 1.6329926694847237,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00047409432156207264,
      "loss": 5.9937,
      "mean_token_accuracy": 0.16466163098812103,
      "num_tokens": 33007562.0,
      "step": 15260
    },
    {
      "entropy": 6.460915374755859,
      "epoch": 1.6335277435924875,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047407650767337256,
      "loss": 5.9899,
      "mean_token_accuracy": 0.15076854676008225,
      "num_tokens": 33018653.0,
      "step": 15265
    },
    {
      "entropy": 6.5140032291412355,
      "epoch": 1.6340628177002516,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004740586880363663,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1495158150792122,
      "num_tokens": 33029470.0,
      "step": 15270
    },
    {
      "entropy": 6.444291830062866,
      "epoch": 1.6345978918080153,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047404086265156835,
      "loss": 5.9443,
      "mean_token_accuracy": 0.16093349009752272,
      "num_tokens": 33039597.0,
      "step": 15275
    },
    {
      "entropy": 6.425167369842529,
      "epoch": 1.6351329659157794,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004740230315194935,
      "loss": 5.9562,
      "mean_token_accuracy": 0.15737417042255403,
      "num_tokens": 33050984.0,
      "step": 15280
    },
    {
      "entropy": 6.401877117156983,
      "epoch": 1.6356680400235433,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047400519464065656,
      "loss": 6.0047,
      "mean_token_accuracy": 0.1556090846657753,
      "num_tokens": 33062250.0,
      "step": 15285
    },
    {
      "entropy": 6.499827480316162,
      "epoch": 1.6362031141313071,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004739873520155726,
      "loss": 6.0576,
      "mean_token_accuracy": 0.1498544044792652,
      "num_tokens": 33072129.0,
      "step": 15290
    },
    {
      "entropy": 6.48159613609314,
      "epoch": 1.6367381882390712,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004739695036447568,
      "loss": 5.9131,
      "mean_token_accuracy": 0.15836820900440216,
      "num_tokens": 33082475.0,
      "step": 15295
    },
    {
      "entropy": 6.40509524345398,
      "epoch": 1.637273262346835,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047395164952872455,
      "loss": 5.9373,
      "mean_token_accuracy": 0.1668820321559906,
      "num_tokens": 33093791.0,
      "step": 15300
    },
    {
      "entropy": 6.46657509803772,
      "epoch": 1.637808336454599,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004739337896679914,
      "loss": 5.9249,
      "mean_token_accuracy": 0.16055576503276825,
      "num_tokens": 33104231.0,
      "step": 15305
    },
    {
      "entropy": 6.430231142044067,
      "epoch": 1.6383434105623629,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047391592406307296,
      "loss": 5.9368,
      "mean_token_accuracy": 0.1555699348449707,
      "num_tokens": 33115118.0,
      "step": 15310
    },
    {
      "entropy": 6.456293535232544,
      "epoch": 1.6388784846701268,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004738980527144852,
      "loss": 5.9275,
      "mean_token_accuracy": 0.1623731330037117,
      "num_tokens": 33126154.0,
      "step": 15315
    },
    {
      "entropy": 6.475072431564331,
      "epoch": 1.6394135587778909,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004738801756227441,
      "loss": 6.015,
      "mean_token_accuracy": 0.15676087588071824,
      "num_tokens": 33137342.0,
      "step": 15320
    },
    {
      "entropy": 6.4362259864807125,
      "epoch": 1.6399486328856545,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047386229278836595,
      "loss": 5.9614,
      "mean_token_accuracy": 0.16141223013401032,
      "num_tokens": 33147661.0,
      "step": 15325
    },
    {
      "entropy": 6.477433586120606,
      "epoch": 1.6404837069934186,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047384440421186695,
      "loss": 6.019,
      "mean_token_accuracy": 0.15445654541254045,
      "num_tokens": 33157106.0,
      "step": 15330
    },
    {
      "entropy": 6.466292858123779,
      "epoch": 1.6410187811011825,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004738265098937638,
      "loss": 5.9988,
      "mean_token_accuracy": 0.15546947866678237,
      "num_tokens": 33169451.0,
      "step": 15335
    },
    {
      "entropy": 6.507438516616821,
      "epoch": 1.6415538552089464,
      "grad_norm": 0.9375,
      "learning_rate": 0.000473808609834573,
      "loss": 6.0058,
      "mean_token_accuracy": 0.15672807842493058,
      "num_tokens": 33180799.0,
      "step": 15340
    },
    {
      "entropy": 6.520973491668701,
      "epoch": 1.6420889293167105,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047379070403481165,
      "loss": 6.0683,
      "mean_token_accuracy": 0.15109288543462754,
      "num_tokens": 33191948.0,
      "step": 15345
    },
    {
      "entropy": 6.461967086791992,
      "epoch": 1.6426240034244743,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047377279249499656,
      "loss": 5.9905,
      "mean_token_accuracy": 0.15147173404693604,
      "num_tokens": 33203536.0,
      "step": 15350
    },
    {
      "entropy": 6.476194715499878,
      "epoch": 1.6431590775322382,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004737548752156451,
      "loss": 5.9706,
      "mean_token_accuracy": 0.16004304736852645,
      "num_tokens": 33214526.0,
      "step": 15355
    },
    {
      "entropy": 6.4550611019134525,
      "epoch": 1.643694151640002,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004737369521972745,
      "loss": 5.9415,
      "mean_token_accuracy": 0.16206754744052887,
      "num_tokens": 33226560.0,
      "step": 15360
    },
    {
      "entropy": 6.423022794723511,
      "epoch": 1.644229225747766,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047371902344040234,
      "loss": 5.9503,
      "mean_token_accuracy": 0.1639052540063858,
      "num_tokens": 33237441.0,
      "step": 15365
    },
    {
      "entropy": 6.464718151092529,
      "epoch": 1.64476429985553,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047370108894554633,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15839987397193908,
      "num_tokens": 33248229.0,
      "step": 15370
    },
    {
      "entropy": 6.43058934211731,
      "epoch": 1.645299373963294,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004736831487132243,
      "loss": 5.8869,
      "mean_token_accuracy": 0.16749035120010375,
      "num_tokens": 33258992.0,
      "step": 15375
    },
    {
      "entropy": 6.405369997024536,
      "epoch": 1.6458344480710578,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004736652027439542,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15819730907678603,
      "num_tokens": 33269583.0,
      "step": 15380
    },
    {
      "entropy": 6.478096437454224,
      "epoch": 1.646369522178822,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047364725103825437,
      "loss": 6.0433,
      "mean_token_accuracy": 0.15106171220541,
      "num_tokens": 33280236.0,
      "step": 15385
    },
    {
      "entropy": 6.4326269149780275,
      "epoch": 1.6469045962865856,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000473629293596643,
      "loss": 5.9575,
      "mean_token_accuracy": 0.15307405292987825,
      "num_tokens": 33291327.0,
      "step": 15390
    },
    {
      "entropy": 6.463659381866455,
      "epoch": 1.6474396703943497,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047361133041963873,
      "loss": 6.0518,
      "mean_token_accuracy": 0.15232043862342834,
      "num_tokens": 33302125.0,
      "step": 15395
    },
    {
      "entropy": 6.466626167297363,
      "epoch": 1.6479747445021136,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004735933615077602,
      "loss": 5.9812,
      "mean_token_accuracy": 0.15567114800214768,
      "num_tokens": 33312576.0,
      "step": 15400
    },
    {
      "entropy": 6.482025623321533,
      "epoch": 1.6485098186098774,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047357538686152627,
      "loss": 5.9513,
      "mean_token_accuracy": 0.15918897837400436,
      "num_tokens": 33322821.0,
      "step": 15405
    },
    {
      "entropy": 6.489942598342895,
      "epoch": 1.6490448927176415,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047355740648145596,
      "loss": 5.9942,
      "mean_token_accuracy": 0.151322340965271,
      "num_tokens": 33333951.0,
      "step": 15410
    },
    {
      "entropy": 6.450347471237182,
      "epoch": 1.6495799668254052,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004735394203680684,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15900604277849198,
      "num_tokens": 33344739.0,
      "step": 15415
    },
    {
      "entropy": 6.5108091831207275,
      "epoch": 1.6501150409331693,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00047352142852188304,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15764969885349273,
      "num_tokens": 33354956.0,
      "step": 15420
    },
    {
      "entropy": 6.422850513458252,
      "epoch": 1.6506501150409332,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004735034309434193,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16206090599298478,
      "num_tokens": 33364809.0,
      "step": 15425
    },
    {
      "entropy": 6.511590385437012,
      "epoch": 1.651185189148697,
      "grad_norm": 1.0,
      "learning_rate": 0.0004734854276331968,
      "loss": 5.9967,
      "mean_token_accuracy": 0.16299385875463485,
      "num_tokens": 33375324.0,
      "step": 15430
    },
    {
      "entropy": 6.473663187026977,
      "epoch": 1.6517202632564612,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047346741859173564,
      "loss": 6.0758,
      "mean_token_accuracy": 0.15059303641319274,
      "num_tokens": 33385963.0,
      "step": 15435
    },
    {
      "entropy": 6.485695123672485,
      "epoch": 1.6522553373642248,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004734494038195555,
      "loss": 6.0631,
      "mean_token_accuracy": 0.15361876040697098,
      "num_tokens": 33395927.0,
      "step": 15440
    },
    {
      "entropy": 6.493703889846802,
      "epoch": 1.652790411471989,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004734313833171768,
      "loss": 5.9573,
      "mean_token_accuracy": 0.16094202250242234,
      "num_tokens": 33406294.0,
      "step": 15445
    },
    {
      "entropy": 6.47003698348999,
      "epoch": 1.6533254855797528,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004734133570851198,
      "loss": 5.9221,
      "mean_token_accuracy": 0.1628483787178993,
      "num_tokens": 33416995.0,
      "step": 15450
    },
    {
      "entropy": 6.37986650466919,
      "epoch": 1.6538605596875167,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004733953251239049,
      "loss": 5.8991,
      "mean_token_accuracy": 0.1607152357697487,
      "num_tokens": 33428087.0,
      "step": 15455
    },
    {
      "entropy": 6.491735601425171,
      "epoch": 1.6543956337952808,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004733772874340529,
      "loss": 6.0629,
      "mean_token_accuracy": 0.15076288729906082,
      "num_tokens": 33439151.0,
      "step": 15460
    },
    {
      "entropy": 6.464688634872436,
      "epoch": 1.6549307079030444,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004733592440160846,
      "loss": 5.9516,
      "mean_token_accuracy": 0.16194438189268112,
      "num_tokens": 33449257.0,
      "step": 15465
    },
    {
      "entropy": 6.399406433105469,
      "epoch": 1.6554657820108085,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047334119487052104,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15699736326932906,
      "num_tokens": 33460472.0,
      "step": 15470
    },
    {
      "entropy": 6.467018699645996,
      "epoch": 1.6560008561185724,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047332313999788334,
      "loss": 5.9306,
      "mean_token_accuracy": 0.15310779064893723,
      "num_tokens": 33470193.0,
      "step": 15475
    },
    {
      "entropy": 6.385219097137451,
      "epoch": 1.6565359302263363,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047330507939869284,
      "loss": 5.8698,
      "mean_token_accuracy": 0.1609860487282276,
      "num_tokens": 33481098.0,
      "step": 15480
    },
    {
      "entropy": 6.4897605895996096,
      "epoch": 1.6570710043341004,
      "grad_norm": 0.953125,
      "learning_rate": 0.000473287013073471,
      "loss": 5.9517,
      "mean_token_accuracy": 0.16119356453418732,
      "num_tokens": 33492514.0,
      "step": 15485
    },
    {
      "entropy": 6.42206974029541,
      "epoch": 1.6576060784418643,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004732689410227396,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15791187286376954,
      "num_tokens": 33504254.0,
      "step": 15490
    },
    {
      "entropy": 6.462930488586426,
      "epoch": 1.6581411525496281,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004732508632470203,
      "loss": 5.958,
      "mean_token_accuracy": 0.15695507377386092,
      "num_tokens": 33515101.0,
      "step": 15495
    },
    {
      "entropy": 6.486672258377075,
      "epoch": 1.658676226657392,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004732327797468352,
      "loss": 5.9735,
      "mean_token_accuracy": 0.15471500009298325,
      "num_tokens": 33525052.0,
      "step": 15500
    },
    {
      "entropy": 6.5187990188598635,
      "epoch": 1.6592113007651559,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004732146905227065,
      "loss": 6.1181,
      "mean_token_accuracy": 0.1431793101131916,
      "num_tokens": 33536841.0,
      "step": 15505
    },
    {
      "entropy": 6.486936807632446,
      "epoch": 1.65974637487292,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004731965955751564,
      "loss": 5.9931,
      "mean_token_accuracy": 0.15605949312448503,
      "num_tokens": 33547292.0,
      "step": 15510
    },
    {
      "entropy": 6.493230247497559,
      "epoch": 1.6602814489806839,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004731784949047075,
      "loss": 5.9382,
      "mean_token_accuracy": 0.15418656766414643,
      "num_tokens": 33557674.0,
      "step": 15515
    },
    {
      "entropy": 6.494319009780884,
      "epoch": 1.6608165230884477,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004731603885118824,
      "loss": 6.009,
      "mean_token_accuracy": 0.1474838986992836,
      "num_tokens": 33568396.0,
      "step": 15520
    },
    {
      "entropy": 6.46639575958252,
      "epoch": 1.6613515971962118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004731422763972039,
      "loss": 6.0427,
      "mean_token_accuracy": 0.15358905419707297,
      "num_tokens": 33579688.0,
      "step": 15525
    },
    {
      "entropy": 6.474834394454956,
      "epoch": 1.6618866713039755,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047312415856119505,
      "loss": 6.0398,
      "mean_token_accuracy": 0.15868439376354218,
      "num_tokens": 33591550.0,
      "step": 15530
    },
    {
      "entropy": 6.514468336105347,
      "epoch": 1.6624217454117396,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047310603500437895,
      "loss": 6.0104,
      "mean_token_accuracy": 0.1601698413491249,
      "num_tokens": 33600792.0,
      "step": 15535
    },
    {
      "entropy": 6.4268392562866214,
      "epoch": 1.6629568195195035,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00047308790572727895,
      "loss": 5.9341,
      "mean_token_accuracy": 0.16094885021448135,
      "num_tokens": 33611873.0,
      "step": 15540
    },
    {
      "entropy": 6.426998472213745,
      "epoch": 1.6634918936272673,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004730697707304184,
      "loss": 5.9761,
      "mean_token_accuracy": 0.15629124492406846,
      "num_tokens": 33622960.0,
      "step": 15545
    },
    {
      "entropy": 6.490062475204468,
      "epoch": 1.6640269677350314,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004730516300143211,
      "loss": 6.0531,
      "mean_token_accuracy": 0.1509164810180664,
      "num_tokens": 33635391.0,
      "step": 15550
    },
    {
      "entropy": 6.504662656784058,
      "epoch": 1.664562041842795,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004730334835795109,
      "loss": 5.9995,
      "mean_token_accuracy": 0.1491445153951645,
      "num_tokens": 33647942.0,
      "step": 15555
    },
    {
      "entropy": 6.519580698013305,
      "epoch": 1.6650971159505592,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004730153314265116,
      "loss": 5.9697,
      "mean_token_accuracy": 0.1562997281551361,
      "num_tokens": 33658425.0,
      "step": 15560
    },
    {
      "entropy": 6.397275304794311,
      "epoch": 1.665632190058323,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004729971735558475,
      "loss": 5.9431,
      "mean_token_accuracy": 0.166290844976902,
      "num_tokens": 33669645.0,
      "step": 15565
    },
    {
      "entropy": 6.499933338165283,
      "epoch": 1.666167264166087,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004729790099680428,
      "loss": 6.0299,
      "mean_token_accuracy": 0.153529342263937,
      "num_tokens": 33680415.0,
      "step": 15570
    },
    {
      "entropy": 6.472375917434692,
      "epoch": 1.666702338273851,
      "grad_norm": 0.91796875,
      "learning_rate": 0.000472960840663622,
      "loss": 6.0447,
      "mean_token_accuracy": 0.15471432656049727,
      "num_tokens": 33691425.0,
      "step": 15575
    },
    {
      "entropy": 6.447407341003418,
      "epoch": 1.6672374123816147,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004729426656431097,
      "loss": 5.9753,
      "mean_token_accuracy": 0.16040861904621123,
      "num_tokens": 33703845.0,
      "step": 15580
    },
    {
      "entropy": 6.52987265586853,
      "epoch": 1.6677724864893788,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004729244849070308,
      "loss": 5.9756,
      "mean_token_accuracy": 0.1585749566555023,
      "num_tokens": 33714737.0,
      "step": 15585
    },
    {
      "entropy": 6.4599363803863525,
      "epoch": 1.6683075605971427,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047290629845591026,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15528347790241243,
      "num_tokens": 33725042.0,
      "step": 15590
    },
    {
      "entropy": 6.431044244766236,
      "epoch": 1.6688426347049066,
      "grad_norm": 1.046875,
      "learning_rate": 0.000472888106290273,
      "loss": 5.9522,
      "mean_token_accuracy": 0.15756509825587273,
      "num_tokens": 33736306.0,
      "step": 15595
    },
    {
      "entropy": 6.510509586334228,
      "epoch": 1.6693777088126707,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047286990841064467,
      "loss": 6.0243,
      "mean_token_accuracy": 0.1485986441373825,
      "num_tokens": 33747327.0,
      "step": 15600
    },
    {
      "entropy": 6.446419620513916,
      "epoch": 1.6699127829204343,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004728517048175504,
      "loss": 5.9705,
      "mean_token_accuracy": 0.16039561182260514,
      "num_tokens": 33757723.0,
      "step": 15605
    },
    {
      "entropy": 6.383727216720581,
      "epoch": 1.6704478570281984,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000472833495511516,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15929416716098785,
      "num_tokens": 33769958.0,
      "step": 15610
    },
    {
      "entropy": 6.456109952926636,
      "epoch": 1.6709829311359623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047281528049306727,
      "loss": 5.9499,
      "mean_token_accuracy": 0.158903044462204,
      "num_tokens": 33780107.0,
      "step": 15615
    },
    {
      "entropy": 6.406916236877441,
      "epoch": 1.6715180052437262,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004727970597627301,
      "loss": 5.9748,
      "mean_token_accuracy": 0.15649791806936264,
      "num_tokens": 33791023.0,
      "step": 15620
    },
    {
      "entropy": 6.427644681930542,
      "epoch": 1.6720530793514903,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004727788333210305,
      "loss": 5.9819,
      "mean_token_accuracy": 0.1639208823442459,
      "num_tokens": 33801195.0,
      "step": 15625
    },
    {
      "entropy": 6.469254016876221,
      "epoch": 1.6725881534592542,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472760601168495,
      "loss": 5.9186,
      "mean_token_accuracy": 0.16766682267189026,
      "num_tokens": 33812957.0,
      "step": 15630
    },
    {
      "entropy": 6.483545112609863,
      "epoch": 1.673123227567018,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047274236330564995,
      "loss": 6.0108,
      "mean_token_accuracy": 0.15684265047311782,
      "num_tokens": 33823874.0,
      "step": 15635
    },
    {
      "entropy": 6.462090587615966,
      "epoch": 1.673658301674782,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004727241197330219,
      "loss": 5.9758,
      "mean_token_accuracy": 0.16183445751667022,
      "num_tokens": 33834284.0,
      "step": 15640
    },
    {
      "entropy": 6.492154312133789,
      "epoch": 1.6741933757825458,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004727058704511376,
      "loss": 6.0053,
      "mean_token_accuracy": 0.1567127987742424,
      "num_tokens": 33845544.0,
      "step": 15645
    },
    {
      "entropy": 6.523851633071899,
      "epoch": 1.67472844989031,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004726876154605242,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15711392164230348,
      "num_tokens": 33855420.0,
      "step": 15650
    },
    {
      "entropy": 6.4521204948425295,
      "epoch": 1.6752635239980738,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004726693547617086,
      "loss": 5.869,
      "mean_token_accuracy": 0.1595511481165886,
      "num_tokens": 33864904.0,
      "step": 15655
    },
    {
      "entropy": 6.4311933517456055,
      "epoch": 1.6757985981058376,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047265108835521825,
      "loss": 5.9279,
      "mean_token_accuracy": 0.15561299473047258,
      "num_tokens": 33875139.0,
      "step": 15660
    },
    {
      "entropy": 6.463284587860107,
      "epoch": 1.6763336722136017,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004726328162415804,
      "loss": 5.949,
      "mean_token_accuracy": 0.16171044409275054,
      "num_tokens": 33886401.0,
      "step": 15665
    },
    {
      "entropy": 6.471650552749634,
      "epoch": 1.6768687463213654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047261453842132277,
      "loss": 5.9919,
      "mean_token_accuracy": 0.16129094809293748,
      "num_tokens": 33896382.0,
      "step": 15670
    },
    {
      "entropy": 6.495286989212036,
      "epoch": 1.6774038204291295,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00047259625489497317,
      "loss": 5.9682,
      "mean_token_accuracy": 0.16032076627016068,
      "num_tokens": 33906403.0,
      "step": 15675
    },
    {
      "entropy": 6.354872369766236,
      "epoch": 1.6779388945368934,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047257796566305946,
      "loss": 5.863,
      "mean_token_accuracy": 0.15996704548597335,
      "num_tokens": 33916992.0,
      "step": 15680
    },
    {
      "entropy": 6.475168991088867,
      "epoch": 1.6784739686446573,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004725596707261097,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15472015142440795,
      "num_tokens": 33927599.0,
      "step": 15685
    },
    {
      "entropy": 6.517701005935669,
      "epoch": 1.6790090427524214,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004725413700846522,
      "loss": 6.0747,
      "mean_token_accuracy": 0.1498353511095047,
      "num_tokens": 33938957.0,
      "step": 15690
    },
    {
      "entropy": 6.503129196166992,
      "epoch": 1.679544116860185,
      "grad_norm": 0.8984375,
      "learning_rate": 0.00047252306373921543,
      "loss": 5.955,
      "mean_token_accuracy": 0.16533460319042206,
      "num_tokens": 33950169.0,
      "step": 15695
    },
    {
      "entropy": 6.479008054733276,
      "epoch": 1.680079190967949,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047250475169032804,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15635529309511184,
      "num_tokens": 33960800.0,
      "step": 15700
    },
    {
      "entropy": 6.466224002838135,
      "epoch": 1.680614265075713,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004724864339385185,
      "loss": 6.0396,
      "mean_token_accuracy": 0.14939001053571702,
      "num_tokens": 33973060.0,
      "step": 15705
    },
    {
      "entropy": 6.524610948562622,
      "epoch": 1.6811493391834769,
      "grad_norm": 1.0,
      "learning_rate": 0.0004724681104843161,
      "loss": 6.0531,
      "mean_token_accuracy": 0.15138935223221778,
      "num_tokens": 33983185.0,
      "step": 15710
    },
    {
      "entropy": 6.483154249191284,
      "epoch": 1.681684413291241,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004724497813282497,
      "loss": 5.9995,
      "mean_token_accuracy": 0.1563175469636917,
      "num_tokens": 33993373.0,
      "step": 15715
    },
    {
      "entropy": 6.350405740737915,
      "epoch": 1.6822194873990046,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004724314464708486,
      "loss": 5.8742,
      "mean_token_accuracy": 0.1722984567284584,
      "num_tokens": 34005793.0,
      "step": 15720
    },
    {
      "entropy": 6.459917259216309,
      "epoch": 1.6827545615067687,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004724131059126423,
      "loss": 6.0357,
      "mean_token_accuracy": 0.14798771142959594,
      "num_tokens": 34016415.0,
      "step": 15725
    },
    {
      "entropy": 6.446478652954101,
      "epoch": 1.6832896356145326,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047239475965416016,
      "loss": 5.9179,
      "mean_token_accuracy": 0.15441478043794632,
      "num_tokens": 34026859.0,
      "step": 15730
    },
    {
      "entropy": 6.475410747528076,
      "epoch": 1.6838247097222965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047237640769593224,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1763911470770836,
      "num_tokens": 34037942.0,
      "step": 15735
    },
    {
      "entropy": 6.4168332576751705,
      "epoch": 1.6843597838300606,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004723580500384882,
      "loss": 5.9089,
      "mean_token_accuracy": 0.1632147029042244,
      "num_tokens": 34048659.0,
      "step": 15740
    },
    {
      "entropy": 6.409747982025147,
      "epoch": 1.6848948579378242,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047233968668235817,
      "loss": 5.9396,
      "mean_token_accuracy": 0.16041334718465805,
      "num_tokens": 34059961.0,
      "step": 15745
    },
    {
      "entropy": 6.445631837844848,
      "epoch": 1.6854299320455883,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047232131762807244,
      "loss": 5.9746,
      "mean_token_accuracy": 0.16187651008367537,
      "num_tokens": 34069978.0,
      "step": 15750
    },
    {
      "entropy": 6.479240798950196,
      "epoch": 1.6859650061533522,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004723029428761614,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1562919795513153,
      "num_tokens": 34080301.0,
      "step": 15755
    },
    {
      "entropy": 6.424789953231811,
      "epoch": 1.686500080261116,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047228456242715564,
      "loss": 5.9738,
      "mean_token_accuracy": 0.15887757539749145,
      "num_tokens": 34092376.0,
      "step": 15760
    },
    {
      "entropy": 6.4707996368408205,
      "epoch": 1.6870351543688802,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004722661762815858,
      "loss": 5.9462,
      "mean_token_accuracy": 0.16000592708587646,
      "num_tokens": 34102761.0,
      "step": 15765
    },
    {
      "entropy": 6.448899650573731,
      "epoch": 1.687570228476644,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004722477844399829,
      "loss": 5.9611,
      "mean_token_accuracy": 0.149532812833786,
      "num_tokens": 34113370.0,
      "step": 15770
    },
    {
      "entropy": 6.459095668792725,
      "epoch": 1.688105302584408,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004722293869028779,
      "loss": 5.9586,
      "mean_token_accuracy": 0.16097499132156373,
      "num_tokens": 34125038.0,
      "step": 15775
    },
    {
      "entropy": 6.50250768661499,
      "epoch": 1.6886403766921718,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004722109836708021,
      "loss": 6.004,
      "mean_token_accuracy": 0.15087870657444,
      "num_tokens": 34135595.0,
      "step": 15780
    },
    {
      "entropy": 6.460846662521362,
      "epoch": 1.6891754507999357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047219257474428685,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15211975276470185,
      "num_tokens": 34146755.0,
      "step": 15785
    },
    {
      "entropy": 6.370594167709351,
      "epoch": 1.6897105249076998,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047217416012386367,
      "loss": 5.8432,
      "mean_token_accuracy": 0.17322086542844772,
      "num_tokens": 34158465.0,
      "step": 15790
    },
    {
      "entropy": 6.455634307861328,
      "epoch": 1.6902455990154637,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004721557398100644,
      "loss": 5.955,
      "mean_token_accuracy": 0.1572364404797554,
      "num_tokens": 34170007.0,
      "step": 15795
    },
    {
      "entropy": 6.486449813842773,
      "epoch": 1.6907806731232276,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047213731380342077,
      "loss": 6.018,
      "mean_token_accuracy": 0.153550586104393,
      "num_tokens": 34181595.0,
      "step": 15800
    },
    {
      "entropy": 6.473370456695557,
      "epoch": 1.6913157472309917,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000472118882104465,
      "loss": 5.9852,
      "mean_token_accuracy": 0.158478182554245,
      "num_tokens": 34191791.0,
      "step": 15805
    },
    {
      "entropy": 6.505572175979614,
      "epoch": 1.6918508213387553,
      "grad_norm": 1.0,
      "learning_rate": 0.00047210044471372915,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16093382388353347,
      "num_tokens": 34202636.0,
      "step": 15810
    },
    {
      "entropy": 6.515819120407104,
      "epoch": 1.6923858954465194,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004720820016317456,
      "loss": 6.0406,
      "mean_token_accuracy": 0.14816571697592734,
      "num_tokens": 34213333.0,
      "step": 15815
    },
    {
      "entropy": 6.496355581283569,
      "epoch": 1.6929209695542833,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000472063552859047,
      "loss": 5.9822,
      "mean_token_accuracy": 0.15828276723623275,
      "num_tokens": 34223679.0,
      "step": 15820
    },
    {
      "entropy": 6.496370458602906,
      "epoch": 1.6934560436620472,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047204509839616604,
      "loss": 6.0989,
      "mean_token_accuracy": 0.14664506688714027,
      "num_tokens": 34234528.0,
      "step": 15825
    },
    {
      "entropy": 6.477035617828369,
      "epoch": 1.6939911177698113,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004720266382436355,
      "loss": 5.921,
      "mean_token_accuracy": 0.16284967362880706,
      "num_tokens": 34246203.0,
      "step": 15830
    },
    {
      "entropy": 6.49503870010376,
      "epoch": 1.694526191877575,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00047200817240198846,
      "loss": 5.9259,
      "mean_token_accuracy": 0.15355359464883805,
      "num_tokens": 34257231.0,
      "step": 15835
    },
    {
      "entropy": 6.46951642036438,
      "epoch": 1.695061265985339,
      "grad_norm": 0.984375,
      "learning_rate": 0.00047198970087175817,
      "loss": 5.9412,
      "mean_token_accuracy": 0.1575847715139389,
      "num_tokens": 34268900.0,
      "step": 15840
    },
    {
      "entropy": 6.478216648101807,
      "epoch": 1.695596340093103,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004719712236534779,
      "loss": 6.0297,
      "mean_token_accuracy": 0.15314692109823227,
      "num_tokens": 34279343.0,
      "step": 15845
    },
    {
      "entropy": 6.467459774017334,
      "epoch": 1.6961314142008668,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004719527407476812,
      "loss": 5.9612,
      "mean_token_accuracy": 0.16561685055494307,
      "num_tokens": 34291952.0,
      "step": 15850
    },
    {
      "entropy": 6.487732887268066,
      "epoch": 1.6966664883086309,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047193425215490186,
      "loss": 5.9828,
      "mean_token_accuracy": 0.15159825533628463,
      "num_tokens": 34302516.0,
      "step": 15855
    },
    {
      "entropy": 6.456855297088623,
      "epoch": 1.6972015624163945,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00047191575787567373,
      "loss": 5.9067,
      "mean_token_accuracy": 0.16178855895996094,
      "num_tokens": 34313094.0,
      "step": 15860
    },
    {
      "entropy": 6.517060708999634,
      "epoch": 1.6977366365241586,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004718972579105306,
      "loss": 6.0355,
      "mean_token_accuracy": 0.15415024310350417,
      "num_tokens": 34324803.0,
      "step": 15865
    },
    {
      "entropy": 6.4638875961303714,
      "epoch": 1.6982717106319225,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004718787522600069,
      "loss": 5.9917,
      "mean_token_accuracy": 0.1525728940963745,
      "num_tokens": 34336386.0,
      "step": 15870
    },
    {
      "entropy": 6.482392597198486,
      "epoch": 1.6988067847396864,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004718602409246369,
      "loss": 6.0062,
      "mean_token_accuracy": 0.15308922678232192,
      "num_tokens": 34347085.0,
      "step": 15875
    },
    {
      "entropy": 6.43738021850586,
      "epoch": 1.6993418588474505,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047184172390495505,
      "loss": 5.9501,
      "mean_token_accuracy": 0.1651800572872162,
      "num_tokens": 34358146.0,
      "step": 15880
    },
    {
      "entropy": 6.463834524154663,
      "epoch": 1.6998769329552141,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004718232012014961,
      "loss": 5.9675,
      "mean_token_accuracy": 0.15645352229475976,
      "num_tokens": 34370403.0,
      "step": 15885
    },
    {
      "entropy": 6.50139856338501,
      "epoch": 1.7004120070629782,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00047180467281479496,
      "loss": 5.9656,
      "mean_token_accuracy": 0.16234969198703766,
      "num_tokens": 34381130.0,
      "step": 15890
    },
    {
      "entropy": 6.452266550064087,
      "epoch": 1.7009470811707421,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004717861387453865,
      "loss": 5.998,
      "mean_token_accuracy": 0.15859352499246598,
      "num_tokens": 34391610.0,
      "step": 15895
    },
    {
      "entropy": 6.479373550415039,
      "epoch": 1.701482155278506,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004717675989938059,
      "loss": 5.9803,
      "mean_token_accuracy": 0.15696441233158112,
      "num_tokens": 34402262.0,
      "step": 15900
    },
    {
      "entropy": 6.483989191055298,
      "epoch": 1.70201722938627,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00047174905356058857,
      "loss": 5.9741,
      "mean_token_accuracy": 0.15317117869853974,
      "num_tokens": 34412707.0,
      "step": 15905
    },
    {
      "entropy": 6.440914726257324,
      "epoch": 1.702552303494034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047173050244626993,
      "loss": 5.902,
      "mean_token_accuracy": 0.1655367910861969,
      "num_tokens": 34423203.0,
      "step": 15910
    },
    {
      "entropy": 6.460349702835083,
      "epoch": 1.7030873776017978,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004717119456513857,
      "loss": 6.0037,
      "mean_token_accuracy": 0.15801714807748796,
      "num_tokens": 34433279.0,
      "step": 15915
    },
    {
      "entropy": 6.482739400863648,
      "epoch": 1.7036224517095617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00047169338317647164,
      "loss": 6.0414,
      "mean_token_accuracy": 0.15491918325424195,
      "num_tokens": 34444601.0,
      "step": 15920
    },
    {
      "entropy": 6.483644151687622,
      "epoch": 1.7041575258173256,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047167481502206383,
      "loss": 5.9793,
      "mean_token_accuracy": 0.1609325498342514,
      "num_tokens": 34455103.0,
      "step": 15925
    },
    {
      "entropy": 6.445603609085083,
      "epoch": 1.7046925999250897,
      "grad_norm": 1.125,
      "learning_rate": 0.00047165624118869834,
      "loss": 6.0028,
      "mean_token_accuracy": 0.1548689715564251,
      "num_tokens": 34465835.0,
      "step": 15930
    },
    {
      "entropy": 6.54382472038269,
      "epoch": 1.7052276740328536,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004716376616769115,
      "loss": 6.0473,
      "mean_token_accuracy": 0.15066453218460082,
      "num_tokens": 34476602.0,
      "step": 15935
    },
    {
      "entropy": 6.51798996925354,
      "epoch": 1.7057627481406175,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004716190764872399,
      "loss": 5.9234,
      "mean_token_accuracy": 0.1533742681145668,
      "num_tokens": 34488026.0,
      "step": 15940
    },
    {
      "entropy": 6.468924903869629,
      "epoch": 1.7062978222483816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00047160048562022006,
      "loss": 6.0231,
      "mean_token_accuracy": 0.16374888122081757,
      "num_tokens": 34497837.0,
      "step": 15945
    },
    {
      "entropy": 6.434396171569825,
      "epoch": 1.7068328963561452,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004715818890763888,
      "loss": 5.9713,
      "mean_token_accuracy": 0.16058547496795655,
      "num_tokens": 34507727.0,
      "step": 15950
    },
    {
      "entropy": 6.475588417053222,
      "epoch": 1.7073679704639093,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047156328685628307,
      "loss": 5.9661,
      "mean_token_accuracy": 0.15381831526756287,
      "num_tokens": 34518595.0,
      "step": 15955
    },
    {
      "entropy": 6.453253316879272,
      "epoch": 1.7079030445716732,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004715446789604401,
      "loss": 5.9037,
      "mean_token_accuracy": 0.16315882354974748,
      "num_tokens": 34529415.0,
      "step": 15960
    },
    {
      "entropy": 6.443467044830323,
      "epoch": 1.708438118679437,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004715260653893972,
      "loss": 5.9458,
      "mean_token_accuracy": 0.16361862868070604,
      "num_tokens": 34539593.0,
      "step": 15965
    },
    {
      "entropy": 6.500147294998169,
      "epoch": 1.7089731927872012,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004715074461436917,
      "loss": 5.9615,
      "mean_token_accuracy": 0.15671059489250183,
      "num_tokens": 34549441.0,
      "step": 15970
    },
    {
      "entropy": 6.469748401641846,
      "epoch": 1.7095082668949648,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047148882122386126,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16266058683395385,
      "num_tokens": 34560211.0,
      "step": 15975
    },
    {
      "entropy": 6.455126523971558,
      "epoch": 1.710043341002729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047147019063044387,
      "loss": 6.0104,
      "mean_token_accuracy": 0.15459270179271697,
      "num_tokens": 34570479.0,
      "step": 15980
    },
    {
      "entropy": 6.491168022155762,
      "epoch": 1.7105784151104928,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047145155436397726,
      "loss": 5.9931,
      "mean_token_accuracy": 0.15780818611383438,
      "num_tokens": 34581382.0,
      "step": 15985
    },
    {
      "entropy": 6.461596059799194,
      "epoch": 1.7111134892182567,
      "grad_norm": 1.0,
      "learning_rate": 0.0004714329124249996,
      "loss": 5.9144,
      "mean_token_accuracy": 0.16592525988817214,
      "num_tokens": 34592131.0,
      "step": 15990
    },
    {
      "entropy": 6.470749282836914,
      "epoch": 1.7116485633260208,
      "grad_norm": 0.89453125,
      "learning_rate": 0.0004714142648140492,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15599738657474518,
      "num_tokens": 34604407.0,
      "step": 15995
    },
    {
      "entropy": 6.417363834381104,
      "epoch": 1.7121836374337844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004713956115316645,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1593409851193428,
      "num_tokens": 34614842.0,
      "step": 16000
    },
    {
      "entropy": 6.358746671676636,
      "epoch": 1.7127187115415485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004713769525783841,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16657116562128066,
      "num_tokens": 34625991.0,
      "step": 16005
    },
    {
      "entropy": 6.429520034790039,
      "epoch": 1.7132537856493124,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004713582879547469,
      "loss": 6.0138,
      "mean_token_accuracy": 0.1556413874030113,
      "num_tokens": 34637317.0,
      "step": 16010
    },
    {
      "entropy": 6.498357105255127,
      "epoch": 1.7137888597570763,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004713396176612916,
      "loss": 6.0051,
      "mean_token_accuracy": 0.15320713222026824,
      "num_tokens": 34648282.0,
      "step": 16015
    },
    {
      "entropy": 6.489234495162964,
      "epoch": 1.7143239338648404,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004713209416985575,
      "loss": 5.9671,
      "mean_token_accuracy": 0.15362469106912613,
      "num_tokens": 34658638.0,
      "step": 16020
    },
    {
      "entropy": 6.45802149772644,
      "epoch": 1.714859007972604,
      "grad_norm": 1.046875,
      "learning_rate": 0.00047130226006708375,
      "loss": 5.9922,
      "mean_token_accuracy": 0.1530597150325775,
      "num_tokens": 34669308.0,
      "step": 16025
    },
    {
      "entropy": 6.427453994750977,
      "epoch": 1.7153940820803681,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047128357276740986,
      "loss": 5.9687,
      "mean_token_accuracy": 0.15341118723154068,
      "num_tokens": 34679513.0,
      "step": 16030
    },
    {
      "entropy": 6.477577209472656,
      "epoch": 1.715929156188132,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004712648798000754,
      "loss": 5.976,
      "mean_token_accuracy": 0.15652416199445723,
      "num_tokens": 34690736.0,
      "step": 16035
    },
    {
      "entropy": 6.488976669311524,
      "epoch": 1.716464230295896,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047124618116562003,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15908450037240982,
      "num_tokens": 34701272.0,
      "step": 16040
    },
    {
      "entropy": 6.404465675354004,
      "epoch": 1.71699930440366,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004712274768645839,
      "loss": 5.866,
      "mean_token_accuracy": 0.1668838694691658,
      "num_tokens": 34712486.0,
      "step": 16045
    },
    {
      "entropy": 6.487472867965698,
      "epoch": 1.7175343785114239,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00047120876689750685,
      "loss": 5.9596,
      "mean_token_accuracy": 0.15615903735160827,
      "num_tokens": 34721979.0,
      "step": 16050
    },
    {
      "entropy": 6.429824686050415,
      "epoch": 1.7180694526191878,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004711900512649292,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15977274924516677,
      "num_tokens": 34733496.0,
      "step": 16055
    },
    {
      "entropy": 6.503298950195313,
      "epoch": 1.7186045267269516,
      "grad_norm": 1.0,
      "learning_rate": 0.0004711713299673914,
      "loss": 5.9691,
      "mean_token_accuracy": 0.1630312368273735,
      "num_tokens": 34744223.0,
      "step": 16060
    },
    {
      "entropy": 6.476534605026245,
      "epoch": 1.7191396008347155,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047115260300543407,
      "loss": 6.0061,
      "mean_token_accuracy": 0.15955363363027572,
      "num_tokens": 34754931.0,
      "step": 16065
    },
    {
      "entropy": 6.401424837112427,
      "epoch": 1.7196746749424796,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004711338703795978,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16308656036853791,
      "num_tokens": 34766285.0,
      "step": 16070
    },
    {
      "entropy": 6.451199674606324,
      "epoch": 1.7202097490502435,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004711151320904237,
      "loss": 5.9616,
      "mean_token_accuracy": 0.16261016875505446,
      "num_tokens": 34776979.0,
      "step": 16075
    },
    {
      "entropy": 6.533099222183227,
      "epoch": 1.7207448231580074,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004710963881384527,
      "loss": 6.0383,
      "mean_token_accuracy": 0.14877405613660813,
      "num_tokens": 34787466.0,
      "step": 16080
    },
    {
      "entropy": 6.525169801712036,
      "epoch": 1.7212798972657715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000471077638524226,
      "loss": 5.94,
      "mean_token_accuracy": 0.16005912572145461,
      "num_tokens": 34797664.0,
      "step": 16085
    },
    {
      "entropy": 6.384826612472534,
      "epoch": 1.7218149713735351,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047105888324828506,
      "loss": 5.9294,
      "mean_token_accuracy": 0.16918128430843354,
      "num_tokens": 34809082.0,
      "step": 16090
    },
    {
      "entropy": 6.45182433128357,
      "epoch": 1.7223500454812992,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047104012231117143,
      "loss": 5.9584,
      "mean_token_accuracy": 0.15742510557174683,
      "num_tokens": 34820223.0,
      "step": 16095
    },
    {
      "entropy": 6.398351240158081,
      "epoch": 1.722885119589063,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047102135571342683,
      "loss": 5.8726,
      "mean_token_accuracy": 0.15768765807151794,
      "num_tokens": 34830428.0,
      "step": 16100
    },
    {
      "entropy": 6.430466175079346,
      "epoch": 1.723420193696827,
      "grad_norm": 1.0,
      "learning_rate": 0.00047100258345559307,
      "loss": 5.9899,
      "mean_token_accuracy": 0.15215041488409042,
      "num_tokens": 34842787.0,
      "step": 16105
    },
    {
      "entropy": 6.478031826019287,
      "epoch": 1.723955267804591,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004709838055382123,
      "loss": 5.9382,
      "mean_token_accuracy": 0.16006767749786377,
      "num_tokens": 34854323.0,
      "step": 16110
    },
    {
      "entropy": 6.462150955200196,
      "epoch": 1.7244903419123547,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004709650219618266,
      "loss": 5.9122,
      "mean_token_accuracy": 0.16196918487548828,
      "num_tokens": 34865279.0,
      "step": 16115
    },
    {
      "entropy": 6.445090484619141,
      "epoch": 1.7250254160201188,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004709462327269785,
      "loss": 5.9505,
      "mean_token_accuracy": 0.15675307512283326,
      "num_tokens": 34875951.0,
      "step": 16120
    },
    {
      "entropy": 6.410007619857788,
      "epoch": 1.7255604901278827,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004709274378342105,
      "loss": 5.9722,
      "mean_token_accuracy": 0.1619311600923538,
      "num_tokens": 34885815.0,
      "step": 16125
    },
    {
      "entropy": 6.480797910690308,
      "epoch": 1.7260955642356466,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00047090863728406517,
      "loss": 5.9787,
      "mean_token_accuracy": 0.1549619808793068,
      "num_tokens": 34897158.0,
      "step": 16130
    },
    {
      "entropy": 6.4970627784729,
      "epoch": 1.7266306383434107,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004708898310770856,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15311579555273055,
      "num_tokens": 34907815.0,
      "step": 16135
    },
    {
      "entropy": 6.458138275146484,
      "epoch": 1.7271657124511743,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004708710192138146,
      "loss": 5.9476,
      "mean_token_accuracy": 0.1650984063744545,
      "num_tokens": 34918820.0,
      "step": 16140
    },
    {
      "entropy": 6.511193132400512,
      "epoch": 1.7277007865589384,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004708522016947954,
      "loss": 6.0195,
      "mean_token_accuracy": 0.15626926571130753,
      "num_tokens": 34929998.0,
      "step": 16145
    },
    {
      "entropy": 6.419228363037109,
      "epoch": 1.7282358606667023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004708333785205715,
      "loss": 5.9105,
      "mean_token_accuracy": 0.17040863037109374,
      "num_tokens": 34940989.0,
      "step": 16150
    },
    {
      "entropy": 6.478556346893311,
      "epoch": 1.7287709347744662,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004708145496916862,
      "loss": 5.9435,
      "mean_token_accuracy": 0.15807853937149047,
      "num_tokens": 34952433.0,
      "step": 16155
    },
    {
      "entropy": 6.491824388504028,
      "epoch": 1.7293060088822303,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047079571520868344,
      "loss": 6.0027,
      "mean_token_accuracy": 0.1559208333492279,
      "num_tokens": 34963074.0,
      "step": 16160
    },
    {
      "entropy": 6.41399998664856,
      "epoch": 1.729841082989994,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00047077687507210674,
      "loss": 5.9249,
      "mean_token_accuracy": 0.16185495257377625,
      "num_tokens": 34974315.0,
      "step": 16165
    },
    {
      "entropy": 6.423493242263794,
      "epoch": 1.730376157097758,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004707580292825004,
      "loss": 5.8868,
      "mean_token_accuracy": 0.174258241802454,
      "num_tokens": 34985059.0,
      "step": 16170
    },
    {
      "entropy": 6.504748773574829,
      "epoch": 1.730911231205522,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004707391778404085,
      "loss": 6.0454,
      "mean_token_accuracy": 0.14586610198020936,
      "num_tokens": 34996156.0,
      "step": 16175
    },
    {
      "entropy": 6.455451440811157,
      "epoch": 1.7314463053132858,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004707203207463752,
      "loss": 5.958,
      "mean_token_accuracy": 0.16190685629844664,
      "num_tokens": 35007410.0,
      "step": 16180
    },
    {
      "entropy": 6.4828263282775875,
      "epoch": 1.73198137942105,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004707014580009453,
      "loss": 5.948,
      "mean_token_accuracy": 0.16246227771043778,
      "num_tokens": 35018492.0,
      "step": 16185
    },
    {
      "entropy": 6.490396404266358,
      "epoch": 1.7325164535288138,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004706825896046632,
      "loss": 5.9951,
      "mean_token_accuracy": 0.1559868112206459,
      "num_tokens": 35029923.0,
      "step": 16190
    },
    {
      "entropy": 6.425391912460327,
      "epoch": 1.7330515276365777,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047066371555807386,
      "loss": 5.921,
      "mean_token_accuracy": 0.15621594041585923,
      "num_tokens": 35040370.0,
      "step": 16195
    },
    {
      "entropy": 6.509308624267578,
      "epoch": 1.7335866017443415,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004706448358617222,
      "loss": 5.9784,
      "mean_token_accuracy": 0.15632479786872863,
      "num_tokens": 35051414.0,
      "step": 16200
    },
    {
      "entropy": 6.522961854934692,
      "epoch": 1.7341216758521054,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00047062595051615343,
      "loss": 5.9318,
      "mean_token_accuracy": 0.16424416452646257,
      "num_tokens": 35061015.0,
      "step": 16205
    },
    {
      "entropy": 6.460383367538452,
      "epoch": 1.7346567499598695,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004706070595219128,
      "loss": 6.0226,
      "mean_token_accuracy": 0.1559004969894886,
      "num_tokens": 35071757.0,
      "step": 16210
    },
    {
      "entropy": 6.477738666534424,
      "epoch": 1.7351918240676334,
      "grad_norm": 0.96875,
      "learning_rate": 0.00047058816287954587,
      "loss": 5.9967,
      "mean_token_accuracy": 0.15721683502197265,
      "num_tokens": 35082912.0,
      "step": 16215
    },
    {
      "entropy": 6.47300615310669,
      "epoch": 1.7357268981753973,
      "grad_norm": 0.953125,
      "learning_rate": 0.00047056926058959813,
      "loss": 5.943,
      "mean_token_accuracy": 0.16645231246948242,
      "num_tokens": 35094103.0,
      "step": 16220
    },
    {
      "entropy": 6.471975708007813,
      "epoch": 1.7362619722831614,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004705503526526155,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16652375608682632,
      "num_tokens": 35105960.0,
      "step": 16225
    },
    {
      "entropy": 6.377423048019409,
      "epoch": 1.736797046390925,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00047053143906914405,
      "loss": 5.9171,
      "mean_token_accuracy": 0.16451659500598909,
      "num_tokens": 35116482.0,
      "step": 16230
    },
    {
      "entropy": 6.450377321243286,
      "epoch": 1.7373321204986891,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047051251983972966,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15620717108249665,
      "num_tokens": 35127976.0,
      "step": 16235
    },
    {
      "entropy": 6.474288511276245,
      "epoch": 1.737867194606453,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004704935949649187,
      "loss": 5.9888,
      "mean_token_accuracy": 0.15860860645771027,
      "num_tokens": 35137712.0,
      "step": 16240
    },
    {
      "entropy": 6.404179048538208,
      "epoch": 1.7384022687142169,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004704746644452577,
      "loss": 5.8965,
      "mean_token_accuracy": 0.1650048702955246,
      "num_tokens": 35148524.0,
      "step": 16245
    },
    {
      "entropy": 6.461119890213013,
      "epoch": 1.738937342821981,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00047045572828129323,
      "loss": 5.9748,
      "mean_token_accuracy": 0.15978893935680388,
      "num_tokens": 35159080.0,
      "step": 16250
    },
    {
      "entropy": 6.339294528961181,
      "epoch": 1.7394724169297446,
      "grad_norm": 1.046875,
      "learning_rate": 0.000470436786473572,
      "loss": 5.8941,
      "mean_token_accuracy": 0.1617966428399086,
      "num_tokens": 35169993.0,
      "step": 16255
    },
    {
      "entropy": 6.412735891342163,
      "epoch": 1.7400074910375087,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004704178390226411,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16028994023799897,
      "num_tokens": 35181413.0,
      "step": 16260
    },
    {
      "entropy": 6.441563701629638,
      "epoch": 1.7405425651452726,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00047039888592904753,
      "loss": 5.9246,
      "mean_token_accuracy": 0.1637858510017395,
      "num_tokens": 35191600.0,
      "step": 16265
    },
    {
      "entropy": 6.479606294631958,
      "epoch": 1.7410776392530365,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047037992719333864,
      "loss": 5.9565,
      "mean_token_accuracy": 0.15645871609449385,
      "num_tokens": 35202535.0,
      "step": 16270
    },
    {
      "entropy": 6.45976619720459,
      "epoch": 1.7416127133608006,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00047036096281606176,
      "loss": 5.9236,
      "mean_token_accuracy": 0.15397179275751113,
      "num_tokens": 35214798.0,
      "step": 16275
    },
    {
      "entropy": 6.427538681030273,
      "epoch": 1.7421477874685642,
      "grad_norm": 1.109375,
      "learning_rate": 0.00047034199279776456,
      "loss": 5.9334,
      "mean_token_accuracy": 0.15961792021989823,
      "num_tokens": 35224870.0,
      "step": 16280
    },
    {
      "entropy": 6.446291828155518,
      "epoch": 1.7426828615763283,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004703230171389948,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15565575510263444,
      "num_tokens": 35235408.0,
      "step": 16285
    },
    {
      "entropy": 6.416685676574707,
      "epoch": 1.7432179356840922,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047030403584030024,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1563032165169716,
      "num_tokens": 35246310.0,
      "step": 16290
    },
    {
      "entropy": 6.446374130249024,
      "epoch": 1.743753009791856,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004702850489022292,
      "loss": 5.9587,
      "mean_token_accuracy": 0.15768592655658722,
      "num_tokens": 35257027.0,
      "step": 16295
    },
    {
      "entropy": 6.527727746963501,
      "epoch": 1.7442880838996202,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00047026605632532976,
      "loss": 5.9363,
      "mean_token_accuracy": 0.16405781507492065,
      "num_tokens": 35267397.0,
      "step": 16300
    },
    {
      "entropy": 6.376235580444336,
      "epoch": 1.7448231580073839,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00047024705811015045,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16150096356868743,
      "num_tokens": 35278775.0,
      "step": 16305
    },
    {
      "entropy": 6.368799257278442,
      "epoch": 1.745358232115148,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004702280542572397,
      "loss": 5.8362,
      "mean_token_accuracy": 0.1673252359032631,
      "num_tokens": 35288818.0,
      "step": 16310
    },
    {
      "entropy": 6.493281650543213,
      "epoch": 1.7458933062229118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004702090447671464,
      "loss": 6.0423,
      "mean_token_accuracy": 0.15007688701152802,
      "num_tokens": 35299385.0,
      "step": 16315
    },
    {
      "entropy": 6.477223634719849,
      "epoch": 1.7464283803306757,
      "grad_norm": 1.0,
      "learning_rate": 0.0004701900296404193,
      "loss": 5.9206,
      "mean_token_accuracy": 0.16464000344276428,
      "num_tokens": 35310626.0,
      "step": 16320
    },
    {
      "entropy": 6.476791572570801,
      "epoch": 1.7469634544384398,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004701710088776075,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1556292936205864,
      "num_tokens": 35321615.0,
      "step": 16325
    },
    {
      "entropy": 6.502606821060181,
      "epoch": 1.7474985285462037,
      "grad_norm": 1.015625,
      "learning_rate": 0.00047015198247926025,
      "loss": 5.9968,
      "mean_token_accuracy": 0.1587006777524948,
      "num_tokens": 35332012.0,
      "step": 16330
    },
    {
      "entropy": 6.504772043228149,
      "epoch": 1.7480336026539676,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000470132950445927,
      "loss": 6.0264,
      "mean_token_accuracy": 0.14854292720556259,
      "num_tokens": 35342153.0,
      "step": 16335
    },
    {
      "entropy": 6.488902568817139,
      "epoch": 1.7485686767617317,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004701139127781571,
      "loss": 5.955,
      "mean_token_accuracy": 0.15843971073627472,
      "num_tokens": 35352858.0,
      "step": 16340
    },
    {
      "entropy": 6.474905061721802,
      "epoch": 1.7491037508694953,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004700948694765005,
      "loss": 5.8844,
      "mean_token_accuracy": 0.16186183393001558,
      "num_tokens": 35363489.0,
      "step": 16345
    },
    {
      "entropy": 6.500679969787598,
      "epoch": 1.7496388249772594,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004700758205415069,
      "loss": 6.0582,
      "mean_token_accuracy": 0.15439574122428895,
      "num_tokens": 35375816.0,
      "step": 16350
    },
    {
      "entropy": 6.4478759765625,
      "epoch": 1.7501738990850233,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00047005676597372645,
      "loss": 5.9664,
      "mean_token_accuracy": 0.15530956238508226,
      "num_tokens": 35386708.0,
      "step": 16355
    },
    {
      "entropy": 6.447930240631104,
      "epoch": 1.7507089731927872,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047003770577370925,
      "loss": 6.018,
      "mean_token_accuracy": 0.15138231068849564,
      "num_tokens": 35397734.0,
      "step": 16360
    },
    {
      "entropy": 6.542027950286865,
      "epoch": 1.7512440473005513,
      "grad_norm": 1.03125,
      "learning_rate": 0.00047001863994200574,
      "loss": 5.9417,
      "mean_token_accuracy": 0.1615550622344017,
      "num_tokens": 35408870.0,
      "step": 16365
    },
    {
      "entropy": 6.457405233383179,
      "epoch": 1.751779121408315,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004699995684791664,
      "loss": 5.8986,
      "mean_token_accuracy": 0.16148957014083862,
      "num_tokens": 35419498.0,
      "step": 16370
    },
    {
      "entropy": 6.395500993728637,
      "epoch": 1.752314195516079,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046998049138574177,
      "loss": 5.9276,
      "mean_token_accuracy": 0.16368987560272216,
      "num_tokens": 35430690.0,
      "step": 16375
    },
    {
      "entropy": 6.399061298370361,
      "epoch": 1.752849269623843,
      "grad_norm": 0.94921875,
      "learning_rate": 0.000469961408662283,
      "loss": 5.8853,
      "mean_token_accuracy": 0.16256203800439833,
      "num_tokens": 35441007.0,
      "step": 16380
    },
    {
      "entropy": 6.419111871719361,
      "epoch": 1.7533843437316068,
      "grad_norm": 0.984375,
      "learning_rate": 0.000469942320309341,
      "loss": 5.9547,
      "mean_token_accuracy": 0.16799886673688888,
      "num_tokens": 35453202.0,
      "step": 16385
    },
    {
      "entropy": 6.435385847091675,
      "epoch": 1.7539194178393709,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046992322632746677,
      "loss": 5.96,
      "mean_token_accuracy": 0.1629553869366646,
      "num_tokens": 35463303.0,
      "step": 16390
    },
    {
      "entropy": 6.489564275741577,
      "epoch": 1.7544544919471345,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046990412671721186,
      "loss": 5.9651,
      "mean_token_accuracy": 0.16163942068815232,
      "num_tokens": 35474104.0,
      "step": 16395
    },
    {
      "entropy": 6.476100111007691,
      "epoch": 1.7549895660548986,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004698850214791276,
      "loss": 5.9619,
      "mean_token_accuracy": 0.1586058147251606,
      "num_tokens": 35486412.0,
      "step": 16400
    },
    {
      "entropy": 6.450214910507202,
      "epoch": 1.7555246401626625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004698659106137658,
      "loss": 5.9278,
      "mean_token_accuracy": 0.15887346863746643,
      "num_tokens": 35497505.0,
      "step": 16405
    },
    {
      "entropy": 6.489379835128784,
      "epoch": 1.7560597142704264,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004698467941216782,
      "loss": 5.9818,
      "mean_token_accuracy": 0.16038403511047364,
      "num_tokens": 35508397.0,
      "step": 16410
    },
    {
      "entropy": 6.516765785217285,
      "epoch": 1.7565947883781905,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046982767200341675,
      "loss": 6.0263,
      "mean_token_accuracy": 0.15081835687160491,
      "num_tokens": 35520261.0,
      "step": 16415
    },
    {
      "entropy": 6.484202766418457,
      "epoch": 1.7571298624859542,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046980854425953365,
      "loss": 5.919,
      "mean_token_accuracy": 0.15565306097269058,
      "num_tokens": 35531816.0,
      "step": 16420
    },
    {
      "entropy": 6.535648488998413,
      "epoch": 1.7576649365937183,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004697894108905813,
      "loss": 6.0036,
      "mean_token_accuracy": 0.15688363760709761,
      "num_tokens": 35542233.0,
      "step": 16425
    },
    {
      "entropy": 6.439237785339356,
      "epoch": 1.7582000107014821,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046977027189711197,
      "loss": 5.938,
      "mean_token_accuracy": 0.15909286886453627,
      "num_tokens": 35552881.0,
      "step": 16430
    },
    {
      "entropy": 6.450898551940918,
      "epoch": 1.758735084809246,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004697511272796785,
      "loss": 5.8938,
      "mean_token_accuracy": 0.1622786119580269,
      "num_tokens": 35562392.0,
      "step": 16435
    },
    {
      "entropy": 6.496842288970948,
      "epoch": 1.75927015891701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004697319770388335,
      "loss": 5.9768,
      "mean_token_accuracy": 0.1569884166121483,
      "num_tokens": 35574587.0,
      "step": 16440
    },
    {
      "entropy": 6.4338541507720945,
      "epoch": 1.759805233024774,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004697128211751301,
      "loss": 5.854,
      "mean_token_accuracy": 0.16133675277233123,
      "num_tokens": 35585178.0,
      "step": 16445
    },
    {
      "entropy": 6.426873779296875,
      "epoch": 1.7603403071325379,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004696936596891213,
      "loss": 5.8974,
      "mean_token_accuracy": 0.15735740214586258,
      "num_tokens": 35595501.0,
      "step": 16450
    },
    {
      "entropy": 6.441808748245239,
      "epoch": 1.7608753812403017,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004696744925813605,
      "loss": 5.937,
      "mean_token_accuracy": 0.15726955533027648,
      "num_tokens": 35606794.0,
      "step": 16455
    },
    {
      "entropy": 6.478846454620362,
      "epoch": 1.7614104553480656,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004696553198524011,
      "loss": 5.9526,
      "mean_token_accuracy": 0.1618535429239273,
      "num_tokens": 35618732.0,
      "step": 16460
    },
    {
      "entropy": 6.471178007125855,
      "epoch": 1.7619455294558297,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004696361415027966,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16715379655361176,
      "num_tokens": 35629802.0,
      "step": 16465
    },
    {
      "entropy": 6.455761528015136,
      "epoch": 1.7624806035635936,
      "grad_norm": 1.0,
      "learning_rate": 0.0004696169575331009,
      "loss": 5.9983,
      "mean_token_accuracy": 0.1563188314437866,
      "num_tokens": 35640826.0,
      "step": 16470
    },
    {
      "entropy": 6.492450904846192,
      "epoch": 1.7630156776713575,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000469597767943868,
      "loss": 5.9909,
      "mean_token_accuracy": 0.1551734447479248,
      "num_tokens": 35651613.0,
      "step": 16475
    },
    {
      "entropy": 6.432581281661987,
      "epoch": 1.7635507517791216,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004695785727356517,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16087203323841096,
      "num_tokens": 35662334.0,
      "step": 16480
    },
    {
      "entropy": 6.492366313934326,
      "epoch": 1.7640858258868852,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004695593719090066,
      "loss": 6.0013,
      "mean_token_accuracy": 0.155238738656044,
      "num_tokens": 35673466.0,
      "step": 16485
    },
    {
      "entropy": 6.4545948028564455,
      "epoch": 1.7646208999946493,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004695401654644869,
      "loss": 5.9608,
      "mean_token_accuracy": 0.16082542538642883,
      "num_tokens": 35684708.0,
      "step": 16490
    },
    {
      "entropy": 6.392144012451172,
      "epoch": 1.7651559741024132,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046952095340264727,
      "loss": 5.8308,
      "mean_token_accuracy": 0.16100406795740127,
      "num_tokens": 35694888.0,
      "step": 16495
    },
    {
      "entropy": 6.490682601928711,
      "epoch": 1.765691048210177,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046950173572404255,
      "loss": 5.9043,
      "mean_token_accuracy": 0.155891115963459,
      "num_tokens": 35704844.0,
      "step": 16500
    },
    {
      "entropy": 6.43798828125,
      "epoch": 1.7662261223179412,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004694825124292275,
      "loss": 5.9569,
      "mean_token_accuracy": 0.16436180472373962,
      "num_tokens": 35715468.0,
      "step": 16505
    },
    {
      "entropy": 6.463957738876343,
      "epoch": 1.7667611964257048,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004694632835187572,
      "loss": 6.02,
      "mean_token_accuracy": 0.1557244300842285,
      "num_tokens": 35725542.0,
      "step": 16510
    },
    {
      "entropy": 6.499829483032227,
      "epoch": 1.767296270533469,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046944404899318685,
      "loss": 5.9335,
      "mean_token_accuracy": 0.16274511218070983,
      "num_tokens": 35737032.0,
      "step": 16515
    },
    {
      "entropy": 6.4897668838500975,
      "epoch": 1.7678313446412328,
      "grad_norm": 0.96484375,
      "learning_rate": 0.000469424808853072,
      "loss": 5.9215,
      "mean_token_accuracy": 0.16356155574321746,
      "num_tokens": 35747742.0,
      "step": 16520
    },
    {
      "entropy": 6.40181884765625,
      "epoch": 1.7683664187489967,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004694055630989681,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16019003242254257,
      "num_tokens": 35757795.0,
      "step": 16525
    },
    {
      "entropy": 6.442936611175537,
      "epoch": 1.7689014928567608,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004693863117314308,
      "loss": 5.9011,
      "mean_token_accuracy": 0.17112863659858704,
      "num_tokens": 35767962.0,
      "step": 16530
    },
    {
      "entropy": 6.4684117317199705,
      "epoch": 1.7694365669645244,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004693670547510161,
      "loss": 6.0065,
      "mean_token_accuracy": 0.15416226536035538,
      "num_tokens": 35778564.0,
      "step": 16535
    },
    {
      "entropy": 6.4753271579742435,
      "epoch": 1.7699716410722885,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046934779215828006,
      "loss": 5.9278,
      "mean_token_accuracy": 0.1685270681977272,
      "num_tokens": 35788417.0,
      "step": 16540
    },
    {
      "entropy": 6.415860366821289,
      "epoch": 1.7705067151800524,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004693285239537788,
      "loss": 5.9694,
      "mean_token_accuracy": 0.1547800451517105,
      "num_tokens": 35799724.0,
      "step": 16545
    },
    {
      "entropy": 6.480711269378662,
      "epoch": 1.7710417892878163,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046930925013806873,
      "loss": 5.8905,
      "mean_token_accuracy": 0.16989519000053405,
      "num_tokens": 35810306.0,
      "step": 16550
    },
    {
      "entropy": 6.5347288131713865,
      "epoch": 1.7715768633955804,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004692899707117064,
      "loss": 5.9364,
      "mean_token_accuracy": 0.1598309487104416,
      "num_tokens": 35821103.0,
      "step": 16555
    },
    {
      "entropy": 6.437076139450073,
      "epoch": 1.772111937503344,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004692706856752484,
      "loss": 5.9658,
      "mean_token_accuracy": 0.15920480340719223,
      "num_tokens": 35832271.0,
      "step": 16560
    },
    {
      "entropy": 6.46893630027771,
      "epoch": 1.7726470116111082,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046925139502925167,
      "loss": 5.9725,
      "mean_token_accuracy": 0.15434140264987944,
      "num_tokens": 35843875.0,
      "step": 16565
    },
    {
      "entropy": 6.483735179901123,
      "epoch": 1.773182085718872,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004692320987742732,
      "loss": 5.9701,
      "mean_token_accuracy": 0.15924373567104338,
      "num_tokens": 35854085.0,
      "step": 16570
    },
    {
      "entropy": 6.504604005813599,
      "epoch": 1.773717159826636,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004692127969108702,
      "loss": 5.9973,
      "mean_token_accuracy": 0.15758387595415116,
      "num_tokens": 35864777.0,
      "step": 16575
    },
    {
      "entropy": 6.4667411804199215,
      "epoch": 1.7742522339344,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004691934894396,
      "loss": 6.0145,
      "mean_token_accuracy": 0.15469406098127364,
      "num_tokens": 35876822.0,
      "step": 16580
    },
    {
      "entropy": 6.486543607711792,
      "epoch": 1.774787308042164,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046917417636102,
      "loss": 5.916,
      "mean_token_accuracy": 0.16214681416749954,
      "num_tokens": 35886997.0,
      "step": 16585
    },
    {
      "entropy": 6.479772090911865,
      "epoch": 1.7753223821499278,
      "grad_norm": 1.03125,
      "learning_rate": 0.000469154857675688,
      "loss": 5.9815,
      "mean_token_accuracy": 0.1602438986301422,
      "num_tokens": 35897871.0,
      "step": 16590
    },
    {
      "entropy": 6.4072075366973875,
      "epoch": 1.7758574562576916,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004691355333841617,
      "loss": 5.9831,
      "mean_token_accuracy": 0.1704223558306694,
      "num_tokens": 35908791.0,
      "step": 16595
    },
    {
      "entropy": 6.512836265563965,
      "epoch": 1.7763925303654555,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004691162034869993,
      "loss": 5.9905,
      "mean_token_accuracy": 0.157682141661644,
      "num_tokens": 35919436.0,
      "step": 16600
    },
    {
      "entropy": 6.524726867675781,
      "epoch": 1.7769276044732196,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004690968679847586,
      "loss": 5.8871,
      "mean_token_accuracy": 0.15885405689477922,
      "num_tokens": 35929851.0,
      "step": 16605
    },
    {
      "entropy": 6.483480787277221,
      "epoch": 1.7774626785809835,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004690775268779983,
      "loss": 5.9171,
      "mean_token_accuracy": 0.15991020649671556,
      "num_tokens": 35940271.0,
      "step": 16610
    },
    {
      "entropy": 6.4169879913330075,
      "epoch": 1.7779977526887474,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004690581801672765,
      "loss": 6.0403,
      "mean_token_accuracy": 0.15173082649707795,
      "num_tokens": 35951460.0,
      "step": 16615
    },
    {
      "entropy": 6.3976952075958256,
      "epoch": 1.7785328267965115,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046903882785315215,
      "loss": 5.9071,
      "mean_token_accuracy": 0.15945867747068404,
      "num_tokens": 35961530.0,
      "step": 16620
    },
    {
      "entropy": 6.43360710144043,
      "epoch": 1.7790679009042751,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004690194699361839,
      "loss": 5.8865,
      "mean_token_accuracy": 0.16908957660198212,
      "num_tokens": 35971085.0,
      "step": 16625
    },
    {
      "entropy": 6.443801069259644,
      "epoch": 1.7796029750120392,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046900010641693073,
      "loss": 5.9583,
      "mean_token_accuracy": 0.15444460585713388,
      "num_tokens": 35981393.0,
      "step": 16630
    },
    {
      "entropy": 6.432351875305176,
      "epoch": 1.7801380491198031,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004689807372959517,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16588265895843507,
      "num_tokens": 35992847.0,
      "step": 16635
    },
    {
      "entropy": 6.427000999450684,
      "epoch": 1.780673123227567,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046896136257380615,
      "loss": 5.9495,
      "mean_token_accuracy": 0.16047777235507965,
      "num_tokens": 36002991.0,
      "step": 16640
    },
    {
      "entropy": 6.42995662689209,
      "epoch": 1.781208197335331,
      "grad_norm": 0.91015625,
      "learning_rate": 0.00046894198225105356,
      "loss": 5.9571,
      "mean_token_accuracy": 0.16235849410295486,
      "num_tokens": 36015140.0,
      "step": 16645
    },
    {
      "entropy": 6.432117938995361,
      "epoch": 1.7817432714430947,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004689225963282534,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16186960488557817,
      "num_tokens": 36026014.0,
      "step": 16650
    },
    {
      "entropy": 6.557337951660156,
      "epoch": 1.7822783455508588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004689032048059656,
      "loss": 6.0307,
      "mean_token_accuracy": 0.15577168613672257,
      "num_tokens": 36035926.0,
      "step": 16655
    },
    {
      "entropy": 6.466844177246093,
      "epoch": 1.7828134196586227,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046888380768475,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1568325787782669,
      "num_tokens": 36046406.0,
      "step": 16660
    },
    {
      "entropy": 6.461503505706787,
      "epoch": 1.7833484937663866,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004688644049651667,
      "loss": 5.9033,
      "mean_token_accuracy": 0.15989567190408707,
      "num_tokens": 36056440.0,
      "step": 16665
    },
    {
      "entropy": 6.419238615036011,
      "epoch": 1.7838835678741507,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046884499664777586,
      "loss": 5.9778,
      "mean_token_accuracy": 0.1539990484714508,
      "num_tokens": 36067244.0,
      "step": 16670
    },
    {
      "entropy": 6.4605100631713865,
      "epoch": 1.7844186419819144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004688255827331381,
      "loss": 5.85,
      "mean_token_accuracy": 0.16841596961021424,
      "num_tokens": 36078110.0,
      "step": 16675
    },
    {
      "entropy": 6.483915519714356,
      "epoch": 1.7849537160896785,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004688061632218137,
      "loss": 5.9454,
      "mean_token_accuracy": 0.15961243212223053,
      "num_tokens": 36088805.0,
      "step": 16680
    },
    {
      "entropy": 6.404408121109009,
      "epoch": 1.7854887901974423,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046878673811436375,
      "loss": 5.9399,
      "mean_token_accuracy": 0.16294525265693666,
      "num_tokens": 36098619.0,
      "step": 16685
    },
    {
      "entropy": 6.50284628868103,
      "epoch": 1.7860238643052062,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004687673074113489,
      "loss": 5.9955,
      "mean_token_accuracy": 0.16108569204807283,
      "num_tokens": 36110604.0,
      "step": 16690
    },
    {
      "entropy": 6.550259780883789,
      "epoch": 1.7865589384129703,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004687478711133303,
      "loss": 5.9923,
      "mean_token_accuracy": 0.15501810759305953,
      "num_tokens": 36121669.0,
      "step": 16695
    },
    {
      "entropy": 6.3199851512908936,
      "epoch": 1.787094012520734,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004687284292208691,
      "loss": 5.7789,
      "mean_token_accuracy": 0.17029794305562973,
      "num_tokens": 36132165.0,
      "step": 16700
    },
    {
      "entropy": 6.4140997409820555,
      "epoch": 1.787629086628498,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046870898173452676,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1643413558602333,
      "num_tokens": 36142830.0,
      "step": 16705
    },
    {
      "entropy": 6.41308364868164,
      "epoch": 1.788164160736262,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004686895286548648,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16497465074062348,
      "num_tokens": 36153812.0,
      "step": 16710
    },
    {
      "entropy": 6.422903919219971,
      "epoch": 1.7886992348440258,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046867006998244494,
      "loss": 5.93,
      "mean_token_accuracy": 0.16233614981174468,
      "num_tokens": 36163599.0,
      "step": 16715
    },
    {
      "entropy": 6.449645280838013,
      "epoch": 1.78923430895179,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046865060571782896,
      "loss": 5.9747,
      "mean_token_accuracy": 0.1519537851214409,
      "num_tokens": 36173815.0,
      "step": 16720
    },
    {
      "entropy": 6.528790807723999,
      "epoch": 1.7897693830595538,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046863113586157894,
      "loss": 6.0331,
      "mean_token_accuracy": 0.15549035668373107,
      "num_tokens": 36186131.0,
      "step": 16725
    },
    {
      "entropy": 6.455544805526733,
      "epoch": 1.7903044571673177,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004686116604142572,
      "loss": 5.9162,
      "mean_token_accuracy": 0.16508626639842988,
      "num_tokens": 36196084.0,
      "step": 16730
    },
    {
      "entropy": 6.36201958656311,
      "epoch": 1.7908395312750816,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004685921793764259,
      "loss": 5.8677,
      "mean_token_accuracy": 0.1725776821374893,
      "num_tokens": 36206948.0,
      "step": 16735
    },
    {
      "entropy": 6.46815676689148,
      "epoch": 1.7913746053828454,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004685726927486476,
      "loss": 6.0056,
      "mean_token_accuracy": 0.15429437905550003,
      "num_tokens": 36217202.0,
      "step": 16740
    },
    {
      "entropy": 6.525591564178467,
      "epoch": 1.7919096794906095,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046855320053148505,
      "loss": 6.0433,
      "mean_token_accuracy": 0.14873735159635543,
      "num_tokens": 36227914.0,
      "step": 16745
    },
    {
      "entropy": 6.518138122558594,
      "epoch": 1.7924447535983734,
      "grad_norm": 1.0,
      "learning_rate": 0.000468533702725501,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15561925023794174,
      "num_tokens": 36239344.0,
      "step": 16750
    },
    {
      "entropy": 6.477685260772705,
      "epoch": 1.7929798277061373,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046851419933125843,
      "loss": 6.0072,
      "mean_token_accuracy": 0.1564896047115326,
      "num_tokens": 36250438.0,
      "step": 16755
    },
    {
      "entropy": 6.484805774688721,
      "epoch": 1.7935149018139014,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004684946903493206,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1517724096775055,
      "num_tokens": 36260545.0,
      "step": 16760
    },
    {
      "entropy": 6.38737530708313,
      "epoch": 1.794049975921665,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004684751757802508,
      "loss": 5.8802,
      "mean_token_accuracy": 0.15823953449726105,
      "num_tokens": 36272577.0,
      "step": 16765
    },
    {
      "entropy": 6.451371526718139,
      "epoch": 1.7945850500294291,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046845565562461243,
      "loss": 5.9125,
      "mean_token_accuracy": 0.16405297666788102,
      "num_tokens": 36284161.0,
      "step": 16770
    },
    {
      "entropy": 6.458127212524414,
      "epoch": 1.795120124137193,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046843612988296926,
      "loss": 5.9338,
      "mean_token_accuracy": 0.15465751886367798,
      "num_tokens": 36294910.0,
      "step": 16775
    },
    {
      "entropy": 6.432353210449219,
      "epoch": 1.795655198244957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004684165985558849,
      "loss": 5.8905,
      "mean_token_accuracy": 0.16392970383167266,
      "num_tokens": 36305388.0,
      "step": 16780
    },
    {
      "entropy": 6.414425897598266,
      "epoch": 1.796190272352721,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004683970616439236,
      "loss": 5.9401,
      "mean_token_accuracy": 0.16578255742788314,
      "num_tokens": 36316696.0,
      "step": 16785
    },
    {
      "entropy": 6.413032674789429,
      "epoch": 1.7967253464604847,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004683775191476492,
      "loss": 5.8877,
      "mean_token_accuracy": 0.1695579931139946,
      "num_tokens": 36327246.0,
      "step": 16790
    },
    {
      "entropy": 6.5049182891845705,
      "epoch": 1.7972604205682488,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004683579710676262,
      "loss": 6.0173,
      "mean_token_accuracy": 0.15417939573526382,
      "num_tokens": 36338089.0,
      "step": 16795
    },
    {
      "entropy": 6.479226446151733,
      "epoch": 1.7977954946760126,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046833841740441886,
      "loss": 5.9759,
      "mean_token_accuracy": 0.1613370269536972,
      "num_tokens": 36350285.0,
      "step": 16800
    },
    {
      "entropy": 6.503708600997925,
      "epoch": 1.7983305687837765,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004683188581585919,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15894935131072999,
      "num_tokens": 36360431.0,
      "step": 16805
    },
    {
      "entropy": 6.461076164245606,
      "epoch": 1.7988656428915406,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004682992933307101,
      "loss": 5.9152,
      "mean_token_accuracy": 0.1636996179819107,
      "num_tokens": 36371257.0,
      "step": 16810
    },
    {
      "entropy": 6.378208160400391,
      "epoch": 1.7994007169993043,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046827972292133835,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16179004460573196,
      "num_tokens": 36382613.0,
      "step": 16815
    },
    {
      "entropy": 6.512180233001709,
      "epoch": 1.7999357911070684,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004682601469310418,
      "loss": 5.9657,
      "mean_token_accuracy": 0.15890199244022368,
      "num_tokens": 36393799.0,
      "step": 16820
    },
    {
      "entropy": 6.466541957855225,
      "epoch": 1.8004708652148322,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046824056536038565,
      "loss": 5.9441,
      "mean_token_accuracy": 0.15570156201720237,
      "num_tokens": 36403914.0,
      "step": 16825
    },
    {
      "entropy": 6.356946563720703,
      "epoch": 1.8010059393225961,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004682209782099353,
      "loss": 5.9012,
      "mean_token_accuracy": 0.16132306307554245,
      "num_tokens": 36415022.0,
      "step": 16830
    },
    {
      "entropy": 6.448938226699829,
      "epoch": 1.8015410134303602,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046820138548025634,
      "loss": 5.8778,
      "mean_token_accuracy": 0.16665878295898437,
      "num_tokens": 36426155.0,
      "step": 16835
    },
    {
      "entropy": 6.466655111312866,
      "epoch": 1.8020760875381239,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046818178717191464,
      "loss": 5.9311,
      "mean_token_accuracy": 0.158075512945652,
      "num_tokens": 36437345.0,
      "step": 16840
    },
    {
      "entropy": 6.4574323177337645,
      "epoch": 1.802611161645888,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004681621832854759,
      "loss": 5.9661,
      "mean_token_accuracy": 0.16535462886095048,
      "num_tokens": 36448634.0,
      "step": 16845
    },
    {
      "entropy": 6.422204256057739,
      "epoch": 1.8031462357536518,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046814257382150627,
      "loss": 5.8965,
      "mean_token_accuracy": 0.1664792701601982,
      "num_tokens": 36458870.0,
      "step": 16850
    },
    {
      "entropy": 6.487599420547485,
      "epoch": 1.8036813098614157,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046812295878057193,
      "loss": 6.0537,
      "mean_token_accuracy": 0.15737884491682053,
      "num_tokens": 36469865.0,
      "step": 16855
    },
    {
      "entropy": 6.468596649169922,
      "epoch": 1.8042163839691798,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004681033381632393,
      "loss": 5.9428,
      "mean_token_accuracy": 0.15757690519094467,
      "num_tokens": 36480502.0,
      "step": 16860
    },
    {
      "entropy": 6.49014949798584,
      "epoch": 1.8047514580769437,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004680837119700749,
      "loss": 5.957,
      "mean_token_accuracy": 0.1585205852985382,
      "num_tokens": 36492356.0,
      "step": 16865
    },
    {
      "entropy": 6.337128448486328,
      "epoch": 1.8052865321847076,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046806408020164553,
      "loss": 5.9298,
      "mean_token_accuracy": 0.16342657059431076,
      "num_tokens": 36504659.0,
      "step": 16870
    },
    {
      "entropy": 6.4387535572052,
      "epoch": 1.8058216062924715,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046804444285851796,
      "loss": 5.9345,
      "mean_token_accuracy": 0.16214554160833358,
      "num_tokens": 36516042.0,
      "step": 16875
    },
    {
      "entropy": 6.522484683990479,
      "epoch": 1.8063566804002353,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046802479994125916,
      "loss": 6.0219,
      "mean_token_accuracy": 0.15437335073947905,
      "num_tokens": 36527417.0,
      "step": 16880
    },
    {
      "entropy": 6.444937610626221,
      "epoch": 1.8068917545079994,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004680051514504364,
      "loss": 5.9329,
      "mean_token_accuracy": 0.16625189930200576,
      "num_tokens": 36538182.0,
      "step": 16885
    },
    {
      "entropy": 6.443659543991089,
      "epoch": 1.8074268286157633,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004679854973866171,
      "loss": 5.9993,
      "mean_token_accuracy": 0.16126487851142884,
      "num_tokens": 36549835.0,
      "step": 16890
    },
    {
      "entropy": 6.423023700714111,
      "epoch": 1.8079619027235272,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046796583775036857,
      "loss": 5.919,
      "mean_token_accuracy": 0.16747631281614303,
      "num_tokens": 36560238.0,
      "step": 16895
    },
    {
      "entropy": 6.4416412830352785,
      "epoch": 1.8084969768312913,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046794617254225865,
      "loss": 5.948,
      "mean_token_accuracy": 0.1589287906885147,
      "num_tokens": 36571267.0,
      "step": 16900
    },
    {
      "entropy": 6.47004075050354,
      "epoch": 1.809032050939055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004679265017628551,
      "loss": 6.0064,
      "mean_token_accuracy": 0.16381992399692535,
      "num_tokens": 36582174.0,
      "step": 16905
    },
    {
      "entropy": 6.528776359558106,
      "epoch": 1.809567125046819,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004679068254127259,
      "loss": 5.9476,
      "mean_token_accuracy": 0.16211193352937697,
      "num_tokens": 36593054.0,
      "step": 16910
    },
    {
      "entropy": 6.4962318420410154,
      "epoch": 1.810102199154583,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046788714349243923,
      "loss": 5.9698,
      "mean_token_accuracy": 0.16107268780469894,
      "num_tokens": 36603962.0,
      "step": 16915
    },
    {
      "entropy": 6.4810038089752195,
      "epoch": 1.8106372732623468,
      "grad_norm": 1.0,
      "learning_rate": 0.0004678674560025634,
      "loss": 5.9389,
      "mean_token_accuracy": 0.16441629827022552,
      "num_tokens": 36616003.0,
      "step": 16920
    },
    {
      "entropy": 6.538247060775757,
      "epoch": 1.811172347370111,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004678477629436669,
      "loss": 6.0312,
      "mean_token_accuracy": 0.14943517595529557,
      "num_tokens": 36627189.0,
      "step": 16925
    },
    {
      "entropy": 6.533910846710205,
      "epoch": 1.8117074214778746,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004678280643163184,
      "loss": 6.0053,
      "mean_token_accuracy": 0.1537737399339676,
      "num_tokens": 36637756.0,
      "step": 16930
    },
    {
      "entropy": 6.438095569610596,
      "epoch": 1.8122424955856387,
      "grad_norm": 1.0,
      "learning_rate": 0.0004678083601210865,
      "loss": 5.9706,
      "mean_token_accuracy": 0.1623497799038887,
      "num_tokens": 36648946.0,
      "step": 16935
    },
    {
      "entropy": 6.511267614364624,
      "epoch": 1.8127775696934025,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004677886503585404,
      "loss": 6.008,
      "mean_token_accuracy": 0.15804363191127777,
      "num_tokens": 36660184.0,
      "step": 16940
    },
    {
      "entropy": 6.548721885681152,
      "epoch": 1.8133126438011664,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046776893502924904,
      "loss": 6.0342,
      "mean_token_accuracy": 0.15542205572128295,
      "num_tokens": 36671346.0,
      "step": 16945
    },
    {
      "entropy": 6.4743359088897705,
      "epoch": 1.8138477179089305,
      "grad_norm": 1.0,
      "learning_rate": 0.0004677492141337818,
      "loss": 5.949,
      "mean_token_accuracy": 0.1603161558508873,
      "num_tokens": 36682917.0,
      "step": 16950
    },
    {
      "entropy": 6.436689758300782,
      "epoch": 1.8143827920166942,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004677294876727081,
      "loss": 6.0152,
      "mean_token_accuracy": 0.1500933960080147,
      "num_tokens": 36693577.0,
      "step": 16955
    },
    {
      "entropy": 6.531511354446411,
      "epoch": 1.8149178661244583,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004677097556465975,
      "loss": 5.9714,
      "mean_token_accuracy": 0.15637680888175964,
      "num_tokens": 36704915.0,
      "step": 16960
    },
    {
      "entropy": 6.512617874145508,
      "epoch": 1.8154529402322221,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00046769001805601983,
      "loss": 6.0809,
      "mean_token_accuracy": 0.14920285791158677,
      "num_tokens": 36716086.0,
      "step": 16965
    },
    {
      "entropy": 6.417675876617432,
      "epoch": 1.815988014339986,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046767027490154494,
      "loss": 5.9312,
      "mean_token_accuracy": 0.15422417521476744,
      "num_tokens": 36726110.0,
      "step": 16970
    },
    {
      "entropy": 6.496168947219848,
      "epoch": 1.8165230884477501,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046765052618374296,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16704937368631362,
      "num_tokens": 36735846.0,
      "step": 16975
    },
    {
      "entropy": 6.472632598876953,
      "epoch": 1.8170581625555138,
      "grad_norm": 0.984375,
      "learning_rate": 0.000467630771903184,
      "loss": 5.9756,
      "mean_token_accuracy": 0.16260184347629547,
      "num_tokens": 36746851.0,
      "step": 16980
    },
    {
      "entropy": 6.4802347183227536,
      "epoch": 1.8175932366632779,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004676110120604387,
      "loss": 5.9423,
      "mean_token_accuracy": 0.15826310813426972,
      "num_tokens": 36758032.0,
      "step": 16985
    },
    {
      "entropy": 6.4893670082092285,
      "epoch": 1.8181283107710418,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004675912466560775,
      "loss": 5.9851,
      "mean_token_accuracy": 0.1616771101951599,
      "num_tokens": 36768939.0,
      "step": 16990
    },
    {
      "entropy": 6.496434879302979,
      "epoch": 1.8186633848788056,
      "grad_norm": 1.125,
      "learning_rate": 0.00046757147569067106,
      "loss": 6.0142,
      "mean_token_accuracy": 0.1550060033798218,
      "num_tokens": 36780338.0,
      "step": 16995
    },
    {
      "entropy": 6.464590358734131,
      "epoch": 1.8191984589865697,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004675516991647903,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15413339510560037,
      "num_tokens": 36791384.0,
      "step": 17000
    },
    {
      "entropy": 6.463540077209473,
      "epoch": 1.8197335330943336,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046753191707900636,
      "loss": 5.9245,
      "mean_token_accuracy": 0.15407318696379663,
      "num_tokens": 36802460.0,
      "step": 17005
    },
    {
      "entropy": 6.505455350875854,
      "epoch": 1.8202686072020975,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046751212943389026,
      "loss": 5.9991,
      "mean_token_accuracy": 0.1523193135857582,
      "num_tokens": 36813599.0,
      "step": 17010
    },
    {
      "entropy": 6.465418815612793,
      "epoch": 1.8208036813098614,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004674923362300135,
      "loss": 5.894,
      "mean_token_accuracy": 0.15861208587884904,
      "num_tokens": 36823100.0,
      "step": 17015
    },
    {
      "entropy": 6.501190853118897,
      "epoch": 1.8213387554176252,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046747253746794767,
      "loss": 5.9977,
      "mean_token_accuracy": 0.1544879212975502,
      "num_tokens": 36834245.0,
      "step": 17020
    },
    {
      "entropy": 6.461848592758178,
      "epoch": 1.8218738295253893,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046745273314826425,
      "loss": 5.899,
      "mean_token_accuracy": 0.16755282878875732,
      "num_tokens": 36844126.0,
      "step": 17025
    },
    {
      "entropy": 6.44581298828125,
      "epoch": 1.8224089036331532,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004674329232715353,
      "loss": 5.8674,
      "mean_token_accuracy": 0.1700243815779686,
      "num_tokens": 36855451.0,
      "step": 17030
    },
    {
      "entropy": 6.45113172531128,
      "epoch": 1.822943977740917,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004674131078383327,
      "loss": 5.9576,
      "mean_token_accuracy": 0.1540319725871086,
      "num_tokens": 36864994.0,
      "step": 17035
    },
    {
      "entropy": 6.4241961479187015,
      "epoch": 1.8234790518486812,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004673932868492286,
      "loss": 5.9182,
      "mean_token_accuracy": 0.1585528552532196,
      "num_tokens": 36875131.0,
      "step": 17040
    },
    {
      "entropy": 6.5025708198547365,
      "epoch": 1.8240141259564449,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046737346030479546,
      "loss": 5.9885,
      "mean_token_accuracy": 0.1572924479842186,
      "num_tokens": 36884789.0,
      "step": 17045
    },
    {
      "entropy": 6.447276210784912,
      "epoch": 1.824549200064209,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046735362820560564,
      "loss": 5.8778,
      "mean_token_accuracy": 0.15598295778036117,
      "num_tokens": 36895199.0,
      "step": 17050
    },
    {
      "entropy": 6.460340929031372,
      "epoch": 1.8250842741719728,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004673337905522318,
      "loss": 5.992,
      "mean_token_accuracy": 0.15293519645929338,
      "num_tokens": 36906495.0,
      "step": 17055
    },
    {
      "entropy": 6.471297359466552,
      "epoch": 1.8256193482797367,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046731394734524686,
      "loss": 5.9601,
      "mean_token_accuracy": 0.1514850378036499,
      "num_tokens": 36916907.0,
      "step": 17060
    },
    {
      "entropy": 6.444206190109253,
      "epoch": 1.8261544223875008,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046729409858522363,
      "loss": 5.9134,
      "mean_token_accuracy": 0.15952033996582032,
      "num_tokens": 36927872.0,
      "step": 17065
    },
    {
      "entropy": 6.468321180343628,
      "epoch": 1.8266894964952645,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004672742442727354,
      "loss": 5.9596,
      "mean_token_accuracy": 0.16096605211496354,
      "num_tokens": 36938575.0,
      "step": 17070
    },
    {
      "entropy": 6.432369613647461,
      "epoch": 1.8272245706030286,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046725438440835536,
      "loss": 5.8526,
      "mean_token_accuracy": 0.17158474400639534,
      "num_tokens": 36949179.0,
      "step": 17075
    },
    {
      "entropy": 6.412952327728272,
      "epoch": 1.8277596447107924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000467234518992657,
      "loss": 5.9027,
      "mean_token_accuracy": 0.16556908786296845,
      "num_tokens": 36959164.0,
      "step": 17080
    },
    {
      "entropy": 6.498966693878174,
      "epoch": 1.8282947188185563,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046721464802621383,
      "loss": 6.0292,
      "mean_token_accuracy": 0.15994127839803696,
      "num_tokens": 36969091.0,
      "step": 17085
    },
    {
      "entropy": 6.484938430786133,
      "epoch": 1.8288297929263204,
      "grad_norm": 0.890625,
      "learning_rate": 0.00046719477150959984,
      "loss": 5.9606,
      "mean_token_accuracy": 0.15471125692129134,
      "num_tokens": 36980481.0,
      "step": 17090
    },
    {
      "entropy": 6.462921905517578,
      "epoch": 1.829364867034084,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046717488944338873,
      "loss": 5.9323,
      "mean_token_accuracy": 0.1598064422607422,
      "num_tokens": 36991349.0,
      "step": 17095
    },
    {
      "entropy": 6.4504095077514645,
      "epoch": 1.8298999411418482,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046715500182815473,
      "loss": 6.0284,
      "mean_token_accuracy": 0.15910299718379975,
      "num_tokens": 37002803.0,
      "step": 17100
    },
    {
      "entropy": 6.498575925827026,
      "epoch": 1.830435015249612,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046713510866447204,
      "loss": 5.9744,
      "mean_token_accuracy": 0.15925669968128203,
      "num_tokens": 37013578.0,
      "step": 17105
    },
    {
      "entropy": 6.456055688858032,
      "epoch": 1.830970089357376,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004671152099529152,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16450800150632858,
      "num_tokens": 37024012.0,
      "step": 17110
    },
    {
      "entropy": 6.423557710647583,
      "epoch": 1.83150516346514,
      "grad_norm": 1.0,
      "learning_rate": 0.00046709530569405855,
      "loss": 5.876,
      "mean_token_accuracy": 0.16302137821912766,
      "num_tokens": 37034149.0,
      "step": 17115
    },
    {
      "entropy": 6.410289716720581,
      "epoch": 1.8320402375729037,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004670753958884769,
      "loss": 5.8747,
      "mean_token_accuracy": 0.16444529443979264,
      "num_tokens": 37045669.0,
      "step": 17120
    },
    {
      "entropy": 6.451217222213745,
      "epoch": 1.8325753116806678,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00046705548053674524,
      "loss": 5.9547,
      "mean_token_accuracy": 0.16052009761333466,
      "num_tokens": 37056611.0,
      "step": 17125
    },
    {
      "entropy": 6.418110609054565,
      "epoch": 1.8331103857884317,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046703555963943855,
      "loss": 5.9455,
      "mean_token_accuracy": 0.15376774817705155,
      "num_tokens": 37067387.0,
      "step": 17130
    },
    {
      "entropy": 6.5577398300170895,
      "epoch": 1.8336454598961955,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046701563319713214,
      "loss": 6.0077,
      "mean_token_accuracy": 0.15382856875658035,
      "num_tokens": 37078151.0,
      "step": 17135
    },
    {
      "entropy": 6.513970279693604,
      "epoch": 1.8341805340039596,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046699570121040126,
      "loss": 5.9827,
      "mean_token_accuracy": 0.16027993559837342,
      "num_tokens": 37088219.0,
      "step": 17140
    },
    {
      "entropy": 6.423409128189087,
      "epoch": 1.8347156081117235,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004669757636798215,
      "loss": 5.8658,
      "mean_token_accuracy": 0.1583323821425438,
      "num_tokens": 37098769.0,
      "step": 17145
    },
    {
      "entropy": 6.4833066940307615,
      "epoch": 1.8352506822194874,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004669558206059684,
      "loss": 6.0213,
      "mean_token_accuracy": 0.14872189238667488,
      "num_tokens": 37110716.0,
      "step": 17150
    },
    {
      "entropy": 6.4609991073608395,
      "epoch": 1.8357857563272513,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004669358719894181,
      "loss": 5.9102,
      "mean_token_accuracy": 0.1585691064596176,
      "num_tokens": 37122785.0,
      "step": 17155
    },
    {
      "entropy": 6.435960912704468,
      "epoch": 1.8363208304350152,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004669159178307465,
      "loss": 5.9386,
      "mean_token_accuracy": 0.15738015621900558,
      "num_tokens": 37133308.0,
      "step": 17160
    },
    {
      "entropy": 6.402177047729492,
      "epoch": 1.8368559045427792,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004668959581305296,
      "loss": 5.8866,
      "mean_token_accuracy": 0.15992464423179625,
      "num_tokens": 37143916.0,
      "step": 17165
    },
    {
      "entropy": 6.426356363296509,
      "epoch": 1.8373909786505431,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046687599288934394,
      "loss": 5.9139,
      "mean_token_accuracy": 0.15923503190279006,
      "num_tokens": 37154335.0,
      "step": 17170
    },
    {
      "entropy": 6.452762699127197,
      "epoch": 1.837926052758307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000466856022107766,
      "loss": 5.9538,
      "mean_token_accuracy": 0.15909279584884645,
      "num_tokens": 37164116.0,
      "step": 17175
    },
    {
      "entropy": 6.493045997619629,
      "epoch": 1.838461126866071,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046683604578637235,
      "loss": 6.0205,
      "mean_token_accuracy": 0.15719804614782334,
      "num_tokens": 37175407.0,
      "step": 17180
    },
    {
      "entropy": 6.440932989120483,
      "epoch": 1.8389962009738348,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046681606392573975,
      "loss": 5.9076,
      "mean_token_accuracy": 0.15843815505504608,
      "num_tokens": 37185339.0,
      "step": 17185
    },
    {
      "entropy": 6.504052066802979,
      "epoch": 1.8395312750815989,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004667960765264454,
      "loss": 6.0033,
      "mean_token_accuracy": 0.15555770695209503,
      "num_tokens": 37195145.0,
      "step": 17190
    },
    {
      "entropy": 6.4788590431213375,
      "epoch": 1.8400663491893627,
      "grad_norm": 1.0,
      "learning_rate": 0.0004667760835890662,
      "loss": 6.0247,
      "mean_token_accuracy": 0.1537323296070099,
      "num_tokens": 37206454.0,
      "step": 17195
    },
    {
      "entropy": 6.469643068313599,
      "epoch": 1.8406014232971266,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046675608511417957,
      "loss": 5.9166,
      "mean_token_accuracy": 0.1637010842561722,
      "num_tokens": 37216646.0,
      "step": 17200
    },
    {
      "entropy": 6.389241790771484,
      "epoch": 1.8411364974048907,
      "grad_norm": 1.0,
      "learning_rate": 0.00046673608110236293,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1708819553256035,
      "num_tokens": 37227368.0,
      "step": 17205
    },
    {
      "entropy": 6.486259984970093,
      "epoch": 1.8416715715126544,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004667160715541939,
      "loss": 5.9925,
      "mean_token_accuracy": 0.15764017701148986,
      "num_tokens": 37238150.0,
      "step": 17210
    },
    {
      "entropy": 6.457462978363037,
      "epoch": 1.8422066456204185,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046669605647025014,
      "loss": 5.9199,
      "mean_token_accuracy": 0.15664153397083283,
      "num_tokens": 37249534.0,
      "step": 17215
    },
    {
      "entropy": 6.504864311218261,
      "epoch": 1.8427417197281823,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004666760358511098,
      "loss": 6.0275,
      "mean_token_accuracy": 0.1589784763753414,
      "num_tokens": 37260633.0,
      "step": 17220
    },
    {
      "entropy": 6.491711044311524,
      "epoch": 1.8432767938359462,
      "grad_norm": 0.95703125,
      "learning_rate": 0.00046665600969735086,
      "loss": 5.9735,
      "mean_token_accuracy": 0.1562628984451294,
      "num_tokens": 37271978.0,
      "step": 17225
    },
    {
      "entropy": 6.405184698104859,
      "epoch": 1.8438118679437103,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004666359780095515,
      "loss": 5.8454,
      "mean_token_accuracy": 0.16568995416164398,
      "num_tokens": 37284130.0,
      "step": 17230
    },
    {
      "entropy": 6.476573467254639,
      "epoch": 1.844346942051474,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004666159407882902,
      "loss": 6.0241,
      "mean_token_accuracy": 0.1524647891521454,
      "num_tokens": 37295648.0,
      "step": 17235
    },
    {
      "entropy": 6.458185720443725,
      "epoch": 1.844882016159238,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004665958980341456,
      "loss": 5.9905,
      "mean_token_accuracy": 0.15826675146818162,
      "num_tokens": 37306715.0,
      "step": 17240
    },
    {
      "entropy": 6.504457712173462,
      "epoch": 1.845417090267002,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046657584974769636,
      "loss": 5.9951,
      "mean_token_accuracy": 0.15971070677042007,
      "num_tokens": 37317605.0,
      "step": 17245
    },
    {
      "entropy": 6.458282947540283,
      "epoch": 1.8459521643747658,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004665557959295213,
      "loss": 5.9272,
      "mean_token_accuracy": 0.16281720101833344,
      "num_tokens": 37328898.0,
      "step": 17250
    },
    {
      "entropy": 6.453275918960571,
      "epoch": 1.84648723848253,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046653573658019963,
      "loss": 5.9024,
      "mean_token_accuracy": 0.16214517056941985,
      "num_tokens": 37340108.0,
      "step": 17255
    },
    {
      "entropy": 6.489439249038696,
      "epoch": 1.8470223125902936,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046651567170031053,
      "loss": 6.0181,
      "mean_token_accuracy": 0.1569735050201416,
      "num_tokens": 37351773.0,
      "step": 17260
    },
    {
      "entropy": 6.459282159805298,
      "epoch": 1.8475573866980577,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004664956012904332,
      "loss": 5.931,
      "mean_token_accuracy": 0.1620381012558937,
      "num_tokens": 37362748.0,
      "step": 17265
    },
    {
      "entropy": 6.46019983291626,
      "epoch": 1.8480924608058216,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004664755253511474,
      "loss": 6.0201,
      "mean_token_accuracy": 0.15298373401165008,
      "num_tokens": 37373487.0,
      "step": 17270
    },
    {
      "entropy": 6.458335733413696,
      "epoch": 1.8486275349135854,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004664554438830326,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16100094318389893,
      "num_tokens": 37383302.0,
      "step": 17275
    },
    {
      "entropy": 6.457779312133789,
      "epoch": 1.8491626090213495,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004664353568866689,
      "loss": 5.91,
      "mean_token_accuracy": 0.1534661665558815,
      "num_tokens": 37393085.0,
      "step": 17280
    },
    {
      "entropy": 6.519992303848267,
      "epoch": 1.8496976831291134,
      "grad_norm": 1.171875,
      "learning_rate": 0.000466415264362636,
      "loss": 5.9814,
      "mean_token_accuracy": 0.1614265576004982,
      "num_tokens": 37403774.0,
      "step": 17285
    },
    {
      "entropy": 6.465917682647705,
      "epoch": 1.8502327572368773,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046639516631151435,
      "loss": 5.9352,
      "mean_token_accuracy": 0.16626698076725005,
      "num_tokens": 37414356.0,
      "step": 17290
    },
    {
      "entropy": 6.471253108978272,
      "epoch": 1.8507678313446412,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046637506273388416,
      "loss": 5.9805,
      "mean_token_accuracy": 0.15715974867343901,
      "num_tokens": 37426381.0,
      "step": 17295
    },
    {
      "entropy": 6.499348402023315,
      "epoch": 1.851302905452405,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046635495363032595,
      "loss": 5.9378,
      "mean_token_accuracy": 0.16127346009016036,
      "num_tokens": 37436259.0,
      "step": 17300
    },
    {
      "entropy": 6.443133306503296,
      "epoch": 1.8518379795601692,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046633483900142026,
      "loss": 5.9208,
      "mean_token_accuracy": 0.16828929781913757,
      "num_tokens": 37447498.0,
      "step": 17305
    },
    {
      "entropy": 6.432978200912475,
      "epoch": 1.852373053667933,
      "grad_norm": 0.984375,
      "learning_rate": 0.000466314718847748,
      "loss": 5.9218,
      "mean_token_accuracy": 0.15934144109487533,
      "num_tokens": 37457323.0,
      "step": 17310
    },
    {
      "entropy": 6.468656015396118,
      "epoch": 1.852908127775697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004662945931698901,
      "loss": 5.9418,
      "mean_token_accuracy": 0.15894611924886703,
      "num_tokens": 37468121.0,
      "step": 17315
    },
    {
      "entropy": 6.495964527130127,
      "epoch": 1.853443201883461,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046627446196842784,
      "loss": 5.9885,
      "mean_token_accuracy": 0.15414802134037017,
      "num_tokens": 37479205.0,
      "step": 17320
    },
    {
      "entropy": 6.525689744949341,
      "epoch": 1.8539782759912247,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004662543252439422,
      "loss": 6.0653,
      "mean_token_accuracy": 0.15618605315685272,
      "num_tokens": 37491015.0,
      "step": 17325
    },
    {
      "entropy": 6.5569323062896725,
      "epoch": 1.8545133500989888,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046623418299701494,
      "loss": 5.9997,
      "mean_token_accuracy": 0.15278897881507875,
      "num_tokens": 37503078.0,
      "step": 17330
    },
    {
      "entropy": 6.484497499465943,
      "epoch": 1.8550484242067526,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004662140352282274,
      "loss": 5.9595,
      "mean_token_accuracy": 0.16609691828489304,
      "num_tokens": 37514623.0,
      "step": 17335
    },
    {
      "entropy": 6.433059597015381,
      "epoch": 1.8555834983145165,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046619388193816155,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15891185253858567,
      "num_tokens": 37525618.0,
      "step": 17340
    },
    {
      "entropy": 6.531949758529663,
      "epoch": 1.8561185724222806,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046617372312739917,
      "loss": 5.9852,
      "mean_token_accuracy": 0.16115886420011521,
      "num_tokens": 37536023.0,
      "step": 17345
    },
    {
      "entropy": 6.557612943649292,
      "epoch": 1.8566536465300443,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004661535587965224,
      "loss": 5.9965,
      "mean_token_accuracy": 0.14983400702476501,
      "num_tokens": 37545886.0,
      "step": 17350
    },
    {
      "entropy": 6.459822273254394,
      "epoch": 1.8571887206378084,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046613338894611347,
      "loss": 5.9137,
      "mean_token_accuracy": 0.15863879770040512,
      "num_tokens": 37557587.0,
      "step": 17355
    },
    {
      "entropy": 6.377936840057373,
      "epoch": 1.8577237947455723,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004661132135767548,
      "loss": 5.8178,
      "mean_token_accuracy": 0.17574200481176377,
      "num_tokens": 37569379.0,
      "step": 17360
    },
    {
      "entropy": 6.552577877044678,
      "epoch": 1.8582588688533361,
      "grad_norm": 0.9375,
      "learning_rate": 0.000466093032689029,
      "loss": 6.0241,
      "mean_token_accuracy": 0.15287581980228424,
      "num_tokens": 37580478.0,
      "step": 17365
    },
    {
      "entropy": 6.415551471710205,
      "epoch": 1.8587939429611002,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004660728462835187,
      "loss": 5.953,
      "mean_token_accuracy": 0.16037695556879045,
      "num_tokens": 37591506.0,
      "step": 17370
    },
    {
      "entropy": 6.400301694869995,
      "epoch": 1.8593290170688639,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004660526543608068,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16831107288599015,
      "num_tokens": 37602074.0,
      "step": 17375
    },
    {
      "entropy": 6.510325479507446,
      "epoch": 1.859864091176628,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004660324569214763,
      "loss": 5.981,
      "mean_token_accuracy": 0.1557241126894951,
      "num_tokens": 37612180.0,
      "step": 17380
    },
    {
      "entropy": 6.47301754951477,
      "epoch": 1.8603991652843919,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004660122539661106,
      "loss": 5.9564,
      "mean_token_accuracy": 0.15704852640628814,
      "num_tokens": 37622588.0,
      "step": 17385
    },
    {
      "entropy": 6.501664495468139,
      "epoch": 1.8609342393921557,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004659920454952928,
      "loss": 5.958,
      "mean_token_accuracy": 0.1596525341272354,
      "num_tokens": 37634323.0,
      "step": 17390
    },
    {
      "entropy": 6.484533357620239,
      "epoch": 1.8614693134999198,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004659718315096065,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15735394954681398,
      "num_tokens": 37645816.0,
      "step": 17395
    },
    {
      "entropy": 6.52896900177002,
      "epoch": 1.8620043876076835,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046595161200963545,
      "loss": 5.9979,
      "mean_token_accuracy": 0.16146398186683655,
      "num_tokens": 37656143.0,
      "step": 17400
    },
    {
      "entropy": 6.477845335006714,
      "epoch": 1.8625394617154476,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046593138699596343,
      "loss": 5.9014,
      "mean_token_accuracy": 0.15622884929180145,
      "num_tokens": 37665959.0,
      "step": 17405
    },
    {
      "entropy": 6.428560781478882,
      "epoch": 1.8630745358232115,
      "grad_norm": 1.5,
      "learning_rate": 0.00046591115646917443,
      "loss": 5.9205,
      "mean_token_accuracy": 0.15840750485658645,
      "num_tokens": 37677176.0,
      "step": 17410
    },
    {
      "entropy": 6.472634887695312,
      "epoch": 1.8636096099309754,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004658909204298526,
      "loss": 5.9445,
      "mean_token_accuracy": 0.1665267214179039,
      "num_tokens": 37687740.0,
      "step": 17415
    },
    {
      "entropy": 6.399833393096924,
      "epoch": 1.8641446840387395,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004658706788785823,
      "loss": 5.9094,
      "mean_token_accuracy": 0.1666820153594017,
      "num_tokens": 37698536.0,
      "step": 17420
    },
    {
      "entropy": 6.501494646072388,
      "epoch": 1.8646797581465033,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046585043181594793,
      "loss": 5.9331,
      "mean_token_accuracy": 0.1640442669391632,
      "num_tokens": 37708795.0,
      "step": 17425
    },
    {
      "entropy": 6.497988367080689,
      "epoch": 1.8652148322542672,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046583017924253426,
      "loss": 5.9305,
      "mean_token_accuracy": 0.16455675214529036,
      "num_tokens": 37719136.0,
      "step": 17430
    },
    {
      "entropy": 6.604510354995727,
      "epoch": 1.8657499063620313,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004658099211589259,
      "loss": 6.0324,
      "mean_token_accuracy": 0.16005798131227494,
      "num_tokens": 37730513.0,
      "step": 17435
    },
    {
      "entropy": 6.457183885574341,
      "epoch": 1.866284980469795,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046578965756570787,
      "loss": 5.9432,
      "mean_token_accuracy": 0.16090741157531738,
      "num_tokens": 37741811.0,
      "step": 17440
    },
    {
      "entropy": 6.377370309829712,
      "epoch": 1.866820054577559,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046576938846346527,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17062472701072692,
      "num_tokens": 37752213.0,
      "step": 17445
    },
    {
      "entropy": 6.333180856704712,
      "epoch": 1.867355128685323,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046574911385278343,
      "loss": 5.784,
      "mean_token_accuracy": 0.173097725212574,
      "num_tokens": 37763253.0,
      "step": 17450
    },
    {
      "entropy": 6.4618408203125,
      "epoch": 1.8678902027930868,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046572883373424776,
      "loss": 5.988,
      "mean_token_accuracy": 0.15351605266332627,
      "num_tokens": 37775087.0,
      "step": 17455
    },
    {
      "entropy": 6.513929319381714,
      "epoch": 1.868425276900851,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046570854810844373,
      "loss": 5.9932,
      "mean_token_accuracy": 0.15291830599308015,
      "num_tokens": 37786119.0,
      "step": 17460
    },
    {
      "entropy": 6.522600698471069,
      "epoch": 1.8689603510086146,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046568825697595724,
      "loss": 5.9893,
      "mean_token_accuracy": 0.15564531832933426,
      "num_tokens": 37797482.0,
      "step": 17465
    },
    {
      "entropy": 6.466575050354004,
      "epoch": 1.8694954251163787,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004656679603373741,
      "loss": 5.9122,
      "mean_token_accuracy": 0.15961637943983079,
      "num_tokens": 37808067.0,
      "step": 17470
    },
    {
      "entropy": 6.483156347274781,
      "epoch": 1.8700304992241426,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004656476581932804,
      "loss": 6.0657,
      "mean_token_accuracy": 0.1511141076683998,
      "num_tokens": 37819762.0,
      "step": 17475
    },
    {
      "entropy": 6.429815435409546,
      "epoch": 1.8705655733319064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004656273505442624,
      "loss": 5.9444,
      "mean_token_accuracy": 0.16189284026622772,
      "num_tokens": 37829510.0,
      "step": 17480
    },
    {
      "entropy": 6.467136240005493,
      "epoch": 1.8711006474396705,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046560703739090633,
      "loss": 6.0191,
      "mean_token_accuracy": 0.15177636593580246,
      "num_tokens": 37839431.0,
      "step": 17485
    },
    {
      "entropy": 6.549655103683472,
      "epoch": 1.8716357215474342,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046558671873379886,
      "loss": 6.0092,
      "mean_token_accuracy": 0.14886450618505478,
      "num_tokens": 37850531.0,
      "step": 17490
    },
    {
      "entropy": 6.562495946884155,
      "epoch": 1.8721707956551983,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004655663945735268,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15029444098472594,
      "num_tokens": 37861964.0,
      "step": 17495
    },
    {
      "entropy": 6.37295184135437,
      "epoch": 1.8727058697629622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004655460649106768,
      "loss": 5.7323,
      "mean_token_accuracy": 0.17583550810813903,
      "num_tokens": 37873598.0,
      "step": 17500
    },
    {
      "entropy": 6.485764455795288,
      "epoch": 1.873240943870726,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046552572974583597,
      "loss": 5.9999,
      "mean_token_accuracy": 0.15087592005729675,
      "num_tokens": 37884643.0,
      "step": 17505
    },
    {
      "entropy": 6.410701274871826,
      "epoch": 1.8737760179784901,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004655053890795914,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15715604647994041,
      "num_tokens": 37896566.0,
      "step": 17510
    },
    {
      "entropy": 6.520356893539429,
      "epoch": 1.8743110920862538,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046548504291253054,
      "loss": 6.022,
      "mean_token_accuracy": 0.1520705744624138,
      "num_tokens": 37907424.0,
      "step": 17515
    },
    {
      "entropy": 6.525497484207153,
      "epoch": 1.874846166194018,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004654646912452408,
      "loss": 5.9641,
      "mean_token_accuracy": 0.16574487388134002,
      "num_tokens": 37917318.0,
      "step": 17520
    },
    {
      "entropy": 6.452678585052491,
      "epoch": 1.8753812403017818,
      "grad_norm": 1.0,
      "learning_rate": 0.0004654443340783099,
      "loss": 5.9191,
      "mean_token_accuracy": 0.1605178713798523,
      "num_tokens": 37928521.0,
      "step": 17525
    },
    {
      "entropy": 6.4554948806762695,
      "epoch": 1.8759163144095456,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046542397141232554,
      "loss": 5.9915,
      "mean_token_accuracy": 0.15985041558742524,
      "num_tokens": 37939930.0,
      "step": 17530
    },
    {
      "entropy": 6.455539512634277,
      "epoch": 1.8764513885173097,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004654036032478759,
      "loss": 5.902,
      "mean_token_accuracy": 0.15926898121833802,
      "num_tokens": 37949690.0,
      "step": 17535
    },
    {
      "entropy": 6.502910852432251,
      "epoch": 1.8769864626250736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046538322958554885,
      "loss": 5.9261,
      "mean_token_accuracy": 0.16178428530693054,
      "num_tokens": 37962162.0,
      "step": 17540
    },
    {
      "entropy": 6.568083763122559,
      "epoch": 1.8775215367328375,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004653628504259329,
      "loss": 5.9733,
      "mean_token_accuracy": 0.15916069149971007,
      "num_tokens": 37972707.0,
      "step": 17545
    },
    {
      "entropy": 6.435571670532227,
      "epoch": 1.8780566108406014,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046534246576961633,
      "loss": 5.9032,
      "mean_token_accuracy": 0.16427022516727446,
      "num_tokens": 37983101.0,
      "step": 17550
    },
    {
      "entropy": 6.332002449035644,
      "epoch": 1.8785916849483653,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046532207561718775,
      "loss": 5.8383,
      "mean_token_accuracy": 0.16688434928655624,
      "num_tokens": 37994025.0,
      "step": 17555
    },
    {
      "entropy": 6.4586193561553955,
      "epoch": 1.8791267590561294,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046530167996923613,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15451473742723465,
      "num_tokens": 38004761.0,
      "step": 17560
    },
    {
      "entropy": 6.528156328201294,
      "epoch": 1.8796618331638932,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004652812788263501,
      "loss": 5.9443,
      "mean_token_accuracy": 0.1629989981651306,
      "num_tokens": 38015129.0,
      "step": 17565
    },
    {
      "entropy": 6.462521934509278,
      "epoch": 1.8801969072716571,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046526087218911894,
      "loss": 5.9668,
      "mean_token_accuracy": 0.16572099477052687,
      "num_tokens": 38025501.0,
      "step": 17570
    },
    {
      "entropy": 6.501592493057251,
      "epoch": 1.8807319813794212,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046524046005813185,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15564179718494414,
      "num_tokens": 38034912.0,
      "step": 17575
    },
    {
      "entropy": 6.433433389663696,
      "epoch": 1.8812670554871849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004652200424339782,
      "loss": 5.895,
      "mean_token_accuracy": 0.16068387851119043,
      "num_tokens": 38045417.0,
      "step": 17580
    },
    {
      "entropy": 6.480678558349609,
      "epoch": 1.881802129594949,
      "grad_norm": 1.0,
      "learning_rate": 0.0004651996193172475,
      "loss": 6.0294,
      "mean_token_accuracy": 0.15243419259786606,
      "num_tokens": 38057033.0,
      "step": 17585
    },
    {
      "entropy": 6.5572985172271725,
      "epoch": 1.8823372037027128,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004651791907085296,
      "loss": 5.9495,
      "mean_token_accuracy": 0.16113544702529908,
      "num_tokens": 38067804.0,
      "step": 17590
    },
    {
      "entropy": 6.507932472229004,
      "epoch": 1.8828722778104767,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004651587566084143,
      "loss": 5.9319,
      "mean_token_accuracy": 0.16618551462888717,
      "num_tokens": 38079246.0,
      "step": 17595
    },
    {
      "entropy": 6.46846022605896,
      "epoch": 1.8834073519182408,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046513831701749155,
      "loss": 6.0886,
      "mean_token_accuracy": 0.15423011928796768,
      "num_tokens": 38090628.0,
      "step": 17600
    },
    {
      "entropy": 6.447257614135742,
      "epoch": 1.8839424260260045,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046511787193635165,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16394680067896844,
      "num_tokens": 38100848.0,
      "step": 17605
    },
    {
      "entropy": 6.525193309783935,
      "epoch": 1.8844775001337686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046509742136558493,
      "loss": 5.9345,
      "mean_token_accuracy": 0.16092996448278427,
      "num_tokens": 38110732.0,
      "step": 17610
    },
    {
      "entropy": 6.403708839416504,
      "epoch": 1.8850125742415325,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046507696530578185,
      "loss": 5.917,
      "mean_token_accuracy": 0.165976482629776,
      "num_tokens": 38121002.0,
      "step": 17615
    },
    {
      "entropy": 6.465030002593994,
      "epoch": 1.8855476483492963,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004650565037575331,
      "loss": 5.9742,
      "mean_token_accuracy": 0.16195354461669922,
      "num_tokens": 38132294.0,
      "step": 17620
    },
    {
      "entropy": 6.444005966186523,
      "epoch": 1.8860827224570604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046503603672142955,
      "loss": 5.852,
      "mean_token_accuracy": 0.16389525532722474,
      "num_tokens": 38142962.0,
      "step": 17625
    },
    {
      "entropy": 6.530234336853027,
      "epoch": 1.886617796564824,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004650155641980621,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15992840230464936,
      "num_tokens": 38153737.0,
      "step": 17630
    },
    {
      "entropy": 6.489470529556274,
      "epoch": 1.8871528706725882,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046499508618802195,
      "loss": 5.9723,
      "mean_token_accuracy": 0.15639637261629105,
      "num_tokens": 38163897.0,
      "step": 17635
    },
    {
      "entropy": 6.40191421508789,
      "epoch": 1.887687944780352,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004649746026919004,
      "loss": 5.9243,
      "mean_token_accuracy": 0.1663943275809288,
      "num_tokens": 38175123.0,
      "step": 17640
    },
    {
      "entropy": 6.4625166893005375,
      "epoch": 1.888223018888116,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046495411371028886,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1658332273364067,
      "num_tokens": 38186100.0,
      "step": 17645
    },
    {
      "entropy": 6.507238435745239,
      "epoch": 1.88875809299588,
      "grad_norm": 1.0,
      "learning_rate": 0.00046493361924377903,
      "loss": 5.9619,
      "mean_token_accuracy": 0.1624431788921356,
      "num_tokens": 38196385.0,
      "step": 17650
    },
    {
      "entropy": 6.428504991531372,
      "epoch": 1.8892931671036437,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004649131192929626,
      "loss": 5.9051,
      "mean_token_accuracy": 0.1562515139579773,
      "num_tokens": 38207260.0,
      "step": 17655
    },
    {
      "entropy": 6.431979417800903,
      "epoch": 1.8898282412114078,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004648926138584315,
      "loss": 5.8265,
      "mean_token_accuracy": 0.16608547270298005,
      "num_tokens": 38217537.0,
      "step": 17660
    },
    {
      "entropy": 6.514480495452881,
      "epoch": 1.8903633153191717,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046487210294077794,
      "loss": 5.9258,
      "mean_token_accuracy": 0.16653878837823868,
      "num_tokens": 38228956.0,
      "step": 17665
    },
    {
      "entropy": 6.359472274780273,
      "epoch": 1.8908983894269356,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046485158654059405,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16674765646457673,
      "num_tokens": 38240236.0,
      "step": 17670
    },
    {
      "entropy": 6.464616346359253,
      "epoch": 1.8914334635346997,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004648310646584723,
      "loss": 6.005,
      "mean_token_accuracy": 0.16295454502105713,
      "num_tokens": 38251739.0,
      "step": 17675
    },
    {
      "entropy": 6.50155668258667,
      "epoch": 1.8919685376424635,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046481053729500516,
      "loss": 5.9202,
      "mean_token_accuracy": 0.15126555413007736,
      "num_tokens": 38262090.0,
      "step": 17680
    },
    {
      "entropy": 6.526460599899292,
      "epoch": 1.8925036117502274,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004647900044507855,
      "loss": 5.9351,
      "mean_token_accuracy": 0.16002599596977235,
      "num_tokens": 38272770.0,
      "step": 17685
    },
    {
      "entropy": 6.457193803787232,
      "epoch": 1.8930386858579913,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004647694661264061,
      "loss": 5.9042,
      "mean_token_accuracy": 0.16314941197633742,
      "num_tokens": 38283449.0,
      "step": 17690
    },
    {
      "entropy": 6.444292116165161,
      "epoch": 1.8935737599657552,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046474892232246,
      "loss": 5.9386,
      "mean_token_accuracy": 0.16346117109060287,
      "num_tokens": 38293831.0,
      "step": 17695
    },
    {
      "entropy": 6.462577867507934,
      "epoch": 1.8941088340735193,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046472837303954053,
      "loss": 5.9184,
      "mean_token_accuracy": 0.15528633818030357,
      "num_tokens": 38304782.0,
      "step": 17700
    },
    {
      "entropy": 6.46515588760376,
      "epoch": 1.8946439081812831,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046470781827824087,
      "loss": 5.9258,
      "mean_token_accuracy": 0.16735129952430725,
      "num_tokens": 38316547.0,
      "step": 17705
    },
    {
      "entropy": 6.489554071426392,
      "epoch": 1.895178982289047,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046468725803915464,
      "loss": 5.9635,
      "mean_token_accuracy": 0.16514952033758162,
      "num_tokens": 38329488.0,
      "step": 17710
    },
    {
      "entropy": 6.389694499969482,
      "epoch": 1.8957140563968111,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046466669232287543,
      "loss": 5.82,
      "mean_token_accuracy": 0.17053742706775665,
      "num_tokens": 38341290.0,
      "step": 17715
    },
    {
      "entropy": 6.46589937210083,
      "epoch": 1.8962491305045748,
      "grad_norm": 1.125,
      "learning_rate": 0.0004646461211299973,
      "loss": 5.9638,
      "mean_token_accuracy": 0.15508825182914734,
      "num_tokens": 38352445.0,
      "step": 17720
    },
    {
      "entropy": 6.430524396896362,
      "epoch": 1.8967842046123389,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004646255444611139,
      "loss": 5.955,
      "mean_token_accuracy": 0.16056891530752182,
      "num_tokens": 38363825.0,
      "step": 17725
    },
    {
      "entropy": 6.427944707870483,
      "epoch": 1.8973192787201028,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00046460496231681957,
      "loss": 6.012,
      "mean_token_accuracy": 0.16275362968444823,
      "num_tokens": 38375454.0,
      "step": 17730
    },
    {
      "entropy": 6.481643915176392,
      "epoch": 1.8978543528278666,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046458437469770866,
      "loss": 5.9733,
      "mean_token_accuracy": 0.15943110287189483,
      "num_tokens": 38385938.0,
      "step": 17735
    },
    {
      "entropy": 6.483687877655029,
      "epoch": 1.8983894269356307,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046456378160437555,
      "loss": 5.8913,
      "mean_token_accuracy": 0.16693678200244905,
      "num_tokens": 38396143.0,
      "step": 17740
    },
    {
      "entropy": 6.422693586349487,
      "epoch": 1.8989245010433944,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004645431830374149,
      "loss": 5.9859,
      "mean_token_accuracy": 0.15001035928726197,
      "num_tokens": 38407445.0,
      "step": 17745
    },
    {
      "entropy": 6.529591798782349,
      "epoch": 1.8994595751511585,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046452257899742144,
      "loss": 6.0173,
      "mean_token_accuracy": 0.145848585665226,
      "num_tokens": 38418438.0,
      "step": 17750
    },
    {
      "entropy": 6.474014377593994,
      "epoch": 1.8999946492589224,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046450196948499015,
      "loss": 5.8973,
      "mean_token_accuracy": 0.156736159324646,
      "num_tokens": 38428821.0,
      "step": 17755
    },
    {
      "entropy": 6.464312601089477,
      "epoch": 1.9005297233666862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004644813545007161,
      "loss": 5.8643,
      "mean_token_accuracy": 0.167196324467659,
      "num_tokens": 38439560.0,
      "step": 17760
    },
    {
      "entropy": 6.480480670928955,
      "epoch": 1.9010647974744503,
      "grad_norm": 1.0,
      "learning_rate": 0.0004644607340451946,
      "loss": 5.9602,
      "mean_token_accuracy": 0.1617266207933426,
      "num_tokens": 38449724.0,
      "step": 17765
    },
    {
      "entropy": 6.474592065811157,
      "epoch": 1.901599871582214,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046444010811902105,
      "loss": 5.9188,
      "mean_token_accuracy": 0.1581224411725998,
      "num_tokens": 38459833.0,
      "step": 17770
    },
    {
      "entropy": 6.4242205142974855,
      "epoch": 1.902134945689978,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046441947672279095,
      "loss": 5.9136,
      "mean_token_accuracy": 0.17115303874015808,
      "num_tokens": 38470682.0,
      "step": 17775
    },
    {
      "entropy": 6.496579170227051,
      "epoch": 1.902670019797742,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004643988398571001,
      "loss": 5.948,
      "mean_token_accuracy": 0.15786376148462294,
      "num_tokens": 38481362.0,
      "step": 17780
    },
    {
      "entropy": 6.510118865966797,
      "epoch": 1.9032050939055059,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004643781975225444,
      "loss": 5.9212,
      "mean_token_accuracy": 0.16369596868753433,
      "num_tokens": 38492941.0,
      "step": 17785
    },
    {
      "entropy": 6.543005847930909,
      "epoch": 1.90374016801327,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046435754971971976,
      "loss": 5.9574,
      "mean_token_accuracy": 0.15618983507156373,
      "num_tokens": 38503823.0,
      "step": 17790
    },
    {
      "entropy": 6.426774644851685,
      "epoch": 1.9042752421210336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046433689644922256,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16940215677022935,
      "num_tokens": 38514482.0,
      "step": 17795
    },
    {
      "entropy": 6.461113834381104,
      "epoch": 1.9048103162287977,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046431623771164897,
      "loss": 5.9604,
      "mean_token_accuracy": 0.1606248661875725,
      "num_tokens": 38525580.0,
      "step": 17800
    },
    {
      "entropy": 6.3908350467681885,
      "epoch": 1.9053453903365616,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004642955735075958,
      "loss": 5.9565,
      "mean_token_accuracy": 0.16313422173261644,
      "num_tokens": 38537025.0,
      "step": 17805
    },
    {
      "entropy": 6.436099338531494,
      "epoch": 1.9058804644443255,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046427490383765943,
      "loss": 5.9042,
      "mean_token_accuracy": 0.16353002488613128,
      "num_tokens": 38548974.0,
      "step": 17810
    },
    {
      "entropy": 6.4351380348205565,
      "epoch": 1.9064155385520896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004642542287024368,
      "loss": 5.8039,
      "mean_token_accuracy": 0.17682133615016937,
      "num_tokens": 38559971.0,
      "step": 17815
    },
    {
      "entropy": 6.4619852066040036,
      "epoch": 1.9069506126598534,
      "grad_norm": 1.078125,
      "learning_rate": 0.000464233548102525,
      "loss": 5.8914,
      "mean_token_accuracy": 0.16204525530338287,
      "num_tokens": 38570326.0,
      "step": 17820
    },
    {
      "entropy": 6.443046712875367,
      "epoch": 1.9074856867676173,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000464212862038521,
      "loss": 5.9117,
      "mean_token_accuracy": 0.16062327474355698,
      "num_tokens": 38580799.0,
      "step": 17825
    },
    {
      "entropy": 6.386077833175659,
      "epoch": 1.9080207608753812,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046419217051102223,
      "loss": 6.0032,
      "mean_token_accuracy": 0.15533971637487412,
      "num_tokens": 38591136.0,
      "step": 17830
    },
    {
      "entropy": 6.480170583724975,
      "epoch": 1.908555834983145,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004641714735206261,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16499500423669816,
      "num_tokens": 38602600.0,
      "step": 17835
    },
    {
      "entropy": 6.462655258178711,
      "epoch": 1.9090909090909092,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046415077106793024,
      "loss": 5.9172,
      "mean_token_accuracy": 0.16294852644205093,
      "num_tokens": 38612641.0,
      "step": 17840
    },
    {
      "entropy": 6.465324068069458,
      "epoch": 1.909625983198673,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004641300631535325,
      "loss": 5.9347,
      "mean_token_accuracy": 0.16009906977415084,
      "num_tokens": 38622873.0,
      "step": 17845
    },
    {
      "entropy": 6.447917318344116,
      "epoch": 1.910161057306437,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004641093497780307,
      "loss": 5.9174,
      "mean_token_accuracy": 0.1654792919754982,
      "num_tokens": 38633523.0,
      "step": 17850
    },
    {
      "entropy": 6.444404077529907,
      "epoch": 1.910696131414201,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004640886309420231,
      "loss": 5.9342,
      "mean_token_accuracy": 0.15472232103347777,
      "num_tokens": 38645635.0,
      "step": 17855
    },
    {
      "entropy": 6.491268825531006,
      "epoch": 1.9112312055219647,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004640679066461078,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1672418937087059,
      "num_tokens": 38656293.0,
      "step": 17860
    },
    {
      "entropy": 6.492960834503174,
      "epoch": 1.9117662796297288,
      "grad_norm": 1.0,
      "learning_rate": 0.0004640471768908832,
      "loss": 6.0217,
      "mean_token_accuracy": 0.15258216932415963,
      "num_tokens": 38667307.0,
      "step": 17865
    },
    {
      "entropy": 6.497925615310669,
      "epoch": 1.9123013537374927,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004640264416769481,
      "loss": 5.8947,
      "mean_token_accuracy": 0.16140078902244567,
      "num_tokens": 38677775.0,
      "step": 17870
    },
    {
      "entropy": 6.451355409622193,
      "epoch": 1.9128364278452565,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046400570100490086,
      "loss": 6.0004,
      "mean_token_accuracy": 0.15699757263064384,
      "num_tokens": 38688652.0,
      "step": 17875
    },
    {
      "entropy": 6.514651441574097,
      "epoch": 1.9133715019530206,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00046398495487534066,
      "loss": 5.9715,
      "mean_token_accuracy": 0.1535099282860756,
      "num_tokens": 38698694.0,
      "step": 17880
    },
    {
      "entropy": 6.544709348678589,
      "epoch": 1.9139065760607843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004639642032888664,
      "loss": 5.9686,
      "mean_token_accuracy": 0.16100693643093109,
      "num_tokens": 38709864.0,
      "step": 17885
    },
    {
      "entropy": 6.483394908905029,
      "epoch": 1.9144416501685484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004639434462460774,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15690088272094727,
      "num_tokens": 38721034.0,
      "step": 17890
    },
    {
      "entropy": 6.463530969619751,
      "epoch": 1.9149767242763123,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046392268374757285,
      "loss": 5.9738,
      "mean_token_accuracy": 0.15669871866703033,
      "num_tokens": 38731874.0,
      "step": 17895
    },
    {
      "entropy": 6.423438262939453,
      "epoch": 1.9155117983840761,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004639019157939524,
      "loss": 5.9085,
      "mean_token_accuracy": 0.1609168142080307,
      "num_tokens": 38742447.0,
      "step": 17900
    },
    {
      "entropy": 6.451863574981689,
      "epoch": 1.9160468724918402,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004638811423858157,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16079128980636598,
      "num_tokens": 38754117.0,
      "step": 17905
    },
    {
      "entropy": 6.457959175109863,
      "epoch": 1.916581946599604,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046386036352376255,
      "loss": 5.9475,
      "mean_token_accuracy": 0.15441786944866182,
      "num_tokens": 38765057.0,
      "step": 17910
    },
    {
      "entropy": 6.45240044593811,
      "epoch": 1.917117020707368,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046383957920839285,
      "loss": 5.8945,
      "mean_token_accuracy": 0.1648472100496292,
      "num_tokens": 38775938.0,
      "step": 17915
    },
    {
      "entropy": 6.4203619956970215,
      "epoch": 1.9176520948151319,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046381878944030697,
      "loss": 5.8352,
      "mean_token_accuracy": 0.1624774530529976,
      "num_tokens": 38787013.0,
      "step": 17920
    },
    {
      "entropy": 6.468729829788208,
      "epoch": 1.9181871689228958,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000463797994220105,
      "loss": 5.8605,
      "mean_token_accuracy": 0.16470538675785065,
      "num_tokens": 38798412.0,
      "step": 17925
    },
    {
      "entropy": 6.484822750091553,
      "epoch": 1.9187222430306599,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046377719354838756,
      "loss": 5.9567,
      "mean_token_accuracy": 0.15588985085487367,
      "num_tokens": 38809983.0,
      "step": 17930
    },
    {
      "entropy": 6.507261562347412,
      "epoch": 1.9192573171384235,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046375638742575506,
      "loss": 5.9386,
      "mean_token_accuracy": 0.15925129055976867,
      "num_tokens": 38820418.0,
      "step": 17935
    },
    {
      "entropy": 6.511274099349976,
      "epoch": 1.9197923912461876,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004637355758528085,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15776665955781938,
      "num_tokens": 38831482.0,
      "step": 17940
    },
    {
      "entropy": 6.545492458343506,
      "epoch": 1.9203274653539515,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004637147588301486,
      "loss": 5.9793,
      "mean_token_accuracy": 0.15374392420053482,
      "num_tokens": 38842110.0,
      "step": 17945
    },
    {
      "entropy": 6.4252448081970215,
      "epoch": 1.9208625394617154,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046369393635837657,
      "loss": 5.8331,
      "mean_token_accuracy": 0.1586639404296875,
      "num_tokens": 38852861.0,
      "step": 17950
    },
    {
      "entropy": 6.463755369186401,
      "epoch": 1.9213976135694795,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046367310843809374,
      "loss": 5.9729,
      "mean_token_accuracy": 0.15403397008776665,
      "num_tokens": 38864067.0,
      "step": 17955
    },
    {
      "entropy": 6.481246566772461,
      "epoch": 1.9219326876772433,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046365227506990137,
      "loss": 5.9322,
      "mean_token_accuracy": 0.15614519715309144,
      "num_tokens": 38875363.0,
      "step": 17960
    },
    {
      "entropy": 6.5344151020050045,
      "epoch": 1.9224677617850072,
      "grad_norm": 1.0,
      "learning_rate": 0.0004636314362544011,
      "loss": 5.9578,
      "mean_token_accuracy": 0.16470584869384766,
      "num_tokens": 38886414.0,
      "step": 17965
    },
    {
      "entropy": 6.503572225570679,
      "epoch": 1.923002835892771,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00046361059199219457,
      "loss": 6.0241,
      "mean_token_accuracy": 0.15635054409503937,
      "num_tokens": 38897708.0,
      "step": 17970
    },
    {
      "entropy": 6.39807710647583,
      "epoch": 1.923537910000535,
      "grad_norm": 1.0,
      "learning_rate": 0.0004635897422838837,
      "loss": 5.8938,
      "mean_token_accuracy": 0.16421412378549577,
      "num_tokens": 38908683.0,
      "step": 17975
    },
    {
      "entropy": 6.388296031951905,
      "epoch": 1.924072984108299,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046356888713007055,
      "loss": 5.8708,
      "mean_token_accuracy": 0.17081587463617326,
      "num_tokens": 38919279.0,
      "step": 17980
    },
    {
      "entropy": 6.531394338607788,
      "epoch": 1.924608058216063,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046354802653135723,
      "loss": 5.9704,
      "mean_token_accuracy": 0.158871853351593,
      "num_tokens": 38929353.0,
      "step": 17985
    },
    {
      "entropy": 6.494729280471802,
      "epoch": 1.9251431323238268,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004635271604883462,
      "loss": 5.9086,
      "mean_token_accuracy": 0.16518499404191972,
      "num_tokens": 38939296.0,
      "step": 17990
    },
    {
      "entropy": 6.420190763473511,
      "epoch": 1.925678206431591,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046350628900163987,
      "loss": 5.925,
      "mean_token_accuracy": 0.16411629915237427,
      "num_tokens": 38950149.0,
      "step": 17995
    },
    {
      "entropy": 6.469804763793945,
      "epoch": 1.9262132805393546,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000463485412071841,
      "loss": 5.9827,
      "mean_token_accuracy": 0.15814982503652572,
      "num_tokens": 38961052.0,
      "step": 18000
    },
    {
      "epoch": 1.9262132805393546,
      "eval_entropy": 6.273194344667091,
      "eval_loss": 6.002620697021484,
      "eval_mean_token_accuracy": 0.16490907197130336,
      "eval_num_tokens": 38961052.0,
      "eval_runtime": 17.5593,
      "eval_samples_per_second": 1690.215,
      "eval_steps_per_second": 211.284,
      "step": 18000
    },
    {
      "entropy": 6.524932336807251,
      "epoch": 1.9267483546471187,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046346452969955214,
      "loss": 5.9709,
      "mean_token_accuracy": 0.15824434757232667,
      "num_tokens": 38972344.0,
      "step": 18005
    },
    {
      "entropy": 6.528006982803345,
      "epoch": 1.9272834287548826,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004634436418853767,
      "loss": 5.9532,
      "mean_token_accuracy": 0.15189577937126159,
      "num_tokens": 38982877.0,
      "step": 18010
    },
    {
      "entropy": 6.508360719680786,
      "epoch": 1.9278185028626464,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046342274862991745,
      "loss": 5.9404,
      "mean_token_accuracy": 0.161869552731514,
      "num_tokens": 38994322.0,
      "step": 18015
    },
    {
      "entropy": 6.4798671245574955,
      "epoch": 1.9283535769704105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046340184993377793,
      "loss": 5.8828,
      "mean_token_accuracy": 0.16501724272966384,
      "num_tokens": 39003447.0,
      "step": 18020
    },
    {
      "entropy": 6.4705610275268555,
      "epoch": 1.9288886510781742,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046338094579756136,
      "loss": 5.9452,
      "mean_token_accuracy": 0.15464228093624116,
      "num_tokens": 39013760.0,
      "step": 18025
    },
    {
      "entropy": 6.330158662796021,
      "epoch": 1.9294237251859383,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00046336003622187143,
      "loss": 5.7663,
      "mean_token_accuracy": 0.1717679962515831,
      "num_tokens": 39025559.0,
      "step": 18030
    },
    {
      "entropy": 6.45264196395874,
      "epoch": 1.9299587992937022,
      "grad_norm": 0.98046875,
      "learning_rate": 0.000463339121207312,
      "loss": 5.9301,
      "mean_token_accuracy": 0.15625325590372086,
      "num_tokens": 39036004.0,
      "step": 18035
    },
    {
      "entropy": 6.483163166046142,
      "epoch": 1.930493873401466,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004633182007544868,
      "loss": 5.9538,
      "mean_token_accuracy": 0.1617903634905815,
      "num_tokens": 39047251.0,
      "step": 18040
    },
    {
      "entropy": 6.497071647644043,
      "epoch": 1.9310289475092302,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046329727486400015,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15793880671262742,
      "num_tokens": 39057255.0,
      "step": 18045
    },
    {
      "entropy": 6.517926740646362,
      "epoch": 1.9315640216169938,
      "grad_norm": 1.0,
      "learning_rate": 0.0004632763435364561,
      "loss": 5.9798,
      "mean_token_accuracy": 0.15558959990739823,
      "num_tokens": 39068186.0,
      "step": 18050
    },
    {
      "entropy": 6.398785829544067,
      "epoch": 1.932099095724758,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046325540677245907,
      "loss": 5.8951,
      "mean_token_accuracy": 0.15932125151157378,
      "num_tokens": 39079173.0,
      "step": 18055
    },
    {
      "entropy": 6.478333330154419,
      "epoch": 1.9326341698325218,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046323446457261365,
      "loss": 5.9696,
      "mean_token_accuracy": 0.16038255840539933,
      "num_tokens": 39089939.0,
      "step": 18060
    },
    {
      "entropy": 6.455627727508545,
      "epoch": 1.9331692439402857,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004632135169375245,
      "loss": 5.9535,
      "mean_token_accuracy": 0.1628076195716858,
      "num_tokens": 39100464.0,
      "step": 18065
    },
    {
      "entropy": 6.518622350692749,
      "epoch": 1.9337043180480498,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004631925638677965,
      "loss": 5.9083,
      "mean_token_accuracy": 0.16252013593912124,
      "num_tokens": 39111161.0,
      "step": 18070
    },
    {
      "entropy": 6.458624649047851,
      "epoch": 1.9342393921558134,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046317160536403463,
      "loss": 5.9263,
      "mean_token_accuracy": 0.16784788966178893,
      "num_tokens": 39122469.0,
      "step": 18075
    },
    {
      "entropy": 6.438329601287842,
      "epoch": 1.9347744662635775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046315064142684416,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16274093687534333,
      "num_tokens": 39131871.0,
      "step": 18080
    },
    {
      "entropy": 6.4832062244415285,
      "epoch": 1.9353095403713414,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004631296720568304,
      "loss": 5.9321,
      "mean_token_accuracy": 0.16050767600536348,
      "num_tokens": 39143061.0,
      "step": 18085
    },
    {
      "entropy": 6.421927642822266,
      "epoch": 1.9358446144791053,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004631086972545987,
      "loss": 5.9547,
      "mean_token_accuracy": 0.1653192386031151,
      "num_tokens": 39152452.0,
      "step": 18090
    },
    {
      "entropy": 6.401672267913819,
      "epoch": 1.9363796885868694,
      "grad_norm": 1.0,
      "learning_rate": 0.0004630877170207548,
      "loss": 5.8986,
      "mean_token_accuracy": 0.16243208944797516,
      "num_tokens": 39163196.0,
      "step": 18095
    },
    {
      "entropy": 6.406721305847168,
      "epoch": 1.9369147626946333,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004630667313559045,
      "loss": 5.8762,
      "mean_token_accuracy": 0.160360486805439,
      "num_tokens": 39173958.0,
      "step": 18100
    },
    {
      "entropy": 6.474235010147095,
      "epoch": 1.9374498368023971,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046304574026065376,
      "loss": 5.9167,
      "mean_token_accuracy": 0.17297814637422562,
      "num_tokens": 39184483.0,
      "step": 18105
    },
    {
      "entropy": 6.457554578781128,
      "epoch": 1.937984910910161,
      "grad_norm": 0.88671875,
      "learning_rate": 0.0004630247437356087,
      "loss": 5.9556,
      "mean_token_accuracy": 0.15626244768500328,
      "num_tokens": 39196500.0,
      "step": 18110
    },
    {
      "entropy": 6.5098470687866214,
      "epoch": 1.9385199850179249,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00046300374178137566,
      "loss": 6.0097,
      "mean_token_accuracy": 0.15174277275800704,
      "num_tokens": 39207492.0,
      "step": 18115
    },
    {
      "entropy": 6.461736583709717,
      "epoch": 1.939055059125689,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004629827343985609,
      "loss": 5.9264,
      "mean_token_accuracy": 0.16412588059902192,
      "num_tokens": 39218204.0,
      "step": 18120
    },
    {
      "entropy": 6.502477359771729,
      "epoch": 1.9395901332334529,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046296172158777116,
      "loss": 5.9837,
      "mean_token_accuracy": 0.1553572252392769,
      "num_tokens": 39229646.0,
      "step": 18125
    },
    {
      "entropy": 6.464808940887451,
      "epoch": 1.9401252073412167,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004629407033496131,
      "loss": 5.9299,
      "mean_token_accuracy": 0.16154939383268357,
      "num_tokens": 39240151.0,
      "step": 18130
    },
    {
      "entropy": 6.489027214050293,
      "epoch": 1.9406602814489808,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004629196796846936,
      "loss": 5.9856,
      "mean_token_accuracy": 0.163087297976017,
      "num_tokens": 39251571.0,
      "step": 18135
    },
    {
      "entropy": 6.517105007171631,
      "epoch": 1.9411953555567445,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046289865059361983,
      "loss": 5.9179,
      "mean_token_accuracy": 0.15798383951187134,
      "num_tokens": 39262937.0,
      "step": 18140
    },
    {
      "entropy": 6.50760326385498,
      "epoch": 1.9417304296645086,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046287761607699887,
      "loss": 6.0069,
      "mean_token_accuracy": 0.15683023929595946,
      "num_tokens": 39274354.0,
      "step": 18145
    },
    {
      "entropy": 6.478582143783569,
      "epoch": 1.9422655037722725,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004628565761354382,
      "loss": 5.8966,
      "mean_token_accuracy": 0.1629406675696373,
      "num_tokens": 39284314.0,
      "step": 18150
    },
    {
      "entropy": 6.446175050735474,
      "epoch": 1.9428005778800364,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046283553076954527,
      "loss": 5.9211,
      "mean_token_accuracy": 0.15867335498332977,
      "num_tokens": 39295516.0,
      "step": 18155
    },
    {
      "entropy": 6.508010149002075,
      "epoch": 1.9433356519878004,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004628144799799277,
      "loss": 5.9155,
      "mean_token_accuracy": 0.16166880130767822,
      "num_tokens": 39307221.0,
      "step": 18160
    },
    {
      "entropy": 6.465686702728272,
      "epoch": 1.943870726095564,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004627934237671935,
      "loss": 5.9567,
      "mean_token_accuracy": 0.15754917711019517,
      "num_tokens": 39318448.0,
      "step": 18165
    },
    {
      "entropy": 6.4157696723937985,
      "epoch": 1.9444058002033282,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046277236213195064,
      "loss": 5.8344,
      "mean_token_accuracy": 0.17190908640623093,
      "num_tokens": 39329384.0,
      "step": 18170
    },
    {
      "entropy": 6.466805124282837,
      "epoch": 1.944940874311092,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046275129507480705,
      "loss": 5.9838,
      "mean_token_accuracy": 0.15652558207511902,
      "num_tokens": 39340342.0,
      "step": 18175
    },
    {
      "entropy": 6.349398612976074,
      "epoch": 1.945475948418856,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004627302225963713,
      "loss": 5.8468,
      "mean_token_accuracy": 0.16820669323205947,
      "num_tokens": 39350825.0,
      "step": 18180
    },
    {
      "entropy": 6.542615747451782,
      "epoch": 1.94601102252662,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046270914469725174,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15706620514392852,
      "num_tokens": 39361986.0,
      "step": 18185
    },
    {
      "entropy": 6.503807783126831,
      "epoch": 1.9465460966343837,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046268806137805704,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1641916111111641,
      "num_tokens": 39372440.0,
      "step": 18190
    },
    {
      "entropy": 6.44639139175415,
      "epoch": 1.9470811707421478,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004626669726393958,
      "loss": 5.9049,
      "mean_token_accuracy": 0.17297554463148118,
      "num_tokens": 39383455.0,
      "step": 18195
    },
    {
      "entropy": 6.439597749710083,
      "epoch": 1.9476162448499117,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004626458784818772,
      "loss": 5.9375,
      "mean_token_accuracy": 0.1618105575442314,
      "num_tokens": 39393511.0,
      "step": 18200
    },
    {
      "entropy": 6.431254196166992,
      "epoch": 1.9481513189576756,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00046262477890611026,
      "loss": 5.9221,
      "mean_token_accuracy": 0.16573905050754548,
      "num_tokens": 39404049.0,
      "step": 18205
    },
    {
      "entropy": 6.492399215698242,
      "epoch": 1.9486863930654397,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046260367391270414,
      "loss": 5.9126,
      "mean_token_accuracy": 0.16284340918064116,
      "num_tokens": 39414771.0,
      "step": 18210
    },
    {
      "entropy": 6.428909111022949,
      "epoch": 1.9492214671732033,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046258256350226835,
      "loss": 5.906,
      "mean_token_accuracy": 0.1671500325202942,
      "num_tokens": 39424869.0,
      "step": 18215
    },
    {
      "entropy": 6.484317207336426,
      "epoch": 1.9497565412809674,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004625614476754123,
      "loss": 5.9754,
      "mean_token_accuracy": 0.1605912961065769,
      "num_tokens": 39435442.0,
      "step": 18220
    },
    {
      "entropy": 6.449296522140503,
      "epoch": 1.9502916153887313,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00046254032643274596,
      "loss": 5.9089,
      "mean_token_accuracy": 0.16540035009384155,
      "num_tokens": 39446704.0,
      "step": 18225
    },
    {
      "entropy": 6.410387563705444,
      "epoch": 1.9508266894964952,
      "grad_norm": 0.9375,
      "learning_rate": 0.00046251919977487887,
      "loss": 5.919,
      "mean_token_accuracy": 0.16384106874465942,
      "num_tokens": 39458079.0,
      "step": 18230
    },
    {
      "entropy": 6.479158926010132,
      "epoch": 1.9513617636042593,
      "grad_norm": 0.984375,
      "learning_rate": 0.00046249806770242135,
      "loss": 5.9133,
      "mean_token_accuracy": 0.16938420832157136,
      "num_tokens": 39469867.0,
      "step": 18235
    },
    {
      "entropy": 6.432933902740478,
      "epoch": 1.9518968377120232,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004624769302159834,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15640245676040648,
      "num_tokens": 39481138.0,
      "step": 18240
    },
    {
      "entropy": 6.439736461639404,
      "epoch": 1.952431911819787,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004624557873161755,
      "loss": 5.8905,
      "mean_token_accuracy": 0.16263058483600618,
      "num_tokens": 39492066.0,
      "step": 18245
    },
    {
      "entropy": 6.530600547790527,
      "epoch": 1.952966985927551,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004624346390036081,
      "loss": 5.9441,
      "mean_token_accuracy": 0.15769539922475814,
      "num_tokens": 39502576.0,
      "step": 18250
    },
    {
      "entropy": 6.4326740264892575,
      "epoch": 1.9535020600353148,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004624134852788918,
      "loss": 5.9028,
      "mean_token_accuracy": 0.15925951898097992,
      "num_tokens": 39513373.0,
      "step": 18255
    },
    {
      "entropy": 6.469168186187744,
      "epoch": 1.954037134143079,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004623923261426374,
      "loss": 5.9675,
      "mean_token_accuracy": 0.1567094311118126,
      "num_tokens": 39523970.0,
      "step": 18260
    },
    {
      "entropy": 6.458700847625733,
      "epoch": 1.9545722082508428,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000462371161595456,
      "loss": 5.9488,
      "mean_token_accuracy": 0.1626107409596443,
      "num_tokens": 39533893.0,
      "step": 18265
    },
    {
      "entropy": 6.455472993850708,
      "epoch": 1.9551072823586066,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004623499916379586,
      "loss": 5.9154,
      "mean_token_accuracy": 0.157929165661335,
      "num_tokens": 39544276.0,
      "step": 18270
    },
    {
      "entropy": 6.490838956832886,
      "epoch": 1.9556423564663707,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004623288162707565,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16644099056720735,
      "num_tokens": 39554443.0,
      "step": 18275
    },
    {
      "entropy": 6.475710010528564,
      "epoch": 1.9561774305741344,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046230763549446116,
      "loss": 5.8585,
      "mean_token_accuracy": 0.1635194569826126,
      "num_tokens": 39564508.0,
      "step": 18280
    },
    {
      "entropy": 6.517408180236816,
      "epoch": 1.9567125046818985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004622864493096842,
      "loss": 5.941,
      "mean_token_accuracy": 0.15448608100414277,
      "num_tokens": 39574757.0,
      "step": 18285
    },
    {
      "entropy": 6.487943649291992,
      "epoch": 1.9572475787896624,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00046226525771703733,
      "loss": 6.001,
      "mean_token_accuracy": 0.1551746428012848,
      "num_tokens": 39585911.0,
      "step": 18290
    },
    {
      "entropy": 6.489192819595337,
      "epoch": 1.9577826528974263,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004622440607171324,
      "loss": 5.917,
      "mean_token_accuracy": 0.15961997359991073,
      "num_tokens": 39597443.0,
      "step": 18295
    },
    {
      "entropy": 6.496833276748657,
      "epoch": 1.9583177270051904,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046222285831058155,
      "loss": 5.9968,
      "mean_token_accuracy": 0.15467360466718674,
      "num_tokens": 39608693.0,
      "step": 18300
    },
    {
      "entropy": 6.4829507827758786,
      "epoch": 1.958852801112954,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000462201650497997,
      "loss": 5.965,
      "mean_token_accuracy": 0.16452267915010452,
      "num_tokens": 39620002.0,
      "step": 18305
    },
    {
      "entropy": 6.45285005569458,
      "epoch": 1.9593878752207181,
      "grad_norm": 1.015625,
      "learning_rate": 0.00046218043727999105,
      "loss": 5.9335,
      "mean_token_accuracy": 0.162069870531559,
      "num_tokens": 39629856.0,
      "step": 18310
    },
    {
      "entropy": 6.379008483886719,
      "epoch": 1.959922949328482,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046215921865717637,
      "loss": 5.8941,
      "mean_token_accuracy": 0.16670921593904495,
      "num_tokens": 39640418.0,
      "step": 18315
    },
    {
      "entropy": 6.538016748428345,
      "epoch": 1.9604580234362459,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004621379946301655,
      "loss": 6.0064,
      "mean_token_accuracy": 0.15381101965904237,
      "num_tokens": 39652871.0,
      "step": 18320
    },
    {
      "entropy": 6.510451126098633,
      "epoch": 1.96099309754401,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004621167651995713,
      "loss": 5.9128,
      "mean_token_accuracy": 0.16266919672489166,
      "num_tokens": 39663864.0,
      "step": 18325
    },
    {
      "entropy": 6.410566425323486,
      "epoch": 1.9615281716517736,
      "grad_norm": 1.0,
      "learning_rate": 0.0004620955303660068,
      "loss": 5.902,
      "mean_token_accuracy": 0.16757538318634033,
      "num_tokens": 39675835.0,
      "step": 18330
    },
    {
      "entropy": 6.481011295318604,
      "epoch": 1.9620632457595377,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046207429013008525,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15804524719715118,
      "num_tokens": 39686799.0,
      "step": 18335
    },
    {
      "entropy": 6.566179895401001,
      "epoch": 1.9625983198673016,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046205304449241976,
      "loss": 5.9248,
      "mean_token_accuracy": 0.16302954405546188,
      "num_tokens": 39697564.0,
      "step": 18340
    },
    {
      "entropy": 6.477157688140869,
      "epoch": 1.9631333939750655,
      "grad_norm": 1.125,
      "learning_rate": 0.0004620317934536239,
      "loss": 5.934,
      "mean_token_accuracy": 0.16124674081802368,
      "num_tokens": 39708064.0,
      "step": 18345
    },
    {
      "entropy": 6.364507389068604,
      "epoch": 1.9636684680828296,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004620105370143112,
      "loss": 5.832,
      "mean_token_accuracy": 0.16664274483919145,
      "num_tokens": 39718076.0,
      "step": 18350
    },
    {
      "entropy": 6.462129878997803,
      "epoch": 1.9642035421905932,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004619892751750957,
      "loss": 5.9075,
      "mean_token_accuracy": 0.16285010874271394,
      "num_tokens": 39729532.0,
      "step": 18355
    },
    {
      "entropy": 6.551578044891357,
      "epoch": 1.9647386162983573,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046196800793659116,
      "loss": 5.9214,
      "mean_token_accuracy": 0.1624340832233429,
      "num_tokens": 39740372.0,
      "step": 18360
    },
    {
      "entropy": 6.449893474578857,
      "epoch": 1.9652736904061212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004619467352994115,
      "loss": 5.7981,
      "mean_token_accuracy": 0.1691265121102333,
      "num_tokens": 39751283.0,
      "step": 18365
    },
    {
      "entropy": 6.437889289855957,
      "epoch": 1.965808764513885,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004619254572641712,
      "loss": 6.0059,
      "mean_token_accuracy": 0.15606431663036346,
      "num_tokens": 39763295.0,
      "step": 18370
    },
    {
      "entropy": 6.467307090759277,
      "epoch": 1.9663438386216492,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046190417383148465,
      "loss": 5.9403,
      "mean_token_accuracy": 0.16092216968536377,
      "num_tokens": 39773849.0,
      "step": 18375
    },
    {
      "entropy": 6.389621114730835,
      "epoch": 1.966878912729413,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004618828850019663,
      "loss": 5.7577,
      "mean_token_accuracy": 0.17260676324367524,
      "num_tokens": 39784366.0,
      "step": 18380
    },
    {
      "entropy": 6.468810796737671,
      "epoch": 1.967413986837177,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00046186159077623093,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15842816531658171,
      "num_tokens": 39794226.0,
      "step": 18385
    },
    {
      "entropy": 6.408194971084595,
      "epoch": 1.9679490609449408,
      "grad_norm": 1.0,
      "learning_rate": 0.0004618402911548933,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1597668632864952,
      "num_tokens": 39805557.0,
      "step": 18390
    },
    {
      "entropy": 6.457835245132446,
      "epoch": 1.9684841350527047,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004618189861385686,
      "loss": 5.8708,
      "mean_token_accuracy": 0.16851210296154023,
      "num_tokens": 39816090.0,
      "step": 18395
    },
    {
      "entropy": 6.490698194503784,
      "epoch": 1.9690192091604688,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004617976757278719,
      "loss": 5.9911,
      "mean_token_accuracy": 0.15069078505039216,
      "num_tokens": 39826662.0,
      "step": 18400
    },
    {
      "entropy": 6.510570287704468,
      "epoch": 1.9695542832682327,
      "grad_norm": 1.125,
      "learning_rate": 0.00046177635992341855,
      "loss": 5.958,
      "mean_token_accuracy": 0.1587168738245964,
      "num_tokens": 39837805.0,
      "step": 18405
    },
    {
      "entropy": 6.512331104278564,
      "epoch": 1.9700893573759966,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004617550387258241,
      "loss": 5.9515,
      "mean_token_accuracy": 0.1533978372812271,
      "num_tokens": 39848052.0,
      "step": 18410
    },
    {
      "entropy": 6.4840271949768065,
      "epoch": 1.9706244314837607,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004617337121357041,
      "loss": 5.9473,
      "mean_token_accuracy": 0.15984692871570588,
      "num_tokens": 39859448.0,
      "step": 18415
    },
    {
      "entropy": 6.501749897003174,
      "epoch": 1.9711595055915243,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004617123801536744,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15918989032506942,
      "num_tokens": 39870053.0,
      "step": 18420
    },
    {
      "entropy": 6.528685140609741,
      "epoch": 1.9716945796992884,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000461691042780351,
      "loss": 6.0148,
      "mean_token_accuracy": 0.1546623170375824,
      "num_tokens": 39881547.0,
      "step": 18425
    },
    {
      "entropy": 6.473186063766479,
      "epoch": 1.9722296538070523,
      "grad_norm": 1.0,
      "learning_rate": 0.00046166970001635,
      "loss": 5.9834,
      "mean_token_accuracy": 0.16102519780397415,
      "num_tokens": 39892974.0,
      "step": 18430
    },
    {
      "entropy": 6.477826452255249,
      "epoch": 1.9727647279148162,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00046164835186228757,
      "loss": 5.9543,
      "mean_token_accuracy": 0.16319599449634553,
      "num_tokens": 39904210.0,
      "step": 18435
    },
    {
      "entropy": 6.4631730079650875,
      "epoch": 1.9732998020225803,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046162699831878026,
      "loss": 5.9158,
      "mean_token_accuracy": 0.16038064658641815,
      "num_tokens": 39915464.0,
      "step": 18440
    },
    {
      "entropy": 6.412339496612549,
      "epoch": 1.973834876130344,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004616056393864446,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1628278061747551,
      "num_tokens": 39927117.0,
      "step": 18445
    },
    {
      "entropy": 6.480841445922851,
      "epoch": 1.974369950238108,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004615842750658972,
      "loss": 5.9432,
      "mean_token_accuracy": 0.161232990026474,
      "num_tokens": 39937363.0,
      "step": 18450
    },
    {
      "entropy": 6.4120166301727295,
      "epoch": 1.974905024345872,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004615629053577552,
      "loss": 5.838,
      "mean_token_accuracy": 0.16420925855636598,
      "num_tokens": 39948934.0,
      "step": 18455
    },
    {
      "entropy": 6.435087442398071,
      "epoch": 1.9754400984536358,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004615415302626355,
      "loss": 5.8759,
      "mean_token_accuracy": 0.16903770267963408,
      "num_tokens": 39959435.0,
      "step": 18460
    },
    {
      "entropy": 6.452201461791992,
      "epoch": 1.9759751725613999,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046152014978115535,
      "loss": 5.9436,
      "mean_token_accuracy": 0.16079553365707397,
      "num_tokens": 39969006.0,
      "step": 18465
    },
    {
      "entropy": 6.463232564926147,
      "epoch": 1.9765102466691635,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000461498763913932,
      "loss": 5.8755,
      "mean_token_accuracy": 0.1586931064724922,
      "num_tokens": 39979264.0,
      "step": 18470
    },
    {
      "entropy": 6.537736558914185,
      "epoch": 1.9770453207769276,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046147737266158314,
      "loss": 6.0285,
      "mean_token_accuracy": 0.15352119207382203,
      "num_tokens": 39989791.0,
      "step": 18475
    },
    {
      "entropy": 6.452395057678222,
      "epoch": 1.9775803948846915,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004614559760247263,
      "loss": 5.8993,
      "mean_token_accuracy": 0.16246129870414733,
      "num_tokens": 39999591.0,
      "step": 18480
    },
    {
      "entropy": 6.438455390930176,
      "epoch": 1.9781154689924554,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004614345740039794,
      "loss": 6.0155,
      "mean_token_accuracy": 0.15853001326322555,
      "num_tokens": 40010811.0,
      "step": 18485
    },
    {
      "entropy": 6.490279006958008,
      "epoch": 1.9786505431002195,
      "grad_norm": 0.9453125,
      "learning_rate": 0.00046141316659996037,
      "loss": 5.9643,
      "mean_token_accuracy": 0.15916822850704193,
      "num_tokens": 40022190.0,
      "step": 18490
    },
    {
      "entropy": 6.493713235855102,
      "epoch": 1.9791856172079834,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00046139175381328725,
      "loss": 5.9315,
      "mean_token_accuracy": 0.15728117972612382,
      "num_tokens": 40033905.0,
      "step": 18495
    },
    {
      "entropy": 6.499391174316406,
      "epoch": 1.9797206913157472,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004613703356445786,
      "loss": 5.9842,
      "mean_token_accuracy": 0.15830520838499068,
      "num_tokens": 40044923.0,
      "step": 18500
    },
    {
      "entropy": 6.455191659927368,
      "epoch": 1.9802557654235111,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004613489120944526,
      "loss": 5.8799,
      "mean_token_accuracy": 0.16706508547067642,
      "num_tokens": 40054988.0,
      "step": 18505
    },
    {
      "entropy": 6.504331731796265,
      "epoch": 1.980790839531275,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004613274831635279,
      "loss": 6.0037,
      "mean_token_accuracy": 0.15722417384386062,
      "num_tokens": 40065565.0,
      "step": 18510
    },
    {
      "entropy": 6.46919355392456,
      "epoch": 1.981325913639039,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046130604885242333,
      "loss": 5.9727,
      "mean_token_accuracy": 0.1632459506392479,
      "num_tokens": 40076282.0,
      "step": 18515
    },
    {
      "entropy": 6.463894271850586,
      "epoch": 1.981860987746803,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004612846091617578,
      "loss": 5.917,
      "mean_token_accuracy": 0.1643184944987297,
      "num_tokens": 40087546.0,
      "step": 18520
    },
    {
      "entropy": 6.488373422622681,
      "epoch": 1.9823960618545668,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004612631640921504,
      "loss": 5.9703,
      "mean_token_accuracy": 0.16306506097316742,
      "num_tokens": 40097937.0,
      "step": 18525
    },
    {
      "entropy": 6.509903144836426,
      "epoch": 1.982931135962331,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046124171364422025,
      "loss": 5.9362,
      "mean_token_accuracy": 0.16380630433559418,
      "num_tokens": 40108639.0,
      "step": 18530
    },
    {
      "entropy": 6.451324939727783,
      "epoch": 1.9834662100700946,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004612202578185868,
      "loss": 5.9296,
      "mean_token_accuracy": 0.1635017514228821,
      "num_tokens": 40119533.0,
      "step": 18535
    },
    {
      "entropy": 6.522223854064942,
      "epoch": 1.9840012841778587,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004611987966158695,
      "loss": 5.9383,
      "mean_token_accuracy": 0.15966727510094642,
      "num_tokens": 40131176.0,
      "step": 18540
    },
    {
      "entropy": 6.490710735321045,
      "epoch": 1.9845363582856226,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046117733003668825,
      "loss": 5.9185,
      "mean_token_accuracy": 0.16538300961256028,
      "num_tokens": 40143293.0,
      "step": 18545
    },
    {
      "entropy": 6.497122001647949,
      "epoch": 1.9850714323933865,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004611558580816628,
      "loss": 5.888,
      "mean_token_accuracy": 0.16010850965976714,
      "num_tokens": 40155312.0,
      "step": 18550
    },
    {
      "entropy": 6.47957935333252,
      "epoch": 1.9856065065011506,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000461134380751413,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1605210170149803,
      "num_tokens": 40166261.0,
      "step": 18555
    },
    {
      "entropy": 6.439353656768799,
      "epoch": 1.9861415806089142,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046111289804655906,
      "loss": 5.888,
      "mean_token_accuracy": 0.16698995977640152,
      "num_tokens": 40176641.0,
      "step": 18560
    },
    {
      "entropy": 6.477392721176147,
      "epoch": 1.9866766547166783,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004610914099677214,
      "loss": 5.9602,
      "mean_token_accuracy": 0.1590610772371292,
      "num_tokens": 40187784.0,
      "step": 18565
    },
    {
      "entropy": 6.434626293182373,
      "epoch": 1.9872117288244422,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004610699165155204,
      "loss": 5.9582,
      "mean_token_accuracy": 0.15608287006616592,
      "num_tokens": 40198219.0,
      "step": 18570
    },
    {
      "entropy": 6.45184965133667,
      "epoch": 1.987746802932206,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046104841769057677,
      "loss": 5.9271,
      "mean_token_accuracy": 0.15973594933748245,
      "num_tokens": 40208589.0,
      "step": 18575
    },
    {
      "entropy": 6.508424472808838,
      "epoch": 1.9882818770399702,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004610269134935112,
      "loss": 5.8977,
      "mean_token_accuracy": 0.16617718786001207,
      "num_tokens": 40219587.0,
      "step": 18580
    },
    {
      "entropy": 6.500035905838013,
      "epoch": 1.9888169511477338,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004610054039249446,
      "loss": 5.9242,
      "mean_token_accuracy": 0.16414874792099,
      "num_tokens": 40230681.0,
      "step": 18585
    },
    {
      "entropy": 6.372467756271362,
      "epoch": 1.989352025255498,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046098388898549816,
      "loss": 5.8907,
      "mean_token_accuracy": 0.1668432503938675,
      "num_tokens": 40241356.0,
      "step": 18590
    },
    {
      "entropy": 6.469804191589356,
      "epoch": 1.9898870993632618,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00046096236867579306,
      "loss": 5.9142,
      "mean_token_accuracy": 0.16379387825727462,
      "num_tokens": 40251652.0,
      "step": 18595
    },
    {
      "entropy": 6.463411140441894,
      "epoch": 1.9904221734710257,
      "grad_norm": 1.078125,
      "learning_rate": 0.00046094084299645065,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16540828943252564,
      "num_tokens": 40262819.0,
      "step": 18600
    },
    {
      "entropy": 6.522915029525757,
      "epoch": 1.9909572475787898,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046091931194809253,
      "loss": 6.0012,
      "mean_token_accuracy": 0.15167615711688995,
      "num_tokens": 40274314.0,
      "step": 18605
    },
    {
      "entropy": 6.513075542449951,
      "epoch": 1.9914923216865534,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004608977755313404,
      "loss": 5.941,
      "mean_token_accuracy": 0.1574372410774231,
      "num_tokens": 40284733.0,
      "step": 18610
    },
    {
      "entropy": 6.45805025100708,
      "epoch": 1.9920273957943175,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004608762337468162,
      "loss": 5.9233,
      "mean_token_accuracy": 0.16204752326011657,
      "num_tokens": 40295484.0,
      "step": 18615
    },
    {
      "entropy": 6.48737359046936,
      "epoch": 1.9925624699020814,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004608546865951418,
      "loss": 5.8961,
      "mean_token_accuracy": 0.16682881861925125,
      "num_tokens": 40305470.0,
      "step": 18620
    },
    {
      "entropy": 6.489409399032593,
      "epoch": 1.9930975440098453,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046083313407693945,
      "loss": 6.033,
      "mean_token_accuracy": 0.15291958302259445,
      "num_tokens": 40316055.0,
      "step": 18625
    },
    {
      "entropy": 6.489158868789673,
      "epoch": 1.9936326181176094,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004608115761928315,
      "loss": 5.9335,
      "mean_token_accuracy": 0.15997691601514816,
      "num_tokens": 40326689.0,
      "step": 18630
    },
    {
      "entropy": 6.4842212200164795,
      "epoch": 1.9941676922253733,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00046079001294344036,
      "loss": 5.9395,
      "mean_token_accuracy": 0.1597072184085846,
      "num_tokens": 40338511.0,
      "step": 18635
    },
    {
      "entropy": 6.454106712341309,
      "epoch": 1.9947027663331371,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004607684443293887,
      "loss": 5.9059,
      "mean_token_accuracy": 0.1590313047170639,
      "num_tokens": 40350395.0,
      "step": 18640
    },
    {
      "entropy": 6.487757444381714,
      "epoch": 1.995237840440901,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004607468703512993,
      "loss": 5.9094,
      "mean_token_accuracy": 0.16866012662649155,
      "num_tokens": 40360166.0,
      "step": 18645
    },
    {
      "entropy": 6.455652856826783,
      "epoch": 1.995772914548665,
      "grad_norm": 1.125,
      "learning_rate": 0.00046072529100979514,
      "loss": 5.9304,
      "mean_token_accuracy": 0.1574041113257408,
      "num_tokens": 40370655.0,
      "step": 18650
    },
    {
      "entropy": 6.476152133941651,
      "epoch": 1.996307988656429,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004607037063054993,
      "loss": 5.9008,
      "mean_token_accuracy": 0.16352151781320573,
      "num_tokens": 40381502.0,
      "step": 18655
    },
    {
      "entropy": 6.5687939643859865,
      "epoch": 1.9968430627641929,
      "grad_norm": 1.03125,
      "learning_rate": 0.000460682116239035,
      "loss": 5.9776,
      "mean_token_accuracy": 0.15785715132951736,
      "num_tokens": 40391815.0,
      "step": 18660
    },
    {
      "entropy": 6.460694694519043,
      "epoch": 1.9973781368719568,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00046066052081102576,
      "loss": 5.9443,
      "mean_token_accuracy": 0.16276349276304244,
      "num_tokens": 40403306.0,
      "step": 18665
    },
    {
      "entropy": 6.451063013076782,
      "epoch": 1.9979132109797209,
      "grad_norm": 0.953125,
      "learning_rate": 0.00046063892002209505,
      "loss": 5.9005,
      "mean_token_accuracy": 0.16664816737174987,
      "num_tokens": 40414210.0,
      "step": 18670
    },
    {
      "entropy": 6.356628561019898,
      "epoch": 1.9984482850874845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004606173138728666,
      "loss": 5.8531,
      "mean_token_accuracy": 0.17519440352916718,
      "num_tokens": 40424368.0,
      "step": 18675
    },
    {
      "entropy": 6.44556303024292,
      "epoch": 1.9989833591952486,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004605957023639643,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16035487353801728,
      "num_tokens": 40436252.0,
      "step": 18680
    },
    {
      "entropy": 6.485825729370117,
      "epoch": 1.9995184333030125,
      "grad_norm": 0.9296875,
      "learning_rate": 0.00046057408549601223,
      "loss": 5.9519,
      "mean_token_accuracy": 0.1635002925992012,
      "num_tokens": 40447256.0,
      "step": 18685
    },
    {
      "entropy": 6.5044247839185925,
      "epoch": 2.0,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004605524632696344,
      "loss": 5.8936,
      "mean_token_accuracy": 0.1655910031663047,
      "num_tokens": 40456516.0,
      "step": 18690
    },
    {
      "entropy": 6.510512638092041,
      "epoch": 2.000535074107764,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004605308356854553,
      "loss": 5.8385,
      "mean_token_accuracy": 0.1650438979268074,
      "num_tokens": 40467563.0,
      "step": 18695
    },
    {
      "entropy": 6.438768672943115,
      "epoch": 2.0010701482155278,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004605092027440994,
      "loss": 5.8778,
      "mean_token_accuracy": 0.1689416915178299,
      "num_tokens": 40478202.0,
      "step": 18700
    },
    {
      "entropy": 6.483799600601197,
      "epoch": 2.001605222323292,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004604875644461914,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16843170672655106,
      "num_tokens": 40489677.0,
      "step": 18705
    },
    {
      "entropy": 6.482054376602173,
      "epoch": 2.0021402964310555,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00046046592079235595,
      "loss": 5.8712,
      "mean_token_accuracy": 0.1641274005174637,
      "num_tokens": 40500694.0,
      "step": 18710
    },
    {
      "entropy": 6.47382607460022,
      "epoch": 2.0026753705388196,
      "grad_norm": 1.125,
      "learning_rate": 0.0004604442717832181,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16604586988687514,
      "num_tokens": 40511539.0,
      "step": 18715
    },
    {
      "entropy": 6.451759910583496,
      "epoch": 2.0032104446465837,
      "grad_norm": 1.25,
      "learning_rate": 0.000460422617419403,
      "loss": 5.8023,
      "mean_token_accuracy": 0.16651538759469986,
      "num_tokens": 40521906.0,
      "step": 18720
    },
    {
      "entropy": 6.523891019821167,
      "epoch": 2.0037455187543474,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046040095770153594,
      "loss": 5.8744,
      "mean_token_accuracy": 0.15731584131717682,
      "num_tokens": 40531454.0,
      "step": 18725
    },
    {
      "entropy": 6.534545755386352,
      "epoch": 2.0042805928621115,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004603792926302422,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15670345723628998,
      "num_tokens": 40541497.0,
      "step": 18730
    },
    {
      "entropy": 6.472694540023804,
      "epoch": 2.004815666969875,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004603576222061475,
      "loss": 5.9015,
      "mean_token_accuracy": 0.16415764540433883,
      "num_tokens": 40552576.0,
      "step": 18735
    },
    {
      "entropy": 6.469368362426758,
      "epoch": 2.005350741077639,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004603359464298774,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16486305594444275,
      "num_tokens": 40562906.0,
      "step": 18740
    },
    {
      "entropy": 6.465398836135864,
      "epoch": 2.0058858151854033,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046031426530205796,
      "loss": 5.768,
      "mean_token_accuracy": 0.17067015767097474,
      "num_tokens": 40574510.0,
      "step": 18745
    },
    {
      "entropy": 6.414217805862426,
      "epoch": 2.006420889293167,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004602925788233152,
      "loss": 5.8311,
      "mean_token_accuracy": 0.1761641427874565,
      "num_tokens": 40586170.0,
      "step": 18750
    },
    {
      "entropy": 6.475377368927002,
      "epoch": 2.006955963400931,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00046027088699427516,
      "loss": 5.8595,
      "mean_token_accuracy": 0.15970857441425323,
      "num_tokens": 40597613.0,
      "step": 18755
    },
    {
      "entropy": 6.515648746490479,
      "epoch": 2.007491037508695,
      "grad_norm": 1.03125,
      "learning_rate": 0.00046024918981556434,
      "loss": 5.9264,
      "mean_token_accuracy": 0.16082891076803207,
      "num_tokens": 40608366.0,
      "step": 18760
    },
    {
      "entropy": 6.447613191604614,
      "epoch": 2.008026111616459,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004602274872878092,
      "loss": 5.8448,
      "mean_token_accuracy": 0.17151678800582887,
      "num_tokens": 40618616.0,
      "step": 18765
    },
    {
      "entropy": 6.3876810550689695,
      "epoch": 2.008561185724223,
      "grad_norm": 0.96875,
      "learning_rate": 0.00046020577941163635,
      "loss": 5.707,
      "mean_token_accuracy": 0.17460259944200515,
      "num_tokens": 40628342.0,
      "step": 18770
    },
    {
      "entropy": 6.448616886138916,
      "epoch": 2.0090962598319866,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046018406618767273,
      "loss": 5.8269,
      "mean_token_accuracy": 0.16514158248901367,
      "num_tokens": 40638644.0,
      "step": 18775
    },
    {
      "entropy": 6.440449190139771,
      "epoch": 2.0096313339397507,
      "grad_norm": 1.0,
      "learning_rate": 0.0004601623476165451,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16634675115346909,
      "num_tokens": 40649945.0,
      "step": 18780
    },
    {
      "entropy": 6.561657094955445,
      "epoch": 2.010166408047515,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00046014062369888084,
      "loss": 5.9544,
      "mean_token_accuracy": 0.1575586587190628,
      "num_tokens": 40660896.0,
      "step": 18785
    },
    {
      "entropy": 6.52346134185791,
      "epoch": 2.0107014821552784,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000460118894435307,
      "loss": 5.912,
      "mean_token_accuracy": 0.16213926076889038,
      "num_tokens": 40672072.0,
      "step": 18790
    },
    {
      "entropy": 6.508182096481323,
      "epoch": 2.0112365562630425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004600971598264511,
      "loss": 5.9166,
      "mean_token_accuracy": 0.1624039426445961,
      "num_tokens": 40682622.0,
      "step": 18795
    },
    {
      "entropy": 6.405517148971557,
      "epoch": 2.011771630370806,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004600754198729407,
      "loss": 5.8304,
      "mean_token_accuracy": 0.1659341886639595,
      "num_tokens": 40693127.0,
      "step": 18800
    },
    {
      "entropy": 6.457127475738526,
      "epoch": 2.0123067044785703,
      "grad_norm": 0.90625,
      "learning_rate": 0.00046005367457540366,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16428034454584123,
      "num_tokens": 40705009.0,
      "step": 18805
    },
    {
      "entropy": 6.504219722747803,
      "epoch": 2.0128417785863344,
      "grad_norm": 1.0,
      "learning_rate": 0.00046003192393446763,
      "loss": 5.8902,
      "mean_token_accuracy": 0.1615692675113678,
      "num_tokens": 40716498.0,
      "step": 18810
    },
    {
      "entropy": 6.474493598937988,
      "epoch": 2.013376852694098,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004600101679507609,
      "loss": 5.8681,
      "mean_token_accuracy": 0.16769025027751921,
      "num_tokens": 40727802.0,
      "step": 18815
    },
    {
      "entropy": 6.527164173126221,
      "epoch": 2.013911926801862,
      "grad_norm": 1.0,
      "learning_rate": 0.0004599884066249115,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16638778150081635,
      "num_tokens": 40738866.0,
      "step": 18820
    },
    {
      "entropy": 6.447434520721435,
      "epoch": 2.014447000909626,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004599666399575479,
      "loss": 5.8562,
      "mean_token_accuracy": 0.16192474961280823,
      "num_tokens": 40750214.0,
      "step": 18825
    },
    {
      "entropy": 6.49119234085083,
      "epoch": 2.01498207501739,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004599448679492986,
      "loss": 5.8559,
      "mean_token_accuracy": 0.16967889666557312,
      "num_tokens": 40759830.0,
      "step": 18830
    },
    {
      "entropy": 6.520104932785034,
      "epoch": 2.015517149125154,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004599230906007921,
      "loss": 5.9049,
      "mean_token_accuracy": 0.16171150356531144,
      "num_tokens": 40770007.0,
      "step": 18835
    },
    {
      "entropy": 6.499899244308471,
      "epoch": 2.0160522232329177,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004599013079126575,
      "loss": 5.8099,
      "mean_token_accuracy": 0.1611242488026619,
      "num_tokens": 40781495.0,
      "step": 18840
    },
    {
      "entropy": 6.4952293872833256,
      "epoch": 2.0165872973406818,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004598795198855235,
      "loss": 5.8169,
      "mean_token_accuracy": 0.16788874715566635,
      "num_tokens": 40792875.0,
      "step": 18845
    },
    {
      "entropy": 6.482176446914673,
      "epoch": 2.0171223714484454,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004598577265200193,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1550595983862877,
      "num_tokens": 40804484.0,
      "step": 18850
    },
    {
      "entropy": 6.464883661270141,
      "epoch": 2.0176574455562095,
      "grad_norm": 1.0,
      "learning_rate": 0.0004598359278167744,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16639561355113983,
      "num_tokens": 40814692.0,
      "step": 18855
    },
    {
      "entropy": 6.478203535079956,
      "epoch": 2.0181925196639736,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045981412377641795,
      "loss": 5.8045,
      "mean_token_accuracy": 0.17130476981401443,
      "num_tokens": 40825443.0,
      "step": 18860
    },
    {
      "entropy": 6.499489068984985,
      "epoch": 2.0187275937717373,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045979231439957966,
      "loss": 5.874,
      "mean_token_accuracy": 0.16265991777181626,
      "num_tokens": 40836699.0,
      "step": 18865
    },
    {
      "entropy": 6.484277057647705,
      "epoch": 2.0192626678795014,
      "grad_norm": 1.125,
      "learning_rate": 0.0004597704996868893,
      "loss": 5.8492,
      "mean_token_accuracy": 0.16351415514945983,
      "num_tokens": 40846736.0,
      "step": 18870
    },
    {
      "entropy": 6.505946207046509,
      "epoch": 2.019797741987265,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004597486796389766,
      "loss": 5.9317,
      "mean_token_accuracy": 0.16407058387994766,
      "num_tokens": 40858362.0,
      "step": 18875
    },
    {
      "entropy": 6.471547222137451,
      "epoch": 2.020332816095029,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045972685425647186,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16968994885683059,
      "num_tokens": 40869346.0,
      "step": 18880
    },
    {
      "entropy": 6.539822721481324,
      "epoch": 2.0208678902027932,
      "grad_norm": 1.125,
      "learning_rate": 0.0004597050235400051,
      "loss": 5.9408,
      "mean_token_accuracy": 0.16143878996372224,
      "num_tokens": 40880003.0,
      "step": 18885
    },
    {
      "entropy": 6.522287750244141,
      "epoch": 2.021402964310557,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045968318749020675,
      "loss": 5.8777,
      "mean_token_accuracy": 0.1632978469133377,
      "num_tokens": 40891515.0,
      "step": 18890
    },
    {
      "entropy": 6.477145481109619,
      "epoch": 2.021938038418321,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004596613461077073,
      "loss": 5.833,
      "mean_token_accuracy": 0.16654694825410843,
      "num_tokens": 40902024.0,
      "step": 18895
    },
    {
      "entropy": 6.463969755172729,
      "epoch": 2.022473112526085,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045963949939313754,
      "loss": 5.8318,
      "mean_token_accuracy": 0.16850030720233916,
      "num_tokens": 40912160.0,
      "step": 18900
    },
    {
      "entropy": 6.496173238754272,
      "epoch": 2.0230081866338487,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000459617647347128,
      "loss": 5.9251,
      "mean_token_accuracy": 0.15282502695918082,
      "num_tokens": 40922516.0,
      "step": 18905
    },
    {
      "entropy": 6.470272207260132,
      "epoch": 2.023543260741613,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045959578997031,
      "loss": 5.8896,
      "mean_token_accuracy": 0.16029372811317444,
      "num_tokens": 40934148.0,
      "step": 18910
    },
    {
      "entropy": 6.52022066116333,
      "epoch": 2.0240783348493765,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045957392726331443,
      "loss": 5.8133,
      "mean_token_accuracy": 0.17105006724596022,
      "num_tokens": 40944409.0,
      "step": 18915
    },
    {
      "entropy": 6.472586727142334,
      "epoch": 2.0246134089571406,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045955205922677264,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1668213739991188,
      "num_tokens": 40955133.0,
      "step": 18920
    },
    {
      "entropy": 6.45203800201416,
      "epoch": 2.0251484830649047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004595301858613161,
      "loss": 5.9442,
      "mean_token_accuracy": 0.16431741863489152,
      "num_tokens": 40965508.0,
      "step": 18925
    },
    {
      "entropy": 6.573894882202149,
      "epoch": 2.0256835571726683,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004595083071675763,
      "loss": 5.932,
      "mean_token_accuracy": 0.16677047312259674,
      "num_tokens": 40975931.0,
      "step": 18930
    },
    {
      "entropy": 6.525880002975464,
      "epoch": 2.0262186312804324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045948642314618514,
      "loss": 5.8781,
      "mean_token_accuracy": 0.16080835908651353,
      "num_tokens": 40985797.0,
      "step": 18935
    },
    {
      "entropy": 6.4700089454650875,
      "epoch": 2.026753705388196,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004594645337977744,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16267758309841157,
      "num_tokens": 40996501.0,
      "step": 18940
    },
    {
      "entropy": 6.482262372970581,
      "epoch": 2.02728877949596,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004594426391229762,
      "loss": 5.8534,
      "mean_token_accuracy": 0.1654360696673393,
      "num_tokens": 41005970.0,
      "step": 18945
    },
    {
      "entropy": 6.45759596824646,
      "epoch": 2.0278238536037243,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004594207391224227,
      "loss": 5.8438,
      "mean_token_accuracy": 0.1636953517794609,
      "num_tokens": 41017208.0,
      "step": 18950
    },
    {
      "entropy": 6.557088375091553,
      "epoch": 2.028358927711488,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00045939883379674624,
      "loss": 5.9959,
      "mean_token_accuracy": 0.15847332701086997,
      "num_tokens": 41028435.0,
      "step": 18955
    },
    {
      "entropy": 6.4631500244140625,
      "epoch": 2.028894001819252,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045937692314657936,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1696695178747177,
      "num_tokens": 41039256.0,
      "step": 18960
    },
    {
      "entropy": 6.439812231063843,
      "epoch": 2.0294290759270157,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004593550071725548,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16237109303474426,
      "num_tokens": 41050527.0,
      "step": 18965
    },
    {
      "entropy": 6.491014862060547,
      "epoch": 2.02996415003478,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045933308587530527,
      "loss": 5.8941,
      "mean_token_accuracy": 0.16085420548915863,
      "num_tokens": 41062259.0,
      "step": 18970
    },
    {
      "entropy": 6.485957050323487,
      "epoch": 2.030499224142544,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004593111592554638,
      "loss": 5.918,
      "mean_token_accuracy": 0.16451694071292877,
      "num_tokens": 41073011.0,
      "step": 18975
    },
    {
      "entropy": 6.455955982208252,
      "epoch": 2.0310342982503076,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004592892273136635,
      "loss": 5.7651,
      "mean_token_accuracy": 0.17509791553020476,
      "num_tokens": 41083482.0,
      "step": 18980
    },
    {
      "entropy": 6.453741264343262,
      "epoch": 2.0315693723580717,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045926729005053764,
      "loss": 5.8359,
      "mean_token_accuracy": 0.1706819698214531,
      "num_tokens": 41094028.0,
      "step": 18985
    },
    {
      "entropy": 6.432194995880127,
      "epoch": 2.0321044464658353,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004592453474667197,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16453876197338105,
      "num_tokens": 41106346.0,
      "step": 18990
    },
    {
      "entropy": 6.518768453598023,
      "epoch": 2.0326395205735994,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004592233995628432,
      "loss": 5.791,
      "mean_token_accuracy": 0.16712938249111176,
      "num_tokens": 41116130.0,
      "step": 18995
    },
    {
      "entropy": 6.490428686141968,
      "epoch": 2.0331745946813635,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000459201446339542,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16928952634334565,
      "num_tokens": 41127103.0,
      "step": 19000
    },
    {
      "entropy": 6.411393499374389,
      "epoch": 2.033709668789127,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004591794877974499,
      "loss": 5.8277,
      "mean_token_accuracy": 0.1634768784046173,
      "num_tokens": 41138258.0,
      "step": 19005
    },
    {
      "entropy": 6.486131811141968,
      "epoch": 2.0342447428968913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045915752393720094,
      "loss": 5.885,
      "mean_token_accuracy": 0.16703638136386872,
      "num_tokens": 41148859.0,
      "step": 19010
    },
    {
      "entropy": 6.4885648727417,
      "epoch": 2.034779817004655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045913555475942937,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16324065178632735,
      "num_tokens": 41158614.0,
      "step": 19015
    },
    {
      "entropy": 6.411256885528564,
      "epoch": 2.035314891112419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004591135802647695,
      "loss": 5.8371,
      "mean_token_accuracy": 0.1752866581082344,
      "num_tokens": 41170215.0,
      "step": 19020
    },
    {
      "entropy": 6.458434438705444,
      "epoch": 2.035849965220183,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004590916004538559,
      "loss": 5.8452,
      "mean_token_accuracy": 0.16274577677249907,
      "num_tokens": 41180784.0,
      "step": 19025
    },
    {
      "entropy": 6.520003843307495,
      "epoch": 2.036385039327947,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045906961532732316,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16188473105430604,
      "num_tokens": 41191305.0,
      "step": 19030
    },
    {
      "entropy": 6.489679384231567,
      "epoch": 2.036920113435711,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004590476248858062,
      "loss": 5.8901,
      "mean_token_accuracy": 0.16224412024021148,
      "num_tokens": 41202388.0,
      "step": 19035
    },
    {
      "entropy": 6.4304992198944095,
      "epoch": 2.037455187543475,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004590256291299399,
      "loss": 5.8443,
      "mean_token_accuracy": 0.16643714159727097,
      "num_tokens": 41212136.0,
      "step": 19040
    },
    {
      "entropy": 6.440283346176147,
      "epoch": 2.0379902616512386,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045900362806035945,
      "loss": 5.8722,
      "mean_token_accuracy": 0.16702118068933486,
      "num_tokens": 41223390.0,
      "step": 19045
    },
    {
      "entropy": 6.481003093719482,
      "epoch": 2.0385253357590027,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045898162167770006,
      "loss": 5.9053,
      "mean_token_accuracy": 0.167675943672657,
      "num_tokens": 41234775.0,
      "step": 19050
    },
    {
      "entropy": 6.491551542282105,
      "epoch": 2.0390604098667664,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004589596099825973,
      "loss": 5.8183,
      "mean_token_accuracy": 0.16556926518678666,
      "num_tokens": 41245627.0,
      "step": 19055
    },
    {
      "entropy": 6.432875871658325,
      "epoch": 2.0395954839745305,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045893759297568654,
      "loss": 5.8882,
      "mean_token_accuracy": 0.16342023611068726,
      "num_tokens": 41255441.0,
      "step": 19060
    },
    {
      "entropy": 6.409200620651245,
      "epoch": 2.0401305580822946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004589155706576036,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16868349611759187,
      "num_tokens": 41268686.0,
      "step": 19065
    },
    {
      "entropy": 6.566879367828369,
      "epoch": 2.0406656321900583,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045889354302898445,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15423219352960588,
      "num_tokens": 41279384.0,
      "step": 19070
    },
    {
      "entropy": 6.571926021575928,
      "epoch": 2.0412007062978224,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004588715100904651,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15895104855298997,
      "num_tokens": 41289923.0,
      "step": 19075
    },
    {
      "entropy": 6.454684400558472,
      "epoch": 2.041735780405586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004588494718426818,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16840842366218567,
      "num_tokens": 41300486.0,
      "step": 19080
    },
    {
      "entropy": 6.4193014144897464,
      "epoch": 2.04227085451335,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004588274282862708,
      "loss": 5.8488,
      "mean_token_accuracy": 0.1669850692152977,
      "num_tokens": 41310507.0,
      "step": 19085
    },
    {
      "entropy": 6.501082515716552,
      "epoch": 2.042805928621114,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045880537942186855,
      "loss": 5.8562,
      "mean_token_accuracy": 0.17119995802640914,
      "num_tokens": 41321284.0,
      "step": 19090
    },
    {
      "entropy": 6.45316333770752,
      "epoch": 2.043341002728878,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004587833252501118,
      "loss": 5.8057,
      "mean_token_accuracy": 0.17230299562215806,
      "num_tokens": 41332037.0,
      "step": 19095
    },
    {
      "entropy": 6.378763151168823,
      "epoch": 2.043876076836642,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004587612657716374,
      "loss": 5.8502,
      "mean_token_accuracy": 0.1675565093755722,
      "num_tokens": 41343480.0,
      "step": 19100
    },
    {
      "entropy": 6.53266716003418,
      "epoch": 2.0444111509444056,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004587392009870824,
      "loss": 5.9066,
      "mean_token_accuracy": 0.15852901488542556,
      "num_tokens": 41353656.0,
      "step": 19105
    },
    {
      "entropy": 6.4966706275939945,
      "epoch": 2.0449462250521697,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004587171308970836,
      "loss": 5.8582,
      "mean_token_accuracy": 0.1642326146364212,
      "num_tokens": 41363640.0,
      "step": 19110
    },
    {
      "entropy": 6.448680019378662,
      "epoch": 2.045481299159934,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045869505550227857,
      "loss": 5.8897,
      "mean_token_accuracy": 0.15792203247547149,
      "num_tokens": 41374194.0,
      "step": 19115
    },
    {
      "entropy": 6.460755014419556,
      "epoch": 2.0460163732676975,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004586729748033046,
      "loss": 5.8577,
      "mean_token_accuracy": 0.17022178769111634,
      "num_tokens": 41384968.0,
      "step": 19120
    },
    {
      "entropy": 6.47335410118103,
      "epoch": 2.0465514473754616,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045865088880079924,
      "loss": 5.7499,
      "mean_token_accuracy": 0.18075418174266816,
      "num_tokens": 41395117.0,
      "step": 19125
    },
    {
      "entropy": 6.5149565696716305,
      "epoch": 2.0470865214832252,
      "grad_norm": 1.0,
      "learning_rate": 0.0004586287974954004,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16422117054462432,
      "num_tokens": 41406366.0,
      "step": 19130
    },
    {
      "entropy": 6.515776109695435,
      "epoch": 2.0476215955909893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004586067008877457,
      "loss": 5.8598,
      "mean_token_accuracy": 0.16343075186014175,
      "num_tokens": 41417388.0,
      "step": 19135
    },
    {
      "entropy": 6.44080491065979,
      "epoch": 2.0481566696987534,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004585845989784734,
      "loss": 5.8249,
      "mean_token_accuracy": 0.16619649827480315,
      "num_tokens": 41428085.0,
      "step": 19140
    },
    {
      "entropy": 6.516266679763794,
      "epoch": 2.048691743806517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045856249176822155,
      "loss": 5.9135,
      "mean_token_accuracy": 0.1633109986782074,
      "num_tokens": 41439117.0,
      "step": 19145
    },
    {
      "entropy": 6.472226476669311,
      "epoch": 2.049226817914281,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045854037925762857,
      "loss": 5.8996,
      "mean_token_accuracy": 0.15708108842372895,
      "num_tokens": 41450042.0,
      "step": 19150
    },
    {
      "entropy": 6.428773069381714,
      "epoch": 2.049761892022045,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045851826144733296,
      "loss": 5.8003,
      "mean_token_accuracy": 0.1717658668756485,
      "num_tokens": 41460554.0,
      "step": 19155
    },
    {
      "entropy": 6.514788389205933,
      "epoch": 2.050296966129809,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045849613833797326,
      "loss": 5.8915,
      "mean_token_accuracy": 0.16608842611312866,
      "num_tokens": 41471403.0,
      "step": 19160
    },
    {
      "entropy": 6.501409244537354,
      "epoch": 2.050832040237573,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045847400993018835,
      "loss": 5.882,
      "mean_token_accuracy": 0.16245611160993575,
      "num_tokens": 41482098.0,
      "step": 19165
    },
    {
      "entropy": 6.448575496673584,
      "epoch": 2.0513671143453367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045845187622461724,
      "loss": 5.8455,
      "mean_token_accuracy": 0.16848851889371871,
      "num_tokens": 41493126.0,
      "step": 19170
    },
    {
      "entropy": 6.524034881591797,
      "epoch": 2.051902188453101,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000458429737221899,
      "loss": 5.9673,
      "mean_token_accuracy": 0.15790134072303771,
      "num_tokens": 41504242.0,
      "step": 19175
    },
    {
      "entropy": 6.477486896514892,
      "epoch": 2.052437262560865,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004584075929226728,
      "loss": 5.843,
      "mean_token_accuracy": 0.16866402328014374,
      "num_tokens": 41515368.0,
      "step": 19180
    },
    {
      "entropy": 6.510952377319336,
      "epoch": 2.0529723366686286,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004583854433275782,
      "loss": 5.8912,
      "mean_token_accuracy": 0.1591535195708275,
      "num_tokens": 41525939.0,
      "step": 19185
    },
    {
      "entropy": 6.467635726928711,
      "epoch": 2.0535074107763926,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045836328843725465,
      "loss": 5.9092,
      "mean_token_accuracy": 0.16421155631542206,
      "num_tokens": 41536505.0,
      "step": 19190
    },
    {
      "entropy": 6.461625385284424,
      "epoch": 2.0540424848841563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045834112825234193,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16639016270637513,
      "num_tokens": 41546293.0,
      "step": 19195
    },
    {
      "entropy": 6.503767967224121,
      "epoch": 2.0545775589919204,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045831896277347985,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16666459739208223,
      "num_tokens": 41556596.0,
      "step": 19200
    },
    {
      "entropy": 6.451248025894165,
      "epoch": 2.0551126330996845,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004582967920013085,
      "loss": 5.8794,
      "mean_token_accuracy": 0.17004042565822602,
      "num_tokens": 41566732.0,
      "step": 19205
    },
    {
      "entropy": 6.492414379119873,
      "epoch": 2.055647707207448,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045827461593646806,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16111411303281784,
      "num_tokens": 41577777.0,
      "step": 19210
    },
    {
      "entropy": 6.477559518814087,
      "epoch": 2.0561827813152123,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004582524345795989,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16588083952665328,
      "num_tokens": 41587978.0,
      "step": 19215
    },
    {
      "entropy": 6.475324440002441,
      "epoch": 2.056717855422976,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004582302479313413,
      "loss": 5.8123,
      "mean_token_accuracy": 0.1686853811144829,
      "num_tokens": 41598977.0,
      "step": 19220
    },
    {
      "entropy": 6.489716815948486,
      "epoch": 2.05725292953074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004582080559923362,
      "loss": 5.8927,
      "mean_token_accuracy": 0.1602018803358078,
      "num_tokens": 41609664.0,
      "step": 19225
    },
    {
      "entropy": 6.507627534866333,
      "epoch": 2.057788003638504,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004581858587632241,
      "loss": 5.8715,
      "mean_token_accuracy": 0.1640459969639778,
      "num_tokens": 41619427.0,
      "step": 19230
    },
    {
      "entropy": 6.4372419834136965,
      "epoch": 2.0583230777462678,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581636562446461,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16239285916090013,
      "num_tokens": 41629551.0,
      "step": 19235
    },
    {
      "entropy": 6.406296253204346,
      "epoch": 2.058858151854032,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581414484372433,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16379759907722474,
      "num_tokens": 41640183.0,
      "step": 19240
    },
    {
      "entropy": 6.465412855148315,
      "epoch": 2.0593932259617955,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004581192353416569,
      "loss": 5.847,
      "mean_token_accuracy": 0.1683710992336273,
      "num_tokens": 41651934.0,
      "step": 19245
    },
    {
      "entropy": 6.412327861785888,
      "epoch": 2.0599283000695596,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004580970169585283,
      "loss": 5.7397,
      "mean_token_accuracy": 0.18560385555028916,
      "num_tokens": 41662418.0,
      "step": 19250
    },
    {
      "entropy": 6.386852598190307,
      "epoch": 2.0604633741773237,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045807479328849906,
      "loss": 5.7741,
      "mean_token_accuracy": 0.17042402625083924,
      "num_tokens": 41673419.0,
      "step": 19255
    },
    {
      "entropy": 6.394661617279053,
      "epoch": 2.0609984482850874,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045805256433221087,
      "loss": 5.8172,
      "mean_token_accuracy": 0.17177309542894365,
      "num_tokens": 41684132.0,
      "step": 19260
    },
    {
      "entropy": 6.507149934768677,
      "epoch": 2.0615335223928515,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004580303300903056,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1648765832185745,
      "num_tokens": 41695051.0,
      "step": 19265
    },
    {
      "entropy": 6.488150978088379,
      "epoch": 2.062068596500615,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045800809056342534,
      "loss": 5.892,
      "mean_token_accuracy": 0.1650779739022255,
      "num_tokens": 41705986.0,
      "step": 19270
    },
    {
      "entropy": 6.465203952789307,
      "epoch": 2.0626036706083792,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045798584575221213,
      "loss": 5.8984,
      "mean_token_accuracy": 0.1633210226893425,
      "num_tokens": 41717082.0,
      "step": 19275
    },
    {
      "entropy": 6.513140630722046,
      "epoch": 2.0631387447161433,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004579635956573084,
      "loss": 5.9077,
      "mean_token_accuracy": 0.1673460751771927,
      "num_tokens": 41728783.0,
      "step": 19280
    },
    {
      "entropy": 6.4859260559082035,
      "epoch": 2.063673818823907,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045794134027935646,
      "loss": 5.8784,
      "mean_token_accuracy": 0.1652866929769516,
      "num_tokens": 41740452.0,
      "step": 19285
    },
    {
      "entropy": 6.477737998962402,
      "epoch": 2.064208892931671,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004579190796189991,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16969979852437972,
      "num_tokens": 41750966.0,
      "step": 19290
    },
    {
      "entropy": 6.516948223114014,
      "epoch": 2.064743967039435,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000457896813676879,
      "loss": 5.8082,
      "mean_token_accuracy": 0.1685435265302658,
      "num_tokens": 41762174.0,
      "step": 19295
    },
    {
      "entropy": 6.474477863311767,
      "epoch": 2.065279041147199,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045787454245363907,
      "loss": 5.9203,
      "mean_token_accuracy": 0.15837727040052413,
      "num_tokens": 41773396.0,
      "step": 19300
    },
    {
      "entropy": 6.476316642761231,
      "epoch": 2.065814115254963,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004578522659499225,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16626913398504256,
      "num_tokens": 41783833.0,
      "step": 19305
    },
    {
      "entropy": 6.450223922729492,
      "epoch": 2.0663491893627266,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045782998416637235,
      "loss": 5.8393,
      "mean_token_accuracy": 0.17160961031913757,
      "num_tokens": 41795363.0,
      "step": 19310
    },
    {
      "entropy": 6.501335573196411,
      "epoch": 2.0668842634704907,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004578076971036321,
      "loss": 5.8943,
      "mean_token_accuracy": 0.1667793795466423,
      "num_tokens": 41807575.0,
      "step": 19315
    },
    {
      "entropy": 6.420790672302246,
      "epoch": 2.067419337578255,
      "grad_norm": 1.125,
      "learning_rate": 0.0004577854047623454,
      "loss": 5.8144,
      "mean_token_accuracy": 0.1656635269522667,
      "num_tokens": 41818496.0,
      "step": 19320
    },
    {
      "entropy": 6.530952310562133,
      "epoch": 2.0679544116860185,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004577631071431557,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1666251987218857,
      "num_tokens": 41829106.0,
      "step": 19325
    },
    {
      "entropy": 6.4277403354644775,
      "epoch": 2.0684894857937826,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045774080424670704,
      "loss": 5.8421,
      "mean_token_accuracy": 0.16808194667100906,
      "num_tokens": 41838095.0,
      "step": 19330
    },
    {
      "entropy": 6.395112705230713,
      "epoch": 2.069024559901546,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004577184960736433,
      "loss": 5.8545,
      "mean_token_accuracy": 0.1562212109565735,
      "num_tokens": 41847980.0,
      "step": 19335
    },
    {
      "entropy": 6.507107353210449,
      "epoch": 2.0695596340093103,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045769618262460864,
      "loss": 5.8668,
      "mean_token_accuracy": 0.16605396419763566,
      "num_tokens": 41858625.0,
      "step": 19340
    },
    {
      "entropy": 6.504085397720337,
      "epoch": 2.0700947081170744,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004576738639002475,
      "loss": 5.8853,
      "mean_token_accuracy": 0.15824336037039757,
      "num_tokens": 41870939.0,
      "step": 19345
    },
    {
      "entropy": 6.528838586807251,
      "epoch": 2.070629782224838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004576515399012041,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1626370444893837,
      "num_tokens": 41882356.0,
      "step": 19350
    },
    {
      "entropy": 6.5101395606994625,
      "epoch": 2.071164856332602,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004576292106281231,
      "loss": 5.8778,
      "mean_token_accuracy": 0.1643811881542206,
      "num_tokens": 41892306.0,
      "step": 19355
    },
    {
      "entropy": 6.505565118789673,
      "epoch": 2.071699930440366,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045760687608164933,
      "loss": 5.9395,
      "mean_token_accuracy": 0.16350977420806884,
      "num_tokens": 41902450.0,
      "step": 19360
    },
    {
      "entropy": 6.453457069396973,
      "epoch": 2.07223500454813,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004575845362624278,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1679016500711441,
      "num_tokens": 41913319.0,
      "step": 19365
    },
    {
      "entropy": 6.488507032394409,
      "epoch": 2.072770078655894,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045756219117110337,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1611151561141014,
      "num_tokens": 41923598.0,
      "step": 19370
    },
    {
      "entropy": 6.513248634338379,
      "epoch": 2.0733051527636577,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045753984080832127,
      "loss": 5.8507,
      "mean_token_accuracy": 0.16608191132545472,
      "num_tokens": 41934482.0,
      "step": 19375
    },
    {
      "entropy": 6.492088985443115,
      "epoch": 2.0738402268714218,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000457517485174727,
      "loss": 5.8574,
      "mean_token_accuracy": 0.1653871491551399,
      "num_tokens": 41945994.0,
      "step": 19380
    },
    {
      "entropy": 6.4923851013183596,
      "epoch": 2.0743753009791854,
      "grad_norm": 1.0,
      "learning_rate": 0.00045749512427096595,
      "loss": 5.8704,
      "mean_token_accuracy": 0.16107353270053865,
      "num_tokens": 41956564.0,
      "step": 19385
    },
    {
      "entropy": 6.461237478256225,
      "epoch": 2.0749103750869495,
      "grad_norm": 1.125,
      "learning_rate": 0.00045747275809768386,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16529469639062883,
      "num_tokens": 41966847.0,
      "step": 19390
    },
    {
      "entropy": 6.506811189651489,
      "epoch": 2.0754454491947136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004574503866555265,
      "loss": 5.9408,
      "mean_token_accuracy": 0.15201306641101836,
      "num_tokens": 41977752.0,
      "step": 19395
    },
    {
      "entropy": 6.525647926330566,
      "epoch": 2.0759805233024773,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045742800994513983,
      "loss": 5.8687,
      "mean_token_accuracy": 0.16600601077079774,
      "num_tokens": 41987837.0,
      "step": 19400
    },
    {
      "entropy": 6.533274364471436,
      "epoch": 2.0765155974102414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045740562796717005,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16282256543636323,
      "num_tokens": 41997988.0,
      "step": 19405
    },
    {
      "entropy": 6.3938316822052,
      "epoch": 2.077050671518005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004573832407222634,
      "loss": 5.7823,
      "mean_token_accuracy": 0.16964642852544784,
      "num_tokens": 42009643.0,
      "step": 19410
    },
    {
      "entropy": 6.462463140487671,
      "epoch": 2.077585745625769,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004573608482110663,
      "loss": 5.8497,
      "mean_token_accuracy": 0.16264649033546447,
      "num_tokens": 42019972.0,
      "step": 19415
    },
    {
      "entropy": 6.464416360855102,
      "epoch": 2.0781208197335332,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00045733845043422535,
      "loss": 5.832,
      "mean_token_accuracy": 0.17043236047029495,
      "num_tokens": 42030536.0,
      "step": 19420
    },
    {
      "entropy": 6.4750933170318605,
      "epoch": 2.078655893841297,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004573160473923872,
      "loss": 5.8539,
      "mean_token_accuracy": 0.16336055845022202,
      "num_tokens": 42041217.0,
      "step": 19425
    },
    {
      "entropy": 6.347858810424805,
      "epoch": 2.079190967949061,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004572936390861988,
      "loss": 5.7542,
      "mean_token_accuracy": 0.16863729804754257,
      "num_tokens": 42051346.0,
      "step": 19430
    },
    {
      "entropy": 6.5534192562103275,
      "epoch": 2.0797260420568247,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004572712255163072,
      "loss": 5.9216,
      "mean_token_accuracy": 0.16346222311258315,
      "num_tokens": 42062718.0,
      "step": 19435
    },
    {
      "entropy": 6.52284951210022,
      "epoch": 2.0802611161645888,
      "grad_norm": 1.0,
      "learning_rate": 0.0004572488066833596,
      "loss": 5.8551,
      "mean_token_accuracy": 0.16144430339336396,
      "num_tokens": 42073587.0,
      "step": 19440
    },
    {
      "entropy": 6.482475090026855,
      "epoch": 2.080796190272353,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045722638258800335,
      "loss": 5.8741,
      "mean_token_accuracy": 0.1722232446074486,
      "num_tokens": 42083763.0,
      "step": 19445
    },
    {
      "entropy": 6.411497688293457,
      "epoch": 2.0813312643801165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004572039532308858,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16639414429664612,
      "num_tokens": 42094062.0,
      "step": 19450
    },
    {
      "entropy": 6.471854639053345,
      "epoch": 2.0818663384878806,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004571815186126549,
      "loss": 5.8722,
      "mean_token_accuracy": 0.163537135720253,
      "num_tokens": 42106136.0,
      "step": 19455
    },
    {
      "entropy": 6.462906217575073,
      "epoch": 2.0824014125956447,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045715907873395805,
      "loss": 5.8817,
      "mean_token_accuracy": 0.15836379528045655,
      "num_tokens": 42116309.0,
      "step": 19460
    },
    {
      "entropy": 6.441565942764282,
      "epoch": 2.0829364867034084,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004571366335954435,
      "loss": 5.8684,
      "mean_token_accuracy": 0.1681746080517769,
      "num_tokens": 42127747.0,
      "step": 19465
    },
    {
      "entropy": 6.386489391326904,
      "epoch": 2.0834715608111725,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004571141831977592,
      "loss": 5.7987,
      "mean_token_accuracy": 0.17530223429203035,
      "num_tokens": 42137992.0,
      "step": 19470
    },
    {
      "entropy": 6.4585182666778564,
      "epoch": 2.084006634918936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004570917275415535,
      "loss": 5.8079,
      "mean_token_accuracy": 0.17035417705774308,
      "num_tokens": 42148833.0,
      "step": 19475
    },
    {
      "entropy": 6.4442754745483395,
      "epoch": 2.0845417090267,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004570692666274747,
      "loss": 5.876,
      "mean_token_accuracy": 0.16558097302913666,
      "num_tokens": 42159276.0,
      "step": 19480
    },
    {
      "entropy": 6.511191987991333,
      "epoch": 2.0850767831344643,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004570468004561715,
      "loss": 5.9325,
      "mean_token_accuracy": 0.16486080437898637,
      "num_tokens": 42171541.0,
      "step": 19485
    },
    {
      "entropy": 6.470981788635254,
      "epoch": 2.085611857242228,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045702432902829245,
      "loss": 5.8805,
      "mean_token_accuracy": 0.1694780945777893,
      "num_tokens": 42182888.0,
      "step": 19490
    },
    {
      "entropy": 6.462714290618896,
      "epoch": 2.086146931349992,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004570018523444865,
      "loss": 5.8161,
      "mean_token_accuracy": 0.17538104504346846,
      "num_tokens": 42193405.0,
      "step": 19495
    },
    {
      "entropy": 6.478708982467651,
      "epoch": 2.0866820054577557,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045697937040540263,
      "loss": 5.8345,
      "mean_token_accuracy": 0.16551922708749772,
      "num_tokens": 42203355.0,
      "step": 19500
    },
    {
      "entropy": 6.436646604537964,
      "epoch": 2.08721707956552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045695688321169,
      "loss": 5.8709,
      "mean_token_accuracy": 0.15502800792455673,
      "num_tokens": 42213191.0,
      "step": 19505
    },
    {
      "entropy": 6.495199537277221,
      "epoch": 2.087752153673284,
      "grad_norm": 1.125,
      "learning_rate": 0.00045693439076399797,
      "loss": 5.9343,
      "mean_token_accuracy": 0.16047230660915374,
      "num_tokens": 42223357.0,
      "step": 19510
    },
    {
      "entropy": 6.514384078979492,
      "epoch": 2.0882872277810476,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045691189306297595,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1668185830116272,
      "num_tokens": 42234917.0,
      "step": 19515
    },
    {
      "entropy": 6.547740936279297,
      "epoch": 2.0888223018888117,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045688939010927353,
      "loss": 5.8821,
      "mean_token_accuracy": 0.16421758979558945,
      "num_tokens": 42245977.0,
      "step": 19520
    },
    {
      "entropy": 6.451862430572509,
      "epoch": 2.0893573759965753,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004568668819035406,
      "loss": 5.8523,
      "mean_token_accuracy": 0.1601534068584442,
      "num_tokens": 42257523.0,
      "step": 19525
    },
    {
      "entropy": 6.46871247291565,
      "epoch": 2.0898924501043394,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045684436844642694,
      "loss": 5.8627,
      "mean_token_accuracy": 0.1697766199707985,
      "num_tokens": 42267727.0,
      "step": 19530
    },
    {
      "entropy": 6.488473176956177,
      "epoch": 2.0904275242121035,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004568218497385828,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16607387214899064,
      "num_tokens": 42278201.0,
      "step": 19535
    },
    {
      "entropy": 6.5041172981262205,
      "epoch": 2.090962598319867,
      "grad_norm": 1.0,
      "learning_rate": 0.0004567993257806582,
      "loss": 5.9553,
      "mean_token_accuracy": 0.15660367012023926,
      "num_tokens": 42288695.0,
      "step": 19540
    },
    {
      "entropy": 6.462008380889893,
      "epoch": 2.0914976724276313,
      "grad_norm": 1.125,
      "learning_rate": 0.0004567767965733036,
      "loss": 5.9215,
      "mean_token_accuracy": 0.1680365726351738,
      "num_tokens": 42299418.0,
      "step": 19545
    },
    {
      "entropy": 6.431255245208741,
      "epoch": 2.092032746535395,
      "grad_norm": 1.375,
      "learning_rate": 0.00045675426211716963,
      "loss": 5.7638,
      "mean_token_accuracy": 0.1732567459344864,
      "num_tokens": 42310901.0,
      "step": 19550
    },
    {
      "entropy": 6.5339454174041744,
      "epoch": 2.092567820643159,
      "grad_norm": 1.125,
      "learning_rate": 0.0004567317224129069,
      "loss": 5.8383,
      "mean_token_accuracy": 0.17024020552635194,
      "num_tokens": 42320055.0,
      "step": 19555
    },
    {
      "entropy": 6.5529656410217285,
      "epoch": 2.093102894750923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004567091774611661,
      "loss": 5.9402,
      "mean_token_accuracy": 0.16383353471755982,
      "num_tokens": 42330393.0,
      "step": 19560
    },
    {
      "entropy": 6.451726818084717,
      "epoch": 2.093637968858687,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045668662726259846,
      "loss": 5.855,
      "mean_token_accuracy": 0.16566202342510222,
      "num_tokens": 42341384.0,
      "step": 19565
    },
    {
      "entropy": 6.434664964675903,
      "epoch": 2.094173042966451,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045666407181785496,
      "loss": 5.7889,
      "mean_token_accuracy": 0.1780661165714264,
      "num_tokens": 42352351.0,
      "step": 19570
    },
    {
      "entropy": 6.3793590545654295,
      "epoch": 2.094708117074215,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004566415111275869,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1667708307504654,
      "num_tokens": 42363912.0,
      "step": 19575
    },
    {
      "entropy": 6.471550798416137,
      "epoch": 2.0952431911819787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004566189451924457,
      "loss": 5.8875,
      "mean_token_accuracy": 0.16412573009729386,
      "num_tokens": 42374638.0,
      "step": 19580
    },
    {
      "entropy": 6.458401823043824,
      "epoch": 2.0957782652897428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045659637401308296,
      "loss": 5.8261,
      "mean_token_accuracy": 0.17475943714380265,
      "num_tokens": 42385976.0,
      "step": 19585
    },
    {
      "entropy": 6.460292005538941,
      "epoch": 2.0963133393975064,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004565737975901505,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16003927141427993,
      "num_tokens": 42395714.0,
      "step": 19590
    },
    {
      "entropy": 6.429117488861084,
      "epoch": 2.0968484135052705,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004565512159243001,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16980321258306502,
      "num_tokens": 42406027.0,
      "step": 19595
    },
    {
      "entropy": 6.518707704544068,
      "epoch": 2.0973834876130346,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004565286290161839,
      "loss": 5.9245,
      "mean_token_accuracy": 0.1615753188729286,
      "num_tokens": 42417756.0,
      "step": 19600
    },
    {
      "entropy": 6.494333505630493,
      "epoch": 2.0979185617207983,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045650603686645403,
      "loss": 5.927,
      "mean_token_accuracy": 0.16082765460014342,
      "num_tokens": 42429321.0,
      "step": 19605
    },
    {
      "entropy": 6.41153883934021,
      "epoch": 2.0984536358285624,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045648343947576284,
      "loss": 5.7514,
      "mean_token_accuracy": 0.1737269401550293,
      "num_tokens": 42440003.0,
      "step": 19610
    },
    {
      "entropy": 6.474349069595337,
      "epoch": 2.098988709936326,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004564608368447628,
      "loss": 5.8529,
      "mean_token_accuracy": 0.1653352752327919,
      "num_tokens": 42450743.0,
      "step": 19615
    },
    {
      "entropy": 6.450653028488159,
      "epoch": 2.09952378404409,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004564382289741067,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16511894911527633,
      "num_tokens": 42460885.0,
      "step": 19620
    },
    {
      "entropy": 6.466882991790771,
      "epoch": 2.1000588581518542,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045641561586444706,
      "loss": 5.8522,
      "mean_token_accuracy": 0.17066492587327958,
      "num_tokens": 42470854.0,
      "step": 19625
    },
    {
      "entropy": 6.469440031051636,
      "epoch": 2.100593932259618,
      "grad_norm": 1.0,
      "learning_rate": 0.00045639299751643714,
      "loss": 5.9305,
      "mean_token_accuracy": 0.16156110316514968,
      "num_tokens": 42482392.0,
      "step": 19630
    },
    {
      "entropy": 6.45660662651062,
      "epoch": 2.101129006367382,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045637037393072996,
      "loss": 5.8794,
      "mean_token_accuracy": 0.157142373919487,
      "num_tokens": 42494156.0,
      "step": 19635
    },
    {
      "entropy": 6.478666830062866,
      "epoch": 2.1016640804751456,
      "grad_norm": 1.125,
      "learning_rate": 0.0004563477451079786,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16507073044776915,
      "num_tokens": 42505185.0,
      "step": 19640
    },
    {
      "entropy": 6.440119409561158,
      "epoch": 2.1021991545829097,
      "grad_norm": 1.0,
      "learning_rate": 0.00045632511104883657,
      "loss": 5.8019,
      "mean_token_accuracy": 0.16433438211679458,
      "num_tokens": 42516415.0,
      "step": 19645
    },
    {
      "entropy": 6.454553890228271,
      "epoch": 2.102734228690674,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004563024717539575,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16787300407886505,
      "num_tokens": 42526945.0,
      "step": 19650
    },
    {
      "entropy": 6.503728818893433,
      "epoch": 2.1032693027984375,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045627982722399506,
      "loss": 5.928,
      "mean_token_accuracy": 0.16380500346422194,
      "num_tokens": 42537629.0,
      "step": 19655
    },
    {
      "entropy": 6.4140904426574705,
      "epoch": 2.1038043769062016,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045625717745960306,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16761477440595626,
      "num_tokens": 42547952.0,
      "step": 19660
    },
    {
      "entropy": 6.444326305389405,
      "epoch": 2.1043394510139652,
      "grad_norm": 0.96484375,
      "learning_rate": 0.00045623452246143547,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16423785090446472,
      "num_tokens": 42559026.0,
      "step": 19665
    },
    {
      "entropy": 6.5385191440582275,
      "epoch": 2.1048745251217293,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004562118622301466,
      "loss": 5.9222,
      "mean_token_accuracy": 0.1639838770031929,
      "num_tokens": 42570046.0,
      "step": 19670
    },
    {
      "entropy": 6.550991058349609,
      "epoch": 2.1054095992294934,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004561891967663906,
      "loss": 5.91,
      "mean_token_accuracy": 0.15914071053266526,
      "num_tokens": 42580440.0,
      "step": 19675
    },
    {
      "entropy": 6.477895641326905,
      "epoch": 2.105944673337257,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045616652607082203,
      "loss": 5.8537,
      "mean_token_accuracy": 0.16097139120101928,
      "num_tokens": 42591872.0,
      "step": 19680
    },
    {
      "entropy": 6.462870407104492,
      "epoch": 2.106479747445021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004561438501440955,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16488580107688905,
      "num_tokens": 42602843.0,
      "step": 19685
    },
    {
      "entropy": 6.417347145080567,
      "epoch": 2.107014821552785,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004561211689868657,
      "loss": 5.8896,
      "mean_token_accuracy": 0.1733737736940384,
      "num_tokens": 42613476.0,
      "step": 19690
    },
    {
      "entropy": 6.381961441040039,
      "epoch": 2.107549895660549,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045609848259978766,
      "loss": 5.761,
      "mean_token_accuracy": 0.1792847380042076,
      "num_tokens": 42624059.0,
      "step": 19695
    },
    {
      "entropy": 6.497478389739991,
      "epoch": 2.108084969768313,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004560757909835164,
      "loss": 5.9356,
      "mean_token_accuracy": 0.1636153444647789,
      "num_tokens": 42633980.0,
      "step": 19700
    },
    {
      "entropy": 6.549127244949341,
      "epoch": 2.1086200438760767,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045605309413870697,
      "loss": 5.9426,
      "mean_token_accuracy": 0.1596073940396309,
      "num_tokens": 42645504.0,
      "step": 19705
    },
    {
      "entropy": 6.5248266696929935,
      "epoch": 2.109155117983841,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004560303920660151,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16221111565828322,
      "num_tokens": 42655403.0,
      "step": 19710
    },
    {
      "entropy": 6.496182060241699,
      "epoch": 2.1096901920916045,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045600768476609596,
      "loss": 5.9078,
      "mean_token_accuracy": 0.16884891539812089,
      "num_tokens": 42666236.0,
      "step": 19715
    },
    {
      "entropy": 6.507245683670044,
      "epoch": 2.1102252661993686,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045598497223960533,
      "loss": 5.9792,
      "mean_token_accuracy": 0.15719814896583556,
      "num_tokens": 42677365.0,
      "step": 19720
    },
    {
      "entropy": 6.488958692550659,
      "epoch": 2.1107603403071327,
      "grad_norm": 1.125,
      "learning_rate": 0.00045596225448719913,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16546755880117417,
      "num_tokens": 42688518.0,
      "step": 19725
    },
    {
      "entropy": 6.4860169887542725,
      "epoch": 2.1112954144148963,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045593953150953326,
      "loss": 5.8774,
      "mean_token_accuracy": 0.1670508563518524,
      "num_tokens": 42699877.0,
      "step": 19730
    },
    {
      "entropy": 6.461322975158692,
      "epoch": 2.1118304885226604,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004559168033072638,
      "loss": 5.8652,
      "mean_token_accuracy": 0.17132428884506226,
      "num_tokens": 42710191.0,
      "step": 19735
    },
    {
      "entropy": 6.481604480743409,
      "epoch": 2.1123655626304245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045589406988104717,
      "loss": 5.9381,
      "mean_token_accuracy": 0.16343845278024674,
      "num_tokens": 42720874.0,
      "step": 19740
    },
    {
      "entropy": 6.480198192596435,
      "epoch": 2.112900636738188,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004558713312315396,
      "loss": 5.8813,
      "mean_token_accuracy": 0.16386687010526657,
      "num_tokens": 42731783.0,
      "step": 19745
    },
    {
      "entropy": 6.451361179351807,
      "epoch": 2.1134357108459523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004558485873593978,
      "loss": 5.8537,
      "mean_token_accuracy": 0.16167689710855485,
      "num_tokens": 42741602.0,
      "step": 19750
    },
    {
      "entropy": 6.528582668304443,
      "epoch": 2.113970784953716,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045582583826527835,
      "loss": 5.8303,
      "mean_token_accuracy": 0.17044297158718108,
      "num_tokens": 42751249.0,
      "step": 19755
    },
    {
      "entropy": 6.4624090671539305,
      "epoch": 2.11450585906148,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004558030839498383,
      "loss": 5.8402,
      "mean_token_accuracy": 0.16705645322799684,
      "num_tokens": 42762083.0,
      "step": 19760
    },
    {
      "entropy": 6.3923351764678955,
      "epoch": 2.115040933169244,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004557803244137347,
      "loss": 5.7124,
      "mean_token_accuracy": 0.1817711502313614,
      "num_tokens": 42773443.0,
      "step": 19765
    },
    {
      "entropy": 6.455114078521729,
      "epoch": 2.115576007277008,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004557575596576245,
      "loss": 5.9013,
      "mean_token_accuracy": 0.15753695666790007,
      "num_tokens": 42784126.0,
      "step": 19770
    },
    {
      "entropy": 6.443821382522583,
      "epoch": 2.116111081384772,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045573478968216526,
      "loss": 5.7946,
      "mean_token_accuracy": 0.17330519407987593,
      "num_tokens": 42795472.0,
      "step": 19775
    },
    {
      "entropy": 6.463377237319946,
      "epoch": 2.1166461554925355,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045571201448801424,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1673652559518814,
      "num_tokens": 42805279.0,
      "step": 19780
    },
    {
      "entropy": 6.487586736679077,
      "epoch": 2.1171812296002996,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004556892340758293,
      "loss": 5.8843,
      "mean_token_accuracy": 0.16532934606075286,
      "num_tokens": 42815680.0,
      "step": 19785
    },
    {
      "entropy": 6.492253065109253,
      "epoch": 2.1177163037080637,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004556664484462681,
      "loss": 5.9444,
      "mean_token_accuracy": 0.1554264523088932,
      "num_tokens": 42826724.0,
      "step": 19790
    },
    {
      "entropy": 6.484016084671021,
      "epoch": 2.1182513778158274,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045564365759998856,
      "loss": 5.8681,
      "mean_token_accuracy": 0.1637852668762207,
      "num_tokens": 42837875.0,
      "step": 19795
    },
    {
      "entropy": 6.45098180770874,
      "epoch": 2.1187864519235915,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004556208615376488,
      "loss": 5.8702,
      "mean_token_accuracy": 0.15743426829576493,
      "num_tokens": 42848714.0,
      "step": 19800
    },
    {
      "entropy": 6.4815489768981935,
      "epoch": 2.119321526031355,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004555980602599071,
      "loss": 5.9254,
      "mean_token_accuracy": 0.16415486857295036,
      "num_tokens": 42859764.0,
      "step": 19805
    },
    {
      "entropy": 6.3967061042785645,
      "epoch": 2.1198566001391193,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004555752537674217,
      "loss": 5.7641,
      "mean_token_accuracy": 0.1690279558300972,
      "num_tokens": 42870346.0,
      "step": 19810
    },
    {
      "entropy": 6.424504804611206,
      "epoch": 2.1203916742468834,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004555524420608513,
      "loss": 5.8708,
      "mean_token_accuracy": 0.16756978780031204,
      "num_tokens": 42882706.0,
      "step": 19815
    },
    {
      "entropy": 6.462999248504639,
      "epoch": 2.120926748354647,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045552962514085443,
      "loss": 5.8581,
      "mean_token_accuracy": 0.1772725060582161,
      "num_tokens": 42893611.0,
      "step": 19820
    },
    {
      "entropy": 6.503946828842163,
      "epoch": 2.121461822462411,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045550680300809003,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16296734735369683,
      "num_tokens": 42904034.0,
      "step": 19825
    },
    {
      "entropy": 6.5209479331970215,
      "epoch": 2.1219968965701748,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045548397566321706,
      "loss": 5.841,
      "mean_token_accuracy": 0.16554945856332778,
      "num_tokens": 42915530.0,
      "step": 19830
    },
    {
      "entropy": 6.4289655685424805,
      "epoch": 2.122531970677939,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004554611431068947,
      "loss": 5.8246,
      "mean_token_accuracy": 0.16544085294008254,
      "num_tokens": 42926696.0,
      "step": 19835
    },
    {
      "entropy": 6.475296211242676,
      "epoch": 2.123067044785703,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004554383053397822,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16922033429145814,
      "num_tokens": 42936985.0,
      "step": 19840
    },
    {
      "entropy": 6.479017543792724,
      "epoch": 2.1236021188934666,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045541546236253896,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16800356656312943,
      "num_tokens": 42947996.0,
      "step": 19845
    },
    {
      "entropy": 6.43947114944458,
      "epoch": 2.1241371930012307,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045539261417582456,
      "loss": 5.7677,
      "mean_token_accuracy": 0.17180561721324922,
      "num_tokens": 42959669.0,
      "step": 19850
    },
    {
      "entropy": 6.441443157196045,
      "epoch": 2.124672267108995,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00045536976078029874,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17119500637054444,
      "num_tokens": 42969687.0,
      "step": 19855
    },
    {
      "entropy": 6.494623613357544,
      "epoch": 2.1252073412167585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004553469021766215,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15882987529039383,
      "num_tokens": 42980752.0,
      "step": 19860
    },
    {
      "entropy": 6.492315721511841,
      "epoch": 2.1257424153245226,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045532403836545286,
      "loss": 5.9489,
      "mean_token_accuracy": 0.16198266297578812,
      "num_tokens": 42991949.0,
      "step": 19865
    },
    {
      "entropy": 6.406301212310791,
      "epoch": 2.1262774894322862,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004553011693474528,
      "loss": 5.8054,
      "mean_token_accuracy": 0.17907581180334092,
      "num_tokens": 43002689.0,
      "step": 19870
    },
    {
      "entropy": 6.4892302513122555,
      "epoch": 2.1268125635400503,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004552782951232819,
      "loss": 5.8622,
      "mean_token_accuracy": 0.16351267844438552,
      "num_tokens": 43012939.0,
      "step": 19875
    },
    {
      "entropy": 6.480828332901001,
      "epoch": 2.1273476376478144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045525541569360053,
      "loss": 5.8289,
      "mean_token_accuracy": 0.17394785732030868,
      "num_tokens": 43023887.0,
      "step": 19880
    },
    {
      "entropy": 6.39893069267273,
      "epoch": 2.127882711755578,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004552325310590694,
      "loss": 5.7636,
      "mean_token_accuracy": 0.1698220416903496,
      "num_tokens": 43034554.0,
      "step": 19885
    },
    {
      "entropy": 6.472014760971069,
      "epoch": 2.128417785863342,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004552096412203492,
      "loss": 5.9227,
      "mean_token_accuracy": 0.16225723922252655,
      "num_tokens": 43046322.0,
      "step": 19890
    },
    {
      "entropy": 6.494905042648315,
      "epoch": 2.128952859971106,
      "grad_norm": 1.125,
      "learning_rate": 0.000455186746178101,
      "loss": 5.802,
      "mean_token_accuracy": 0.17069389671087265,
      "num_tokens": 43055981.0,
      "step": 19895
    },
    {
      "entropy": 6.519297981262207,
      "epoch": 2.12948793407887,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045516384593298584,
      "loss": 5.8661,
      "mean_token_accuracy": 0.15961899906396865,
      "num_tokens": 43067217.0,
      "step": 19900
    },
    {
      "entropy": 6.471814441680908,
      "epoch": 2.130023008186634,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045514094048566486,
      "loss": 5.8668,
      "mean_token_accuracy": 0.1638723075389862,
      "num_tokens": 43078312.0,
      "step": 19905
    },
    {
      "entropy": 6.32650465965271,
      "epoch": 2.1305580822943977,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045511802983679955,
      "loss": 5.6818,
      "mean_token_accuracy": 0.19024215042591094,
      "num_tokens": 43090210.0,
      "step": 19910
    },
    {
      "entropy": 6.456379222869873,
      "epoch": 2.131093156402162,
      "grad_norm": 1.125,
      "learning_rate": 0.0004550951139870515,
      "loss": 5.81,
      "mean_token_accuracy": 0.17028167247772216,
      "num_tokens": 43101282.0,
      "step": 19915
    },
    {
      "entropy": 6.468144464492798,
      "epoch": 2.1316282305099254,
      "grad_norm": 0.96875,
      "learning_rate": 0.00045507219293708227,
      "loss": 5.9529,
      "mean_token_accuracy": 0.1642445892095566,
      "num_tokens": 43112856.0,
      "step": 19920
    },
    {
      "entropy": 6.458612537384033,
      "epoch": 2.1321633046176895,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045504926668755384,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16859571188688277,
      "num_tokens": 43123898.0,
      "step": 19925
    },
    {
      "entropy": 6.487282228469849,
      "epoch": 2.1326983787254536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004550263352391281,
      "loss": 5.8728,
      "mean_token_accuracy": 0.16914221197366713,
      "num_tokens": 43134651.0,
      "step": 19930
    },
    {
      "entropy": 6.484668254852295,
      "epoch": 2.1332334528332173,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004550033985924672,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16201801598072052,
      "num_tokens": 43145385.0,
      "step": 19935
    },
    {
      "entropy": 6.485315036773682,
      "epoch": 2.1337685269409814,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004549804567482335,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16419271230697632,
      "num_tokens": 43157356.0,
      "step": 19940
    },
    {
      "entropy": 6.485888338088989,
      "epoch": 2.134303601048745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045495750970708944,
      "loss": 5.8765,
      "mean_token_accuracy": 0.1689384922385216,
      "num_tokens": 43168289.0,
      "step": 19945
    },
    {
      "entropy": 6.3840028762817385,
      "epoch": 2.134838675156509,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045493455746969746,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16823284178972245,
      "num_tokens": 43177796.0,
      "step": 19950
    },
    {
      "entropy": 6.412529706954956,
      "epoch": 2.1353737492642733,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004549116000367205,
      "loss": 5.841,
      "mean_token_accuracy": 0.16815511137247086,
      "num_tokens": 43188846.0,
      "step": 19955
    },
    {
      "entropy": 6.535986328125,
      "epoch": 2.135908823372037,
      "grad_norm": 1.0,
      "learning_rate": 0.00045488863740882133,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1677407369017601,
      "num_tokens": 43199852.0,
      "step": 19960
    },
    {
      "entropy": 6.481540155410767,
      "epoch": 2.136443897479801,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045486566958666307,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1670091062784195,
      "num_tokens": 43210591.0,
      "step": 19965
    },
    {
      "entropy": 6.4823895454406735,
      "epoch": 2.136978971587565,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045484269657090887,
      "loss": 5.8271,
      "mean_token_accuracy": 0.15958329290151596,
      "num_tokens": 43221760.0,
      "step": 19970
    },
    {
      "entropy": 6.510053586959839,
      "epoch": 2.1375140456953288,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045481971836222205,
      "loss": 5.9236,
      "mean_token_accuracy": 0.15849653929471968,
      "num_tokens": 43233693.0,
      "step": 19975
    },
    {
      "entropy": 6.4962057113647464,
      "epoch": 2.138049119803093,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045479673496126615,
      "loss": 5.8979,
      "mean_token_accuracy": 0.1664629966020584,
      "num_tokens": 43244565.0,
      "step": 19980
    },
    {
      "entropy": 6.4309244632720945,
      "epoch": 2.1385841939108565,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045477374636870485,
      "loss": 5.8198,
      "mean_token_accuracy": 0.16839751601219177,
      "num_tokens": 43254913.0,
      "step": 19985
    },
    {
      "entropy": 6.518014764785766,
      "epoch": 2.1391192680186206,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045475075258520175,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16478254199028014,
      "num_tokens": 43265525.0,
      "step": 19990
    },
    {
      "entropy": 6.529369258880616,
      "epoch": 2.1396543421263843,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004547277536114211,
      "loss": 5.9525,
      "mean_token_accuracy": 0.15578901022672653,
      "num_tokens": 43276954.0,
      "step": 19995
    },
    {
      "entropy": 6.496371173858643,
      "epoch": 2.1401894162341484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004547047494480267,
      "loss": 5.8673,
      "mean_token_accuracy": 0.16268888115882874,
      "num_tokens": 43287491.0,
      "step": 20000
    },
    {
      "entropy": 6.453723287582397,
      "epoch": 2.1407244903419125,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004546817400956829,
      "loss": 5.9051,
      "mean_token_accuracy": 0.15941646844148635,
      "num_tokens": 43298341.0,
      "step": 20005
    },
    {
      "entropy": 6.410027551651001,
      "epoch": 2.141259564449676,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045465872555505417,
      "loss": 5.8293,
      "mean_token_accuracy": 0.17429626137018203,
      "num_tokens": 43309450.0,
      "step": 20010
    },
    {
      "entropy": 6.500331735610962,
      "epoch": 2.1417946385574402,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000454635705826805,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1593199536204338,
      "num_tokens": 43320126.0,
      "step": 20015
    },
    {
      "entropy": 6.5411436557769775,
      "epoch": 2.1423297126652043,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004546126809116,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15641611367464064,
      "num_tokens": 43331347.0,
      "step": 20020
    },
    {
      "entropy": 6.509288740158081,
      "epoch": 2.142864786772968,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004545896508101041,
      "loss": 5.8463,
      "mean_token_accuracy": 0.1651618719100952,
      "num_tokens": 43341351.0,
      "step": 20025
    },
    {
      "entropy": 6.518145656585693,
      "epoch": 2.143399860880732,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004545666155229823,
      "loss": 5.9296,
      "mean_token_accuracy": 0.15940297842025758,
      "num_tokens": 43352471.0,
      "step": 20030
    },
    {
      "entropy": 6.5085005283355715,
      "epoch": 2.1439349349884957,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045454357505089965,
      "loss": 5.7948,
      "mean_token_accuracy": 0.1669934630393982,
      "num_tokens": 43363041.0,
      "step": 20035
    },
    {
      "entropy": 6.4582839012146,
      "epoch": 2.14447000909626,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004545205293945215,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16373032033443452,
      "num_tokens": 43375034.0,
      "step": 20040
    },
    {
      "entropy": 6.528272581100464,
      "epoch": 2.145005083204024,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004544974785545134,
      "loss": 5.9505,
      "mean_token_accuracy": 0.15853089988231658,
      "num_tokens": 43386110.0,
      "step": 20045
    },
    {
      "entropy": 6.495435285568237,
      "epoch": 2.1455401573117876,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544744225315407,
      "loss": 5.8748,
      "mean_token_accuracy": 0.1646910309791565,
      "num_tokens": 43396150.0,
      "step": 20050
    },
    {
      "entropy": 6.4745840549469,
      "epoch": 2.1460752314195517,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004544513613262693,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16283731460571288,
      "num_tokens": 43407237.0,
      "step": 20055
    },
    {
      "entropy": 6.424422073364258,
      "epoch": 2.1466103055273154,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004544282949393652,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17108509987592696,
      "num_tokens": 43417626.0,
      "step": 20060
    },
    {
      "entropy": 6.452867937088013,
      "epoch": 2.1471453796350795,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004544052233714941,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16590890437364578,
      "num_tokens": 43427368.0,
      "step": 20065
    },
    {
      "entropy": 6.449078321456909,
      "epoch": 2.1476804537428436,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045438214662332245,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16396653652191162,
      "num_tokens": 43438159.0,
      "step": 20070
    },
    {
      "entropy": 6.482664108276367,
      "epoch": 2.148215527850607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045435906469551657,
      "loss": 5.8866,
      "mean_token_accuracy": 0.1636155441403389,
      "num_tokens": 43447403.0,
      "step": 20075
    },
    {
      "entropy": 6.4264003276824955,
      "epoch": 2.1487506019583713,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045433597758874286,
      "loss": 5.7684,
      "mean_token_accuracy": 0.17984655797481536,
      "num_tokens": 43457713.0,
      "step": 20080
    },
    {
      "entropy": 6.494991636276245,
      "epoch": 2.149285676066135,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000454312885303668,
      "loss": 5.898,
      "mean_token_accuracy": 0.16288311183452606,
      "num_tokens": 43468520.0,
      "step": 20085
    },
    {
      "entropy": 6.529734373092651,
      "epoch": 2.149820750173899,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045428978784095873,
      "loss": 5.9078,
      "mean_token_accuracy": 0.16344235539436341,
      "num_tokens": 43480501.0,
      "step": 20090
    },
    {
      "entropy": 6.4901337146759035,
      "epoch": 2.150355824281663,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045426668520128204,
      "loss": 5.8557,
      "mean_token_accuracy": 0.1666504204273224,
      "num_tokens": 43491297.0,
      "step": 20095
    },
    {
      "entropy": 6.49347038269043,
      "epoch": 2.150890898389427,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004542435773853051,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16233655363321303,
      "num_tokens": 43501212.0,
      "step": 20100
    },
    {
      "entropy": 6.454028844833374,
      "epoch": 2.151425972497191,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000454220464393695,
      "loss": 5.9137,
      "mean_token_accuracy": 0.15748606473207474,
      "num_tokens": 43511136.0,
      "step": 20105
    },
    {
      "entropy": 6.5523059368133545,
      "epoch": 2.1519610466049546,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004541973462271192,
      "loss": 5.924,
      "mean_token_accuracy": 0.16211857348680497,
      "num_tokens": 43522006.0,
      "step": 20110
    },
    {
      "entropy": 6.468632698059082,
      "epoch": 2.1524961207127187,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004541742228862451,
      "loss": 5.8746,
      "mean_token_accuracy": 0.1661093831062317,
      "num_tokens": 43532746.0,
      "step": 20115
    },
    {
      "entropy": 6.352495098114014,
      "epoch": 2.1530311948204828,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045415109437174064,
      "loss": 5.8445,
      "mean_token_accuracy": 0.1679021120071411,
      "num_tokens": 43543707.0,
      "step": 20120
    },
    {
      "entropy": 6.489015388488769,
      "epoch": 2.1535662689282464,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045412796068427337,
      "loss": 5.843,
      "mean_token_accuracy": 0.16258713752031326,
      "num_tokens": 43555029.0,
      "step": 20125
    },
    {
      "entropy": 6.536959409713745,
      "epoch": 2.1541013430360105,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045410482182451153,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16469907909631729,
      "num_tokens": 43565639.0,
      "step": 20130
    },
    {
      "entropy": 6.425443077087403,
      "epoch": 2.1546364171437746,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045408167779312306,
      "loss": 5.7944,
      "mean_token_accuracy": 0.1722143918275833,
      "num_tokens": 43576333.0,
      "step": 20135
    },
    {
      "entropy": 6.47620849609375,
      "epoch": 2.1551714912515383,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045405852859077626,
      "loss": 5.8354,
      "mean_token_accuracy": 0.1641660988330841,
      "num_tokens": 43587012.0,
      "step": 20140
    },
    {
      "entropy": 6.459071063995362,
      "epoch": 2.1557065653593024,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045403537421813975,
      "loss": 5.7829,
      "mean_token_accuracy": 0.17521045953035355,
      "num_tokens": 43597779.0,
      "step": 20145
    },
    {
      "entropy": 6.484965181350708,
      "epoch": 2.156241639467066,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045401221467588184,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16601230949163437,
      "num_tokens": 43608631.0,
      "step": 20150
    },
    {
      "entropy": 6.458986282348633,
      "epoch": 2.15677671357483,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004539890499646714,
      "loss": 5.8051,
      "mean_token_accuracy": 0.1735161989927292,
      "num_tokens": 43620254.0,
      "step": 20155
    },
    {
      "entropy": 6.4482419967651365,
      "epoch": 2.1573117876825942,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004539658800851773,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1680944472551346,
      "num_tokens": 43630986.0,
      "step": 20160
    },
    {
      "entropy": 6.450579023361206,
      "epoch": 2.157846861790358,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004539427050380686,
      "loss": 5.7937,
      "mean_token_accuracy": 0.16388647109270096,
      "num_tokens": 43641260.0,
      "step": 20165
    },
    {
      "entropy": 6.506737518310547,
      "epoch": 2.158381935898122,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045391952482401444,
      "loss": 5.9473,
      "mean_token_accuracy": 0.16380299925804137,
      "num_tokens": 43652093.0,
      "step": 20170
    },
    {
      "entropy": 6.455452728271484,
      "epoch": 2.1589170100058857,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045389633944368406,
      "loss": 5.8602,
      "mean_token_accuracy": 0.16357452124357225,
      "num_tokens": 43662799.0,
      "step": 20175
    },
    {
      "entropy": 6.4199299812316895,
      "epoch": 2.1594520841136498,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004538731488977471,
      "loss": 5.8016,
      "mean_token_accuracy": 0.16757601499557495,
      "num_tokens": 43674198.0,
      "step": 20180
    },
    {
      "entropy": 6.486158323287964,
      "epoch": 2.159987158221414,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004538499531868731,
      "loss": 5.8698,
      "mean_token_accuracy": 0.1655881494283676,
      "num_tokens": 43684841.0,
      "step": 20185
    },
    {
      "entropy": 6.506292104721069,
      "epoch": 2.1605222323291775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045382675231173174,
      "loss": 5.948,
      "mean_token_accuracy": 0.1600341483950615,
      "num_tokens": 43695903.0,
      "step": 20190
    },
    {
      "entropy": 6.486567831039428,
      "epoch": 2.1610573064369416,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004538035462729931,
      "loss": 5.8577,
      "mean_token_accuracy": 0.1640781953930855,
      "num_tokens": 43706740.0,
      "step": 20195
    },
    {
      "entropy": 6.432819890975952,
      "epoch": 2.1615923805447053,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004537803350713273,
      "loss": 5.8076,
      "mean_token_accuracy": 0.16696966290473939,
      "num_tokens": 43717124.0,
      "step": 20200
    },
    {
      "entropy": 6.396174716949463,
      "epoch": 2.1621274546524694,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004537571187074043,
      "loss": 5.8177,
      "mean_token_accuracy": 0.1734643667936325,
      "num_tokens": 43727346.0,
      "step": 20205
    },
    {
      "entropy": 6.538708448410034,
      "epoch": 2.1626625287602335,
      "grad_norm": 1.125,
      "learning_rate": 0.00045373389718189464,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15156850665807725,
      "num_tokens": 43738297.0,
      "step": 20210
    },
    {
      "entropy": 6.5128693103790285,
      "epoch": 2.163197602867997,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004537106704954689,
      "loss": 5.8707,
      "mean_token_accuracy": 0.15931367725133896,
      "num_tokens": 43749543.0,
      "step": 20215
    },
    {
      "entropy": 6.510898113250732,
      "epoch": 2.163732676975761,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045368743864879755,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16607716381549836,
      "num_tokens": 43760324.0,
      "step": 20220
    },
    {
      "entropy": 6.5256859302520756,
      "epoch": 2.164267751083525,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045366420164255156,
      "loss": 5.9233,
      "mean_token_accuracy": 0.16102688759565353,
      "num_tokens": 43770391.0,
      "step": 20225
    },
    {
      "entropy": 6.430365133285522,
      "epoch": 2.164802825191289,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004536409594774019,
      "loss": 5.8251,
      "mean_token_accuracy": 0.15968013405799866,
      "num_tokens": 43781516.0,
      "step": 20230
    },
    {
      "entropy": 6.4198507308959964,
      "epoch": 2.165337899299053,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045361771215401953,
      "loss": 5.8573,
      "mean_token_accuracy": 0.1723635345697403,
      "num_tokens": 43794152.0,
      "step": 20235
    },
    {
      "entropy": 6.51537127494812,
      "epoch": 2.1658729734068167,
      "grad_norm": 1.125,
      "learning_rate": 0.00045359445967307595,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16747324019670487,
      "num_tokens": 43805712.0,
      "step": 20240
    },
    {
      "entropy": 6.502164316177368,
      "epoch": 2.166408047514581,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004535712020352423,
      "loss": 5.9168,
      "mean_token_accuracy": 0.16027720123529435,
      "num_tokens": 43816985.0,
      "step": 20245
    },
    {
      "entropy": 6.465137052536011,
      "epoch": 2.166943121622345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004535479392411904,
      "loss": 5.8237,
      "mean_token_accuracy": 0.16319603770971297,
      "num_tokens": 43828048.0,
      "step": 20250
    },
    {
      "entropy": 6.491486215591431,
      "epoch": 2.1674781957301086,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045352467129159183,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16337529122829436,
      "num_tokens": 43838507.0,
      "step": 20255
    },
    {
      "entropy": 6.390494441986084,
      "epoch": 2.1680132698378727,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045350139818711846,
      "loss": 5.7701,
      "mean_token_accuracy": 0.17159541696310043,
      "num_tokens": 43849328.0,
      "step": 20260
    },
    {
      "entropy": 6.4341095924377445,
      "epoch": 2.1685483439456363,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004534781199284423,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16150501668453215,
      "num_tokens": 43860311.0,
      "step": 20265
    },
    {
      "entropy": 6.525981283187866,
      "epoch": 2.1690834180534004,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004534548365162354,
      "loss": 5.9276,
      "mean_token_accuracy": 0.16084444522857666,
      "num_tokens": 43871167.0,
      "step": 20270
    },
    {
      "entropy": 6.465597200393677,
      "epoch": 2.169618492161164,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045343154795117024,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17480555325746536,
      "num_tokens": 43882332.0,
      "step": 20275
    },
    {
      "entropy": 6.507810068130493,
      "epoch": 2.170153566268928,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004534082542339192,
      "loss": 5.7693,
      "mean_token_accuracy": 0.1666393205523491,
      "num_tokens": 43892971.0,
      "step": 20280
    },
    {
      "entropy": 6.457220602035522,
      "epoch": 2.1706886403766923,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045338495536515497,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16927892416715623,
      "num_tokens": 43904305.0,
      "step": 20285
    },
    {
      "entropy": 6.330482912063599,
      "epoch": 2.171223714484456,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004533616513455502,
      "loss": 5.8299,
      "mean_token_accuracy": 0.16147409826517106,
      "num_tokens": 43914431.0,
      "step": 20290
    },
    {
      "entropy": 6.433372592926025,
      "epoch": 2.17175878859222,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045333834217577775,
      "loss": 5.8598,
      "mean_token_accuracy": 0.17268452644348145,
      "num_tokens": 43925350.0,
      "step": 20295
    },
    {
      "entropy": 6.425912809371948,
      "epoch": 2.172293862699984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045331502785651075,
      "loss": 5.89,
      "mean_token_accuracy": 0.16455530375242233,
      "num_tokens": 43936194.0,
      "step": 20300
    },
    {
      "entropy": 6.512571382522583,
      "epoch": 2.172828936807748,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004532917083884224,
      "loss": 5.8696,
      "mean_token_accuracy": 0.160905858874321,
      "num_tokens": 43946692.0,
      "step": 20305
    },
    {
      "entropy": 6.478751611709595,
      "epoch": 2.173364010915512,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000453268383772186,
      "loss": 5.8145,
      "mean_token_accuracy": 0.16756239980459214,
      "num_tokens": 43957493.0,
      "step": 20310
    },
    {
      "entropy": 6.505754518508911,
      "epoch": 2.1738990850232756,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045324505400847506,
      "loss": 5.9315,
      "mean_token_accuracy": 0.16247908920049667,
      "num_tokens": 43970203.0,
      "step": 20315
    },
    {
      "entropy": 6.496106576919556,
      "epoch": 2.1744341591310397,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045322171909796327,
      "loss": 5.8449,
      "mean_token_accuracy": 0.1616058111190796,
      "num_tokens": 43980778.0,
      "step": 20320
    },
    {
      "entropy": 6.484720516204834,
      "epoch": 2.1749692332388038,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045319837904132435,
      "loss": 5.8209,
      "mean_token_accuracy": 0.17534621804952621,
      "num_tokens": 43992239.0,
      "step": 20325
    },
    {
      "entropy": 6.506054306030274,
      "epoch": 2.1755043073465674,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004531750338392323,
      "loss": 5.8819,
      "mean_token_accuracy": 0.1661924570798874,
      "num_tokens": 44001882.0,
      "step": 20330
    },
    {
      "entropy": 6.441217851638794,
      "epoch": 2.1760393814543315,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004531516834923611,
      "loss": 5.8569,
      "mean_token_accuracy": 0.1628911480307579,
      "num_tokens": 44013192.0,
      "step": 20335
    },
    {
      "entropy": 6.549660062789917,
      "epoch": 2.176574455562095,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004531283280013852,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15757093727588653,
      "num_tokens": 44023072.0,
      "step": 20340
    },
    {
      "entropy": 6.491693782806396,
      "epoch": 2.1771095296698593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045310496736697875,
      "loss": 5.8859,
      "mean_token_accuracy": 0.16530305445194243,
      "num_tokens": 44034238.0,
      "step": 20345
    },
    {
      "entropy": 6.50402340888977,
      "epoch": 2.1776446037776234,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045308160158981646,
      "loss": 5.8654,
      "mean_token_accuracy": 0.1661301463842392,
      "num_tokens": 44044392.0,
      "step": 20350
    },
    {
      "entropy": 6.557263040542603,
      "epoch": 2.178179677885387,
      "grad_norm": 1.109375,
      "learning_rate": 0.000453058230670573,
      "loss": 5.8968,
      "mean_token_accuracy": 0.16225262284278869,
      "num_tokens": 44056129.0,
      "step": 20355
    },
    {
      "entropy": 6.429604148864746,
      "epoch": 2.178714751993151,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045303485460992303,
      "loss": 5.8267,
      "mean_token_accuracy": 0.17296368330717088,
      "num_tokens": 44067600.0,
      "step": 20360
    },
    {
      "entropy": 6.446758365631103,
      "epoch": 2.1792498261009148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045301147340854177,
      "loss": 5.7413,
      "mean_token_accuracy": 0.17326935529708862,
      "num_tokens": 44077373.0,
      "step": 20365
    },
    {
      "entropy": 6.515975618362427,
      "epoch": 2.179784900208679,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045298808706710416,
      "loss": 5.9425,
      "mean_token_accuracy": 0.16317231953144073,
      "num_tokens": 44087422.0,
      "step": 20370
    },
    {
      "entropy": 6.478044033050537,
      "epoch": 2.180319974316443,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045296469558628555,
      "loss": 5.8936,
      "mean_token_accuracy": 0.16340119242668152,
      "num_tokens": 44098524.0,
      "step": 20375
    },
    {
      "entropy": 6.400179147720337,
      "epoch": 2.1808550484242066,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045294129896676134,
      "loss": 5.8321,
      "mean_token_accuracy": 0.17177040576934816,
      "num_tokens": 44108340.0,
      "step": 20380
    },
    {
      "entropy": 6.479645538330078,
      "epoch": 2.1813901225319707,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004529178972092071,
      "loss": 5.8558,
      "mean_token_accuracy": 0.16096516847610473,
      "num_tokens": 44118985.0,
      "step": 20385
    },
    {
      "entropy": 6.522010135650635,
      "epoch": 2.1819251966397344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045289449031429873,
      "loss": 5.8606,
      "mean_token_accuracy": 0.17317609190940858,
      "num_tokens": 44129126.0,
      "step": 20390
    },
    {
      "entropy": 6.5171044826507565,
      "epoch": 2.1824602707474985,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045287107828271184,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16763114631175996,
      "num_tokens": 44139731.0,
      "step": 20395
    },
    {
      "entropy": 6.476385927200317,
      "epoch": 2.1829953448552626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004528476611151226,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1622420459985733,
      "num_tokens": 44149924.0,
      "step": 20400
    },
    {
      "entropy": 6.45866961479187,
      "epoch": 2.1835304189630262,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004528242388122071,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16382619589567185,
      "num_tokens": 44161155.0,
      "step": 20405
    },
    {
      "entropy": 6.464540767669678,
      "epoch": 2.1840654930707903,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004528008113746417,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16506076902151107,
      "num_tokens": 44171563.0,
      "step": 20410
    },
    {
      "entropy": 6.434492015838623,
      "epoch": 2.1846005671785544,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045277737880310284,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16791648864746095,
      "num_tokens": 44182719.0,
      "step": 20415
    },
    {
      "entropy": 6.453215837478638,
      "epoch": 2.185135641286318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045275394109826715,
      "loss": 5.8072,
      "mean_token_accuracy": 0.17150099128484725,
      "num_tokens": 44192214.0,
      "step": 20420
    },
    {
      "entropy": 6.4589684963226315,
      "epoch": 2.185670715394082,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045273049826081145,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16372417360544206,
      "num_tokens": 44204074.0,
      "step": 20425
    },
    {
      "entropy": 6.551078271865845,
      "epoch": 2.186205789501846,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004527070502914125,
      "loss": 5.9678,
      "mean_token_accuracy": 0.1606735870242119,
      "num_tokens": 44214402.0,
      "step": 20430
    },
    {
      "entropy": 6.508745384216309,
      "epoch": 2.18674086360961,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004526835971907475,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16587917804718016,
      "num_tokens": 44224479.0,
      "step": 20435
    },
    {
      "entropy": 6.4783398628234865,
      "epoch": 2.187275937717374,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045266013895949356,
      "loss": 5.8933,
      "mean_token_accuracy": 0.1598525434732437,
      "num_tokens": 44234069.0,
      "step": 20440
    },
    {
      "entropy": 6.472596216201782,
      "epoch": 2.1878110118251377,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004526366755983281,
      "loss": 5.8716,
      "mean_token_accuracy": 0.1592169463634491,
      "num_tokens": 44244978.0,
      "step": 20445
    },
    {
      "entropy": 6.483738756179809,
      "epoch": 2.188346085932902,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004526132071079286,
      "loss": 5.8257,
      "mean_token_accuracy": 0.17532628327608107,
      "num_tokens": 44255657.0,
      "step": 20450
    },
    {
      "entropy": 6.414804458618164,
      "epoch": 2.1888811600406655,
      "grad_norm": 0.984375,
      "learning_rate": 0.00045258973348897267,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16419113278388978,
      "num_tokens": 44267370.0,
      "step": 20455
    },
    {
      "entropy": 6.439362096786499,
      "epoch": 2.1894162341484296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004525662547421382,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16552397161722182,
      "num_tokens": 44277855.0,
      "step": 20460
    },
    {
      "entropy": 6.493360996246338,
      "epoch": 2.1899513082561937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000452542770868103,
      "loss": 5.8315,
      "mean_token_accuracy": 0.17035821974277496,
      "num_tokens": 44288233.0,
      "step": 20465
    },
    {
      "entropy": 6.4315392017364506,
      "epoch": 2.1904863823639573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045251928186754533,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16926713436841964,
      "num_tokens": 44298054.0,
      "step": 20470
    },
    {
      "entropy": 6.459919023513794,
      "epoch": 2.1910214564717214,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004524957877411432,
      "loss": 5.8415,
      "mean_token_accuracy": 0.1677771896123886,
      "num_tokens": 44309435.0,
      "step": 20475
    },
    {
      "entropy": 6.496394920349121,
      "epoch": 2.191556530579485,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004524722884895753,
      "loss": 5.9203,
      "mean_token_accuracy": 0.16100984960794448,
      "num_tokens": 44320502.0,
      "step": 20480
    },
    {
      "entropy": 6.432992839813233,
      "epoch": 2.192091604687249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045244878411352,
      "loss": 5.7585,
      "mean_token_accuracy": 0.17735180407762527,
      "num_tokens": 44330911.0,
      "step": 20485
    },
    {
      "entropy": 6.4874450206756595,
      "epoch": 2.1926266787950133,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004524252746136559,
      "loss": 5.8868,
      "mean_token_accuracy": 0.160236756503582,
      "num_tokens": 44342159.0,
      "step": 20490
    },
    {
      "entropy": 6.475741958618164,
      "epoch": 2.193161752902777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045240175999066194,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16896373331546782,
      "num_tokens": 44352293.0,
      "step": 20495
    },
    {
      "entropy": 6.485591459274292,
      "epoch": 2.193696827010541,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045237824024521714,
      "loss": 5.9877,
      "mean_token_accuracy": 0.16158706396818162,
      "num_tokens": 44362321.0,
      "step": 20500
    },
    {
      "entropy": 6.4720508575439455,
      "epoch": 2.1942319011183047,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004523547153780005,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16433738321065902,
      "num_tokens": 44372682.0,
      "step": 20505
    },
    {
      "entropy": 6.502557802200317,
      "epoch": 2.194766975226069,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004523311853896915,
      "loss": 5.87,
      "mean_token_accuracy": 0.1665627419948578,
      "num_tokens": 44383482.0,
      "step": 20510
    },
    {
      "entropy": 6.337419271469116,
      "epoch": 2.195302049333833,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045230765028096925,
      "loss": 5.6714,
      "mean_token_accuracy": 0.19091199040412904,
      "num_tokens": 44395333.0,
      "step": 20515
    },
    {
      "entropy": 6.474304628372193,
      "epoch": 2.1958371234415965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045228411005251366,
      "loss": 5.9367,
      "mean_token_accuracy": 0.15517167299985885,
      "num_tokens": 44406210.0,
      "step": 20520
    },
    {
      "entropy": 6.523835611343384,
      "epoch": 2.1963721975493606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004522605647050043,
      "loss": 5.8792,
      "mean_token_accuracy": 0.1619505763053894,
      "num_tokens": 44415834.0,
      "step": 20525
    },
    {
      "entropy": 6.592779350280762,
      "epoch": 2.1969072716571247,
      "grad_norm": 1.0,
      "learning_rate": 0.00045223701423912103,
      "loss": 5.9301,
      "mean_token_accuracy": 0.16160725951194763,
      "num_tokens": 44427902.0,
      "step": 20530
    },
    {
      "entropy": 6.488599491119385,
      "epoch": 2.1974423457648884,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045221345865554386,
      "loss": 5.7843,
      "mean_token_accuracy": 0.16939454674720764,
      "num_tokens": 44437406.0,
      "step": 20535
    },
    {
      "entropy": 6.458600330352783,
      "epoch": 2.1979774198726525,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045218989795495294,
      "loss": 5.8116,
      "mean_token_accuracy": 0.16834816634654998,
      "num_tokens": 44446889.0,
      "step": 20540
    },
    {
      "entropy": 6.509384441375732,
      "epoch": 2.198512493980416,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045216633213802866,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16027392894029618,
      "num_tokens": 44457132.0,
      "step": 20545
    },
    {
      "entropy": 6.473583602905274,
      "epoch": 2.1990475680881802,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045214276120545137,
      "loss": 5.8884,
      "mean_token_accuracy": 0.16436221748590468,
      "num_tokens": 44468006.0,
      "step": 20550
    },
    {
      "entropy": 6.446438026428223,
      "epoch": 2.199582642195944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004521191851579019,
      "loss": 5.7752,
      "mean_token_accuracy": 0.17192772328853606,
      "num_tokens": 44478650.0,
      "step": 20555
    },
    {
      "entropy": 6.435380554199218,
      "epoch": 2.200117716303708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004520956039960607,
      "loss": 5.8985,
      "mean_token_accuracy": 0.16293962448835372,
      "num_tokens": 44489842.0,
      "step": 20560
    },
    {
      "entropy": 6.464346599578858,
      "epoch": 2.200652790411472,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045207201772060887,
      "loss": 5.8276,
      "mean_token_accuracy": 0.16762628704309462,
      "num_tokens": 44500444.0,
      "step": 20565
    },
    {
      "entropy": 6.4245658874511715,
      "epoch": 2.2011878645192358,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004520484263322274,
      "loss": 5.7852,
      "mean_token_accuracy": 0.1703597828745842,
      "num_tokens": 44510711.0,
      "step": 20570
    },
    {
      "entropy": 6.41108250617981,
      "epoch": 2.201722938627,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004520248298315975,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1678318202495575,
      "num_tokens": 44522343.0,
      "step": 20575
    },
    {
      "entropy": 6.459940385818482,
      "epoch": 2.202258012734764,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045200122821940057,
      "loss": 5.8867,
      "mean_token_accuracy": 0.1647458091378212,
      "num_tokens": 44533218.0,
      "step": 20580
    },
    {
      "entropy": 6.533621835708618,
      "epoch": 2.2027930868425276,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045197762149631795,
      "loss": 5.9143,
      "mean_token_accuracy": 0.1661139503121376,
      "num_tokens": 44543406.0,
      "step": 20585
    },
    {
      "entropy": 6.537712717056275,
      "epoch": 2.2033281609502917,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045195400966303146,
      "loss": 5.9219,
      "mean_token_accuracy": 0.16877577006816863,
      "num_tokens": 44555329.0,
      "step": 20590
    },
    {
      "entropy": 6.521209383010865,
      "epoch": 2.2038632350580554,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004519303927202228,
      "loss": 5.9349,
      "mean_token_accuracy": 0.1640845701098442,
      "num_tokens": 44565191.0,
      "step": 20595
    },
    {
      "entropy": 6.454711055755615,
      "epoch": 2.2043983091658195,
      "grad_norm": 1.03125,
      "learning_rate": 0.00045190677066857384,
      "loss": 5.8126,
      "mean_token_accuracy": 0.17683560103178025,
      "num_tokens": 44576374.0,
      "step": 20600
    },
    {
      "entropy": 6.425337457656861,
      "epoch": 2.2049333832735836,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004518831435087668,
      "loss": 5.7989,
      "mean_token_accuracy": 0.17098885178565978,
      "num_tokens": 44586533.0,
      "step": 20605
    },
    {
      "entropy": 6.392602968215942,
      "epoch": 2.2054684573813472,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004518595112414838,
      "loss": 5.7236,
      "mean_token_accuracy": 0.17752385437488555,
      "num_tokens": 44596753.0,
      "step": 20610
    },
    {
      "entropy": 6.523242950439453,
      "epoch": 2.2060035314891113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004518358738674073,
      "loss": 5.9738,
      "mean_token_accuracy": 0.15560027509927749,
      "num_tokens": 44607632.0,
      "step": 20615
    },
    {
      "entropy": 6.508864831924439,
      "epoch": 2.206538605596875,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004518122313872198,
      "loss": 5.8599,
      "mean_token_accuracy": 0.16673669964075089,
      "num_tokens": 44618786.0,
      "step": 20620
    },
    {
      "entropy": 6.456614637374878,
      "epoch": 2.207073679704639,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045178858380160396,
      "loss": 5.845,
      "mean_token_accuracy": 0.1629694566130638,
      "num_tokens": 44628760.0,
      "step": 20625
    },
    {
      "entropy": 6.538754415512085,
      "epoch": 2.207608753812403,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004517649311112426,
      "loss": 5.9085,
      "mean_token_accuracy": 0.16276949346065522,
      "num_tokens": 44641206.0,
      "step": 20630
    },
    {
      "entropy": 6.524659013748169,
      "epoch": 2.208143827920167,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004517412733168187,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16688226014375687,
      "num_tokens": 44651923.0,
      "step": 20635
    },
    {
      "entropy": 6.4422706127166744,
      "epoch": 2.208678902027931,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045171761041901533,
      "loss": 5.8266,
      "mean_token_accuracy": 0.1642281726002693,
      "num_tokens": 44662375.0,
      "step": 20640
    },
    {
      "entropy": 6.332964611053467,
      "epoch": 2.2092139761356946,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004516939424185158,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17337203472852708,
      "num_tokens": 44673201.0,
      "step": 20645
    },
    {
      "entropy": 6.496636390686035,
      "epoch": 2.2097490502434587,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004516702693160035,
      "loss": 5.8515,
      "mean_token_accuracy": 0.15964802950620652,
      "num_tokens": 44684300.0,
      "step": 20650
    },
    {
      "entropy": 6.5248558044433596,
      "epoch": 2.210284124351223,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004516465911121621,
      "loss": 5.896,
      "mean_token_accuracy": 0.16303883194923402,
      "num_tokens": 44694603.0,
      "step": 20655
    },
    {
      "entropy": 6.456915712356567,
      "epoch": 2.2108191984589864,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004516229078076751,
      "loss": 5.874,
      "mean_token_accuracy": 0.16800688803195954,
      "num_tokens": 44704461.0,
      "step": 20660
    },
    {
      "entropy": 6.488613796234131,
      "epoch": 2.2113542725667505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004515992194032265,
      "loss": 5.9155,
      "mean_token_accuracy": 0.16932720839977264,
      "num_tokens": 44715784.0,
      "step": 20665
    },
    {
      "entropy": 6.479518985748291,
      "epoch": 2.211889346674514,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045157552589950023,
      "loss": 5.7952,
      "mean_token_accuracy": 0.17483361065387726,
      "num_tokens": 44726025.0,
      "step": 20670
    },
    {
      "entropy": 6.406489610671997,
      "epoch": 2.2124244207822783,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004515518272971805,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1794299840927124,
      "num_tokens": 44736602.0,
      "step": 20675
    },
    {
      "entropy": 6.486744213104248,
      "epoch": 2.2129594948900424,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004515281235969515,
      "loss": 5.9615,
      "mean_token_accuracy": 0.1558422178030014,
      "num_tokens": 44746260.0,
      "step": 20680
    },
    {
      "entropy": 6.47368106842041,
      "epoch": 2.213494568997806,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004515044147994978,
      "loss": 5.8412,
      "mean_token_accuracy": 0.1679094836115837,
      "num_tokens": 44756894.0,
      "step": 20685
    },
    {
      "entropy": 6.486376333236694,
      "epoch": 2.21402964310557,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004514807009055039,
      "loss": 5.8988,
      "mean_token_accuracy": 0.1609685629606247,
      "num_tokens": 44768483.0,
      "step": 20690
    },
    {
      "entropy": 6.459349536895752,
      "epoch": 2.2145647172133343,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045145698191565456,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16516641080379485,
      "num_tokens": 44778843.0,
      "step": 20695
    },
    {
      "entropy": 6.482363557815551,
      "epoch": 2.215099791321098,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045143325783063466,
      "loss": 5.8037,
      "mean_token_accuracy": 0.1701285108923912,
      "num_tokens": 44788659.0,
      "step": 20700
    },
    {
      "entropy": 6.490698719024659,
      "epoch": 2.215634865428862,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004514095286511293,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16360589116811752,
      "num_tokens": 44799730.0,
      "step": 20705
    },
    {
      "entropy": 6.4248593807220455,
      "epoch": 2.2161699395366257,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004513857943778235,
      "loss": 5.8452,
      "mean_token_accuracy": 0.172635617852211,
      "num_tokens": 44811679.0,
      "step": 20710
    },
    {
      "entropy": 6.4315417289733885,
      "epoch": 2.2167050136443898,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004513620550114027,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16716813147068024,
      "num_tokens": 44821996.0,
      "step": 20715
    },
    {
      "entropy": 6.511249208450318,
      "epoch": 2.217240087752154,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00045133831055255234,
      "loss": 5.9172,
      "mean_token_accuracy": 0.16535038948059083,
      "num_tokens": 44833713.0,
      "step": 20720
    },
    {
      "entropy": 6.514973592758179,
      "epoch": 2.2177751618599175,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004513145610019581,
      "loss": 5.9042,
      "mean_token_accuracy": 0.1659704327583313,
      "num_tokens": 44844751.0,
      "step": 20725
    },
    {
      "entropy": 6.5036561489105225,
      "epoch": 2.2183102359676816,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045129080636030554,
      "loss": 5.9228,
      "mean_token_accuracy": 0.1612008646130562,
      "num_tokens": 44856089.0,
      "step": 20730
    },
    {
      "entropy": 6.471727132797241,
      "epoch": 2.2188453100754453,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004512670466282808,
      "loss": 5.8772,
      "mean_token_accuracy": 0.1636594295501709,
      "num_tokens": 44867313.0,
      "step": 20735
    },
    {
      "entropy": 6.50086464881897,
      "epoch": 2.2193803841832094,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045124328180656993,
      "loss": 5.8661,
      "mean_token_accuracy": 0.16183009892702102,
      "num_tokens": 44878430.0,
      "step": 20740
    },
    {
      "entropy": 6.4665018081665036,
      "epoch": 2.2199154582909735,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045121951189585897,
      "loss": 5.8029,
      "mean_token_accuracy": 0.17250235378742218,
      "num_tokens": 44888746.0,
      "step": 20745
    },
    {
      "entropy": 6.503683948516846,
      "epoch": 2.220450532398737,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045119573689683445,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1603775292634964,
      "num_tokens": 44899412.0,
      "step": 20750
    },
    {
      "entropy": 6.47873330116272,
      "epoch": 2.2209856065065012,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045117195681018273,
      "loss": 5.8471,
      "mean_token_accuracy": 0.17003336697816848,
      "num_tokens": 44910362.0,
      "step": 20755
    },
    {
      "entropy": 6.46144061088562,
      "epoch": 2.221520680614265,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004511481716365905,
      "loss": 5.8229,
      "mean_token_accuracy": 0.16260856240987778,
      "num_tokens": 44920552.0,
      "step": 20760
    },
    {
      "entropy": 6.439069175720215,
      "epoch": 2.222055754722029,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004511243813767446,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16264118254184723,
      "num_tokens": 44932090.0,
      "step": 20765
    },
    {
      "entropy": 6.5542564868927,
      "epoch": 2.222590828829793,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000451100586031332,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16415328532457352,
      "num_tokens": 44944122.0,
      "step": 20770
    },
    {
      "entropy": 6.529652547836304,
      "epoch": 2.2231259029375567,
      "grad_norm": 0.953125,
      "learning_rate": 0.00045107678560103953,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16434241235256195,
      "num_tokens": 44955104.0,
      "step": 20775
    },
    {
      "entropy": 6.4908332347869875,
      "epoch": 2.223660977045321,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00045105298008655485,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16115287244319915,
      "num_tokens": 44965495.0,
      "step": 20780
    },
    {
      "entropy": 6.471954202651977,
      "epoch": 2.2241960511530845,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045102916948856493,
      "loss": 5.9141,
      "mean_token_accuracy": 0.15933675467967987,
      "num_tokens": 44976627.0,
      "step": 20785
    },
    {
      "entropy": 6.527413940429687,
      "epoch": 2.2247311252608486,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004510053538077576,
      "loss": 5.9312,
      "mean_token_accuracy": 0.16076175272464752,
      "num_tokens": 44987955.0,
      "step": 20790
    },
    {
      "entropy": 6.534368562698364,
      "epoch": 2.2252661993686127,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004509815330448203,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16706618964672088,
      "num_tokens": 44998124.0,
      "step": 20795
    },
    {
      "entropy": 6.512688016891479,
      "epoch": 2.2258012734763764,
      "grad_norm": 1.015625,
      "learning_rate": 0.000450957707200441,
      "loss": 5.9265,
      "mean_token_accuracy": 0.15971988588571548,
      "num_tokens": 45007881.0,
      "step": 20800
    },
    {
      "entropy": 6.491861200332641,
      "epoch": 2.2263363475841405,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004509338762753076,
      "loss": 5.8533,
      "mean_token_accuracy": 0.16691645234823227,
      "num_tokens": 45018994.0,
      "step": 20805
    },
    {
      "entropy": 6.528700542449951,
      "epoch": 2.2268714216919046,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004509100402701083,
      "loss": 5.9059,
      "mean_token_accuracy": 0.16486834287643432,
      "num_tokens": 45030109.0,
      "step": 20810
    },
    {
      "entropy": 6.4071653366088865,
      "epoch": 2.227406495799668,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045088619918553136,
      "loss": 5.7962,
      "mean_token_accuracy": 0.1827381283044815,
      "num_tokens": 45039872.0,
      "step": 20815
    },
    {
      "entropy": 6.5377555847167965,
      "epoch": 2.2279415699074323,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004508623530222651,
      "loss": 5.8067,
      "mean_token_accuracy": 0.17789260298013687,
      "num_tokens": 45049216.0,
      "step": 20820
    },
    {
      "entropy": 6.437960004806518,
      "epoch": 2.228476644015196,
      "grad_norm": 1.046875,
      "learning_rate": 0.00045083850178099805,
      "loss": 5.7827,
      "mean_token_accuracy": 0.1769940137863159,
      "num_tokens": 45059191.0,
      "step": 20825
    },
    {
      "entropy": 6.422557830810547,
      "epoch": 2.22901171812296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000450814645462419,
      "loss": 5.824,
      "mean_token_accuracy": 0.16965683102607726,
      "num_tokens": 45069212.0,
      "step": 20830
    },
    {
      "entropy": 6.485573434829712,
      "epoch": 2.229546792230724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045079078406721676,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16728191822767258,
      "num_tokens": 45080248.0,
      "step": 20835
    },
    {
      "entropy": 6.4722065925598145,
      "epoch": 2.230081866338488,
      "grad_norm": 1.015625,
      "learning_rate": 0.00045076691759608035,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16131263524293898,
      "num_tokens": 45091315.0,
      "step": 20840
    },
    {
      "entropy": 6.466313314437866,
      "epoch": 2.230616940446252,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045074304604969886,
      "loss": 5.8475,
      "mean_token_accuracy": 0.1649936705827713,
      "num_tokens": 45102674.0,
      "step": 20845
    },
    {
      "entropy": 6.442714595794678,
      "epoch": 2.2311520145540156,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045071916942876163,
      "loss": 5.8402,
      "mean_token_accuracy": 0.16662923693656922,
      "num_tokens": 45113171.0,
      "step": 20850
    },
    {
      "entropy": 6.473045110702515,
      "epoch": 2.2316870886617797,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004506952877339581,
      "loss": 5.8288,
      "mean_token_accuracy": 0.1669688791036606,
      "num_tokens": 45122912.0,
      "step": 20855
    },
    {
      "entropy": 6.511230897903443,
      "epoch": 2.2322221627695438,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004506714009659778,
      "loss": 5.8575,
      "mean_token_accuracy": 0.1702345058321953,
      "num_tokens": 45133839.0,
      "step": 20860
    },
    {
      "entropy": 6.488212299346924,
      "epoch": 2.2327572368773074,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004506475091255105,
      "loss": 5.8524,
      "mean_token_accuracy": 0.16949136704206466,
      "num_tokens": 45144914.0,
      "step": 20865
    },
    {
      "entropy": 6.478864049911499,
      "epoch": 2.2332923109850715,
      "grad_norm": 1.078125,
      "learning_rate": 0.000450623612213246,
      "loss": 5.8511,
      "mean_token_accuracy": 0.16739283353090287,
      "num_tokens": 45155194.0,
      "step": 20870
    },
    {
      "entropy": 6.504784727096558,
      "epoch": 2.233827385092835,
      "grad_norm": 1.125,
      "learning_rate": 0.00045059971022987446,
      "loss": 5.9585,
      "mean_token_accuracy": 0.15813899636268616,
      "num_tokens": 45166771.0,
      "step": 20875
    },
    {
      "entropy": 6.52585015296936,
      "epoch": 2.2343624592005993,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045057580317608585,
      "loss": 5.8103,
      "mean_token_accuracy": 0.16512493938207626,
      "num_tokens": 45177236.0,
      "step": 20880
    },
    {
      "entropy": 6.512587261199951,
      "epoch": 2.2348975333083634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045055189105257064,
      "loss": 5.8385,
      "mean_token_accuracy": 0.162494558095932,
      "num_tokens": 45187382.0,
      "step": 20885
    },
    {
      "entropy": 6.490521812438965,
      "epoch": 2.235432607416127,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045052797386001936,
      "loss": 5.8659,
      "mean_token_accuracy": 0.1681062698364258,
      "num_tokens": 45197287.0,
      "step": 20890
    },
    {
      "entropy": 6.548718452453613,
      "epoch": 2.235967681523891,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004505040515991224,
      "loss": 5.8288,
      "mean_token_accuracy": 0.16956937611103057,
      "num_tokens": 45208737.0,
      "step": 20895
    },
    {
      "entropy": 6.539750242233277,
      "epoch": 2.236502755631655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045048012427057057,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15730833560228347,
      "num_tokens": 45219586.0,
      "step": 20900
    },
    {
      "entropy": 6.544824552536011,
      "epoch": 2.237037829739419,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004504561918750549,
      "loss": 5.9243,
      "mean_token_accuracy": 0.15545279532670975,
      "num_tokens": 45230730.0,
      "step": 20905
    },
    {
      "entropy": 6.4899688243865965,
      "epoch": 2.237572903847183,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004504322544132663,
      "loss": 5.7754,
      "mean_token_accuracy": 0.18073288947343827,
      "num_tokens": 45241330.0,
      "step": 20910
    },
    {
      "entropy": 6.423538637161255,
      "epoch": 2.2381079779549466,
      "grad_norm": 1.109375,
      "learning_rate": 0.000450408311885896,
      "loss": 5.8659,
      "mean_token_accuracy": 0.16664992421865463,
      "num_tokens": 45251439.0,
      "step": 20915
    },
    {
      "entropy": 6.421528720855713,
      "epoch": 2.2386430520627107,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045038436429363537,
      "loss": 5.6709,
      "mean_token_accuracy": 0.18157233893871308,
      "num_tokens": 45262078.0,
      "step": 20920
    },
    {
      "entropy": 6.4975049018859865,
      "epoch": 2.2391781261704744,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045036041163717586,
      "loss": 5.8919,
      "mean_token_accuracy": 0.16317496746778487,
      "num_tokens": 45272466.0,
      "step": 20925
    },
    {
      "entropy": 6.49933819770813,
      "epoch": 2.2397132002782385,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004503364539172091,
      "loss": 5.8085,
      "mean_token_accuracy": 0.16944617927074432,
      "num_tokens": 45281830.0,
      "step": 20930
    },
    {
      "entropy": 6.464972734451294,
      "epoch": 2.2402482743860026,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004503124911344268,
      "loss": 5.8842,
      "mean_token_accuracy": 0.15397200956940651,
      "num_tokens": 45293026.0,
      "step": 20935
    },
    {
      "entropy": 6.4903429508209225,
      "epoch": 2.2407833484937663,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00045028852328952106,
      "loss": 5.848,
      "mean_token_accuracy": 0.1651660293340683,
      "num_tokens": 45303532.0,
      "step": 20940
    },
    {
      "entropy": 6.3957594871521,
      "epoch": 2.2413184226015304,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004502645503831838,
      "loss": 5.7297,
      "mean_token_accuracy": 0.17807627022266387,
      "num_tokens": 45314448.0,
      "step": 20945
    },
    {
      "entropy": 6.471572589874268,
      "epoch": 2.241853496709294,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045024057241610727,
      "loss": 5.8916,
      "mean_token_accuracy": 0.15321228206157683,
      "num_tokens": 45323194.0,
      "step": 20950
    },
    {
      "entropy": 6.465305757522583,
      "epoch": 2.242388570817058,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004502165893889839,
      "loss": 5.7618,
      "mean_token_accuracy": 0.18128467947244645,
      "num_tokens": 45333464.0,
      "step": 20955
    },
    {
      "entropy": 6.450857019424438,
      "epoch": 2.242923644924822,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004501926013025061,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16979770511388778,
      "num_tokens": 45344364.0,
      "step": 20960
    },
    {
      "entropy": 6.469534921646118,
      "epoch": 2.243458719032586,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00045016860815736646,
      "loss": 5.8132,
      "mean_token_accuracy": 0.1685362532734871,
      "num_tokens": 45355202.0,
      "step": 20965
    },
    {
      "entropy": 6.4369956970214846,
      "epoch": 2.24399379314035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004501446099542579,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16932329833507537,
      "num_tokens": 45365696.0,
      "step": 20970
    },
    {
      "entropy": 6.415595388412475,
      "epoch": 2.244528867248114,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004501206066938734,
      "loss": 5.7679,
      "mean_token_accuracy": 0.1761318027973175,
      "num_tokens": 45376609.0,
      "step": 20975
    },
    {
      "entropy": 6.5396120071411135,
      "epoch": 2.2450639413558777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045009659837690606,
      "loss": 5.9259,
      "mean_token_accuracy": 0.15419590771198272,
      "num_tokens": 45386158.0,
      "step": 20980
    },
    {
      "entropy": 6.570333051681518,
      "epoch": 2.245599015463642,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000450072585004049,
      "loss": 5.92,
      "mean_token_accuracy": 0.1637445107102394,
      "num_tokens": 45396824.0,
      "step": 20985
    },
    {
      "entropy": 6.524670171737671,
      "epoch": 2.2461340895714055,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004500485665759955,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16473570913076402,
      "num_tokens": 45407219.0,
      "step": 20990
    },
    {
      "entropy": 6.471851015090943,
      "epoch": 2.2466691636791696,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004500245430934394,
      "loss": 5.8744,
      "mean_token_accuracy": 0.1578470140695572,
      "num_tokens": 45418572.0,
      "step": 20995
    },
    {
      "entropy": 6.4355127811431885,
      "epoch": 2.2472042377869337,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045000051455707413,
      "loss": 5.7981,
      "mean_token_accuracy": 0.1677699565887451,
      "num_tokens": 45429643.0,
      "step": 21000
    },
    {
      "epoch": 2.2472042377869337,
      "eval_entropy": 6.256859263062799,
      "eval_loss": 5.970716953277588,
      "eval_mean_token_accuracy": 0.16837227823378423,
      "eval_num_tokens": 45429643.0,
      "eval_runtime": 17.413,
      "eval_samples_per_second": 1704.421,
      "eval_steps_per_second": 213.06,
      "step": 21000
    },
    {
      "entropy": 6.510654640197754,
      "epoch": 2.2477393118946973,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004499764809675937,
      "loss": 5.8317,
      "mean_token_accuracy": 0.16893653571605682,
      "num_tokens": 45440183.0,
      "step": 21005
    },
    {
      "entropy": 6.444044399261474,
      "epoch": 2.2482743860024614,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004499524423256918,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1672471582889557,
      "num_tokens": 45451252.0,
      "step": 21010
    },
    {
      "entropy": 6.426980638504029,
      "epoch": 2.248809460110225,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044992839863206284,
      "loss": 5.884,
      "mean_token_accuracy": 0.1625429093837738,
      "num_tokens": 45462336.0,
      "step": 21015
    },
    {
      "entropy": 6.465860986709595,
      "epoch": 2.249344534217989,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044990434988740095,
      "loss": 5.8581,
      "mean_token_accuracy": 0.16377479881048201,
      "num_tokens": 45472526.0,
      "step": 21020
    },
    {
      "entropy": 6.523887300491333,
      "epoch": 2.2498796083257533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004498802960924005,
      "loss": 5.9149,
      "mean_token_accuracy": 0.15668225288391113,
      "num_tokens": 45483281.0,
      "step": 21025
    },
    {
      "entropy": 6.497504806518554,
      "epoch": 2.250414682433517,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004498562372477561,
      "loss": 5.8243,
      "mean_token_accuracy": 0.1705967977643013,
      "num_tokens": 45494258.0,
      "step": 21030
    },
    {
      "entropy": 6.508712100982666,
      "epoch": 2.250949756541281,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004498321733541624,
      "loss": 5.867,
      "mean_token_accuracy": 0.16417690068483354,
      "num_tokens": 45504233.0,
      "step": 21035
    },
    {
      "entropy": 6.454278993606567,
      "epoch": 2.2514848306490447,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044980810441231424,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16491375118494034,
      "num_tokens": 45513902.0,
      "step": 21040
    },
    {
      "entropy": 6.483533763885498,
      "epoch": 2.252019904756809,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004497840304229066,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16572380363941192,
      "num_tokens": 45524111.0,
      "step": 21045
    },
    {
      "entropy": 6.465489673614502,
      "epoch": 2.252554978864573,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044975995138663466,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16826518923044204,
      "num_tokens": 45534816.0,
      "step": 21050
    },
    {
      "entropy": 6.505864524841309,
      "epoch": 2.2530900529723366,
      "grad_norm": 1.0,
      "learning_rate": 0.00044973586730419373,
      "loss": 5.8297,
      "mean_token_accuracy": 0.17163622081279756,
      "num_tokens": 45546012.0,
      "step": 21055
    },
    {
      "entropy": 6.431884241104126,
      "epoch": 2.2536251270801007,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004497117781762792,
      "loss": 5.8385,
      "mean_token_accuracy": 0.169489885866642,
      "num_tokens": 45555402.0,
      "step": 21060
    },
    {
      "entropy": 6.4304986000061035,
      "epoch": 2.2541602011878643,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004496876840035866,
      "loss": 5.8736,
      "mean_token_accuracy": 0.16070492267608644,
      "num_tokens": 45565537.0,
      "step": 21065
    },
    {
      "entropy": 6.512145709991455,
      "epoch": 2.2546952752956284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044966358478681163,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16737861931324005,
      "num_tokens": 45576449.0,
      "step": 21070
    },
    {
      "entropy": 6.604936122894287,
      "epoch": 2.2552303494033925,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004496394805266502,
      "loss": 5.882,
      "mean_token_accuracy": 0.15773238241672516,
      "num_tokens": 45586477.0,
      "step": 21075
    },
    {
      "entropy": 6.442500352859497,
      "epoch": 2.255765423511156,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044961537122379844,
      "loss": 5.8636,
      "mean_token_accuracy": 0.1642209395766258,
      "num_tokens": 45597279.0,
      "step": 21080
    },
    {
      "entropy": 6.44842562675476,
      "epoch": 2.2563004976189203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004495912568789522,
      "loss": 5.9032,
      "mean_token_accuracy": 0.1619752287864685,
      "num_tokens": 45608216.0,
      "step": 21085
    },
    {
      "entropy": 6.466652917861938,
      "epoch": 2.2568355717266844,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044956713749280805,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16703361570835112,
      "num_tokens": 45620665.0,
      "step": 21090
    },
    {
      "entropy": 6.508062744140625,
      "epoch": 2.257370645834448,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044954301306606234,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16586200445890426,
      "num_tokens": 45631635.0,
      "step": 21095
    },
    {
      "entropy": 6.381379747390747,
      "epoch": 2.257905719942212,
      "grad_norm": 1.0,
      "learning_rate": 0.00044951888359941156,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17220815867185593,
      "num_tokens": 45642507.0,
      "step": 21100
    },
    {
      "entropy": 6.469487428665161,
      "epoch": 2.2584407940499758,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044949474909355264,
      "loss": 5.8582,
      "mean_token_accuracy": 0.16551116555929185,
      "num_tokens": 45653123.0,
      "step": 21105
    },
    {
      "entropy": 6.495351552963257,
      "epoch": 2.25897586815774,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044947060954918233,
      "loss": 5.8442,
      "mean_token_accuracy": 0.17096397429704666,
      "num_tokens": 45663228.0,
      "step": 21110
    },
    {
      "entropy": 6.485017871856689,
      "epoch": 2.2595109422655035,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004494464649669976,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16830089539289475,
      "num_tokens": 45673389.0,
      "step": 21115
    },
    {
      "entropy": 6.483903408050537,
      "epoch": 2.2600460163732676,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044942231534769587,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16321436762809755,
      "num_tokens": 45684803.0,
      "step": 21120
    },
    {
      "entropy": 6.478506660461425,
      "epoch": 2.2605810904810317,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044939816069197415,
      "loss": 5.9139,
      "mean_token_accuracy": 0.1636238619685173,
      "num_tokens": 45695164.0,
      "step": 21125
    },
    {
      "entropy": 6.493110513687133,
      "epoch": 2.2611161645887954,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044937400100053017,
      "loss": 5.8976,
      "mean_token_accuracy": 0.16862733364105226,
      "num_tokens": 45706169.0,
      "step": 21130
    },
    {
      "entropy": 6.521122932434082,
      "epoch": 2.2616512386965595,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004493498362740613,
      "loss": 5.9618,
      "mean_token_accuracy": 0.15492086559534074,
      "num_tokens": 45717299.0,
      "step": 21135
    },
    {
      "entropy": 6.422291421890259,
      "epoch": 2.2621863128043236,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044932566651326546,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16635026037693024,
      "num_tokens": 45728815.0,
      "step": 21140
    },
    {
      "entropy": 6.4986124515533445,
      "epoch": 2.2627213869120872,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004493014917188405,
      "loss": 5.8879,
      "mean_token_accuracy": 0.16745662242174147,
      "num_tokens": 45740293.0,
      "step": 21145
    },
    {
      "entropy": 6.527046585083008,
      "epoch": 2.2632564610198513,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044927731189148447,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16667712926864625,
      "num_tokens": 45751243.0,
      "step": 21150
    },
    {
      "entropy": 6.395951414108277,
      "epoch": 2.263791535127615,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044925312703189546,
      "loss": 5.815,
      "mean_token_accuracy": 0.16917366832494735,
      "num_tokens": 45762477.0,
      "step": 21155
    },
    {
      "entropy": 6.4934361457824705,
      "epoch": 2.264326609235379,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000449228937140772,
      "loss": 5.846,
      "mean_token_accuracy": 0.16823104918003082,
      "num_tokens": 45772929.0,
      "step": 21160
    },
    {
      "entropy": 6.473659324645996,
      "epoch": 2.264861683343143,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044920474221881243,
      "loss": 5.8616,
      "mean_token_accuracy": 0.16943355053663253,
      "num_tokens": 45782999.0,
      "step": 21165
    },
    {
      "entropy": 6.425304174423218,
      "epoch": 2.265396757450907,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044918054226671545,
      "loss": 5.8533,
      "mean_token_accuracy": 0.17226449251174927,
      "num_tokens": 45793020.0,
      "step": 21170
    },
    {
      "entropy": 6.4763346195220945,
      "epoch": 2.265931831558671,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044915633728517975,
      "loss": 5.8064,
      "mean_token_accuracy": 0.17684617042541503,
      "num_tokens": 45803020.0,
      "step": 21175
    },
    {
      "entropy": 6.472933626174926,
      "epoch": 2.2664669056664346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004491321272749042,
      "loss": 5.8835,
      "mean_token_accuracy": 0.17065804153680803,
      "num_tokens": 45812700.0,
      "step": 21180
    },
    {
      "entropy": 6.4705041408538815,
      "epoch": 2.2670019797741987,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044910791223658805,
      "loss": 5.9099,
      "mean_token_accuracy": 0.1581263482570648,
      "num_tokens": 45823097.0,
      "step": 21185
    },
    {
      "entropy": 6.475236701965332,
      "epoch": 2.267537053881963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044908369217093035,
      "loss": 5.8356,
      "mean_token_accuracy": 0.1611158937215805,
      "num_tokens": 45832935.0,
      "step": 21190
    },
    {
      "entropy": 6.520775604248047,
      "epoch": 2.2680721279897265,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044905946707863057,
      "loss": 5.9477,
      "mean_token_accuracy": 0.15655047595500945,
      "num_tokens": 45844296.0,
      "step": 21195
    },
    {
      "entropy": 6.501375722885132,
      "epoch": 2.2686072020974906,
      "grad_norm": 1.0,
      "learning_rate": 0.0004490352369603881,
      "loss": 5.864,
      "mean_token_accuracy": 0.1613788679242134,
      "num_tokens": 45854942.0,
      "step": 21200
    },
    {
      "entropy": 6.483459758758545,
      "epoch": 2.2691422762052547,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044901100181690256,
      "loss": 5.9213,
      "mean_token_accuracy": 0.16506835222244262,
      "num_tokens": 45866427.0,
      "step": 21205
    },
    {
      "entropy": 6.468076515197754,
      "epoch": 2.2696773503130183,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044898676164887387,
      "loss": 5.894,
      "mean_token_accuracy": 0.16583137661218644,
      "num_tokens": 45877955.0,
      "step": 21210
    },
    {
      "entropy": 6.405737209320068,
      "epoch": 2.2702124244207824,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044896251645700185,
      "loss": 5.76,
      "mean_token_accuracy": 0.17447579503059388,
      "num_tokens": 45888668.0,
      "step": 21215
    },
    {
      "entropy": 6.462586688995361,
      "epoch": 2.270747498528546,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004489382662419866,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16536019295454024,
      "num_tokens": 45898819.0,
      "step": 21220
    },
    {
      "entropy": 6.492992305755616,
      "epoch": 2.27128257263631,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004489140110045283,
      "loss": 5.8988,
      "mean_token_accuracy": 0.16567114889621734,
      "num_tokens": 45909390.0,
      "step": 21225
    },
    {
      "entropy": 6.481360960006714,
      "epoch": 2.271817646744074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004488897507453275,
      "loss": 5.8374,
      "mean_token_accuracy": 0.16485968828201295,
      "num_tokens": 45919697.0,
      "step": 21230
    },
    {
      "entropy": 6.520775318145752,
      "epoch": 2.272352720851838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044886548546508453,
      "loss": 5.9312,
      "mean_token_accuracy": 0.16901127398014068,
      "num_tokens": 45929458.0,
      "step": 21235
    },
    {
      "entropy": 6.429954290390015,
      "epoch": 2.272887794959602,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004488412151645001,
      "loss": 5.7827,
      "mean_token_accuracy": 0.17274929732084274,
      "num_tokens": 45939373.0,
      "step": 21240
    },
    {
      "entropy": 6.414415407180786,
      "epoch": 2.2734228690673657,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044881693984427504,
      "loss": 5.8615,
      "mean_token_accuracy": 0.1649525061249733,
      "num_tokens": 45949954.0,
      "step": 21245
    },
    {
      "entropy": 6.500512170791626,
      "epoch": 2.27395794317513,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004487926595051102,
      "loss": 5.901,
      "mean_token_accuracy": 0.16436106860637664,
      "num_tokens": 45960147.0,
      "step": 21250
    },
    {
      "entropy": 6.490777540206909,
      "epoch": 2.274493017282894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004487683741477068,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16781248897314072,
      "num_tokens": 45970718.0,
      "step": 21255
    },
    {
      "entropy": 6.488932275772095,
      "epoch": 2.2750280913906575,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000448744083772766,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16861318200826644,
      "num_tokens": 45981631.0,
      "step": 21260
    },
    {
      "entropy": 6.516122102737427,
      "epoch": 2.2755631654984216,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044871978838098917,
      "loss": 5.9096,
      "mean_token_accuracy": 0.16141709834337234,
      "num_tokens": 45992295.0,
      "step": 21265
    },
    {
      "entropy": 6.5384105205535885,
      "epoch": 2.2760982396061853,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004486954879730778,
      "loss": 5.9029,
      "mean_token_accuracy": 0.15982945412397384,
      "num_tokens": 46002222.0,
      "step": 21270
    },
    {
      "entropy": 6.4364406108856205,
      "epoch": 2.2766333137139494,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004486711825497337,
      "loss": 5.7369,
      "mean_token_accuracy": 0.18445081263780594,
      "num_tokens": 46012393.0,
      "step": 21275
    },
    {
      "entropy": 6.453106689453125,
      "epoch": 2.2771683878217135,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044864687211165863,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16758042573928833,
      "num_tokens": 46022095.0,
      "step": 21280
    },
    {
      "entropy": 6.461852312088013,
      "epoch": 2.277703461929477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044862255665955447,
      "loss": 5.8455,
      "mean_token_accuracy": 0.1621552973985672,
      "num_tokens": 46032973.0,
      "step": 21285
    },
    {
      "entropy": 6.504717397689819,
      "epoch": 2.2782385360372412,
      "grad_norm": 1.125,
      "learning_rate": 0.0004485982361941234,
      "loss": 5.8309,
      "mean_token_accuracy": 0.1709066554903984,
      "num_tokens": 46043430.0,
      "step": 21290
    },
    {
      "entropy": 6.450718593597412,
      "epoch": 2.278773610145005,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004485739107160677,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16860976368188857,
      "num_tokens": 46054580.0,
      "step": 21295
    },
    {
      "entropy": 6.466533184051514,
      "epoch": 2.279308684252769,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00044854958022608974,
      "loss": 5.8453,
      "mean_token_accuracy": 0.17548623085021972,
      "num_tokens": 46064898.0,
      "step": 21300
    },
    {
      "entropy": 6.445337390899658,
      "epoch": 2.279843758360533,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044852524472489196,
      "loss": 5.8251,
      "mean_token_accuracy": 0.17019203603267669,
      "num_tokens": 46075985.0,
      "step": 21305
    },
    {
      "entropy": 6.4809667587280275,
      "epoch": 2.2803788324682968,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004485009042131771,
      "loss": 5.9188,
      "mean_token_accuracy": 0.16150370687246324,
      "num_tokens": 46087544.0,
      "step": 21310
    },
    {
      "entropy": 6.518268060684204,
      "epoch": 2.280913906576061,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044847655869164813,
      "loss": 5.8538,
      "mean_token_accuracy": 0.15945089906454085,
      "num_tokens": 46097613.0,
      "step": 21315
    },
    {
      "entropy": 6.414216136932373,
      "epoch": 2.281448980683825,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004484522081610078,
      "loss": 5.776,
      "mean_token_accuracy": 0.1822814628481865,
      "num_tokens": 46107728.0,
      "step": 21320
    },
    {
      "entropy": 6.4331718444824215,
      "epoch": 2.2819840547915886,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044842785262195936,
      "loss": 5.8068,
      "mean_token_accuracy": 0.16877790689468383,
      "num_tokens": 46118309.0,
      "step": 21325
    },
    {
      "entropy": 6.4482639789581295,
      "epoch": 2.2825191288993527,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044840349207520604,
      "loss": 5.8503,
      "mean_token_accuracy": 0.1716996267437935,
      "num_tokens": 46128425.0,
      "step": 21330
    },
    {
      "entropy": 6.561215400695801,
      "epoch": 2.2830542030071164,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004483791265214513,
      "loss": 5.9398,
      "mean_token_accuracy": 0.16440824121236802,
      "num_tokens": 46139506.0,
      "step": 21335
    },
    {
      "entropy": 6.501630353927612,
      "epoch": 2.2835892771148805,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044835475596139855,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16622481495141983,
      "num_tokens": 46151226.0,
      "step": 21340
    },
    {
      "entropy": 6.510273838043213,
      "epoch": 2.284124351222644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004483303803957516,
      "loss": 5.8854,
      "mean_token_accuracy": 0.1661148875951767,
      "num_tokens": 46162749.0,
      "step": 21345
    },
    {
      "entropy": 6.485588455200196,
      "epoch": 2.2846594253304082,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004483059998252144,
      "loss": 5.884,
      "mean_token_accuracy": 0.16539423912763596,
      "num_tokens": 46174802.0,
      "step": 21350
    },
    {
      "entropy": 6.491522264480591,
      "epoch": 2.2851944994381723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044828161425049067,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16875047087669373,
      "num_tokens": 46185579.0,
      "step": 21355
    },
    {
      "entropy": 6.489670896530152,
      "epoch": 2.285729573545936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044825722367228477,
      "loss": 5.9655,
      "mean_token_accuracy": 0.15513585060834884,
      "num_tokens": 46196399.0,
      "step": 21360
    },
    {
      "entropy": 6.532983875274658,
      "epoch": 2.2862646476537,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044823282809130084,
      "loss": 5.8745,
      "mean_token_accuracy": 0.15694854706525802,
      "num_tokens": 46206931.0,
      "step": 21365
    },
    {
      "entropy": 6.499423980712891,
      "epoch": 2.286799721761464,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044820842750824334,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16731313914060592,
      "num_tokens": 46217327.0,
      "step": 21370
    },
    {
      "entropy": 6.492517948150635,
      "epoch": 2.287334795869228,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00044818402192381684,
      "loss": 5.8553,
      "mean_token_accuracy": 0.16869425326585769,
      "num_tokens": 46227981.0,
      "step": 21375
    },
    {
      "entropy": 6.482279682159424,
      "epoch": 2.287869869976992,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044815961133872606,
      "loss": 5.9174,
      "mean_token_accuracy": 0.1569733127951622,
      "num_tokens": 46238500.0,
      "step": 21380
    },
    {
      "entropy": 6.4291346073150635,
      "epoch": 2.2884049440847556,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044813519575367586,
      "loss": 5.7943,
      "mean_token_accuracy": 0.17415079176425935,
      "num_tokens": 46249530.0,
      "step": 21385
    },
    {
      "entropy": 6.509825372695923,
      "epoch": 2.2889400181925197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004481107751693711,
      "loss": 5.7444,
      "mean_token_accuracy": 0.17290020138025283,
      "num_tokens": 46260272.0,
      "step": 21390
    },
    {
      "entropy": 6.4266492366790775,
      "epoch": 2.2894750923002833,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004480863495865171,
      "loss": 5.8293,
      "mean_token_accuracy": 0.16299197524785997,
      "num_tokens": 46271805.0,
      "step": 21395
    },
    {
      "entropy": 6.468395805358886,
      "epoch": 2.2900101664080474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004480619190058191,
      "loss": 5.8786,
      "mean_token_accuracy": 0.167424838244915,
      "num_tokens": 46282211.0,
      "step": 21400
    },
    {
      "entropy": 6.441889810562134,
      "epoch": 2.2905452405158115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044803748342798247,
      "loss": 5.7896,
      "mean_token_accuracy": 0.1692647933959961,
      "num_tokens": 46292287.0,
      "step": 21405
    },
    {
      "entropy": 6.469284391403198,
      "epoch": 2.291080314623575,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044801304285371286,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16234617978334426,
      "num_tokens": 46302978.0,
      "step": 21410
    },
    {
      "entropy": 6.476571941375733,
      "epoch": 2.2916153887313393,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004479885972837159,
      "loss": 5.8969,
      "mean_token_accuracy": 0.1646575391292572,
      "num_tokens": 46313958.0,
      "step": 21415
    },
    {
      "entropy": 6.575262022018433,
      "epoch": 2.2921504628391034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004479641467186975,
      "loss": 5.9067,
      "mean_token_accuracy": 0.1584774598479271,
      "num_tokens": 46325262.0,
      "step": 21420
    },
    {
      "entropy": 6.516370964050293,
      "epoch": 2.292685536946867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044793969115936375,
      "loss": 5.9099,
      "mean_token_accuracy": 0.1625306412577629,
      "num_tokens": 46336773.0,
      "step": 21425
    },
    {
      "entropy": 6.444751834869384,
      "epoch": 2.293220611054631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004479152306064207,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16370812952518463,
      "num_tokens": 46346616.0,
      "step": 21430
    },
    {
      "entropy": 6.454352855682373,
      "epoch": 2.293755685162395,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004478907650605745,
      "loss": 5.8789,
      "mean_token_accuracy": 0.1692926123738289,
      "num_tokens": 46358564.0,
      "step": 21435
    },
    {
      "entropy": 6.4461390495300295,
      "epoch": 2.294290759270159,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044786629452253196,
      "loss": 5.7395,
      "mean_token_accuracy": 0.17540887147188186,
      "num_tokens": 46369207.0,
      "step": 21440
    },
    {
      "entropy": 6.504999446868896,
      "epoch": 2.294825833377923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004478418189929994,
      "loss": 5.8593,
      "mean_token_accuracy": 0.17063624858856202,
      "num_tokens": 46379382.0,
      "step": 21445
    },
    {
      "entropy": 6.430526256561279,
      "epoch": 2.2953609074856867,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044781733847268357,
      "loss": 5.9394,
      "mean_token_accuracy": 0.15819693654775618,
      "num_tokens": 46390738.0,
      "step": 21450
    },
    {
      "entropy": 6.446515226364136,
      "epoch": 2.2958959815934508,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044779285296229137,
      "loss": 5.7582,
      "mean_token_accuracy": 0.17367409467697142,
      "num_tokens": 46401740.0,
      "step": 21455
    },
    {
      "entropy": 6.534186124801636,
      "epoch": 2.2964310557012144,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044776836246252985,
      "loss": 5.8069,
      "mean_token_accuracy": 0.17123168855905532,
      "num_tokens": 46412009.0,
      "step": 21460
    },
    {
      "entropy": 6.5191107273101805,
      "epoch": 2.2969661298089785,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004477438669741061,
      "loss": 5.9089,
      "mean_token_accuracy": 0.16583413630723953,
      "num_tokens": 46422550.0,
      "step": 21465
    },
    {
      "entropy": 6.475829839706421,
      "epoch": 2.2975012039167426,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004477193664977275,
      "loss": 5.8999,
      "mean_token_accuracy": 0.16567476838827133,
      "num_tokens": 46432797.0,
      "step": 21470
    },
    {
      "entropy": 6.41486759185791,
      "epoch": 2.2980362780245063,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004476948610341015,
      "loss": 5.6984,
      "mean_token_accuracy": 0.17708563059568405,
      "num_tokens": 46444638.0,
      "step": 21475
    },
    {
      "entropy": 6.491079044342041,
      "epoch": 2.2985713521322704,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044767035058393564,
      "loss": 5.8059,
      "mean_token_accuracy": 0.17105662375688552,
      "num_tokens": 46455404.0,
      "step": 21480
    },
    {
      "entropy": 6.475593900680542,
      "epoch": 2.2991064262400345,
      "grad_norm": 0.97265625,
      "learning_rate": 0.00044764583514793766,
      "loss": 5.8955,
      "mean_token_accuracy": 0.16358212977647782,
      "num_tokens": 46466821.0,
      "step": 21485
    },
    {
      "entropy": 6.524421644210816,
      "epoch": 2.299641500347798,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004476213147268154,
      "loss": 5.861,
      "mean_token_accuracy": 0.16895377933979033,
      "num_tokens": 46477222.0,
      "step": 21490
    },
    {
      "entropy": 6.499584913253784,
      "epoch": 2.3001765744555622,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044759678932127693,
      "loss": 5.8186,
      "mean_token_accuracy": 0.16453190445899962,
      "num_tokens": 46488596.0,
      "step": 21495
    },
    {
      "entropy": 6.482390117645264,
      "epoch": 2.300711648563326,
      "grad_norm": 1.125,
      "learning_rate": 0.00044757225893203047,
      "loss": 5.8779,
      "mean_token_accuracy": 0.16494361609220504,
      "num_tokens": 46499799.0,
      "step": 21500
    },
    {
      "entropy": 6.520649003982544,
      "epoch": 2.30124672267109,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004475477235597843,
      "loss": 5.9028,
      "mean_token_accuracy": 0.1711495876312256,
      "num_tokens": 46511077.0,
      "step": 21505
    },
    {
      "entropy": 6.500854969024658,
      "epoch": 2.3017817967788536,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004475231832052469,
      "loss": 5.9305,
      "mean_token_accuracy": 0.15790300220251083,
      "num_tokens": 46523517.0,
      "step": 21510
    },
    {
      "entropy": 6.489237594604492,
      "epoch": 2.3023168708866177,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004474986378691267,
      "loss": 5.8239,
      "mean_token_accuracy": 0.17285002022981644,
      "num_tokens": 46533198.0,
      "step": 21515
    },
    {
      "entropy": 6.5430701732635494,
      "epoch": 2.302851944994382,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044747408755213265,
      "loss": 5.8443,
      "mean_token_accuracy": 0.16709946542978288,
      "num_tokens": 46544659.0,
      "step": 21520
    },
    {
      "entropy": 6.47597370147705,
      "epoch": 2.3033870191021455,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044744953225497354,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16060105860233306,
      "num_tokens": 46555897.0,
      "step": 21525
    },
    {
      "entropy": 6.48106632232666,
      "epoch": 2.3039220932099096,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044742497197835845,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16564506739377977,
      "num_tokens": 46565695.0,
      "step": 21530
    },
    {
      "entropy": 6.53315110206604,
      "epoch": 2.3044571673176737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004474004067229965,
      "loss": 5.8973,
      "mean_token_accuracy": 0.16309992372989654,
      "num_tokens": 46576620.0,
      "step": 21535
    },
    {
      "entropy": 6.588476705551147,
      "epoch": 2.3049922414254373,
      "grad_norm": 0.984375,
      "learning_rate": 0.000447375836489597,
      "loss": 5.948,
      "mean_token_accuracy": 0.15898308604955674,
      "num_tokens": 46588264.0,
      "step": 21540
    },
    {
      "entropy": 6.539448595046997,
      "epoch": 2.3055273155332014,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044735126127886944,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16986375153064728,
      "num_tokens": 46600180.0,
      "step": 21545
    },
    {
      "entropy": 6.48881950378418,
      "epoch": 2.306062389640965,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004473266810915235,
      "loss": 5.8602,
      "mean_token_accuracy": 0.1665474072098732,
      "num_tokens": 46610391.0,
      "step": 21550
    },
    {
      "entropy": 6.519261503219605,
      "epoch": 2.306597463748729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004473020959282688,
      "loss": 5.9108,
      "mean_token_accuracy": 0.16793932914733886,
      "num_tokens": 46622955.0,
      "step": 21555
    },
    {
      "entropy": 6.529154109954834,
      "epoch": 2.3071325378564933,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044727750578981535,
      "loss": 5.8805,
      "mean_token_accuracy": 0.16604702174663544,
      "num_tokens": 46632361.0,
      "step": 21560
    },
    {
      "entropy": 6.465843915939331,
      "epoch": 2.307667611964257,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004472529106768731,
      "loss": 5.836,
      "mean_token_accuracy": 0.16943798661231996,
      "num_tokens": 46642693.0,
      "step": 21565
    },
    {
      "entropy": 6.433224391937256,
      "epoch": 2.308202686072021,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004472283105901522,
      "loss": 5.7574,
      "mean_token_accuracy": 0.16932720988988875,
      "num_tokens": 46653251.0,
      "step": 21570
    },
    {
      "entropy": 6.534247493743896,
      "epoch": 2.3087377601797847,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004472037055303631,
      "loss": 5.918,
      "mean_token_accuracy": 0.15541287362575532,
      "num_tokens": 46664751.0,
      "step": 21575
    },
    {
      "entropy": 6.477377128601074,
      "epoch": 2.309272834287549,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004471790954982162,
      "loss": 5.7771,
      "mean_token_accuracy": 0.1657135859131813,
      "num_tokens": 46676825.0,
      "step": 21580
    },
    {
      "entropy": 6.487316370010376,
      "epoch": 2.309807908395313,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004471544804944221,
      "loss": 5.8003,
      "mean_token_accuracy": 0.1729530081152916,
      "num_tokens": 46687202.0,
      "step": 21585
    },
    {
      "entropy": 6.502920150756836,
      "epoch": 2.3103429825030766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004471298605196916,
      "loss": 5.9408,
      "mean_token_accuracy": 0.1660833790898323,
      "num_tokens": 46697687.0,
      "step": 21590
    },
    {
      "entropy": 6.485231542587281,
      "epoch": 2.3108780566108407,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044710523557473544,
      "loss": 5.7965,
      "mean_token_accuracy": 0.17151385247707368,
      "num_tokens": 46709088.0,
      "step": 21595
    },
    {
      "entropy": 6.502265405654907,
      "epoch": 2.3114131307186048,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044708060566026487,
      "loss": 5.8935,
      "mean_token_accuracy": 0.16616518199443817,
      "num_tokens": 46721675.0,
      "step": 21600
    },
    {
      "entropy": 6.491225910186768,
      "epoch": 2.3119482048263684,
      "grad_norm": 1.0,
      "learning_rate": 0.000447055970776991,
      "loss": 5.9244,
      "mean_token_accuracy": 0.1629437252879143,
      "num_tokens": 46731854.0,
      "step": 21605
    },
    {
      "entropy": 6.491027927398681,
      "epoch": 2.3124832789341325,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044703133092562514,
      "loss": 5.9095,
      "mean_token_accuracy": 0.1625854566693306,
      "num_tokens": 46742815.0,
      "step": 21610
    },
    {
      "entropy": 6.509676170349121,
      "epoch": 2.313018353041896,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044700668610687884,
      "loss": 5.8576,
      "mean_token_accuracy": 0.17413725405931474,
      "num_tokens": 46753012.0,
      "step": 21615
    },
    {
      "entropy": 6.435381555557251,
      "epoch": 2.3135534271496603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044698203632146353,
      "loss": 5.8401,
      "mean_token_accuracy": 0.1691618397831917,
      "num_tokens": 46763701.0,
      "step": 21620
    },
    {
      "entropy": 6.511702919006348,
      "epoch": 2.314088501257424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004469573815700911,
      "loss": 5.8951,
      "mean_token_accuracy": 0.16223306506872176,
      "num_tokens": 46775407.0,
      "step": 21625
    },
    {
      "entropy": 6.450700044631958,
      "epoch": 2.314623575365188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044693272185347357,
      "loss": 5.7832,
      "mean_token_accuracy": 0.16860665380954742,
      "num_tokens": 46786282.0,
      "step": 21630
    },
    {
      "entropy": 6.469667100906372,
      "epoch": 2.315158649472952,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004469080571723228,
      "loss": 5.8786,
      "mean_token_accuracy": 0.1657257378101349,
      "num_tokens": 46796768.0,
      "step": 21635
    },
    {
      "entropy": 6.4272199153900145,
      "epoch": 2.315693723580716,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044688338752735095,
      "loss": 5.7944,
      "mean_token_accuracy": 0.16916946023702623,
      "num_tokens": 46806538.0,
      "step": 21640
    },
    {
      "entropy": 6.418496894836426,
      "epoch": 2.31622879768848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004468587129192705,
      "loss": 5.7821,
      "mean_token_accuracy": 0.1761409178376198,
      "num_tokens": 46816692.0,
      "step": 21645
    },
    {
      "entropy": 6.448935270309448,
      "epoch": 2.316763871796244,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044683403334879384,
      "loss": 5.8607,
      "mean_token_accuracy": 0.17273238599300383,
      "num_tokens": 46827551.0,
      "step": 21650
    },
    {
      "entropy": 6.519799852371216,
      "epoch": 2.3172989459040076,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004468093488166337,
      "loss": 5.9749,
      "mean_token_accuracy": 0.15675267726182937,
      "num_tokens": 46840168.0,
      "step": 21655
    },
    {
      "entropy": 6.520435333251953,
      "epoch": 2.3178340200117717,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044678465932350273,
      "loss": 5.8465,
      "mean_token_accuracy": 0.1690165862441063,
      "num_tokens": 46851086.0,
      "step": 21660
    },
    {
      "entropy": 6.41790623664856,
      "epoch": 2.3183690941195354,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044675996487011373,
      "loss": 5.7943,
      "mean_token_accuracy": 0.17407871782779694,
      "num_tokens": 46861830.0,
      "step": 21665
    },
    {
      "entropy": 6.499821329116822,
      "epoch": 2.3189041682272995,
      "grad_norm": 1.125,
      "learning_rate": 0.00044673526545718,
      "loss": 5.9005,
      "mean_token_accuracy": 0.1670697957277298,
      "num_tokens": 46871348.0,
      "step": 21670
    },
    {
      "entropy": 6.460634994506836,
      "epoch": 2.3194392423350636,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004467105610854146,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16179295629262924,
      "num_tokens": 46882737.0,
      "step": 21675
    },
    {
      "entropy": 6.537662506103516,
      "epoch": 2.3199743164428273,
      "grad_norm": 0.984375,
      "learning_rate": 0.00044668585175553085,
      "loss": 5.932,
      "mean_token_accuracy": 0.1539579689502716,
      "num_tokens": 46893616.0,
      "step": 21680
    },
    {
      "entropy": 6.510444021224975,
      "epoch": 2.3205093905505914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004466611374682423,
      "loss": 5.852,
      "mean_token_accuracy": 0.1655426189303398,
      "num_tokens": 46904546.0,
      "step": 21685
    },
    {
      "entropy": 6.438604354858398,
      "epoch": 2.321044464658355,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044663641822426244,
      "loss": 5.8384,
      "mean_token_accuracy": 0.16705650091171265,
      "num_tokens": 46915513.0,
      "step": 21690
    },
    {
      "entropy": 6.5424659729003904,
      "epoch": 2.321579538766119,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044661169402430517,
      "loss": 5.9015,
      "mean_token_accuracy": 0.16826484501361846,
      "num_tokens": 46925535.0,
      "step": 21695
    },
    {
      "entropy": 6.449262952804565,
      "epoch": 2.322114612873883,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044658696486908443,
      "loss": 5.8914,
      "mean_token_accuracy": 0.1666087195277214,
      "num_tokens": 46936222.0,
      "step": 21700
    },
    {
      "entropy": 6.430657815933228,
      "epoch": 2.322649686981647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004465622307593141,
      "loss": 5.7889,
      "mean_token_accuracy": 0.17486222088336945,
      "num_tokens": 46946855.0,
      "step": 21705
    },
    {
      "entropy": 6.493397378921509,
      "epoch": 2.323184761089411,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044653749169570845,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16840722858905793,
      "num_tokens": 46958246.0,
      "step": 21710
    },
    {
      "entropy": 6.440023136138916,
      "epoch": 2.3237198351971746,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044651274767898186,
      "loss": 5.8582,
      "mean_token_accuracy": 0.16783496588468552,
      "num_tokens": 46969237.0,
      "step": 21715
    },
    {
      "entropy": 6.524897670745849,
      "epoch": 2.3242549093049387,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004464879987098488,
      "loss": 5.9252,
      "mean_token_accuracy": 0.1692347928881645,
      "num_tokens": 46979897.0,
      "step": 21720
    },
    {
      "entropy": 6.553484106063843,
      "epoch": 2.324789983412703,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004464632447890238,
      "loss": 5.8158,
      "mean_token_accuracy": 0.17113833129405975,
      "num_tokens": 46989833.0,
      "step": 21725
    },
    {
      "entropy": 6.432345151901245,
      "epoch": 2.3253250575204665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044643848591722183,
      "loss": 5.8225,
      "mean_token_accuracy": 0.17391082346439363,
      "num_tokens": 46999822.0,
      "step": 21730
    },
    {
      "entropy": 6.429431819915772,
      "epoch": 2.3258601316282306,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044641372209515764,
      "loss": 5.829,
      "mean_token_accuracy": 0.1705512821674347,
      "num_tokens": 47010176.0,
      "step": 21735
    },
    {
      "entropy": 6.423533868789673,
      "epoch": 2.3263952057359942,
      "grad_norm": 1.125,
      "learning_rate": 0.0004463889533235462,
      "loss": 5.8277,
      "mean_token_accuracy": 0.17135369181632995,
      "num_tokens": 47020184.0,
      "step": 21740
    },
    {
      "entropy": 6.516287088394165,
      "epoch": 2.3269302798437583,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044636417960310294,
      "loss": 5.9191,
      "mean_token_accuracy": 0.16408784091472625,
      "num_tokens": 47032066.0,
      "step": 21745
    },
    {
      "entropy": 6.5498942852020265,
      "epoch": 2.3274653539515224,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004463394009345431,
      "loss": 5.9349,
      "mean_token_accuracy": 0.16045622974634172,
      "num_tokens": 47043571.0,
      "step": 21750
    },
    {
      "entropy": 6.530133199691773,
      "epoch": 2.328000428059286,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044631461731858195,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16530881524086,
      "num_tokens": 47053813.0,
      "step": 21755
    },
    {
      "entropy": 6.50972990989685,
      "epoch": 2.32853550216705,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044628982875593546,
      "loss": 5.8916,
      "mean_token_accuracy": 0.16227002590894699,
      "num_tokens": 47063671.0,
      "step": 21760
    },
    {
      "entropy": 6.394323873519897,
      "epoch": 2.3290705762748143,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044626503524731917,
      "loss": 5.7824,
      "mean_token_accuracy": 0.17499128133058547,
      "num_tokens": 47073537.0,
      "step": 21765
    },
    {
      "entropy": 6.424983978271484,
      "epoch": 2.329605650382578,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004462402367934491,
      "loss": 5.8667,
      "mean_token_accuracy": 0.16861234456300736,
      "num_tokens": 47084272.0,
      "step": 21770
    },
    {
      "entropy": 6.4960705757141115,
      "epoch": 2.330140724490342,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004462154333950412,
      "loss": 5.9013,
      "mean_token_accuracy": 0.16609022617340088,
      "num_tokens": 47095198.0,
      "step": 21775
    },
    {
      "entropy": 6.531994152069092,
      "epoch": 2.3306757985981057,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004461906250528117,
      "loss": 5.9121,
      "mean_token_accuracy": 0.1629682019352913,
      "num_tokens": 47106540.0,
      "step": 21780
    },
    {
      "entropy": 6.559878730773926,
      "epoch": 2.33121087270587,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000446165811767477,
      "loss": 5.9029,
      "mean_token_accuracy": 0.1722935453057289,
      "num_tokens": 47117301.0,
      "step": 21785
    },
    {
      "entropy": 6.51632285118103,
      "epoch": 2.3317459468136335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004461409935397536,
      "loss": 5.9043,
      "mean_token_accuracy": 0.1681519106030464,
      "num_tokens": 47127058.0,
      "step": 21790
    },
    {
      "entropy": 6.450133848190307,
      "epoch": 2.3322810209213976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044611617037035794,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16815730482339858,
      "num_tokens": 47137182.0,
      "step": 21795
    },
    {
      "entropy": 6.486263704299927,
      "epoch": 2.3328160950291617,
      "grad_norm": 0.984375,
      "learning_rate": 0.000446091342260007,
      "loss": 5.8912,
      "mean_token_accuracy": 0.16219987124204635,
      "num_tokens": 47148591.0,
      "step": 21800
    },
    {
      "entropy": 6.496916580200195,
      "epoch": 2.3333511691369253,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044606650920941753,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16308656558394433,
      "num_tokens": 47159014.0,
      "step": 21805
    },
    {
      "entropy": 6.502030229568481,
      "epoch": 2.3338862432446894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004460416712193067,
      "loss": 5.8904,
      "mean_token_accuracy": 0.16338668763637543,
      "num_tokens": 47168760.0,
      "step": 21810
    },
    {
      "entropy": 6.491846084594727,
      "epoch": 2.3344213173524535,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004460168282903916,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16330858021974565,
      "num_tokens": 47180902.0,
      "step": 21815
    },
    {
      "entropy": 6.477505350112915,
      "epoch": 2.334956391460217,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044599198042338963,
      "loss": 5.775,
      "mean_token_accuracy": 0.1746992588043213,
      "num_tokens": 47190458.0,
      "step": 21820
    },
    {
      "entropy": 6.46680269241333,
      "epoch": 2.3354914655679813,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044596712761901824,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16478560268878936,
      "num_tokens": 47201099.0,
      "step": 21825
    },
    {
      "entropy": 6.430627012252808,
      "epoch": 2.336026539675745,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004459422698779951,
      "loss": 5.8396,
      "mean_token_accuracy": 0.17092858850955964,
      "num_tokens": 47211284.0,
      "step": 21830
    },
    {
      "entropy": 6.486549806594849,
      "epoch": 2.336561613783509,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004459174072010379,
      "loss": 5.9201,
      "mean_token_accuracy": 0.16473333835601806,
      "num_tokens": 47222875.0,
      "step": 21835
    },
    {
      "entropy": 6.489226818084717,
      "epoch": 2.337096687891273,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004458925395888646,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1601955845952034,
      "num_tokens": 47234238.0,
      "step": 21840
    },
    {
      "entropy": 6.516684532165527,
      "epoch": 2.3376317619990368,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044586766704219326,
      "loss": 5.8882,
      "mean_token_accuracy": 0.16580738723278046,
      "num_tokens": 47244634.0,
      "step": 21845
    },
    {
      "entropy": 6.534788036346436,
      "epoch": 2.338166836106801,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000445842789561742,
      "loss": 5.8958,
      "mean_token_accuracy": 0.16102438122034074,
      "num_tokens": 47256538.0,
      "step": 21850
    },
    {
      "entropy": 6.416242551803589,
      "epoch": 2.3387019102145645,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004458179071482292,
      "loss": 5.7492,
      "mean_token_accuracy": 0.180821792781353,
      "num_tokens": 47266373.0,
      "step": 21855
    },
    {
      "entropy": 6.45223298072815,
      "epoch": 2.3392369843223286,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044579301980237333,
      "loss": 5.922,
      "mean_token_accuracy": 0.1568940222263336,
      "num_tokens": 47276441.0,
      "step": 21860
    },
    {
      "entropy": 6.491032314300537,
      "epoch": 2.3397720584300927,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004457681275248931,
      "loss": 5.9083,
      "mean_token_accuracy": 0.16336438655853272,
      "num_tokens": 47287299.0,
      "step": 21865
    },
    {
      "entropy": 6.4218645095825195,
      "epoch": 2.3403071325378564,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044574323031650715,
      "loss": 5.7873,
      "mean_token_accuracy": 0.16646311730146407,
      "num_tokens": 47298007.0,
      "step": 21870
    },
    {
      "entropy": 6.405284881591797,
      "epoch": 2.3408422066456205,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004457183281779343,
      "loss": 5.747,
      "mean_token_accuracy": 0.17829046100378038,
      "num_tokens": 47309729.0,
      "step": 21875
    },
    {
      "entropy": 6.497087907791138,
      "epoch": 2.3413772807533846,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044569342110989385,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16778986006975175,
      "num_tokens": 47319050.0,
      "step": 21880
    },
    {
      "entropy": 6.5246974468231205,
      "epoch": 2.3419123548611482,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004456685091131048,
      "loss": 5.9797,
      "mean_token_accuracy": 0.16263773292303085,
      "num_tokens": 47330365.0,
      "step": 21885
    },
    {
      "entropy": 6.584377002716065,
      "epoch": 2.3424474289689123,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044564359218828654,
      "loss": 5.8858,
      "mean_token_accuracy": 0.1602620378136635,
      "num_tokens": 47341225.0,
      "step": 21890
    },
    {
      "entropy": 6.470265054702759,
      "epoch": 2.342982503076676,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044561867033615854,
      "loss": 5.8173,
      "mean_token_accuracy": 0.17341943234205245,
      "num_tokens": 47352996.0,
      "step": 21895
    },
    {
      "entropy": 6.401006364822388,
      "epoch": 2.34351757718444,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004455937435574404,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17151351869106293,
      "num_tokens": 47363577.0,
      "step": 21900
    },
    {
      "entropy": 6.510982513427734,
      "epoch": 2.3440526512922037,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044556881185285185,
      "loss": 5.9909,
      "mean_token_accuracy": 0.1604711189866066,
      "num_tokens": 47374389.0,
      "step": 21905
    },
    {
      "entropy": 6.460742568969726,
      "epoch": 2.344587725399968,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044554387522311286,
      "loss": 5.8154,
      "mean_token_accuracy": 0.17220281213521957,
      "num_tokens": 47385624.0,
      "step": 21910
    },
    {
      "entropy": 6.523248338699341,
      "epoch": 2.345122799507732,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004455189336689435,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1632843419909477,
      "num_tokens": 47396611.0,
      "step": 21915
    },
    {
      "entropy": 6.546470308303833,
      "epoch": 2.3456578736154956,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044549398719106374,
      "loss": 5.8765,
      "mean_token_accuracy": 0.16329316049814224,
      "num_tokens": 47406693.0,
      "step": 21920
    },
    {
      "entropy": 6.468829345703125,
      "epoch": 2.3461929477232597,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004454690357901942,
      "loss": 5.7868,
      "mean_token_accuracy": 0.1682095170021057,
      "num_tokens": 47417528.0,
      "step": 21925
    },
    {
      "entropy": 6.499433851242065,
      "epoch": 2.346728021831024,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004454440794670552,
      "loss": 5.9178,
      "mean_token_accuracy": 0.16592097878456116,
      "num_tokens": 47430032.0,
      "step": 21930
    },
    {
      "entropy": 6.510506725311279,
      "epoch": 2.3472630959387875,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044541911822236724,
      "loss": 5.9433,
      "mean_token_accuracy": 0.1603750228881836,
      "num_tokens": 47441002.0,
      "step": 21935
    },
    {
      "entropy": 6.479095029830932,
      "epoch": 2.3477981700465516,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004453941520568513,
      "loss": 5.7967,
      "mean_token_accuracy": 0.17677695602178572,
      "num_tokens": 47451716.0,
      "step": 21940
    },
    {
      "entropy": 6.479981803894043,
      "epoch": 2.348333244154315,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004453691809712281,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16581775844097138,
      "num_tokens": 47463039.0,
      "step": 21945
    },
    {
      "entropy": 6.4180436611175535,
      "epoch": 2.3488683182620793,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044534420496621877,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1759392350912094,
      "num_tokens": 47473836.0,
      "step": 21950
    },
    {
      "entropy": 6.4484560012817385,
      "epoch": 2.3494033923698434,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044531922404254445,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16561149656772614,
      "num_tokens": 47485103.0,
      "step": 21955
    },
    {
      "entropy": 6.401976156234741,
      "epoch": 2.349938466477607,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004452942382009265,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17344778925180435,
      "num_tokens": 47495509.0,
      "step": 21960
    },
    {
      "entropy": 6.48512601852417,
      "epoch": 2.350473540585371,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004452692474420863,
      "loss": 5.8428,
      "mean_token_accuracy": 0.17599835842847825,
      "num_tokens": 47506062.0,
      "step": 21965
    },
    {
      "entropy": 6.423240709304809,
      "epoch": 2.351008614693135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004452442517667455,
      "loss": 5.8047,
      "mean_token_accuracy": 0.17291782796382904,
      "num_tokens": 47515632.0,
      "step": 21970
    },
    {
      "entropy": 6.415789985656739,
      "epoch": 2.351543688800899,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004452192511756259,
      "loss": 5.7538,
      "mean_token_accuracy": 0.17480533868074416,
      "num_tokens": 47525412.0,
      "step": 21975
    },
    {
      "entropy": 6.432691478729248,
      "epoch": 2.352078762908663,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044519424566944935,
      "loss": 5.8402,
      "mean_token_accuracy": 0.1682863026857376,
      "num_tokens": 47536094.0,
      "step": 21980
    },
    {
      "entropy": 6.453028821945191,
      "epoch": 2.3526138370164267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044516923524893777,
      "loss": 5.8026,
      "mean_token_accuracy": 0.16969103962183,
      "num_tokens": 47546893.0,
      "step": 21985
    },
    {
      "entropy": 6.467904996871948,
      "epoch": 2.353148911124191,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044514421991481354,
      "loss": 5.8164,
      "mean_token_accuracy": 0.1718533754348755,
      "num_tokens": 47557117.0,
      "step": 21990
    },
    {
      "entropy": 6.453710412979126,
      "epoch": 2.3536839852319544,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044511919966779885,
      "loss": 5.7909,
      "mean_token_accuracy": 0.168911674618721,
      "num_tokens": 47567878.0,
      "step": 21995
    },
    {
      "entropy": 6.513557434082031,
      "epoch": 2.3542190593397185,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044509417450861615,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16148020029067994,
      "num_tokens": 47578185.0,
      "step": 22000
    },
    {
      "entropy": 6.528646469116211,
      "epoch": 2.3547541334474826,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004450691444379881,
      "loss": 5.9672,
      "mean_token_accuracy": 0.1571372076869011,
      "num_tokens": 47590042.0,
      "step": 22005
    },
    {
      "entropy": 6.515891122817993,
      "epoch": 2.3552892075552463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044504410945663735,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16483661979436875,
      "num_tokens": 47601225.0,
      "step": 22010
    },
    {
      "entropy": 6.453355026245117,
      "epoch": 2.3558242816630104,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044501906956528686,
      "loss": 5.873,
      "mean_token_accuracy": 0.1633631020784378,
      "num_tokens": 47612596.0,
      "step": 22015
    },
    {
      "entropy": 6.523511075973511,
      "epoch": 2.356359355770774,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004449940247646596,
      "loss": 5.928,
      "mean_token_accuracy": 0.16437840908765794,
      "num_tokens": 47623156.0,
      "step": 22020
    },
    {
      "entropy": 6.462477397918701,
      "epoch": 2.356894429878538,
      "grad_norm": 1.0,
      "learning_rate": 0.00044496897505547886,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16403736919164658,
      "num_tokens": 47633743.0,
      "step": 22025
    },
    {
      "entropy": 6.483136892318726,
      "epoch": 2.3574295039863022,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004449439204384677,
      "loss": 5.9074,
      "mean_token_accuracy": 0.16287122815847396,
      "num_tokens": 47643914.0,
      "step": 22030
    },
    {
      "entropy": 6.504411268234253,
      "epoch": 2.357964578094066,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044491886091434983,
      "loss": 5.8789,
      "mean_token_accuracy": 0.16399407833814622,
      "num_tokens": 47655187.0,
      "step": 22035
    },
    {
      "entropy": 6.508800840377807,
      "epoch": 2.35849965220183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004448937964838487,
      "loss": 5.8569,
      "mean_token_accuracy": 0.1658545821905136,
      "num_tokens": 47667193.0,
      "step": 22040
    },
    {
      "entropy": 6.48462438583374,
      "epoch": 2.359034726309594,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004448687271476881,
      "loss": 5.8967,
      "mean_token_accuracy": 0.16406976878643037,
      "num_tokens": 47678116.0,
      "step": 22045
    },
    {
      "entropy": 6.508517551422119,
      "epoch": 2.3595698004173578,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004448436529065918,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16815751045942307,
      "num_tokens": 47689883.0,
      "step": 22050
    },
    {
      "entropy": 6.487256622314453,
      "epoch": 2.360104874525122,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004448185737612839,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16420442014932632,
      "num_tokens": 47699939.0,
      "step": 22055
    },
    {
      "entropy": 6.399065351486206,
      "epoch": 2.3606399486328855,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044479348971248863,
      "loss": 5.8001,
      "mean_token_accuracy": 0.1755296394228935,
      "num_tokens": 47710477.0,
      "step": 22060
    },
    {
      "entropy": 6.510415267944336,
      "epoch": 2.3611750227406496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004447684007609301,
      "loss": 5.8663,
      "mean_token_accuracy": 0.16414888948202133,
      "num_tokens": 47720548.0,
      "step": 22065
    },
    {
      "entropy": 6.508013582229614,
      "epoch": 2.3617100968484133,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044474330690733293,
      "loss": 5.8491,
      "mean_token_accuracy": 0.163398177921772,
      "num_tokens": 47732135.0,
      "step": 22070
    },
    {
      "entropy": 6.485276985168457,
      "epoch": 2.3622451709561774,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004447182081524216,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16303360909223558,
      "num_tokens": 47742934.0,
      "step": 22075
    },
    {
      "entropy": 6.441720819473266,
      "epoch": 2.3627802450639415,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044469310449692085,
      "loss": 5.8748,
      "mean_token_accuracy": 0.165049047768116,
      "num_tokens": 47754586.0,
      "step": 22080
    },
    {
      "entropy": 6.49192533493042,
      "epoch": 2.363315319171705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044466799594155556,
      "loss": 5.8729,
      "mean_token_accuracy": 0.16245600134134291,
      "num_tokens": 47765564.0,
      "step": 22085
    },
    {
      "entropy": 6.427829837799072,
      "epoch": 2.363850393279469,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044464288248705075,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16884429454803468,
      "num_tokens": 47776191.0,
      "step": 22090
    },
    {
      "entropy": 6.5709208965301515,
      "epoch": 2.3643854673872333,
      "grad_norm": 1.0,
      "learning_rate": 0.00044461776413413153,
      "loss": 5.9581,
      "mean_token_accuracy": 0.1605585977435112,
      "num_tokens": 47787193.0,
      "step": 22095
    },
    {
      "entropy": 6.576856565475464,
      "epoch": 2.364920541494997,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004445926408835232,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16081370562314987,
      "num_tokens": 47798576.0,
      "step": 22100
    },
    {
      "entropy": 6.3913637638092045,
      "epoch": 2.365455615602761,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004445675127359511,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17675468623638152,
      "num_tokens": 47809519.0,
      "step": 22105
    },
    {
      "entropy": 6.442627429962158,
      "epoch": 2.3659906897105247,
      "grad_norm": 1.140625,
      "learning_rate": 0.000444542379692141,
      "loss": 5.8237,
      "mean_token_accuracy": 0.171942800283432,
      "num_tokens": 47820618.0,
      "step": 22110
    },
    {
      "entropy": 6.503458023071289,
      "epoch": 2.366525763818289,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004445172417528186,
      "loss": 5.896,
      "mean_token_accuracy": 0.16439107209444045,
      "num_tokens": 47831756.0,
      "step": 22115
    },
    {
      "entropy": 6.51939582824707,
      "epoch": 2.367060837926053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044449209891870954,
      "loss": 5.8644,
      "mean_token_accuracy": 0.16158444285392762,
      "num_tokens": 47841974.0,
      "step": 22120
    },
    {
      "entropy": 6.410629987716675,
      "epoch": 2.3675959120338166,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044446695119053993,
      "loss": 5.7869,
      "mean_token_accuracy": 0.16878886967897416,
      "num_tokens": 47852194.0,
      "step": 22125
    },
    {
      "entropy": 6.490573501586914,
      "epoch": 2.3681309861415807,
      "grad_norm": 1.09375,
      "learning_rate": 0.000444441798569036,
      "loss": 5.9389,
      "mean_token_accuracy": 0.16081197559833527,
      "num_tokens": 47863921.0,
      "step": 22130
    },
    {
      "entropy": 6.5719915390014645,
      "epoch": 2.3686660602493443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044441664105492393,
      "loss": 6.0546,
      "mean_token_accuracy": 0.15439448058605193,
      "num_tokens": 47875520.0,
      "step": 22135
    },
    {
      "entropy": 6.521930837631226,
      "epoch": 2.3692011343571084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004443914786489301,
      "loss": 5.8,
      "mean_token_accuracy": 0.16876946538686752,
      "num_tokens": 47886047.0,
      "step": 22140
    },
    {
      "entropy": 6.372669506072998,
      "epoch": 2.3697362084648725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044436631135178113,
      "loss": 5.7856,
      "mean_token_accuracy": 0.17274439036846162,
      "num_tokens": 47896449.0,
      "step": 22145
    },
    {
      "entropy": 6.509721183776856,
      "epoch": 2.370271282572636,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044434113916420377,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16444920748472214,
      "num_tokens": 47908613.0,
      "step": 22150
    },
    {
      "entropy": 6.57553129196167,
      "epoch": 2.3708063566804003,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044431596208692475,
      "loss": 5.9567,
      "mean_token_accuracy": 0.1520371876657009,
      "num_tokens": 47919474.0,
      "step": 22155
    },
    {
      "entropy": 6.449701881408691,
      "epoch": 2.3713414307881644,
      "grad_norm": 1.0,
      "learning_rate": 0.0004442907801206712,
      "loss": 5.8485,
      "mean_token_accuracy": 0.1685652866959572,
      "num_tokens": 47930929.0,
      "step": 22160
    },
    {
      "entropy": 6.454047298431396,
      "epoch": 2.371876504895928,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044426559326617013,
      "loss": 5.7981,
      "mean_token_accuracy": 0.174635611474514,
      "num_tokens": 47942050.0,
      "step": 22165
    },
    {
      "entropy": 6.522723960876465,
      "epoch": 2.372411579003692,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004442404015241488,
      "loss": 5.9271,
      "mean_token_accuracy": 0.15920935720205306,
      "num_tokens": 47953162.0,
      "step": 22170
    },
    {
      "entropy": 6.5607178688049315,
      "epoch": 2.372946653111456,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044421520489533467,
      "loss": 5.8832,
      "mean_token_accuracy": 0.16630488336086274,
      "num_tokens": 47963140.0,
      "step": 22175
    },
    {
      "entropy": 6.467858028411865,
      "epoch": 2.37348172721922,
      "grad_norm": 1.765625,
      "learning_rate": 0.00044419000338045527,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16710296496748925,
      "num_tokens": 47975622.0,
      "step": 22180
    },
    {
      "entropy": 6.5080865859985355,
      "epoch": 2.3740168013269836,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004441647969802383,
      "loss": 5.8748,
      "mean_token_accuracy": 0.159488283097744,
      "num_tokens": 47986450.0,
      "step": 22185
    },
    {
      "entropy": 6.555475234985352,
      "epoch": 2.3745518754347477,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004441395856954117,
      "loss": 5.9386,
      "mean_token_accuracy": 0.1593143090605736,
      "num_tokens": 47997441.0,
      "step": 22190
    },
    {
      "entropy": 6.524975442886353,
      "epoch": 2.3750869495425118,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044411436952670325,
      "loss": 5.8405,
      "mean_token_accuracy": 0.1615590751171112,
      "num_tokens": 48008995.0,
      "step": 22195
    },
    {
      "entropy": 6.51609239578247,
      "epoch": 2.3756220236502754,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044408914847484103,
      "loss": 5.8668,
      "mean_token_accuracy": 0.16854875087738036,
      "num_tokens": 48019975.0,
      "step": 22200
    },
    {
      "entropy": 6.5305956363677975,
      "epoch": 2.3761570977580395,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004440639225405536,
      "loss": 5.8661,
      "mean_token_accuracy": 0.1624830335378647,
      "num_tokens": 48031278.0,
      "step": 22205
    },
    {
      "entropy": 6.456838989257813,
      "epoch": 2.3766921718658036,
      "grad_norm": 1.0625,
      "learning_rate": 0.000444038691724569,
      "loss": 5.858,
      "mean_token_accuracy": 0.16932945996522902,
      "num_tokens": 48042050.0,
      "step": 22210
    },
    {
      "entropy": 6.390295028686523,
      "epoch": 2.3772272459735673,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044401345602761605,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1789153680205345,
      "num_tokens": 48052833.0,
      "step": 22215
    },
    {
      "entropy": 6.439375352859497,
      "epoch": 2.3777623200813314,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044398821545042327,
      "loss": 5.8412,
      "mean_token_accuracy": 0.1672619476914406,
      "num_tokens": 48063425.0,
      "step": 22220
    },
    {
      "entropy": 6.437749767303467,
      "epoch": 2.378297394189095,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044396296999371946,
      "loss": 5.8098,
      "mean_token_accuracy": 0.17153306156396866,
      "num_tokens": 48074941.0,
      "step": 22225
    },
    {
      "entropy": 6.389747428894043,
      "epoch": 2.378832468296859,
      "grad_norm": 2.125,
      "learning_rate": 0.00044393771965823376,
      "loss": 5.6804,
      "mean_token_accuracy": 0.17552227079868316,
      "num_tokens": 48085224.0,
      "step": 22230
    },
    {
      "entropy": 6.49615159034729,
      "epoch": 2.3793675424046232,
      "grad_norm": 1.015625,
      "learning_rate": 0.000443912464444695,
      "loss": 5.8686,
      "mean_token_accuracy": 0.16418653428554536,
      "num_tokens": 48096509.0,
      "step": 22235
    },
    {
      "entropy": 6.447915077209473,
      "epoch": 2.379902616512387,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004438872043538327,
      "loss": 5.8202,
      "mean_token_accuracy": 0.16724435538053511,
      "num_tokens": 48106839.0,
      "step": 22240
    },
    {
      "entropy": 6.534123468399048,
      "epoch": 2.380437690620151,
      "grad_norm": 1.078125,
      "learning_rate": 0.000443861939386376,
      "loss": 5.9022,
      "mean_token_accuracy": 0.16472362875938415,
      "num_tokens": 48118209.0,
      "step": 22245
    },
    {
      "entropy": 6.511884927749634,
      "epoch": 2.3809727647279146,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004438366695430546,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16286516040563584,
      "num_tokens": 48128651.0,
      "step": 22250
    },
    {
      "entropy": 6.517772817611695,
      "epoch": 2.3815078388356787,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044381139482459805,
      "loss": 5.9088,
      "mean_token_accuracy": 0.1637333557009697,
      "num_tokens": 48139612.0,
      "step": 22255
    },
    {
      "entropy": 6.512114572525024,
      "epoch": 2.382042912943443,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044378611523173615,
      "loss": 5.8073,
      "mean_token_accuracy": 0.16861847639083863,
      "num_tokens": 48150557.0,
      "step": 22260
    },
    {
      "entropy": 6.444575548171997,
      "epoch": 2.3825779870512065,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004437608307651989,
      "loss": 5.7629,
      "mean_token_accuracy": 0.17384880185127258,
      "num_tokens": 48160784.0,
      "step": 22265
    },
    {
      "entropy": 6.425744581222534,
      "epoch": 2.3831130611589706,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004437355414257165,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16512642204761505,
      "num_tokens": 48171075.0,
      "step": 22270
    },
    {
      "entropy": 6.503937911987305,
      "epoch": 2.3836481352667347,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00044371024721401894,
      "loss": 5.8944,
      "mean_token_accuracy": 0.15893257707357406,
      "num_tokens": 48182713.0,
      "step": 22275
    },
    {
      "entropy": 6.4966319561004635,
      "epoch": 2.3841832093744983,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004436849481308367,
      "loss": 5.8282,
      "mean_token_accuracy": 0.17403660714626312,
      "num_tokens": 48194198.0,
      "step": 22280
    },
    {
      "entropy": 6.4700275421142575,
      "epoch": 2.3847182834822624,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044365964417690035,
      "loss": 5.7788,
      "mean_token_accuracy": 0.17006421983242034,
      "num_tokens": 48205650.0,
      "step": 22285
    },
    {
      "entropy": 6.410359334945679,
      "epoch": 2.385253357590026,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004436343353529404,
      "loss": 5.8081,
      "mean_token_accuracy": 0.17207445949316025,
      "num_tokens": 48216335.0,
      "step": 22290
    },
    {
      "entropy": 6.462364530563354,
      "epoch": 2.38578843169779,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044360902165968774,
      "loss": 5.834,
      "mean_token_accuracy": 0.16749124079942704,
      "num_tokens": 48227206.0,
      "step": 22295
    },
    {
      "entropy": 6.46773886680603,
      "epoch": 2.386323505805554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004435837030978733,
      "loss": 5.8315,
      "mean_token_accuracy": 0.1714819386601448,
      "num_tokens": 48237948.0,
      "step": 22300
    },
    {
      "entropy": 6.463465452194214,
      "epoch": 2.386858579913318,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004435583796682281,
      "loss": 5.8528,
      "mean_token_accuracy": 0.17822350561618805,
      "num_tokens": 48250086.0,
      "step": 22305
    },
    {
      "entropy": 6.503418684005737,
      "epoch": 2.387393654021082,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004435330513714834,
      "loss": 5.867,
      "mean_token_accuracy": 0.16837407052516937,
      "num_tokens": 48261004.0,
      "step": 22310
    },
    {
      "entropy": 6.472817277908325,
      "epoch": 2.3879287281288457,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004435077182083705,
      "loss": 5.8866,
      "mean_token_accuracy": 0.16605350524187087,
      "num_tokens": 48271314.0,
      "step": 22315
    },
    {
      "entropy": 6.418709373474121,
      "epoch": 2.38846380223661,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044348238017962095,
      "loss": 5.7882,
      "mean_token_accuracy": 0.1727592095732689,
      "num_tokens": 48281899.0,
      "step": 22320
    },
    {
      "entropy": 6.468342971801758,
      "epoch": 2.388998876344374,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004434570372859663,
      "loss": 5.8568,
      "mean_token_accuracy": 0.16882280260324478,
      "num_tokens": 48292925.0,
      "step": 22325
    },
    {
      "entropy": 6.549630403518677,
      "epoch": 2.3895339504521376,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004434316895281384,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16526953876018524,
      "num_tokens": 48303138.0,
      "step": 22330
    },
    {
      "entropy": 6.414926815032959,
      "epoch": 2.3900690245599017,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004434063369068691,
      "loss": 5.74,
      "mean_token_accuracy": 0.18532718122005462,
      "num_tokens": 48313586.0,
      "step": 22335
    },
    {
      "entropy": 6.387601137161255,
      "epoch": 2.3906040986676653,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044338097942289045,
      "loss": 5.8096,
      "mean_token_accuracy": 0.17318292409181596,
      "num_tokens": 48324756.0,
      "step": 22340
    },
    {
      "entropy": 6.485189008712768,
      "epoch": 2.3911391727754294,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004433556170769348,
      "loss": 5.8954,
      "mean_token_accuracy": 0.1585668668150902,
      "num_tokens": 48336012.0,
      "step": 22345
    },
    {
      "entropy": 6.53205041885376,
      "epoch": 2.391674246883193,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004433302498697343,
      "loss": 5.9238,
      "mean_token_accuracy": 0.15887781828641892,
      "num_tokens": 48348923.0,
      "step": 22350
    },
    {
      "entropy": 6.479787588119507,
      "epoch": 2.392209320990957,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044330487780202144,
      "loss": 5.7985,
      "mean_token_accuracy": 0.18304580748081206,
      "num_tokens": 48359708.0,
      "step": 22355
    },
    {
      "entropy": 6.463601684570312,
      "epoch": 2.3927443950987213,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044327950087452895,
      "loss": 5.8532,
      "mean_token_accuracy": 0.1646541878581047,
      "num_tokens": 48370497.0,
      "step": 22360
    },
    {
      "entropy": 6.409730672836304,
      "epoch": 2.393279469206485,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004432541190879895,
      "loss": 5.8595,
      "mean_token_accuracy": 0.1616755172610283,
      "num_tokens": 48381020.0,
      "step": 22365
    },
    {
      "entropy": 6.494356727600097,
      "epoch": 2.393814543314249,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044322873244313603,
      "loss": 5.8193,
      "mean_token_accuracy": 0.16827249974012376,
      "num_tokens": 48391696.0,
      "step": 22370
    },
    {
      "entropy": 6.481193399429321,
      "epoch": 2.394349617422013,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044320334094070154,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16706612557172776,
      "num_tokens": 48402403.0,
      "step": 22375
    },
    {
      "entropy": 6.449653673171997,
      "epoch": 2.394884691529777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004431779445814192,
      "loss": 5.8634,
      "mean_token_accuracy": 0.17364323884248734,
      "num_tokens": 48413579.0,
      "step": 22380
    },
    {
      "entropy": 6.531477737426758,
      "epoch": 2.395419765637541,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044315254336602236,
      "loss": 5.9125,
      "mean_token_accuracy": 0.15922873914241792,
      "num_tokens": 48423814.0,
      "step": 22385
    },
    {
      "entropy": 6.526466417312622,
      "epoch": 2.3959548397453045,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044312713729524443,
      "loss": 5.9886,
      "mean_token_accuracy": 0.15871059894561768,
      "num_tokens": 48433937.0,
      "step": 22390
    },
    {
      "entropy": 6.502141761779785,
      "epoch": 2.3964899138530686,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044310172636981905,
      "loss": 5.8495,
      "mean_token_accuracy": 0.1707923024892807,
      "num_tokens": 48444397.0,
      "step": 22395
    },
    {
      "entropy": 6.51025710105896,
      "epoch": 2.3970249879608327,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004430763105904799,
      "loss": 5.7385,
      "mean_token_accuracy": 0.17507847547531127,
      "num_tokens": 48454566.0,
      "step": 22400
    },
    {
      "entropy": 6.434056806564331,
      "epoch": 2.3975600620685964,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000443050889957961,
      "loss": 5.7661,
      "mean_token_accuracy": 0.1734343945980072,
      "num_tokens": 48465129.0,
      "step": 22405
    },
    {
      "entropy": 6.4193846702575685,
      "epoch": 2.3980951361763605,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044302546447299625,
      "loss": 5.8135,
      "mean_token_accuracy": 0.1669488787651062,
      "num_tokens": 48475793.0,
      "step": 22410
    },
    {
      "entropy": 6.446265649795532,
      "epoch": 2.398630210284124,
      "grad_norm": 1.125,
      "learning_rate": 0.0004430000341363198,
      "loss": 5.7449,
      "mean_token_accuracy": 0.1786079078912735,
      "num_tokens": 48486543.0,
      "step": 22415
    },
    {
      "entropy": 6.522268915176392,
      "epoch": 2.3991652843918883,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004429745989486659,
      "loss": 5.9096,
      "mean_token_accuracy": 0.16936013698577881,
      "num_tokens": 48498426.0,
      "step": 22420
    },
    {
      "entropy": 6.488598775863648,
      "epoch": 2.3997003584996524,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004429491589107691,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1713315188884735,
      "num_tokens": 48508763.0,
      "step": 22425
    },
    {
      "entropy": 6.516184091567993,
      "epoch": 2.400235432607416,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004429237140233639,
      "loss": 5.9056,
      "mean_token_accuracy": 0.16344597786664963,
      "num_tokens": 48519112.0,
      "step": 22430
    },
    {
      "entropy": 6.498443126678467,
      "epoch": 2.40077050671518,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004428982642871852,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1734997183084488,
      "num_tokens": 48530341.0,
      "step": 22435
    },
    {
      "entropy": 6.436092567443848,
      "epoch": 2.401305580822944,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004428728097029676,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17183464914560317,
      "num_tokens": 48541962.0,
      "step": 22440
    },
    {
      "entropy": 6.541606044769287,
      "epoch": 2.401840654930708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044284735027144624,
      "loss": 5.9321,
      "mean_token_accuracy": 0.15864285826683044,
      "num_tokens": 48554206.0,
      "step": 22445
    },
    {
      "entropy": 6.438163900375367,
      "epoch": 2.402375729038472,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004428218859933562,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1672838509082794,
      "num_tokens": 48563919.0,
      "step": 22450
    },
    {
      "entropy": 6.45432710647583,
      "epoch": 2.4029108031462356,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044279641686943277,
      "loss": 5.8894,
      "mean_token_accuracy": 0.17314953804016114,
      "num_tokens": 48575388.0,
      "step": 22455
    },
    {
      "entropy": 6.516183996200562,
      "epoch": 2.4034458772539997,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00044277094290041144,
      "loss": 5.8593,
      "mean_token_accuracy": 0.1699663668870926,
      "num_tokens": 48587492.0,
      "step": 22460
    },
    {
      "entropy": 6.399788665771484,
      "epoch": 2.4039809513617634,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044274546408702763,
      "loss": 5.686,
      "mean_token_accuracy": 0.18751525282859802,
      "num_tokens": 48598757.0,
      "step": 22465
    },
    {
      "entropy": 6.451042604446411,
      "epoch": 2.4045160254695275,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044271998043001723,
      "loss": 5.8465,
      "mean_token_accuracy": 0.16941637694835662,
      "num_tokens": 48609912.0,
      "step": 22470
    },
    {
      "entropy": 6.536119508743286,
      "epoch": 2.4050510995772916,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004426944919301158,
      "loss": 5.9616,
      "mean_token_accuracy": 0.1640952482819557,
      "num_tokens": 48621614.0,
      "step": 22475
    },
    {
      "entropy": 6.461690616607666,
      "epoch": 2.4055861736850552,
      "grad_norm": 1.015625,
      "learning_rate": 0.00044266899858805964,
      "loss": 5.8549,
      "mean_token_accuracy": 0.17188275158405303,
      "num_tokens": 48632539.0,
      "step": 22480
    },
    {
      "entropy": 6.568011331558227,
      "epoch": 2.4061212477928193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044264350040458463,
      "loss": 5.9556,
      "mean_token_accuracy": 0.16254207491874695,
      "num_tokens": 48644157.0,
      "step": 22485
    },
    {
      "entropy": 6.5479896068573,
      "epoch": 2.4066563219005834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044261799738042713,
      "loss": 5.9406,
      "mean_token_accuracy": 0.16438670307397843,
      "num_tokens": 48654928.0,
      "step": 22490
    },
    {
      "entropy": 6.547058248519898,
      "epoch": 2.407191396008347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004425924895163235,
      "loss": 5.87,
      "mean_token_accuracy": 0.17153672128915787,
      "num_tokens": 48665183.0,
      "step": 22495
    },
    {
      "entropy": 6.450140190124512,
      "epoch": 2.407726470116111,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044256697681301034,
      "loss": 5.8193,
      "mean_token_accuracy": 0.17480300962924958,
      "num_tokens": 48675233.0,
      "step": 22500
    },
    {
      "entropy": 6.440717506408691,
      "epoch": 2.408261544223875,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004425414592712242,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16418114304542542,
      "num_tokens": 48685881.0,
      "step": 22505
    },
    {
      "entropy": 6.495183658599854,
      "epoch": 2.408796618331639,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044251593689170196,
      "loss": 5.8992,
      "mean_token_accuracy": 0.16176283210515977,
      "num_tokens": 48696402.0,
      "step": 22510
    },
    {
      "entropy": 6.4499585151672365,
      "epoch": 2.409331692439403,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044249040967518063,
      "loss": 5.9187,
      "mean_token_accuracy": 0.16743779182434082,
      "num_tokens": 48708041.0,
      "step": 22515
    },
    {
      "entropy": 6.488186502456665,
      "epoch": 2.4098667665471667,
      "grad_norm": 1.125,
      "learning_rate": 0.00044246487762239717,
      "loss": 5.8169,
      "mean_token_accuracy": 0.17089991718530656,
      "num_tokens": 48718836.0,
      "step": 22520
    },
    {
      "entropy": 6.516123914718628,
      "epoch": 2.410401840654931,
      "grad_norm": 1.03125,
      "learning_rate": 0.000442439340734089,
      "loss": 5.8653,
      "mean_token_accuracy": 0.16580672413110734,
      "num_tokens": 48729115.0,
      "step": 22525
    },
    {
      "entropy": 6.4641398906707765,
      "epoch": 2.4109369147626944,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044241379901099334,
      "loss": 5.8449,
      "mean_token_accuracy": 0.16976746618747712,
      "num_tokens": 48740037.0,
      "step": 22530
    },
    {
      "entropy": 6.453396224975586,
      "epoch": 2.4114719888704585,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044238825245384775,
      "loss": 5.865,
      "mean_token_accuracy": 0.16769579350948333,
      "num_tokens": 48750894.0,
      "step": 22535
    },
    {
      "entropy": 6.505610513687134,
      "epoch": 2.4120070629782226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004423627010633899,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17161150127649308,
      "num_tokens": 48761566.0,
      "step": 22540
    },
    {
      "entropy": 6.340556240081787,
      "epoch": 2.4125421370859863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044233714484035765,
      "loss": 5.7603,
      "mean_token_accuracy": 0.17223970144987105,
      "num_tokens": 48772971.0,
      "step": 22545
    },
    {
      "entropy": 6.398986005783081,
      "epoch": 2.4130772111937504,
      "grad_norm": 0.96875,
      "learning_rate": 0.00044231158378548873,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16572082787752151,
      "num_tokens": 48784548.0,
      "step": 22550
    },
    {
      "entropy": 6.3498705387115475,
      "epoch": 2.4136122853015145,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004422860178995215,
      "loss": 5.6902,
      "mean_token_accuracy": 0.17637497186660767,
      "num_tokens": 48795228.0,
      "step": 22555
    },
    {
      "entropy": 6.453208160400391,
      "epoch": 2.414147359409278,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044226044718319383,
      "loss": 5.802,
      "mean_token_accuracy": 0.17268977761268617,
      "num_tokens": 48806497.0,
      "step": 22560
    },
    {
      "entropy": 6.496035861968994,
      "epoch": 2.4146824335170423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004422348716372444,
      "loss": 5.8365,
      "mean_token_accuracy": 0.17173880487680435,
      "num_tokens": 48816402.0,
      "step": 22565
    },
    {
      "entropy": 6.5258567333221436,
      "epoch": 2.415217507624806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004422092912624115,
      "loss": 5.7837,
      "mean_token_accuracy": 0.171484375,
      "num_tokens": 48826151.0,
      "step": 22570
    },
    {
      "entropy": 6.4531769275665285,
      "epoch": 2.41575258173257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044218370605943376,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16923686861991882,
      "num_tokens": 48836458.0,
      "step": 22575
    },
    {
      "entropy": 6.47174825668335,
      "epoch": 2.4162876558403337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044215811602905003,
      "loss": 5.8687,
      "mean_token_accuracy": 0.16414408683776854,
      "num_tokens": 48845863.0,
      "step": 22580
    },
    {
      "entropy": 6.456540298461914,
      "epoch": 2.4168227299480978,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044213252117199914,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16434946060180664,
      "num_tokens": 48856453.0,
      "step": 22585
    },
    {
      "entropy": 6.487018013000489,
      "epoch": 2.417357804055862,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044210692148902034,
      "loss": 5.8218,
      "mean_token_accuracy": 0.16543214172124862,
      "num_tokens": 48867013.0,
      "step": 22590
    },
    {
      "entropy": 6.474529647827149,
      "epoch": 2.4178928781636255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004420813169808525,
      "loss": 5.8458,
      "mean_token_accuracy": 0.1646002158522606,
      "num_tokens": 48876455.0,
      "step": 22595
    },
    {
      "entropy": 6.481237983703613,
      "epoch": 2.4184279522713896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004420557076482352,
      "loss": 5.8811,
      "mean_token_accuracy": 0.1657540649175644,
      "num_tokens": 48888223.0,
      "step": 22600
    },
    {
      "entropy": 6.506170654296875,
      "epoch": 2.4189630263791537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044203009349190784,
      "loss": 5.8683,
      "mean_token_accuracy": 0.16968309432268142,
      "num_tokens": 48899287.0,
      "step": 22605
    },
    {
      "entropy": 6.512514543533325,
      "epoch": 2.4194981004869174,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004420044745126099,
      "loss": 5.8896,
      "mean_token_accuracy": 0.16383208483457565,
      "num_tokens": 48910740.0,
      "step": 22610
    },
    {
      "entropy": 6.406011343002319,
      "epoch": 2.4200331745946815,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044197885071108127,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17091156244277955,
      "num_tokens": 48921925.0,
      "step": 22615
    },
    {
      "entropy": 6.498166418075561,
      "epoch": 2.420568248702445,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004419532220880619,
      "loss": 5.9664,
      "mean_token_accuracy": 0.15854809582233428,
      "num_tokens": 48933164.0,
      "step": 22620
    },
    {
      "entropy": 6.54214997291565,
      "epoch": 2.4211033228102092,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044192758864429157,
      "loss": 5.9216,
      "mean_token_accuracy": 0.16434021592140197,
      "num_tokens": 48943285.0,
      "step": 22625
    },
    {
      "entropy": 6.502472686767578,
      "epoch": 2.421638396917973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004419019503805106,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16002110242843628,
      "num_tokens": 48952852.0,
      "step": 22630
    },
    {
      "entropy": 6.48378734588623,
      "epoch": 2.422173471025737,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044187630729745924,
      "loss": 5.9432,
      "mean_token_accuracy": 0.15828063786029817,
      "num_tokens": 48964339.0,
      "step": 22635
    },
    {
      "entropy": 6.539627361297607,
      "epoch": 2.422708545133501,
      "grad_norm": 1.515625,
      "learning_rate": 0.00044185065939587805,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16828398853540422,
      "num_tokens": 48974754.0,
      "step": 22640
    },
    {
      "entropy": 6.528960132598877,
      "epoch": 2.4232436192412647,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044182500667650746,
      "loss": 5.8671,
      "mean_token_accuracy": 0.1715221032500267,
      "num_tokens": 48985596.0,
      "step": 22645
    },
    {
      "entropy": 6.541462659835815,
      "epoch": 2.423778693349029,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004417993491400882,
      "loss": 5.8999,
      "mean_token_accuracy": 0.16636165082454682,
      "num_tokens": 48997400.0,
      "step": 22650
    },
    {
      "entropy": 6.468556594848633,
      "epoch": 2.424313767456793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044177368678736126,
      "loss": 5.8256,
      "mean_token_accuracy": 0.16837923973798752,
      "num_tokens": 49007525.0,
      "step": 22655
    },
    {
      "entropy": 6.4439270973205565,
      "epoch": 2.4248488415645566,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00044174801961906744,
      "loss": 5.8561,
      "mean_token_accuracy": 0.17125612497329712,
      "num_tokens": 49018205.0,
      "step": 22660
    },
    {
      "entropy": 6.523239088058472,
      "epoch": 2.4253839156723207,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044172234763594805,
      "loss": 5.8897,
      "mean_token_accuracy": 0.16048525124788285,
      "num_tokens": 49028760.0,
      "step": 22665
    },
    {
      "entropy": 6.512425374984741,
      "epoch": 2.4259189897800844,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004416966708387443,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16694998890161514,
      "num_tokens": 49040734.0,
      "step": 22670
    },
    {
      "entropy": 6.459326219558716,
      "epoch": 2.4264540638878485,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044167098922819764,
      "loss": 5.8128,
      "mean_token_accuracy": 0.16231640726327895,
      "num_tokens": 49051474.0,
      "step": 22675
    },
    {
      "entropy": 6.414677000045776,
      "epoch": 2.4269891379956126,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044164530280504954,
      "loss": 5.8232,
      "mean_token_accuracy": 0.1782454863190651,
      "num_tokens": 49061697.0,
      "step": 22680
    },
    {
      "entropy": 6.4463742733001705,
      "epoch": 2.427524212103376,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044161961157004176,
      "loss": 5.8739,
      "mean_token_accuracy": 0.16809922605752944,
      "num_tokens": 49072124.0,
      "step": 22685
    },
    {
      "entropy": 6.48224196434021,
      "epoch": 2.4280592862111403,
      "grad_norm": 1.140625,
      "learning_rate": 0.000441593915523916,
      "loss": 5.8558,
      "mean_token_accuracy": 0.16546070128679274,
      "num_tokens": 49082653.0,
      "step": 22690
    },
    {
      "entropy": 6.53511962890625,
      "epoch": 2.428594360318904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044156821466741447,
      "loss": 5.9327,
      "mean_token_accuracy": 0.15964449048042298,
      "num_tokens": 49092481.0,
      "step": 22695
    },
    {
      "entropy": 6.489990329742431,
      "epoch": 2.429129434426668,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004415425090012792,
      "loss": 5.9425,
      "mean_token_accuracy": 0.16099707335233687,
      "num_tokens": 49103189.0,
      "step": 22700
    },
    {
      "entropy": 6.520315885543823,
      "epoch": 2.429664508534432,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044151679852625226,
      "loss": 5.9035,
      "mean_token_accuracy": 0.1655298501253128,
      "num_tokens": 49114071.0,
      "step": 22705
    },
    {
      "entropy": 6.560464191436767,
      "epoch": 2.430199582642196,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004414910832430762,
      "loss": 5.8047,
      "mean_token_accuracy": 0.17160235196352006,
      "num_tokens": 49125521.0,
      "step": 22710
    },
    {
      "entropy": 6.550657320022583,
      "epoch": 2.43073465674996,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004414653631524935,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16226189136505126,
      "num_tokens": 49136162.0,
      "step": 22715
    },
    {
      "entropy": 6.531015682220459,
      "epoch": 2.431269730857724,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004414396382552468,
      "loss": 5.8461,
      "mean_token_accuracy": 0.1643773466348648,
      "num_tokens": 49146127.0,
      "step": 22720
    },
    {
      "entropy": 6.439239835739135,
      "epoch": 2.4318048049654877,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044141390855207904,
      "loss": 5.8076,
      "mean_token_accuracy": 0.17702715396881102,
      "num_tokens": 49156619.0,
      "step": 22725
    },
    {
      "entropy": 6.4639434814453125,
      "epoch": 2.4323398790732518,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044138817404373295,
      "loss": 5.9453,
      "mean_token_accuracy": 0.1595514804124832,
      "num_tokens": 49168578.0,
      "step": 22730
    },
    {
      "entropy": 6.569987106323242,
      "epoch": 2.4328749531810154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004413624347309517,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16808344274759293,
      "num_tokens": 49178845.0,
      "step": 22735
    },
    {
      "entropy": 6.532905197143554,
      "epoch": 2.4334100272887795,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044133669061447855,
      "loss": 5.8938,
      "mean_token_accuracy": 0.1677004560828209,
      "num_tokens": 49189067.0,
      "step": 22740
    },
    {
      "entropy": 6.511377382278442,
      "epoch": 2.433945101396543,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044131094169505694,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16293689906597136,
      "num_tokens": 49200368.0,
      "step": 22745
    },
    {
      "entropy": 6.451305484771728,
      "epoch": 2.4344801755043073,
      "grad_norm": 1.0,
      "learning_rate": 0.00044128518797343014,
      "loss": 5.8535,
      "mean_token_accuracy": 0.16226119995117189,
      "num_tokens": 49211630.0,
      "step": 22750
    },
    {
      "entropy": 6.5478840351104735,
      "epoch": 2.4350152496120714,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000441259429450342,
      "loss": 5.9134,
      "mean_token_accuracy": 0.16570124477148057,
      "num_tokens": 49223092.0,
      "step": 22755
    },
    {
      "entropy": 6.509630155563355,
      "epoch": 2.435550323719835,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044123366612653617,
      "loss": 5.8556,
      "mean_token_accuracy": 0.17087456881999968,
      "num_tokens": 49233963.0,
      "step": 22760
    },
    {
      "entropy": 6.55901608467102,
      "epoch": 2.436085397827599,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004412078980027565,
      "loss": 5.8987,
      "mean_token_accuracy": 0.16685618013143538,
      "num_tokens": 49243834.0,
      "step": 22765
    },
    {
      "entropy": 6.441530179977417,
      "epoch": 2.4366204719353632,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004411821250797472,
      "loss": 5.731,
      "mean_token_accuracy": 0.1730647176504135,
      "num_tokens": 49254274.0,
      "step": 22770
    },
    {
      "entropy": 6.425611400604248,
      "epoch": 2.437155546043127,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004411563473582524,
      "loss": 5.7849,
      "mean_token_accuracy": 0.17400356531143188,
      "num_tokens": 49264734.0,
      "step": 22775
    },
    {
      "entropy": 6.47057695388794,
      "epoch": 2.437690620150891,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004411305648390163,
      "loss": 5.8159,
      "mean_token_accuracy": 0.1675855726003647,
      "num_tokens": 49274708.0,
      "step": 22780
    },
    {
      "entropy": 6.454339408874512,
      "epoch": 2.4382256942586547,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004411047775227836,
      "loss": 5.8599,
      "mean_token_accuracy": 0.1728945940732956,
      "num_tokens": 49286318.0,
      "step": 22785
    },
    {
      "entropy": 6.552979421615601,
      "epoch": 2.4387607683664188,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00044107898541029885,
      "loss": 5.8838,
      "mean_token_accuracy": 0.16629405319690704,
      "num_tokens": 49296948.0,
      "step": 22790
    },
    {
      "entropy": 6.449427747726441,
      "epoch": 2.439295842474183,
      "grad_norm": 1.125,
      "learning_rate": 0.0004410531885023066,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16825269311666488,
      "num_tokens": 49307060.0,
      "step": 22795
    },
    {
      "entropy": 6.493971157073974,
      "epoch": 2.4398309165819465,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044102738679955195,
      "loss": 5.8583,
      "mean_token_accuracy": 0.16317052245140076,
      "num_tokens": 49318449.0,
      "step": 22800
    },
    {
      "entropy": 6.478823852539063,
      "epoch": 2.4403659906897106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004410015803027798,
      "loss": 5.8937,
      "mean_token_accuracy": 0.16687073558568954,
      "num_tokens": 49329567.0,
      "step": 22805
    },
    {
      "entropy": 6.522231769561768,
      "epoch": 2.4409010647974743,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044097576901273535,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1681257739663124,
      "num_tokens": 49340854.0,
      "step": 22810
    },
    {
      "entropy": 6.503978586196899,
      "epoch": 2.4414361389052384,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004409499529301639,
      "loss": 5.9063,
      "mean_token_accuracy": 0.16672515124082565,
      "num_tokens": 49352749.0,
      "step": 22815
    },
    {
      "entropy": 6.533674716949463,
      "epoch": 2.4419712130130025,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004409241320558109,
      "loss": 5.9093,
      "mean_token_accuracy": 0.1607213482260704,
      "num_tokens": 49364703.0,
      "step": 22820
    },
    {
      "entropy": 6.480649852752686,
      "epoch": 2.442506287120766,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044089830639042183,
      "loss": 5.7623,
      "mean_token_accuracy": 0.1690144956111908,
      "num_tokens": 49375873.0,
      "step": 22825
    },
    {
      "entropy": 6.474816513061524,
      "epoch": 2.44304136122853,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044087247593474255,
      "loss": 5.9266,
      "mean_token_accuracy": 0.1676350563764572,
      "num_tokens": 49386578.0,
      "step": 22830
    },
    {
      "entropy": 6.48452091217041,
      "epoch": 2.4435764353362943,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004408466406895188,
      "loss": 5.8592,
      "mean_token_accuracy": 0.1663139656186104,
      "num_tokens": 49398394.0,
      "step": 22835
    },
    {
      "entropy": 6.429463958740234,
      "epoch": 2.444111509444058,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004408208006554966,
      "loss": 5.7979,
      "mean_token_accuracy": 0.18173613101243974,
      "num_tokens": 49408758.0,
      "step": 22840
    },
    {
      "entropy": 6.484366178512573,
      "epoch": 2.444646583551822,
      "grad_norm": 1.125,
      "learning_rate": 0.0004407949558334221,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16850806027650833,
      "num_tokens": 49420447.0,
      "step": 22845
    },
    {
      "entropy": 6.497842788696289,
      "epoch": 2.4451816576595857,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044076910622404153,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1682829052209854,
      "num_tokens": 49430246.0,
      "step": 22850
    },
    {
      "entropy": 6.511043691635132,
      "epoch": 2.44571673176735,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00044074325182810135,
      "loss": 5.8323,
      "mean_token_accuracy": 0.1738967478275299,
      "num_tokens": 49441272.0,
      "step": 22855
    },
    {
      "entropy": 6.47623987197876,
      "epoch": 2.4462518058751135,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044071739264634803,
      "loss": 5.8885,
      "mean_token_accuracy": 0.1596525028347969,
      "num_tokens": 49452778.0,
      "step": 22860
    },
    {
      "entropy": 6.460806751251221,
      "epoch": 2.4467868799828776,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004406915286795283,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16810843199491501,
      "num_tokens": 49463594.0,
      "step": 22865
    },
    {
      "entropy": 6.481576633453369,
      "epoch": 2.4473219540906417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004406656599283889,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16139936894178392,
      "num_tokens": 49473649.0,
      "step": 22870
    },
    {
      "entropy": 6.459052658081054,
      "epoch": 2.4478570281984053,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004406397863936769,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16848415583372117,
      "num_tokens": 49485112.0,
      "step": 22875
    },
    {
      "entropy": 6.51873288154602,
      "epoch": 2.4483921023061694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004406139080761394,
      "loss": 5.9241,
      "mean_token_accuracy": 0.1656185731291771,
      "num_tokens": 49495455.0,
      "step": 22880
    },
    {
      "entropy": 6.467900228500366,
      "epoch": 2.4489271764139335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004405880249765235,
      "loss": 5.8651,
      "mean_token_accuracy": 0.17258595526218415,
      "num_tokens": 49506495.0,
      "step": 22885
    },
    {
      "entropy": 6.453693866729736,
      "epoch": 2.449462250521697,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044056213709557667,
      "loss": 5.8863,
      "mean_token_accuracy": 0.16435530483722688,
      "num_tokens": 49516384.0,
      "step": 22890
    },
    {
      "entropy": 6.453031063079834,
      "epoch": 2.4499973246294613,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044053624443404646,
      "loss": 5.8394,
      "mean_token_accuracy": 0.170610973238945,
      "num_tokens": 49527972.0,
      "step": 22895
    },
    {
      "entropy": 6.472983932495117,
      "epoch": 2.450532398737225,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004405103469926803,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16576625406742096,
      "num_tokens": 49540643.0,
      "step": 22900
    },
    {
      "entropy": 6.4607383728027346,
      "epoch": 2.451067472844989,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044048444477222626,
      "loss": 5.7882,
      "mean_token_accuracy": 0.1727369874715805,
      "num_tokens": 49550968.0,
      "step": 22905
    },
    {
      "entropy": 6.477924823760986,
      "epoch": 2.451602546952753,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004404585377734321,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16792201846837998,
      "num_tokens": 49562381.0,
      "step": 22910
    },
    {
      "entropy": 6.4731768608093265,
      "epoch": 2.452137621060517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004404326259970459,
      "loss": 5.8889,
      "mean_token_accuracy": 0.1646314635872841,
      "num_tokens": 49573414.0,
      "step": 22915
    },
    {
      "entropy": 6.490538692474365,
      "epoch": 2.452672695168281,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004404067094438159,
      "loss": 5.8858,
      "mean_token_accuracy": 0.17001922726631163,
      "num_tokens": 49583772.0,
      "step": 22920
    },
    {
      "entropy": 6.503818321228027,
      "epoch": 2.4532077692760446,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004403807881144903,
      "loss": 5.9277,
      "mean_token_accuracy": 0.16005913764238358,
      "num_tokens": 49594751.0,
      "step": 22925
    },
    {
      "entropy": 6.458941459655762,
      "epoch": 2.4537428433838087,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044035486200981786,
      "loss": 5.8189,
      "mean_token_accuracy": 0.1683187425136566,
      "num_tokens": 49605048.0,
      "step": 22930
    },
    {
      "entropy": 6.517916297912597,
      "epoch": 2.4542779174915728,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044032893113054683,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1605648472905159,
      "num_tokens": 49616582.0,
      "step": 22935
    },
    {
      "entropy": 6.531847524642944,
      "epoch": 2.4548129915993364,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004403029954774263,
      "loss": 5.8409,
      "mean_token_accuracy": 0.16616763323545455,
      "num_tokens": 49626974.0,
      "step": 22940
    },
    {
      "entropy": 6.453983163833618,
      "epoch": 2.4553480657071005,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004402770550512049,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16209321618080139,
      "num_tokens": 49637580.0,
      "step": 22945
    },
    {
      "entropy": 6.542248630523682,
      "epoch": 2.455883139814864,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402511098526318,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16500818878412246,
      "num_tokens": 49648543.0,
      "step": 22950
    },
    {
      "entropy": 6.529124689102173,
      "epoch": 2.4564182139226283,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004402251598824561,
      "loss": 5.8897,
      "mean_token_accuracy": 0.1682015210390091,
      "num_tokens": 49659057.0,
      "step": 22955
    },
    {
      "entropy": 6.4719054222106935,
      "epoch": 2.4569532880303924,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004401992051414272,
      "loss": 5.8435,
      "mean_token_accuracy": 0.17100395858287812,
      "num_tokens": 49671226.0,
      "step": 22960
    },
    {
      "entropy": 6.475981760025024,
      "epoch": 2.457488362138156,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044017324563029435,
      "loss": 5.8551,
      "mean_token_accuracy": 0.17114123404026033,
      "num_tokens": 49682063.0,
      "step": 22965
    },
    {
      "entropy": 6.416031455993652,
      "epoch": 2.45802343624592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004401472813498074,
      "loss": 5.8138,
      "mean_token_accuracy": 0.1735602244734764,
      "num_tokens": 49693768.0,
      "step": 22970
    },
    {
      "entropy": 6.499839162826538,
      "epoch": 2.458558510353684,
      "grad_norm": 1.125,
      "learning_rate": 0.00044012131230071575,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16353389173746108,
      "num_tokens": 49703807.0,
      "step": 22975
    },
    {
      "entropy": 6.504425382614135,
      "epoch": 2.459093584461448,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004400953384837695,
      "loss": 5.8465,
      "mean_token_accuracy": 0.1641426682472229,
      "num_tokens": 49714945.0,
      "step": 22980
    },
    {
      "entropy": 6.48578143119812,
      "epoch": 2.459628658569212,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004400693598997185,
      "loss": 5.7831,
      "mean_token_accuracy": 0.17760445475578307,
      "num_tokens": 49727267.0,
      "step": 22985
    },
    {
      "entropy": 6.4551708698272705,
      "epoch": 2.4601637326769756,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044004337654931293,
      "loss": 5.8295,
      "mean_token_accuracy": 0.16553349792957306,
      "num_tokens": 49738370.0,
      "step": 22990
    },
    {
      "entropy": 6.508518648147583,
      "epoch": 2.4606988067847397,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044001738843330306,
      "loss": 5.8762,
      "mean_token_accuracy": 0.16452356725931166,
      "num_tokens": 49750113.0,
      "step": 22995
    },
    {
      "entropy": 6.495656681060791,
      "epoch": 2.461233880892504,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043999139555243936,
      "loss": 5.9431,
      "mean_token_accuracy": 0.158562371134758,
      "num_tokens": 49761163.0,
      "step": 23000
    },
    {
      "entropy": 6.486612987518311,
      "epoch": 2.4617689550002675,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004399653979074722,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16212462931871413,
      "num_tokens": 49771185.0,
      "step": 23005
    },
    {
      "entropy": 6.521358919143677,
      "epoch": 2.4623040291080316,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043993939549915245,
      "loss": 5.8621,
      "mean_token_accuracy": 0.16605806946754456,
      "num_tokens": 49781202.0,
      "step": 23010
    },
    {
      "entropy": 6.48941822052002,
      "epoch": 2.4628391032157952,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043991338832823074,
      "loss": 5.8659,
      "mean_token_accuracy": 0.166810604929924,
      "num_tokens": 49790971.0,
      "step": 23015
    },
    {
      "entropy": 6.454057264328003,
      "epoch": 2.4633741773235593,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004398873763954582,
      "loss": 5.8024,
      "mean_token_accuracy": 0.1702205717563629,
      "num_tokens": 49800942.0,
      "step": 23020
    },
    {
      "entropy": 6.431850385665894,
      "epoch": 2.463909251431323,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043986135970158573,
      "loss": 5.8059,
      "mean_token_accuracy": 0.1670064777135849,
      "num_tokens": 49812043.0,
      "step": 23025
    },
    {
      "entropy": 6.3762359619140625,
      "epoch": 2.464444325539087,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043983533824736476,
      "loss": 5.7905,
      "mean_token_accuracy": 0.18067671805620195,
      "num_tokens": 49823045.0,
      "step": 23030
    },
    {
      "entropy": 6.537246751785278,
      "epoch": 2.464979399646851,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004398093120335465,
      "loss": 5.9204,
      "mean_token_accuracy": 0.16623048931360246,
      "num_tokens": 49833014.0,
      "step": 23035
    },
    {
      "entropy": 6.492333602905274,
      "epoch": 2.465514473754615,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043978328106088254,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1679843544960022,
      "num_tokens": 49843171.0,
      "step": 23040
    },
    {
      "entropy": 6.460606861114502,
      "epoch": 2.466049547862379,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004397572453301246,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17469244301319123,
      "num_tokens": 49853772.0,
      "step": 23045
    },
    {
      "entropy": 6.475110483169556,
      "epoch": 2.466584621970143,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043973120484202416,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16731531769037247,
      "num_tokens": 49865379.0,
      "step": 23050
    },
    {
      "entropy": 6.5002028465271,
      "epoch": 2.4671196960779067,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043970515959733343,
      "loss": 5.9103,
      "mean_token_accuracy": 0.1629177913069725,
      "num_tokens": 49876453.0,
      "step": 23055
    },
    {
      "entropy": 6.4282502174377445,
      "epoch": 2.467654770185671,
      "grad_norm": 1.0,
      "learning_rate": 0.00043967910959680435,
      "loss": 5.7824,
      "mean_token_accuracy": 0.18048057556152344,
      "num_tokens": 49886675.0,
      "step": 23060
    },
    {
      "entropy": 6.4419440746307375,
      "epoch": 2.4681898442934345,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004396530548411891,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1704205647110939,
      "num_tokens": 49896946.0,
      "step": 23065
    },
    {
      "entropy": 6.519670104980468,
      "epoch": 2.4687249184011986,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004396269953312401,
      "loss": 5.9317,
      "mean_token_accuracy": 0.16057479977607728,
      "num_tokens": 49906989.0,
      "step": 23070
    },
    {
      "entropy": 6.404631805419922,
      "epoch": 2.4692599925089627,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043960093106770964,
      "loss": 5.7204,
      "mean_token_accuracy": 0.16963956505060196,
      "num_tokens": 49917861.0,
      "step": 23075
    },
    {
      "entropy": 6.453144264221192,
      "epoch": 2.4697950666167263,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043957486205135047,
      "loss": 5.8801,
      "mean_token_accuracy": 0.1627962589263916,
      "num_tokens": 49930529.0,
      "step": 23080
    },
    {
      "entropy": 6.472833919525146,
      "epoch": 2.4703301407244904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004395487882829153,
      "loss": 5.8092,
      "mean_token_accuracy": 0.16989320367574692,
      "num_tokens": 49943067.0,
      "step": 23085
    },
    {
      "entropy": 6.478387069702149,
      "epoch": 2.470865214832254,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043952270976315707,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16346391588449477,
      "num_tokens": 49953648.0,
      "step": 23090
    },
    {
      "entropy": 6.548659944534302,
      "epoch": 2.471400288940018,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043949662649282853,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15896454453468323,
      "num_tokens": 49963571.0,
      "step": 23095
    },
    {
      "entropy": 6.468227434158325,
      "epoch": 2.4719353630477823,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043947053847268313,
      "loss": 5.8401,
      "mean_token_accuracy": 0.17807146161794662,
      "num_tokens": 49974627.0,
      "step": 23100
    },
    {
      "entropy": 6.53054633140564,
      "epoch": 2.472470437155546,
      "grad_norm": 1.09375,
      "learning_rate": 0.000439444445703474,
      "loss": 5.9066,
      "mean_token_accuracy": 0.16694648414850236,
      "num_tokens": 49986659.0,
      "step": 23105
    },
    {
      "entropy": 6.4633934020996096,
      "epoch": 2.47300551126331,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043941834818595464,
      "loss": 5.838,
      "mean_token_accuracy": 0.17352975010871888,
      "num_tokens": 49997099.0,
      "step": 23110
    },
    {
      "entropy": 6.489995431900025,
      "epoch": 2.473540585371074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043939224592087864,
      "loss": 5.8421,
      "mean_token_accuracy": 0.1709866926074028,
      "num_tokens": 50008148.0,
      "step": 23115
    },
    {
      "entropy": 6.5418132781982425,
      "epoch": 2.474075659478838,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043936613890899955,
      "loss": 5.9653,
      "mean_token_accuracy": 0.15835631787776946,
      "num_tokens": 50019335.0,
      "step": 23120
    },
    {
      "entropy": 6.479639959335327,
      "epoch": 2.474610733586602,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004393400271510713,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16992992013692856,
      "num_tokens": 50029513.0,
      "step": 23125
    },
    {
      "entropy": 6.535048723220825,
      "epoch": 2.4751458076943655,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043931391064784786,
      "loss": 5.9144,
      "mean_token_accuracy": 0.1616342321038246,
      "num_tokens": 50039353.0,
      "step": 23130
    },
    {
      "entropy": 6.465950059890747,
      "epoch": 2.4756808818021296,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043928778940008334,
      "loss": 5.9115,
      "mean_token_accuracy": 0.1647768422961235,
      "num_tokens": 50050260.0,
      "step": 23135
    },
    {
      "entropy": 6.423762702941895,
      "epoch": 2.4762159559098933,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043926166340853203,
      "loss": 5.7365,
      "mean_token_accuracy": 0.16860306039452552,
      "num_tokens": 50061384.0,
      "step": 23140
    },
    {
      "entropy": 6.455498504638672,
      "epoch": 2.4767510300176574,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004392355326739483,
      "loss": 5.8085,
      "mean_token_accuracy": 0.1730794668197632,
      "num_tokens": 50071984.0,
      "step": 23145
    },
    {
      "entropy": 6.4545611381530765,
      "epoch": 2.4772861041254215,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043920939719708656,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17167263925075532,
      "num_tokens": 50083225.0,
      "step": 23150
    },
    {
      "entropy": 6.467261219024659,
      "epoch": 2.477821178233185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004391832569787016,
      "loss": 5.821,
      "mean_token_accuracy": 0.1743471547961235,
      "num_tokens": 50093240.0,
      "step": 23155
    },
    {
      "entropy": 6.454530858993531,
      "epoch": 2.4783562523409493,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004391571120195481,
      "loss": 5.8752,
      "mean_token_accuracy": 0.16499096155166626,
      "num_tokens": 50103870.0,
      "step": 23160
    },
    {
      "entropy": 6.443850421905518,
      "epoch": 2.4788913264487134,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004391309623203811,
      "loss": 5.9232,
      "mean_token_accuracy": 0.16234916001558303,
      "num_tokens": 50115231.0,
      "step": 23165
    },
    {
      "entropy": 6.499606418609619,
      "epoch": 2.479426400556477,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004391048078819556,
      "loss": 5.7617,
      "mean_token_accuracy": 0.17446935325860977,
      "num_tokens": 50125592.0,
      "step": 23170
    },
    {
      "entropy": 6.446830749511719,
      "epoch": 2.479961474664241,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004390786487050269,
      "loss": 5.771,
      "mean_token_accuracy": 0.17092553973197938,
      "num_tokens": 50136713.0,
      "step": 23175
    },
    {
      "entropy": 6.483169460296631,
      "epoch": 2.4804965487720048,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004390524847903502,
      "loss": 5.869,
      "mean_token_accuracy": 0.17030581682920456,
      "num_tokens": 50148645.0,
      "step": 23180
    },
    {
      "entropy": 6.514463281631469,
      "epoch": 2.481031622879769,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004390263161386811,
      "loss": 5.8243,
      "mean_token_accuracy": 0.17060547024011613,
      "num_tokens": 50159973.0,
      "step": 23185
    },
    {
      "entropy": 6.505413341522217,
      "epoch": 2.481566696987533,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043900014275077517,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16263652890920638,
      "num_tokens": 50170591.0,
      "step": 23190
    },
    {
      "entropy": 6.46008152961731,
      "epoch": 2.4821017710952966,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043897396462738817,
      "loss": 5.8732,
      "mean_token_accuracy": 0.1715746968984604,
      "num_tokens": 50182414.0,
      "step": 23195
    },
    {
      "entropy": 6.563867664337158,
      "epoch": 2.4826368452030607,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004389477817692759,
      "loss": 5.9128,
      "mean_token_accuracy": 0.1664452701807022,
      "num_tokens": 50192567.0,
      "step": 23200
    },
    {
      "entropy": 6.580123376846314,
      "epoch": 2.4831719193108244,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004389215941771946,
      "loss": 5.8793,
      "mean_token_accuracy": 0.16621657609939575,
      "num_tokens": 50203142.0,
      "step": 23205
    },
    {
      "entropy": 6.494157886505127,
      "epoch": 2.4837069934185885,
      "grad_norm": 1.125,
      "learning_rate": 0.00043889540185190024,
      "loss": 5.8762,
      "mean_token_accuracy": 0.15906044244766235,
      "num_tokens": 50214854.0,
      "step": 23210
    },
    {
      "entropy": 6.495627737045288,
      "epoch": 2.4842420675263526,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004388692047941492,
      "loss": 5.808,
      "mean_token_accuracy": 0.16576708257198333,
      "num_tokens": 50225999.0,
      "step": 23215
    },
    {
      "entropy": 6.579906892776489,
      "epoch": 2.4847771416341162,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004388430030046978,
      "loss": 5.926,
      "mean_token_accuracy": 0.15851307660341263,
      "num_tokens": 50236657.0,
      "step": 23220
    },
    {
      "entropy": 6.559229993820191,
      "epoch": 2.4853122157418803,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004388167964843029,
      "loss": 5.8628,
      "mean_token_accuracy": 0.16428677439689637,
      "num_tokens": 50246457.0,
      "step": 23225
    },
    {
      "entropy": 6.49316439628601,
      "epoch": 2.485847289849644,
      "grad_norm": 1.125,
      "learning_rate": 0.0004387905852337209,
      "loss": 5.8749,
      "mean_token_accuracy": 0.16425037980079651,
      "num_tokens": 50257020.0,
      "step": 23230
    },
    {
      "entropy": 6.519448566436767,
      "epoch": 2.486382363957408,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043876436925370873,
      "loss": 5.7721,
      "mean_token_accuracy": 0.16868528574705124,
      "num_tokens": 50267173.0,
      "step": 23235
    },
    {
      "entropy": 6.463965320587159,
      "epoch": 2.486917438065172,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004387381485450235,
      "loss": 5.9082,
      "mean_token_accuracy": 0.16605817377567292,
      "num_tokens": 50278352.0,
      "step": 23240
    },
    {
      "entropy": 6.513075685501098,
      "epoch": 2.487452512172936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043871192310842216,
      "loss": 5.9641,
      "mean_token_accuracy": 0.15771486163139342,
      "num_tokens": 50290301.0,
      "step": 23245
    },
    {
      "entropy": 6.573231172561646,
      "epoch": 2.4879875862807,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004386856929446621,
      "loss": 5.9394,
      "mean_token_accuracy": 0.1569827526807785,
      "num_tokens": 50301707.0,
      "step": 23250
    },
    {
      "entropy": 6.54427661895752,
      "epoch": 2.4885226603884636,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004386594580545006,
      "loss": 5.8717,
      "mean_token_accuracy": 0.16798582822084426,
      "num_tokens": 50313334.0,
      "step": 23255
    },
    {
      "entropy": 6.536630773544312,
      "epoch": 2.4890577344962277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004386332184386953,
      "loss": 5.8651,
      "mean_token_accuracy": 0.16764302104711531,
      "num_tokens": 50324228.0,
      "step": 23260
    },
    {
      "entropy": 6.4105775356292725,
      "epoch": 2.489592808603992,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043860697409800383,
      "loss": 5.6936,
      "mean_token_accuracy": 0.17700487226247788,
      "num_tokens": 50335001.0,
      "step": 23265
    },
    {
      "entropy": 6.435513496398926,
      "epoch": 2.4901278827117554,
      "grad_norm": 1.125,
      "learning_rate": 0.00043858072503318393,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16983411461114883,
      "num_tokens": 50346415.0,
      "step": 23270
    },
    {
      "entropy": 6.467947006225586,
      "epoch": 2.4906629568195195,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043855447124499356,
      "loss": 5.9101,
      "mean_token_accuracy": 0.16489588916301728,
      "num_tokens": 50357625.0,
      "step": 23275
    },
    {
      "entropy": 6.481111764907837,
      "epoch": 2.4911980309272836,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043852821273419087,
      "loss": 5.7856,
      "mean_token_accuracy": 0.17845466881990432,
      "num_tokens": 50367672.0,
      "step": 23280
    },
    {
      "entropy": 6.494016313552857,
      "epoch": 2.4917331050350473,
      "grad_norm": 1.09375,
      "learning_rate": 0.000438501949501534,
      "loss": 5.8181,
      "mean_token_accuracy": 0.17169232666492462,
      "num_tokens": 50378220.0,
      "step": 23285
    },
    {
      "entropy": 6.505207967758179,
      "epoch": 2.4922681791428114,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043847568154778134,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1742689609527588,
      "num_tokens": 50389517.0,
      "step": 23290
    },
    {
      "entropy": 6.463148593902588,
      "epoch": 2.492803253250575,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004384494088736913,
      "loss": 5.8629,
      "mean_token_accuracy": 0.16740390062332153,
      "num_tokens": 50401155.0,
      "step": 23295
    },
    {
      "entropy": 6.435260868072509,
      "epoch": 2.493338327358339,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004384231314800226,
      "loss": 5.8424,
      "mean_token_accuracy": 0.17122751325368882,
      "num_tokens": 50412609.0,
      "step": 23300
    },
    {
      "entropy": 6.5483293533325195,
      "epoch": 2.493873401466103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004383968493675339,
      "loss": 5.8601,
      "mean_token_accuracy": 0.164106085896492,
      "num_tokens": 50422665.0,
      "step": 23305
    },
    {
      "entropy": 6.543108892440796,
      "epoch": 2.494408475573867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004383705625369841,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16341825127601622,
      "num_tokens": 50434012.0,
      "step": 23310
    },
    {
      "entropy": 6.510788774490356,
      "epoch": 2.494943549681631,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004383442709891323,
      "loss": 5.8608,
      "mean_token_accuracy": 0.1633867546916008,
      "num_tokens": 50444270.0,
      "step": 23315
    },
    {
      "entropy": 6.421707487106323,
      "epoch": 2.4954786237893947,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043831797472473763,
      "loss": 5.7294,
      "mean_token_accuracy": 0.17748444378376008,
      "num_tokens": 50455236.0,
      "step": 23320
    },
    {
      "entropy": 6.465208101272583,
      "epoch": 2.4960136978971588,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004382916737445594,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16659135967493058,
      "num_tokens": 50466431.0,
      "step": 23325
    },
    {
      "entropy": 6.553368854522705,
      "epoch": 2.496548772004923,
      "grad_norm": 1.046875,
      "learning_rate": 0.000438265368049357,
      "loss": 5.9322,
      "mean_token_accuracy": 0.15760879963636398,
      "num_tokens": 50477211.0,
      "step": 23330
    },
    {
      "entropy": 6.502098226547242,
      "epoch": 2.4970838461126865,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043823905763989003,
      "loss": 5.8243,
      "mean_token_accuracy": 0.1653080999851227,
      "num_tokens": 50488668.0,
      "step": 23335
    },
    {
      "entropy": 6.470821714401245,
      "epoch": 2.4976189202204506,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043821274251691823,
      "loss": 5.7916,
      "mean_token_accuracy": 0.16735200881958007,
      "num_tokens": 50498031.0,
      "step": 23340
    },
    {
      "entropy": 6.414916896820069,
      "epoch": 2.4981539943282143,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043818642268120153,
      "loss": 5.8141,
      "mean_token_accuracy": 0.17081369012594222,
      "num_tokens": 50508567.0,
      "step": 23345
    },
    {
      "entropy": 6.4382565975189205,
      "epoch": 2.4986890684359784,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043816009813349964,
      "loss": 5.7921,
      "mean_token_accuracy": 0.1771507129073143,
      "num_tokens": 50519057.0,
      "step": 23350
    },
    {
      "entropy": 6.505366039276123,
      "epoch": 2.4992241425437425,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000438133768874573,
      "loss": 5.8446,
      "mean_token_accuracy": 0.1703565910458565,
      "num_tokens": 50530572.0,
      "step": 23355
    },
    {
      "entropy": 6.499741411209106,
      "epoch": 2.499759216651506,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043810743490518155,
      "loss": 5.9023,
      "mean_token_accuracy": 0.16066293567419052,
      "num_tokens": 50541195.0,
      "step": 23360
    },
    {
      "entropy": 6.536323404312133,
      "epoch": 2.5002942907592702,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043808109622608596,
      "loss": 5.9632,
      "mean_token_accuracy": 0.15797372609376908,
      "num_tokens": 50552655.0,
      "step": 23365
    },
    {
      "entropy": 6.53928165435791,
      "epoch": 2.500829364867034,
      "grad_norm": 1.125,
      "learning_rate": 0.00043805475283804657,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16533986032009124,
      "num_tokens": 50562848.0,
      "step": 23370
    },
    {
      "entropy": 6.492247247695923,
      "epoch": 2.501364438974798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004380284047418241,
      "loss": 5.8513,
      "mean_token_accuracy": 0.17432734817266465,
      "num_tokens": 50572690.0,
      "step": 23375
    },
    {
      "entropy": 6.413250398635864,
      "epoch": 2.501899513082562,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004380020519381794,
      "loss": 5.7835,
      "mean_token_accuracy": 0.1748422145843506,
      "num_tokens": 50583826.0,
      "step": 23380
    },
    {
      "entropy": 6.402146339416504,
      "epoch": 2.5024345871903257,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043797569442787334,
      "loss": 5.7713,
      "mean_token_accuracy": 0.1698264643549919,
      "num_tokens": 50595148.0,
      "step": 23385
    },
    {
      "entropy": 6.495045614242554,
      "epoch": 2.50296966129809,
      "grad_norm": 1.0625,
      "learning_rate": 0.000437949332211667,
      "loss": 5.9574,
      "mean_token_accuracy": 0.15920255035161973,
      "num_tokens": 50606568.0,
      "step": 23390
    },
    {
      "entropy": 6.586487197875977,
      "epoch": 2.503504735405854,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043792296529032156,
      "loss": 5.8805,
      "mean_token_accuracy": 0.1594464123249054,
      "num_tokens": 50618144.0,
      "step": 23395
    },
    {
      "entropy": 6.5810747146606445,
      "epoch": 2.5040398095136176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004378965936645984,
      "loss": 5.8354,
      "mean_token_accuracy": 0.1714294284582138,
      "num_tokens": 50629721.0,
      "step": 23400
    },
    {
      "entropy": 6.484992456436157,
      "epoch": 2.5045748836213817,
      "grad_norm": 1.125,
      "learning_rate": 0.00043787021733525904,
      "loss": 5.8904,
      "mean_token_accuracy": 0.16467204093933105,
      "num_tokens": 50640482.0,
      "step": 23405
    },
    {
      "entropy": 6.499088335037231,
      "epoch": 2.5051099577291454,
      "grad_norm": 1.09375,
      "learning_rate": 0.000437843836303065,
      "loss": 5.9406,
      "mean_token_accuracy": 0.15610195249319075,
      "num_tokens": 50652261.0,
      "step": 23410
    },
    {
      "entropy": 6.4427845001220705,
      "epoch": 2.5056450318369095,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004378174505687781,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17099595218896865,
      "num_tokens": 50662397.0,
      "step": 23415
    },
    {
      "entropy": 6.4348719120025635,
      "epoch": 2.506180105944673,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004377910601331601,
      "loss": 5.8144,
      "mean_token_accuracy": 0.16891078203916549,
      "num_tokens": 50673494.0,
      "step": 23420
    },
    {
      "entropy": 6.387500429153443,
      "epoch": 2.506715180052437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004377646649969733,
      "loss": 5.7797,
      "mean_token_accuracy": 0.17810856252908708,
      "num_tokens": 50683963.0,
      "step": 23425
    },
    {
      "entropy": 6.466680335998535,
      "epoch": 2.5072502541602013,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043773826516097945,
      "loss": 5.8581,
      "mean_token_accuracy": 0.1665113940834999,
      "num_tokens": 50694449.0,
      "step": 23430
    },
    {
      "entropy": 6.5149421215057375,
      "epoch": 2.507785328267965,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043771186062594114,
      "loss": 5.8266,
      "mean_token_accuracy": 0.17603527307510375,
      "num_tokens": 50705877.0,
      "step": 23435
    },
    {
      "entropy": 6.488853406906128,
      "epoch": 2.508320402375729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043768545139262076,
      "loss": 5.857,
      "mean_token_accuracy": 0.1684902250766754,
      "num_tokens": 50716581.0,
      "step": 23440
    },
    {
      "entropy": 6.472997760772705,
      "epoch": 2.508855476483493,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004376590374617809,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16627235412597657,
      "num_tokens": 50727948.0,
      "step": 23445
    },
    {
      "entropy": 6.344044589996338,
      "epoch": 2.509390550591257,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043763261883418396,
      "loss": 5.7041,
      "mean_token_accuracy": 0.1883593261241913,
      "num_tokens": 50739189.0,
      "step": 23450
    },
    {
      "entropy": 6.441882562637329,
      "epoch": 2.509925624699021,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004376061955105932,
      "loss": 5.839,
      "mean_token_accuracy": 0.1650082141160965,
      "num_tokens": 50749439.0,
      "step": 23455
    },
    {
      "entropy": 6.504968690872192,
      "epoch": 2.5104606988067846,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043757976749177137,
      "loss": 5.9308,
      "mean_token_accuracy": 0.16332645416259767,
      "num_tokens": 50761463.0,
      "step": 23460
    },
    {
      "entropy": 6.537862491607666,
      "epoch": 2.5109957729145487,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004375533347784816,
      "loss": 5.9308,
      "mean_token_accuracy": 0.16367720663547516,
      "num_tokens": 50771359.0,
      "step": 23465
    },
    {
      "entropy": 6.488256931304932,
      "epoch": 2.5115308470223123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043752689737148705,
      "loss": 5.8674,
      "mean_token_accuracy": 0.16099714636802673,
      "num_tokens": 50781669.0,
      "step": 23470
    },
    {
      "entropy": 6.502426242828369,
      "epoch": 2.5120659211300764,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043750045527155126,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16813524216413497,
      "num_tokens": 50792217.0,
      "step": 23475
    },
    {
      "entropy": 6.504700946807861,
      "epoch": 2.5126009952378405,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004374740084794376,
      "loss": 5.8766,
      "mean_token_accuracy": 0.16314024329185486,
      "num_tokens": 50805288.0,
      "step": 23480
    },
    {
      "entropy": 6.43148283958435,
      "epoch": 2.513136069345604,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004374475569959098,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16978089213371278,
      "num_tokens": 50815563.0,
      "step": 23485
    },
    {
      "entropy": 6.447844362258911,
      "epoch": 2.5136711434533683,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004374211008217316,
      "loss": 5.8058,
      "mean_token_accuracy": 0.17379246652126312,
      "num_tokens": 50826108.0,
      "step": 23490
    },
    {
      "entropy": 6.44561448097229,
      "epoch": 2.5142062175611324,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043739463995766696,
      "loss": 5.817,
      "mean_token_accuracy": 0.16885231286287308,
      "num_tokens": 50837180.0,
      "step": 23495
    },
    {
      "entropy": 6.423863697052002,
      "epoch": 2.514741291668896,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004373681744044799,
      "loss": 5.8226,
      "mean_token_accuracy": 0.17217445820569993,
      "num_tokens": 50847430.0,
      "step": 23500
    },
    {
      "entropy": 6.512261962890625,
      "epoch": 2.51527636577666,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043734170416293467,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16849461644887925,
      "num_tokens": 50858100.0,
      "step": 23505
    },
    {
      "entropy": 6.50587739944458,
      "epoch": 2.5158114398844242,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043731522923379554,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1615135908126831,
      "num_tokens": 50868549.0,
      "step": 23510
    },
    {
      "entropy": 6.519756650924682,
      "epoch": 2.516346513992188,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004372887496178269,
      "loss": 5.8387,
      "mean_token_accuracy": 0.16749815940856932,
      "num_tokens": 50879522.0,
      "step": 23515
    },
    {
      "entropy": 6.458217620849609,
      "epoch": 2.516881588099952,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004372622653157935,
      "loss": 5.848,
      "mean_token_accuracy": 0.17194699496030807,
      "num_tokens": 50891139.0,
      "step": 23520
    },
    {
      "entropy": 6.456025171279907,
      "epoch": 2.5174166622077156,
      "grad_norm": 1.125,
      "learning_rate": 0.0004372357763284599,
      "loss": 5.8829,
      "mean_token_accuracy": 0.16179523169994353,
      "num_tokens": 50901267.0,
      "step": 23525
    },
    {
      "entropy": 6.50121579170227,
      "epoch": 2.5179517363154797,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004372092826565911,
      "loss": 5.7696,
      "mean_token_accuracy": 0.1755378797650337,
      "num_tokens": 50912137.0,
      "step": 23530
    },
    {
      "entropy": 6.470375347137451,
      "epoch": 2.5184868104232434,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004371827843009521,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16875196993350983,
      "num_tokens": 50922724.0,
      "step": 23535
    },
    {
      "entropy": 6.486111116409302,
      "epoch": 2.5190218845310075,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043715628126230795,
      "loss": 5.8699,
      "mean_token_accuracy": 0.16180576086044313,
      "num_tokens": 50934617.0,
      "step": 23540
    },
    {
      "entropy": 6.5267236709594725,
      "epoch": 2.5195569586387716,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004371297735414239,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1637105703353882,
      "num_tokens": 50945179.0,
      "step": 23545
    },
    {
      "entropy": 6.490875482559204,
      "epoch": 2.5200920327465353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043710326113906555,
      "loss": 5.8891,
      "mean_token_accuracy": 0.1705836370587349,
      "num_tokens": 50955383.0,
      "step": 23550
    },
    {
      "entropy": 6.5112481117248535,
      "epoch": 2.5206271068542994,
      "grad_norm": 1.125,
      "learning_rate": 0.0004370767440559983,
      "loss": 5.799,
      "mean_token_accuracy": 0.17422621548175812,
      "num_tokens": 50966107.0,
      "step": 23555
    },
    {
      "entropy": 6.41521258354187,
      "epoch": 2.5211621809620635,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043705022229298775,
      "loss": 5.7305,
      "mean_token_accuracy": 0.18969931602478027,
      "num_tokens": 50977042.0,
      "step": 23560
    },
    {
      "entropy": 6.461200475692749,
      "epoch": 2.521697255069827,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004370236958507998,
      "loss": 5.9116,
      "mean_token_accuracy": 0.1717265009880066,
      "num_tokens": 50986999.0,
      "step": 23565
    },
    {
      "entropy": 6.463965368270874,
      "epoch": 2.522232329177591,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004369971647302005,
      "loss": 5.8416,
      "mean_token_accuracy": 0.16396985501050948,
      "num_tokens": 50997824.0,
      "step": 23570
    },
    {
      "entropy": 6.574843835830689,
      "epoch": 2.522767403285355,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004369706289319558,
      "loss": 5.9122,
      "mean_token_accuracy": 0.16399952173233032,
      "num_tokens": 51009001.0,
      "step": 23575
    },
    {
      "entropy": 6.4940837860107425,
      "epoch": 2.523302477393119,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043694408845683185,
      "loss": 5.8272,
      "mean_token_accuracy": 0.16722618341445922,
      "num_tokens": 51019222.0,
      "step": 23580
    },
    {
      "entropy": 6.435578060150147,
      "epoch": 2.5238375515008826,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004369175433055952,
      "loss": 5.8203,
      "mean_token_accuracy": 0.16909260600805281,
      "num_tokens": 51029473.0,
      "step": 23585
    },
    {
      "entropy": 6.466200447082519,
      "epoch": 2.5243726256086467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043689099347901227,
      "loss": 5.8666,
      "mean_token_accuracy": 0.1666961133480072,
      "num_tokens": 51040667.0,
      "step": 23590
    },
    {
      "entropy": 6.434639883041382,
      "epoch": 2.524907699716411,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043686443897784954,
      "loss": 5.751,
      "mean_token_accuracy": 0.17269023209810258,
      "num_tokens": 51050386.0,
      "step": 23595
    },
    {
      "entropy": 6.461810827255249,
      "epoch": 2.5254427738241745,
      "grad_norm": 1.140625,
      "learning_rate": 0.000436837879802874,
      "loss": 5.8863,
      "mean_token_accuracy": 0.16667189449071884,
      "num_tokens": 51060752.0,
      "step": 23600
    },
    {
      "entropy": 6.438376951217651,
      "epoch": 2.5259778479319386,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043681131595485234,
      "loss": 5.8696,
      "mean_token_accuracy": 0.1632821410894394,
      "num_tokens": 51072467.0,
      "step": 23605
    },
    {
      "entropy": 6.5563154220581055,
      "epoch": 2.5265129220397027,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004367847474345517,
      "loss": 5.9388,
      "mean_token_accuracy": 0.1615274578332901,
      "num_tokens": 51083404.0,
      "step": 23610
    },
    {
      "entropy": 6.524729442596436,
      "epoch": 2.5270479961474663,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043675817424273925,
      "loss": 5.8433,
      "mean_token_accuracy": 0.17226932793855668,
      "num_tokens": 51094392.0,
      "step": 23615
    },
    {
      "entropy": 6.5307234764099125,
      "epoch": 2.5275830702552304,
      "grad_norm": 1.125,
      "learning_rate": 0.00043673159638018215,
      "loss": 5.836,
      "mean_token_accuracy": 0.17042842209339143,
      "num_tokens": 51104706.0,
      "step": 23620
    },
    {
      "entropy": 6.473104619979859,
      "epoch": 2.5281181443629945,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000436705013847648,
      "loss": 5.9036,
      "mean_token_accuracy": 0.16310847997665406,
      "num_tokens": 51114881.0,
      "step": 23625
    },
    {
      "entropy": 6.53145751953125,
      "epoch": 2.528653218470758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004366784266459043,
      "loss": 5.9072,
      "mean_token_accuracy": 0.16330372989177705,
      "num_tokens": 51125705.0,
      "step": 23630
    },
    {
      "entropy": 6.5418425559997555,
      "epoch": 2.5291882925785223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043665183477571874,
      "loss": 5.9087,
      "mean_token_accuracy": 0.16295332014560698,
      "num_tokens": 51136460.0,
      "step": 23635
    },
    {
      "entropy": 6.455497074127197,
      "epoch": 2.529723366686286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004366252382378592,
      "loss": 5.8442,
      "mean_token_accuracy": 0.17003264874219895,
      "num_tokens": 51146206.0,
      "step": 23640
    },
    {
      "entropy": 6.462458610534668,
      "epoch": 2.53025844079405,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004365986370330935,
      "loss": 5.9471,
      "mean_token_accuracy": 0.1625384047627449,
      "num_tokens": 51156650.0,
      "step": 23645
    },
    {
      "entropy": 6.497457027435303,
      "epoch": 2.5307935149018137,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043657203116218994,
      "loss": 5.9186,
      "mean_token_accuracy": 0.16501009464263916,
      "num_tokens": 51168401.0,
      "step": 23650
    },
    {
      "entropy": 6.536757564544677,
      "epoch": 2.531328589009578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043654542062591674,
      "loss": 5.8878,
      "mean_token_accuracy": 0.15894696712493897,
      "num_tokens": 51178571.0,
      "step": 23655
    },
    {
      "entropy": 6.52230658531189,
      "epoch": 2.531863663117342,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043651880542504214,
      "loss": 5.924,
      "mean_token_accuracy": 0.159299036860466,
      "num_tokens": 51189143.0,
      "step": 23660
    },
    {
      "entropy": 6.487399291992188,
      "epoch": 2.5323987372251056,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043649218556033475,
      "loss": 5.8811,
      "mean_token_accuracy": 0.16783400177955626,
      "num_tokens": 51200552.0,
      "step": 23665
    },
    {
      "entropy": 6.529806327819824,
      "epoch": 2.5329338113328697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004364655610325632,
      "loss": 5.9,
      "mean_token_accuracy": 0.1633798211812973,
      "num_tokens": 51210493.0,
      "step": 23670
    },
    {
      "entropy": 6.5160153865814205,
      "epoch": 2.5334688854406338,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004364389318424962,
      "loss": 5.8813,
      "mean_token_accuracy": 0.16537481546401978,
      "num_tokens": 51220535.0,
      "step": 23675
    },
    {
      "entropy": 6.467274618148804,
      "epoch": 2.5340039595483974,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043641229799090285,
      "loss": 5.9354,
      "mean_token_accuracy": 0.15889283418655395,
      "num_tokens": 51232078.0,
      "step": 23680
    },
    {
      "entropy": 6.51124119758606,
      "epoch": 2.5345390336561615,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000436385659478552,
      "loss": 5.8846,
      "mean_token_accuracy": 0.1743737831711769,
      "num_tokens": 51242750.0,
      "step": 23685
    },
    {
      "entropy": 6.6086986541748045,
      "epoch": 2.535074107763925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043635901630621293,
      "loss": 5.9117,
      "mean_token_accuracy": 0.16255918145179749,
      "num_tokens": 51254295.0,
      "step": 23690
    },
    {
      "entropy": 6.479697322845459,
      "epoch": 2.5356091818716893,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043633236847465497,
      "loss": 5.861,
      "mean_token_accuracy": 0.16950491517782212,
      "num_tokens": 51265060.0,
      "step": 23695
    },
    {
      "entropy": 6.407068586349487,
      "epoch": 2.536144255979453,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004363057159846475,
      "loss": 5.7797,
      "mean_token_accuracy": 0.16911096572875978,
      "num_tokens": 51275623.0,
      "step": 23700
    },
    {
      "entropy": 6.447988271713257,
      "epoch": 2.536679330087217,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004362790588369602,
      "loss": 5.7642,
      "mean_token_accuracy": 0.17667825222015382,
      "num_tokens": 51285830.0,
      "step": 23705
    },
    {
      "entropy": 6.522962951660157,
      "epoch": 2.537214404194981,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004362523970323627,
      "loss": 5.88,
      "mean_token_accuracy": 0.1660325139760971,
      "num_tokens": 51297047.0,
      "step": 23710
    },
    {
      "entropy": 6.488448762893677,
      "epoch": 2.5377494783027448,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043622573057162484,
      "loss": 5.8143,
      "mean_token_accuracy": 0.17077960520982743,
      "num_tokens": 51307254.0,
      "step": 23715
    },
    {
      "entropy": 6.517378139495849,
      "epoch": 2.538284552410509,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004361990594555167,
      "loss": 5.97,
      "mean_token_accuracy": 0.15426816791296005,
      "num_tokens": 51317796.0,
      "step": 23720
    },
    {
      "entropy": 6.536872673034668,
      "epoch": 2.538819626518273,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043617238368480845,
      "loss": 5.9564,
      "mean_token_accuracy": 0.16186587512493134,
      "num_tokens": 51330172.0,
      "step": 23725
    },
    {
      "entropy": 6.506312513351441,
      "epoch": 2.5393547006260366,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043614570326027015,
      "loss": 5.8011,
      "mean_token_accuracy": 0.17344185560941697,
      "num_tokens": 51340844.0,
      "step": 23730
    },
    {
      "entropy": 6.486125946044922,
      "epoch": 2.5398897747338007,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004361190181826724,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16895855069160462,
      "num_tokens": 51352078.0,
      "step": 23735
    },
    {
      "entropy": 6.499806451797485,
      "epoch": 2.540424848841565,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043609232845278565,
      "loss": 5.9438,
      "mean_token_accuracy": 0.1538054659962654,
      "num_tokens": 51364121.0,
      "step": 23740
    },
    {
      "entropy": 6.493271684646606,
      "epoch": 2.5409599229493285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004360656340713805,
      "loss": 5.8862,
      "mean_token_accuracy": 0.17439073026180268,
      "num_tokens": 51375046.0,
      "step": 23745
    },
    {
      "entropy": 6.48885703086853,
      "epoch": 2.541494997057092,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004360389350392278,
      "loss": 5.8025,
      "mean_token_accuracy": 0.1650024965405464,
      "num_tokens": 51386363.0,
      "step": 23750
    },
    {
      "entropy": 6.419749450683594,
      "epoch": 2.5420300711648562,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043601223135709846,
      "loss": 5.7813,
      "mean_token_accuracy": 0.17530083954334258,
      "num_tokens": 51398036.0,
      "step": 23755
    },
    {
      "entropy": 6.4764996528625485,
      "epoch": 2.5425651452726203,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004359855230257636,
      "loss": 5.8693,
      "mean_token_accuracy": 0.16315210312604905,
      "num_tokens": 51408769.0,
      "step": 23760
    },
    {
      "entropy": 6.481798553466797,
      "epoch": 2.543100219380384,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043595881004599444,
      "loss": 5.8947,
      "mean_token_accuracy": 0.1643972650170326,
      "num_tokens": 51420315.0,
      "step": 23765
    },
    {
      "entropy": 6.47357029914856,
      "epoch": 2.543635293488148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043593209241856216,
      "loss": 5.8172,
      "mean_token_accuracy": 0.17637567967176437,
      "num_tokens": 51432283.0,
      "step": 23770
    },
    {
      "entropy": 6.507818412780762,
      "epoch": 2.544170367595912,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043590537014423833,
      "loss": 5.8478,
      "mean_token_accuracy": 0.1744784250855446,
      "num_tokens": 51443151.0,
      "step": 23775
    },
    {
      "entropy": 6.401148128509521,
      "epoch": 2.544705441703676,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004358786432237946,
      "loss": 5.7778,
      "mean_token_accuracy": 0.16762554943561553,
      "num_tokens": 51453964.0,
      "step": 23780
    },
    {
      "entropy": 6.485669374465942,
      "epoch": 2.54524051581144,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004358519116580026,
      "loss": 5.8152,
      "mean_token_accuracy": 0.16888062059879302,
      "num_tokens": 51464236.0,
      "step": 23785
    },
    {
      "entropy": 6.402811479568482,
      "epoch": 2.545775589919204,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004358251754476344,
      "loss": 5.8056,
      "mean_token_accuracy": 0.1757721021771431,
      "num_tokens": 51474388.0,
      "step": 23790
    },
    {
      "entropy": 6.455102157592774,
      "epoch": 2.5463106640269677,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004357984345934617,
      "loss": 5.8248,
      "mean_token_accuracy": 0.15862708389759064,
      "num_tokens": 51485209.0,
      "step": 23795
    },
    {
      "entropy": 6.4282303810119625,
      "epoch": 2.546845738134732,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004357716890962568,
      "loss": 5.7898,
      "mean_token_accuracy": 0.1697897046804428,
      "num_tokens": 51495993.0,
      "step": 23800
    },
    {
      "entropy": 6.458822965621948,
      "epoch": 2.5473808122424955,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000435744938956792,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17241765409708024,
      "num_tokens": 51506282.0,
      "step": 23805
    },
    {
      "entropy": 6.506618070602417,
      "epoch": 2.5479158863502596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043571818417583976,
      "loss": 5.8966,
      "mean_token_accuracy": 0.16533890962600709,
      "num_tokens": 51517728.0,
      "step": 23810
    },
    {
      "entropy": 6.479338979721069,
      "epoch": 2.548450960458023,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004356914247541724,
      "loss": 5.8341,
      "mean_token_accuracy": 0.16847689002752303,
      "num_tokens": 51528769.0,
      "step": 23815
    },
    {
      "entropy": 6.472897720336914,
      "epoch": 2.5489860345657873,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043566466069256283,
      "loss": 5.8244,
      "mean_token_accuracy": 0.1712278574705124,
      "num_tokens": 51540148.0,
      "step": 23820
    },
    {
      "entropy": 6.468016576766968,
      "epoch": 2.5495211086735514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004356378919917837,
      "loss": 5.8126,
      "mean_token_accuracy": 0.17377597242593765,
      "num_tokens": 51550657.0,
      "step": 23825
    },
    {
      "entropy": 6.504455471038819,
      "epoch": 2.550056182781315,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004356111186526081,
      "loss": 5.8433,
      "mean_token_accuracy": 0.1722799777984619,
      "num_tokens": 51560399.0,
      "step": 23830
    },
    {
      "entropy": 6.478572750091553,
      "epoch": 2.550591256889079,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043558434067580894,
      "loss": 5.887,
      "mean_token_accuracy": 0.16424966156482695,
      "num_tokens": 51571463.0,
      "step": 23835
    },
    {
      "entropy": 6.455742883682251,
      "epoch": 2.5511263309968433,
      "grad_norm": 0.96875,
      "learning_rate": 0.00043555755806215946,
      "loss": 5.8008,
      "mean_token_accuracy": 0.17660776674747466,
      "num_tokens": 51583136.0,
      "step": 23840
    },
    {
      "entropy": 6.539184617996216,
      "epoch": 2.551661405104607,
      "grad_norm": 1.0,
      "learning_rate": 0.00043553077081243314,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16063306480646133,
      "num_tokens": 51594397.0,
      "step": 23845
    },
    {
      "entropy": 6.528384780883789,
      "epoch": 2.552196479212371,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004355039789274034,
      "loss": 5.9431,
      "mean_token_accuracy": 0.16338956356048584,
      "num_tokens": 51605788.0,
      "step": 23850
    },
    {
      "entropy": 6.513995504379272,
      "epoch": 2.5527315533201347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004354771824078437,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16666921824216843,
      "num_tokens": 51615693.0,
      "step": 23855
    },
    {
      "entropy": 6.442201757431031,
      "epoch": 2.553266627427899,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043545038125452793,
      "loss": 5.7593,
      "mean_token_accuracy": 0.17332127392292024,
      "num_tokens": 51626607.0,
      "step": 23860
    },
    {
      "entropy": 6.462056827545166,
      "epoch": 2.5538017015356624,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043542357546823003,
      "loss": 5.8515,
      "mean_token_accuracy": 0.16773085445165634,
      "num_tokens": 51636905.0,
      "step": 23865
    },
    {
      "entropy": 6.444929552078247,
      "epoch": 2.5543367756434265,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043539676504972393,
      "loss": 5.8333,
      "mean_token_accuracy": 0.16650621294975282,
      "num_tokens": 51648142.0,
      "step": 23870
    },
    {
      "entropy": 6.576227474212646,
      "epoch": 2.5548718497511906,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004353699499997837,
      "loss": 5.8272,
      "mean_token_accuracy": 0.16700329333543779,
      "num_tokens": 51659583.0,
      "step": 23875
    },
    {
      "entropy": 6.522205877304077,
      "epoch": 2.5554069238589543,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004353431303191837,
      "loss": 5.9332,
      "mean_token_accuracy": 0.16275508999824523,
      "num_tokens": 51670582.0,
      "step": 23880
    },
    {
      "entropy": 6.558988904953003,
      "epoch": 2.5559419979667184,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043531630600869846,
      "loss": 5.9075,
      "mean_token_accuracy": 0.16118451356887817,
      "num_tokens": 51680773.0,
      "step": 23885
    },
    {
      "entropy": 6.51259217262268,
      "epoch": 2.5564770720744825,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004352894770691024,
      "loss": 5.8864,
      "mean_token_accuracy": 0.16588906943798065,
      "num_tokens": 51690680.0,
      "step": 23890
    },
    {
      "entropy": 6.5203039169311525,
      "epoch": 2.557012146182246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004352626435011701,
      "loss": 5.8534,
      "mean_token_accuracy": 0.16917362660169602,
      "num_tokens": 51700910.0,
      "step": 23895
    },
    {
      "entropy": 6.494070768356323,
      "epoch": 2.5575472202900102,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043523580530567654,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16796368211507798,
      "num_tokens": 51712331.0,
      "step": 23900
    },
    {
      "entropy": 6.420089530944824,
      "epoch": 2.5580822943977743,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004352089624833966,
      "loss": 5.7612,
      "mean_token_accuracy": 0.1733874574303627,
      "num_tokens": 51722805.0,
      "step": 23905
    },
    {
      "entropy": 6.488110971450806,
      "epoch": 2.558617368505538,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043518211503510535,
      "loss": 5.8278,
      "mean_token_accuracy": 0.17454640567302704,
      "num_tokens": 51733909.0,
      "step": 23910
    },
    {
      "entropy": 6.50665168762207,
      "epoch": 2.559152442613302,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043515526296157813,
      "loss": 5.8842,
      "mean_token_accuracy": 0.1593624994158745,
      "num_tokens": 51745062.0,
      "step": 23915
    },
    {
      "entropy": 6.482795000076294,
      "epoch": 2.5596875167210658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004351284062635901,
      "loss": 5.9033,
      "mean_token_accuracy": 0.16618989408016205,
      "num_tokens": 51756235.0,
      "step": 23920
    },
    {
      "entropy": 6.545911693572998,
      "epoch": 2.56022259082883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000435101544941917,
      "loss": 5.85,
      "mean_token_accuracy": 0.16460319757461547,
      "num_tokens": 51765689.0,
      "step": 23925
    },
    {
      "entropy": 6.487536334991455,
      "epoch": 2.5607576649365935,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004350746789973341,
      "loss": 5.7156,
      "mean_token_accuracy": 0.16793793439865112,
      "num_tokens": 51777090.0,
      "step": 23930
    },
    {
      "entropy": 6.484965229034424,
      "epoch": 2.5612927390443576,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004350478084306174,
      "loss": 5.8775,
      "mean_token_accuracy": 0.16193293780088425,
      "num_tokens": 51787360.0,
      "step": 23935
    },
    {
      "entropy": 6.422898530960083,
      "epoch": 2.5618278131521217,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004350209332425428,
      "loss": 5.8097,
      "mean_token_accuracy": 0.169431072473526,
      "num_tokens": 51799075.0,
      "step": 23940
    },
    {
      "entropy": 6.384306335449219,
      "epoch": 2.5623628872598854,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043499405343388613,
      "loss": 5.6888,
      "mean_token_accuracy": 0.18263380974531174,
      "num_tokens": 51810024.0,
      "step": 23945
    },
    {
      "entropy": 6.4647351741790775,
      "epoch": 2.5628979613676495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004349671690054238,
      "loss": 5.817,
      "mean_token_accuracy": 0.1623155191540718,
      "num_tokens": 51820710.0,
      "step": 23950
    },
    {
      "entropy": 6.477718782424927,
      "epoch": 2.5634330354754136,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043494027995793176,
      "loss": 5.8327,
      "mean_token_accuracy": 0.17112138718366623,
      "num_tokens": 51831451.0,
      "step": 23955
    },
    {
      "entropy": 6.355104303359985,
      "epoch": 2.5639681095831772,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004349133862921867,
      "loss": 5.7753,
      "mean_token_accuracy": 0.18032821118831635,
      "num_tokens": 51841483.0,
      "step": 23960
    },
    {
      "entropy": 6.480193567276001,
      "epoch": 2.5645031836909413,
      "grad_norm": 1.125,
      "learning_rate": 0.00043488648800896516,
      "loss": 5.8219,
      "mean_token_accuracy": 0.17155596762895584,
      "num_tokens": 51852838.0,
      "step": 23965
    },
    {
      "entropy": 6.468820524215698,
      "epoch": 2.565038257798705,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004348595851090436,
      "loss": 5.8579,
      "mean_token_accuracy": 0.1640855059027672,
      "num_tokens": 51863851.0,
      "step": 23970
    },
    {
      "entropy": 6.474511384963989,
      "epoch": 2.565573331906469,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043483267759319907,
      "loss": 5.7772,
      "mean_token_accuracy": 0.17176942825317382,
      "num_tokens": 51874380.0,
      "step": 23975
    },
    {
      "entropy": 6.509739351272583,
      "epoch": 2.5661084060142327,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004348057654622084,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16756340712308884,
      "num_tokens": 51884594.0,
      "step": 23980
    },
    {
      "entropy": 6.511013174057007,
      "epoch": 2.566643480121997,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004347788487168487,
      "loss": 5.9318,
      "mean_token_accuracy": 0.16200787872076033,
      "num_tokens": 51897679.0,
      "step": 23985
    },
    {
      "entropy": 6.599509572982788,
      "epoch": 2.567178554229761,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004347519273578972,
      "loss": 5.9958,
      "mean_token_accuracy": 0.15727077871561052,
      "num_tokens": 51907501.0,
      "step": 23990
    },
    {
      "entropy": 6.494534301757812,
      "epoch": 2.5677136283375246,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043472500138613124,
      "loss": 5.8502,
      "mean_token_accuracy": 0.16256376653909682,
      "num_tokens": 51917732.0,
      "step": 23995
    },
    {
      "entropy": 6.481424808502197,
      "epoch": 2.5682487024452887,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004346980708023284,
      "loss": 5.9112,
      "mean_token_accuracy": 0.163826385140419,
      "num_tokens": 51929038.0,
      "step": 24000
    },
    {
      "epoch": 2.5682487024452887,
      "eval_entropy": 6.246094875644159,
      "eval_loss": 5.942140579223633,
      "eval_mean_token_accuracy": 0.17128227438396118,
      "eval_num_tokens": 51929038.0,
      "eval_runtime": 17.673,
      "eval_samples_per_second": 1679.344,
      "eval_steps_per_second": 209.925,
      "step": 24000
    },
    {
      "entropy": 6.528321981430054,
      "epoch": 2.568783776553053,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004346711356072661,
      "loss": 5.9551,
      "mean_token_accuracy": 0.15530667901039125,
      "num_tokens": 51939481.0,
      "step": 24005
    },
    {
      "entropy": 6.615501499176025,
      "epoch": 2.5693188506608164,
      "grad_norm": 1.125,
      "learning_rate": 0.0004346441958017222,
      "loss": 5.9092,
      "mean_token_accuracy": 0.1626884564757347,
      "num_tokens": 51951393.0,
      "step": 24010
    },
    {
      "entropy": 6.568589210510254,
      "epoch": 2.5698539247685805,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004346172513864746,
      "loss": 5.9361,
      "mean_token_accuracy": 0.15947612076997758,
      "num_tokens": 51962897.0,
      "step": 24015
    },
    {
      "entropy": 6.509376764297485,
      "epoch": 2.5703889988763446,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043459030236230137,
      "loss": 5.8718,
      "mean_token_accuracy": 0.1644015684723854,
      "num_tokens": 51974443.0,
      "step": 24020
    },
    {
      "entropy": 6.433397054672241,
      "epoch": 2.5709240729841083,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004345633487299805,
      "loss": 5.8659,
      "mean_token_accuracy": 0.16950739473104476,
      "num_tokens": 51985158.0,
      "step": 24025
    },
    {
      "entropy": 6.545836305618286,
      "epoch": 2.571459147091872,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043453639049029035,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15801543444395066,
      "num_tokens": 51994956.0,
      "step": 24030
    },
    {
      "entropy": 6.540965843200683,
      "epoch": 2.571994221199636,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043450942764400937,
      "loss": 5.7489,
      "mean_token_accuracy": 0.17454442381858826,
      "num_tokens": 52004922.0,
      "step": 24035
    },
    {
      "entropy": 6.431773662567139,
      "epoch": 2.5725292953074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004344824601919161,
      "loss": 5.8007,
      "mean_token_accuracy": 0.17699980735778809,
      "num_tokens": 52014684.0,
      "step": 24040
    },
    {
      "entropy": 6.4730345726013185,
      "epoch": 2.573064369415164,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043445548813478914,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17918358892202377,
      "num_tokens": 52025283.0,
      "step": 24045
    },
    {
      "entropy": 6.5161745071411135,
      "epoch": 2.573599443522928,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004344285114734074,
      "loss": 5.916,
      "mean_token_accuracy": 0.1673520177602768,
      "num_tokens": 52036513.0,
      "step": 24050
    },
    {
      "entropy": 6.528909063339233,
      "epoch": 2.574134517630692,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043440153020854987,
      "loss": 5.7916,
      "mean_token_accuracy": 0.17205137759447098,
      "num_tokens": 52047230.0,
      "step": 24055
    },
    {
      "entropy": 6.50579981803894,
      "epoch": 2.5746695917384557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004343745443409954,
      "loss": 5.8655,
      "mean_token_accuracy": 0.1688064903020859,
      "num_tokens": 52058889.0,
      "step": 24060
    },
    {
      "entropy": 6.460259532928466,
      "epoch": 2.5752046658462198,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043434755387152354,
      "loss": 5.8144,
      "mean_token_accuracy": 0.1710692584514618,
      "num_tokens": 52068478.0,
      "step": 24065
    },
    {
      "entropy": 6.449825382232666,
      "epoch": 2.575739739953984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004343205588009133,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16937859058380128,
      "num_tokens": 52079756.0,
      "step": 24070
    },
    {
      "entropy": 6.529866313934326,
      "epoch": 2.5762748140617475,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043429355912994437,
      "loss": 5.8458,
      "mean_token_accuracy": 0.1704728588461876,
      "num_tokens": 52090452.0,
      "step": 24075
    },
    {
      "entropy": 6.46786470413208,
      "epoch": 2.5768098881695116,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004342665548593964,
      "loss": 5.7755,
      "mean_token_accuracy": 0.16981011331081391,
      "num_tokens": 52101134.0,
      "step": 24080
    },
    {
      "entropy": 6.474951553344726,
      "epoch": 2.5773449622772753,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004342395459900489,
      "loss": 5.8034,
      "mean_token_accuracy": 0.17088887095451355,
      "num_tokens": 52111529.0,
      "step": 24085
    },
    {
      "entropy": 6.420427465438843,
      "epoch": 2.5778800363850394,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043421253252268194,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16996812969446182,
      "num_tokens": 52122779.0,
      "step": 24090
    },
    {
      "entropy": 6.492587900161743,
      "epoch": 2.578415110492803,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004341855144580754,
      "loss": 5.8057,
      "mean_token_accuracy": 0.16402738839387893,
      "num_tokens": 52134140.0,
      "step": 24095
    },
    {
      "entropy": 6.484058094024658,
      "epoch": 2.578950184600567,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043415849179700956,
      "loss": 5.8155,
      "mean_token_accuracy": 0.16996650099754335,
      "num_tokens": 52146033.0,
      "step": 24100
    },
    {
      "entropy": 6.483802938461304,
      "epoch": 2.5794852587083312,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004341314645402647,
      "loss": 5.8284,
      "mean_token_accuracy": 0.16691495031118392,
      "num_tokens": 52157516.0,
      "step": 24105
    },
    {
      "entropy": 6.476871633529663,
      "epoch": 2.580020332816095,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004341044326886211,
      "loss": 5.8001,
      "mean_token_accuracy": 0.17904454916715623,
      "num_tokens": 52167523.0,
      "step": 24110
    },
    {
      "entropy": 6.388882255554199,
      "epoch": 2.580555406923859,
      "grad_norm": 1.125,
      "learning_rate": 0.0004340773962428593,
      "loss": 5.7488,
      "mean_token_accuracy": 0.17883386760950087,
      "num_tokens": 52178235.0,
      "step": 24115
    },
    {
      "entropy": 6.460281324386597,
      "epoch": 2.581090481031623,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043405035520376013,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17641050219535828,
      "num_tokens": 52188586.0,
      "step": 24120
    },
    {
      "entropy": 6.496337842941284,
      "epoch": 2.5816255551393867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004340233095721043,
      "loss": 5.8388,
      "mean_token_accuracy": 0.1730636864900589,
      "num_tokens": 52199081.0,
      "step": 24125
    },
    {
      "entropy": 6.489830160140992,
      "epoch": 2.582160629247151,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004339962593486727,
      "loss": 5.8863,
      "mean_token_accuracy": 0.16583990901708603,
      "num_tokens": 52210324.0,
      "step": 24130
    },
    {
      "entropy": 6.529017114639283,
      "epoch": 2.5826957033549145,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043396920453424656,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16555933952331542,
      "num_tokens": 52221973.0,
      "step": 24135
    },
    {
      "entropy": 6.4329170227050785,
      "epoch": 2.5832307774626786,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043394214512960683,
      "loss": 5.7952,
      "mean_token_accuracy": 0.16738942563533782,
      "num_tokens": 52232560.0,
      "step": 24140
    },
    {
      "entropy": 6.432302951812744,
      "epoch": 2.5837658515704423,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004339150811355351,
      "loss": 5.7522,
      "mean_token_accuracy": 0.17248117923736572,
      "num_tokens": 52243416.0,
      "step": 24145
    },
    {
      "entropy": 6.420765066146851,
      "epoch": 2.5843009256782064,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004338880125528127,
      "loss": 5.8299,
      "mean_token_accuracy": 0.17080971598625183,
      "num_tokens": 52254213.0,
      "step": 24150
    },
    {
      "entropy": 6.402639961242675,
      "epoch": 2.5848359997859705,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043386093938222126,
      "loss": 5.7991,
      "mean_token_accuracy": 0.16703277975320815,
      "num_tokens": 52265381.0,
      "step": 24155
    },
    {
      "entropy": 6.469441843032837,
      "epoch": 2.585371073893734,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004338338616245425,
      "loss": 5.7542,
      "mean_token_accuracy": 0.17471615821123124,
      "num_tokens": 52276407.0,
      "step": 24160
    },
    {
      "entropy": 6.487953042984008,
      "epoch": 2.585906148001498,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043380677928055834,
      "loss": 5.7969,
      "mean_token_accuracy": 0.1680793896317482,
      "num_tokens": 52286484.0,
      "step": 24165
    },
    {
      "entropy": 6.496300601959229,
      "epoch": 2.5864412221092623,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043377969235105075,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1814131796360016,
      "num_tokens": 52297674.0,
      "step": 24170
    },
    {
      "entropy": 6.5267857074737545,
      "epoch": 2.586976296217026,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004337526008368018,
      "loss": 5.9889,
      "mean_token_accuracy": 0.15604786723852157,
      "num_tokens": 52309100.0,
      "step": 24175
    },
    {
      "entropy": 6.521353149414063,
      "epoch": 2.58751137032479,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004337255047385938,
      "loss": 5.9149,
      "mean_token_accuracy": 0.16376935690641403,
      "num_tokens": 52320002.0,
      "step": 24180
    },
    {
      "entropy": 6.515598583221435,
      "epoch": 2.588046444432554,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043369840405720923,
      "loss": 5.81,
      "mean_token_accuracy": 0.16819993257522584,
      "num_tokens": 52330073.0,
      "step": 24185
    },
    {
      "entropy": 6.531065654754639,
      "epoch": 2.588581518540318,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043367129879343054,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16262570917606353,
      "num_tokens": 52340472.0,
      "step": 24190
    },
    {
      "entropy": 6.4435680389404295,
      "epoch": 2.589116592648082,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043364418894804033,
      "loss": 5.8702,
      "mean_token_accuracy": 0.1649083450436592,
      "num_tokens": 52350654.0,
      "step": 24195
    },
    {
      "entropy": 6.426727056503296,
      "epoch": 2.5896516667558456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043361707452182146,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16626553982496262,
      "num_tokens": 52361412.0,
      "step": 24200
    },
    {
      "entropy": 6.491952037811279,
      "epoch": 2.5901867408636097,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043358995551555693,
      "loss": 5.7997,
      "mean_token_accuracy": 0.17008461207151412,
      "num_tokens": 52372030.0,
      "step": 24205
    },
    {
      "entropy": 6.4797522068023685,
      "epoch": 2.5907218149713733,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043356283193002966,
      "loss": 5.8826,
      "mean_token_accuracy": 0.17062922567129135,
      "num_tokens": 52382965.0,
      "step": 24210
    },
    {
      "entropy": 6.468506622314453,
      "epoch": 2.5912568890791374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004335357037660229,
      "loss": 5.7732,
      "mean_token_accuracy": 0.17656420469284057,
      "num_tokens": 52393959.0,
      "step": 24215
    },
    {
      "entropy": 6.538433265686035,
      "epoch": 2.5917919631869015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043350857102432,
      "loss": 5.9099,
      "mean_token_accuracy": 0.16203110292553902,
      "num_tokens": 52405142.0,
      "step": 24220
    },
    {
      "entropy": 6.495584487915039,
      "epoch": 2.592327037294665,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004334814337057043,
      "loss": 5.8754,
      "mean_token_accuracy": 0.1605286568403244,
      "num_tokens": 52415927.0,
      "step": 24225
    },
    {
      "entropy": 6.472766447067261,
      "epoch": 2.5928621114024293,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043345429181095957,
      "loss": 5.789,
      "mean_token_accuracy": 0.17478866577148439,
      "num_tokens": 52426475.0,
      "step": 24230
    },
    {
      "entropy": 6.4724650382995605,
      "epoch": 2.5933971855101934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043342714534086945,
      "loss": 5.8209,
      "mean_token_accuracy": 0.17247231304645538,
      "num_tokens": 52437239.0,
      "step": 24235
    },
    {
      "entropy": 6.49750394821167,
      "epoch": 2.593932259617957,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004333999942962177,
      "loss": 5.8644,
      "mean_token_accuracy": 0.161271633207798,
      "num_tokens": 52447323.0,
      "step": 24240
    },
    {
      "entropy": 6.5336998462677,
      "epoch": 2.594467333725721,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043337283867778836,
      "loss": 5.9142,
      "mean_token_accuracy": 0.167860247194767,
      "num_tokens": 52458465.0,
      "step": 24245
    },
    {
      "entropy": 6.4886126041412355,
      "epoch": 2.595002407833485,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043334567848636563,
      "loss": 5.8944,
      "mean_token_accuracy": 0.16244422793388366,
      "num_tokens": 52468809.0,
      "step": 24250
    },
    {
      "entropy": 6.523831748962403,
      "epoch": 2.595537481941249,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043331851372273365,
      "loss": 5.8676,
      "mean_token_accuracy": 0.16373403668403624,
      "num_tokens": 52479456.0,
      "step": 24255
    },
    {
      "entropy": 6.494037055969239,
      "epoch": 2.5960725560490125,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043329134438767683,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1719989150762558,
      "num_tokens": 52490146.0,
      "step": 24260
    },
    {
      "entropy": 6.529480743408203,
      "epoch": 2.5966076301567766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004332641704819797,
      "loss": 5.9017,
      "mean_token_accuracy": 0.163737915456295,
      "num_tokens": 52499875.0,
      "step": 24265
    },
    {
      "entropy": 6.452851057052612,
      "epoch": 2.5971427042645407,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043323699200642686,
      "loss": 5.7393,
      "mean_token_accuracy": 0.17367866337299348,
      "num_tokens": 52510297.0,
      "step": 24270
    },
    {
      "entropy": 6.306313133239746,
      "epoch": 2.5976777783723044,
      "grad_norm": 1.0,
      "learning_rate": 0.0004332098089618031,
      "loss": 5.6894,
      "mean_token_accuracy": 0.18564457595348358,
      "num_tokens": 52523111.0,
      "step": 24275
    },
    {
      "entropy": 6.438412046432495,
      "epoch": 2.5982128524800685,
      "grad_norm": 1.25,
      "learning_rate": 0.00043318262134889334,
      "loss": 5.8189,
      "mean_token_accuracy": 0.1788093015551567,
      "num_tokens": 52534097.0,
      "step": 24280
    },
    {
      "entropy": 6.488639974594117,
      "epoch": 2.5987479265878326,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004331554291684827,
      "loss": 5.87,
      "mean_token_accuracy": 0.16787887066602708,
      "num_tokens": 52544985.0,
      "step": 24285
    },
    {
      "entropy": 6.559153318405151,
      "epoch": 2.5992830006955963,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004331282324213561,
      "loss": 5.854,
      "mean_token_accuracy": 0.17110853791236877,
      "num_tokens": 52555772.0,
      "step": 24290
    },
    {
      "entropy": 6.453483438491821,
      "epoch": 2.5998180748033604,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004331010311082992,
      "loss": 5.8482,
      "mean_token_accuracy": 0.166809543967247,
      "num_tokens": 52566578.0,
      "step": 24295
    },
    {
      "entropy": 6.445932149887085,
      "epoch": 2.6003531489111245,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043307382523009717,
      "loss": 5.9141,
      "mean_token_accuracy": 0.15865110903978347,
      "num_tokens": 52578551.0,
      "step": 24300
    },
    {
      "entropy": 6.517331552505493,
      "epoch": 2.600888223018888,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004330466147875356,
      "loss": 5.8188,
      "mean_token_accuracy": 0.17244610637426377,
      "num_tokens": 52589225.0,
      "step": 24305
    },
    {
      "entropy": 6.488640546798706,
      "epoch": 2.6014232971266518,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043301939978140024,
      "loss": 5.7633,
      "mean_token_accuracy": 0.17592981159687043,
      "num_tokens": 52599572.0,
      "step": 24310
    },
    {
      "entropy": 6.5210412502288815,
      "epoch": 2.601958371234416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043299218021247693,
      "loss": 5.867,
      "mean_token_accuracy": 0.1689134031534195,
      "num_tokens": 52610562.0,
      "step": 24315
    },
    {
      "entropy": 6.51394829750061,
      "epoch": 2.60249344534218,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004329649560815517,
      "loss": 5.936,
      "mean_token_accuracy": 0.1604168236255646,
      "num_tokens": 52621078.0,
      "step": 24320
    },
    {
      "entropy": 6.468183374404907,
      "epoch": 2.6030285194499436,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004329377273894104,
      "loss": 5.834,
      "mean_token_accuracy": 0.17429707795381547,
      "num_tokens": 52632336.0,
      "step": 24325
    },
    {
      "entropy": 6.375121212005615,
      "epoch": 2.6035635935577077,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004329104941368396,
      "loss": 5.7293,
      "mean_token_accuracy": 0.18177442252635956,
      "num_tokens": 52643701.0,
      "step": 24330
    },
    {
      "entropy": 6.37853479385376,
      "epoch": 2.604098667665472,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004328832563246254,
      "loss": 5.774,
      "mean_token_accuracy": 0.18020970970392228,
      "num_tokens": 52654052.0,
      "step": 24335
    },
    {
      "entropy": 6.446176242828369,
      "epoch": 2.6046337417732355,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043285601395355426,
      "loss": 5.738,
      "mean_token_accuracy": 0.18140121996402742,
      "num_tokens": 52665033.0,
      "step": 24340
    },
    {
      "entropy": 6.53370885848999,
      "epoch": 2.6051688158809996,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043282876702441296,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16478959769010543,
      "num_tokens": 52674726.0,
      "step": 24345
    },
    {
      "entropy": 6.478613948822021,
      "epoch": 2.6057038899887637,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00043280151553798824,
      "loss": 5.8664,
      "mean_token_accuracy": 0.17012593150138855,
      "num_tokens": 52687221.0,
      "step": 24350
    },
    {
      "entropy": 6.461548233032227,
      "epoch": 2.6062389640965273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004327742594950669,
      "loss": 5.888,
      "mean_token_accuracy": 0.17210280150175095,
      "num_tokens": 52697902.0,
      "step": 24355
    },
    {
      "entropy": 6.484062147140503,
      "epoch": 2.6067740382042914,
      "grad_norm": 1.203125,
      "learning_rate": 0.000432746998896436,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1707917869091034,
      "num_tokens": 52707845.0,
      "step": 24360
    },
    {
      "entropy": 6.462534618377686,
      "epoch": 2.607309112312055,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004327197337428827,
      "loss": 5.7723,
      "mean_token_accuracy": 0.17089948654174805,
      "num_tokens": 52719254.0,
      "step": 24365
    },
    {
      "entropy": 6.483975219726562,
      "epoch": 2.607844186419819,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004326924640351942,
      "loss": 5.8594,
      "mean_token_accuracy": 0.16400874108076097,
      "num_tokens": 52729765.0,
      "step": 24370
    },
    {
      "entropy": 6.491088819503784,
      "epoch": 2.608379260527583,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043266518977415797,
      "loss": 5.8141,
      "mean_token_accuracy": 0.1699533596634865,
      "num_tokens": 52740997.0,
      "step": 24375
    },
    {
      "entropy": 6.507480192184448,
      "epoch": 2.608914334635347,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004326379109605615,
      "loss": 5.9647,
      "mean_token_accuracy": 0.16042214035987853,
      "num_tokens": 52753107.0,
      "step": 24380
    },
    {
      "entropy": 6.443166494369507,
      "epoch": 2.609449408743111,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004326106275951926,
      "loss": 5.7751,
      "mean_token_accuracy": 0.1642197161912918,
      "num_tokens": 52765453.0,
      "step": 24385
    },
    {
      "entropy": 6.5265179634094235,
      "epoch": 2.6099844828508747,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004325833396788389,
      "loss": 5.8593,
      "mean_token_accuracy": 0.1701074481010437,
      "num_tokens": 52776145.0,
      "step": 24390
    },
    {
      "entropy": 6.5216938018798825,
      "epoch": 2.610519556958639,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004325560472122884,
      "loss": 5.894,
      "mean_token_accuracy": 0.16259075999259948,
      "num_tokens": 52788404.0,
      "step": 24395
    },
    {
      "entropy": 6.5248359680175785,
      "epoch": 2.611054631066403,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004325287501963293,
      "loss": 5.8903,
      "mean_token_accuracy": 0.17089541852474213,
      "num_tokens": 52800481.0,
      "step": 24400
    },
    {
      "entropy": 6.5086887836456295,
      "epoch": 2.6115897051741666,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325014486317495,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16979362815618515,
      "num_tokens": 52811539.0,
      "step": 24405
    },
    {
      "entropy": 6.508788824081421,
      "epoch": 2.6121247792819307,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324741425193376,
      "loss": 5.8606,
      "mean_token_accuracy": 0.1724521279335022,
      "num_tokens": 52822330.0,
      "step": 24410
    },
    {
      "entropy": 6.506738710403442,
      "epoch": 2.6126598533896943,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000432446831859882,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16785401850938797,
      "num_tokens": 52832447.0,
      "step": 24415
    },
    {
      "entropy": 6.431128025054932,
      "epoch": 2.6131949274974584,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004324195166541712,
      "loss": 5.798,
      "mean_token_accuracy": 0.17242011576890945,
      "num_tokens": 52843089.0,
      "step": 24420
    },
    {
      "entropy": 6.5355918407440186,
      "epoch": 2.613730001605222,
      "grad_norm": 1.15625,
      "learning_rate": 0.000432392196902994,
      "loss": 5.8608,
      "mean_token_accuracy": 0.1681615576148033,
      "num_tokens": 52854235.0,
      "step": 24425
    },
    {
      "entropy": 6.539495992660522,
      "epoch": 2.614265075712986,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004323648726071392,
      "loss": 5.923,
      "mean_token_accuracy": 0.16695099622011184,
      "num_tokens": 52864693.0,
      "step": 24430
    },
    {
      "entropy": 6.493974876403809,
      "epoch": 2.6148001498207503,
      "grad_norm": 1.25,
      "learning_rate": 0.00043233754376739583,
      "loss": 5.8994,
      "mean_token_accuracy": 0.16051026731729506,
      "num_tokens": 52876410.0,
      "step": 24435
    },
    {
      "entropy": 6.473673915863037,
      "epoch": 2.615335223928514,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043231021038455297,
      "loss": 5.8723,
      "mean_token_accuracy": 0.1693979948759079,
      "num_tokens": 52887416.0,
      "step": 24440
    },
    {
      "entropy": 6.508442354202271,
      "epoch": 2.615870298036278,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043228287245939995,
      "loss": 5.833,
      "mean_token_accuracy": 0.17217568159103394,
      "num_tokens": 52899191.0,
      "step": 24445
    },
    {
      "entropy": 6.528335857391357,
      "epoch": 2.616405372144042,
      "grad_norm": 1.109375,
      "learning_rate": 0.000432255529992726,
      "loss": 5.8965,
      "mean_token_accuracy": 0.16627942770719528,
      "num_tokens": 52910479.0,
      "step": 24450
    },
    {
      "entropy": 6.522810268402099,
      "epoch": 2.6169404462518058,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043222818298532074,
      "loss": 5.8075,
      "mean_token_accuracy": 0.16910222172737122,
      "num_tokens": 52921121.0,
      "step": 24455
    },
    {
      "entropy": 6.468616008758545,
      "epoch": 2.61747552035957,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043220083143797385,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16274162158370017,
      "num_tokens": 52931604.0,
      "step": 24460
    },
    {
      "entropy": 6.424640750885009,
      "epoch": 2.618010594467334,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000432173475351475,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16202551275491714,
      "num_tokens": 52942200.0,
      "step": 24465
    },
    {
      "entropy": 6.480301189422607,
      "epoch": 2.6185456685750976,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004321461147266141,
      "loss": 5.8353,
      "mean_token_accuracy": 0.16690368801355362,
      "num_tokens": 52954350.0,
      "step": 24470
    },
    {
      "entropy": 6.497279644012451,
      "epoch": 2.6190807426828617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043211874956418134,
      "loss": 5.8525,
      "mean_token_accuracy": 0.1788276821374893,
      "num_tokens": 52965683.0,
      "step": 24475
    },
    {
      "entropy": 6.4870672702789305,
      "epoch": 2.6196158167906254,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004320913798649667,
      "loss": 5.8485,
      "mean_token_accuracy": 0.1698323130607605,
      "num_tokens": 52975463.0,
      "step": 24480
    },
    {
      "entropy": 6.444397926330566,
      "epoch": 2.6201508908983895,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004320640056297605,
      "loss": 5.843,
      "mean_token_accuracy": 0.17257820516824723,
      "num_tokens": 52986115.0,
      "step": 24485
    },
    {
      "entropy": 6.492922687530518,
      "epoch": 2.620685965006153,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004320366268593532,
      "loss": 5.8376,
      "mean_token_accuracy": 0.16835754364728928,
      "num_tokens": 52995770.0,
      "step": 24490
    },
    {
      "entropy": 6.498361015319825,
      "epoch": 2.6212210391139172,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043200924355453544,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16593496054410933,
      "num_tokens": 53006490.0,
      "step": 24495
    },
    {
      "entropy": 6.465487909317017,
      "epoch": 2.6217561132216813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004319818557160978,
      "loss": 5.7396,
      "mean_token_accuracy": 0.1765119642019272,
      "num_tokens": 53016129.0,
      "step": 24500
    },
    {
      "entropy": 6.473629093170166,
      "epoch": 2.622291187329445,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043195446334483116,
      "loss": 5.824,
      "mean_token_accuracy": 0.170784991979599,
      "num_tokens": 53026568.0,
      "step": 24505
    },
    {
      "entropy": 6.490683603286743,
      "epoch": 2.622826261437209,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043192706644152646,
      "loss": 5.798,
      "mean_token_accuracy": 0.174462029337883,
      "num_tokens": 53036488.0,
      "step": 24510
    },
    {
      "entropy": 6.5438232421875,
      "epoch": 2.623361335544973,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004318996650069747,
      "loss": 5.8613,
      "mean_token_accuracy": 0.16755947917699815,
      "num_tokens": 53047195.0,
      "step": 24515
    },
    {
      "entropy": 6.525188207626343,
      "epoch": 2.623896409652737,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004318722590419672,
      "loss": 5.9349,
      "mean_token_accuracy": 0.15790549516677857,
      "num_tokens": 53058602.0,
      "step": 24520
    },
    {
      "entropy": 6.527998208999634,
      "epoch": 2.624431483760501,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004318448485472953,
      "loss": 5.904,
      "mean_token_accuracy": 0.16398152858018875,
      "num_tokens": 53070822.0,
      "step": 24525
    },
    {
      "entropy": 6.451700830459595,
      "epoch": 2.6249665578682646,
      "grad_norm": 1.640625,
      "learning_rate": 0.00043181743352375035,
      "loss": 5.8469,
      "mean_token_accuracy": 0.1685868576169014,
      "num_tokens": 53082166.0,
      "step": 24530
    },
    {
      "entropy": 6.514977693557739,
      "epoch": 2.6255016319760287,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000431790013972124,
      "loss": 6.017,
      "mean_token_accuracy": 0.16159932613372802,
      "num_tokens": 53093688.0,
      "step": 24535
    },
    {
      "entropy": 6.500256967544556,
      "epoch": 2.6260367060837924,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004317625898932082,
      "loss": 5.8477,
      "mean_token_accuracy": 0.17211787700653075,
      "num_tokens": 53104793.0,
      "step": 24540
    },
    {
      "entropy": 6.517303228378296,
      "epoch": 2.6265717801915565,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004317351612877946,
      "loss": 5.8662,
      "mean_token_accuracy": 0.17047973573207856,
      "num_tokens": 53115835.0,
      "step": 24545
    },
    {
      "entropy": 6.453346252441406,
      "epoch": 2.6271068542993206,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004317077281566751,
      "loss": 5.7329,
      "mean_token_accuracy": 0.180822491645813,
      "num_tokens": 53127153.0,
      "step": 24550
    },
    {
      "entropy": 6.429151725769043,
      "epoch": 2.627641928407084,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043168029050064214,
      "loss": 5.7714,
      "mean_token_accuracy": 0.1734142005443573,
      "num_tokens": 53137323.0,
      "step": 24555
    },
    {
      "entropy": 6.47566556930542,
      "epoch": 2.6281770025148483,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043165284832048773,
      "loss": 5.9222,
      "mean_token_accuracy": 0.16612547487020493,
      "num_tokens": 53146858.0,
      "step": 24560
    },
    {
      "entropy": 6.420253705978394,
      "epoch": 2.6287120766226124,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043162540161700434,
      "loss": 5.8651,
      "mean_token_accuracy": 0.16648954302072524,
      "num_tokens": 53157225.0,
      "step": 24565
    },
    {
      "entropy": 6.501774406433105,
      "epoch": 2.629247150730376,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043159795039098453,
      "loss": 5.8434,
      "mean_token_accuracy": 0.16852261275053024,
      "num_tokens": 53167571.0,
      "step": 24570
    },
    {
      "entropy": 6.486942720413208,
      "epoch": 2.62978222483814,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004315704946432209,
      "loss": 5.7986,
      "mean_token_accuracy": 0.17439942210912704,
      "num_tokens": 53178562.0,
      "step": 24575
    },
    {
      "entropy": 6.529079961776733,
      "epoch": 2.6303172989459043,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043154303437450617,
      "loss": 5.918,
      "mean_token_accuracy": 0.1614033207297325,
      "num_tokens": 53188833.0,
      "step": 24580
    },
    {
      "entropy": 6.512873601913452,
      "epoch": 2.630852373053668,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043151556958563337,
      "loss": 5.8737,
      "mean_token_accuracy": 0.16899704486131667,
      "num_tokens": 53199596.0,
      "step": 24585
    },
    {
      "entropy": 6.526838397979736,
      "epoch": 2.6313874471614316,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004314881002773955,
      "loss": 5.8465,
      "mean_token_accuracy": 0.16576128005981444,
      "num_tokens": 53209942.0,
      "step": 24590
    },
    {
      "entropy": 6.545229816436768,
      "epoch": 2.6319225212691957,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004314606264505857,
      "loss": 5.9311,
      "mean_token_accuracy": 0.16597942858934403,
      "num_tokens": 53221546.0,
      "step": 24595
    },
    {
      "entropy": 6.530277729034424,
      "epoch": 2.63245759537696,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004314331481059973,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16575371325016022,
      "num_tokens": 53232343.0,
      "step": 24600
    },
    {
      "entropy": 6.471540832519532,
      "epoch": 2.6329926694847234,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004314056652444237,
      "loss": 5.8635,
      "mean_token_accuracy": 0.17661373764276506,
      "num_tokens": 53242820.0,
      "step": 24605
    },
    {
      "entropy": 6.57922511100769,
      "epoch": 2.6335277435924875,
      "grad_norm": 1.125,
      "learning_rate": 0.00043137817786665863,
      "loss": 5.8689,
      "mean_token_accuracy": 0.17170093357563018,
      "num_tokens": 53253503.0,
      "step": 24610
    },
    {
      "entropy": 6.571479654312133,
      "epoch": 2.6340628177002516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043135068597349545,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16944090276956558,
      "num_tokens": 53264160.0,
      "step": 24615
    },
    {
      "entropy": 6.498293828964234,
      "epoch": 2.6345978918080153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004313231895657283,
      "loss": 5.8515,
      "mean_token_accuracy": 0.16336987763643265,
      "num_tokens": 53275002.0,
      "step": 24620
    },
    {
      "entropy": 6.419729137420655,
      "epoch": 2.6351329659157794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043129568864415103,
      "loss": 5.7582,
      "mean_token_accuracy": 0.17796950936317443,
      "num_tokens": 53284216.0,
      "step": 24625
    },
    {
      "entropy": 6.4468522548675535,
      "epoch": 2.6356680400235435,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043126818320955766,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17550437301397323,
      "num_tokens": 53294277.0,
      "step": 24630
    },
    {
      "entropy": 6.499338150024414,
      "epoch": 2.636203114131307,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004312406732627425,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1642283320426941,
      "num_tokens": 53304448.0,
      "step": 24635
    },
    {
      "entropy": 6.47317590713501,
      "epoch": 2.6367381882390712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043121315880449983,
      "loss": 5.8121,
      "mean_token_accuracy": 0.16713852137327195,
      "num_tokens": 53315765.0,
      "step": 24640
    },
    {
      "entropy": 6.3820915699005125,
      "epoch": 2.637273262346835,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043118563983562415,
      "loss": 5.7096,
      "mean_token_accuracy": 0.17920370995998383,
      "num_tokens": 53326529.0,
      "step": 24645
    },
    {
      "entropy": 6.423469209671021,
      "epoch": 2.637808336454599,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043115811635691006,
      "loss": 5.7135,
      "mean_token_accuracy": 0.17902804166078568,
      "num_tokens": 53336132.0,
      "step": 24650
    },
    {
      "entropy": 6.428691291809082,
      "epoch": 2.6383434105623627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004311305883691524,
      "loss": 5.8564,
      "mean_token_accuracy": 0.17140955626964569,
      "num_tokens": 53347284.0,
      "step": 24655
    },
    {
      "entropy": 6.4407639503479,
      "epoch": 2.6388784846701268,
      "grad_norm": 1.125,
      "learning_rate": 0.0004311030558731458,
      "loss": 5.8653,
      "mean_token_accuracy": 0.172186042368412,
      "num_tokens": 53357872.0,
      "step": 24660
    },
    {
      "entropy": 6.481744194030762,
      "epoch": 2.639413558777891,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043107551886968545,
      "loss": 5.7985,
      "mean_token_accuracy": 0.1751430869102478,
      "num_tokens": 53370311.0,
      "step": 24665
    },
    {
      "entropy": 6.4570966243743895,
      "epoch": 2.6399486328856545,
      "grad_norm": 1.125,
      "learning_rate": 0.00043104797735956643,
      "loss": 5.7045,
      "mean_token_accuracy": 0.18929157555103301,
      "num_tokens": 53380320.0,
      "step": 24670
    },
    {
      "entropy": 6.520108556747436,
      "epoch": 2.6404837069934186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000431020431343584,
      "loss": 5.9588,
      "mean_token_accuracy": 0.1626199647784233,
      "num_tokens": 53391311.0,
      "step": 24675
    },
    {
      "entropy": 6.457558059692383,
      "epoch": 2.6410187811011827,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004309928808225335,
      "loss": 5.9291,
      "mean_token_accuracy": 0.16280593276023864,
      "num_tokens": 53402367.0,
      "step": 24680
    },
    {
      "entropy": 6.462678050994873,
      "epoch": 2.6415538552089464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004309653257972105,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16738794147968292,
      "num_tokens": 53414253.0,
      "step": 24685
    },
    {
      "entropy": 6.53260350227356,
      "epoch": 2.6420889293167105,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043093776626841075,
      "loss": 5.8491,
      "mean_token_accuracy": 0.1742934063076973,
      "num_tokens": 53425727.0,
      "step": 24690
    },
    {
      "entropy": 6.532307147979736,
      "epoch": 2.6426240034244746,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043091020223692974,
      "loss": 5.8603,
      "mean_token_accuracy": 0.1693821832537651,
      "num_tokens": 53435402.0,
      "step": 24695
    },
    {
      "entropy": 6.451574230194092,
      "epoch": 2.643159077532238,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043088263370356364,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17745147198438643,
      "num_tokens": 53447421.0,
      "step": 24700
    },
    {
      "entropy": 6.479432392120361,
      "epoch": 2.643694151640002,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004308550606691084,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1787162870168686,
      "num_tokens": 53458346.0,
      "step": 24705
    },
    {
      "entropy": 6.469180583953857,
      "epoch": 2.644229225747766,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004308274831343601,
      "loss": 5.8522,
      "mean_token_accuracy": 0.17411617040634156,
      "num_tokens": 53469638.0,
      "step": 24710
    },
    {
      "entropy": 6.553719234466553,
      "epoch": 2.64476429985553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043079990110011525,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1662727952003479,
      "num_tokens": 53480645.0,
      "step": 24715
    },
    {
      "entropy": 6.506572675704956,
      "epoch": 2.6452993739632937,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043077231456717006,
      "loss": 5.8354,
      "mean_token_accuracy": 0.17024810910224913,
      "num_tokens": 53491958.0,
      "step": 24720
    },
    {
      "entropy": 6.4893004417419435,
      "epoch": 2.645834448071058,
      "grad_norm": 1.125,
      "learning_rate": 0.00043074472353632124,
      "loss": 5.7985,
      "mean_token_accuracy": 0.17130138128995895,
      "num_tokens": 53502183.0,
      "step": 24725
    },
    {
      "entropy": 6.458721065521241,
      "epoch": 2.646369522178822,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004307171280083653,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16620871126651765,
      "num_tokens": 53513373.0,
      "step": 24730
    },
    {
      "entropy": 6.431275701522827,
      "epoch": 2.6469045962865856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004306895279840992,
      "loss": 5.7464,
      "mean_token_accuracy": 0.17533878237009048,
      "num_tokens": 53523668.0,
      "step": 24735
    },
    {
      "entropy": 6.536407327651977,
      "epoch": 2.6474396703943497,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004306619234643199,
      "loss": 5.8689,
      "mean_token_accuracy": 0.16812394559383392,
      "num_tokens": 53535642.0,
      "step": 24740
    },
    {
      "entropy": 6.34388313293457,
      "epoch": 2.647974744502114,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043063431444982447,
      "loss": 5.612,
      "mean_token_accuracy": 0.18642182350158693,
      "num_tokens": 53546192.0,
      "step": 24745
    },
    {
      "entropy": 6.43807373046875,
      "epoch": 2.6485098186098774,
      "grad_norm": 1.125,
      "learning_rate": 0.00043060670094141005,
      "loss": 5.8178,
      "mean_token_accuracy": 0.17271072417497635,
      "num_tokens": 53557375.0,
      "step": 24750
    },
    {
      "entropy": 6.5330808639526365,
      "epoch": 2.6490448927176415,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043057908293987404,
      "loss": 5.8879,
      "mean_token_accuracy": 0.16451967209577562,
      "num_tokens": 53568434.0,
      "step": 24755
    },
    {
      "entropy": 6.491071939468384,
      "epoch": 2.649579966825405,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004305514604460138,
      "loss": 5.8231,
      "mean_token_accuracy": 0.17302296906709672,
      "num_tokens": 53579095.0,
      "step": 24760
    },
    {
      "entropy": 6.461151552200318,
      "epoch": 2.6501150409331693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000430523833460627,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16958325505256652,
      "num_tokens": 53590082.0,
      "step": 24765
    },
    {
      "entropy": 6.491979598999023,
      "epoch": 2.650650115040933,
      "grad_norm": 1.25,
      "learning_rate": 0.00043049620198451147,
      "loss": 5.9035,
      "mean_token_accuracy": 0.16561015397310258,
      "num_tokens": 53600421.0,
      "step": 24770
    },
    {
      "entropy": 6.383750486373901,
      "epoch": 2.651185189148697,
      "grad_norm": 1.125,
      "learning_rate": 0.00043046856601846483,
      "loss": 5.772,
      "mean_token_accuracy": 0.17940492928028107,
      "num_tokens": 53610994.0,
      "step": 24775
    },
    {
      "entropy": 6.50364933013916,
      "epoch": 2.651720263256461,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043044092556328517,
      "loss": 5.8651,
      "mean_token_accuracy": 0.16764414608478545,
      "num_tokens": 53622205.0,
      "step": 24780
    },
    {
      "entropy": 6.451356410980225,
      "epoch": 2.652255337364225,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043041328061977073,
      "loss": 5.7927,
      "mean_token_accuracy": 0.17521774768829346,
      "num_tokens": 53632440.0,
      "step": 24785
    },
    {
      "entropy": 6.440392112731933,
      "epoch": 2.652790411471989,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043038563118871965,
      "loss": 5.8005,
      "mean_token_accuracy": 0.1716696098446846,
      "num_tokens": 53642443.0,
      "step": 24790
    },
    {
      "entropy": 6.586670780181885,
      "epoch": 2.653325485579753,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004303579772709303,
      "loss": 5.9769,
      "mean_token_accuracy": 0.15685766488313674,
      "num_tokens": 53653671.0,
      "step": 24795
    },
    {
      "entropy": 6.517107677459717,
      "epoch": 2.6538605596875167,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043033031886720107,
      "loss": 5.7833,
      "mean_token_accuracy": 0.17525545358657837,
      "num_tokens": 53664874.0,
      "step": 24800
    },
    {
      "entropy": 6.503127479553223,
      "epoch": 2.6543956337952808,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004303026559783309,
      "loss": 5.8712,
      "mean_token_accuracy": 0.16410054713487626,
      "num_tokens": 53675623.0,
      "step": 24805
    },
    {
      "entropy": 6.518387365341186,
      "epoch": 2.6549307079030444,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004302749886051182,
      "loss": 5.8502,
      "mean_token_accuracy": 0.16392042189836503,
      "num_tokens": 53685754.0,
      "step": 24810
    },
    {
      "entropy": 6.46727294921875,
      "epoch": 2.6554657820108085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004302473167483621,
      "loss": 5.7503,
      "mean_token_accuracy": 0.17042961716651917,
      "num_tokens": 53696925.0,
      "step": 24815
    },
    {
      "entropy": 6.475355052947998,
      "epoch": 2.656000856118572,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043021964040886154,
      "loss": 5.8143,
      "mean_token_accuracy": 0.1690433457493782,
      "num_tokens": 53708046.0,
      "step": 24820
    },
    {
      "entropy": 6.392120885848999,
      "epoch": 2.6565359302263363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004301919595874156,
      "loss": 5.7048,
      "mean_token_accuracy": 0.18533088713884355,
      "num_tokens": 53719179.0,
      "step": 24825
    },
    {
      "entropy": 6.4848379611969,
      "epoch": 2.6570710043341004,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004301642742848237,
      "loss": 5.8535,
      "mean_token_accuracy": 0.17006590515375136,
      "num_tokens": 53731523.0,
      "step": 24830
    },
    {
      "entropy": 6.546422481536865,
      "epoch": 2.657606078441864,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004301365845018853,
      "loss": 5.9599,
      "mean_token_accuracy": 0.1590213105082512,
      "num_tokens": 53742072.0,
      "step": 24835
    },
    {
      "entropy": 6.523851013183593,
      "epoch": 2.658141152549628,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004301088902393996,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16473958194255828,
      "num_tokens": 53754695.0,
      "step": 24840
    },
    {
      "entropy": 6.431696891784668,
      "epoch": 2.6586762266573922,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004300811914981666,
      "loss": 5.7558,
      "mean_token_accuracy": 0.18142683953046798,
      "num_tokens": 53764587.0,
      "step": 24845
    },
    {
      "entropy": 6.5478120803833,
      "epoch": 2.659211300765156,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000430053488278986,
      "loss": 5.9374,
      "mean_token_accuracy": 0.1579838737845421,
      "num_tokens": 53775609.0,
      "step": 24850
    },
    {
      "entropy": 6.458367443084716,
      "epoch": 2.65974637487292,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043002578058265776,
      "loss": 5.8192,
      "mean_token_accuracy": 0.17234109789133073,
      "num_tokens": 53786495.0,
      "step": 24855
    },
    {
      "entropy": 6.483634948730469,
      "epoch": 2.660281448980684,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004299980684099818,
      "loss": 5.8304,
      "mean_token_accuracy": 0.1714881479740143,
      "num_tokens": 53797058.0,
      "step": 24860
    },
    {
      "entropy": 6.556471586227417,
      "epoch": 2.6608165230884477,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042997035176175834,
      "loss": 5.862,
      "mean_token_accuracy": 0.168348428606987,
      "num_tokens": 53808148.0,
      "step": 24865
    },
    {
      "entropy": 6.487131786346436,
      "epoch": 2.661351597196212,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004299426306387879,
      "loss": 5.778,
      "mean_token_accuracy": 0.174246084690094,
      "num_tokens": 53818375.0,
      "step": 24870
    },
    {
      "entropy": 6.529974317550659,
      "epoch": 2.6618866713039755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004299149050418707,
      "loss": 5.8988,
      "mean_token_accuracy": 0.16705805957317352,
      "num_tokens": 53831273.0,
      "step": 24875
    },
    {
      "entropy": 6.48136715888977,
      "epoch": 2.6624217454117396,
      "grad_norm": 1.125,
      "learning_rate": 0.00042988717497180733,
      "loss": 5.8003,
      "mean_token_accuracy": 0.17100610136985778,
      "num_tokens": 53842492.0,
      "step": 24880
    },
    {
      "entropy": 6.496687650680542,
      "epoch": 2.6629568195195032,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042985944042939867,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1671614557504654,
      "num_tokens": 53853017.0,
      "step": 24885
    },
    {
      "entropy": 6.560709142684937,
      "epoch": 2.6634918936272673,
      "grad_norm": 1.25,
      "learning_rate": 0.00042983170141544535,
      "loss": 5.9283,
      "mean_token_accuracy": 0.1673526257276535,
      "num_tokens": 53863606.0,
      "step": 24890
    },
    {
      "entropy": 6.510554552078247,
      "epoch": 2.6640269677350314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004298039579307484,
      "loss": 5.8573,
      "mean_token_accuracy": 0.16996636241674423,
      "num_tokens": 53874689.0,
      "step": 24895
    },
    {
      "entropy": 6.536205768585205,
      "epoch": 2.664562041842795,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000429776209976109,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17557397931814195,
      "num_tokens": 53885247.0,
      "step": 24900
    },
    {
      "entropy": 6.487976980209351,
      "epoch": 2.665097115950559,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004297484575523282,
      "loss": 5.8959,
      "mean_token_accuracy": 0.16575776785612106,
      "num_tokens": 53896829.0,
      "step": 24905
    },
    {
      "entropy": 6.4522346496582035,
      "epoch": 2.6656321900583233,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042972070066020745,
      "loss": 5.7863,
      "mean_token_accuracy": 0.17409384697675706,
      "num_tokens": 53907236.0,
      "step": 24910
    },
    {
      "entropy": 6.535991907119751,
      "epoch": 2.666167264166087,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004296929393005482,
      "loss": 5.852,
      "mean_token_accuracy": 0.16546559780836106,
      "num_tokens": 53918420.0,
      "step": 24915
    },
    {
      "entropy": 6.4294883728027346,
      "epoch": 2.666702338273851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004296651734741521,
      "loss": 5.7918,
      "mean_token_accuracy": 0.18101671636104583,
      "num_tokens": 53928949.0,
      "step": 24920
    },
    {
      "entropy": 6.488697576522827,
      "epoch": 2.6672374123816147,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042963740318182085,
      "loss": 5.9037,
      "mean_token_accuracy": 0.1645342156291008,
      "num_tokens": 53939332.0,
      "step": 24925
    },
    {
      "entropy": 6.4832343578338625,
      "epoch": 2.667772486489379,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004296096284243563,
      "loss": 5.8212,
      "mean_token_accuracy": 0.17199569642543794,
      "num_tokens": 53950703.0,
      "step": 24930
    },
    {
      "entropy": 6.578641414642334,
      "epoch": 2.6683075605971425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004295818492025605,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16285451352596284,
      "num_tokens": 53964061.0,
      "step": 24935
    },
    {
      "entropy": 6.491735219955444,
      "epoch": 2.6688426347049066,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004295540655172355,
      "loss": 5.9145,
      "mean_token_accuracy": 0.16970796287059783,
      "num_tokens": 53974980.0,
      "step": 24940
    },
    {
      "entropy": 6.402004146575928,
      "epoch": 2.6693777088126707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004295262773691835,
      "loss": 5.7737,
      "mean_token_accuracy": 0.17489724457263947,
      "num_tokens": 53985437.0,
      "step": 24945
    },
    {
      "entropy": 6.5118732929229735,
      "epoch": 2.6699127829204343,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042949848475920706,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16710035353899003,
      "num_tokens": 53996952.0,
      "step": 24950
    },
    {
      "entropy": 6.536525535583496,
      "epoch": 2.6704478570281984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004294706876881085,
      "loss": 5.8107,
      "mean_token_accuracy": 0.1743543937802315,
      "num_tokens": 54007727.0,
      "step": 24955
    },
    {
      "entropy": 6.423408460617066,
      "epoch": 2.6709829311359625,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042944288615669055,
      "loss": 5.7186,
      "mean_token_accuracy": 0.18078698366880416,
      "num_tokens": 54018275.0,
      "step": 24960
    },
    {
      "entropy": 6.458427810668946,
      "epoch": 2.671518005243726,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042941508016575597,
      "loss": 5.947,
      "mean_token_accuracy": 0.165918031334877,
      "num_tokens": 54030376.0,
      "step": 24965
    },
    {
      "entropy": 6.479463720321656,
      "epoch": 2.6720530793514903,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004293872697161077,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1658268854022026,
      "num_tokens": 54041487.0,
      "step": 24970
    },
    {
      "entropy": 6.528324699401855,
      "epoch": 2.6725881534592544,
      "grad_norm": 1.25,
      "learning_rate": 0.00042935945480854866,
      "loss": 5.799,
      "mean_token_accuracy": 0.18159285485744475,
      "num_tokens": 54051273.0,
      "step": 24975
    },
    {
      "entropy": 6.475639867782593,
      "epoch": 2.673123227567018,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004293316354438821,
      "loss": 5.7887,
      "mean_token_accuracy": 0.17276590466499328,
      "num_tokens": 54061910.0,
      "step": 24980
    },
    {
      "entropy": 6.509027290344238,
      "epoch": 2.6736583016747817,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004293038116229112,
      "loss": 5.9072,
      "mean_token_accuracy": 0.1643783852458,
      "num_tokens": 54072980.0,
      "step": 24985
    },
    {
      "entropy": 6.470298719406128,
      "epoch": 2.674193375782546,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004292759833464394,
      "loss": 5.8199,
      "mean_token_accuracy": 0.17258767187595367,
      "num_tokens": 54083027.0,
      "step": 24990
    },
    {
      "entropy": 6.424474287033081,
      "epoch": 2.67472844989031,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042924815061527033,
      "loss": 5.775,
      "mean_token_accuracy": 0.17036359161138534,
      "num_tokens": 54094068.0,
      "step": 24995
    },
    {
      "entropy": 6.535147190093994,
      "epoch": 2.6752635239980735,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004292203134302076,
      "loss": 5.8437,
      "mean_token_accuracy": 0.16679221838712693,
      "num_tokens": 54104683.0,
      "step": 25000
    },
    {
      "entropy": 6.491231679916382,
      "epoch": 2.6757985981058376,
      "grad_norm": 1.171875,
      "learning_rate": 0.000429192471792055,
      "loss": 5.7977,
      "mean_token_accuracy": 0.1696283519268036,
      "num_tokens": 54115357.0,
      "step": 25005
    },
    {
      "entropy": 6.511938095092773,
      "epoch": 2.6763336722136017,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004291646257016163,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16545387208461762,
      "num_tokens": 54126410.0,
      "step": 25010
    },
    {
      "entropy": 6.445648908615112,
      "epoch": 2.6768687463213654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042913677515969586,
      "loss": 5.7629,
      "mean_token_accuracy": 0.1768421247601509,
      "num_tokens": 54137279.0,
      "step": 25015
    },
    {
      "entropy": 6.480887937545776,
      "epoch": 2.6774038204291295,
      "grad_norm": 1.25,
      "learning_rate": 0.0004291089201670976,
      "loss": 5.8679,
      "mean_token_accuracy": 0.17247879952192308,
      "num_tokens": 54146521.0,
      "step": 25020
    },
    {
      "entropy": 6.5315749168396,
      "epoch": 2.6779388945368936,
      "grad_norm": 1.125,
      "learning_rate": 0.00042908106072462595,
      "loss": 5.9089,
      "mean_token_accuracy": 0.15747358053922653,
      "num_tokens": 54157277.0,
      "step": 25025
    },
    {
      "entropy": 6.545756244659424,
      "epoch": 2.6784739686446573,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004290531968330853,
      "loss": 5.9278,
      "mean_token_accuracy": 0.16060391068458557,
      "num_tokens": 54168078.0,
      "step": 25030
    },
    {
      "entropy": 6.562824440002442,
      "epoch": 2.6790090427524214,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004290253284932803,
      "loss": 5.8741,
      "mean_token_accuracy": 0.16845367699861527,
      "num_tokens": 54178567.0,
      "step": 25035
    },
    {
      "entropy": 6.483724117279053,
      "epoch": 2.679544116860185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004289974557060155,
      "loss": 5.8209,
      "mean_token_accuracy": 0.1669696241617203,
      "num_tokens": 54189485.0,
      "step": 25040
    },
    {
      "entropy": 6.567569780349731,
      "epoch": 2.680079190967949,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042896957847209593,
      "loss": 5.9255,
      "mean_token_accuracy": 0.16425575017929078,
      "num_tokens": 54199351.0,
      "step": 25045
    },
    {
      "entropy": 6.520831108093262,
      "epoch": 2.6806142650757128,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004289416967923264,
      "loss": 5.8647,
      "mean_token_accuracy": 0.16455538272857667,
      "num_tokens": 54209001.0,
      "step": 25050
    },
    {
      "entropy": 6.548098707199097,
      "epoch": 2.681149339183477,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042891381066751186,
      "loss": 5.9025,
      "mean_token_accuracy": 0.1649604395031929,
      "num_tokens": 54219783.0,
      "step": 25055
    },
    {
      "entropy": 6.530045509338379,
      "epoch": 2.681684413291241,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042888592009845776,
      "loss": 5.8522,
      "mean_token_accuracy": 0.17061162292957305,
      "num_tokens": 54231337.0,
      "step": 25060
    },
    {
      "entropy": 6.512972259521485,
      "epoch": 2.6822194873990046,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004288580250859694,
      "loss": 5.8668,
      "mean_token_accuracy": 0.17322382926940919,
      "num_tokens": 54241650.0,
      "step": 25065
    },
    {
      "entropy": 6.4891845703125,
      "epoch": 2.6827545615067687,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004288301256308521,
      "loss": 5.8857,
      "mean_token_accuracy": 0.16169488728046416,
      "num_tokens": 54252223.0,
      "step": 25070
    },
    {
      "entropy": 6.52498197555542,
      "epoch": 2.683289635614533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042880222173391155,
      "loss": 5.7956,
      "mean_token_accuracy": 0.17561921924352647,
      "num_tokens": 54262469.0,
      "step": 25075
    },
    {
      "entropy": 6.509916830062866,
      "epoch": 2.6838247097222965,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004287743133959535,
      "loss": 5.844,
      "mean_token_accuracy": 0.1692213460803032,
      "num_tokens": 54272885.0,
      "step": 25080
    },
    {
      "entropy": 6.488689422607422,
      "epoch": 2.6843597838300606,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004287464006177837,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16562828719615935,
      "num_tokens": 54283327.0,
      "step": 25085
    },
    {
      "entropy": 6.472698497772217,
      "epoch": 2.6848948579378242,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004287184834002082,
      "loss": 5.7843,
      "mean_token_accuracy": 0.17515695542097093,
      "num_tokens": 54294761.0,
      "step": 25090
    },
    {
      "entropy": 6.545413494110107,
      "epoch": 2.6854299320455883,
      "grad_norm": 1.125,
      "learning_rate": 0.00042869056174403303,
      "loss": 5.8526,
      "mean_token_accuracy": 0.17302776128053665,
      "num_tokens": 54305205.0,
      "step": 25095
    },
    {
      "entropy": 6.521032094955444,
      "epoch": 2.685965006153352,
      "grad_norm": 1.125,
      "learning_rate": 0.0004286626356500647,
      "loss": 5.873,
      "mean_token_accuracy": 0.16858349293470382,
      "num_tokens": 54315902.0,
      "step": 25100
    },
    {
      "entropy": 6.51122899055481,
      "epoch": 2.686500080261116,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004286347051191091,
      "loss": 5.8659,
      "mean_token_accuracy": 0.168125618994236,
      "num_tokens": 54327345.0,
      "step": 25105
    },
    {
      "entropy": 6.528494501113892,
      "epoch": 2.68703515436888,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286067701519731,
      "loss": 5.9246,
      "mean_token_accuracy": 0.1634068787097931,
      "num_tokens": 54337548.0,
      "step": 25110
    },
    {
      "entropy": 6.543246555328369,
      "epoch": 2.687570228476644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042857883074946324,
      "loss": 5.8814,
      "mean_token_accuracy": 0.16154275983572006,
      "num_tokens": 54348384.0,
      "step": 25115
    },
    {
      "entropy": 6.545587587356567,
      "epoch": 2.688105302584408,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004285508869123862,
      "loss": 5.843,
      "mean_token_accuracy": 0.16249530464410783,
      "num_tokens": 54358245.0,
      "step": 25120
    },
    {
      "entropy": 6.4134774684906,
      "epoch": 2.688640376692172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042852293864154885,
      "loss": 5.7973,
      "mean_token_accuracy": 0.16773589849472045,
      "num_tokens": 54370421.0,
      "step": 25125
    },
    {
      "entropy": 6.524346876144409,
      "epoch": 2.6891754507999357,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042849498593775827,
      "loss": 5.8,
      "mean_token_accuracy": 0.1672895848751068,
      "num_tokens": 54381368.0,
      "step": 25130
    },
    {
      "entropy": 6.5182935237884525,
      "epoch": 2.6897105249077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042846702880182147,
      "loss": 5.845,
      "mean_token_accuracy": 0.16478429436683656,
      "num_tokens": 54392226.0,
      "step": 25135
    },
    {
      "entropy": 6.458339548110962,
      "epoch": 2.690245599015464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004284390672345458,
      "loss": 5.8464,
      "mean_token_accuracy": 0.16413207352161407,
      "num_tokens": 54402295.0,
      "step": 25140
    },
    {
      "entropy": 6.5140852451324465,
      "epoch": 2.6907806731232276,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004284111012367387,
      "loss": 5.8988,
      "mean_token_accuracy": 0.16267080903053283,
      "num_tokens": 54413808.0,
      "step": 25145
    },
    {
      "entropy": 6.52843451499939,
      "epoch": 2.6913157472309917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042838313080920755,
      "loss": 5.9013,
      "mean_token_accuracy": 0.1648396298289299,
      "num_tokens": 54425048.0,
      "step": 25150
    },
    {
      "entropy": 6.573152160644531,
      "epoch": 2.6918508213387553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042835515595276,
      "loss": 5.9111,
      "mean_token_accuracy": 0.16282006353139877,
      "num_tokens": 54435908.0,
      "step": 25155
    },
    {
      "entropy": 6.432372617721557,
      "epoch": 2.6923858954465194,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004283271766682039,
      "loss": 5.8199,
      "mean_token_accuracy": 0.17771984487771988,
      "num_tokens": 54447594.0,
      "step": 25160
    },
    {
      "entropy": 6.437346696853638,
      "epoch": 2.692920969554283,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004282991929563471,
      "loss": 5.7694,
      "mean_token_accuracy": 0.1718212753534317,
      "num_tokens": 54458918.0,
      "step": 25165
    },
    {
      "entropy": 6.442570590972901,
      "epoch": 2.693456043662047,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004282712048179977,
      "loss": 5.7815,
      "mean_token_accuracy": 0.1762408956885338,
      "num_tokens": 54469423.0,
      "step": 25170
    },
    {
      "entropy": 6.518641233444214,
      "epoch": 2.6939911177698113,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004282432122539638,
      "loss": 5.9114,
      "mean_token_accuracy": 0.16925821900367738,
      "num_tokens": 54481895.0,
      "step": 25175
    },
    {
      "entropy": 6.537839126586914,
      "epoch": 2.694526191877575,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042821521526505363,
      "loss": 5.8119,
      "mean_token_accuracy": 0.16898961663246154,
      "num_tokens": 54492233.0,
      "step": 25180
    },
    {
      "entropy": 6.53663477897644,
      "epoch": 2.695061265985339,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004281872138520757,
      "loss": 5.8072,
      "mean_token_accuracy": 0.17124247550964355,
      "num_tokens": 54502366.0,
      "step": 25185
    },
    {
      "entropy": 6.45349349975586,
      "epoch": 2.695596340093103,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042815920801583845,
      "loss": 5.8426,
      "mean_token_accuracy": 0.17050666958093644,
      "num_tokens": 54512565.0,
      "step": 25190
    },
    {
      "entropy": 6.4871245384216305,
      "epoch": 2.6961314142008668,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042813119775715066,
      "loss": 5.8331,
      "mean_token_accuracy": 0.17221689522266387,
      "num_tokens": 54522999.0,
      "step": 25195
    },
    {
      "entropy": 6.498887538909912,
      "epoch": 2.696666488308631,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042810318307682095,
      "loss": 5.8487,
      "mean_token_accuracy": 0.16688903272151948,
      "num_tokens": 54533776.0,
      "step": 25200
    },
    {
      "entropy": 6.494991827011108,
      "epoch": 2.6972015624163945,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042807516397565845,
      "loss": 5.8376,
      "mean_token_accuracy": 0.17001082748174667,
      "num_tokens": 54544972.0,
      "step": 25205
    },
    {
      "entropy": 6.492968368530273,
      "epoch": 2.6977366365241586,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004280471404544721,
      "loss": 5.8351,
      "mean_token_accuracy": 0.16909375935792922,
      "num_tokens": 54556034.0,
      "step": 25210
    },
    {
      "entropy": 6.525446987152099,
      "epoch": 2.6982717106319223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042801911251407095,
      "loss": 5.7835,
      "mean_token_accuracy": 0.17678810358047486,
      "num_tokens": 54566662.0,
      "step": 25215
    },
    {
      "entropy": 6.491478061676025,
      "epoch": 2.6988067847396864,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004279910801552645,
      "loss": 5.8337,
      "mean_token_accuracy": 0.1662793219089508,
      "num_tokens": 54578889.0,
      "step": 25220
    },
    {
      "entropy": 6.566757011413574,
      "epoch": 2.6993418588474505,
      "grad_norm": 1.125,
      "learning_rate": 0.0004279630433788621,
      "loss": 5.9091,
      "mean_token_accuracy": 0.1697377935051918,
      "num_tokens": 54590456.0,
      "step": 25225
    },
    {
      "entropy": 6.560050964355469,
      "epoch": 2.699876932955214,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004279350021856734,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1634875863790512,
      "num_tokens": 54602327.0,
      "step": 25230
    },
    {
      "entropy": 6.539851140975952,
      "epoch": 2.7004120070629782,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042790695657650787,
      "loss": 5.8281,
      "mean_token_accuracy": 0.16578231006860733,
      "num_tokens": 54612999.0,
      "step": 25235
    },
    {
      "entropy": 6.478379535675049,
      "epoch": 2.7009470811707423,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042787890655217553,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16513761729002,
      "num_tokens": 54624332.0,
      "step": 25240
    },
    {
      "entropy": 6.430241584777832,
      "epoch": 2.701482155278506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042785085211348624,
      "loss": 5.7955,
      "mean_token_accuracy": 0.17026522755622864,
      "num_tokens": 54635679.0,
      "step": 25245
    },
    {
      "entropy": 6.507163667678833,
      "epoch": 2.70201722938627,
      "grad_norm": 1.125,
      "learning_rate": 0.00042782279326125005,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1698881670832634,
      "num_tokens": 54646344.0,
      "step": 25250
    },
    {
      "entropy": 6.567301225662232,
      "epoch": 2.702552303494034,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042779472999627724,
      "loss": 5.8729,
      "mean_token_accuracy": 0.1649868071079254,
      "num_tokens": 54657395.0,
      "step": 25255
    },
    {
      "entropy": 6.456261920928955,
      "epoch": 2.703087377601798,
      "grad_norm": 1.15625,
      "learning_rate": 0.000427766662319378,
      "loss": 5.8303,
      "mean_token_accuracy": 0.17323897778987885,
      "num_tokens": 54668755.0,
      "step": 25260
    },
    {
      "entropy": 6.510009670257569,
      "epoch": 2.7036224517095615,
      "grad_norm": 1.125,
      "learning_rate": 0.00042773859023136285,
      "loss": 5.8182,
      "mean_token_accuracy": 0.17065616250038146,
      "num_tokens": 54679170.0,
      "step": 25265
    },
    {
      "entropy": 6.493105936050415,
      "epoch": 2.7041575258173256,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004277105137330424,
      "loss": 5.846,
      "mean_token_accuracy": 0.16557424515485764,
      "num_tokens": 54688728.0,
      "step": 25270
    },
    {
      "entropy": 6.455563306808472,
      "epoch": 2.7046925999250897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042768243282522735,
      "loss": 5.85,
      "mean_token_accuracy": 0.17051410377025605,
      "num_tokens": 54700236.0,
      "step": 25275
    },
    {
      "entropy": 6.507881832122803,
      "epoch": 2.7052276740328534,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042765434750872847,
      "loss": 5.7972,
      "mean_token_accuracy": 0.17246685475111007,
      "num_tokens": 54710586.0,
      "step": 25280
    },
    {
      "entropy": 6.49490532875061,
      "epoch": 2.7057627481406175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042762625778435676,
      "loss": 5.7794,
      "mean_token_accuracy": 0.1749851182103157,
      "num_tokens": 54721159.0,
      "step": 25285
    },
    {
      "entropy": 6.497365474700928,
      "epoch": 2.7062978222483816,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004275981636529233,
      "loss": 5.8648,
      "mean_token_accuracy": 0.16782157272100448,
      "num_tokens": 54731593.0,
      "step": 25290
    },
    {
      "entropy": 6.457200717926026,
      "epoch": 2.706832896356145,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004275700651152393,
      "loss": 5.7866,
      "mean_token_accuracy": 0.17611806243658065,
      "num_tokens": 54742890.0,
      "step": 25295
    },
    {
      "entropy": 6.405351543426514,
      "epoch": 2.7073679704639093,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042754196217211616,
      "loss": 5.8703,
      "mean_token_accuracy": 0.17105623483657836,
      "num_tokens": 54753344.0,
      "step": 25300
    },
    {
      "entropy": 6.513272047042847,
      "epoch": 2.7079030445716734,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042751385482436523,
      "loss": 5.9235,
      "mean_token_accuracy": 0.16382136642932893,
      "num_tokens": 54763157.0,
      "step": 25305
    },
    {
      "entropy": 6.521832990646362,
      "epoch": 2.708438118679437,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042748574307279824,
      "loss": 5.8102,
      "mean_token_accuracy": 0.1708103358745575,
      "num_tokens": 54772515.0,
      "step": 25310
    },
    {
      "entropy": 6.482995843887329,
      "epoch": 2.708973192787201,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004274576269182268,
      "loss": 5.7922,
      "mean_token_accuracy": 0.17092054337263107,
      "num_tokens": 54783178.0,
      "step": 25315
    },
    {
      "entropy": 6.537156009674073,
      "epoch": 2.709508266894965,
      "grad_norm": 1.125,
      "learning_rate": 0.0004274295063614629,
      "loss": 5.9127,
      "mean_token_accuracy": 0.16822676360607147,
      "num_tokens": 54794709.0,
      "step": 25320
    },
    {
      "entropy": 6.515727806091308,
      "epoch": 2.710043341002729,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004274013814033184,
      "loss": 5.845,
      "mean_token_accuracy": 0.16979381889104844,
      "num_tokens": 54804821.0,
      "step": 25325
    },
    {
      "entropy": 6.539466810226441,
      "epoch": 2.7105784151104926,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042737325204460523,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1646232783794403,
      "num_tokens": 54815291.0,
      "step": 25330
    },
    {
      "entropy": 6.3160240173339846,
      "epoch": 2.7111134892182567,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427345118286136,
      "loss": 5.6689,
      "mean_token_accuracy": 0.18492710441350937,
      "num_tokens": 54826891.0,
      "step": 25335
    },
    {
      "entropy": 6.52972936630249,
      "epoch": 2.7116485633260208,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004273169801287228,
      "loss": 5.8553,
      "mean_token_accuracy": 0.1690157562494278,
      "num_tokens": 54837545.0,
      "step": 25340
    },
    {
      "entropy": 6.535535621643066,
      "epoch": 2.7121836374337844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042728883757317825,
      "loss": 5.9642,
      "mean_token_accuracy": 0.15842620730400087,
      "num_tokens": 54848028.0,
      "step": 25345
    },
    {
      "entropy": 6.518591451644897,
      "epoch": 2.7127187115415485,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042726069062031495,
      "loss": 5.8993,
      "mean_token_accuracy": 0.1651954546570778,
      "num_tokens": 54858679.0,
      "step": 25350
    },
    {
      "entropy": 6.509331178665161,
      "epoch": 2.7132537856493126,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042723253927094565,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1735717847943306,
      "num_tokens": 54870892.0,
      "step": 25355
    },
    {
      "entropy": 6.498851776123047,
      "epoch": 2.7137888597570763,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000427204383525883,
      "loss": 5.7619,
      "mean_token_accuracy": 0.17019272595643997,
      "num_tokens": 54881131.0,
      "step": 25360
    },
    {
      "entropy": 6.515372610092163,
      "epoch": 2.7143239338648404,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004271762233859403,
      "loss": 5.8918,
      "mean_token_accuracy": 0.16488134115934372,
      "num_tokens": 54891532.0,
      "step": 25365
    },
    {
      "entropy": 6.458114862442017,
      "epoch": 2.714859007972604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042714805885193046,
      "loss": 5.8081,
      "mean_token_accuracy": 0.17634636163711548,
      "num_tokens": 54902945.0,
      "step": 25370
    },
    {
      "entropy": 6.4901947498321535,
      "epoch": 2.715394082080368,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042711988992466684,
      "loss": 5.8563,
      "mean_token_accuracy": 0.16191485822200774,
      "num_tokens": 54913943.0,
      "step": 25375
    },
    {
      "entropy": 6.492808485031128,
      "epoch": 2.715929156188132,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042709171660496275,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16900673657655715,
      "num_tokens": 54924241.0,
      "step": 25380
    },
    {
      "entropy": 6.474675512313842,
      "epoch": 2.716464230295896,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004270635388936317,
      "loss": 5.8317,
      "mean_token_accuracy": 0.17011313289403915,
      "num_tokens": 54934083.0,
      "step": 25385
    },
    {
      "entropy": 6.542514085769653,
      "epoch": 2.71699930440366,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042703535679148734,
      "loss": 5.9239,
      "mean_token_accuracy": 0.16141712069511413,
      "num_tokens": 54944928.0,
      "step": 25390
    },
    {
      "entropy": 6.443210458755493,
      "epoch": 2.7175343785114237,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042700717029934345,
      "loss": 5.7264,
      "mean_token_accuracy": 0.17753425389528274,
      "num_tokens": 54955715.0,
      "step": 25395
    },
    {
      "entropy": 6.508191204071045,
      "epoch": 2.7180694526191878,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004269789794180139,
      "loss": 5.8939,
      "mean_token_accuracy": 0.16787009090185165,
      "num_tokens": 54966149.0,
      "step": 25400
    },
    {
      "entropy": 6.503817939758301,
      "epoch": 2.718604526726952,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042695078414831263,
      "loss": 5.8679,
      "mean_token_accuracy": 0.16678803265094758,
      "num_tokens": 54976854.0,
      "step": 25405
    },
    {
      "entropy": 6.494216966629028,
      "epoch": 2.7191396008347155,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004269225844910538,
      "loss": 5.87,
      "mean_token_accuracy": 0.17005829364061356,
      "num_tokens": 54987377.0,
      "step": 25410
    },
    {
      "entropy": 6.510860872268677,
      "epoch": 2.7196746749424796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004268943804470518,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16398942172527314,
      "num_tokens": 54999488.0,
      "step": 25415
    },
    {
      "entropy": 6.6251304149627686,
      "epoch": 2.7202097490502437,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042686617201712074,
      "loss": 5.9367,
      "mean_token_accuracy": 0.15756019353866577,
      "num_tokens": 55011517.0,
      "step": 25420
    },
    {
      "entropy": 6.609432315826416,
      "epoch": 2.7207448231580074,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004268379592020754,
      "loss": 5.8748,
      "mean_token_accuracy": 0.1661329448223114,
      "num_tokens": 55022133.0,
      "step": 25425
    },
    {
      "entropy": 6.4293866634368895,
      "epoch": 2.7212798972657715,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042680974200273036,
      "loss": 5.838,
      "mean_token_accuracy": 0.17131779938936234,
      "num_tokens": 55032684.0,
      "step": 25430
    },
    {
      "entropy": 6.555597829818725,
      "epoch": 2.721814971373535,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004267815204199003,
      "loss": 5.9122,
      "mean_token_accuracy": 0.17100197672843934,
      "num_tokens": 55042943.0,
      "step": 25435
    },
    {
      "entropy": 6.500936889648438,
      "epoch": 2.722350045481299,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004267532944544002,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16956137269735336,
      "num_tokens": 55053805.0,
      "step": 25440
    },
    {
      "entropy": 6.44835352897644,
      "epoch": 2.722885119589063,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042672506410704495,
      "loss": 5.847,
      "mean_token_accuracy": 0.16764627248048783,
      "num_tokens": 55066611.0,
      "step": 25445
    },
    {
      "entropy": 6.5429503440856935,
      "epoch": 2.723420193696827,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004266968293786499,
      "loss": 5.8735,
      "mean_token_accuracy": 0.1638684540987015,
      "num_tokens": 55077954.0,
      "step": 25450
    },
    {
      "entropy": 6.54258131980896,
      "epoch": 2.723955267804591,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004266685902700302,
      "loss": 5.8884,
      "mean_token_accuracy": 0.17225332856178283,
      "num_tokens": 55090296.0,
      "step": 25455
    },
    {
      "entropy": 6.57615065574646,
      "epoch": 2.7244903419123547,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004266403467820013,
      "loss": 5.876,
      "mean_token_accuracy": 0.16612754464149476,
      "num_tokens": 55100664.0,
      "step": 25460
    },
    {
      "entropy": 6.458496856689453,
      "epoch": 2.725025416020119,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004266120989153786,
      "loss": 5.7575,
      "mean_token_accuracy": 0.1726003423333168,
      "num_tokens": 55111094.0,
      "step": 25465
    },
    {
      "entropy": 6.535148239135742,
      "epoch": 2.725560490127883,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042658384667097787,
      "loss": 5.889,
      "mean_token_accuracy": 0.1643130823969841,
      "num_tokens": 55121877.0,
      "step": 25470
    },
    {
      "entropy": 6.465945434570313,
      "epoch": 2.7260955642356466,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004265555900496149,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17038530111312866,
      "num_tokens": 55133087.0,
      "step": 25475
    },
    {
      "entropy": 6.523959016799926,
      "epoch": 2.7266306383434107,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042652732905210555,
      "loss": 5.8864,
      "mean_token_accuracy": 0.16265678256750107,
      "num_tokens": 55144257.0,
      "step": 25480
    },
    {
      "entropy": 6.482910060882569,
      "epoch": 2.7271657124511743,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004264990636792659,
      "loss": 5.7555,
      "mean_token_accuracy": 0.1778413698077202,
      "num_tokens": 55154389.0,
      "step": 25485
    },
    {
      "entropy": 6.430528259277343,
      "epoch": 2.7277007865589384,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000426470793931912,
      "loss": 5.7594,
      "mean_token_accuracy": 0.18311949968338012,
      "num_tokens": 55165734.0,
      "step": 25490
    },
    {
      "entropy": 6.361513137817383,
      "epoch": 2.728235860666702,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004264425198108603,
      "loss": 5.7467,
      "mean_token_accuracy": 0.17742665410041808,
      "num_tokens": 55176448.0,
      "step": 25495
    },
    {
      "entropy": 6.451157093048096,
      "epoch": 2.728770934774466,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042641424131692707,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1708410143852234,
      "num_tokens": 55187515.0,
      "step": 25500
    },
    {
      "entropy": 6.539643049240112,
      "epoch": 2.7293060088822303,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004263859584509289,
      "loss": 5.9057,
      "mean_token_accuracy": 0.166774682700634,
      "num_tokens": 55199236.0,
      "step": 25505
    },
    {
      "entropy": 6.50022177696228,
      "epoch": 2.729841082989994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004263576712136825,
      "loss": 5.857,
      "mean_token_accuracy": 0.17455334067344666,
      "num_tokens": 55210151.0,
      "step": 25510
    },
    {
      "entropy": 6.464462327957153,
      "epoch": 2.730376157097758,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042632937960600454,
      "loss": 5.827,
      "mean_token_accuracy": 0.16319289356470107,
      "num_tokens": 55221322.0,
      "step": 25515
    },
    {
      "entropy": 6.43232536315918,
      "epoch": 2.730911231205522,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042630108362871205,
      "loss": 5.7686,
      "mean_token_accuracy": 0.1778502121567726,
      "num_tokens": 55231901.0,
      "step": 25520
    },
    {
      "entropy": 6.528699016571045,
      "epoch": 2.731446305313286,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042627278328262203,
      "loss": 5.8052,
      "mean_token_accuracy": 0.176513671875,
      "num_tokens": 55243154.0,
      "step": 25525
    },
    {
      "entropy": 6.473900032043457,
      "epoch": 2.73198137942105,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042624447856855164,
      "loss": 5.8273,
      "mean_token_accuracy": 0.16868849396705626,
      "num_tokens": 55253485.0,
      "step": 25530
    },
    {
      "entropy": 6.537059211730957,
      "epoch": 2.732516453528814,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042621616948731816,
      "loss": 5.8862,
      "mean_token_accuracy": 0.16402463465929032,
      "num_tokens": 55264936.0,
      "step": 25535
    },
    {
      "entropy": 6.498049545288086,
      "epoch": 2.7330515276365777,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004261878560397391,
      "loss": 5.8488,
      "mean_token_accuracy": 0.17241866439580916,
      "num_tokens": 55276138.0,
      "step": 25540
    },
    {
      "entropy": 6.54367299079895,
      "epoch": 2.7335866017443413,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042615953822663195,
      "loss": 5.8886,
      "mean_token_accuracy": 0.16705030351877212,
      "num_tokens": 55287398.0,
      "step": 25545
    },
    {
      "entropy": 6.614708471298218,
      "epoch": 2.7341216758521054,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042613121604881426,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15802060663700104,
      "num_tokens": 55298806.0,
      "step": 25550
    },
    {
      "entropy": 6.469089937210083,
      "epoch": 2.7346567499598695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004261028895071041,
      "loss": 5.8132,
      "mean_token_accuracy": 0.17133119106292724,
      "num_tokens": 55309976.0,
      "step": 25555
    },
    {
      "entropy": 6.4923093795776365,
      "epoch": 2.735191824067633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042607455860231916,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16577669829130173,
      "num_tokens": 55320806.0,
      "step": 25560
    },
    {
      "entropy": 6.497726011276245,
      "epoch": 2.7357268981753973,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004260462233352775,
      "loss": 5.8814,
      "mean_token_accuracy": 0.16817573010921477,
      "num_tokens": 55332153.0,
      "step": 25565
    },
    {
      "entropy": 6.502553796768188,
      "epoch": 2.7362619722831614,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042601788370679746,
      "loss": 5.866,
      "mean_token_accuracy": 0.16395435631275176,
      "num_tokens": 55343595.0,
      "step": 25570
    },
    {
      "entropy": 6.473385381698608,
      "epoch": 2.736797046390925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042598953971769723,
      "loss": 5.8796,
      "mean_token_accuracy": 0.16388111114501952,
      "num_tokens": 55353891.0,
      "step": 25575
    },
    {
      "entropy": 6.486343955993652,
      "epoch": 2.737332120498689,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004259611913687952,
      "loss": 5.8061,
      "mean_token_accuracy": 0.16900498867034913,
      "num_tokens": 55364743.0,
      "step": 25580
    },
    {
      "entropy": 6.570917129516602,
      "epoch": 2.7378671946064532,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042593283866091,
      "loss": 5.917,
      "mean_token_accuracy": 0.16105240285396577,
      "num_tokens": 55375122.0,
      "step": 25585
    },
    {
      "entropy": 6.490377569198609,
      "epoch": 2.738402268714217,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042590448159486035,
      "loss": 5.8571,
      "mean_token_accuracy": 0.16294703483581544,
      "num_tokens": 55387281.0,
      "step": 25590
    },
    {
      "entropy": 6.544297981262207,
      "epoch": 2.738937342821981,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042587612017146496,
      "loss": 5.876,
      "mean_token_accuracy": 0.15671471655368804,
      "num_tokens": 55397859.0,
      "step": 25595
    },
    {
      "entropy": 6.512187862396241,
      "epoch": 2.7394724169297446,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042584775439154276,
      "loss": 5.8114,
      "mean_token_accuracy": 0.1708010643720627,
      "num_tokens": 55408169.0,
      "step": 25600
    },
    {
      "entropy": 6.485558938980103,
      "epoch": 2.7400074910375087,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004258193842559129,
      "loss": 5.7824,
      "mean_token_accuracy": 0.16613163352012633,
      "num_tokens": 55419342.0,
      "step": 25605
    },
    {
      "entropy": 6.41924090385437,
      "epoch": 2.7405425651452724,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004257910097653945,
      "loss": 5.7659,
      "mean_token_accuracy": 0.17381515800952912,
      "num_tokens": 55430271.0,
      "step": 25610
    },
    {
      "entropy": 6.525516366958618,
      "epoch": 2.7410776392530365,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004257626309208069,
      "loss": 5.8137,
      "mean_token_accuracy": 0.16813075989484788,
      "num_tokens": 55441901.0,
      "step": 25615
    },
    {
      "entropy": 6.498415803909301,
      "epoch": 2.7416127133608006,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004257342477229695,
      "loss": 5.848,
      "mean_token_accuracy": 0.16712346374988557,
      "num_tokens": 55451743.0,
      "step": 25620
    },
    {
      "entropy": 6.47631196975708,
      "epoch": 2.7421477874685642,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042570586017270196,
      "loss": 5.8216,
      "mean_token_accuracy": 0.1729650765657425,
      "num_tokens": 55462805.0,
      "step": 25625
    },
    {
      "entropy": 6.446879720687866,
      "epoch": 2.7426828615763283,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042567746827082377,
      "loss": 5.8997,
      "mean_token_accuracy": 0.16776878386735916,
      "num_tokens": 55474277.0,
      "step": 25630
    },
    {
      "entropy": 6.522888422012329,
      "epoch": 2.7432179356840924,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004256490720181549,
      "loss": 5.8179,
      "mean_token_accuracy": 0.16526043862104417,
      "num_tokens": 55485257.0,
      "step": 25635
    },
    {
      "entropy": 6.432376146316528,
      "epoch": 2.743753009791856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042562067141551523,
      "loss": 5.768,
      "mean_token_accuracy": 0.17390390634536743,
      "num_tokens": 55495835.0,
      "step": 25640
    },
    {
      "entropy": 6.5857549667358395,
      "epoch": 2.74428808389962,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042559226646372486,
      "loss": 5.9613,
      "mean_token_accuracy": 0.1620474487543106,
      "num_tokens": 55506256.0,
      "step": 25645
    },
    {
      "entropy": 6.551511430740357,
      "epoch": 2.744823158007384,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000425563857163604,
      "loss": 5.8079,
      "mean_token_accuracy": 0.18005023449659346,
      "num_tokens": 55516821.0,
      "step": 25650
    },
    {
      "entropy": 6.479223585128784,
      "epoch": 2.745358232115148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042553544351597286,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16163473427295685,
      "num_tokens": 55527153.0,
      "step": 25655
    },
    {
      "entropy": 6.474881362915039,
      "epoch": 2.7458933062229116,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042550702552165195,
      "loss": 5.911,
      "mean_token_accuracy": 0.16439063996076583,
      "num_tokens": 55538219.0,
      "step": 25660
    },
    {
      "entropy": 6.55819935798645,
      "epoch": 2.7464283803306757,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042547860318146183,
      "loss": 5.8613,
      "mean_token_accuracy": 0.16559073328971863,
      "num_tokens": 55548733.0,
      "step": 25665
    },
    {
      "entropy": 6.5381254196167,
      "epoch": 2.74696345443844,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042545017649622314,
      "loss": 5.8382,
      "mean_token_accuracy": 0.1688027262687683,
      "num_tokens": 55559746.0,
      "step": 25670
    },
    {
      "entropy": 6.49302830696106,
      "epoch": 2.7474985285462035,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004254217454667569,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1708567038178444,
      "num_tokens": 55570322.0,
      "step": 25675
    },
    {
      "entropy": 6.36444993019104,
      "epoch": 2.7480336026539676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004253933100938837,
      "loss": 5.7518,
      "mean_token_accuracy": 0.17264339774847032,
      "num_tokens": 55580715.0,
      "step": 25680
    },
    {
      "entropy": 6.505426836013794,
      "epoch": 2.7485686767617317,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004253648703784249,
      "loss": 5.7868,
      "mean_token_accuracy": 0.17573439776897432,
      "num_tokens": 55591685.0,
      "step": 25685
    },
    {
      "entropy": 6.535365200042724,
      "epoch": 2.7491037508694953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004253364263212016,
      "loss": 5.819,
      "mean_token_accuracy": 0.16745562255382537,
      "num_tokens": 55602288.0,
      "step": 25690
    },
    {
      "entropy": 6.524036693572998,
      "epoch": 2.7496388249772594,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004253079779230352,
      "loss": 5.9194,
      "mean_token_accuracy": 0.16610848605632783,
      "num_tokens": 55613011.0,
      "step": 25695
    },
    {
      "entropy": 6.466824960708618,
      "epoch": 2.7501738990850235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004252795251847469,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1706088200211525,
      "num_tokens": 55623630.0,
      "step": 25700
    },
    {
      "entropy": 6.443159437179565,
      "epoch": 2.750708973192787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042525106810715845,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17211644500494003,
      "num_tokens": 55634764.0,
      "step": 25705
    },
    {
      "entropy": 6.562702274322509,
      "epoch": 2.7512440473005513,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042522260669109157,
      "loss": 5.8634,
      "mean_token_accuracy": 0.1751796633005142,
      "num_tokens": 55645085.0,
      "step": 25710
    },
    {
      "entropy": 6.5640825748443605,
      "epoch": 2.751779121408315,
      "grad_norm": 1.09375,
      "learning_rate": 0.000425194140937368,
      "loss": 5.9336,
      "mean_token_accuracy": 0.15877878814935684,
      "num_tokens": 55656249.0,
      "step": 25715
    },
    {
      "entropy": 6.560854387283325,
      "epoch": 2.752314195516079,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042516567084680977,
      "loss": 5.883,
      "mean_token_accuracy": 0.17108915001153946,
      "num_tokens": 55668138.0,
      "step": 25720
    },
    {
      "entropy": 6.544164705276489,
      "epoch": 2.7528492696238427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004251371964202389,
      "loss": 5.8391,
      "mean_token_accuracy": 0.1690256342291832,
      "num_tokens": 55678747.0,
      "step": 25725
    },
    {
      "entropy": 6.5391875267028805,
      "epoch": 2.753384343731607,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004251087176584775,
      "loss": 5.9165,
      "mean_token_accuracy": 0.1635986402630806,
      "num_tokens": 55690266.0,
      "step": 25730
    },
    {
      "entropy": 6.498034715652466,
      "epoch": 2.753919417839371,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042508023456234805,
      "loss": 5.8495,
      "mean_token_accuracy": 0.1696402221918106,
      "num_tokens": 55700229.0,
      "step": 25735
    },
    {
      "entropy": 6.50631856918335,
      "epoch": 2.7544544919471345,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042505174713267286,
      "loss": 5.8867,
      "mean_token_accuracy": 0.16258366107940675,
      "num_tokens": 55711564.0,
      "step": 25740
    },
    {
      "entropy": 6.53729395866394,
      "epoch": 2.7549895660548986,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004250232553702746,
      "loss": 5.8083,
      "mean_token_accuracy": 0.17675295025110244,
      "num_tokens": 55722618.0,
      "step": 25745
    },
    {
      "entropy": 6.453394079208374,
      "epoch": 2.7555246401626627,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004249947592759759,
      "loss": 5.7917,
      "mean_token_accuracy": 0.1746248334646225,
      "num_tokens": 55732828.0,
      "step": 25750
    },
    {
      "entropy": 6.4476244926452635,
      "epoch": 2.7560597142704264,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004249662588505996,
      "loss": 5.8003,
      "mean_token_accuracy": 0.16857294291257857,
      "num_tokens": 55742283.0,
      "step": 25755
    },
    {
      "entropy": 6.492388725280762,
      "epoch": 2.7565947883781905,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042493775409496867,
      "loss": 5.838,
      "mean_token_accuracy": 0.1708940088748932,
      "num_tokens": 55752883.0,
      "step": 25760
    },
    {
      "entropy": 6.517801284790039,
      "epoch": 2.757129862485954,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004249092450099062,
      "loss": 5.8474,
      "mean_token_accuracy": 0.17100439220666885,
      "num_tokens": 55762730.0,
      "step": 25765
    },
    {
      "entropy": 6.405041456222534,
      "epoch": 2.7576649365937183,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004248807315962353,
      "loss": 5.8,
      "mean_token_accuracy": 0.17058038264513015,
      "num_tokens": 55774068.0,
      "step": 25770
    },
    {
      "entropy": 6.482345199584961,
      "epoch": 2.758200010701482,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004248522138547793,
      "loss": 5.8764,
      "mean_token_accuracy": 0.16008334010839462,
      "num_tokens": 55784896.0,
      "step": 25775
    },
    {
      "entropy": 6.492315101623535,
      "epoch": 2.758735084809246,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004248236917863617,
      "loss": 5.8275,
      "mean_token_accuracy": 0.1730072721838951,
      "num_tokens": 55796186.0,
      "step": 25780
    },
    {
      "entropy": 6.449084234237671,
      "epoch": 2.75927015891701,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042479516539180605,
      "loss": 5.7797,
      "mean_token_accuracy": 0.179116889834404,
      "num_tokens": 55806439.0,
      "step": 25785
    },
    {
      "entropy": 6.488138628005982,
      "epoch": 2.7598052330247738,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000424766634671936,
      "loss": 5.8605,
      "mean_token_accuracy": 0.1582298070192337,
      "num_tokens": 55817955.0,
      "step": 25790
    },
    {
      "entropy": 6.492053842544555,
      "epoch": 2.760340307132538,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042473809962757553,
      "loss": 5.8379,
      "mean_token_accuracy": 0.17173564285039902,
      "num_tokens": 55828876.0,
      "step": 25795
    },
    {
      "entropy": 6.483574676513672,
      "epoch": 2.760875381240302,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004247095602595485,
      "loss": 5.7712,
      "mean_token_accuracy": 0.18065893948078154,
      "num_tokens": 55839721.0,
      "step": 25800
    },
    {
      "entropy": 6.433148002624511,
      "epoch": 2.7614104553480656,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004246810165686788,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16965566128492354,
      "num_tokens": 55850427.0,
      "step": 25805
    },
    {
      "entropy": 6.5212219715118405,
      "epoch": 2.7619455294558297,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042465246855579093,
      "loss": 5.8943,
      "mean_token_accuracy": 0.16293561458587646,
      "num_tokens": 55860245.0,
      "step": 25810
    },
    {
      "entropy": 6.5336585521698,
      "epoch": 2.762480603563594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042462391622170897,
      "loss": 5.8898,
      "mean_token_accuracy": 0.16794337779283525,
      "num_tokens": 55869957.0,
      "step": 25815
    },
    {
      "entropy": 6.530043888092041,
      "epoch": 2.7630156776713575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004245953595672575,
      "loss": 5.911,
      "mean_token_accuracy": 0.16023056954145432,
      "num_tokens": 55881159.0,
      "step": 25820
    },
    {
      "entropy": 6.603977966308594,
      "epoch": 2.7635507517791216,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000424566798593261,
      "loss": 5.8558,
      "mean_token_accuracy": 0.1641255348920822,
      "num_tokens": 55891830.0,
      "step": 25825
    },
    {
      "entropy": 6.553379058837891,
      "epoch": 2.7640858258868852,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042453823330054427,
      "loss": 5.8608,
      "mean_token_accuracy": 0.16496307998895646,
      "num_tokens": 55901904.0,
      "step": 25830
    },
    {
      "entropy": 6.493668937683106,
      "epoch": 2.7646208999946493,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000424509663689932,
      "loss": 5.789,
      "mean_token_accuracy": 0.16870740354061126,
      "num_tokens": 55913808.0,
      "step": 25835
    },
    {
      "entropy": 6.4606622695922855,
      "epoch": 2.765155974102413,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004244810897622492,
      "loss": 5.821,
      "mean_token_accuracy": 0.1717306762933731,
      "num_tokens": 55924997.0,
      "step": 25840
    },
    {
      "entropy": 6.538218975067139,
      "epoch": 2.765691048210177,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000424452511518321,
      "loss": 5.9452,
      "mean_token_accuracy": 0.15800642371177673,
      "num_tokens": 55935623.0,
      "step": 25845
    },
    {
      "entropy": 6.543405342102051,
      "epoch": 2.766226122317941,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004244239289589724,
      "loss": 5.8933,
      "mean_token_accuracy": 0.16538061797618867,
      "num_tokens": 55946733.0,
      "step": 25850
    },
    {
      "entropy": 6.502685928344727,
      "epoch": 2.766761196425705,
      "grad_norm": 1.1875,
      "learning_rate": 0.000424395342085029,
      "loss": 5.9235,
      "mean_token_accuracy": 0.1601630836725235,
      "num_tokens": 55956158.0,
      "step": 25855
    },
    {
      "entropy": 6.506714534759522,
      "epoch": 2.767296270533469,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042436675089731603,
      "loss": 5.8189,
      "mean_token_accuracy": 0.17353506833314897,
      "num_tokens": 55966442.0,
      "step": 25860
    },
    {
      "entropy": 6.539748954772949,
      "epoch": 2.767831344641233,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004243381553966591,
      "loss": 5.8869,
      "mean_token_accuracy": 0.1672050565481186,
      "num_tokens": 55977314.0,
      "step": 25865
    },
    {
      "entropy": 6.507328367233276,
      "epoch": 2.7683664187489967,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042430955558388396,
      "loss": 5.8023,
      "mean_token_accuracy": 0.17265867441892624,
      "num_tokens": 55987457.0,
      "step": 25870
    },
    {
      "entropy": 6.496155834197998,
      "epoch": 2.768901492856761,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042428095145981634,
      "loss": 5.8068,
      "mean_token_accuracy": 0.17683081030845643,
      "num_tokens": 55998319.0,
      "step": 25875
    },
    {
      "entropy": 6.4558204174041744,
      "epoch": 2.7694365669645244,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042425234302528224,
      "loss": 5.7303,
      "mean_token_accuracy": 0.17595356553792954,
      "num_tokens": 56008654.0,
      "step": 25880
    },
    {
      "entropy": 6.505675554275513,
      "epoch": 2.7699716410722885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004242237302811077,
      "loss": 5.918,
      "mean_token_accuracy": 0.1625065714120865,
      "num_tokens": 56019793.0,
      "step": 25885
    },
    {
      "entropy": 6.531436395645142,
      "epoch": 2.770506715180052,
      "grad_norm": 1.09375,
      "learning_rate": 0.000424195113228119,
      "loss": 5.8465,
      "mean_token_accuracy": 0.17086572647094728,
      "num_tokens": 56031051.0,
      "step": 25890
    },
    {
      "entropy": 6.552368211746216,
      "epoch": 2.7710417892878163,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004241664918671423,
      "loss": 5.9317,
      "mean_token_accuracy": 0.16699955016374587,
      "num_tokens": 56041210.0,
      "step": 25895
    },
    {
      "entropy": 6.487704420089722,
      "epoch": 2.7715768633955804,
      "grad_norm": 1.171875,
      "learning_rate": 0.000424137866199004,
      "loss": 5.8085,
      "mean_token_accuracy": 0.16960131078958512,
      "num_tokens": 56051601.0,
      "step": 25900
    },
    {
      "entropy": 6.490559816360474,
      "epoch": 2.772111937503344,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004241092362245309,
      "loss": 5.7529,
      "mean_token_accuracy": 0.1758995071053505,
      "num_tokens": 56062669.0,
      "step": 25905
    },
    {
      "entropy": 6.458813810348511,
      "epoch": 2.772647011611108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042408060194454956,
      "loss": 5.8918,
      "mean_token_accuracy": 0.16910582333803176,
      "num_tokens": 56074414.0,
      "step": 25910
    },
    {
      "entropy": 6.491256380081177,
      "epoch": 2.7731820857188723,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042405196335988676,
      "loss": 5.8536,
      "mean_token_accuracy": 0.16593457907438278,
      "num_tokens": 56085428.0,
      "step": 25915
    },
    {
      "entropy": 6.522317409515381,
      "epoch": 2.773717159826636,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042402332047136943,
      "loss": 5.8681,
      "mean_token_accuracy": 0.15988942980766296,
      "num_tokens": 56097403.0,
      "step": 25920
    },
    {
      "entropy": 6.478246641159058,
      "epoch": 2.7742522339344,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004239946732798247,
      "loss": 5.7529,
      "mean_token_accuracy": 0.17795479595661162,
      "num_tokens": 56107781.0,
      "step": 25925
    },
    {
      "entropy": 6.422014427185059,
      "epoch": 2.774787308042164,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042396602178607973,
      "loss": 5.7265,
      "mean_token_accuracy": 0.1826716497540474,
      "num_tokens": 56118698.0,
      "step": 25930
    },
    {
      "entropy": 6.482721471786499,
      "epoch": 2.7753223821499278,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004239373659909618,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16545447558164597,
      "num_tokens": 56128397.0,
      "step": 25935
    },
    {
      "entropy": 6.459578323364258,
      "epoch": 2.7758574562576914,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004239087058952984,
      "loss": 5.8239,
      "mean_token_accuracy": 0.1746899351477623,
      "num_tokens": 56138798.0,
      "step": 25940
    },
    {
      "entropy": 6.522713041305542,
      "epoch": 2.7763925303654555,
      "grad_norm": 1.0625,
      "learning_rate": 0.000423880041499917,
      "loss": 5.7834,
      "mean_token_accuracy": 0.17857991605997087,
      "num_tokens": 56149936.0,
      "step": 25945
    },
    {
      "entropy": 6.526859712600708,
      "epoch": 2.7769276044732196,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004238513728056453,
      "loss": 5.8575,
      "mean_token_accuracy": 0.17207858562469483,
      "num_tokens": 56161433.0,
      "step": 25950
    },
    {
      "entropy": 6.5603619575500485,
      "epoch": 2.7774626785809833,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042382269981331116,
      "loss": 5.8905,
      "mean_token_accuracy": 0.15811391025781632,
      "num_tokens": 56172312.0,
      "step": 25955
    },
    {
      "entropy": 6.488846254348755,
      "epoch": 2.7779977526887474,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042379402252374244,
      "loss": 5.7802,
      "mean_token_accuracy": 0.17501093298196793,
      "num_tokens": 56183887.0,
      "step": 25960
    },
    {
      "entropy": 6.467304992675781,
      "epoch": 2.7785328267965115,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042376534093776727,
      "loss": 5.8157,
      "mean_token_accuracy": 0.17699937671422958,
      "num_tokens": 56194172.0,
      "step": 25965
    },
    {
      "entropy": 6.489737701416016,
      "epoch": 2.779067900904275,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004237366550562137,
      "loss": 5.9443,
      "mean_token_accuracy": 0.15797322392463684,
      "num_tokens": 56204569.0,
      "step": 25970
    },
    {
      "entropy": 6.475646066665649,
      "epoch": 2.7796029750120392,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042370796487991013,
      "loss": 5.8306,
      "mean_token_accuracy": 0.17092911452054976,
      "num_tokens": 56215875.0,
      "step": 25975
    },
    {
      "entropy": 6.529653549194336,
      "epoch": 2.7801380491198033,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042367927040968514,
      "loss": 5.8675,
      "mean_token_accuracy": 0.16182407289743422,
      "num_tokens": 56227633.0,
      "step": 25980
    },
    {
      "entropy": 6.384666538238525,
      "epoch": 2.780673123227567,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004236505716463669,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17705947458744048,
      "num_tokens": 56239420.0,
      "step": 25985
    },
    {
      "entropy": 6.486491870880127,
      "epoch": 2.781208197335331,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004236218685907844,
      "loss": 5.8174,
      "mean_token_accuracy": 0.16681949198246002,
      "num_tokens": 56250538.0,
      "step": 25990
    },
    {
      "entropy": 6.473531579971313,
      "epoch": 2.7817432714430947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004235931612437663,
      "loss": 5.8718,
      "mean_token_accuracy": 0.16893790811300277,
      "num_tokens": 56261423.0,
      "step": 25995
    },
    {
      "entropy": 6.474535989761352,
      "epoch": 2.782278345550859,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004235644496061416,
      "loss": 5.8727,
      "mean_token_accuracy": 0.16766170561313629,
      "num_tokens": 56272581.0,
      "step": 26000
    },
    {
      "entropy": 6.49563193321228,
      "epoch": 2.7828134196586225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004235357336787392,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1737714409828186,
      "num_tokens": 56283703.0,
      "step": 26005
    },
    {
      "entropy": 6.554462051391601,
      "epoch": 2.7833484937663866,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042350701346238853,
      "loss": 5.9039,
      "mean_token_accuracy": 0.16458913832902908,
      "num_tokens": 56294657.0,
      "step": 26010
    },
    {
      "entropy": 6.505699253082275,
      "epoch": 2.7838835678741507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004234782889579185,
      "loss": 5.7718,
      "mean_token_accuracy": 0.174188594520092,
      "num_tokens": 56305555.0,
      "step": 26015
    },
    {
      "entropy": 6.532172632217407,
      "epoch": 2.7844186419819144,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042344956016615885,
      "loss": 5.898,
      "mean_token_accuracy": 0.16682716012001036,
      "num_tokens": 56316716.0,
      "step": 26020
    },
    {
      "entropy": 6.507379484176636,
      "epoch": 2.7849537160896785,
      "grad_norm": 1.25,
      "learning_rate": 0.0004234208270879389,
      "loss": 5.7756,
      "mean_token_accuracy": 0.1682429701089859,
      "num_tokens": 56326582.0,
      "step": 26025
    },
    {
      "entropy": 6.498872995376587,
      "epoch": 2.7854887901974426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042339208972408865,
      "loss": 5.8623,
      "mean_token_accuracy": 0.17156561315059662,
      "num_tokens": 56336687.0,
      "step": 26030
    },
    {
      "entropy": 6.4134996891021725,
      "epoch": 2.786023864305206,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004233633480754375,
      "loss": 5.6789,
      "mean_token_accuracy": 0.18656487464904786,
      "num_tokens": 56347493.0,
      "step": 26035
    },
    {
      "entropy": 6.494147872924804,
      "epoch": 2.7865589384129703,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042333460214281557,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17364689707756042,
      "num_tokens": 56359259.0,
      "step": 26040
    },
    {
      "entropy": 6.503252410888672,
      "epoch": 2.787094012520734,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042330585192705284,
      "loss": 5.8158,
      "mean_token_accuracy": 0.16431496143341065,
      "num_tokens": 56370502.0,
      "step": 26045
    },
    {
      "entropy": 6.442932748794556,
      "epoch": 2.787629086628498,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042327709742897944,
      "loss": 5.8566,
      "mean_token_accuracy": 0.16760501265525818,
      "num_tokens": 56381047.0,
      "step": 26050
    },
    {
      "entropy": 6.435987854003907,
      "epoch": 2.7881641607362617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042324833864942576,
      "loss": 5.7681,
      "mean_token_accuracy": 0.1756642520427704,
      "num_tokens": 56392304.0,
      "step": 26055
    },
    {
      "entropy": 6.4989289283752445,
      "epoch": 2.788699234844026,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004232195755892221,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1758020967245102,
      "num_tokens": 56402480.0,
      "step": 26060
    },
    {
      "entropy": 6.5030604839324955,
      "epoch": 2.78923430895179,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042319080824919895,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16887935549020766,
      "num_tokens": 56413659.0,
      "step": 26065
    },
    {
      "entropy": 6.4925915718078615,
      "epoch": 2.7897693830595536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004231620366301871,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1642981216311455,
      "num_tokens": 56425272.0,
      "step": 26070
    },
    {
      "entropy": 6.566420650482177,
      "epoch": 2.7903044571673177,
      "grad_norm": 1.125,
      "learning_rate": 0.00042313326073301733,
      "loss": 5.8622,
      "mean_token_accuracy": 0.16651221215724946,
      "num_tokens": 56436696.0,
      "step": 26075
    },
    {
      "entropy": 6.548642778396607,
      "epoch": 2.7908395312750818,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004231044805585204,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16750627309083937,
      "num_tokens": 56446666.0,
      "step": 26080
    },
    {
      "entropy": 6.473950815200806,
      "epoch": 2.7913746053828454,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004230756961075274,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16975184679031372,
      "num_tokens": 56457082.0,
      "step": 26085
    },
    {
      "entropy": 6.428369522094727,
      "epoch": 2.7919096794906095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00042304690738086944,
      "loss": 5.7906,
      "mean_token_accuracy": 0.17374304831027984,
      "num_tokens": 56468492.0,
      "step": 26090
    },
    {
      "entropy": 6.568636131286621,
      "epoch": 2.7924447535983736,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004230181143793779,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1659633368253708,
      "num_tokens": 56479353.0,
      "step": 26095
    },
    {
      "entropy": 6.549053192138672,
      "epoch": 2.7929798277061373,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042298931710388414,
      "loss": 5.8502,
      "mean_token_accuracy": 0.17011522501707077,
      "num_tokens": 56490105.0,
      "step": 26100
    },
    {
      "entropy": 6.5404034614562985,
      "epoch": 2.7935149018139014,
      "grad_norm": 1.125,
      "learning_rate": 0.0004229605155552195,
      "loss": 5.8896,
      "mean_token_accuracy": 0.16312743127346038,
      "num_tokens": 56501528.0,
      "step": 26105
    },
    {
      "entropy": 6.550677824020386,
      "epoch": 2.794049975921665,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042293170973421593,
      "loss": 5.9523,
      "mean_token_accuracy": 0.1602058708667755,
      "num_tokens": 56512952.0,
      "step": 26110
    },
    {
      "entropy": 6.532808780670166,
      "epoch": 2.794585050029429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042290289964170495,
      "loss": 5.8441,
      "mean_token_accuracy": 0.17126893848180771,
      "num_tokens": 56523332.0,
      "step": 26115
    },
    {
      "entropy": 6.540939712524414,
      "epoch": 2.795120124137193,
      "grad_norm": 1.125,
      "learning_rate": 0.0004228740852785185,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16761828511953353,
      "num_tokens": 56534777.0,
      "step": 26120
    },
    {
      "entropy": 6.507453155517578,
      "epoch": 2.795655198244957,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042284526664548866,
      "loss": 5.8358,
      "mean_token_accuracy": 0.1637384369969368,
      "num_tokens": 56545706.0,
      "step": 26125
    },
    {
      "entropy": 6.507033109664917,
      "epoch": 2.796190272352721,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042281644374344754,
      "loss": 5.8334,
      "mean_token_accuracy": 0.164088873565197,
      "num_tokens": 56557014.0,
      "step": 26130
    },
    {
      "entropy": 6.485359859466553,
      "epoch": 2.7967253464604847,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042278761657322736,
      "loss": 5.8679,
      "mean_token_accuracy": 0.17043450325727463,
      "num_tokens": 56569707.0,
      "step": 26135
    },
    {
      "entropy": 6.529911136627197,
      "epoch": 2.7972604205682488,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227587851356605,
      "loss": 5.7992,
      "mean_token_accuracy": 0.17023918628692628,
      "num_tokens": 56579924.0,
      "step": 26140
    },
    {
      "entropy": 6.56852011680603,
      "epoch": 2.797795494676013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227299494315795,
      "loss": 5.8763,
      "mean_token_accuracy": 0.16993345469236373,
      "num_tokens": 56590301.0,
      "step": 26145
    },
    {
      "entropy": 6.550875234603882,
      "epoch": 2.7983305687837765,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042270110946181693,
      "loss": 5.8229,
      "mean_token_accuracy": 0.1685004338622093,
      "num_tokens": 56600717.0,
      "step": 26150
    },
    {
      "entropy": 6.562976360321045,
      "epoch": 2.7988656428915406,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042267226522720556,
      "loss": 5.8598,
      "mean_token_accuracy": 0.16793892085552214,
      "num_tokens": 56611547.0,
      "step": 26155
    },
    {
      "entropy": 6.507299184799194,
      "epoch": 2.7994007169993043,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004226434167285784,
      "loss": 5.77,
      "mean_token_accuracy": 0.17138452380895614,
      "num_tokens": 56622542.0,
      "step": 26160
    },
    {
      "entropy": 6.487933731079101,
      "epoch": 2.7999357911070684,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004226145639667683,
      "loss": 5.8188,
      "mean_token_accuracy": 0.17738782316446305,
      "num_tokens": 56633359.0,
      "step": 26165
    },
    {
      "entropy": 6.407403182983399,
      "epoch": 2.800470865214832,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004225857069426083,
      "loss": 5.7145,
      "mean_token_accuracy": 0.1849641516804695,
      "num_tokens": 56644101.0,
      "step": 26170
    },
    {
      "entropy": 6.482030439376831,
      "epoch": 2.801005939322596,
      "grad_norm": 1.03125,
      "learning_rate": 0.00042255684565693183,
      "loss": 5.7873,
      "mean_token_accuracy": 0.16536916494369508,
      "num_tokens": 56654630.0,
      "step": 26175
    },
    {
      "entropy": 6.480303859710693,
      "epoch": 2.80154101343036,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004225279801105722,
      "loss": 5.7982,
      "mean_token_accuracy": 0.16779685616493226,
      "num_tokens": 56665826.0,
      "step": 26180
    },
    {
      "entropy": 6.494302558898926,
      "epoch": 2.802076087538124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004224991103043628,
      "loss": 5.8374,
      "mean_token_accuracy": 0.17120614051818847,
      "num_tokens": 56676431.0,
      "step": 26185
    },
    {
      "entropy": 6.538093328475952,
      "epoch": 2.802611161645888,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042247023623913745,
      "loss": 5.8293,
      "mean_token_accuracy": 0.17257019877433777,
      "num_tokens": 56687297.0,
      "step": 26190
    },
    {
      "entropy": 6.471608018875122,
      "epoch": 2.803146235753652,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004224413579157296,
      "loss": 5.869,
      "mean_token_accuracy": 0.1722571760416031,
      "num_tokens": 56698392.0,
      "step": 26195
    },
    {
      "entropy": 6.521790790557861,
      "epoch": 2.8036813098614157,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004224124753349733,
      "loss": 5.8785,
      "mean_token_accuracy": 0.168235182762146,
      "num_tokens": 56710588.0,
      "step": 26200
    },
    {
      "entropy": 6.420947885513305,
      "epoch": 2.80421638396918,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004223835884977025,
      "loss": 5.7201,
      "mean_token_accuracy": 0.17860404402017593,
      "num_tokens": 56723143.0,
      "step": 26205
    },
    {
      "entropy": 6.525055980682373,
      "epoch": 2.804751458076944,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004223546974047513,
      "loss": 5.8068,
      "mean_token_accuracy": 0.17227760553359986,
      "num_tokens": 56734097.0,
      "step": 26210
    },
    {
      "entropy": 6.550083112716675,
      "epoch": 2.8052865321847076,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004223258020569539,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16578950732946396,
      "num_tokens": 56745311.0,
      "step": 26215
    },
    {
      "entropy": 6.56948561668396,
      "epoch": 2.8058216062924712,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004222969024551447,
      "loss": 5.9285,
      "mean_token_accuracy": 0.17033571302890776,
      "num_tokens": 56755495.0,
      "step": 26220
    },
    {
      "entropy": 6.5143238544464115,
      "epoch": 2.8063566804002353,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042226799860015817,
      "loss": 5.8117,
      "mean_token_accuracy": 0.17301410138607026,
      "num_tokens": 56766107.0,
      "step": 26225
    },
    {
      "entropy": 6.455198526382446,
      "epoch": 2.8068917545079994,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004222390904928288,
      "loss": 5.7933,
      "mean_token_accuracy": 0.17905282825231553,
      "num_tokens": 56776877.0,
      "step": 26230
    },
    {
      "entropy": 6.501850175857544,
      "epoch": 2.807426828615763,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042221017813399146,
      "loss": 5.8863,
      "mean_token_accuracy": 0.17134655863046647,
      "num_tokens": 56786439.0,
      "step": 26235
    },
    {
      "entropy": 6.491676616668701,
      "epoch": 2.807961902723527,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004221812615244809,
      "loss": 5.8745,
      "mean_token_accuracy": 0.16119641959667205,
      "num_tokens": 56797433.0,
      "step": 26240
    },
    {
      "entropy": 6.595850276947021,
      "epoch": 2.8084969768312913,
      "grad_norm": 1.125,
      "learning_rate": 0.00042215234066513203,
      "loss": 5.9058,
      "mean_token_accuracy": 0.1622065305709839,
      "num_tokens": 56808270.0,
      "step": 26245
    },
    {
      "entropy": 6.603894567489624,
      "epoch": 2.809032050939055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042212341555678,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16557940244674682,
      "num_tokens": 56818590.0,
      "step": 26250
    },
    {
      "entropy": 6.5223503589630125,
      "epoch": 2.809567125046819,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004220944862002601,
      "loss": 5.8824,
      "mean_token_accuracy": 0.1626641720533371,
      "num_tokens": 56828799.0,
      "step": 26255
    },
    {
      "entropy": 6.470403814315796,
      "epoch": 2.810102199154583,
      "grad_norm": 1.125,
      "learning_rate": 0.00042206555259640744,
      "loss": 5.8758,
      "mean_token_accuracy": 0.16367677301168443,
      "num_tokens": 56839409.0,
      "step": 26260
    },
    {
      "entropy": 6.439814853668213,
      "epoch": 2.810637273262347,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042203661474605775,
      "loss": 5.784,
      "mean_token_accuracy": 0.17282844185829163,
      "num_tokens": 56849650.0,
      "step": 26265
    },
    {
      "entropy": 6.510466909408569,
      "epoch": 2.811172347370111,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004220076726500464,
      "loss": 5.8575,
      "mean_token_accuracy": 0.1620931938290596,
      "num_tokens": 56860437.0,
      "step": 26270
    },
    {
      "entropy": 6.499694013595581,
      "epoch": 2.8117074214778746,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042197872630920927,
      "loss": 5.8574,
      "mean_token_accuracy": 0.172714164853096,
      "num_tokens": 56870314.0,
      "step": 26275
    },
    {
      "entropy": 6.504363250732422,
      "epoch": 2.8122424955856387,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042194977572438194,
      "loss": 5.863,
      "mean_token_accuracy": 0.17279110848903656,
      "num_tokens": 56881675.0,
      "step": 26280
    },
    {
      "entropy": 6.55975375175476,
      "epoch": 2.8127775696934023,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042192082089640055,
      "loss": 5.9347,
      "mean_token_accuracy": 0.16011327505111694,
      "num_tokens": 56893148.0,
      "step": 26285
    },
    {
      "entropy": 6.546499729156494,
      "epoch": 2.8133126438011664,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004218918618261011,
      "loss": 5.7923,
      "mean_token_accuracy": 0.17324396967887878,
      "num_tokens": 56904008.0,
      "step": 26290
    },
    {
      "entropy": 6.527633571624756,
      "epoch": 2.8138477179089305,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042186289851431976,
      "loss": 5.8197,
      "mean_token_accuracy": 0.1709178328514099,
      "num_tokens": 56914731.0,
      "step": 26295
    },
    {
      "entropy": 6.584592676162719,
      "epoch": 2.814382792016694,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042183393096189285,
      "loss": 5.9508,
      "mean_token_accuracy": 0.163165120780468,
      "num_tokens": 56926037.0,
      "step": 26300
    },
    {
      "entropy": 6.445376014709472,
      "epoch": 2.8149178661244583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042180495916965686,
      "loss": 5.7624,
      "mean_token_accuracy": 0.17790987938642502,
      "num_tokens": 56936456.0,
      "step": 26305
    },
    {
      "entropy": 6.486808347702026,
      "epoch": 2.8154529402322224,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004217759831384483,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17486654967069626,
      "num_tokens": 56947929.0,
      "step": 26310
    },
    {
      "entropy": 6.486575508117676,
      "epoch": 2.815988014339986,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042174700286910376,
      "loss": 5.808,
      "mean_token_accuracy": 0.1736774206161499,
      "num_tokens": 56958153.0,
      "step": 26315
    },
    {
      "entropy": 6.567714500427246,
      "epoch": 2.81652308844775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004217180183624602,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16303154528141023,
      "num_tokens": 56968609.0,
      "step": 26320
    },
    {
      "entropy": 6.532463026046753,
      "epoch": 2.817058162555514,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004216890296193545,
      "loss": 5.884,
      "mean_token_accuracy": 0.1726097881793976,
      "num_tokens": 56978434.0,
      "step": 26325
    },
    {
      "entropy": 6.526335382461548,
      "epoch": 2.817593236663278,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004216600366406236,
      "loss": 5.8418,
      "mean_token_accuracy": 0.17260701060295106,
      "num_tokens": 56988826.0,
      "step": 26330
    },
    {
      "entropy": 6.507293605804444,
      "epoch": 2.8181283107710415,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042163103942710487,
      "loss": 5.8741,
      "mean_token_accuracy": 0.1678992584347725,
      "num_tokens": 57000385.0,
      "step": 26335
    },
    {
      "entropy": 6.453782844543457,
      "epoch": 2.8186633848788056,
      "grad_norm": 1.8046875,
      "learning_rate": 0.0004216020379796354,
      "loss": 5.8278,
      "mean_token_accuracy": 0.17806876301765442,
      "num_tokens": 57012137.0,
      "step": 26340
    },
    {
      "entropy": 6.519626522064209,
      "epoch": 2.8191984589865697,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042157303229905277,
      "loss": 5.9078,
      "mean_token_accuracy": 0.16525439023971558,
      "num_tokens": 57022533.0,
      "step": 26345
    },
    {
      "entropy": 6.552776575088501,
      "epoch": 2.8197335330943334,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042154402238619434,
      "loss": 5.8758,
      "mean_token_accuracy": 0.16218078434467315,
      "num_tokens": 57034438.0,
      "step": 26350
    },
    {
      "entropy": 6.568913555145263,
      "epoch": 2.8202686072020975,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042151500824189785,
      "loss": 5.8869,
      "mean_token_accuracy": 0.164529649913311,
      "num_tokens": 57046343.0,
      "step": 26355
    },
    {
      "entropy": 6.532766532897949,
      "epoch": 2.8208036813098616,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00042148598986700117,
      "loss": 5.8736,
      "mean_token_accuracy": 0.16228617429733277,
      "num_tokens": 57058269.0,
      "step": 26360
    },
    {
      "entropy": 6.5044067859649655,
      "epoch": 2.8213387554176252,
      "grad_norm": 1.125,
      "learning_rate": 0.00042145696726234207,
      "loss": 5.808,
      "mean_token_accuracy": 0.1749272584915161,
      "num_tokens": 57069192.0,
      "step": 26365
    },
    {
      "entropy": 6.54943528175354,
      "epoch": 2.8218738295253893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004214279404287586,
      "loss": 5.8723,
      "mean_token_accuracy": 0.17139019668102265,
      "num_tokens": 57080021.0,
      "step": 26370
    },
    {
      "entropy": 6.514299774169922,
      "epoch": 2.8224089036331534,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004213989093670889,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1700558692216873,
      "num_tokens": 57090336.0,
      "step": 26375
    },
    {
      "entropy": 6.494061470031738,
      "epoch": 2.822943977740917,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004213698740781713,
      "loss": 5.8183,
      "mean_token_accuracy": 0.1704588234424591,
      "num_tokens": 57101942.0,
      "step": 26380
    },
    {
      "entropy": 6.47497968673706,
      "epoch": 2.823479051848681,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004213408345628442,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1684766322374344,
      "num_tokens": 57112353.0,
      "step": 26385
    },
    {
      "entropy": 6.36620192527771,
      "epoch": 2.824014125956445,
      "grad_norm": 1.140625,
      "learning_rate": 0.000421311790821946,
      "loss": 5.7064,
      "mean_token_accuracy": 0.19547300040721893,
      "num_tokens": 57123796.0,
      "step": 26390
    },
    {
      "entropy": 6.492906856536865,
      "epoch": 2.824549200064209,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042128274285631545,
      "loss": 5.8389,
      "mean_token_accuracy": 0.16935500800609588,
      "num_tokens": 57134463.0,
      "step": 26395
    },
    {
      "entropy": 6.4836681365966795,
      "epoch": 2.8250842741719726,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042125369066679124,
      "loss": 5.7555,
      "mean_token_accuracy": 0.17516842633485794,
      "num_tokens": 57146964.0,
      "step": 26400
    },
    {
      "entropy": 6.558551979064942,
      "epoch": 2.8256193482797367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042122463425421224,
      "loss": 5.829,
      "mean_token_accuracy": 0.1674622893333435,
      "num_tokens": 57157322.0,
      "step": 26405
    },
    {
      "entropy": 6.424867820739746,
      "epoch": 2.826154422387501,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004211955736194175,
      "loss": 5.7596,
      "mean_token_accuracy": 0.17285378128290177,
      "num_tokens": 57168758.0,
      "step": 26410
    },
    {
      "entropy": 6.4669106006622314,
      "epoch": 2.8266894964952645,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042116650876324616,
      "loss": 5.8049,
      "mean_token_accuracy": 0.1757404774427414,
      "num_tokens": 57180322.0,
      "step": 26415
    },
    {
      "entropy": 6.534886884689331,
      "epoch": 2.8272245706030286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004211374396865373,
      "loss": 5.8249,
      "mean_token_accuracy": 0.16469650566577912,
      "num_tokens": 57190836.0,
      "step": 26420
    },
    {
      "entropy": 6.456713914871216,
      "epoch": 2.8277596447107927,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004211083663901305,
      "loss": 5.8144,
      "mean_token_accuracy": 0.16592526882886888,
      "num_tokens": 57202439.0,
      "step": 26425
    },
    {
      "entropy": 6.56335973739624,
      "epoch": 2.8282947188185563,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004210792888748651,
      "loss": 5.9656,
      "mean_token_accuracy": 0.15755388885736465,
      "num_tokens": 57213335.0,
      "step": 26430
    },
    {
      "entropy": 6.521729469299316,
      "epoch": 2.8288297929263204,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042105020714158076,
      "loss": 5.8699,
      "mean_token_accuracy": 0.16481989473104477,
      "num_tokens": 57223840.0,
      "step": 26435
    },
    {
      "entropy": 6.534981107711792,
      "epoch": 2.829364867034084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042102112119111725,
      "loss": 5.8707,
      "mean_token_accuracy": 0.16651411205530167,
      "num_tokens": 57234702.0,
      "step": 26440
    },
    {
      "entropy": 6.435993671417236,
      "epoch": 2.829899941141848,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004209920310243144,
      "loss": 5.7674,
      "mean_token_accuracy": 0.17474970370531082,
      "num_tokens": 57247415.0,
      "step": 26445
    },
    {
      "entropy": 6.426545667648315,
      "epoch": 2.830435015249612,
      "grad_norm": 1.25,
      "learning_rate": 0.0004209629366420121,
      "loss": 5.7841,
      "mean_token_accuracy": 0.17408457249403,
      "num_tokens": 57257768.0,
      "step": 26450
    },
    {
      "entropy": 6.557960128784179,
      "epoch": 2.830970089357376,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004209338380450506,
      "loss": 5.9753,
      "mean_token_accuracy": 0.16313797384500503,
      "num_tokens": 57268114.0,
      "step": 26455
    },
    {
      "entropy": 6.560983085632325,
      "epoch": 2.83150516346514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042090473523426994,
      "loss": 5.8975,
      "mean_token_accuracy": 0.16317763179540634,
      "num_tokens": 57278680.0,
      "step": 26460
    },
    {
      "entropy": 6.577279138565063,
      "epoch": 2.8320402375729037,
      "grad_norm": 1.125,
      "learning_rate": 0.00042087562821051055,
      "loss": 5.9135,
      "mean_token_accuracy": 0.1623005136847496,
      "num_tokens": 57289823.0,
      "step": 26465
    },
    {
      "entropy": 6.5240857124328615,
      "epoch": 2.832575311680668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042084651697461297,
      "loss": 5.8286,
      "mean_token_accuracy": 0.16584232449531555,
      "num_tokens": 57300137.0,
      "step": 26470
    },
    {
      "entropy": 6.489007186889649,
      "epoch": 2.833110385788432,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042081740152741763,
      "loss": 5.7579,
      "mean_token_accuracy": 0.1779188871383667,
      "num_tokens": 57311160.0,
      "step": 26475
    },
    {
      "entropy": 6.518664169311523,
      "epoch": 2.8336454598961955,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004207882818697654,
      "loss": 5.8305,
      "mean_token_accuracy": 0.17107800245285035,
      "num_tokens": 57322708.0,
      "step": 26480
    },
    {
      "entropy": 6.4869238376617435,
      "epoch": 2.8341805340039596,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000420759158002497,
      "loss": 5.7915,
      "mean_token_accuracy": 0.17050898671150208,
      "num_tokens": 57333788.0,
      "step": 26485
    },
    {
      "entropy": 6.531880855560303,
      "epoch": 2.8347156081117237,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042073002992645337,
      "loss": 5.8454,
      "mean_token_accuracy": 0.16931153535842897,
      "num_tokens": 57344806.0,
      "step": 26490
    },
    {
      "entropy": 6.555361986160278,
      "epoch": 2.8352506822194874,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004207008976424756,
      "loss": 5.9098,
      "mean_token_accuracy": 0.16410670727491378,
      "num_tokens": 57355086.0,
      "step": 26495
    },
    {
      "entropy": 6.4664164066314695,
      "epoch": 2.835785756327251,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004206717611514049,
      "loss": 5.7381,
      "mean_token_accuracy": 0.18020674139261245,
      "num_tokens": 57364837.0,
      "step": 26500
    },
    {
      "entropy": 6.520678758621216,
      "epoch": 2.836320830435015,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004206426204540825,
      "loss": 5.8489,
      "mean_token_accuracy": 0.1707681268453598,
      "num_tokens": 57376049.0,
      "step": 26505
    },
    {
      "entropy": 6.542647981643677,
      "epoch": 2.8368559045427792,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042061347555135,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16215457916259765,
      "num_tokens": 57387829.0,
      "step": 26510
    },
    {
      "entropy": 6.530613899230957,
      "epoch": 2.837390978650543,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004205843264440488,
      "loss": 5.8139,
      "mean_token_accuracy": 0.17437983453273773,
      "num_tokens": 57397165.0,
      "step": 26515
    },
    {
      "entropy": 6.4723801612854,
      "epoch": 2.837926052758307,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042055517313302064,
      "loss": 5.8039,
      "mean_token_accuracy": 0.1722429022192955,
      "num_tokens": 57408602.0,
      "step": 26520
    },
    {
      "entropy": 6.476287984848023,
      "epoch": 2.838461126866071,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004205260156191073,
      "loss": 5.7688,
      "mean_token_accuracy": 0.17672402560710906,
      "num_tokens": 57419393.0,
      "step": 26525
    },
    {
      "entropy": 6.495133781433106,
      "epoch": 2.8389962009738348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042049685390315074,
      "loss": 5.8559,
      "mean_token_accuracy": 0.1636224552989006,
      "num_tokens": 57431238.0,
      "step": 26530
    },
    {
      "entropy": 6.501330375671387,
      "epoch": 2.839531275081599,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042046768798599293,
      "loss": 5.8029,
      "mean_token_accuracy": 0.17608294934034346,
      "num_tokens": 57442080.0,
      "step": 26535
    },
    {
      "entropy": 6.4949370384216305,
      "epoch": 2.840066349189363,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004204385178684761,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16514965295791625,
      "num_tokens": 57452539.0,
      "step": 26540
    },
    {
      "entropy": 6.526943683624268,
      "epoch": 2.8406014232971266,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042040934355144245,
      "loss": 5.9401,
      "mean_token_accuracy": 0.16001699566841127,
      "num_tokens": 57464936.0,
      "step": 26545
    },
    {
      "entropy": 6.482459449768067,
      "epoch": 2.8411364974048907,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004203801650357346,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16893817931413652,
      "num_tokens": 57477460.0,
      "step": 26550
    },
    {
      "entropy": 6.508335542678833,
      "epoch": 2.8416715715126544,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004203509823221947,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16931583732366562,
      "num_tokens": 57487707.0,
      "step": 26555
    },
    {
      "entropy": 6.498920536041259,
      "epoch": 2.8422066456204185,
      "grad_norm": 1.125,
      "learning_rate": 0.0004203217954116657,
      "loss": 5.8882,
      "mean_token_accuracy": 0.16574712842702866,
      "num_tokens": 57498107.0,
      "step": 26560
    },
    {
      "entropy": 6.511497974395752,
      "epoch": 2.842741719728182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042029260430499033,
      "loss": 5.7911,
      "mean_token_accuracy": 0.1724663943052292,
      "num_tokens": 57509035.0,
      "step": 26565
    },
    {
      "entropy": 6.52892484664917,
      "epoch": 2.8432767938359462,
      "grad_norm": 1.0,
      "learning_rate": 0.00042026340900301135,
      "loss": 5.8776,
      "mean_token_accuracy": 0.16490163654088974,
      "num_tokens": 57521234.0,
      "step": 26570
    },
    {
      "entropy": 6.517225217819214,
      "epoch": 2.8438118679437103,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004202342095065719,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17254748940467834,
      "num_tokens": 57532294.0,
      "step": 26575
    },
    {
      "entropy": 6.560593843460083,
      "epoch": 2.844346942051474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000420205005816515,
      "loss": 5.8413,
      "mean_token_accuracy": 0.17309950590133666,
      "num_tokens": 57542047.0,
      "step": 26580
    },
    {
      "entropy": 6.42896990776062,
      "epoch": 2.844882016159238,
      "grad_norm": 1.140625,
      "learning_rate": 0.000420175797933684,
      "loss": 5.8463,
      "mean_token_accuracy": 0.17177257239818572,
      "num_tokens": 57553254.0,
      "step": 26585
    },
    {
      "entropy": 6.474917364120484,
      "epoch": 2.845417090267002,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042014658585892223,
      "loss": 5.788,
      "mean_token_accuracy": 0.18162437230348588,
      "num_tokens": 57564321.0,
      "step": 26590
    },
    {
      "entropy": 6.476011085510254,
      "epoch": 2.845952164374766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042011736959307323,
      "loss": 5.7414,
      "mean_token_accuracy": 0.1693373829126358,
      "num_tokens": 57575018.0,
      "step": 26595
    },
    {
      "entropy": 6.461613559722901,
      "epoch": 2.84648723848253,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042008814913698054,
      "loss": 5.8246,
      "mean_token_accuracy": 0.16508539766073227,
      "num_tokens": 57585698.0,
      "step": 26600
    },
    {
      "entropy": 6.4750700950622555,
      "epoch": 2.8470223125902936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004200589244914879,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17367863059043884,
      "num_tokens": 57595363.0,
      "step": 26605
    },
    {
      "entropy": 6.455188465118408,
      "epoch": 2.8475573866980577,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042002969565743925,
      "loss": 5.8072,
      "mean_token_accuracy": 0.17978428602218627,
      "num_tokens": 57605908.0,
      "step": 26610
    },
    {
      "entropy": 6.463802528381348,
      "epoch": 2.8480924608058213,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004200004626356785,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1804230645298958,
      "num_tokens": 57616172.0,
      "step": 26615
    },
    {
      "entropy": 6.4981614112854,
      "epoch": 2.8486275349135854,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041997122542704974,
      "loss": 5.7959,
      "mean_token_accuracy": 0.17419852465391159,
      "num_tokens": 57626770.0,
      "step": 26620
    },
    {
      "entropy": 6.432469654083252,
      "epoch": 2.8491626090213495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004199419840323972,
      "loss": 5.7176,
      "mean_token_accuracy": 0.17992641180753707,
      "num_tokens": 57636576.0,
      "step": 26625
    },
    {
      "entropy": 6.409944868087768,
      "epoch": 2.849697683129113,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041991273845256536,
      "loss": 5.8084,
      "mean_token_accuracy": 0.17217391282320021,
      "num_tokens": 57646431.0,
      "step": 26630
    },
    {
      "entropy": 6.442438220977783,
      "epoch": 2.8502327572368773,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041988348868839836,
      "loss": 5.8041,
      "mean_token_accuracy": 0.17318031936883926,
      "num_tokens": 57657412.0,
      "step": 26635
    },
    {
      "entropy": 6.5337615489959715,
      "epoch": 2.8507678313446414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004198542347407412,
      "loss": 5.8371,
      "mean_token_accuracy": 0.17313409447669983,
      "num_tokens": 57667434.0,
      "step": 26640
    },
    {
      "entropy": 6.505516815185547,
      "epoch": 2.851302905452405,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004198249766104382,
      "loss": 5.8467,
      "mean_token_accuracy": 0.1723646938800812,
      "num_tokens": 57677640.0,
      "step": 26645
    },
    {
      "entropy": 6.483448839187622,
      "epoch": 2.851837979560169,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004197957142983344,
      "loss": 5.8514,
      "mean_token_accuracy": 0.1678123354911804,
      "num_tokens": 57687909.0,
      "step": 26650
    },
    {
      "entropy": 6.564454984664917,
      "epoch": 2.8523730536679333,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004197664478052747,
      "loss": 5.9577,
      "mean_token_accuracy": 0.15795026272535323,
      "num_tokens": 57700374.0,
      "step": 26655
    },
    {
      "entropy": 6.5580870628356935,
      "epoch": 2.852908127775697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004197371771321042,
      "loss": 5.8259,
      "mean_token_accuracy": 0.17053146064281463,
      "num_tokens": 57711577.0,
      "step": 26660
    },
    {
      "entropy": 6.452719688415527,
      "epoch": 2.853443201883461,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041970790227966804,
      "loss": 5.7661,
      "mean_token_accuracy": 0.17981648594141006,
      "num_tokens": 57721259.0,
      "step": 26665
    },
    {
      "entropy": 6.423259162902832,
      "epoch": 2.8539782759912247,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041967862324881154,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16970555633306503,
      "num_tokens": 57731499.0,
      "step": 26670
    },
    {
      "entropy": 6.549926662445069,
      "epoch": 2.8545133500989888,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004196493400403801,
      "loss": 5.9255,
      "mean_token_accuracy": 0.16070932894945145,
      "num_tokens": 57742343.0,
      "step": 26675
    },
    {
      "entropy": 6.503125143051148,
      "epoch": 2.8550484242067524,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004196200526552193,
      "loss": 5.8389,
      "mean_token_accuracy": 0.17062205374240874,
      "num_tokens": 57752942.0,
      "step": 26680
    },
    {
      "entropy": 6.530670261383056,
      "epoch": 2.8555834983145165,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004195907610941748,
      "loss": 5.8455,
      "mean_token_accuracy": 0.17119980603456497,
      "num_tokens": 57763623.0,
      "step": 26685
    },
    {
      "entropy": 6.465994024276734,
      "epoch": 2.8561185724222806,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004195614653580924,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16890647858381272,
      "num_tokens": 57774538.0,
      "step": 26690
    },
    {
      "entropy": 6.486776876449585,
      "epoch": 2.8566536465300443,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041953216544781796,
      "loss": 5.8327,
      "mean_token_accuracy": 0.16495385020971298,
      "num_tokens": 57785420.0,
      "step": 26695
    },
    {
      "entropy": 6.477488565444946,
      "epoch": 2.8571887206378084,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041950286136419756,
      "loss": 5.8289,
      "mean_token_accuracy": 0.16357463449239731,
      "num_tokens": 57796488.0,
      "step": 26700
    },
    {
      "entropy": 6.531862592697143,
      "epoch": 2.8577237947455725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041947355310807737,
      "loss": 5.8193,
      "mean_token_accuracy": 0.16954318732023238,
      "num_tokens": 57806391.0,
      "step": 26705
    },
    {
      "entropy": 6.468953466415405,
      "epoch": 2.858258868853336,
      "grad_norm": 1.125,
      "learning_rate": 0.00041944424068030364,
      "loss": 5.8458,
      "mean_token_accuracy": 0.16397044509649278,
      "num_tokens": 57817410.0,
      "step": 26710
    },
    {
      "entropy": 6.512260150909424,
      "epoch": 2.8587939429611002,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041941492408172277,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1709974303841591,
      "num_tokens": 57828582.0,
      "step": 26715
    },
    {
      "entropy": 6.467840909957886,
      "epoch": 2.859329017068864,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041938560331318124,
      "loss": 5.7812,
      "mean_token_accuracy": 0.18148132115602494,
      "num_tokens": 57838962.0,
      "step": 26720
    },
    {
      "entropy": 6.544838285446167,
      "epoch": 2.859864091176628,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004193562783755257,
      "loss": 5.8376,
      "mean_token_accuracy": 0.17214002162218095,
      "num_tokens": 57849335.0,
      "step": 26725
    },
    {
      "entropy": 6.521011686325073,
      "epoch": 2.8603991652843916,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004193269492696029,
      "loss": 5.8309,
      "mean_token_accuracy": 0.17162279039621353,
      "num_tokens": 57860904.0,
      "step": 26730
    },
    {
      "entropy": 6.506897020339966,
      "epoch": 2.8609342393921557,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004192976159962597,
      "loss": 5.8507,
      "mean_token_accuracy": 0.16605626344680785,
      "num_tokens": 57870861.0,
      "step": 26735
    },
    {
      "entropy": 6.4800214767456055,
      "epoch": 2.86146931349992,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004192682785563432,
      "loss": 5.867,
      "mean_token_accuracy": 0.1655280038714409,
      "num_tokens": 57881590.0,
      "step": 26740
    },
    {
      "entropy": 6.508210134506226,
      "epoch": 2.8620043876076835,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004192389369507004,
      "loss": 5.7962,
      "mean_token_accuracy": 0.17699856758117677,
      "num_tokens": 57892736.0,
      "step": 26745
    },
    {
      "entropy": 6.543175792694091,
      "epoch": 2.8625394617154476,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004192095911801785,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16621468663215638,
      "num_tokens": 57904340.0,
      "step": 26750
    },
    {
      "entropy": 6.517156553268433,
      "epoch": 2.8630745358232117,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000419180241245625,
      "loss": 5.8287,
      "mean_token_accuracy": 0.17286199629306792,
      "num_tokens": 57916369.0,
      "step": 26755
    },
    {
      "entropy": 6.511870527267456,
      "epoch": 2.8636096099309754,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004191508871478873,
      "loss": 5.9067,
      "mean_token_accuracy": 0.16237364262342452,
      "num_tokens": 57926867.0,
      "step": 26760
    },
    {
      "entropy": 6.514552593231201,
      "epoch": 2.8641446840387395,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041912152888781283,
      "loss": 5.7733,
      "mean_token_accuracy": 0.16661376655101776,
      "num_tokens": 57938408.0,
      "step": 26765
    },
    {
      "entropy": 6.537756872177124,
      "epoch": 2.8646797581465036,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041909216646624967,
      "loss": 5.799,
      "mean_token_accuracy": 0.17540223449468612,
      "num_tokens": 57948821.0,
      "step": 26770
    },
    {
      "entropy": 6.523450326919556,
      "epoch": 2.865214832254267,
      "grad_norm": 1.0,
      "learning_rate": 0.0004190627998840453,
      "loss": 5.8989,
      "mean_token_accuracy": 0.16614235788583756,
      "num_tokens": 57959589.0,
      "step": 26775
    },
    {
      "entropy": 6.487152576446533,
      "epoch": 2.8657499063620313,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004190334291420479,
      "loss": 5.804,
      "mean_token_accuracy": 0.17093757092952727,
      "num_tokens": 57970508.0,
      "step": 26780
    },
    {
      "entropy": 6.516197633743286,
      "epoch": 2.866284980469795,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004190040542411055,
      "loss": 5.8194,
      "mean_token_accuracy": 0.1720832645893097,
      "num_tokens": 57980411.0,
      "step": 26785
    },
    {
      "entropy": 6.497263479232788,
      "epoch": 2.866820054577559,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041897467518206614,
      "loss": 5.8587,
      "mean_token_accuracy": 0.1711605116724968,
      "num_tokens": 57991457.0,
      "step": 26790
    },
    {
      "entropy": 6.538822603225708,
      "epoch": 2.8673551286853227,
      "grad_norm": 1.125,
      "learning_rate": 0.0004189452919657784,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15901913791894912,
      "num_tokens": 58001860.0,
      "step": 26795
    },
    {
      "entropy": 6.555558538436889,
      "epoch": 2.867890202793087,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041891590459309055,
      "loss": 5.9097,
      "mean_token_accuracy": 0.1613926738500595,
      "num_tokens": 58011958.0,
      "step": 26800
    },
    {
      "entropy": 6.489037704467774,
      "epoch": 2.868425276900851,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041888651306485117,
      "loss": 5.8164,
      "mean_token_accuracy": 0.17008043974637985,
      "num_tokens": 58022216.0,
      "step": 26805
    },
    {
      "entropy": 6.514347076416016,
      "epoch": 2.8689603510086146,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041885711738190895,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16997911632061005,
      "num_tokens": 58032502.0,
      "step": 26810
    },
    {
      "entropy": 6.4453576564788815,
      "epoch": 2.8694954251163787,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004188277175451126,
      "loss": 5.8458,
      "mean_token_accuracy": 0.17206878066062928,
      "num_tokens": 58043944.0,
      "step": 26815
    },
    {
      "entropy": 6.498595142364502,
      "epoch": 2.8700304992241428,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041879831355531123,
      "loss": 5.9087,
      "mean_token_accuracy": 0.16456774473190308,
      "num_tokens": 58055857.0,
      "step": 26820
    },
    {
      "entropy": 6.501140832901001,
      "epoch": 2.8705655733319064,
      "grad_norm": 1.125,
      "learning_rate": 0.0004187689054133537,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17018964588642121,
      "num_tokens": 58067070.0,
      "step": 26825
    },
    {
      "entropy": 6.469398927688599,
      "epoch": 2.8711006474396705,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004187394931200893,
      "loss": 5.6764,
      "mean_token_accuracy": 0.1830156311392784,
      "num_tokens": 58077922.0,
      "step": 26830
    },
    {
      "entropy": 6.5225982666015625,
      "epoch": 2.871635721547434,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004187100766763671,
      "loss": 5.8024,
      "mean_token_accuracy": 0.17097950279712676,
      "num_tokens": 58089698.0,
      "step": 26835
    },
    {
      "entropy": 6.556467723846436,
      "epoch": 2.8721707956551983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041868065608303674,
      "loss": 5.8001,
      "mean_token_accuracy": 0.17433367669582367,
      "num_tokens": 58100640.0,
      "step": 26840
    },
    {
      "entropy": 6.488606262207031,
      "epoch": 2.872705869762962,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004186512313409476,
      "loss": 5.8731,
      "mean_token_accuracy": 0.16693221032619476,
      "num_tokens": 58112810.0,
      "step": 26845
    },
    {
      "entropy": 6.575130891799927,
      "epoch": 2.873240943870726,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004186218024509493,
      "loss": 5.8979,
      "mean_token_accuracy": 0.1676798641681671,
      "num_tokens": 58123818.0,
      "step": 26850
    },
    {
      "entropy": 6.574738216400147,
      "epoch": 2.87377601797849,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041859236941389175,
      "loss": 5.859,
      "mean_token_accuracy": 0.16879161298274994,
      "num_tokens": 58134785.0,
      "step": 26855
    },
    {
      "entropy": 6.5258142948150635,
      "epoch": 2.874311092086254,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004185629322306246,
      "loss": 5.8538,
      "mean_token_accuracy": 0.17425422370433807,
      "num_tokens": 58144987.0,
      "step": 26860
    },
    {
      "entropy": 6.511073350906372,
      "epoch": 2.874846166194018,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041853349090199805,
      "loss": 5.802,
      "mean_token_accuracy": 0.17430074959993364,
      "num_tokens": 58155031.0,
      "step": 26865
    },
    {
      "entropy": 6.488924646377564,
      "epoch": 2.875381240301782,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000418504045428862,
      "loss": 5.796,
      "mean_token_accuracy": 0.17291359603405,
      "num_tokens": 58165216.0,
      "step": 26870
    },
    {
      "entropy": 6.53068437576294,
      "epoch": 2.8759163144095456,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004184745958120669,
      "loss": 5.8405,
      "mean_token_accuracy": 0.17511154115200042,
      "num_tokens": 58176329.0,
      "step": 26875
    },
    {
      "entropy": 6.555236864089966,
      "epoch": 2.8764513885173097,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041844514205246293,
      "loss": 5.8543,
      "mean_token_accuracy": 0.16746763288974761,
      "num_tokens": 58187058.0,
      "step": 26880
    },
    {
      "entropy": 6.518997526168823,
      "epoch": 2.876986462625074,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004184156841509007,
      "loss": 5.8508,
      "mean_token_accuracy": 0.1703554719686508,
      "num_tokens": 58196579.0,
      "step": 26885
    },
    {
      "entropy": 6.40214467048645,
      "epoch": 2.8775215367328375,
      "grad_norm": 1.125,
      "learning_rate": 0.00041838622210823076,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17343718260526658,
      "num_tokens": 58206689.0,
      "step": 26890
    },
    {
      "entropy": 6.486832904815674,
      "epoch": 2.878056610840601,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004183567559253037,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16956965625286102,
      "num_tokens": 58218093.0,
      "step": 26895
    },
    {
      "entropy": 6.543793535232544,
      "epoch": 2.8785916849483653,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041832728560297055,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16466719806194305,
      "num_tokens": 58228587.0,
      "step": 26900
    },
    {
      "entropy": 6.4970564365386965,
      "epoch": 2.8791267590561294,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041829781114208214,
      "loss": 5.7506,
      "mean_token_accuracy": 0.17292326986789702,
      "num_tokens": 58240500.0,
      "step": 26905
    },
    {
      "entropy": 6.45921220779419,
      "epoch": 2.879661833163893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041826833254348956,
      "loss": 5.8098,
      "mean_token_accuracy": 0.16611901223659514,
      "num_tokens": 58252253.0,
      "step": 26910
    },
    {
      "entropy": 6.537049341201782,
      "epoch": 2.880196907271657,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041823884980804406,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16698794960975646,
      "num_tokens": 58263356.0,
      "step": 26915
    },
    {
      "entropy": 6.429355955123901,
      "epoch": 2.880731981379421,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041820936293659685,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17403949350118636,
      "num_tokens": 58273270.0,
      "step": 26920
    },
    {
      "entropy": 6.511911916732788,
      "epoch": 2.881267055487185,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041817987192999947,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16176581680774688,
      "num_tokens": 58284669.0,
      "step": 26925
    },
    {
      "entropy": 6.545569229125976,
      "epoch": 2.881802129594949,
      "grad_norm": 1.25,
      "learning_rate": 0.00041815037678910334,
      "loss": 5.9203,
      "mean_token_accuracy": 0.1705167219042778,
      "num_tokens": 58294791.0,
      "step": 26930
    },
    {
      "entropy": 6.535962533950806,
      "epoch": 2.882337203702713,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004181208775147602,
      "loss": 5.866,
      "mean_token_accuracy": 0.1691754564642906,
      "num_tokens": 58305789.0,
      "step": 26935
    },
    {
      "entropy": 6.5269859790802,
      "epoch": 2.8828722778104767,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004180913741078218,
      "loss": 5.8388,
      "mean_token_accuracy": 0.1702611044049263,
      "num_tokens": 58317547.0,
      "step": 26940
    },
    {
      "entropy": 6.517317104339599,
      "epoch": 2.883407351918241,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004180618665691402,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16964101195335388,
      "num_tokens": 58328408.0,
      "step": 26945
    },
    {
      "entropy": 6.3612336158752445,
      "epoch": 2.8839424260260045,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004180323548995672,
      "loss": 5.7178,
      "mean_token_accuracy": 0.18455341905355455,
      "num_tokens": 58339356.0,
      "step": 26950
    },
    {
      "entropy": 6.4419557571411135,
      "epoch": 2.8844775001337686,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041800283909995505,
      "loss": 5.7713,
      "mean_token_accuracy": 0.1789791226387024,
      "num_tokens": 58349016.0,
      "step": 26955
    },
    {
      "entropy": 6.517625999450684,
      "epoch": 2.8850125742415322,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004179733191711561,
      "loss": 5.9114,
      "mean_token_accuracy": 0.1584542766213417,
      "num_tokens": 58360188.0,
      "step": 26960
    },
    {
      "entropy": 6.577715682983398,
      "epoch": 2.8855476483492963,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041794379511402263,
      "loss": 5.8766,
      "mean_token_accuracy": 0.1638513594865799,
      "num_tokens": 58370733.0,
      "step": 26965
    },
    {
      "entropy": 6.503220558166504,
      "epoch": 2.8860827224570604,
      "grad_norm": 1.125,
      "learning_rate": 0.0004179142669294071,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16740099638700484,
      "num_tokens": 58382249.0,
      "step": 26970
    },
    {
      "entropy": 6.490060424804687,
      "epoch": 2.886617796564824,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041788473461816224,
      "loss": 5.7813,
      "mean_token_accuracy": 0.17220530807971954,
      "num_tokens": 58393598.0,
      "step": 26975
    },
    {
      "entropy": 6.503634405136109,
      "epoch": 2.887152870672588,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041785519818114073,
      "loss": 5.8295,
      "mean_token_accuracy": 0.16628775000572205,
      "num_tokens": 58403920.0,
      "step": 26980
    },
    {
      "entropy": 6.475538349151611,
      "epoch": 2.8876879447803523,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041782565761919546,
      "loss": 5.8671,
      "mean_token_accuracy": 0.16254540532827377,
      "num_tokens": 58413691.0,
      "step": 26985
    },
    {
      "entropy": 6.561694765090943,
      "epoch": 2.888223018888116,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041779611293317936,
      "loss": 5.8701,
      "mean_token_accuracy": 0.16567191183567048,
      "num_tokens": 58425995.0,
      "step": 26990
    },
    {
      "entropy": 6.522514772415161,
      "epoch": 2.88875809299588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004177665641239456,
      "loss": 5.8278,
      "mean_token_accuracy": 0.16866870373487472,
      "num_tokens": 58435984.0,
      "step": 26995
    },
    {
      "entropy": 6.491448831558228,
      "epoch": 2.8892931671036437,
      "grad_norm": 1.125,
      "learning_rate": 0.00041773701119234734,
      "loss": 5.8719,
      "mean_token_accuracy": 0.16593987196683885,
      "num_tokens": 58447630.0,
      "step": 27000
    },
    {
      "epoch": 2.8892931671036437,
      "eval_entropy": 6.249952104599328,
      "eval_loss": 5.919104099273682,
      "eval_mean_token_accuracy": 0.17370621865424465,
      "eval_num_tokens": 58447630.0,
      "eval_runtime": 17.6124,
      "eval_samples_per_second": 1685.116,
      "eval_steps_per_second": 210.647,
      "step": 27000
    },
    {
      "entropy": 6.547634935379028,
      "epoch": 2.889828241211408,
      "grad_norm": 1.21875,
      "learning_rate": 0.000417707454139238,
      "loss": 5.9032,
      "mean_token_accuracy": 0.1624731555581093,
      "num_tokens": 58458055.0,
      "step": 27005
    },
    {
      "entropy": 6.535300302505493,
      "epoch": 2.8903633153191715,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004176778929654709,
      "loss": 5.8678,
      "mean_token_accuracy": 0.16417768895626067,
      "num_tokens": 58469115.0,
      "step": 27010
    },
    {
      "entropy": 6.49657883644104,
      "epoch": 2.8908983894269356,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004176483276718997,
      "loss": 5.7957,
      "mean_token_accuracy": 0.1791925996541977,
      "num_tokens": 58480038.0,
      "step": 27015
    },
    {
      "entropy": 6.514646863937378,
      "epoch": 2.8914334635346997,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004176187582593781,
      "loss": 5.9058,
      "mean_token_accuracy": 0.16774394363164902,
      "num_tokens": 58490413.0,
      "step": 27020
    },
    {
      "entropy": 6.508981513977051,
      "epoch": 2.8919685376424633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004175891847287599,
      "loss": 5.8534,
      "mean_token_accuracy": 0.17019887119531632,
      "num_tokens": 58500162.0,
      "step": 27025
    },
    {
      "entropy": 6.440421485900879,
      "epoch": 2.8925036117502274,
      "grad_norm": 1.140625,
      "learning_rate": 0.000417559607080899,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1805241122841835,
      "num_tokens": 58511174.0,
      "step": 27030
    },
    {
      "entropy": 6.5441569805145265,
      "epoch": 2.8930386858579915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041753002531664954,
      "loss": 5.8854,
      "mean_token_accuracy": 0.1640648603439331,
      "num_tokens": 58521891.0,
      "step": 27035
    },
    {
      "entropy": 6.512217473983765,
      "epoch": 2.893573759965755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004175004394368656,
      "loss": 5.8703,
      "mean_token_accuracy": 0.16520286202430726,
      "num_tokens": 58532029.0,
      "step": 27040
    },
    {
      "entropy": 6.528636074066162,
      "epoch": 2.8941088340735193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041747084944240154,
      "loss": 5.8691,
      "mean_token_accuracy": 0.1688498392701149,
      "num_tokens": 58542568.0,
      "step": 27045
    },
    {
      "entropy": 6.550668668746948,
      "epoch": 2.8946439081812834,
      "grad_norm": 1.125,
      "learning_rate": 0.0004174412553341116,
      "loss": 5.8257,
      "mean_token_accuracy": 0.1727073982357979,
      "num_tokens": 58553396.0,
      "step": 27050
    },
    {
      "entropy": 6.492252588272095,
      "epoch": 2.895178982289047,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004174116571128506,
      "loss": 5.8586,
      "mean_token_accuracy": 0.17057157754898072,
      "num_tokens": 58564792.0,
      "step": 27055
    },
    {
      "entropy": 6.510940074920654,
      "epoch": 2.895714056396811,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041738205477947284,
      "loss": 5.899,
      "mean_token_accuracy": 0.16881562173366546,
      "num_tokens": 58575666.0,
      "step": 27060
    },
    {
      "entropy": 6.459005832672119,
      "epoch": 2.8962491305045748,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041735244833483325,
      "loss": 5.7874,
      "mean_token_accuracy": 0.1730884850025177,
      "num_tokens": 58587411.0,
      "step": 27065
    },
    {
      "entropy": 6.50862488746643,
      "epoch": 2.896784204612339,
      "grad_norm": 1.125,
      "learning_rate": 0.00041732283777978687,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1662924960255623,
      "num_tokens": 58598283.0,
      "step": 27070
    },
    {
      "entropy": 6.507575416564942,
      "epoch": 2.8973192787201025,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004172932231151885,
      "loss": 5.8192,
      "mean_token_accuracy": 0.1744409903883934,
      "num_tokens": 58608982.0,
      "step": 27075
    },
    {
      "entropy": 6.5299290180206295,
      "epoch": 2.8978543528278666,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004172636043418932,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16884707510471345,
      "num_tokens": 58619022.0,
      "step": 27080
    },
    {
      "entropy": 6.449179029464721,
      "epoch": 2.8983894269356307,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004172339814607564,
      "loss": 5.7675,
      "mean_token_accuracy": 0.1696891650557518,
      "num_tokens": 58630559.0,
      "step": 27085
    },
    {
      "entropy": 6.5142944812774655,
      "epoch": 2.8989245010433944,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004172043544726334,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1633377902209759,
      "num_tokens": 58641346.0,
      "step": 27090
    },
    {
      "entropy": 6.538030099868775,
      "epoch": 2.8994595751511585,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004171747233783796,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16701075583696365,
      "num_tokens": 58652697.0,
      "step": 27095
    },
    {
      "entropy": 6.52331953048706,
      "epoch": 2.8999946492589226,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004171450881788507,
      "loss": 5.8397,
      "mean_token_accuracy": 0.1613471880555153,
      "num_tokens": 58665574.0,
      "step": 27100
    },
    {
      "entropy": 6.508168125152588,
      "epoch": 2.9005297233666862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041711544887490233,
      "loss": 5.8087,
      "mean_token_accuracy": 0.1732803001999855,
      "num_tokens": 58676079.0,
      "step": 27105
    },
    {
      "entropy": 6.503347873687744,
      "epoch": 2.9010647974744503,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004170858054673903,
      "loss": 5.8386,
      "mean_token_accuracy": 0.1691518098115921,
      "num_tokens": 58688465.0,
      "step": 27110
    },
    {
      "entropy": 6.518226861953735,
      "epoch": 2.901599871582214,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004170561579571706,
      "loss": 5.8372,
      "mean_token_accuracy": 0.17220430821180344,
      "num_tokens": 58698772.0,
      "step": 27115
    },
    {
      "entropy": 6.572803258895874,
      "epoch": 2.902134945689978,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041702650634509934,
      "loss": 5.9663,
      "mean_token_accuracy": 0.1599682703614235,
      "num_tokens": 58709505.0,
      "step": 27120
    },
    {
      "entropy": 6.512079191207886,
      "epoch": 2.9026700197977418,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041699685063203263,
      "loss": 5.8442,
      "mean_token_accuracy": 0.17408496588468553,
      "num_tokens": 58718652.0,
      "step": 27125
    },
    {
      "entropy": 6.5388082504272464,
      "epoch": 2.903205093905506,
      "grad_norm": 1.0,
      "learning_rate": 0.00041696719081882685,
      "loss": 5.8978,
      "mean_token_accuracy": 0.1652471363544464,
      "num_tokens": 58730207.0,
      "step": 27130
    },
    {
      "entropy": 6.54032416343689,
      "epoch": 2.90374016801327,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041693752690633837,
      "loss": 5.8177,
      "mean_token_accuracy": 0.1713273361325264,
      "num_tokens": 58741418.0,
      "step": 27135
    },
    {
      "entropy": 6.559287786483765,
      "epoch": 2.9042752421210336,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004169078588954237,
      "loss": 5.916,
      "mean_token_accuracy": 0.1640244796872139,
      "num_tokens": 58753535.0,
      "step": 27140
    },
    {
      "entropy": 6.536944913864136,
      "epoch": 2.9048103162287977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004168781867869396,
      "loss": 5.8107,
      "mean_token_accuracy": 0.1683312952518463,
      "num_tokens": 58766220.0,
      "step": 27145
    },
    {
      "entropy": 6.539760255813599,
      "epoch": 2.905345390336562,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041684851058174284,
      "loss": 5.8321,
      "mean_token_accuracy": 0.1661725103855133,
      "num_tokens": 58776795.0,
      "step": 27150
    },
    {
      "entropy": 6.533239030838013,
      "epoch": 2.9058804644443255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004168188302806902,
      "loss": 5.8494,
      "mean_token_accuracy": 0.16924497336149216,
      "num_tokens": 58787495.0,
      "step": 27155
    },
    {
      "entropy": 6.53238730430603,
      "epoch": 2.9064155385520896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004167891458846387,
      "loss": 5.8415,
      "mean_token_accuracy": 0.16523651778697968,
      "num_tokens": 58798518.0,
      "step": 27160
    },
    {
      "entropy": 6.53167691230774,
      "epoch": 2.9069506126598537,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041675945739444567,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16810425221920014,
      "num_tokens": 58808849.0,
      "step": 27165
    },
    {
      "entropy": 6.535613012313843,
      "epoch": 2.9074856867676173,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004167297648109682,
      "loss": 5.8238,
      "mean_token_accuracy": 0.1684948220849037,
      "num_tokens": 58821110.0,
      "step": 27170
    },
    {
      "entropy": 6.55921745300293,
      "epoch": 2.908020760875381,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041670006813506366,
      "loss": 5.888,
      "mean_token_accuracy": 0.16831713914871216,
      "num_tokens": 58831433.0,
      "step": 27175
    },
    {
      "entropy": 6.552947664260865,
      "epoch": 2.908555834983145,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041667036736758954,
      "loss": 5.891,
      "mean_token_accuracy": 0.1664306789636612,
      "num_tokens": 58842031.0,
      "step": 27180
    },
    {
      "entropy": 6.456445980072021,
      "epoch": 2.909090909090909,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041664066250940353,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16886749267578124,
      "num_tokens": 58852767.0,
      "step": 27185
    },
    {
      "entropy": 6.529803466796875,
      "epoch": 2.909625983198673,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041661095356136326,
      "loss": 5.8496,
      "mean_token_accuracy": 0.16827345490455628,
      "num_tokens": 58864147.0,
      "step": 27190
    },
    {
      "entropy": 6.52880072593689,
      "epoch": 2.910161057306437,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041658124052432655,
      "loss": 5.8279,
      "mean_token_accuracy": 0.17267894148826599,
      "num_tokens": 58874059.0,
      "step": 27195
    },
    {
      "entropy": 6.507177829742432,
      "epoch": 2.910696131414201,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004165515233991514,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16970134377479554,
      "num_tokens": 58884433.0,
      "step": 27200
    },
    {
      "entropy": 6.562891006469727,
      "epoch": 2.9112312055219647,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004165218021866961,
      "loss": 5.9376,
      "mean_token_accuracy": 0.158113469183445,
      "num_tokens": 58896062.0,
      "step": 27205
    },
    {
      "entropy": 6.508063888549804,
      "epoch": 2.911766279629729,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004164920768878184,
      "loss": 5.8497,
      "mean_token_accuracy": 0.16528478413820266,
      "num_tokens": 58906926.0,
      "step": 27210
    },
    {
      "entropy": 6.470512342453003,
      "epoch": 2.912301353737493,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041646234750337697,
      "loss": 5.8045,
      "mean_token_accuracy": 0.172328183054924,
      "num_tokens": 58919230.0,
      "step": 27215
    },
    {
      "entropy": 6.523137331008911,
      "epoch": 2.9128364278452565,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004164326140342302,
      "loss": 5.8842,
      "mean_token_accuracy": 0.16409091055393218,
      "num_tokens": 58929094.0,
      "step": 27220
    },
    {
      "entropy": 6.536241579055786,
      "epoch": 2.9133715019530206,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041640287648123645,
      "loss": 5.8457,
      "mean_token_accuracy": 0.17607535421848297,
      "num_tokens": 58939952.0,
      "step": 27225
    },
    {
      "entropy": 6.566717672348022,
      "epoch": 2.9139065760607843,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041637313484525464,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1725477620959282,
      "num_tokens": 58950520.0,
      "step": 27230
    },
    {
      "entropy": 6.587516736984253,
      "epoch": 2.9144416501685484,
      "grad_norm": 1.25,
      "learning_rate": 0.0004163433891271434,
      "loss": 5.8874,
      "mean_token_accuracy": 0.16510043144226075,
      "num_tokens": 58960578.0,
      "step": 27235
    },
    {
      "entropy": 6.5473775386810305,
      "epoch": 2.914976724276312,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041631363932776154,
      "loss": 5.9139,
      "mean_token_accuracy": 0.16419365257024765,
      "num_tokens": 58972188.0,
      "step": 27240
    },
    {
      "entropy": 6.5734649181365965,
      "epoch": 2.915511798384076,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041628388544796835,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1766703724861145,
      "num_tokens": 58982852.0,
      "step": 27245
    },
    {
      "entropy": 6.485908126831054,
      "epoch": 2.9160468724918402,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004162541274886228,
      "loss": 5.7583,
      "mean_token_accuracy": 0.17881738394498825,
      "num_tokens": 58993713.0,
      "step": 27250
    },
    {
      "entropy": 6.422953319549561,
      "epoch": 2.916581946599604,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041622436545058405,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1768077000975609,
      "num_tokens": 59004021.0,
      "step": 27255
    },
    {
      "entropy": 6.391600179672241,
      "epoch": 2.917117020707368,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041619459933471174,
      "loss": 5.6992,
      "mean_token_accuracy": 0.19364406615495683,
      "num_tokens": 59016592.0,
      "step": 27260
    },
    {
      "entropy": 6.566645526885987,
      "epoch": 2.917652094815132,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004161648291418651,
      "loss": 5.7935,
      "mean_token_accuracy": 0.17517084777355194,
      "num_tokens": 59027116.0,
      "step": 27265
    },
    {
      "entropy": 6.532719898223877,
      "epoch": 2.9181871689228958,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004161350548729039,
      "loss": 5.87,
      "mean_token_accuracy": 0.1675683319568634,
      "num_tokens": 59036917.0,
      "step": 27270
    },
    {
      "entropy": 6.517678451538086,
      "epoch": 2.91872224303066,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004161052765286878,
      "loss": 5.819,
      "mean_token_accuracy": 0.16960546523332595,
      "num_tokens": 59046588.0,
      "step": 27275
    },
    {
      "entropy": 6.523439931869507,
      "epoch": 2.9192573171384235,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004160754941100767,
      "loss": 5.7752,
      "mean_token_accuracy": 0.17297797799110412,
      "num_tokens": 59057710.0,
      "step": 27280
    },
    {
      "entropy": 6.580919456481934,
      "epoch": 2.9197923912461876,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041604570761793054,
      "loss": 5.8974,
      "mean_token_accuracy": 0.1638222813606262,
      "num_tokens": 59068795.0,
      "step": 27285
    },
    {
      "entropy": 6.519959306716919,
      "epoch": 2.9203274653539513,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004160159170531094,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17266649454832078,
      "num_tokens": 59079266.0,
      "step": 27290
    },
    {
      "entropy": 6.552259206771851,
      "epoch": 2.9208625394617154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004159861224164734,
      "loss": 5.9431,
      "mean_token_accuracy": 0.16684074997901915,
      "num_tokens": 59088827.0,
      "step": 27295
    },
    {
      "entropy": 6.461137437820435,
      "epoch": 2.9213976135694795,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000415956323708883,
      "loss": 5.8725,
      "mean_token_accuracy": 0.16903913468122483,
      "num_tokens": 59099097.0,
      "step": 27300
    },
    {
      "entropy": 6.5426253318786625,
      "epoch": 2.921932687677243,
      "grad_norm": 1.125,
      "learning_rate": 0.00041592652093119846,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1750080928206444,
      "num_tokens": 59110175.0,
      "step": 27305
    },
    {
      "entropy": 6.579403686523437,
      "epoch": 2.9224677617850072,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041589671408428055,
      "loss": 5.8726,
      "mean_token_accuracy": 0.16818459779024125,
      "num_tokens": 59121133.0,
      "step": 27310
    },
    {
      "entropy": 6.535346364974975,
      "epoch": 2.9230028358927713,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041586690316898977,
      "loss": 5.8691,
      "mean_token_accuracy": 0.16591157615184784,
      "num_tokens": 59132750.0,
      "step": 27315
    },
    {
      "entropy": 6.5130870819091795,
      "epoch": 2.923537910000535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041583708818618693,
      "loss": 5.8417,
      "mean_token_accuracy": 0.17179012596607207,
      "num_tokens": 59143488.0,
      "step": 27320
    },
    {
      "entropy": 6.434557962417602,
      "epoch": 2.924072984108299,
      "grad_norm": 1.125,
      "learning_rate": 0.000415807269136733,
      "loss": 5.8111,
      "mean_token_accuracy": 0.16871285289525986,
      "num_tokens": 59154118.0,
      "step": 27325
    },
    {
      "entropy": 6.5125792026519775,
      "epoch": 2.924608058216063,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041577744602148893,
      "loss": 5.8699,
      "mean_token_accuracy": 0.17001231014728546,
      "num_tokens": 59164736.0,
      "step": 27330
    },
    {
      "entropy": 6.507899045944214,
      "epoch": 2.925143132323827,
      "grad_norm": 1.046875,
      "learning_rate": 0.000415747618841316,
      "loss": 5.8769,
      "mean_token_accuracy": 0.16230118721723558,
      "num_tokens": 59175895.0,
      "step": 27335
    },
    {
      "entropy": 6.501858282089233,
      "epoch": 2.925678206431591,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004157177875970752,
      "loss": 5.8265,
      "mean_token_accuracy": 0.17246880531311035,
      "num_tokens": 59186692.0,
      "step": 27340
    },
    {
      "entropy": 6.535362911224365,
      "epoch": 2.9262132805393546,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004156879522896281,
      "loss": 5.8183,
      "mean_token_accuracy": 0.1758524462580681,
      "num_tokens": 59197731.0,
      "step": 27345
    },
    {
      "entropy": 6.5147181987762455,
      "epoch": 2.9267483546471187,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004156581129198362,
      "loss": 5.8182,
      "mean_token_accuracy": 0.16914924681186677,
      "num_tokens": 59209470.0,
      "step": 27350
    },
    {
      "entropy": 6.441955518722534,
      "epoch": 2.9272834287548823,
      "grad_norm": 1.15625,
      "learning_rate": 0.000415628269488561,
      "loss": 5.8263,
      "mean_token_accuracy": 0.1693178802728653,
      "num_tokens": 59220085.0,
      "step": 27355
    },
    {
      "entropy": 6.464104509353637,
      "epoch": 2.9278185028626464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004155984219966643,
      "loss": 5.8139,
      "mean_token_accuracy": 0.1730867624282837,
      "num_tokens": 59231862.0,
      "step": 27360
    },
    {
      "entropy": 6.489051675796508,
      "epoch": 2.9283535769704105,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041556857044500796,
      "loss": 5.8667,
      "mean_token_accuracy": 0.1683574825525284,
      "num_tokens": 59242771.0,
      "step": 27365
    },
    {
      "entropy": 6.502635335922241,
      "epoch": 2.928888651078174,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004155387148344538,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16763778030872345,
      "num_tokens": 59253456.0,
      "step": 27370
    },
    {
      "entropy": 6.555328512191773,
      "epoch": 2.9294237251859383,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041550885516586405,
      "loss": 5.8591,
      "mean_token_accuracy": 0.16327004581689836,
      "num_tokens": 59265307.0,
      "step": 27375
    },
    {
      "entropy": 6.486687755584716,
      "epoch": 2.9299587992937024,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041547899144010086,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17462557554244995,
      "num_tokens": 59276920.0,
      "step": 27380
    },
    {
      "entropy": 6.534366655349731,
      "epoch": 2.930493873401466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041544912365802655,
      "loss": 5.8181,
      "mean_token_accuracy": 0.17255858778953553,
      "num_tokens": 59288370.0,
      "step": 27385
    },
    {
      "entropy": 6.409578514099121,
      "epoch": 2.93102894750923,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004154192518205035,
      "loss": 5.7133,
      "mean_token_accuracy": 0.18239875435829161,
      "num_tokens": 59299719.0,
      "step": 27390
    },
    {
      "entropy": 6.525975847244263,
      "epoch": 2.931564021616994,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041538937592839427,
      "loss": 5.829,
      "mean_token_accuracy": 0.17090270221233367,
      "num_tokens": 59310409.0,
      "step": 27395
    },
    {
      "entropy": 6.49151725769043,
      "epoch": 2.932099095724758,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041535949598256157,
      "loss": 5.7574,
      "mean_token_accuracy": 0.17285608500242233,
      "num_tokens": 59321429.0,
      "step": 27400
    },
    {
      "entropy": 6.602522325515747,
      "epoch": 2.9326341698325216,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041532961198386813,
      "loss": 5.8752,
      "mean_token_accuracy": 0.1674613043665886,
      "num_tokens": 59332095.0,
      "step": 27405
    },
    {
      "entropy": 6.503559684753418,
      "epoch": 2.9331692439402857,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004152997239331769,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16554013788700103,
      "num_tokens": 59344002.0,
      "step": 27410
    },
    {
      "entropy": 6.511908626556396,
      "epoch": 2.9337043180480498,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004152698318313509,
      "loss": 5.826,
      "mean_token_accuracy": 0.1733454391360283,
      "num_tokens": 59355148.0,
      "step": 27415
    },
    {
      "entropy": 6.469069480895996,
      "epoch": 2.9342393921558134,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004152399356792531,
      "loss": 5.7732,
      "mean_token_accuracy": 0.17106448709964753,
      "num_tokens": 59365916.0,
      "step": 27420
    },
    {
      "entropy": 6.484485054016114,
      "epoch": 2.9347744662635775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041521003547774694,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16916528046131135,
      "num_tokens": 59376843.0,
      "step": 27425
    },
    {
      "entropy": 6.5467616558074955,
      "epoch": 2.9353095403713416,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004151801312276957,
      "loss": 5.9051,
      "mean_token_accuracy": 0.16567988693714142,
      "num_tokens": 59388572.0,
      "step": 27430
    },
    {
      "entropy": 6.589218664169311,
      "epoch": 2.9358446144791053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004151502229299628,
      "loss": 5.8371,
      "mean_token_accuracy": 0.17156889736652375,
      "num_tokens": 59399122.0,
      "step": 27435
    },
    {
      "entropy": 6.505823755264283,
      "epoch": 2.9363796885868694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000415120310585412,
      "loss": 5.7937,
      "mean_token_accuracy": 0.17163701951503754,
      "num_tokens": 59409565.0,
      "step": 27440
    },
    {
      "entropy": 6.482821083068847,
      "epoch": 2.9369147626946335,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004150903941949068,
      "loss": 5.8108,
      "mean_token_accuracy": 0.17587636262178422,
      "num_tokens": 59420465.0,
      "step": 27445
    },
    {
      "entropy": 6.542292499542237,
      "epoch": 2.937449836802397,
      "grad_norm": 1.125,
      "learning_rate": 0.00041506047375931127,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16811376214027404,
      "num_tokens": 59431457.0,
      "step": 27450
    },
    {
      "entropy": 6.490768098831177,
      "epoch": 2.937984910910161,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004150305492794892,
      "loss": 5.8277,
      "mean_token_accuracy": 0.17425801903009414,
      "num_tokens": 59442257.0,
      "step": 27455
    },
    {
      "entropy": 6.572396516799927,
      "epoch": 2.938519985017925,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004150006207563047,
      "loss": 5.8507,
      "mean_token_accuracy": 0.17113209664821624,
      "num_tokens": 59453634.0,
      "step": 27460
    },
    {
      "entropy": 6.557479333877564,
      "epoch": 2.939055059125689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041497068819062197,
      "loss": 5.9078,
      "mean_token_accuracy": 0.16659074276685715,
      "num_tokens": 59463085.0,
      "step": 27465
    },
    {
      "entropy": 6.5212640285491945,
      "epoch": 2.9395901332334526,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004149407515833052,
      "loss": 5.7902,
      "mean_token_accuracy": 0.1688513532280922,
      "num_tokens": 59473195.0,
      "step": 27470
    },
    {
      "entropy": 6.507944488525391,
      "epoch": 2.9401252073412167,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000414910810935219,
      "loss": 5.8836,
      "mean_token_accuracy": 0.16712315678596495,
      "num_tokens": 59483669.0,
      "step": 27475
    },
    {
      "entropy": 6.558891296386719,
      "epoch": 2.940660281448981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041488086624722777,
      "loss": 5.8167,
      "mean_token_accuracy": 0.1718631848692894,
      "num_tokens": 59495036.0,
      "step": 27480
    },
    {
      "entropy": 6.503674745559692,
      "epoch": 2.9411953555567445,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041485091752019613,
      "loss": 5.7892,
      "mean_token_accuracy": 0.17367982864379883,
      "num_tokens": 59505130.0,
      "step": 27485
    },
    {
      "entropy": 6.418827104568481,
      "epoch": 2.9417304296645086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041482096475498894,
      "loss": 5.7015,
      "mean_token_accuracy": 0.18377583473920822,
      "num_tokens": 59516275.0,
      "step": 27490
    },
    {
      "entropy": 6.589940166473388,
      "epoch": 2.9422655037722727,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041479100795247106,
      "loss": 5.8898,
      "mean_token_accuracy": 0.17452985942363738,
      "num_tokens": 59526978.0,
      "step": 27495
    },
    {
      "entropy": 6.544177436828614,
      "epoch": 2.9428005778800364,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004147610471135074,
      "loss": 5.8577,
      "mean_token_accuracy": 0.1748403027653694,
      "num_tokens": 59536771.0,
      "step": 27500
    },
    {
      "entropy": 6.4838450908660885,
      "epoch": 2.9433356519878004,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00041473108223896314,
      "loss": 5.8108,
      "mean_token_accuracy": 0.16807738989591597,
      "num_tokens": 59548371.0,
      "step": 27505
    },
    {
      "entropy": 6.5316907405853275,
      "epoch": 2.943870726095564,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041470111332970354,
      "loss": 5.8546,
      "mean_token_accuracy": 0.1678283080458641,
      "num_tokens": 59558578.0,
      "step": 27510
    },
    {
      "entropy": 6.4851393699646,
      "epoch": 2.944405800203328,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000414671140386594,
      "loss": 5.8431,
      "mean_token_accuracy": 0.16974422484636306,
      "num_tokens": 59569023.0,
      "step": 27515
    },
    {
      "entropy": 6.518186235427857,
      "epoch": 2.944940874311092,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004146411634104999,
      "loss": 5.8744,
      "mean_token_accuracy": 0.17105051428079604,
      "num_tokens": 59580296.0,
      "step": 27520
    },
    {
      "entropy": 6.498822355270386,
      "epoch": 2.945475948418856,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041461118240228675,
      "loss": 5.7489,
      "mean_token_accuracy": 0.17163839787244797,
      "num_tokens": 59591298.0,
      "step": 27525
    },
    {
      "entropy": 6.581059122085572,
      "epoch": 2.94601102252662,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041458119736282033,
      "loss": 5.8349,
      "mean_token_accuracy": 0.17082142382860183,
      "num_tokens": 59602575.0,
      "step": 27530
    },
    {
      "entropy": 6.560162830352783,
      "epoch": 2.9465460966343837,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041455120829296657,
      "loss": 5.829,
      "mean_token_accuracy": 0.16472222208976744,
      "num_tokens": 59613287.0,
      "step": 27535
    },
    {
      "entropy": 6.540136909484863,
      "epoch": 2.947081170742148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004145212151935911,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16319344192743301,
      "num_tokens": 59623263.0,
      "step": 27540
    },
    {
      "entropy": 6.568040657043457,
      "epoch": 2.947616244849912,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004144912180655603,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16853083819150924,
      "num_tokens": 59634416.0,
      "step": 27545
    },
    {
      "entropy": 6.542325973510742,
      "epoch": 2.9481513189576756,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004144612169097401,
      "loss": 5.848,
      "mean_token_accuracy": 0.16472143530845643,
      "num_tokens": 59643866.0,
      "step": 27550
    },
    {
      "entropy": 6.537867498397827,
      "epoch": 2.9486863930654397,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004144312117269969,
      "loss": 5.8256,
      "mean_token_accuracy": 0.16968811601400374,
      "num_tokens": 59654113.0,
      "step": 27555
    },
    {
      "entropy": 6.500804090499878,
      "epoch": 2.9492214671732033,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041440120251819704,
      "loss": 5.8067,
      "mean_token_accuracy": 0.1681987538933754,
      "num_tokens": 59663837.0,
      "step": 27560
    },
    {
      "entropy": 6.339911794662475,
      "epoch": 2.9497565412809674,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000414371189284207,
      "loss": 5.5993,
      "mean_token_accuracy": 0.18340162485837935,
      "num_tokens": 59673782.0,
      "step": 27565
    },
    {
      "entropy": 6.466681623458863,
      "epoch": 2.950291615388731,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041434117202589346,
      "loss": 5.7863,
      "mean_token_accuracy": 0.17581849694252014,
      "num_tokens": 59684775.0,
      "step": 27570
    },
    {
      "entropy": 6.490397500991821,
      "epoch": 2.950826689496495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004143111507441232,
      "loss": 5.8227,
      "mean_token_accuracy": 0.17142179012298583,
      "num_tokens": 59695193.0,
      "step": 27575
    },
    {
      "entropy": 6.516656494140625,
      "epoch": 2.9513617636042593,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041428112543976303,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1660060316324234,
      "num_tokens": 59705707.0,
      "step": 27580
    },
    {
      "entropy": 6.543127346038818,
      "epoch": 2.951896837712023,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041425109611367987,
      "loss": 5.8974,
      "mean_token_accuracy": 0.16551271677017212,
      "num_tokens": 59715882.0,
      "step": 27585
    },
    {
      "entropy": 6.555527973175049,
      "epoch": 2.952431911819787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004142210627667409,
      "loss": 5.9164,
      "mean_token_accuracy": 0.16115127354860306,
      "num_tokens": 59726690.0,
      "step": 27590
    },
    {
      "entropy": 6.513081789016724,
      "epoch": 2.952966985927551,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041419102539981344,
      "loss": 5.8137,
      "mean_token_accuracy": 0.1684042051434517,
      "num_tokens": 59737052.0,
      "step": 27595
    },
    {
      "entropy": 6.508711814880371,
      "epoch": 2.953502060035315,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004141609840137646,
      "loss": 5.8604,
      "mean_token_accuracy": 0.16855444759130478,
      "num_tokens": 59747989.0,
      "step": 27600
    },
    {
      "entropy": 6.499968576431274,
      "epoch": 2.954037134143079,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004141309386094619,
      "loss": 5.8292,
      "mean_token_accuracy": 0.17261068075895308,
      "num_tokens": 59759169.0,
      "step": 27605
    },
    {
      "entropy": 6.526378583908081,
      "epoch": 2.954572208250843,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004141008891877728,
      "loss": 5.8544,
      "mean_token_accuracy": 0.16888293027877807,
      "num_tokens": 59769837.0,
      "step": 27610
    },
    {
      "entropy": 6.483110952377319,
      "epoch": 2.9551072823586066,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041407083574956515,
      "loss": 5.8015,
      "mean_token_accuracy": 0.1737779602408409,
      "num_tokens": 59780671.0,
      "step": 27615
    },
    {
      "entropy": 6.48542046546936,
      "epoch": 2.9556423564663707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004140407782957067,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17951976209878923,
      "num_tokens": 59792006.0,
      "step": 27620
    },
    {
      "entropy": 6.511893796920776,
      "epoch": 2.9561774305741344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004140107168270653,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16883885115385056,
      "num_tokens": 59802627.0,
      "step": 27625
    },
    {
      "entropy": 6.503259992599487,
      "epoch": 2.9567125046818985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041398065134450897,
      "loss": 5.8511,
      "mean_token_accuracy": 0.17411347478628159,
      "num_tokens": 59813048.0,
      "step": 27630
    },
    {
      "entropy": 6.537734127044677,
      "epoch": 2.957247578789662,
      "grad_norm": 1.125,
      "learning_rate": 0.0004139505818489058,
      "loss": 5.8479,
      "mean_token_accuracy": 0.17791009545326233,
      "num_tokens": 59823730.0,
      "step": 27635
    },
    {
      "entropy": 6.5643877506256105,
      "epoch": 2.9577826528974263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041392050834112415,
      "loss": 5.8879,
      "mean_token_accuracy": 0.17509416341781617,
      "num_tokens": 59835054.0,
      "step": 27640
    },
    {
      "entropy": 6.564706945419312,
      "epoch": 2.9583177270051904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004138904308220324,
      "loss": 5.8919,
      "mean_token_accuracy": 0.16301138699054718,
      "num_tokens": 59845641.0,
      "step": 27645
    },
    {
      "entropy": 6.512484836578369,
      "epoch": 2.958852801112954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041386034929249895,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16328477263450622,
      "num_tokens": 59856255.0,
      "step": 27650
    },
    {
      "entropy": 6.467221307754516,
      "epoch": 2.959387875220718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004138302637533924,
      "loss": 5.7456,
      "mean_token_accuracy": 0.17852499932050706,
      "num_tokens": 59867215.0,
      "step": 27655
    },
    {
      "entropy": 6.540746974945068,
      "epoch": 2.959922949328482,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004138001742055815,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16664886027574538,
      "num_tokens": 59877690.0,
      "step": 27660
    },
    {
      "entropy": 6.541539192199707,
      "epoch": 2.960458023436246,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004137700806499351,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1732930988073349,
      "num_tokens": 59888183.0,
      "step": 27665
    },
    {
      "entropy": 6.47396149635315,
      "epoch": 2.96099309754401,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041373998308732204,
      "loss": 5.8145,
      "mean_token_accuracy": 0.17131813019514083,
      "num_tokens": 59898573.0,
      "step": 27670
    },
    {
      "entropy": 6.487097072601318,
      "epoch": 2.9615281716517736,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041370988151861153,
      "loss": 5.8895,
      "mean_token_accuracy": 0.16919139474630357,
      "num_tokens": 59909605.0,
      "step": 27675
    },
    {
      "entropy": 6.520481157302856,
      "epoch": 2.9620632457595377,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041367977594467263,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16982742846012117,
      "num_tokens": 59920808.0,
      "step": 27680
    },
    {
      "entropy": 6.492253637313842,
      "epoch": 2.9625983198673014,
      "grad_norm": 1.125,
      "learning_rate": 0.00041364966636637477,
      "loss": 5.8698,
      "mean_token_accuracy": 0.16860713064670563,
      "num_tokens": 59931302.0,
      "step": 27685
    },
    {
      "entropy": 6.564207410812378,
      "epoch": 2.9631333939750655,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041361955278458725,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16408952474594116,
      "num_tokens": 59942542.0,
      "step": 27690
    },
    {
      "entropy": 6.545034646987915,
      "epoch": 2.9636684680828296,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004135894352001796,
      "loss": 5.7967,
      "mean_token_accuracy": 0.17194267064332963,
      "num_tokens": 59953547.0,
      "step": 27695
    },
    {
      "entropy": 6.5429542541503904,
      "epoch": 2.9642035421905932,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004135593136140215,
      "loss": 5.8568,
      "mean_token_accuracy": 0.17249720990657808,
      "num_tokens": 59964828.0,
      "step": 27700
    },
    {
      "entropy": 6.480683374404907,
      "epoch": 2.9647386162983573,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041352918802698265,
      "loss": 5.7995,
      "mean_token_accuracy": 0.17939026802778243,
      "num_tokens": 59974989.0,
      "step": 27705
    },
    {
      "entropy": 6.461118888854981,
      "epoch": 2.9652736904061214,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000413499058439933,
      "loss": 5.7654,
      "mean_token_accuracy": 0.1771106779575348,
      "num_tokens": 59985247.0,
      "step": 27710
    },
    {
      "entropy": 6.5196329116821286,
      "epoch": 2.965808764513885,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004134689248537425,
      "loss": 5.8452,
      "mean_token_accuracy": 0.166631019115448,
      "num_tokens": 59995145.0,
      "step": 27715
    },
    {
      "entropy": 6.475072050094605,
      "epoch": 2.966343838621649,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041343878726928125,
      "loss": 5.7571,
      "mean_token_accuracy": 0.182059183716774,
      "num_tokens": 60006400.0,
      "step": 27720
    },
    {
      "entropy": 6.48921275138855,
      "epoch": 2.9668789127294133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004134086456874194,
      "loss": 5.7967,
      "mean_token_accuracy": 0.17679100334644318,
      "num_tokens": 60017408.0,
      "step": 27725
    },
    {
      "entropy": 6.511163711547852,
      "epoch": 2.967413986837177,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041337850010902743,
      "loss": 5.837,
      "mean_token_accuracy": 0.16999782919883727,
      "num_tokens": 60030138.0,
      "step": 27730
    },
    {
      "entropy": 6.46955132484436,
      "epoch": 2.9679490609449406,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041334835053497573,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1673674017190933,
      "num_tokens": 60040966.0,
      "step": 27735
    },
    {
      "entropy": 6.552483034133911,
      "epoch": 2.9684841350527047,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004133181969661349,
      "loss": 5.8567,
      "mean_token_accuracy": 0.17027245908975602,
      "num_tokens": 60053090.0,
      "step": 27740
    },
    {
      "entropy": 6.5984502792358395,
      "epoch": 2.969019209160469,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041328803940337546,
      "loss": 5.8962,
      "mean_token_accuracy": 0.16103436201810836,
      "num_tokens": 60063967.0,
      "step": 27745
    },
    {
      "entropy": 6.532171773910522,
      "epoch": 2.9695542832682325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004132578778475683,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17140624076128005,
      "num_tokens": 60075025.0,
      "step": 27750
    },
    {
      "entropy": 6.556490421295166,
      "epoch": 2.9700893573759966,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004132277122995844,
      "loss": 5.8375,
      "mean_token_accuracy": 0.16603638976812363,
      "num_tokens": 60084530.0,
      "step": 27755
    },
    {
      "entropy": 6.5117823600769045,
      "epoch": 2.9706244314837607,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004131975427602948,
      "loss": 5.7929,
      "mean_token_accuracy": 0.1715244546532631,
      "num_tokens": 60095809.0,
      "step": 27760
    },
    {
      "entropy": 6.467450284957886,
      "epoch": 2.9711595055915243,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004131673692305704,
      "loss": 5.8338,
      "mean_token_accuracy": 0.18093820065259933,
      "num_tokens": 60106559.0,
      "step": 27765
    },
    {
      "entropy": 6.588932847976684,
      "epoch": 2.9716945796992884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004131371917112827,
      "loss": 5.9065,
      "mean_token_accuracy": 0.16006378680467606,
      "num_tokens": 60117557.0,
      "step": 27770
    },
    {
      "entropy": 6.572182941436767,
      "epoch": 2.9722296538070525,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004131070102033031,
      "loss": 5.8985,
      "mean_token_accuracy": 0.16570233702659606,
      "num_tokens": 60129211.0,
      "step": 27775
    },
    {
      "entropy": 6.524394512176514,
      "epoch": 2.972764727914816,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041307682470750293,
      "loss": 5.8059,
      "mean_token_accuracy": 0.1756237640976906,
      "num_tokens": 60139345.0,
      "step": 27780
    },
    {
      "entropy": 6.470071601867676,
      "epoch": 2.9732998020225803,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004130466352247538,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16995588093996047,
      "num_tokens": 60148881.0,
      "step": 27785
    },
    {
      "entropy": 6.448166704177856,
      "epoch": 2.973834876130344,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041301644175592756,
      "loss": 5.7438,
      "mean_token_accuracy": 0.16993479430675507,
      "num_tokens": 60159660.0,
      "step": 27790
    },
    {
      "entropy": 6.486830234527588,
      "epoch": 2.974369950238108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041298624430189583,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17815747559070588,
      "num_tokens": 60170649.0,
      "step": 27795
    },
    {
      "entropy": 6.4488355159759525,
      "epoch": 2.9749050243458717,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004129560428635308,
      "loss": 5.7249,
      "mean_token_accuracy": 0.17495835721492767,
      "num_tokens": 60181419.0,
      "step": 27800
    },
    {
      "entropy": 6.517814826965332,
      "epoch": 2.9754400984536358,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004129258374417044,
      "loss": 5.8783,
      "mean_token_accuracy": 0.16560710370540618,
      "num_tokens": 60192466.0,
      "step": 27805
    },
    {
      "entropy": 6.508512878417969,
      "epoch": 2.9759751725614,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004128956280372888,
      "loss": 5.8123,
      "mean_token_accuracy": 0.1738770142197609,
      "num_tokens": 60203081.0,
      "step": 27810
    },
    {
      "entropy": 6.558508396148682,
      "epoch": 2.9765102466691635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041286541465115635,
      "loss": 5.8437,
      "mean_token_accuracy": 0.16180558353662491,
      "num_tokens": 60214203.0,
      "step": 27815
    },
    {
      "entropy": 6.481377983093262,
      "epoch": 2.9770453207769276,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004128351972841793,
      "loss": 5.8164,
      "mean_token_accuracy": 0.16919700801372528,
      "num_tokens": 60224904.0,
      "step": 27820
    },
    {
      "entropy": 6.528039503097534,
      "epoch": 2.9775803948846917,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041280497593723046,
      "loss": 5.8688,
      "mean_token_accuracy": 0.1641887202858925,
      "num_tokens": 60234962.0,
      "step": 27825
    },
    {
      "entropy": 6.498012065887451,
      "epoch": 2.9781154689924554,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041277475061118223,
      "loss": 5.7978,
      "mean_token_accuracy": 0.17014726400375366,
      "num_tokens": 60245758.0,
      "step": 27830
    },
    {
      "entropy": 6.523113250732422,
      "epoch": 2.9786505431002195,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041274452130690746,
      "loss": 5.848,
      "mean_token_accuracy": 0.16936099529266357,
      "num_tokens": 60256809.0,
      "step": 27835
    },
    {
      "entropy": 6.492101001739502,
      "epoch": 2.9791856172079836,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004127142880252789,
      "loss": 5.8033,
      "mean_token_accuracy": 0.17262800484895707,
      "num_tokens": 60268031.0,
      "step": 27840
    },
    {
      "entropy": 6.529950046539307,
      "epoch": 2.9797206913157472,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004126840507671698,
      "loss": 5.9449,
      "mean_token_accuracy": 0.16435965448617934,
      "num_tokens": 60278526.0,
      "step": 27845
    },
    {
      "entropy": 6.637952423095703,
      "epoch": 2.980255765423511,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041265380953345293,
      "loss": 5.9083,
      "mean_token_accuracy": 0.16399327367544175,
      "num_tokens": 60289412.0,
      "step": 27850
    },
    {
      "entropy": 6.506182241439819,
      "epoch": 2.980790839531275,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041262356432500175,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1693725362420082,
      "num_tokens": 60301242.0,
      "step": 27855
    },
    {
      "entropy": 6.486252927780152,
      "epoch": 2.981325913639039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041259331514268945,
      "loss": 5.8086,
      "mean_token_accuracy": 0.17140536159276962,
      "num_tokens": 60311625.0,
      "step": 27860
    },
    {
      "entropy": 6.53452696800232,
      "epoch": 2.9818609877468027,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041256306198738946,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16743030548095703,
      "num_tokens": 60321973.0,
      "step": 27865
    },
    {
      "entropy": 6.569035243988037,
      "epoch": 2.982396061854567,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041253280485997545,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16618963629007338,
      "num_tokens": 60332445.0,
      "step": 27870
    },
    {
      "entropy": 6.498869466781616,
      "epoch": 2.982931135962331,
      "grad_norm": 1.140625,
      "learning_rate": 0.000412502543761321,
      "loss": 5.7748,
      "mean_token_accuracy": 0.18169355392456055,
      "num_tokens": 60342378.0,
      "step": 27875
    },
    {
      "entropy": 6.513909006118775,
      "epoch": 2.9834662100700946,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004124722786922999,
      "loss": 5.852,
      "mean_token_accuracy": 0.1712527647614479,
      "num_tokens": 60353502.0,
      "step": 27880
    },
    {
      "entropy": 6.481444358825684,
      "epoch": 2.9840012841778587,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004124420096537861,
      "loss": 5.8307,
      "mean_token_accuracy": 0.1721061199903488,
      "num_tokens": 60364243.0,
      "step": 27885
    },
    {
      "entropy": 6.552316808700562,
      "epoch": 2.984536358285623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004124117366466536,
      "loss": 5.8987,
      "mean_token_accuracy": 0.1652245417237282,
      "num_tokens": 60375354.0,
      "step": 27890
    },
    {
      "entropy": 6.482083797454834,
      "epoch": 2.9850714323933865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041238145967177647,
      "loss": 5.7754,
      "mean_token_accuracy": 0.17232565432786942,
      "num_tokens": 60385474.0,
      "step": 27895
    },
    {
      "entropy": 6.490542316436768,
      "epoch": 2.9856065065011506,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004123511787300291,
      "loss": 5.8044,
      "mean_token_accuracy": 0.17047758251428605,
      "num_tokens": 60395339.0,
      "step": 27900
    },
    {
      "entropy": 6.490913772583008,
      "epoch": 2.986141580608914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004123208938222856,
      "loss": 5.8035,
      "mean_token_accuracy": 0.174322646856308,
      "num_tokens": 60406202.0,
      "step": 27905
    },
    {
      "entropy": 6.5281798362731935,
      "epoch": 2.9866766547166783,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041229060494942074,
      "loss": 5.8256,
      "mean_token_accuracy": 0.17446219623088838,
      "num_tokens": 60417251.0,
      "step": 27910
    },
    {
      "entropy": 6.569406938552857,
      "epoch": 2.987211728824442,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041226031211230883,
      "loss": 5.8412,
      "mean_token_accuracy": 0.17285431325435638,
      "num_tokens": 60427299.0,
      "step": 27915
    },
    {
      "entropy": 6.566664123535157,
      "epoch": 2.987746802932206,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041223001531182475,
      "loss": 5.8438,
      "mean_token_accuracy": 0.17277057021856307,
      "num_tokens": 60438076.0,
      "step": 27920
    },
    {
      "entropy": 6.437691783905029,
      "epoch": 2.98828187703997,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041219971454884325,
      "loss": 5.7257,
      "mean_token_accuracy": 0.17967863827943803,
      "num_tokens": 60449479.0,
      "step": 27925
    },
    {
      "entropy": 6.560141086578369,
      "epoch": 2.988816951147734,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004121694098242392,
      "loss": 5.8746,
      "mean_token_accuracy": 0.16653197407722473,
      "num_tokens": 60460664.0,
      "step": 27930
    },
    {
      "entropy": 6.524281454086304,
      "epoch": 2.989352025255498,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004121391011388878,
      "loss": 5.8134,
      "mean_token_accuracy": 0.1687292292714119,
      "num_tokens": 60470673.0,
      "step": 27935
    },
    {
      "entropy": 6.534603786468506,
      "epoch": 2.989887099363262,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041210878849366414,
      "loss": 5.826,
      "mean_token_accuracy": 0.1695399761199951,
      "num_tokens": 60481378.0,
      "step": 27940
    },
    {
      "entropy": 6.475687408447266,
      "epoch": 2.9904221734710257,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004120784718894434,
      "loss": 5.8346,
      "mean_token_accuracy": 0.17037393748760224,
      "num_tokens": 60491264.0,
      "step": 27945
    },
    {
      "entropy": 6.422441148757935,
      "epoch": 2.99095724757879,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004120481513271011,
      "loss": 5.7836,
      "mean_token_accuracy": 0.17424727231264114,
      "num_tokens": 60502053.0,
      "step": 27950
    },
    {
      "entropy": 6.4860834121704105,
      "epoch": 2.9914923216865534,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004120178268075127,
      "loss": 5.8202,
      "mean_token_accuracy": 0.17186488062143326,
      "num_tokens": 60511995.0,
      "step": 27955
    },
    {
      "entropy": 6.541752004623413,
      "epoch": 2.9920273957943175,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004119874983315538,
      "loss": 5.8853,
      "mean_token_accuracy": 0.16697009801864623,
      "num_tokens": 60523127.0,
      "step": 27960
    },
    {
      "entropy": 6.5301250457763675,
      "epoch": 2.992562469902081,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041195716590010015,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16797596365213394,
      "num_tokens": 60533448.0,
      "step": 27965
    },
    {
      "entropy": 6.555624008178711,
      "epoch": 2.9930975440098453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004119268295140275,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16265274137258529,
      "num_tokens": 60544369.0,
      "step": 27970
    },
    {
      "entropy": 6.493061971664429,
      "epoch": 2.9936326181176094,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000411896489174212,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1662489280104637,
      "num_tokens": 60555518.0,
      "step": 27975
    },
    {
      "entropy": 6.429067945480346,
      "epoch": 2.994167692225373,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004118661448815295,
      "loss": 5.7446,
      "mean_token_accuracy": 0.18139430582523347,
      "num_tokens": 60566358.0,
      "step": 27980
    },
    {
      "entropy": 6.509699583053589,
      "epoch": 2.994702766333137,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041183579663685646,
      "loss": 5.8346,
      "mean_token_accuracy": 0.1758900299668312,
      "num_tokens": 60577935.0,
      "step": 27985
    },
    {
      "entropy": 6.537249612808227,
      "epoch": 2.9952378404409012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004118054444410688,
      "loss": 5.7971,
      "mean_token_accuracy": 0.16662731021642685,
      "num_tokens": 60588690.0,
      "step": 27990
    },
    {
      "entropy": 6.472266817092896,
      "epoch": 2.995772914548665,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041177508829504334,
      "loss": 5.7753,
      "mean_token_accuracy": 0.1799527272582054,
      "num_tokens": 60599090.0,
      "step": 27995
    },
    {
      "entropy": 6.514400053024292,
      "epoch": 2.996307988656429,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041174472819965636,
      "loss": 5.8498,
      "mean_token_accuracy": 0.16781714856624602,
      "num_tokens": 60609166.0,
      "step": 28000
    },
    {
      "entropy": 6.499955081939698,
      "epoch": 2.996843062764193,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041171436415578456,
      "loss": 5.8123,
      "mean_token_accuracy": 0.17279104441404342,
      "num_tokens": 60619769.0,
      "step": 28005
    },
    {
      "entropy": 6.482716274261475,
      "epoch": 2.9973781368719568,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004116839961643046,
      "loss": 5.8521,
      "mean_token_accuracy": 0.1638320952653885,
      "num_tokens": 60630984.0,
      "step": 28010
    },
    {
      "entropy": 6.513526105880738,
      "epoch": 2.997913210979721,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041165362422609366,
      "loss": 5.8668,
      "mean_token_accuracy": 0.1654451832175255,
      "num_tokens": 60642019.0,
      "step": 28015
    },
    {
      "entropy": 6.5598376274108885,
      "epoch": 2.9984482850874845,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041162324834202843,
      "loss": 5.8872,
      "mean_token_accuracy": 0.16744959056377412,
      "num_tokens": 60653359.0,
      "step": 28020
    },
    {
      "entropy": 6.459599208831787,
      "epoch": 2.9989833591952486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004115928685129861,
      "loss": 5.6987,
      "mean_token_accuracy": 0.1795207068324089,
      "num_tokens": 60663627.0,
      "step": 28025
    },
    {
      "entropy": 6.490934991836548,
      "epoch": 2.9995184333030123,
      "grad_norm": 1.125,
      "learning_rate": 0.00041156248473984383,
      "loss": 5.8701,
      "mean_token_accuracy": 0.1724000468850136,
      "num_tokens": 60675641.0,
      "step": 28030
    },
    {
      "entropy": 6.466282208760579,
      "epoch": 3.0,
      "grad_norm": 2.484375,
      "learning_rate": 0.0004115320970234791,
      "loss": 5.7919,
      "mean_token_accuracy": 0.1730655100610521,
      "num_tokens": 60684774.0,
      "step": 28035
    },
    {
      "entropy": 6.54905333518982,
      "epoch": 3.000535074107764,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041150170536476924,
      "loss": 5.8135,
      "mean_token_accuracy": 0.17183969914913177,
      "num_tokens": 60696477.0,
      "step": 28040
    },
    {
      "entropy": 6.50560564994812,
      "epoch": 3.0010701482155278,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041147130976459175,
      "loss": 5.691,
      "mean_token_accuracy": 0.17886096388101577,
      "num_tokens": 60709214.0,
      "step": 28045
    },
    {
      "entropy": 6.457072782516479,
      "epoch": 3.001605222323292,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004114409102238245,
      "loss": 5.7263,
      "mean_token_accuracy": 0.17777204513549805,
      "num_tokens": 60719938.0,
      "step": 28050
    },
    {
      "entropy": 6.368877840042114,
      "epoch": 3.0021402964310555,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041141050674334495,
      "loss": 5.6549,
      "mean_token_accuracy": 0.17952644377946853,
      "num_tokens": 60730131.0,
      "step": 28055
    },
    {
      "entropy": 6.507490921020508,
      "epoch": 3.0026753705388196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004113800993240313,
      "loss": 5.7898,
      "mean_token_accuracy": 0.17867080718278885,
      "num_tokens": 60741652.0,
      "step": 28060
    },
    {
      "entropy": 6.43840537071228,
      "epoch": 3.0032104446465837,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041134968796676145,
      "loss": 5.691,
      "mean_token_accuracy": 0.18502683341503143,
      "num_tokens": 60752945.0,
      "step": 28065
    },
    {
      "entropy": 6.521778297424317,
      "epoch": 3.0037455187543474,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041131927267241356,
      "loss": 5.8144,
      "mean_token_accuracy": 0.17109619826078415,
      "num_tokens": 60762695.0,
      "step": 28070
    },
    {
      "entropy": 6.51127405166626,
      "epoch": 3.0042805928621115,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041128885344186573,
      "loss": 5.8005,
      "mean_token_accuracy": 0.1771702840924263,
      "num_tokens": 60773158.0,
      "step": 28075
    },
    {
      "entropy": 6.474840307235718,
      "epoch": 3.004815666969875,
      "grad_norm": 1.125,
      "learning_rate": 0.00041125843027599643,
      "loss": 5.8265,
      "mean_token_accuracy": 0.17235568016767502,
      "num_tokens": 60784171.0,
      "step": 28080
    },
    {
      "entropy": 6.523806810379028,
      "epoch": 3.005350741077639,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041122800317568414,
      "loss": 5.8044,
      "mean_token_accuracy": 0.17086901068687438,
      "num_tokens": 60794961.0,
      "step": 28085
    },
    {
      "entropy": 6.570693635940552,
      "epoch": 3.0058858151854033,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004111975721418074,
      "loss": 5.7452,
      "mean_token_accuracy": 0.17526083439588547,
      "num_tokens": 60805749.0,
      "step": 28090
    },
    {
      "entropy": 6.401798915863037,
      "epoch": 3.006420889293167,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041116713717524484,
      "loss": 5.7167,
      "mean_token_accuracy": 0.17890365421772003,
      "num_tokens": 60815701.0,
      "step": 28095
    },
    {
      "entropy": 6.480127763748169,
      "epoch": 3.006955963400931,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004111366982768754,
      "loss": 5.7946,
      "mean_token_accuracy": 0.17419916540384292,
      "num_tokens": 60826656.0,
      "step": 28100
    },
    {
      "entropy": 6.551935863494873,
      "epoch": 3.007491037508695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041110625544757786,
      "loss": 5.769,
      "mean_token_accuracy": 0.17158049196004868,
      "num_tokens": 60837839.0,
      "step": 28105
    },
    {
      "entropy": 6.539544725418091,
      "epoch": 3.008026111616459,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004110758086882314,
      "loss": 5.8125,
      "mean_token_accuracy": 0.17258890271186828,
      "num_tokens": 60848323.0,
      "step": 28110
    },
    {
      "entropy": 6.4504414081573485,
      "epoch": 3.008561185724223,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000411045357999715,
      "loss": 5.7742,
      "mean_token_accuracy": 0.1767011657357216,
      "num_tokens": 60859049.0,
      "step": 28115
    },
    {
      "entropy": 6.526238298416137,
      "epoch": 3.0090962598319866,
      "grad_norm": 1.125,
      "learning_rate": 0.0004110149033829081,
      "loss": 5.8205,
      "mean_token_accuracy": 0.17570076435804366,
      "num_tokens": 60870451.0,
      "step": 28120
    },
    {
      "entropy": 6.564167308807373,
      "epoch": 3.0096313339397507,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041098444483868995,
      "loss": 5.819,
      "mean_token_accuracy": 0.1778815820813179,
      "num_tokens": 60880530.0,
      "step": 28125
    },
    {
      "entropy": 6.505560827255249,
      "epoch": 3.010166408047515,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041095398236793996,
      "loss": 5.7699,
      "mean_token_accuracy": 0.1712062641978264,
      "num_tokens": 60892170.0,
      "step": 28130
    },
    {
      "entropy": 6.480824613571167,
      "epoch": 3.0107014821552784,
      "grad_norm": 1.1875,
      "learning_rate": 0.000410923515971538,
      "loss": 5.8063,
      "mean_token_accuracy": 0.17329958975315093,
      "num_tokens": 60902880.0,
      "step": 28135
    },
    {
      "entropy": 6.546639728546142,
      "epoch": 3.0112365562630425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004108930456503635,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17464360147714614,
      "num_tokens": 60912860.0,
      "step": 28140
    },
    {
      "entropy": 6.541254901885987,
      "epoch": 3.011771630370806,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041086257140529646,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16789917796850204,
      "num_tokens": 60923303.0,
      "step": 28145
    },
    {
      "entropy": 6.528084182739258,
      "epoch": 3.0123067044785703,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004108320932372167,
      "loss": 5.8197,
      "mean_token_accuracy": 0.1745293691754341,
      "num_tokens": 60935077.0,
      "step": 28150
    },
    {
      "entropy": 6.562550449371338,
      "epoch": 3.0128417785863344,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004108016111470044,
      "loss": 5.8518,
      "mean_token_accuracy": 0.1675170749425888,
      "num_tokens": 60945883.0,
      "step": 28155
    },
    {
      "entropy": 6.516099166870117,
      "epoch": 3.013376852694098,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004107711251355396,
      "loss": 5.7599,
      "mean_token_accuracy": 0.17212636321783065,
      "num_tokens": 60957714.0,
      "step": 28160
    },
    {
      "entropy": 6.478168535232544,
      "epoch": 3.013911926801862,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004107406352037027,
      "loss": 5.7127,
      "mean_token_accuracy": 0.17668879628181458,
      "num_tokens": 60968570.0,
      "step": 28165
    },
    {
      "entropy": 6.469879770278931,
      "epoch": 3.014447000909626,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000410710141352374,
      "loss": 5.7644,
      "mean_token_accuracy": 0.17626895755529404,
      "num_tokens": 60979079.0,
      "step": 28170
    },
    {
      "entropy": 6.519896650314331,
      "epoch": 3.01498207501739,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041067964358243404,
      "loss": 5.7925,
      "mean_token_accuracy": 0.17016825824975967,
      "num_tokens": 60990308.0,
      "step": 28175
    },
    {
      "entropy": 6.468511724472046,
      "epoch": 3.015517149125154,
      "grad_norm": 1.25,
      "learning_rate": 0.0004106491418947635,
      "loss": 5.7019,
      "mean_token_accuracy": 0.1743561863899231,
      "num_tokens": 61000819.0,
      "step": 28180
    },
    {
      "entropy": 6.49314661026001,
      "epoch": 3.0160522232329177,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000410618636290243,
      "loss": 5.8266,
      "mean_token_accuracy": 0.17056282460689545,
      "num_tokens": 61011071.0,
      "step": 28185
    },
    {
      "entropy": 6.5623290061950685,
      "epoch": 3.0165872973406818,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041058812676975337,
      "loss": 5.8154,
      "mean_token_accuracy": 0.17275755554437638,
      "num_tokens": 61022499.0,
      "step": 28190
    },
    {
      "entropy": 6.506626939773559,
      "epoch": 3.0171223714484454,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004105576133341757,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1706804156303406,
      "num_tokens": 61033250.0,
      "step": 28195
    },
    {
      "entropy": 6.420826864242554,
      "epoch": 3.0176574455562095,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041052709598439095,
      "loss": 5.7516,
      "mean_token_accuracy": 0.18080193251371385,
      "num_tokens": 61043222.0,
      "step": 28200
    },
    {
      "entropy": 6.511083269119263,
      "epoch": 3.0181925196639736,
      "grad_norm": 1.125,
      "learning_rate": 0.0004104965747212804,
      "loss": 5.8368,
      "mean_token_accuracy": 0.172726608812809,
      "num_tokens": 61054830.0,
      "step": 28205
    },
    {
      "entropy": 6.533406972885132,
      "epoch": 3.0187275937717373,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004104660495457252,
      "loss": 5.8439,
      "mean_token_accuracy": 0.17413374036550522,
      "num_tokens": 61066385.0,
      "step": 28210
    },
    {
      "entropy": 6.614725589752197,
      "epoch": 3.0192626678795014,
      "grad_norm": 1.15625,
      "learning_rate": 0.000410435520458607,
      "loss": 5.9376,
      "mean_token_accuracy": 0.16615804582834243,
      "num_tokens": 61077434.0,
      "step": 28215
    },
    {
      "entropy": 6.534903717041016,
      "epoch": 3.019797741987265,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041040498746080714,
      "loss": 5.8288,
      "mean_token_accuracy": 0.1665145143866539,
      "num_tokens": 61087661.0,
      "step": 28220
    },
    {
      "entropy": 6.558472776412964,
      "epoch": 3.020332816095029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041037445055320724,
      "loss": 5.8615,
      "mean_token_accuracy": 0.16560629457235337,
      "num_tokens": 61098364.0,
      "step": 28225
    },
    {
      "entropy": 6.479405117034912,
      "epoch": 3.0208678902027932,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041034390973668916,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17079298496246337,
      "num_tokens": 61109061.0,
      "step": 28230
    },
    {
      "entropy": 6.568967580795288,
      "epoch": 3.021402964310557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004103133650121346,
      "loss": 5.7376,
      "mean_token_accuracy": 0.17631596475839614,
      "num_tokens": 61119191.0,
      "step": 28235
    },
    {
      "entropy": 6.499605417251587,
      "epoch": 3.021938038418321,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041028281638042576,
      "loss": 5.778,
      "mean_token_accuracy": 0.17865729779005052,
      "num_tokens": 61130494.0,
      "step": 28240
    },
    {
      "entropy": 6.500240755081177,
      "epoch": 3.022473112526085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004102522638424445,
      "loss": 5.7807,
      "mean_token_accuracy": 0.1704638347029686,
      "num_tokens": 61141660.0,
      "step": 28245
    },
    {
      "entropy": 6.493406915664673,
      "epoch": 3.0230081866338487,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041022170739907324,
      "loss": 5.7419,
      "mean_token_accuracy": 0.1663590505719185,
      "num_tokens": 61152864.0,
      "step": 28250
    },
    {
      "entropy": 6.554328584671021,
      "epoch": 3.023543260741613,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004101911470511941,
      "loss": 5.8192,
      "mean_token_accuracy": 0.16747113019227983,
      "num_tokens": 61163877.0,
      "step": 28255
    },
    {
      "entropy": 6.507947778701782,
      "epoch": 3.0240783348493765,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004101605827996896,
      "loss": 5.7439,
      "mean_token_accuracy": 0.18273259252309798,
      "num_tokens": 61173771.0,
      "step": 28260
    },
    {
      "entropy": 6.450010490417481,
      "epoch": 3.0246134089571406,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004101300146454423,
      "loss": 5.7343,
      "mean_token_accuracy": 0.18483681976795197,
      "num_tokens": 61183652.0,
      "step": 28265
    },
    {
      "entropy": 6.425204038619995,
      "epoch": 3.0251484830649047,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041009944258933473,
      "loss": 5.7496,
      "mean_token_accuracy": 0.1816396325826645,
      "num_tokens": 61195119.0,
      "step": 28270
    },
    {
      "entropy": 6.45196270942688,
      "epoch": 3.0256835571726683,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041006886663224983,
      "loss": 5.796,
      "mean_token_accuracy": 0.1721801921725273,
      "num_tokens": 61206736.0,
      "step": 28275
    },
    {
      "entropy": 6.571006107330322,
      "epoch": 3.0262186312804324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041003828677507037,
      "loss": 5.8631,
      "mean_token_accuracy": 0.16423740535974501,
      "num_tokens": 61217064.0,
      "step": 28280
    },
    {
      "entropy": 6.452839946746826,
      "epoch": 3.026753705388196,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041000770301867934,
      "loss": 5.701,
      "mean_token_accuracy": 0.18210299611091613,
      "num_tokens": 61228520.0,
      "step": 28285
    },
    {
      "entropy": 6.503003358840942,
      "epoch": 3.02728877949596,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004099771153639599,
      "loss": 5.7474,
      "mean_token_accuracy": 0.1738388016819954,
      "num_tokens": 61239835.0,
      "step": 28290
    },
    {
      "entropy": 6.487002325057984,
      "epoch": 3.0278238536037243,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004099465238117951,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17520090639591218,
      "num_tokens": 61250996.0,
      "step": 28295
    },
    {
      "entropy": 6.51951789855957,
      "epoch": 3.028358927711488,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004099159283630685,
      "loss": 5.8447,
      "mean_token_accuracy": 0.17496441751718522,
      "num_tokens": 61261384.0,
      "step": 28300
    },
    {
      "entropy": 6.500483131408691,
      "epoch": 3.028894001819252,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040988532901866336,
      "loss": 5.7804,
      "mean_token_accuracy": 0.1703416272997856,
      "num_tokens": 61271519.0,
      "step": 28305
    },
    {
      "entropy": 6.514999485015869,
      "epoch": 3.0294290759270157,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004098547257794633,
      "loss": 5.7541,
      "mean_token_accuracy": 0.18188953250646592,
      "num_tokens": 61281769.0,
      "step": 28310
    },
    {
      "entropy": 6.539353752136231,
      "epoch": 3.02996415003478,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040982411864635206,
      "loss": 5.8254,
      "mean_token_accuracy": 0.17238755524158478,
      "num_tokens": 61293081.0,
      "step": 28315
    },
    {
      "entropy": 6.442961740493774,
      "epoch": 3.030499224142544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004097935076202133,
      "loss": 5.7067,
      "mean_token_accuracy": 0.17746802270412446,
      "num_tokens": 61303374.0,
      "step": 28320
    },
    {
      "entropy": 6.492518186569214,
      "epoch": 3.0310342982503076,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040976289270193097,
      "loss": 5.8227,
      "mean_token_accuracy": 0.16654905676841736,
      "num_tokens": 61314700.0,
      "step": 28325
    },
    {
      "entropy": 6.510337495803833,
      "epoch": 3.0315693723580717,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040973227389238897,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1773533582687378,
      "num_tokens": 61325463.0,
      "step": 28330
    },
    {
      "entropy": 6.530129337310791,
      "epoch": 3.0321044464658353,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004097016511924716,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17572682499885559,
      "num_tokens": 61337069.0,
      "step": 28335
    },
    {
      "entropy": 6.508240985870361,
      "epoch": 3.0326395205735994,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004096710246030629,
      "loss": 5.8205,
      "mean_token_accuracy": 0.17088890075683594,
      "num_tokens": 61347407.0,
      "step": 28340
    },
    {
      "entropy": 6.452776622772217,
      "epoch": 3.0331745946813635,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040964039412504734,
      "loss": 5.7842,
      "mean_token_accuracy": 0.17320681363344193,
      "num_tokens": 61357950.0,
      "step": 28345
    },
    {
      "entropy": 6.509570646286011,
      "epoch": 3.033709668789127,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040960975975930925,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16780277639627456,
      "num_tokens": 61368394.0,
      "step": 28350
    },
    {
      "entropy": 6.502308559417725,
      "epoch": 3.0342447428968913,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004095791215067333,
      "loss": 5.8282,
      "mean_token_accuracy": 0.16903205662965776,
      "num_tokens": 61379763.0,
      "step": 28355
    },
    {
      "entropy": 6.470074653625488,
      "epoch": 3.034779817004655,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040954847936820423,
      "loss": 5.7435,
      "mean_token_accuracy": 0.1728125438094139,
      "num_tokens": 61390404.0,
      "step": 28360
    },
    {
      "entropy": 6.488880348205567,
      "epoch": 3.035314891112419,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040951783334460657,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17218090146780013,
      "num_tokens": 61401979.0,
      "step": 28365
    },
    {
      "entropy": 6.536883020401001,
      "epoch": 3.035849965220183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004094871834368254,
      "loss": 5.8004,
      "mean_token_accuracy": 0.17075002640485765,
      "num_tokens": 61413306.0,
      "step": 28370
    },
    {
      "entropy": 6.419172191619873,
      "epoch": 3.036385039327947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004094565296457457,
      "loss": 5.6071,
      "mean_token_accuracy": 0.18229532688856126,
      "num_tokens": 61423371.0,
      "step": 28375
    },
    {
      "entropy": 6.530802631378174,
      "epoch": 3.036920113435711,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040942587197225275,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16977821588516234,
      "num_tokens": 61435163.0,
      "step": 28380
    },
    {
      "entropy": 6.498631429672241,
      "epoch": 3.037455187543475,
      "grad_norm": 1.25,
      "learning_rate": 0.0004093952104172315,
      "loss": 5.7746,
      "mean_token_accuracy": 0.17660239785909654,
      "num_tokens": 61445485.0,
      "step": 28385
    },
    {
      "entropy": 6.526083850860596,
      "epoch": 3.0379902616512386,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004093645449815674,
      "loss": 5.7421,
      "mean_token_accuracy": 0.17356383055448532,
      "num_tokens": 61455765.0,
      "step": 28390
    },
    {
      "entropy": 6.464996194839477,
      "epoch": 3.0385253357590027,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040933387566614605,
      "loss": 5.7552,
      "mean_token_accuracy": 0.17575369030237198,
      "num_tokens": 61467337.0,
      "step": 28395
    },
    {
      "entropy": 6.569513607025146,
      "epoch": 3.0390604098667664,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040930320247185284,
      "loss": 5.8477,
      "mean_token_accuracy": 0.16823221147060394,
      "num_tokens": 61478568.0,
      "step": 28400
    },
    {
      "entropy": 6.606893157958984,
      "epoch": 3.0395954839745305,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004092725253995736,
      "loss": 5.8243,
      "mean_token_accuracy": 0.1698472410440445,
      "num_tokens": 61488456.0,
      "step": 28405
    },
    {
      "entropy": 6.528544569015503,
      "epoch": 3.0401305580822946,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040924184445019404,
      "loss": 5.7829,
      "mean_token_accuracy": 0.18164141476154327,
      "num_tokens": 61499996.0,
      "step": 28410
    },
    {
      "entropy": 6.549058675765991,
      "epoch": 3.0406656321900583,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040921115962460003,
      "loss": 5.763,
      "mean_token_accuracy": 0.17309048622846604,
      "num_tokens": 61511271.0,
      "step": 28415
    },
    {
      "entropy": 6.482731819152832,
      "epoch": 3.0412007062978224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004091804709236777,
      "loss": 5.8022,
      "mean_token_accuracy": 0.17249492257833482,
      "num_tokens": 61522284.0,
      "step": 28420
    },
    {
      "entropy": 6.430984020233154,
      "epoch": 3.041735780405586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004091497783483131,
      "loss": 5.6862,
      "mean_token_accuracy": 0.18530830591917039,
      "num_tokens": 61532494.0,
      "step": 28425
    },
    {
      "entropy": 6.486722612380982,
      "epoch": 3.04227085451335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004091190818993925,
      "loss": 5.7743,
      "mean_token_accuracy": 0.1699101835489273,
      "num_tokens": 61543017.0,
      "step": 28430
    },
    {
      "entropy": 6.517398643493652,
      "epoch": 3.042805928621114,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004090883815778023,
      "loss": 5.795,
      "mean_token_accuracy": 0.17542723566293716,
      "num_tokens": 61552984.0,
      "step": 28435
    },
    {
      "entropy": 6.525396490097046,
      "epoch": 3.043341002728878,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004090576773844288,
      "loss": 5.7441,
      "mean_token_accuracy": 0.17682447731494905,
      "num_tokens": 61564045.0,
      "step": 28440
    },
    {
      "entropy": 6.441538333892822,
      "epoch": 3.043876076836642,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040902696932015886,
      "loss": 5.7666,
      "mean_token_accuracy": 0.17685320526361464,
      "num_tokens": 61575369.0,
      "step": 28445
    },
    {
      "entropy": 6.489666700363159,
      "epoch": 3.0444111509444056,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040899625738587895,
      "loss": 5.8387,
      "mean_token_accuracy": 0.17030191347002982,
      "num_tokens": 61586221.0,
      "step": 28450
    },
    {
      "entropy": 6.5479003429412845,
      "epoch": 3.0449462250521697,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040896554158247587,
      "loss": 5.8491,
      "mean_token_accuracy": 0.1675662651658058,
      "num_tokens": 61597091.0,
      "step": 28455
    },
    {
      "entropy": 6.523387908935547,
      "epoch": 3.045481299159934,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040893482191083677,
      "loss": 5.8175,
      "mean_token_accuracy": 0.17323042899370195,
      "num_tokens": 61608026.0,
      "step": 28460
    },
    {
      "entropy": 6.532258605957031,
      "epoch": 3.0460163732676975,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040890409837184835,
      "loss": 5.8306,
      "mean_token_accuracy": 0.16488841325044631,
      "num_tokens": 61620078.0,
      "step": 28465
    },
    {
      "entropy": 6.420398235321045,
      "epoch": 3.0465514473754616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004088733709663979,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1716204583644867,
      "num_tokens": 61630839.0,
      "step": 28470
    },
    {
      "entropy": 6.467930030822754,
      "epoch": 3.0470865214832252,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004088426396953728,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17930959463119506,
      "num_tokens": 61640927.0,
      "step": 28475
    },
    {
      "entropy": 6.486407804489136,
      "epoch": 3.0476215955909893,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004088119045596602,
      "loss": 5.78,
      "mean_token_accuracy": 0.17732743322849273,
      "num_tokens": 61650631.0,
      "step": 28480
    },
    {
      "entropy": 6.527324104309082,
      "epoch": 3.0481566696987534,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004087811655601477,
      "loss": 5.836,
      "mean_token_accuracy": 0.17298216819763185,
      "num_tokens": 61661493.0,
      "step": 28485
    },
    {
      "entropy": 6.538991212844849,
      "epoch": 3.048691743806517,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040875042269772284,
      "loss": 5.8665,
      "mean_token_accuracy": 0.1693005919456482,
      "num_tokens": 61673116.0,
      "step": 28490
    },
    {
      "entropy": 6.483036756515503,
      "epoch": 3.049226817914281,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040871967597327333,
      "loss": 5.7237,
      "mean_token_accuracy": 0.17751851826906204,
      "num_tokens": 61683471.0,
      "step": 28495
    },
    {
      "entropy": 6.501787233352661,
      "epoch": 3.049761892022045,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040868892538768693,
      "loss": 5.8683,
      "mean_token_accuracy": 0.17196546047925948,
      "num_tokens": 61693264.0,
      "step": 28500
    },
    {
      "entropy": 6.470983409881592,
      "epoch": 3.050296966129809,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004086581709418516,
      "loss": 5.7585,
      "mean_token_accuracy": 0.17828162759542465,
      "num_tokens": 61703566.0,
      "step": 28505
    },
    {
      "entropy": 6.483247852325439,
      "epoch": 3.050832040237573,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040862741263665537,
      "loss": 5.7452,
      "mean_token_accuracy": 0.1793535217642784,
      "num_tokens": 61714981.0,
      "step": 28510
    },
    {
      "entropy": 6.441078329086304,
      "epoch": 3.0513671143453367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040859665047298635,
      "loss": 5.7558,
      "mean_token_accuracy": 0.18179795742034913,
      "num_tokens": 61726036.0,
      "step": 28515
    },
    {
      "entropy": 6.5135047912597654,
      "epoch": 3.051902188453101,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004085658844517329,
      "loss": 5.8585,
      "mean_token_accuracy": 0.169424906373024,
      "num_tokens": 61736614.0,
      "step": 28520
    },
    {
      "entropy": 6.456226539611817,
      "epoch": 3.052437262560865,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004085351145737832,
      "loss": 5.6852,
      "mean_token_accuracy": 0.18202015161514282,
      "num_tokens": 61746817.0,
      "step": 28525
    },
    {
      "entropy": 6.429451560974121,
      "epoch": 3.0529723366686286,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004085043408400259,
      "loss": 5.6405,
      "mean_token_accuracy": 0.1825092241168022,
      "num_tokens": 61757714.0,
      "step": 28530
    },
    {
      "entropy": 6.488070249557495,
      "epoch": 3.0535074107763926,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004084735632513495,
      "loss": 5.7578,
      "mean_token_accuracy": 0.1790920838713646,
      "num_tokens": 61768719.0,
      "step": 28535
    },
    {
      "entropy": 6.489954662322998,
      "epoch": 3.0540424848841563,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040844278180864274,
      "loss": 5.793,
      "mean_token_accuracy": 0.17138969153165817,
      "num_tokens": 61778942.0,
      "step": 28540
    },
    {
      "entropy": 6.444582748413086,
      "epoch": 3.0545775589919204,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004084119965127944,
      "loss": 5.7286,
      "mean_token_accuracy": 0.1820797324180603,
      "num_tokens": 61789427.0,
      "step": 28545
    },
    {
      "entropy": 6.49300446510315,
      "epoch": 3.0551126330996845,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004083812073646933,
      "loss": 5.7374,
      "mean_token_accuracy": 0.1742815300822258,
      "num_tokens": 61799702.0,
      "step": 28550
    },
    {
      "entropy": 6.4444831848144535,
      "epoch": 3.055647707207448,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040835041436522874,
      "loss": 5.7185,
      "mean_token_accuracy": 0.1862908586859703,
      "num_tokens": 61809346.0,
      "step": 28555
    },
    {
      "entropy": 6.519741535186768,
      "epoch": 3.0561827813152123,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040831961751528964,
      "loss": 5.7877,
      "mean_token_accuracy": 0.17493139654397966,
      "num_tokens": 61819762.0,
      "step": 28560
    },
    {
      "entropy": 6.494769954681397,
      "epoch": 3.056717855422976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040828881681576534,
      "loss": 5.8002,
      "mean_token_accuracy": 0.16678137332201004,
      "num_tokens": 61830624.0,
      "step": 28565
    },
    {
      "entropy": 6.554408407211303,
      "epoch": 3.05725292953074,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004082580122675451,
      "loss": 5.7893,
      "mean_token_accuracy": 0.17783936560153962,
      "num_tokens": 61841684.0,
      "step": 28570
    },
    {
      "entropy": 6.424530839920044,
      "epoch": 3.057788003638504,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040822720387151856,
      "loss": 5.7463,
      "mean_token_accuracy": 0.1805101990699768,
      "num_tokens": 61852239.0,
      "step": 28575
    },
    {
      "entropy": 6.5100852966308596,
      "epoch": 3.0583230777462678,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004081963916285752,
      "loss": 5.8478,
      "mean_token_accuracy": 0.16472634077072143,
      "num_tokens": 61863014.0,
      "step": 28580
    },
    {
      "entropy": 6.485585880279541,
      "epoch": 3.058858151854032,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040816557553960474,
      "loss": 5.6989,
      "mean_token_accuracy": 0.1771360456943512,
      "num_tokens": 61873819.0,
      "step": 28585
    },
    {
      "entropy": 6.512574625015259,
      "epoch": 3.0593932259617955,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040813475560549705,
      "loss": 5.7536,
      "mean_token_accuracy": 0.1792510434985161,
      "num_tokens": 61885866.0,
      "step": 28590
    },
    {
      "entropy": 6.460404872894287,
      "epoch": 3.0599283000695596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000408103931827142,
      "loss": 5.7177,
      "mean_token_accuracy": 0.18226378560066223,
      "num_tokens": 61895836.0,
      "step": 28595
    },
    {
      "entropy": 6.370234251022339,
      "epoch": 3.0604633741773237,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004080731042054296,
      "loss": 5.6014,
      "mean_token_accuracy": 0.19397284984588622,
      "num_tokens": 61906284.0,
      "step": 28600
    },
    {
      "entropy": 6.4438292503356935,
      "epoch": 3.0609984482850874,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040804227274125,
      "loss": 5.7753,
      "mean_token_accuracy": 0.17441086173057557,
      "num_tokens": 61917024.0,
      "step": 28605
    },
    {
      "entropy": 6.512355375289917,
      "epoch": 3.0615335223928515,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00040801143743549343,
      "loss": 5.7416,
      "mean_token_accuracy": 0.17496391534805297,
      "num_tokens": 61927631.0,
      "step": 28610
    },
    {
      "entropy": 6.523974943161011,
      "epoch": 3.062068596500615,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040798059828905036,
      "loss": 5.7365,
      "mean_token_accuracy": 0.1791718617081642,
      "num_tokens": 61938443.0,
      "step": 28615
    },
    {
      "entropy": 6.474092721939087,
      "epoch": 3.0626036706083792,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004079497553028112,
      "loss": 5.7273,
      "mean_token_accuracy": 0.17499710321426393,
      "num_tokens": 61949148.0,
      "step": 28620
    },
    {
      "entropy": 6.488492345809936,
      "epoch": 3.0631387447161433,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004079189084776665,
      "loss": 5.7809,
      "mean_token_accuracy": 0.184085014462471,
      "num_tokens": 61959052.0,
      "step": 28625
    },
    {
      "entropy": 6.485363101959228,
      "epoch": 3.063673818823907,
      "grad_norm": 1.15625,
      "learning_rate": 0.000407888057814507,
      "loss": 5.7433,
      "mean_token_accuracy": 0.17763131111860275,
      "num_tokens": 61969906.0,
      "step": 28630
    },
    {
      "entropy": 6.486834526062012,
      "epoch": 3.064208892931671,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040785720331422355,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1666905999183655,
      "num_tokens": 61981511.0,
      "step": 28635
    },
    {
      "entropy": 6.520400285720825,
      "epoch": 3.064743967039435,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000407826344977707,
      "loss": 5.7923,
      "mean_token_accuracy": 0.1768509790301323,
      "num_tokens": 61993404.0,
      "step": 28640
    },
    {
      "entropy": 6.5337120532989506,
      "epoch": 3.065279041147199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004077954828058484,
      "loss": 5.8213,
      "mean_token_accuracy": 0.17611646503210068,
      "num_tokens": 62004216.0,
      "step": 28645
    },
    {
      "entropy": 6.516099071502685,
      "epoch": 3.065814115254963,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004077646167995389,
      "loss": 5.8334,
      "mean_token_accuracy": 0.16341103464365006,
      "num_tokens": 62014841.0,
      "step": 28650
    },
    {
      "entropy": 6.497279024124145,
      "epoch": 3.0663491893627266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040773374695966976,
      "loss": 5.7813,
      "mean_token_accuracy": 0.17568275034427644,
      "num_tokens": 62025024.0,
      "step": 28655
    },
    {
      "entropy": 6.449977540969849,
      "epoch": 3.0668842634704907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004077028732871323,
      "loss": 5.7616,
      "mean_token_accuracy": 0.17661577761173247,
      "num_tokens": 62037892.0,
      "step": 28660
    },
    {
      "entropy": 6.554068946838379,
      "epoch": 3.067419337578255,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040767199578281793,
      "loss": 5.7855,
      "mean_token_accuracy": 0.1718478098511696,
      "num_tokens": 62048226.0,
      "step": 28665
    },
    {
      "entropy": 6.610101985931396,
      "epoch": 3.0679544116860185,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040764111444761847,
      "loss": 5.8287,
      "mean_token_accuracy": 0.17380648851394653,
      "num_tokens": 62059339.0,
      "step": 28670
    },
    {
      "entropy": 6.524333667755127,
      "epoch": 3.0684894857937826,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040761022928242544,
      "loss": 5.775,
      "mean_token_accuracy": 0.17117052227258683,
      "num_tokens": 62069008.0,
      "step": 28675
    },
    {
      "entropy": 6.445465183258056,
      "epoch": 3.069024559901546,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040757934028813066,
      "loss": 5.7683,
      "mean_token_accuracy": 0.1772102490067482,
      "num_tokens": 62079405.0,
      "step": 28680
    },
    {
      "entropy": 6.485877418518067,
      "epoch": 3.0695596340093103,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000407548447465626,
      "loss": 5.6908,
      "mean_token_accuracy": 0.17861379384994508,
      "num_tokens": 62089928.0,
      "step": 28685
    },
    {
      "entropy": 6.482310390472412,
      "epoch": 3.0700947081170744,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004075175508158036,
      "loss": 5.7632,
      "mean_token_accuracy": 0.17254854440689088,
      "num_tokens": 62100956.0,
      "step": 28690
    },
    {
      "entropy": 6.521884489059448,
      "epoch": 3.070629782224838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040748665033955557,
      "loss": 5.7545,
      "mean_token_accuracy": 0.18128394931554795,
      "num_tokens": 62111673.0,
      "step": 28695
    },
    {
      "entropy": 6.535765600204468,
      "epoch": 3.071164856332602,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004074557460377741,
      "loss": 5.8173,
      "mean_token_accuracy": 0.1726458802819252,
      "num_tokens": 62122273.0,
      "step": 28700
    },
    {
      "entropy": 6.508139371871948,
      "epoch": 3.071699930440366,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004074248379113516,
      "loss": 5.7736,
      "mean_token_accuracy": 0.17247259616851807,
      "num_tokens": 62132836.0,
      "step": 28705
    },
    {
      "entropy": 6.564722967147827,
      "epoch": 3.07223500454813,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004073939259611805,
      "loss": 5.808,
      "mean_token_accuracy": 0.16787478029727937,
      "num_tokens": 62144228.0,
      "step": 28710
    },
    {
      "entropy": 6.469367027282715,
      "epoch": 3.072770078655894,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004073630101881534,
      "loss": 5.7482,
      "mean_token_accuracy": 0.18245832324028016,
      "num_tokens": 62155126.0,
      "step": 28715
    },
    {
      "entropy": 6.48128023147583,
      "epoch": 3.0733051527636577,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004073320905931629,
      "loss": 5.8071,
      "mean_token_accuracy": 0.17016823291778566,
      "num_tokens": 62165973.0,
      "step": 28720
    },
    {
      "entropy": 6.489196300506592,
      "epoch": 3.0738402268714218,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004073011671771019,
      "loss": 5.7894,
      "mean_token_accuracy": 0.1750078782439232,
      "num_tokens": 62176258.0,
      "step": 28725
    },
    {
      "entropy": 6.508648538589478,
      "epoch": 3.0743753009791854,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040727023994086345,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1730268493294716,
      "num_tokens": 62186078.0,
      "step": 28730
    },
    {
      "entropy": 6.5286116123199465,
      "epoch": 3.0749103750869495,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004072393088853402,
      "loss": 5.7999,
      "mean_token_accuracy": 0.17359113842248916,
      "num_tokens": 62197951.0,
      "step": 28735
    },
    {
      "entropy": 6.542830133438111,
      "epoch": 3.0754454491947136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040720837401142556,
      "loss": 5.7624,
      "mean_token_accuracy": 0.1733967989683151,
      "num_tokens": 62210187.0,
      "step": 28740
    },
    {
      "entropy": 6.552098226547241,
      "epoch": 3.0759805233024773,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004071774353200128,
      "loss": 5.8239,
      "mean_token_accuracy": 0.17345292568206788,
      "num_tokens": 62220561.0,
      "step": 28745
    },
    {
      "entropy": 6.530724430084229,
      "epoch": 3.0765155974102414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000407146492811995,
      "loss": 5.8165,
      "mean_token_accuracy": 0.16903402656316757,
      "num_tokens": 62232378.0,
      "step": 28750
    },
    {
      "entropy": 6.484766674041748,
      "epoch": 3.077050671518005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004071155464882659,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17828771471977234,
      "num_tokens": 62244823.0,
      "step": 28755
    },
    {
      "entropy": 6.4885039806365965,
      "epoch": 3.077585745625769,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004070845963497189,
      "loss": 5.7419,
      "mean_token_accuracy": 0.1748846009373665,
      "num_tokens": 62255070.0,
      "step": 28760
    },
    {
      "entropy": 6.540177679061889,
      "epoch": 3.0781208197335332,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004070536423972478,
      "loss": 5.854,
      "mean_token_accuracy": 0.1680970773100853,
      "num_tokens": 62267081.0,
      "step": 28765
    },
    {
      "entropy": 6.556751537322998,
      "epoch": 3.078655893841297,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004070226846317463,
      "loss": 5.8607,
      "mean_token_accuracy": 0.17275039702653885,
      "num_tokens": 62277512.0,
      "step": 28770
    },
    {
      "entropy": 6.5385887145996096,
      "epoch": 3.079190967949061,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040699172305410833,
      "loss": 5.7626,
      "mean_token_accuracy": 0.1772739589214325,
      "num_tokens": 62287827.0,
      "step": 28775
    },
    {
      "entropy": 6.501900911331177,
      "epoch": 3.0797260420568247,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040696075766522797,
      "loss": 5.8097,
      "mean_token_accuracy": 0.16723703891038894,
      "num_tokens": 62299977.0,
      "step": 28780
    },
    {
      "entropy": 6.526979589462281,
      "epoch": 3.0802611161645888,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004069297884659992,
      "loss": 5.7511,
      "mean_token_accuracy": 0.1774143636226654,
      "num_tokens": 62310157.0,
      "step": 28785
    },
    {
      "entropy": 6.4803849220275875,
      "epoch": 3.080796190272353,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004068988154573163,
      "loss": 5.7056,
      "mean_token_accuracy": 0.18509093970060347,
      "num_tokens": 62320420.0,
      "step": 28790
    },
    {
      "entropy": 6.482977867126465,
      "epoch": 3.0813312643801165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040686783864007367,
      "loss": 5.8213,
      "mean_token_accuracy": 0.17097791731357576,
      "num_tokens": 62330551.0,
      "step": 28795
    },
    {
      "entropy": 6.565534496307373,
      "epoch": 3.0818663384878806,
      "grad_norm": 1.125,
      "learning_rate": 0.0004068368580151658,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16953985691070556,
      "num_tokens": 62341131.0,
      "step": 28800
    },
    {
      "entropy": 6.55127329826355,
      "epoch": 3.0824014125956447,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004068058735834871,
      "loss": 5.8686,
      "mean_token_accuracy": 0.17406694144010543,
      "num_tokens": 62352432.0,
      "step": 28805
    },
    {
      "entropy": 6.555922555923462,
      "epoch": 3.0829364867034084,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004067748853459323,
      "loss": 5.88,
      "mean_token_accuracy": 0.16459882110357285,
      "num_tokens": 62364029.0,
      "step": 28810
    },
    {
      "entropy": 6.5726258754730225,
      "epoch": 3.0834715608111725,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004067438933033963,
      "loss": 5.7764,
      "mean_token_accuracy": 0.1725756824016571,
      "num_tokens": 62373804.0,
      "step": 28815
    },
    {
      "entropy": 6.450080823898316,
      "epoch": 3.084006634918936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040671289745677387,
      "loss": 5.6653,
      "mean_token_accuracy": 0.19057219624519348,
      "num_tokens": 62384400.0,
      "step": 28820
    },
    {
      "entropy": 6.538520050048828,
      "epoch": 3.0845417090267,
      "grad_norm": 1.25,
      "learning_rate": 0.00040668189780695995,
      "loss": 5.878,
      "mean_token_accuracy": 0.16607534885406494,
      "num_tokens": 62395228.0,
      "step": 28825
    },
    {
      "entropy": 6.514668989181518,
      "epoch": 3.0850767831344643,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004066508943548498,
      "loss": 5.798,
      "mean_token_accuracy": 0.17469402253627778,
      "num_tokens": 62405458.0,
      "step": 28830
    },
    {
      "entropy": 6.578116035461425,
      "epoch": 3.085611857242228,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040661988710133856,
      "loss": 5.8526,
      "mean_token_accuracy": 0.17396719753742218,
      "num_tokens": 62416838.0,
      "step": 28835
    },
    {
      "entropy": 6.53657283782959,
      "epoch": 3.086146931349992,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040658887604732154,
      "loss": 5.7122,
      "mean_token_accuracy": 0.17210971266031266,
      "num_tokens": 62427680.0,
      "step": 28840
    },
    {
      "entropy": 6.479504823684692,
      "epoch": 3.0866820054577557,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004065578611936943,
      "loss": 5.7463,
      "mean_token_accuracy": 0.17506371885538102,
      "num_tokens": 62438785.0,
      "step": 28845
    },
    {
      "entropy": 6.502626752853393,
      "epoch": 3.08721707956552,
      "grad_norm": 1.125,
      "learning_rate": 0.00040652684254135227,
      "loss": 5.8331,
      "mean_token_accuracy": 0.17184577584266664,
      "num_tokens": 62450362.0,
      "step": 28850
    },
    {
      "entropy": 6.568906927108765,
      "epoch": 3.087752153673284,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004064958200911911,
      "loss": 5.812,
      "mean_token_accuracy": 0.1737629786133766,
      "num_tokens": 62461962.0,
      "step": 28855
    },
    {
      "entropy": 6.480419063568116,
      "epoch": 3.0882872277810476,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004064647938441066,
      "loss": 5.7162,
      "mean_token_accuracy": 0.18762401789426802,
      "num_tokens": 62473290.0,
      "step": 28860
    },
    {
      "entropy": 6.581043338775634,
      "epoch": 3.0888223018888117,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040643376380099466,
      "loss": 5.9955,
      "mean_token_accuracy": 0.15859522223472594,
      "num_tokens": 62484219.0,
      "step": 28865
    },
    {
      "entropy": 6.519306278228759,
      "epoch": 3.0893573759965753,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004064027299627514,
      "loss": 5.7735,
      "mean_token_accuracy": 0.17654892206192016,
      "num_tokens": 62494542.0,
      "step": 28870
    },
    {
      "entropy": 6.554988861083984,
      "epoch": 3.0898924501043394,
      "grad_norm": 1.125,
      "learning_rate": 0.00040637169233027264,
      "loss": 5.7608,
      "mean_token_accuracy": 0.17720619440078736,
      "num_tokens": 62505351.0,
      "step": 28875
    },
    {
      "entropy": 6.469031429290771,
      "epoch": 3.0904275242121035,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040634065090445473,
      "loss": 5.7781,
      "mean_token_accuracy": 0.18627230823040009,
      "num_tokens": 62515848.0,
      "step": 28880
    },
    {
      "entropy": 6.552205228805542,
      "epoch": 3.090962598319867,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000406309605686194,
      "loss": 5.899,
      "mean_token_accuracy": 0.15730353742837905,
      "num_tokens": 62527042.0,
      "step": 28885
    },
    {
      "entropy": 6.539748859405518,
      "epoch": 3.0914976724276313,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004062785566763869,
      "loss": 5.8345,
      "mean_token_accuracy": 0.1699354663491249,
      "num_tokens": 62537830.0,
      "step": 28890
    },
    {
      "entropy": 6.595416069030762,
      "epoch": 3.092032746535395,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004062475038759298,
      "loss": 5.8609,
      "mean_token_accuracy": 0.1614289551973343,
      "num_tokens": 62548391.0,
      "step": 28895
    },
    {
      "entropy": 6.5523786544799805,
      "epoch": 3.092567820643159,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004062164472857196,
      "loss": 5.7339,
      "mean_token_accuracy": 0.17267246097326278,
      "num_tokens": 62557832.0,
      "step": 28900
    },
    {
      "entropy": 6.499335050582886,
      "epoch": 3.093102894750923,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040618538690665286,
      "loss": 5.848,
      "mean_token_accuracy": 0.17211569845676422,
      "num_tokens": 62568451.0,
      "step": 28905
    },
    {
      "entropy": 6.423787450790405,
      "epoch": 3.093637968858687,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040615432273962657,
      "loss": 5.6627,
      "mean_token_accuracy": 0.18878394216299058,
      "num_tokens": 62579230.0,
      "step": 28910
    },
    {
      "entropy": 6.489891576766968,
      "epoch": 3.094173042966451,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004061232547855376,
      "loss": 5.7922,
      "mean_token_accuracy": 0.17264098823070526,
      "num_tokens": 62589853.0,
      "step": 28915
    },
    {
      "entropy": 6.5673075199127195,
      "epoch": 3.094708117074215,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004060921830452831,
      "loss": 5.8106,
      "mean_token_accuracy": 0.17874808460474015,
      "num_tokens": 62600929.0,
      "step": 28920
    },
    {
      "entropy": 6.457261276245117,
      "epoch": 3.0952431911819787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004060611075197601,
      "loss": 5.7128,
      "mean_token_accuracy": 0.17777159214019775,
      "num_tokens": 62611583.0,
      "step": 28925
    },
    {
      "entropy": 6.57890305519104,
      "epoch": 3.0957782652897428,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004060300282098661,
      "loss": 5.8509,
      "mean_token_accuracy": 0.17113660275936127,
      "num_tokens": 62622609.0,
      "step": 28930
    },
    {
      "entropy": 6.5769775390625,
      "epoch": 3.0963133393975064,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040599894511649854,
      "loss": 5.8848,
      "mean_token_accuracy": 0.16121521294116975,
      "num_tokens": 62633859.0,
      "step": 28935
    },
    {
      "entropy": 6.558881092071533,
      "epoch": 3.0968484135052705,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004059678582405548,
      "loss": 5.7642,
      "mean_token_accuracy": 0.17512628585100173,
      "num_tokens": 62644409.0,
      "step": 28940
    },
    {
      "entropy": 6.578620862960816,
      "epoch": 3.0973834876130346,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004059367675829325,
      "loss": 5.8138,
      "mean_token_accuracy": 0.17860572636127472,
      "num_tokens": 62653703.0,
      "step": 28945
    },
    {
      "entropy": 6.45905442237854,
      "epoch": 3.0979185617207983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040590567314452945,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17062199711799622,
      "num_tokens": 62664413.0,
      "step": 28950
    },
    {
      "entropy": 6.411220598220825,
      "epoch": 3.0984536358285624,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040587457492624357,
      "loss": 5.7056,
      "mean_token_accuracy": 0.18048202991485596,
      "num_tokens": 62674520.0,
      "step": 28955
    },
    {
      "entropy": 6.493005418777466,
      "epoch": 3.098988709936326,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040584347292897265,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16986860930919648,
      "num_tokens": 62685584.0,
      "step": 28960
    },
    {
      "entropy": 6.514594888687133,
      "epoch": 3.09952378404409,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004058123671536149,
      "loss": 5.711,
      "mean_token_accuracy": 0.17865030169487,
      "num_tokens": 62695999.0,
      "step": 28965
    },
    {
      "entropy": 6.542278623580932,
      "epoch": 3.1000588581518542,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004057812576010684,
      "loss": 5.8328,
      "mean_token_accuracy": 0.17003643959760667,
      "num_tokens": 62706486.0,
      "step": 28970
    },
    {
      "entropy": 6.527051591873169,
      "epoch": 3.100593932259618,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040575014427223145,
      "loss": 5.7104,
      "mean_token_accuracy": 0.1821370393037796,
      "num_tokens": 62716667.0,
      "step": 28975
    },
    {
      "entropy": 6.544526195526123,
      "epoch": 3.101129006367382,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040571902716800246,
      "loss": 5.8053,
      "mean_token_accuracy": 0.17154130041599275,
      "num_tokens": 62728101.0,
      "step": 28980
    },
    {
      "entropy": 6.522299098968506,
      "epoch": 3.1016640804751456,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040568790628927996,
      "loss": 5.7981,
      "mean_token_accuracy": 0.1708145558834076,
      "num_tokens": 62739796.0,
      "step": 28985
    },
    {
      "entropy": 6.516091203689575,
      "epoch": 3.1021991545829097,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040565678163696246,
      "loss": 5.7659,
      "mean_token_accuracy": 0.17603325396776198,
      "num_tokens": 62751270.0,
      "step": 28990
    },
    {
      "entropy": 6.56228494644165,
      "epoch": 3.102734228690674,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004056256532119488,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1699285924434662,
      "num_tokens": 62761890.0,
      "step": 28995
    },
    {
      "entropy": 6.572594928741455,
      "epoch": 3.1032693027984375,
      "grad_norm": 1.125,
      "learning_rate": 0.0004055945210151378,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17630544006824495,
      "num_tokens": 62773177.0,
      "step": 29000
    },
    {
      "entropy": 6.561953830718994,
      "epoch": 3.1038043769062016,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004055633850474283,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16861829161643982,
      "num_tokens": 62784511.0,
      "step": 29005
    },
    {
      "entropy": 6.527856969833374,
      "epoch": 3.1043394510139652,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004055322453097194,
      "loss": 5.8671,
      "mean_token_accuracy": 0.16834179311990738,
      "num_tokens": 62795618.0,
      "step": 29010
    },
    {
      "entropy": 6.596693134307861,
      "epoch": 3.1048745251217293,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040550110180291024,
      "loss": 5.8707,
      "mean_token_accuracy": 0.16500944495201111,
      "num_tokens": 62805936.0,
      "step": 29015
    },
    {
      "entropy": 6.593443298339844,
      "epoch": 3.1054095992294934,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004054699545279001,
      "loss": 5.8456,
      "mean_token_accuracy": 0.17067540287971497,
      "num_tokens": 62815511.0,
      "step": 29020
    },
    {
      "entropy": 6.557674837112427,
      "epoch": 3.105944673337257,
      "grad_norm": 1.125,
      "learning_rate": 0.0004054388034855884,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16926479786634446,
      "num_tokens": 62826382.0,
      "step": 29025
    },
    {
      "entropy": 6.561861658096314,
      "epoch": 3.106479747445021,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040540764867687463,
      "loss": 5.9123,
      "mean_token_accuracy": 0.16692616194486617,
      "num_tokens": 62836546.0,
      "step": 29030
    },
    {
      "entropy": 6.606876182556152,
      "epoch": 3.107014821552785,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040537649010265826,
      "loss": 5.8223,
      "mean_token_accuracy": 0.16721284240484238,
      "num_tokens": 62847407.0,
      "step": 29035
    },
    {
      "entropy": 6.621703147888184,
      "epoch": 3.107549895660549,
      "grad_norm": 1.328125,
      "learning_rate": 0.000405345327763839,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16865794509649276,
      "num_tokens": 62857139.0,
      "step": 29040
    },
    {
      "entropy": 6.493870782852173,
      "epoch": 3.108084969768313,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040531416166131686,
      "loss": 5.7302,
      "mean_token_accuracy": 0.17665206640958786,
      "num_tokens": 62867674.0,
      "step": 29045
    },
    {
      "entropy": 6.487670373916626,
      "epoch": 3.1086200438760767,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040528299179599155,
      "loss": 5.7633,
      "mean_token_accuracy": 0.17245515286922455,
      "num_tokens": 62878204.0,
      "step": 29050
    },
    {
      "entropy": 6.507610273361206,
      "epoch": 3.109155117983841,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004052518181687631,
      "loss": 5.7838,
      "mean_token_accuracy": 0.17002652138471602,
      "num_tokens": 62888737.0,
      "step": 29055
    },
    {
      "entropy": 6.454417037963867,
      "epoch": 3.1096901920916045,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040522064078053174,
      "loss": 5.7409,
      "mean_token_accuracy": 0.18540246933698654,
      "num_tokens": 62899481.0,
      "step": 29060
    },
    {
      "entropy": 6.591441345214844,
      "epoch": 3.1102252661993686,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040518945963219767,
      "loss": 5.8475,
      "mean_token_accuracy": 0.16632325798273087,
      "num_tokens": 62910298.0,
      "step": 29065
    },
    {
      "entropy": 6.529539108276367,
      "epoch": 3.1107603403071327,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040515827472466134,
      "loss": 5.6757,
      "mean_token_accuracy": 0.18249710649251938,
      "num_tokens": 62921132.0,
      "step": 29070
    },
    {
      "entropy": 6.487858533859253,
      "epoch": 3.1112954144148963,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004051270860588229,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1764824390411377,
      "num_tokens": 62932107.0,
      "step": 29075
    },
    {
      "entropy": 6.502076053619385,
      "epoch": 3.1118304885226604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004050958936355834,
      "loss": 5.7276,
      "mean_token_accuracy": 0.1712874189019203,
      "num_tokens": 62943177.0,
      "step": 29080
    },
    {
      "entropy": 6.512970018386841,
      "epoch": 3.1123655626304245,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040506469745584314,
      "loss": 5.8146,
      "mean_token_accuracy": 0.17604104280471802,
      "num_tokens": 62954017.0,
      "step": 29085
    },
    {
      "entropy": 6.518638467788696,
      "epoch": 3.112900636738188,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040503349752050294,
      "loss": 5.8161,
      "mean_token_accuracy": 0.175652015209198,
      "num_tokens": 62964352.0,
      "step": 29090
    },
    {
      "entropy": 6.454585981369019,
      "epoch": 3.1134357108459523,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004050022938304639,
      "loss": 5.6792,
      "mean_token_accuracy": 0.18048778176307678,
      "num_tokens": 62974414.0,
      "step": 29095
    },
    {
      "entropy": 6.509975814819336,
      "epoch": 3.113970784953716,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004049710863866268,
      "loss": 5.8331,
      "mean_token_accuracy": 0.16873296946287156,
      "num_tokens": 62985874.0,
      "step": 29100
    },
    {
      "entropy": 6.533389949798584,
      "epoch": 3.11450585906148,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004049398751898929,
      "loss": 5.7809,
      "mean_token_accuracy": 0.17397657185792922,
      "num_tokens": 62996553.0,
      "step": 29105
    },
    {
      "entropy": 6.542629623413086,
      "epoch": 3.115040933169244,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004049086602411633,
      "loss": 5.8296,
      "mean_token_accuracy": 0.16825563311576844,
      "num_tokens": 63008094.0,
      "step": 29110
    },
    {
      "entropy": 6.558885335922241,
      "epoch": 3.115576007277008,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040487744154133945,
      "loss": 5.7997,
      "mean_token_accuracy": 0.1646008610725403,
      "num_tokens": 63018517.0,
      "step": 29115
    },
    {
      "entropy": 6.493367147445679,
      "epoch": 3.116111081384772,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040484621909132277,
      "loss": 5.7374,
      "mean_token_accuracy": 0.17370732724666596,
      "num_tokens": 63029759.0,
      "step": 29120
    },
    {
      "entropy": 6.5511391162872314,
      "epoch": 3.1166461554925355,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004048149928920146,
      "loss": 5.8062,
      "mean_token_accuracy": 0.17202549874782563,
      "num_tokens": 63039862.0,
      "step": 29125
    },
    {
      "entropy": 6.571433448791504,
      "epoch": 3.1171812296002996,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004047837629443169,
      "loss": 5.8601,
      "mean_token_accuracy": 0.16787311136722566,
      "num_tokens": 63050298.0,
      "step": 29130
    },
    {
      "entropy": 6.325559377670288,
      "epoch": 3.1177163037080637,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040475252924913126,
      "loss": 5.5791,
      "mean_token_accuracy": 0.2053312584757805,
      "num_tokens": 63061280.0,
      "step": 29135
    },
    {
      "entropy": 6.5402497291564945,
      "epoch": 3.1182513778158274,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004047212918073595,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17489349395036696,
      "num_tokens": 63071188.0,
      "step": 29140
    },
    {
      "entropy": 6.512634706497193,
      "epoch": 3.1187864519235915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040469005061990364,
      "loss": 5.7705,
      "mean_token_accuracy": 0.17814851552248,
      "num_tokens": 63081804.0,
      "step": 29145
    },
    {
      "entropy": 6.50342206954956,
      "epoch": 3.119321526031355,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004046588056876659,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16903730034828185,
      "num_tokens": 63092671.0,
      "step": 29150
    },
    {
      "entropy": 6.517676591873169,
      "epoch": 3.1198566001391193,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040462755701154834,
      "loss": 5.8094,
      "mean_token_accuracy": 0.1734837457537651,
      "num_tokens": 63104867.0,
      "step": 29155
    },
    {
      "entropy": 6.48099308013916,
      "epoch": 3.1203916742468834,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040459630459245327,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17703667730093003,
      "num_tokens": 63115491.0,
      "step": 29160
    },
    {
      "entropy": 6.465249681472779,
      "epoch": 3.120926748354647,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045650484312831,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17867670357227325,
      "num_tokens": 63127240.0,
      "step": 29165
    },
    {
      "entropy": 6.5084600925445555,
      "epoch": 3.121461822462411,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004045337885289404,
      "loss": 5.7638,
      "mean_token_accuracy": 0.17672971785068511,
      "num_tokens": 63137180.0,
      "step": 29170
    },
    {
      "entropy": 6.525170755386353,
      "epoch": 3.1219968965701748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004045025248863277,
      "loss": 5.8732,
      "mean_token_accuracy": 0.17036725133657454,
      "num_tokens": 63147772.0,
      "step": 29175
    },
    {
      "entropy": 6.509455823898316,
      "epoch": 3.122531970677939,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004044712575043479,
      "loss": 5.7899,
      "mean_token_accuracy": 0.1788125216960907,
      "num_tokens": 63158592.0,
      "step": 29180
    },
    {
      "entropy": 6.465343236923218,
      "epoch": 3.123067044785703,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040443998638390363,
      "loss": 5.7311,
      "mean_token_accuracy": 0.17735537588596345,
      "num_tokens": 63168260.0,
      "step": 29185
    },
    {
      "entropy": 6.5115457534790036,
      "epoch": 3.1236021188934666,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040440871152589797,
      "loss": 5.7902,
      "mean_token_accuracy": 0.1693448469042778,
      "num_tokens": 63178650.0,
      "step": 29190
    },
    {
      "entropy": 6.563362646102905,
      "epoch": 3.1241371930012307,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040437743293123403,
      "loss": 5.7706,
      "mean_token_accuracy": 0.17505760937929155,
      "num_tokens": 63188491.0,
      "step": 29195
    },
    {
      "entropy": 6.499308300018311,
      "epoch": 3.124672267108995,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040434615060081473,
      "loss": 5.7286,
      "mean_token_accuracy": 0.18016821295022964,
      "num_tokens": 63198915.0,
      "step": 29200
    },
    {
      "entropy": 6.483736562728882,
      "epoch": 3.1252073412167585,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004043148645355436,
      "loss": 5.7408,
      "mean_token_accuracy": 0.1765717700123787,
      "num_tokens": 63210010.0,
      "step": 29205
    },
    {
      "entropy": 6.5052801132202145,
      "epoch": 3.1257424153245226,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000404283574736324,
      "loss": 5.821,
      "mean_token_accuracy": 0.17635394781827926,
      "num_tokens": 63220233.0,
      "step": 29210
    },
    {
      "entropy": 6.440524578094482,
      "epoch": 3.1262774894322862,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004042522812040592,
      "loss": 5.706,
      "mean_token_accuracy": 0.17830491065979004,
      "num_tokens": 63231258.0,
      "step": 29215
    },
    {
      "entropy": 6.402766180038452,
      "epoch": 3.1268125635400503,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000404220983939653,
      "loss": 5.7384,
      "mean_token_accuracy": 0.17167857438325881,
      "num_tokens": 63241781.0,
      "step": 29220
    },
    {
      "entropy": 6.540172100067139,
      "epoch": 3.1273476376478144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004041896829440091,
      "loss": 5.8026,
      "mean_token_accuracy": 0.17431037425994872,
      "num_tokens": 63253359.0,
      "step": 29225
    },
    {
      "entropy": 6.557604265213013,
      "epoch": 3.127882711755578,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004041583782180312,
      "loss": 5.807,
      "mean_token_accuracy": 0.1752866193652153,
      "num_tokens": 63263983.0,
      "step": 29230
    },
    {
      "entropy": 6.468903970718384,
      "epoch": 3.128417785863342,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004041270697626233,
      "loss": 5.6745,
      "mean_token_accuracy": 0.18917917609214782,
      "num_tokens": 63274503.0,
      "step": 29235
    },
    {
      "entropy": 6.566156339645386,
      "epoch": 3.128952859971106,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004040957575786894,
      "loss": 5.8388,
      "mean_token_accuracy": 0.16801310777664186,
      "num_tokens": 63285863.0,
      "step": 29240
    },
    {
      "entropy": 6.541887521743774,
      "epoch": 3.12948793407887,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004040644416671337,
      "loss": 5.8241,
      "mean_token_accuracy": 0.1740554928779602,
      "num_tokens": 63297089.0,
      "step": 29245
    },
    {
      "entropy": 6.5330424308776855,
      "epoch": 3.130023008186634,
      "grad_norm": 1.25,
      "learning_rate": 0.00040403312202886026,
      "loss": 5.765,
      "mean_token_accuracy": 0.1752301722764969,
      "num_tokens": 63307591.0,
      "step": 29250
    },
    {
      "entropy": 6.504920816421508,
      "epoch": 3.1305580822943977,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004040017986647736,
      "loss": 5.7324,
      "mean_token_accuracy": 0.18220083713531493,
      "num_tokens": 63317895.0,
      "step": 29255
    },
    {
      "entropy": 6.562329435348511,
      "epoch": 3.131093156402162,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040397047157577823,
      "loss": 5.8009,
      "mean_token_accuracy": 0.17264579683542253,
      "num_tokens": 63327829.0,
      "step": 29260
    },
    {
      "entropy": 6.538815116882324,
      "epoch": 3.1316282305099254,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004039391407627785,
      "loss": 5.819,
      "mean_token_accuracy": 0.17516257464885712,
      "num_tokens": 63337842.0,
      "step": 29265
    },
    {
      "entropy": 6.494863224029541,
      "epoch": 3.1321633046176895,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040390780622667923,
      "loss": 5.7189,
      "mean_token_accuracy": 0.17824832946062089,
      "num_tokens": 63349103.0,
      "step": 29270
    },
    {
      "entropy": 6.495817804336548,
      "epoch": 3.1326983787254536,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040387646796838524,
      "loss": 5.8174,
      "mean_token_accuracy": 0.16870780736207963,
      "num_tokens": 63359172.0,
      "step": 29275
    },
    {
      "entropy": 6.52702751159668,
      "epoch": 3.1332334528332173,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040384512598880127,
      "loss": 5.7811,
      "mean_token_accuracy": 0.17387524843215943,
      "num_tokens": 63370397.0,
      "step": 29280
    },
    {
      "entropy": 6.49996600151062,
      "epoch": 3.1337685269409814,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040381378028883234,
      "loss": 5.7776,
      "mean_token_accuracy": 0.16713542342185975,
      "num_tokens": 63380592.0,
      "step": 29285
    },
    {
      "entropy": 6.552072954177857,
      "epoch": 3.134303601048745,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040378243086938373,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1665787234902382,
      "num_tokens": 63392115.0,
      "step": 29290
    },
    {
      "entropy": 6.484538984298706,
      "epoch": 3.134838675156509,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004037510777313604,
      "loss": 5.7695,
      "mean_token_accuracy": 0.17785045355558396,
      "num_tokens": 63403287.0,
      "step": 29295
    },
    {
      "entropy": 6.51608943939209,
      "epoch": 3.1353737492642733,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004037197208756679,
      "loss": 5.7841,
      "mean_token_accuracy": 0.18150755316019057,
      "num_tokens": 63415248.0,
      "step": 29300
    },
    {
      "entropy": 6.553620672225952,
      "epoch": 3.135908823372037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004036883603032115,
      "loss": 5.7998,
      "mean_token_accuracy": 0.17141447067260743,
      "num_tokens": 63425693.0,
      "step": 29305
    },
    {
      "entropy": 6.492815637588501,
      "epoch": 3.136443897479801,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004036569960148968,
      "loss": 5.7339,
      "mean_token_accuracy": 0.17359143793582915,
      "num_tokens": 63437711.0,
      "step": 29310
    },
    {
      "entropy": 6.487862396240234,
      "epoch": 3.136978971587565,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040362562801162937,
      "loss": 5.7711,
      "mean_token_accuracy": 0.17155006676912307,
      "num_tokens": 63448190.0,
      "step": 29315
    },
    {
      "entropy": 6.54687123298645,
      "epoch": 3.1375140456953288,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040359425629431505,
      "loss": 5.7896,
      "mean_token_accuracy": 0.17577204704284669,
      "num_tokens": 63459285.0,
      "step": 29320
    },
    {
      "entropy": 6.546291971206665,
      "epoch": 3.138049119803093,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040356288086385966,
      "loss": 5.8104,
      "mean_token_accuracy": 0.17096975147724153,
      "num_tokens": 63469093.0,
      "step": 29325
    },
    {
      "entropy": 6.484145450592041,
      "epoch": 3.1385841939108565,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040353150172116916,
      "loss": 5.755,
      "mean_token_accuracy": 0.17801473885774613,
      "num_tokens": 63479108.0,
      "step": 29330
    },
    {
      "entropy": 6.487549114227295,
      "epoch": 3.1391192680186206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040350011886714957,
      "loss": 5.8494,
      "mean_token_accuracy": 0.17320305854082108,
      "num_tokens": 63490656.0,
      "step": 29335
    },
    {
      "entropy": 6.497954654693603,
      "epoch": 3.1396543421263843,
      "grad_norm": 1.125,
      "learning_rate": 0.00040346873230270706,
      "loss": 5.7375,
      "mean_token_accuracy": 0.18382011204957963,
      "num_tokens": 63501261.0,
      "step": 29340
    },
    {
      "entropy": 6.571120548248291,
      "epoch": 3.1401894162341484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000403437342028748,
      "loss": 5.7976,
      "mean_token_accuracy": 0.16900175362825393,
      "num_tokens": 63511926.0,
      "step": 29345
    },
    {
      "entropy": 6.537301635742187,
      "epoch": 3.1407244903419125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040340594804617874,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1728455051779747,
      "num_tokens": 63523140.0,
      "step": 29350
    },
    {
      "entropy": 6.530235147476196,
      "epoch": 3.141259564449676,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040337455035590583,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16760366261005402,
      "num_tokens": 63532745.0,
      "step": 29355
    },
    {
      "entropy": 6.495395088195801,
      "epoch": 3.1417946385574402,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040334314895883574,
      "loss": 5.8023,
      "mean_token_accuracy": 0.17695892304182054,
      "num_tokens": 63544144.0,
      "step": 29360
    },
    {
      "entropy": 6.59240837097168,
      "epoch": 3.1423297126652043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040331174385587533,
      "loss": 5.8516,
      "mean_token_accuracy": 0.1668265014886856,
      "num_tokens": 63554615.0,
      "step": 29365
    },
    {
      "entropy": 6.57316746711731,
      "epoch": 3.142864786772968,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004032803350479313,
      "loss": 5.8422,
      "mean_token_accuracy": 0.1683745875954628,
      "num_tokens": 63565620.0,
      "step": 29370
    },
    {
      "entropy": 6.4368997573852536,
      "epoch": 3.143399860880732,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004032489225359106,
      "loss": 5.7315,
      "mean_token_accuracy": 0.1830444887280464,
      "num_tokens": 63575787.0,
      "step": 29375
    },
    {
      "entropy": 6.420780181884766,
      "epoch": 3.1439349349884957,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004032175063207203,
      "loss": 5.824,
      "mean_token_accuracy": 0.17314514964818956,
      "num_tokens": 63586644.0,
      "step": 29380
    },
    {
      "entropy": 6.5191277980804445,
      "epoch": 3.14447000909626,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040318608640326756,
      "loss": 5.7802,
      "mean_token_accuracy": 0.17035354524850846,
      "num_tokens": 63596592.0,
      "step": 29385
    },
    {
      "entropy": 6.617358779907226,
      "epoch": 3.145005083204024,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040315466278445955,
      "loss": 5.8366,
      "mean_token_accuracy": 0.16980407387018204,
      "num_tokens": 63607263.0,
      "step": 29390
    },
    {
      "entropy": 6.4681953430175785,
      "epoch": 3.1455401573117876,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004031232354652037,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17822767794132233,
      "num_tokens": 63618494.0,
      "step": 29395
    },
    {
      "entropy": 6.445747900009155,
      "epoch": 3.1460752314195517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040309180444640734,
      "loss": 5.698,
      "mean_token_accuracy": 0.17630740106105805,
      "num_tokens": 63629100.0,
      "step": 29400
    },
    {
      "entropy": 6.425060987472534,
      "epoch": 3.1466103055273154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040306036972897823,
      "loss": 5.7475,
      "mean_token_accuracy": 0.17457664310932158,
      "num_tokens": 63640088.0,
      "step": 29405
    },
    {
      "entropy": 6.439825868606567,
      "epoch": 3.1471453796350795,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004030289313138239,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17624555826187133,
      "num_tokens": 63650262.0,
      "step": 29410
    },
    {
      "entropy": 6.566877937316894,
      "epoch": 3.1476804537428436,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040299748920185214,
      "loss": 5.8721,
      "mean_token_accuracy": 0.16742943823337555,
      "num_tokens": 63661919.0,
      "step": 29415
    },
    {
      "entropy": 6.545003986358642,
      "epoch": 3.148215527850607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040296604339397086,
      "loss": 5.7961,
      "mean_token_accuracy": 0.1676165819168091,
      "num_tokens": 63672185.0,
      "step": 29420
    },
    {
      "entropy": 6.55153455734253,
      "epoch": 3.1487506019583713,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004029345938910881,
      "loss": 5.8486,
      "mean_token_accuracy": 0.16474187821149827,
      "num_tokens": 63682948.0,
      "step": 29425
    },
    {
      "entropy": 6.547875690460205,
      "epoch": 3.149285676066135,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040290314069411193,
      "loss": 5.9037,
      "mean_token_accuracy": 0.16564375162124634,
      "num_tokens": 63693227.0,
      "step": 29430
    },
    {
      "entropy": 6.527328968048096,
      "epoch": 3.149820750173899,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004028716838039505,
      "loss": 5.8077,
      "mean_token_accuracy": 0.17630944699048995,
      "num_tokens": 63705315.0,
      "step": 29435
    },
    {
      "entropy": 6.540395927429199,
      "epoch": 3.150355824281663,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004028402232215122,
      "loss": 5.7955,
      "mean_token_accuracy": 0.1759384974837303,
      "num_tokens": 63715924.0,
      "step": 29440
    },
    {
      "entropy": 6.491977167129517,
      "epoch": 3.150890898389427,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004028087589477054,
      "loss": 5.7805,
      "mean_token_accuracy": 0.17222736924886703,
      "num_tokens": 63726889.0,
      "step": 29445
    },
    {
      "entropy": 6.573023509979248,
      "epoch": 3.151425972497191,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004027772909834387,
      "loss": 5.8332,
      "mean_token_accuracy": 0.170660537481308,
      "num_tokens": 63737102.0,
      "step": 29450
    },
    {
      "entropy": 6.541289949417115,
      "epoch": 3.1519610466049546,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004027458193296206,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17655352354049683,
      "num_tokens": 63747079.0,
      "step": 29455
    },
    {
      "entropy": 6.54916467666626,
      "epoch": 3.1524961207127187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004027143439871599,
      "loss": 5.8361,
      "mean_token_accuracy": 0.17092261612415313,
      "num_tokens": 63758852.0,
      "step": 29460
    },
    {
      "entropy": 6.5396466732025145,
      "epoch": 3.1530311948204828,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004026828649569656,
      "loss": 5.7608,
      "mean_token_accuracy": 0.17736270427703857,
      "num_tokens": 63770056.0,
      "step": 29465
    },
    {
      "entropy": 6.562098121643066,
      "epoch": 3.1535662689282464,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040265138223994647,
      "loss": 5.8343,
      "mean_token_accuracy": 0.17255962491035462,
      "num_tokens": 63781241.0,
      "step": 29470
    },
    {
      "entropy": 6.509048175811768,
      "epoch": 3.1541013430360105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004026198958370115,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17109998613595961,
      "num_tokens": 63792393.0,
      "step": 29475
    },
    {
      "entropy": 6.518115282058716,
      "epoch": 3.1546364171437746,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004025884057490701,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1681907907128334,
      "num_tokens": 63802474.0,
      "step": 29480
    },
    {
      "entropy": 6.530253791809082,
      "epoch": 3.1551714912515383,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004025569119770313,
      "loss": 5.7424,
      "mean_token_accuracy": 0.17746728509664536,
      "num_tokens": 63812247.0,
      "step": 29485
    },
    {
      "entropy": 6.551263475418091,
      "epoch": 3.1557065653593024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040252541452180467,
      "loss": 5.8353,
      "mean_token_accuracy": 0.1752293661236763,
      "num_tokens": 63822230.0,
      "step": 29490
    },
    {
      "entropy": 6.516825151443482,
      "epoch": 3.156241639467066,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004024939133842996,
      "loss": 5.8464,
      "mean_token_accuracy": 0.16982854753732682,
      "num_tokens": 63832594.0,
      "step": 29495
    },
    {
      "entropy": 6.461235857009887,
      "epoch": 3.15677671357483,
      "grad_norm": 1.125,
      "learning_rate": 0.0004024624085654257,
      "loss": 5.7961,
      "mean_token_accuracy": 0.16748285740613938,
      "num_tokens": 63844602.0,
      "step": 29500
    },
    {
      "entropy": 6.4688544273376465,
      "epoch": 3.1573117876825942,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004024309000660927,
      "loss": 5.7138,
      "mean_token_accuracy": 0.18221824765205383,
      "num_tokens": 63854683.0,
      "step": 29505
    },
    {
      "entropy": 6.493501091003418,
      "epoch": 3.157846861790358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040239938788721026,
      "loss": 5.7136,
      "mean_token_accuracy": 0.18170934170484543,
      "num_tokens": 63865889.0,
      "step": 29510
    },
    {
      "entropy": 6.504979276657105,
      "epoch": 3.158381935898122,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040236787202968847,
      "loss": 5.7613,
      "mean_token_accuracy": 0.17649233043193818,
      "num_tokens": 63875490.0,
      "step": 29515
    },
    {
      "entropy": 6.5183494091033936,
      "epoch": 3.1589170100058857,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004023363524944373,
      "loss": 5.8193,
      "mean_token_accuracy": 0.18023934811353684,
      "num_tokens": 63886319.0,
      "step": 29520
    },
    {
      "entropy": 6.455697107315063,
      "epoch": 3.1594520841136498,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004023048292823667,
      "loss": 5.7055,
      "mean_token_accuracy": 0.1734362319111824,
      "num_tokens": 63897588.0,
      "step": 29525
    },
    {
      "entropy": 6.4300275325775145,
      "epoch": 3.159987158221414,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004022733023943872,
      "loss": 5.7896,
      "mean_token_accuracy": 0.1722234696149826,
      "num_tokens": 63908350.0,
      "step": 29530
    },
    {
      "entropy": 6.491633939743042,
      "epoch": 3.1605222323291775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004022417718314089,
      "loss": 5.8407,
      "mean_token_accuracy": 0.1671878233551979,
      "num_tokens": 63918744.0,
      "step": 29535
    },
    {
      "entropy": 6.491337585449219,
      "epoch": 3.1610573064369416,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004022102375943424,
      "loss": 5.7464,
      "mean_token_accuracy": 0.17665988951921463,
      "num_tokens": 63928593.0,
      "step": 29540
    },
    {
      "entropy": 6.444918251037597,
      "epoch": 3.1615923805447053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004021786996840981,
      "loss": 5.7445,
      "mean_token_accuracy": 0.18405836075544357,
      "num_tokens": 63939912.0,
      "step": 29545
    },
    {
      "entropy": 6.497038221359253,
      "epoch": 3.1621274546524694,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040214715810158667,
      "loss": 5.7947,
      "mean_token_accuracy": 0.1776667818427086,
      "num_tokens": 63951356.0,
      "step": 29550
    },
    {
      "entropy": 6.552052307128906,
      "epoch": 3.1626625287602335,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000402115612847719,
      "loss": 5.7843,
      "mean_token_accuracy": 0.1773966670036316,
      "num_tokens": 63961252.0,
      "step": 29555
    },
    {
      "entropy": 6.51564736366272,
      "epoch": 3.163197602867997,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004020840639234059,
      "loss": 5.8166,
      "mean_token_accuracy": 0.17745290845632553,
      "num_tokens": 63972090.0,
      "step": 29560
    },
    {
      "entropy": 6.512389707565307,
      "epoch": 3.163732676975761,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040205251132955825,
      "loss": 5.8117,
      "mean_token_accuracy": 0.17783918976783752,
      "num_tokens": 63983092.0,
      "step": 29565
    },
    {
      "entropy": 6.474357604980469,
      "epoch": 3.164267751083525,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040202095506708727,
      "loss": 5.7795,
      "mean_token_accuracy": 0.17892174571752548,
      "num_tokens": 63992527.0,
      "step": 29570
    },
    {
      "entropy": 6.524525737762451,
      "epoch": 3.164802825191289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000401989395136904,
      "loss": 5.7806,
      "mean_token_accuracy": 0.17506806850433348,
      "num_tokens": 64003938.0,
      "step": 29575
    },
    {
      "entropy": 6.510895442962647,
      "epoch": 3.165337899299053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004019578315399197,
      "loss": 5.7325,
      "mean_token_accuracy": 0.17738165557384492,
      "num_tokens": 64013925.0,
      "step": 29580
    },
    {
      "entropy": 6.554517936706543,
      "epoch": 3.1658729734068167,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004019262642770459,
      "loss": 5.9247,
      "mean_token_accuracy": 0.1645405501127243,
      "num_tokens": 64024503.0,
      "step": 29585
    },
    {
      "entropy": 6.521755790710449,
      "epoch": 3.166408047514581,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040189469334919416,
      "loss": 5.7936,
      "mean_token_accuracy": 0.1684940919280052,
      "num_tokens": 64035650.0,
      "step": 29590
    },
    {
      "entropy": 6.519572591781616,
      "epoch": 3.166943121622345,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004018631187572759,
      "loss": 5.8165,
      "mean_token_accuracy": 0.17030999213457107,
      "num_tokens": 64047032.0,
      "step": 29595
    },
    {
      "entropy": 6.433445167541504,
      "epoch": 3.1674781957301086,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040183154050220295,
      "loss": 5.6661,
      "mean_token_accuracy": 0.18912891149520875,
      "num_tokens": 64057578.0,
      "step": 29600
    },
    {
      "entropy": 6.605439138412476,
      "epoch": 3.1680132698378727,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000401799958584887,
      "loss": 5.9146,
      "mean_token_accuracy": 0.16152915805578233,
      "num_tokens": 64068679.0,
      "step": 29605
    },
    {
      "entropy": 6.514319515228271,
      "epoch": 3.1685483439456363,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040176837300624015,
      "loss": 5.8127,
      "mean_token_accuracy": 0.17213257998228074,
      "num_tokens": 64078572.0,
      "step": 29610
    },
    {
      "entropy": 6.499879217147827,
      "epoch": 3.1690834180534004,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004017367837671743,
      "loss": 5.7285,
      "mean_token_accuracy": 0.17369577884674073,
      "num_tokens": 64088507.0,
      "step": 29615
    },
    {
      "entropy": 6.4605508804321286,
      "epoch": 3.169618492161164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040170519086860166,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1770574927330017,
      "num_tokens": 64099405.0,
      "step": 29620
    },
    {
      "entropy": 6.45126314163208,
      "epoch": 3.170153566268928,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040167359431143444,
      "loss": 5.7531,
      "mean_token_accuracy": 0.1809648111462593,
      "num_tokens": 64109554.0,
      "step": 29625
    },
    {
      "entropy": 6.4998125553131105,
      "epoch": 3.1706886403766923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000401641994096585,
      "loss": 5.7192,
      "mean_token_accuracy": 0.1793619304895401,
      "num_tokens": 64119817.0,
      "step": 29630
    },
    {
      "entropy": 6.518010234832763,
      "epoch": 3.171223714484456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004016103902249658,
      "loss": 5.8184,
      "mean_token_accuracy": 0.17400465458631514,
      "num_tokens": 64131001.0,
      "step": 29635
    },
    {
      "entropy": 6.4997968673706055,
      "epoch": 3.17175878859222,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004015787826974892,
      "loss": 5.7711,
      "mean_token_accuracy": 0.17586710304021835,
      "num_tokens": 64142072.0,
      "step": 29640
    },
    {
      "entropy": 6.494489002227783,
      "epoch": 3.172293862699984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004015471715150682,
      "loss": 5.8144,
      "mean_token_accuracy": 0.17298004031181335,
      "num_tokens": 64153097.0,
      "step": 29645
    },
    {
      "entropy": 6.56265959739685,
      "epoch": 3.172828936807748,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004015155566786154,
      "loss": 5.8714,
      "mean_token_accuracy": 0.17052744925022126,
      "num_tokens": 64164952.0,
      "step": 29650
    },
    {
      "entropy": 6.532533454895019,
      "epoch": 3.173364010915512,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040148393818904366,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16712563186883928,
      "num_tokens": 64175754.0,
      "step": 29655
    },
    {
      "entropy": 6.5944112777709964,
      "epoch": 3.1738990850232756,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004014523160472659,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16929504424333572,
      "num_tokens": 64186888.0,
      "step": 29660
    },
    {
      "entropy": 6.518522167205811,
      "epoch": 3.1744341591310397,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040142069025419537,
      "loss": 5.7759,
      "mean_token_accuracy": 0.17433404326438903,
      "num_tokens": 64197362.0,
      "step": 29665
    },
    {
      "entropy": 6.5462054252624515,
      "epoch": 3.1749692332388038,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004013890608107451,
      "loss": 5.8218,
      "mean_token_accuracy": 0.17521124035120011,
      "num_tokens": 64208667.0,
      "step": 29670
    },
    {
      "entropy": 6.59048638343811,
      "epoch": 3.1755043073465674,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004013574277178285,
      "loss": 5.8765,
      "mean_token_accuracy": 0.16389450132846833,
      "num_tokens": 64219525.0,
      "step": 29675
    },
    {
      "entropy": 6.520057106018067,
      "epoch": 3.1760393814543315,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040132579097635897,
      "loss": 5.7418,
      "mean_token_accuracy": 0.1776918038725853,
      "num_tokens": 64228916.0,
      "step": 29680
    },
    {
      "entropy": 6.477375888824463,
      "epoch": 3.176574455562095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040129415058724993,
      "loss": 5.7918,
      "mean_token_accuracy": 0.17181513458490372,
      "num_tokens": 64241467.0,
      "step": 29685
    },
    {
      "entropy": 6.558459711074829,
      "epoch": 3.1771095296698593,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040126250655141507,
      "loss": 5.7002,
      "mean_token_accuracy": 0.1814234063029289,
      "num_tokens": 64251330.0,
      "step": 29690
    },
    {
      "entropy": 6.507724142074585,
      "epoch": 3.1776446037776234,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040123085886976804,
      "loss": 5.7334,
      "mean_token_accuracy": 0.18356698304414748,
      "num_tokens": 64261991.0,
      "step": 29695
    },
    {
      "entropy": 6.526256418228149,
      "epoch": 3.178179677885387,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040119920754322277,
      "loss": 5.8125,
      "mean_token_accuracy": 0.1678932160139084,
      "num_tokens": 64272402.0,
      "step": 29700
    },
    {
      "entropy": 6.490857076644898,
      "epoch": 3.178714751993151,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040116755257269297,
      "loss": 5.75,
      "mean_token_accuracy": 0.17647744417190553,
      "num_tokens": 64283390.0,
      "step": 29705
    },
    {
      "entropy": 6.51210241317749,
      "epoch": 3.1792498261009148,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040113589395909295,
      "loss": 5.7774,
      "mean_token_accuracy": 0.17448538094758986,
      "num_tokens": 64294280.0,
      "step": 29710
    },
    {
      "entropy": 6.584578180313111,
      "epoch": 3.179784900208679,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004011042317033367,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16575801074504853,
      "num_tokens": 64305927.0,
      "step": 29715
    },
    {
      "entropy": 6.484233570098877,
      "epoch": 3.180319974316443,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004010725658063385,
      "loss": 5.716,
      "mean_token_accuracy": 0.1837887704372406,
      "num_tokens": 64316319.0,
      "step": 29720
    },
    {
      "entropy": 6.573418998718262,
      "epoch": 3.1808550484242066,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004010408962690126,
      "loss": 5.8153,
      "mean_token_accuracy": 0.17305251955986023,
      "num_tokens": 64326958.0,
      "step": 29725
    },
    {
      "entropy": 6.582005739212036,
      "epoch": 3.1813901225319707,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040100922309227357,
      "loss": 5.8281,
      "mean_token_accuracy": 0.16512832939624786,
      "num_tokens": 64338433.0,
      "step": 29730
    },
    {
      "entropy": 6.458195209503174,
      "epoch": 3.1819251966397344,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040097754627703593,
      "loss": 5.6556,
      "mean_token_accuracy": 0.19062540233135222,
      "num_tokens": 64349188.0,
      "step": 29735
    },
    {
      "entropy": 6.513427782058716,
      "epoch": 3.1824602707474985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004009458658242143,
      "loss": 5.8084,
      "mean_token_accuracy": 0.1712687149643898,
      "num_tokens": 64359837.0,
      "step": 29740
    },
    {
      "entropy": 6.50833158493042,
      "epoch": 3.1829953448552626,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040091418173472356,
      "loss": 5.7956,
      "mean_token_accuracy": 0.17668959200382234,
      "num_tokens": 64370927.0,
      "step": 29745
    },
    {
      "entropy": 6.445657062530517,
      "epoch": 3.1835304189630262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004008824940094785,
      "loss": 5.7415,
      "mean_token_accuracy": 0.1718342885375023,
      "num_tokens": 64381683.0,
      "step": 29750
    },
    {
      "entropy": 6.455428171157837,
      "epoch": 3.1840654930707903,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004008508026493941,
      "loss": 5.7713,
      "mean_token_accuracy": 0.17385755628347396,
      "num_tokens": 64394063.0,
      "step": 29755
    },
    {
      "entropy": 6.523726320266723,
      "epoch": 3.1846005671785544,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040081910765538543,
      "loss": 5.7825,
      "mean_token_accuracy": 0.17109575718641282,
      "num_tokens": 64405183.0,
      "step": 29760
    },
    {
      "entropy": 6.507354640960694,
      "epoch": 3.185135641286318,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040078740902836776,
      "loss": 5.7534,
      "mean_token_accuracy": 0.18300582766532897,
      "num_tokens": 64415325.0,
      "step": 29765
    },
    {
      "entropy": 6.5559930324554445,
      "epoch": 3.185670715394082,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040075570676925633,
      "loss": 5.8551,
      "mean_token_accuracy": 0.16605243235826492,
      "num_tokens": 64426240.0,
      "step": 29770
    },
    {
      "entropy": 6.490569019317627,
      "epoch": 3.186205789501846,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004007240008789664,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16935409158468245,
      "num_tokens": 64436664.0,
      "step": 29775
    },
    {
      "entropy": 6.529998922348023,
      "epoch": 3.18674086360961,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040069229135841376,
      "loss": 5.8993,
      "mean_token_accuracy": 0.16537918746471406,
      "num_tokens": 64448731.0,
      "step": 29780
    },
    {
      "entropy": 6.615342998504639,
      "epoch": 3.187275937717374,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004006605782085139,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16525874882936478,
      "num_tokens": 64460916.0,
      "step": 29785
    },
    {
      "entropy": 6.585196161270142,
      "epoch": 3.1878110118251377,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004006288614301824,
      "loss": 5.8532,
      "mean_token_accuracy": 0.1671522468328476,
      "num_tokens": 64472114.0,
      "step": 29790
    },
    {
      "entropy": 6.561496829986572,
      "epoch": 3.188346085932902,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040059714102433526,
      "loss": 5.7931,
      "mean_token_accuracy": 0.17520830333232879,
      "num_tokens": 64482465.0,
      "step": 29795
    },
    {
      "entropy": 6.46219573020935,
      "epoch": 3.1888811600406655,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004005654169918883,
      "loss": 5.7375,
      "mean_token_accuracy": 0.1784728839993477,
      "num_tokens": 64493285.0,
      "step": 29800
    },
    {
      "entropy": 6.530496263504029,
      "epoch": 3.1894162341484296,
      "grad_norm": 1.25,
      "learning_rate": 0.00040053368933375756,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16947079300880433,
      "num_tokens": 64504037.0,
      "step": 29805
    },
    {
      "entropy": 6.53930025100708,
      "epoch": 3.1899513082561937,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040050195805085907,
      "loss": 5.744,
      "mean_token_accuracy": 0.1796402543783188,
      "num_tokens": 64513900.0,
      "step": 29810
    },
    {
      "entropy": 6.509256601333618,
      "epoch": 3.1904863823639573,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040047022314410936,
      "loss": 5.8315,
      "mean_token_accuracy": 0.17493016719818116,
      "num_tokens": 64525273.0,
      "step": 29815
    },
    {
      "entropy": 6.548234748840332,
      "epoch": 3.1910214564717214,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040043848461442447,
      "loss": 5.8118,
      "mean_token_accuracy": 0.1727464720606804,
      "num_tokens": 64535929.0,
      "step": 29820
    },
    {
      "entropy": 6.558199834823609,
      "epoch": 3.191556530579485,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000400406742462721,
      "loss": 5.8133,
      "mean_token_accuracy": 0.1739219084382057,
      "num_tokens": 64546975.0,
      "step": 29825
    },
    {
      "entropy": 6.589832878112793,
      "epoch": 3.192091604687249,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040037499668991544,
      "loss": 5.8911,
      "mean_token_accuracy": 0.16215020418167114,
      "num_tokens": 64558295.0,
      "step": 29830
    },
    {
      "entropy": 6.514469623565674,
      "epoch": 3.1926266787950133,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040034324729692453,
      "loss": 5.698,
      "mean_token_accuracy": 0.18393871635198594,
      "num_tokens": 64569638.0,
      "step": 29835
    },
    {
      "entropy": 6.569105815887451,
      "epoch": 3.193161752902777,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004003114942846649,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16979342699050903,
      "num_tokens": 64580693.0,
      "step": 29840
    },
    {
      "entropy": 6.48675765991211,
      "epoch": 3.193696827010541,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040027973765405353,
      "loss": 5.787,
      "mean_token_accuracy": 0.17697692811489105,
      "num_tokens": 64590947.0,
      "step": 29845
    },
    {
      "entropy": 6.581830549240112,
      "epoch": 3.1942319011183047,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004002479774060073,
      "loss": 5.8699,
      "mean_token_accuracy": 0.16514147818088531,
      "num_tokens": 64602340.0,
      "step": 29850
    },
    {
      "entropy": 6.555405378341675,
      "epoch": 3.194766975226069,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004002162135414433,
      "loss": 5.7898,
      "mean_token_accuracy": 0.17366590946912766,
      "num_tokens": 64614410.0,
      "step": 29855
    },
    {
      "entropy": 6.616706132888794,
      "epoch": 3.195302049333833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004001844460612788,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16835221499204636,
      "num_tokens": 64624893.0,
      "step": 29860
    },
    {
      "entropy": 6.554902029037476,
      "epoch": 3.1958371234415965,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040015267496643097,
      "loss": 5.805,
      "mean_token_accuracy": 0.17325272113084794,
      "num_tokens": 64636172.0,
      "step": 29865
    },
    {
      "entropy": 6.506310844421387,
      "epoch": 3.1963721975493606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004001209002578172,
      "loss": 5.8117,
      "mean_token_accuracy": 0.1653382360935211,
      "num_tokens": 64647605.0,
      "step": 29870
    },
    {
      "entropy": 6.564558982849121,
      "epoch": 3.1969072716571247,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004000891219363551,
      "loss": 5.838,
      "mean_token_accuracy": 0.16542019546031952,
      "num_tokens": 64657496.0,
      "step": 29875
    },
    {
      "entropy": 6.536395072937012,
      "epoch": 3.1974423457648884,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040005734000296225,
      "loss": 5.7605,
      "mean_token_accuracy": 0.1797284632921219,
      "num_tokens": 64668905.0,
      "step": 29880
    },
    {
      "entropy": 6.4829834461212155,
      "epoch": 3.1979774198726525,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004000255544585561,
      "loss": 5.7715,
      "mean_token_accuracy": 0.1731960415840149,
      "num_tokens": 64679207.0,
      "step": 29885
    },
    {
      "entropy": 6.504586601257325,
      "epoch": 3.198512493980416,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003999937653040547,
      "loss": 5.7921,
      "mean_token_accuracy": 0.17385823130607606,
      "num_tokens": 64689536.0,
      "step": 29890
    },
    {
      "entropy": 6.586478233337402,
      "epoch": 3.1990475680881802,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003999619725403759,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15552898794412612,
      "num_tokens": 64700304.0,
      "step": 29895
    },
    {
      "entropy": 6.51788444519043,
      "epoch": 3.199582642195944,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039993017616843775,
      "loss": 5.7497,
      "mean_token_accuracy": 0.17525475174188615,
      "num_tokens": 64711657.0,
      "step": 29900
    },
    {
      "entropy": 6.4499835014343265,
      "epoch": 3.200117716303708,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003998983761891582,
      "loss": 5.7433,
      "mean_token_accuracy": 0.18160052746534347,
      "num_tokens": 64722697.0,
      "step": 29905
    },
    {
      "entropy": 6.44803147315979,
      "epoch": 3.200652790411472,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039986657260345567,
      "loss": 5.6996,
      "mean_token_accuracy": 0.18721374422311782,
      "num_tokens": 64732823.0,
      "step": 29910
    },
    {
      "entropy": 6.4979198455810545,
      "epoch": 3.2011878645192358,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003998347654122484,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17135123759508133,
      "num_tokens": 64744471.0,
      "step": 29915
    },
    {
      "entropy": 6.500985527038575,
      "epoch": 3.201722938627,
      "grad_norm": 1.25,
      "learning_rate": 0.00039980295461645483,
      "loss": 5.7809,
      "mean_token_accuracy": 0.1737814337015152,
      "num_tokens": 64754079.0,
      "step": 29920
    },
    {
      "entropy": 6.5569030284881595,
      "epoch": 3.202258012734764,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039977114021699344,
      "loss": 5.7886,
      "mean_token_accuracy": 0.16930270045995713,
      "num_tokens": 64764489.0,
      "step": 29925
    },
    {
      "entropy": 6.480095767974854,
      "epoch": 3.2027930868425276,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039973932221478293,
      "loss": 5.7343,
      "mean_token_accuracy": 0.18349780291318893,
      "num_tokens": 64775512.0,
      "step": 29930
    },
    {
      "entropy": 6.441008567810059,
      "epoch": 3.2033281609502917,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039970750061074206,
      "loss": 5.7551,
      "mean_token_accuracy": 0.16967564821243286,
      "num_tokens": 64787466.0,
      "step": 29935
    },
    {
      "entropy": 6.50834641456604,
      "epoch": 3.2038632350580554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039967567540578956,
      "loss": 5.7812,
      "mean_token_accuracy": 0.17806417793035506,
      "num_tokens": 64797380.0,
      "step": 29940
    },
    {
      "entropy": 6.5212420463562015,
      "epoch": 3.2043983091658195,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003996438466008446,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16993957459926606,
      "num_tokens": 64807609.0,
      "step": 29945
    },
    {
      "entropy": 6.4609699726104735,
      "epoch": 3.2049333832735836,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003996120141968259,
      "loss": 5.7203,
      "mean_token_accuracy": 0.1800501525402069,
      "num_tokens": 64818601.0,
      "step": 29950
    },
    {
      "entropy": 6.523859071731567,
      "epoch": 3.2054684573813472,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039958017819465285,
      "loss": 5.904,
      "mean_token_accuracy": 0.16863629072904587,
      "num_tokens": 64829597.0,
      "step": 29955
    },
    {
      "entropy": 6.495908737182617,
      "epoch": 3.2060035314891113,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039954833859524473,
      "loss": 5.7685,
      "mean_token_accuracy": 0.17725849598646165,
      "num_tokens": 64841293.0,
      "step": 29960
    },
    {
      "entropy": 6.56892876625061,
      "epoch": 3.206538605596875,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039951649539952076,
      "loss": 5.8174,
      "mean_token_accuracy": 0.17320433706045152,
      "num_tokens": 64852724.0,
      "step": 29965
    },
    {
      "entropy": 6.536759853363037,
      "epoch": 3.207073679704639,
      "grad_norm": 1.375,
      "learning_rate": 0.00039948464860840047,
      "loss": 5.7858,
      "mean_token_accuracy": 0.17873647511005403,
      "num_tokens": 64864915.0,
      "step": 29970
    },
    {
      "entropy": 6.506737184524536,
      "epoch": 3.207608753812403,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003994527982228035,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17387575805187225,
      "num_tokens": 64876431.0,
      "step": 29975
    },
    {
      "entropy": 6.517050504684448,
      "epoch": 3.208143827920167,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003994209442436495,
      "loss": 5.8084,
      "mean_token_accuracy": 0.17307513505220412,
      "num_tokens": 64887492.0,
      "step": 29980
    },
    {
      "entropy": 6.551008415222168,
      "epoch": 3.208678902027931,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039938908667185816,
      "loss": 5.8247,
      "mean_token_accuracy": 0.17464588582515717,
      "num_tokens": 64897894.0,
      "step": 29985
    },
    {
      "entropy": 6.564755058288574,
      "epoch": 3.2092139761356946,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039935722550834944,
      "loss": 5.8073,
      "mean_token_accuracy": 0.1684461861848831,
      "num_tokens": 64909071.0,
      "step": 29990
    },
    {
      "entropy": 6.548113203048706,
      "epoch": 3.2097490502434587,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003993253607540433,
      "loss": 5.8189,
      "mean_token_accuracy": 0.1755647838115692,
      "num_tokens": 64919434.0,
      "step": 29995
    },
    {
      "entropy": 6.442232656478882,
      "epoch": 3.210284124351223,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003992934924098598,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17352006286382676,
      "num_tokens": 64929897.0,
      "step": 30000
    },
    {
      "epoch": 3.210284124351223,
      "eval_entropy": 6.226400606587248,
      "eval_loss": 5.907739162445068,
      "eval_mean_token_accuracy": 0.17601743047590043,
      "eval_num_tokens": 64929897.0,
      "eval_runtime": 17.621,
      "eval_samples_per_second": 1684.297,
      "eval_steps_per_second": 210.544,
      "step": 30000
    },
    {
      "entropy": 6.426018762588501,
      "epoch": 3.2108191984589864,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003992616204767192,
      "loss": 5.6778,
      "mean_token_accuracy": 0.18850973546504973,
      "num_tokens": 64940219.0,
      "step": 30005
    },
    {
      "entropy": 6.487470626831055,
      "epoch": 3.2113542725667505,
      "grad_norm": 1.25,
      "learning_rate": 0.0003992297449555419,
      "loss": 5.7086,
      "mean_token_accuracy": 0.18036542534828187,
      "num_tokens": 64950326.0,
      "step": 30010
    },
    {
      "entropy": 6.520320034027099,
      "epoch": 3.211889346674514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039919786584724796,
      "loss": 5.8438,
      "mean_token_accuracy": 0.17411854416131972,
      "num_tokens": 64960873.0,
      "step": 30015
    },
    {
      "entropy": 6.457997608184814,
      "epoch": 3.2124244207822783,
      "grad_norm": 1.125,
      "learning_rate": 0.00039916598315275825,
      "loss": 5.7977,
      "mean_token_accuracy": 0.1806625872850418,
      "num_tokens": 64972138.0,
      "step": 30020
    },
    {
      "entropy": 6.558462619781494,
      "epoch": 3.2129594948900424,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003991340968729932,
      "loss": 5.8495,
      "mean_token_accuracy": 0.1688713476061821,
      "num_tokens": 64983002.0,
      "step": 30025
    },
    {
      "entropy": 6.535273742675781,
      "epoch": 3.213494568997806,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039910220700887345,
      "loss": 5.7734,
      "mean_token_accuracy": 0.1764858603477478,
      "num_tokens": 64993973.0,
      "step": 30030
    },
    {
      "entropy": 6.572481870651245,
      "epoch": 3.21402964310557,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039907031356131993,
      "loss": 5.8452,
      "mean_token_accuracy": 0.16507502496242524,
      "num_tokens": 65004711.0,
      "step": 30035
    },
    {
      "entropy": 6.53096194267273,
      "epoch": 3.2145647172133343,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039903841653125366,
      "loss": 5.8058,
      "mean_token_accuracy": 0.16729237884283066,
      "num_tokens": 65016426.0,
      "step": 30040
    },
    {
      "entropy": 6.439546871185303,
      "epoch": 3.215099791321098,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003990065159195955,
      "loss": 5.7317,
      "mean_token_accuracy": 0.18553781658411025,
      "num_tokens": 65027999.0,
      "step": 30045
    },
    {
      "entropy": 6.503052616119385,
      "epoch": 3.215634865428862,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003989746117272665,
      "loss": 5.7384,
      "mean_token_accuracy": 0.17342305332422256,
      "num_tokens": 65038216.0,
      "step": 30050
    },
    {
      "entropy": 6.4232072830200195,
      "epoch": 3.2161699395366257,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003989427039551881,
      "loss": 5.6968,
      "mean_token_accuracy": 0.1808346316218376,
      "num_tokens": 65049726.0,
      "step": 30055
    },
    {
      "entropy": 6.528130722045899,
      "epoch": 3.2167050136443898,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039891079260428157,
      "loss": 5.8387,
      "mean_token_accuracy": 0.1642815202474594,
      "num_tokens": 65060675.0,
      "step": 30060
    },
    {
      "entropy": 6.577535486221313,
      "epoch": 3.217240087752154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039887887767546824,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16981590986251832,
      "num_tokens": 65072378.0,
      "step": 30065
    },
    {
      "entropy": 6.613910675048828,
      "epoch": 3.2177751618599175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003988469591696698,
      "loss": 5.8083,
      "mean_token_accuracy": 0.17623913586139678,
      "num_tokens": 65082587.0,
      "step": 30070
    },
    {
      "entropy": 6.5899457931518555,
      "epoch": 3.2183102359676816,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003988150370878078,
      "loss": 5.7981,
      "mean_token_accuracy": 0.17565345019102097,
      "num_tokens": 65093210.0,
      "step": 30075
    },
    {
      "entropy": 6.507211399078369,
      "epoch": 3.2188453100754453,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000398783111430804,
      "loss": 5.7758,
      "mean_token_accuracy": 0.17527124732732774,
      "num_tokens": 65103074.0,
      "step": 30080
    },
    {
      "entropy": 6.515868616104126,
      "epoch": 3.2193803841832094,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039875118219958016,
      "loss": 5.8114,
      "mean_token_accuracy": 0.16947975754737854,
      "num_tokens": 65113994.0,
      "step": 30085
    },
    {
      "entropy": 6.511314058303833,
      "epoch": 3.2199154582909735,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003987192493950585,
      "loss": 5.7695,
      "mean_token_accuracy": 0.1768835335969925,
      "num_tokens": 65125381.0,
      "step": 30090
    },
    {
      "entropy": 6.52994933128357,
      "epoch": 3.220450532398737,
      "grad_norm": 1.25,
      "learning_rate": 0.00039868731301816076,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17421692311763765,
      "num_tokens": 65135374.0,
      "step": 30095
    },
    {
      "entropy": 6.48056173324585,
      "epoch": 3.2209856065065012,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039865537306980924,
      "loss": 5.6964,
      "mean_token_accuracy": 0.18088581711053847,
      "num_tokens": 65145514.0,
      "step": 30100
    },
    {
      "entropy": 6.488104200363159,
      "epoch": 3.221520680614265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003986234295509262,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1735217824578285,
      "num_tokens": 65156424.0,
      "step": 30105
    },
    {
      "entropy": 6.608654451370239,
      "epoch": 3.222055754722029,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039859148246243405,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16630660891532897,
      "num_tokens": 65167774.0,
      "step": 30110
    },
    {
      "entropy": 6.549910020828247,
      "epoch": 3.222590828829793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039855953180525515,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16868988275527955,
      "num_tokens": 65179552.0,
      "step": 30115
    },
    {
      "entropy": 6.57720103263855,
      "epoch": 3.2231259029375567,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039852757758031217,
      "loss": 5.8199,
      "mean_token_accuracy": 0.17929213494062424,
      "num_tokens": 65190386.0,
      "step": 30120
    },
    {
      "entropy": 6.555870628356933,
      "epoch": 3.223660977045321,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039849561978852775,
      "loss": 5.7816,
      "mean_token_accuracy": 0.17708924561738967,
      "num_tokens": 65201989.0,
      "step": 30125
    },
    {
      "entropy": 6.504936265945434,
      "epoch": 3.2241960511530845,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003984636584308248,
      "loss": 5.7452,
      "mean_token_accuracy": 0.17452987879514695,
      "num_tokens": 65212862.0,
      "step": 30130
    },
    {
      "entropy": 6.525814437866211,
      "epoch": 3.2247311252608486,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039843169350812585,
      "loss": 5.7792,
      "mean_token_accuracy": 0.17442704737186432,
      "num_tokens": 65223381.0,
      "step": 30135
    },
    {
      "entropy": 6.497701644897461,
      "epoch": 3.2252661993686127,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003983997250213542,
      "loss": 5.8119,
      "mean_token_accuracy": 0.17424905449151992,
      "num_tokens": 65234882.0,
      "step": 30140
    },
    {
      "entropy": 6.560753679275512,
      "epoch": 3.2258012734763764,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039836775297143277,
      "loss": 5.8435,
      "mean_token_accuracy": 0.1706228122115135,
      "num_tokens": 65244421.0,
      "step": 30145
    },
    {
      "entropy": 6.527552795410156,
      "epoch": 3.2263363475841405,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039833577735928484,
      "loss": 5.7594,
      "mean_token_accuracy": 0.1766971081495285,
      "num_tokens": 65255536.0,
      "step": 30150
    },
    {
      "entropy": 6.492141628265381,
      "epoch": 3.2268714216919046,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003983037981858337,
      "loss": 5.8419,
      "mean_token_accuracy": 0.17559577226638795,
      "num_tokens": 65265776.0,
      "step": 30155
    },
    {
      "entropy": 6.509293508529663,
      "epoch": 3.227406495799668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039827181545200264,
      "loss": 5.787,
      "mean_token_accuracy": 0.17087354958057405,
      "num_tokens": 65276003.0,
      "step": 30160
    },
    {
      "entropy": 6.562046718597412,
      "epoch": 3.2279415699074323,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003982398291587153,
      "loss": 5.8249,
      "mean_token_accuracy": 0.1721820831298828,
      "num_tokens": 65286817.0,
      "step": 30165
    },
    {
      "entropy": 6.50113878250122,
      "epoch": 3.228476644015196,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003982078393068952,
      "loss": 5.7273,
      "mean_token_accuracy": 0.17957308739423752,
      "num_tokens": 65298142.0,
      "step": 30170
    },
    {
      "entropy": 6.475490999221802,
      "epoch": 3.22901171812296,
      "grad_norm": 1.25,
      "learning_rate": 0.000398175845897466,
      "loss": 5.7734,
      "mean_token_accuracy": 0.18406370431184768,
      "num_tokens": 65308802.0,
      "step": 30175
    },
    {
      "entropy": 6.4874265670776365,
      "epoch": 3.229546792230724,
      "grad_norm": 1.125,
      "learning_rate": 0.0003981438489313516,
      "loss": 5.8219,
      "mean_token_accuracy": 0.16342270225286484,
      "num_tokens": 65319749.0,
      "step": 30180
    },
    {
      "entropy": 6.50525975227356,
      "epoch": 3.230081866338488,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039811184840947585,
      "loss": 5.8237,
      "mean_token_accuracy": 0.17769753485918044,
      "num_tokens": 65331519.0,
      "step": 30185
    },
    {
      "entropy": 6.573365259170532,
      "epoch": 3.230616940446252,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003980798443327629,
      "loss": 5.8577,
      "mean_token_accuracy": 0.16836510598659515,
      "num_tokens": 65342197.0,
      "step": 30190
    },
    {
      "entropy": 6.49190583229065,
      "epoch": 3.2311520145540156,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039804783670213656,
      "loss": 5.7907,
      "mean_token_accuracy": 0.17121997028589248,
      "num_tokens": 65353429.0,
      "step": 30195
    },
    {
      "entropy": 6.520119333267212,
      "epoch": 3.2316870886617797,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039801582551852126,
      "loss": 5.853,
      "mean_token_accuracy": 0.16692281514406204,
      "num_tokens": 65363836.0,
      "step": 30200
    },
    {
      "entropy": 6.56801962852478,
      "epoch": 3.2322221627695438,
      "grad_norm": 1.125,
      "learning_rate": 0.0003979838107828414,
      "loss": 5.8042,
      "mean_token_accuracy": 0.17628134936094284,
      "num_tokens": 65374329.0,
      "step": 30205
    },
    {
      "entropy": 6.574864435195923,
      "epoch": 3.2327572368773074,
      "grad_norm": 1.125,
      "learning_rate": 0.00039795179249602113,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17337048202753066,
      "num_tokens": 65385686.0,
      "step": 30210
    },
    {
      "entropy": 6.508423519134522,
      "epoch": 3.2332923109850715,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039791977065898525,
      "loss": 5.8395,
      "mean_token_accuracy": 0.1629524201154709,
      "num_tokens": 65397181.0,
      "step": 30215
    },
    {
      "entropy": 6.509705448150635,
      "epoch": 3.233827385092835,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039788774527265824,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16869853287935258,
      "num_tokens": 65407935.0,
      "step": 30220
    },
    {
      "entropy": 6.505736684799194,
      "epoch": 3.2343624592005993,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003978557163379649,
      "loss": 5.7545,
      "mean_token_accuracy": 0.1731651559472084,
      "num_tokens": 65418699.0,
      "step": 30225
    },
    {
      "entropy": 6.537900018692016,
      "epoch": 3.2348975333083634,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003978236838558298,
      "loss": 5.768,
      "mean_token_accuracy": 0.18106157034635545,
      "num_tokens": 65429783.0,
      "step": 30230
    },
    {
      "entropy": 6.421227598190308,
      "epoch": 3.235432607416127,
      "grad_norm": 1.25,
      "learning_rate": 0.0003977916478271783,
      "loss": 5.6861,
      "mean_token_accuracy": 0.18611322939395905,
      "num_tokens": 65441001.0,
      "step": 30235
    },
    {
      "entropy": 6.5314099311828615,
      "epoch": 3.235967681523891,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003977596082529352,
      "loss": 5.8158,
      "mean_token_accuracy": 0.17076822370290756,
      "num_tokens": 65451219.0,
      "step": 30240
    },
    {
      "entropy": 6.555176401138306,
      "epoch": 3.236502755631655,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003977275651340256,
      "loss": 5.7994,
      "mean_token_accuracy": 0.17574578076601027,
      "num_tokens": 65461081.0,
      "step": 30245
    },
    {
      "entropy": 6.57248272895813,
      "epoch": 3.237037829739419,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003976955184713748,
      "loss": 5.789,
      "mean_token_accuracy": 0.1728888764977455,
      "num_tokens": 65470391.0,
      "step": 30250
    },
    {
      "entropy": 6.499985694885254,
      "epoch": 3.237572903847183,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039766346826590823,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17384597212076186,
      "num_tokens": 65482199.0,
      "step": 30255
    },
    {
      "entropy": 6.511939573287964,
      "epoch": 3.2381079779549466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003976314145185511,
      "loss": 5.782,
      "mean_token_accuracy": 0.17597583830356597,
      "num_tokens": 65492497.0,
      "step": 30260
    },
    {
      "entropy": 6.444980335235596,
      "epoch": 3.2386430520627107,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039759935723022923,
      "loss": 5.6827,
      "mean_token_accuracy": 0.1819048061966896,
      "num_tokens": 65503769.0,
      "step": 30265
    },
    {
      "entropy": 6.51513557434082,
      "epoch": 3.2391781261704744,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039756729640186815,
      "loss": 5.8618,
      "mean_token_accuracy": 0.17182498425245285,
      "num_tokens": 65513582.0,
      "step": 30270
    },
    {
      "entropy": 6.517272186279297,
      "epoch": 3.2397132002782385,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003975352320343935,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1755541630089283,
      "num_tokens": 65523829.0,
      "step": 30275
    },
    {
      "entropy": 6.564678144454956,
      "epoch": 3.2402482743860026,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039750316412873134,
      "loss": 5.7666,
      "mean_token_accuracy": 0.1766463667154312,
      "num_tokens": 65534893.0,
      "step": 30280
    },
    {
      "entropy": 6.500364780426025,
      "epoch": 3.2407833484937663,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003974710926858075,
      "loss": 5.6553,
      "mean_token_accuracy": 0.19131432473659515,
      "num_tokens": 65544551.0,
      "step": 30285
    },
    {
      "entropy": 6.4897504329681395,
      "epoch": 3.2413184226015304,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003974390177065481,
      "loss": 5.758,
      "mean_token_accuracy": 0.1734618991613388,
      "num_tokens": 65554991.0,
      "step": 30290
    },
    {
      "entropy": 6.530030584335327,
      "epoch": 3.241853496709294,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003974069391918792,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17553792148828506,
      "num_tokens": 65565784.0,
      "step": 30295
    },
    {
      "entropy": 6.525287246704101,
      "epoch": 3.242388570817058,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039737485714272727,
      "loss": 5.753,
      "mean_token_accuracy": 0.17882264107465745,
      "num_tokens": 65575545.0,
      "step": 30300
    },
    {
      "entropy": 6.533629417419434,
      "epoch": 3.242923644924822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003973427715600184,
      "loss": 5.8476,
      "mean_token_accuracy": 0.16451952308416368,
      "num_tokens": 65585987.0,
      "step": 30305
    },
    {
      "entropy": 6.539637088775635,
      "epoch": 3.243458719032586,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003973106824446792,
      "loss": 5.8222,
      "mean_token_accuracy": 0.16773179322481155,
      "num_tokens": 65596991.0,
      "step": 30310
    },
    {
      "entropy": 6.543335676193237,
      "epoch": 3.24399379314035,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039727858979763634,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17656747549772261,
      "num_tokens": 65608161.0,
      "step": 30315
    },
    {
      "entropy": 6.576534414291382,
      "epoch": 3.244528867248114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039724649361981627,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17677524089813232,
      "num_tokens": 65619859.0,
      "step": 30320
    },
    {
      "entropy": 6.537126636505127,
      "epoch": 3.2450639413558777,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003972143939121459,
      "loss": 5.8128,
      "mean_token_accuracy": 0.1649240270256996,
      "num_tokens": 65631400.0,
      "step": 30325
    },
    {
      "entropy": 6.575593757629394,
      "epoch": 3.245599015463642,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003971822906755521,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16450818330049516,
      "num_tokens": 65643519.0,
      "step": 30330
    },
    {
      "entropy": 6.462250185012818,
      "epoch": 3.2461340895714055,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039715018391096185,
      "loss": 5.6842,
      "mean_token_accuracy": 0.17712902426719665,
      "num_tokens": 65654628.0,
      "step": 30335
    },
    {
      "entropy": 6.490246868133545,
      "epoch": 3.2466691636791696,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003971180736193022,
      "loss": 5.7966,
      "mean_token_accuracy": 0.17393025010824203,
      "num_tokens": 65665910.0,
      "step": 30340
    },
    {
      "entropy": 6.598823976516724,
      "epoch": 3.2472042377869337,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039708595980150025,
      "loss": 5.9044,
      "mean_token_accuracy": 0.16142098605632782,
      "num_tokens": 65677126.0,
      "step": 30345
    },
    {
      "entropy": 6.567230081558227,
      "epoch": 3.2477393118946973,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003970538424584834,
      "loss": 5.806,
      "mean_token_accuracy": 0.1659995883703232,
      "num_tokens": 65687609.0,
      "step": 30350
    },
    {
      "entropy": 6.558445835113526,
      "epoch": 3.2482743860024614,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039702172159117907,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17404680848121643,
      "num_tokens": 65699585.0,
      "step": 30355
    },
    {
      "entropy": 6.542458391189575,
      "epoch": 3.248809460110225,
      "grad_norm": 1.25,
      "learning_rate": 0.0003969895972005146,
      "loss": 5.8768,
      "mean_token_accuracy": 0.16828937232494354,
      "num_tokens": 65710455.0,
      "step": 30360
    },
    {
      "entropy": 6.574936389923096,
      "epoch": 3.249344534217989,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003969574692874177,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17026889473199844,
      "num_tokens": 65721797.0,
      "step": 30365
    },
    {
      "entropy": 6.462852430343628,
      "epoch": 3.2498796083257533,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00039692533785281605,
      "loss": 5.7041,
      "mean_token_accuracy": 0.18572403937578202,
      "num_tokens": 65733571.0,
      "step": 30370
    },
    {
      "entropy": 6.5036173343658445,
      "epoch": 3.250414682433517,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003968932028976373,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16809643059968948,
      "num_tokens": 65744501.0,
      "step": 30375
    },
    {
      "entropy": 6.500042247772217,
      "epoch": 3.250949756541281,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039686106442280946,
      "loss": 5.7738,
      "mean_token_accuracy": 0.1758960798382759,
      "num_tokens": 65755428.0,
      "step": 30380
    },
    {
      "entropy": 6.56920018196106,
      "epoch": 3.2514848306490447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003968289224292606,
      "loss": 5.7732,
      "mean_token_accuracy": 0.173471836745739,
      "num_tokens": 65765535.0,
      "step": 30385
    },
    {
      "entropy": 6.4971351146698,
      "epoch": 3.252019904756809,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039679677691791866,
      "loss": 5.7269,
      "mean_token_accuracy": 0.1749537080526352,
      "num_tokens": 65776084.0,
      "step": 30390
    },
    {
      "entropy": 6.421518039703369,
      "epoch": 3.252554978864573,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003967646278897119,
      "loss": 5.7118,
      "mean_token_accuracy": 0.17806440591812134,
      "num_tokens": 65787470.0,
      "step": 30395
    },
    {
      "entropy": 6.4760754108428955,
      "epoch": 3.2530900529723366,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003967324753455687,
      "loss": 5.7705,
      "mean_token_accuracy": 0.17750952541828155,
      "num_tokens": 65796942.0,
      "step": 30400
    },
    {
      "entropy": 6.40193190574646,
      "epoch": 3.2536251270801007,
      "grad_norm": 1.125,
      "learning_rate": 0.00039670031928641723,
      "loss": 5.7084,
      "mean_token_accuracy": 0.178384467959404,
      "num_tokens": 65806902.0,
      "step": 30405
    },
    {
      "entropy": 6.570516729354859,
      "epoch": 3.2541602011878643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039666815971318623,
      "loss": 5.7805,
      "mean_token_accuracy": 0.1742369845509529,
      "num_tokens": 65817689.0,
      "step": 30410
    },
    {
      "entropy": 6.546777009963989,
      "epoch": 3.2546952752956284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039663599662680414,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17639646977186202,
      "num_tokens": 65827995.0,
      "step": 30415
    },
    {
      "entropy": 6.5527314186096195,
      "epoch": 3.2552303494033925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003966038300281999,
      "loss": 5.8971,
      "mean_token_accuracy": 0.17234011739492416,
      "num_tokens": 65838449.0,
      "step": 30420
    },
    {
      "entropy": 6.506424236297607,
      "epoch": 3.255765423511156,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000396571659918302,
      "loss": 5.7543,
      "mean_token_accuracy": 0.18360019326210023,
      "num_tokens": 65847898.0,
      "step": 30425
    },
    {
      "entropy": 6.4501708984375,
      "epoch": 3.2563004976189203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003965394862980396,
      "loss": 5.6777,
      "mean_token_accuracy": 0.18355841785669327,
      "num_tokens": 65859876.0,
      "step": 30430
    },
    {
      "entropy": 6.555815172195435,
      "epoch": 3.2568355717266844,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003965073091683416,
      "loss": 5.7593,
      "mean_token_accuracy": 0.17628485560417176,
      "num_tokens": 65869875.0,
      "step": 30435
    },
    {
      "entropy": 6.492120170593262,
      "epoch": 3.257370645834448,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039647512853013703,
      "loss": 5.7505,
      "mean_token_accuracy": 0.18395413756370543,
      "num_tokens": 65881351.0,
      "step": 30440
    },
    {
      "entropy": 6.477956962585449,
      "epoch": 3.257905719942212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003964429443843552,
      "loss": 5.8227,
      "mean_token_accuracy": 0.17333156913518905,
      "num_tokens": 65892533.0,
      "step": 30445
    },
    {
      "entropy": 6.502191352844238,
      "epoch": 3.2584407940499758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039641075673192564,
      "loss": 5.7093,
      "mean_token_accuracy": 0.17154584378004073,
      "num_tokens": 65903754.0,
      "step": 30450
    },
    {
      "entropy": 6.468652677536011,
      "epoch": 3.25897586815774,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039637856557377734,
      "loss": 5.6373,
      "mean_token_accuracy": 0.1837247520685196,
      "num_tokens": 65915461.0,
      "step": 30455
    },
    {
      "entropy": 6.494789123535156,
      "epoch": 3.2595109422655035,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039634637091083997,
      "loss": 5.801,
      "mean_token_accuracy": 0.17311160117387772,
      "num_tokens": 65927108.0,
      "step": 30460
    },
    {
      "entropy": 6.526828002929688,
      "epoch": 3.2600460163732676,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039631417274404334,
      "loss": 5.7699,
      "mean_token_accuracy": 0.1722302883863449,
      "num_tokens": 65937965.0,
      "step": 30465
    },
    {
      "entropy": 6.482498216629028,
      "epoch": 3.2605810904810317,
      "grad_norm": 1.875,
      "learning_rate": 0.00039628197107431694,
      "loss": 5.7336,
      "mean_token_accuracy": 0.1817318841814995,
      "num_tokens": 65948393.0,
      "step": 30470
    },
    {
      "entropy": 6.480117893218994,
      "epoch": 3.2611161645887954,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039624976590259064,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17554758340120316,
      "num_tokens": 65958960.0,
      "step": 30475
    },
    {
      "entropy": 6.4995192527771,
      "epoch": 3.2616512386965595,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039621755722979446,
      "loss": 5.7465,
      "mean_token_accuracy": 0.17026008665561676,
      "num_tokens": 65970937.0,
      "step": 30480
    },
    {
      "entropy": 6.561894178390503,
      "epoch": 3.2621863128043236,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003961853450568583,
      "loss": 5.8105,
      "mean_token_accuracy": 0.17281523942947388,
      "num_tokens": 65982444.0,
      "step": 30485
    },
    {
      "entropy": 6.5826949119567875,
      "epoch": 3.2627213869120872,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039615312938471236,
      "loss": 5.7993,
      "mean_token_accuracy": 0.1705068364739418,
      "num_tokens": 65994392.0,
      "step": 30490
    },
    {
      "entropy": 6.5563091278076175,
      "epoch": 3.2632564610198513,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039612091021428683,
      "loss": 5.8318,
      "mean_token_accuracy": 0.17191933393478392,
      "num_tokens": 66004710.0,
      "step": 30495
    },
    {
      "entropy": 6.519987773895264,
      "epoch": 3.263791535127615,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003960886875465121,
      "loss": 5.8398,
      "mean_token_accuracy": 0.16744038909673692,
      "num_tokens": 66015402.0,
      "step": 30500
    },
    {
      "entropy": 6.480180501937866,
      "epoch": 3.264326609235379,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039605646138231835,
      "loss": 5.7521,
      "mean_token_accuracy": 0.1770490899682045,
      "num_tokens": 66026795.0,
      "step": 30505
    },
    {
      "entropy": 6.549743604660034,
      "epoch": 3.264861683343143,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039602423172263644,
      "loss": 5.7448,
      "mean_token_accuracy": 0.17897251844406128,
      "num_tokens": 66037339.0,
      "step": 30510
    },
    {
      "entropy": 6.539910221099854,
      "epoch": 3.265396757450907,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003959919985683968,
      "loss": 5.8009,
      "mean_token_accuracy": 0.17355556339025496,
      "num_tokens": 66047415.0,
      "step": 30515
    },
    {
      "entropy": 6.565978097915649,
      "epoch": 3.265931831558671,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039595976192053023,
      "loss": 5.8032,
      "mean_token_accuracy": 0.17865266054868698,
      "num_tokens": 66057475.0,
      "step": 30520
    },
    {
      "entropy": 6.570211553573609,
      "epoch": 3.2664669056664346,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003959275217799675,
      "loss": 5.7192,
      "mean_token_accuracy": 0.17497231662273408,
      "num_tokens": 66067018.0,
      "step": 30525
    },
    {
      "entropy": 6.470949649810791,
      "epoch": 3.2670019797741987,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003958952781476396,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17143121510744094,
      "num_tokens": 66077822.0,
      "step": 30530
    },
    {
      "entropy": 6.490734386444092,
      "epoch": 3.267537053881963,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003958630310244775,
      "loss": 5.7957,
      "mean_token_accuracy": 0.1760551229119301,
      "num_tokens": 66088849.0,
      "step": 30535
    },
    {
      "entropy": 6.594285392761231,
      "epoch": 3.2680721279897265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039583078041141237,
      "loss": 5.8122,
      "mean_token_accuracy": 0.17215616255998611,
      "num_tokens": 66098516.0,
      "step": 30540
    },
    {
      "entropy": 6.564117193222046,
      "epoch": 3.2686072020974906,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003957985263093754,
      "loss": 5.8131,
      "mean_token_accuracy": 0.17385694086551667,
      "num_tokens": 66109202.0,
      "step": 30545
    },
    {
      "entropy": 6.5495625019073485,
      "epoch": 3.2691422762052547,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000395766268719298,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17527477890253068,
      "num_tokens": 66119445.0,
      "step": 30550
    },
    {
      "entropy": 6.501364088058471,
      "epoch": 3.2696773503130183,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039573400764211157,
      "loss": 5.8154,
      "mean_token_accuracy": 0.17825432419776915,
      "num_tokens": 66129922.0,
      "step": 30555
    },
    {
      "entropy": 6.526250267028809,
      "epoch": 3.2702124244207824,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003957017430787476,
      "loss": 5.8329,
      "mean_token_accuracy": 0.1680040642619133,
      "num_tokens": 66141118.0,
      "step": 30560
    },
    {
      "entropy": 6.587417459487915,
      "epoch": 3.270747498528546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039566947503013786,
      "loss": 5.84,
      "mean_token_accuracy": 0.17780213356018065,
      "num_tokens": 66152415.0,
      "step": 30565
    },
    {
      "entropy": 6.4964008808135985,
      "epoch": 3.27128257263631,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003956372034972139,
      "loss": 5.7281,
      "mean_token_accuracy": 0.17787067592144012,
      "num_tokens": 66163086.0,
      "step": 30570
    },
    {
      "entropy": 6.476133155822754,
      "epoch": 3.271817646744074,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039560492848090764,
      "loss": 5.7994,
      "mean_token_accuracy": 0.17125921547412873,
      "num_tokens": 66173020.0,
      "step": 30575
    },
    {
      "entropy": 6.542879962921143,
      "epoch": 3.272352720851838,
      "grad_norm": 1.1875,
      "learning_rate": 0.000395572649982151,
      "loss": 5.792,
      "mean_token_accuracy": 0.17256194353103638,
      "num_tokens": 66183979.0,
      "step": 30580
    },
    {
      "entropy": 6.583937549591065,
      "epoch": 3.272887794959602,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039554036800187604,
      "loss": 5.7211,
      "mean_token_accuracy": 0.18100347518920898,
      "num_tokens": 66194703.0,
      "step": 30585
    },
    {
      "entropy": 6.525908708572388,
      "epoch": 3.2734228690673657,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039550808254101497,
      "loss": 5.7894,
      "mean_token_accuracy": 0.17182048708200454,
      "num_tokens": 66206041.0,
      "step": 30590
    },
    {
      "entropy": 6.519519233703614,
      "epoch": 3.27395794317513,
      "grad_norm": 1.125,
      "learning_rate": 0.00039547579360049977,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1718544915318489,
      "num_tokens": 66218197.0,
      "step": 30595
    },
    {
      "entropy": 6.541883754730224,
      "epoch": 3.274493017282894,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039544350118126315,
      "loss": 5.7825,
      "mean_token_accuracy": 0.1704266309738159,
      "num_tokens": 66229261.0,
      "step": 30600
    },
    {
      "entropy": 6.552662324905396,
      "epoch": 3.2750280913906575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003954112052842373,
      "loss": 5.7162,
      "mean_token_accuracy": 0.18575700670480727,
      "num_tokens": 66240162.0,
      "step": 30605
    },
    {
      "entropy": 6.44815583229065,
      "epoch": 3.2755631654984216,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003953789059103548,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1776721805334091,
      "num_tokens": 66250284.0,
      "step": 30610
    },
    {
      "entropy": 6.559983015060425,
      "epoch": 3.2760982396061853,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039534660306054836,
      "loss": 5.7972,
      "mean_token_accuracy": 0.17418185174465178,
      "num_tokens": 66261164.0,
      "step": 30615
    },
    {
      "entropy": 6.532353305816651,
      "epoch": 3.2766333137139494,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039531429673575057,
      "loss": 5.7382,
      "mean_token_accuracy": 0.1776348352432251,
      "num_tokens": 66272201.0,
      "step": 30620
    },
    {
      "entropy": 6.55341796875,
      "epoch": 3.2771683878217135,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039528198693689444,
      "loss": 5.8186,
      "mean_token_accuracy": 0.180548594892025,
      "num_tokens": 66281297.0,
      "step": 30625
    },
    {
      "entropy": 6.458272504806518,
      "epoch": 3.277703461929477,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003952496736649129,
      "loss": 5.7596,
      "mean_token_accuracy": 0.17097925394773483,
      "num_tokens": 66290990.0,
      "step": 30630
    },
    {
      "entropy": 6.517618894577026,
      "epoch": 3.2782385360372412,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003952173569207388,
      "loss": 5.7859,
      "mean_token_accuracy": 0.17030699402093888,
      "num_tokens": 66301971.0,
      "step": 30635
    },
    {
      "entropy": 6.5747908592224125,
      "epoch": 3.278773610145005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003951850367053054,
      "loss": 5.848,
      "mean_token_accuracy": 0.16590647101402284,
      "num_tokens": 66313576.0,
      "step": 30640
    },
    {
      "entropy": 6.6119684219360355,
      "epoch": 3.279308684252769,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003951527130195461,
      "loss": 5.8246,
      "mean_token_accuracy": 0.1677570939064026,
      "num_tokens": 66325367.0,
      "step": 30645
    },
    {
      "entropy": 6.5035358428955075,
      "epoch": 3.279843758360533,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039512038586439404,
      "loss": 5.7234,
      "mean_token_accuracy": 0.18213865607976915,
      "num_tokens": 66335686.0,
      "step": 30650
    },
    {
      "entropy": 6.4842182159423825,
      "epoch": 3.2803788324682968,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039508805524078263,
      "loss": 5.7363,
      "mean_token_accuracy": 0.181266126036644,
      "num_tokens": 66345963.0,
      "step": 30655
    },
    {
      "entropy": 6.50182147026062,
      "epoch": 3.280913906576061,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039505572114964554,
      "loss": 5.8196,
      "mean_token_accuracy": 0.17615783214569092,
      "num_tokens": 66357781.0,
      "step": 30660
    },
    {
      "entropy": 6.575131464004516,
      "epoch": 3.281448980683825,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003950233835919164,
      "loss": 5.8063,
      "mean_token_accuracy": 0.17856026738882064,
      "num_tokens": 66369360.0,
      "step": 30665
    },
    {
      "entropy": 6.498838901519775,
      "epoch": 3.2819840547915886,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003949910425685289,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17081296294927598,
      "num_tokens": 66381481.0,
      "step": 30670
    },
    {
      "entropy": 6.486058282852173,
      "epoch": 3.2825191288993527,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003949586980804169,
      "loss": 5.7454,
      "mean_token_accuracy": 0.17773913443088532,
      "num_tokens": 66391469.0,
      "step": 30675
    },
    {
      "entropy": 6.52692289352417,
      "epoch": 3.2830542030071164,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039492635012851443,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16788618713617326,
      "num_tokens": 66402094.0,
      "step": 30680
    },
    {
      "entropy": 6.541940307617187,
      "epoch": 3.2835892771148805,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039489399871375536,
      "loss": 5.759,
      "mean_token_accuracy": 0.17374165505170822,
      "num_tokens": 66412149.0,
      "step": 30685
    },
    {
      "entropy": 6.559129333496093,
      "epoch": 3.284124351222644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003948616438370739,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16826678663492203,
      "num_tokens": 66423502.0,
      "step": 30690
    },
    {
      "entropy": 6.566130018234253,
      "epoch": 3.2846594253304082,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003948292854994045,
      "loss": 5.7215,
      "mean_token_accuracy": 0.18049391955137253,
      "num_tokens": 66433684.0,
      "step": 30695
    },
    {
      "entropy": 6.430932903289795,
      "epoch": 3.2851944994381723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039479692370168116,
      "loss": 5.7107,
      "mean_token_accuracy": 0.1855793997645378,
      "num_tokens": 66444728.0,
      "step": 30700
    },
    {
      "entropy": 6.47430944442749,
      "epoch": 3.285729573545936,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039476455844483847,
      "loss": 5.7872,
      "mean_token_accuracy": 0.17503116577863692,
      "num_tokens": 66456117.0,
      "step": 30705
    },
    {
      "entropy": 6.514514923095703,
      "epoch": 3.2862646476537,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003947321897298111,
      "loss": 5.8189,
      "mean_token_accuracy": 0.17299216836690903,
      "num_tokens": 66466467.0,
      "step": 30710
    },
    {
      "entropy": 6.568634653091431,
      "epoch": 3.286799721761464,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003946998175575335,
      "loss": 5.8641,
      "mean_token_accuracy": 0.16405883878469468,
      "num_tokens": 66477702.0,
      "step": 30715
    },
    {
      "entropy": 6.547069072723389,
      "epoch": 3.287334795869228,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039466744192894046,
      "loss": 5.7829,
      "mean_token_accuracy": 0.179475037753582,
      "num_tokens": 66489359.0,
      "step": 30720
    },
    {
      "entropy": 6.516627311706543,
      "epoch": 3.287869869976992,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003946350628449669,
      "loss": 5.7199,
      "mean_token_accuracy": 0.18187360763549804,
      "num_tokens": 66500337.0,
      "step": 30725
    },
    {
      "entropy": 6.524885082244873,
      "epoch": 3.2884049440847556,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039460268030654767,
      "loss": 5.807,
      "mean_token_accuracy": 0.17781000435352326,
      "num_tokens": 66510010.0,
      "step": 30730
    },
    {
      "entropy": 6.518221473693847,
      "epoch": 3.2889400181925197,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039457029431461794,
      "loss": 5.7837,
      "mean_token_accuracy": 0.17688011527061462,
      "num_tokens": 66521570.0,
      "step": 30735
    },
    {
      "entropy": 6.527279090881348,
      "epoch": 3.2894750923002833,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003945379048701127,
      "loss": 5.7511,
      "mean_token_accuracy": 0.1842545598745346,
      "num_tokens": 66532825.0,
      "step": 30740
    },
    {
      "entropy": 6.554337120056152,
      "epoch": 3.2900101664080474,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039450551197396725,
      "loss": 5.7722,
      "mean_token_accuracy": 0.17492623925209044,
      "num_tokens": 66543265.0,
      "step": 30745
    },
    {
      "entropy": 6.560320615768433,
      "epoch": 3.2905452405158115,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039447311562711693,
      "loss": 5.9137,
      "mean_token_accuracy": 0.1614852726459503,
      "num_tokens": 66553589.0,
      "step": 30750
    },
    {
      "entropy": 6.487175369262696,
      "epoch": 3.291080314623575,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039444071583049717,
      "loss": 5.7687,
      "mean_token_accuracy": 0.17758281975984574,
      "num_tokens": 66564679.0,
      "step": 30755
    },
    {
      "entropy": 6.556028938293457,
      "epoch": 3.2916153887313393,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039440831258504363,
      "loss": 5.7882,
      "mean_token_accuracy": 0.17823981493711472,
      "num_tokens": 66576408.0,
      "step": 30760
    },
    {
      "entropy": 6.56840705871582,
      "epoch": 3.2921504628391034,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039437590589169177,
      "loss": 5.8586,
      "mean_token_accuracy": 0.16819994151592255,
      "num_tokens": 66587535.0,
      "step": 30765
    },
    {
      "entropy": 6.526656818389893,
      "epoch": 3.292685536946867,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039434349575137736,
      "loss": 5.8243,
      "mean_token_accuracy": 0.17253679186105728,
      "num_tokens": 66598493.0,
      "step": 30770
    },
    {
      "entropy": 6.5272198677062985,
      "epoch": 3.293220611054631,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003943110821650364,
      "loss": 5.7418,
      "mean_token_accuracy": 0.17470531463623046,
      "num_tokens": 66609366.0,
      "step": 30775
    },
    {
      "entropy": 6.494835424423218,
      "epoch": 3.293755685162395,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003942786651336046,
      "loss": 5.6718,
      "mean_token_accuracy": 0.18733179867267608,
      "num_tokens": 66619902.0,
      "step": 30780
    },
    {
      "entropy": 6.521139621734619,
      "epoch": 3.294290759270159,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003942462446580181,
      "loss": 5.8828,
      "mean_token_accuracy": 0.16206735670566558,
      "num_tokens": 66630629.0,
      "step": 30785
    },
    {
      "entropy": 6.55422945022583,
      "epoch": 3.294825833377923,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039421382073921306,
      "loss": 5.8224,
      "mean_token_accuracy": 0.1719958543777466,
      "num_tokens": 66640720.0,
      "step": 30790
    },
    {
      "entropy": 6.591320514678955,
      "epoch": 3.2953609074856867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003941813933781257,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1652525320649147,
      "num_tokens": 66653021.0,
      "step": 30795
    },
    {
      "entropy": 6.576294994354248,
      "epoch": 3.2958959815934508,
      "grad_norm": 1.125,
      "learning_rate": 0.0003941489625756923,
      "loss": 5.8035,
      "mean_token_accuracy": 0.17007844150066376,
      "num_tokens": 66663997.0,
      "step": 30800
    },
    {
      "entropy": 6.502322864532471,
      "epoch": 3.2964310557012144,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039411652833284944,
      "loss": 5.7652,
      "mean_token_accuracy": 0.16671838909387587,
      "num_tokens": 66674632.0,
      "step": 30805
    },
    {
      "entropy": 6.491587209701538,
      "epoch": 3.2969661298089785,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003940840906505336,
      "loss": 5.7869,
      "mean_token_accuracy": 0.17664833962917328,
      "num_tokens": 66685289.0,
      "step": 30810
    },
    {
      "entropy": 6.572604370117188,
      "epoch": 3.2975012039167426,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003940516495296813,
      "loss": 5.8343,
      "mean_token_accuracy": 0.17413963079452516,
      "num_tokens": 66696448.0,
      "step": 30815
    },
    {
      "entropy": 6.599546718597412,
      "epoch": 3.2980362780245063,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039401920497122934,
      "loss": 5.8317,
      "mean_token_accuracy": 0.17280329763889313,
      "num_tokens": 66706186.0,
      "step": 30820
    },
    {
      "entropy": 6.5582092761993405,
      "epoch": 3.2985713521322704,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039398675697611464,
      "loss": 5.8744,
      "mean_token_accuracy": 0.16326080411672592,
      "num_tokens": 66717586.0,
      "step": 30825
    },
    {
      "entropy": 6.548464775085449,
      "epoch": 3.2991064262400345,
      "grad_norm": 1.15625,
      "learning_rate": 0.000393954305545274,
      "loss": 5.7693,
      "mean_token_accuracy": 0.1714773342013359,
      "num_tokens": 66727985.0,
      "step": 30830
    },
    {
      "entropy": 6.50860824584961,
      "epoch": 3.299641500347798,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039392185067964444,
      "loss": 5.79,
      "mean_token_accuracy": 0.17132668942213058,
      "num_tokens": 66738966.0,
      "step": 30835
    },
    {
      "entropy": 6.468781852722168,
      "epoch": 3.3001765744555622,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039388939238016327,
      "loss": 5.7437,
      "mean_token_accuracy": 0.1799051970243454,
      "num_tokens": 66750442.0,
      "step": 30840
    },
    {
      "entropy": 6.541925096511841,
      "epoch": 3.300711648563326,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003938569306477675,
      "loss": 5.8441,
      "mean_token_accuracy": 0.1710111081600189,
      "num_tokens": 66761246.0,
      "step": 30845
    },
    {
      "entropy": 6.576329278945923,
      "epoch": 3.30124672267109,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039382446548339464,
      "loss": 5.8293,
      "mean_token_accuracy": 0.1705258458852768,
      "num_tokens": 66771711.0,
      "step": 30850
    },
    {
      "entropy": 6.428521251678466,
      "epoch": 3.3017817967788536,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003937919968879819,
      "loss": 5.7144,
      "mean_token_accuracy": 0.18568397760391236,
      "num_tokens": 66783185.0,
      "step": 30855
    },
    {
      "entropy": 6.5912223815917965,
      "epoch": 3.3023168708866177,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003937595248624671,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1694817677140236,
      "num_tokens": 66794541.0,
      "step": 30860
    },
    {
      "entropy": 6.545398616790772,
      "epoch": 3.302851944994382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003937270494077876,
      "loss": 5.7663,
      "mean_token_accuracy": 0.17346570193767546,
      "num_tokens": 66805617.0,
      "step": 30865
    },
    {
      "entropy": 6.55237135887146,
      "epoch": 3.3033870191021455,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039369457052488134,
      "loss": 5.7934,
      "mean_token_accuracy": 0.17692652046680452,
      "num_tokens": 66814784.0,
      "step": 30870
    },
    {
      "entropy": 6.512675619125366,
      "epoch": 3.3039220932099096,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039366208821468596,
      "loss": 5.8168,
      "mean_token_accuracy": 0.17613626271486282,
      "num_tokens": 66825416.0,
      "step": 30875
    },
    {
      "entropy": 6.515519523620606,
      "epoch": 3.3044571673176737,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003936296024781394,
      "loss": 5.7702,
      "mean_token_accuracy": 0.17493868321180345,
      "num_tokens": 66836471.0,
      "step": 30880
    },
    {
      "entropy": 6.475852918624878,
      "epoch": 3.3049922414254373,
      "grad_norm": 1.125,
      "learning_rate": 0.00039359711331617993,
      "loss": 5.7419,
      "mean_token_accuracy": 0.1798856183886528,
      "num_tokens": 66848911.0,
      "step": 30885
    },
    {
      "entropy": 6.5493754863739015,
      "epoch": 3.3055273155332014,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039356462072974533,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17417349964380263,
      "num_tokens": 66860095.0,
      "step": 30890
    },
    {
      "entropy": 6.421298313140869,
      "epoch": 3.306062389640965,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039353212471977394,
      "loss": 5.6693,
      "mean_token_accuracy": 0.18389239460229873,
      "num_tokens": 66872676.0,
      "step": 30895
    },
    {
      "entropy": 6.540817022323608,
      "epoch": 3.306597463748729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003934996252872042,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16817086189985275,
      "num_tokens": 66883686.0,
      "step": 30900
    },
    {
      "entropy": 6.492053556442261,
      "epoch": 3.3071325378564933,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003934671224329744,
      "loss": 5.8512,
      "mean_token_accuracy": 0.16984010934829713,
      "num_tokens": 66895971.0,
      "step": 30905
    },
    {
      "entropy": 6.55046820640564,
      "epoch": 3.307667611964257,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003934346161580231,
      "loss": 5.8279,
      "mean_token_accuracy": 0.17184566110372543,
      "num_tokens": 66907840.0,
      "step": 30910
    },
    {
      "entropy": 6.60574517250061,
      "epoch": 3.308202686072021,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039340210646328887,
      "loss": 5.8511,
      "mean_token_accuracy": 0.1640530377626419,
      "num_tokens": 66918956.0,
      "step": 30915
    },
    {
      "entropy": 6.49932632446289,
      "epoch": 3.3087377601797847,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003933695933497105,
      "loss": 5.7496,
      "mean_token_accuracy": 0.17904710322618483,
      "num_tokens": 66929653.0,
      "step": 30920
    },
    {
      "entropy": 6.482755851745606,
      "epoch": 3.309272834287549,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039333707681822664,
      "loss": 5.7619,
      "mean_token_accuracy": 0.180931855738163,
      "num_tokens": 66939268.0,
      "step": 30925
    },
    {
      "entropy": 6.441131925582885,
      "epoch": 3.309807908395313,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039330455686977645,
      "loss": 5.8172,
      "mean_token_accuracy": 0.17331265956163405,
      "num_tokens": 66949669.0,
      "step": 30930
    },
    {
      "entropy": 6.590733814239502,
      "epoch": 3.3103429825030766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003932720335052987,
      "loss": 5.7864,
      "mean_token_accuracy": 0.18112822920083999,
      "num_tokens": 66959882.0,
      "step": 30935
    },
    {
      "entropy": 6.559532165527344,
      "epoch": 3.3108780566108407,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003932395067257327,
      "loss": 5.8459,
      "mean_token_accuracy": 0.1664731666445732,
      "num_tokens": 66972848.0,
      "step": 30940
    },
    {
      "entropy": 6.496605443954468,
      "epoch": 3.3114131307186048,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003932069765320175,
      "loss": 5.7329,
      "mean_token_accuracy": 0.18280128687620162,
      "num_tokens": 66983289.0,
      "step": 30945
    },
    {
      "entropy": 6.511435174942017,
      "epoch": 3.3119482048263684,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039317444292509253,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17350761741399764,
      "num_tokens": 66994194.0,
      "step": 30950
    },
    {
      "entropy": 6.5306161403656,
      "epoch": 3.3124832789341325,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039314190590589696,
      "loss": 5.7702,
      "mean_token_accuracy": 0.1787925988435745,
      "num_tokens": 67005889.0,
      "step": 30955
    },
    {
      "entropy": 6.473460865020752,
      "epoch": 3.313018353041896,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003931093654753706,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1788189172744751,
      "num_tokens": 67017237.0,
      "step": 30960
    },
    {
      "entropy": 6.551612806320191,
      "epoch": 3.3135534271496603,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003930768216344529,
      "loss": 5.8093,
      "mean_token_accuracy": 0.17421575486660004,
      "num_tokens": 67027637.0,
      "step": 30965
    },
    {
      "entropy": 6.497594261169434,
      "epoch": 3.314088501257424,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039304427438408354,
      "loss": 5.8248,
      "mean_token_accuracy": 0.17168600708246232,
      "num_tokens": 67039039.0,
      "step": 30970
    },
    {
      "entropy": 6.539510202407837,
      "epoch": 3.314623575365188,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039301172372520235,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16912096738815308,
      "num_tokens": 67048944.0,
      "step": 30975
    },
    {
      "entropy": 6.503313732147217,
      "epoch": 3.315158649472952,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003929791696587492,
      "loss": 5.8022,
      "mean_token_accuracy": 0.17134348899126053,
      "num_tokens": 67059724.0,
      "step": 30980
    },
    {
      "entropy": 6.494256544113159,
      "epoch": 3.315693723580716,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003929466121856641,
      "loss": 5.7272,
      "mean_token_accuracy": 0.17650569677352906,
      "num_tokens": 67070128.0,
      "step": 30985
    },
    {
      "entropy": 6.58201150894165,
      "epoch": 3.31622879768848,
      "grad_norm": 1.359375,
      "learning_rate": 0.00039291405130688725,
      "loss": 5.8077,
      "mean_token_accuracy": 0.17063163965940475,
      "num_tokens": 67080739.0,
      "step": 30990
    },
    {
      "entropy": 6.5332245349884035,
      "epoch": 3.316763871796244,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039288148702335863,
      "loss": 5.7769,
      "mean_token_accuracy": 0.17281210124492646,
      "num_tokens": 67091020.0,
      "step": 30995
    },
    {
      "entropy": 6.478491544723511,
      "epoch": 3.3172989459040076,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003928489193360186,
      "loss": 5.7849,
      "mean_token_accuracy": 0.16921785473823547,
      "num_tokens": 67101201.0,
      "step": 31000
    },
    {
      "entropy": 6.501805543899536,
      "epoch": 3.3178340200117717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003928163482458078,
      "loss": 5.7999,
      "mean_token_accuracy": 0.18377386778593063,
      "num_tokens": 67112829.0,
      "step": 31005
    },
    {
      "entropy": 6.601340484619141,
      "epoch": 3.3183690941195354,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039278377375366624,
      "loss": 5.8368,
      "mean_token_accuracy": 0.17458169907331467,
      "num_tokens": 67123626.0,
      "step": 31010
    },
    {
      "entropy": 6.554849338531494,
      "epoch": 3.3189041682272995,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039275119586053487,
      "loss": 5.7979,
      "mean_token_accuracy": 0.17268275022506713,
      "num_tokens": 67134858.0,
      "step": 31015
    },
    {
      "entropy": 6.455008840560913,
      "epoch": 3.3194392423350636,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003927186145673543,
      "loss": 5.7066,
      "mean_token_accuracy": 0.17778818309307098,
      "num_tokens": 67144496.0,
      "step": 31020
    },
    {
      "entropy": 6.500628662109375,
      "epoch": 3.3199743164428273,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039268602987506526,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1782758504152298,
      "num_tokens": 67155443.0,
      "step": 31025
    },
    {
      "entropy": 6.597900295257569,
      "epoch": 3.3205093905505914,
      "grad_norm": 1.25,
      "learning_rate": 0.0003926534417846086,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16984661370515824,
      "num_tokens": 67167604.0,
      "step": 31030
    },
    {
      "entropy": 6.552270221710205,
      "epoch": 3.321044464658355,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039262085029692536,
      "loss": 5.7592,
      "mean_token_accuracy": 0.17379634231328964,
      "num_tokens": 67177915.0,
      "step": 31035
    },
    {
      "entropy": 6.507746934890747,
      "epoch": 3.321579538766119,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039258825541295666,
      "loss": 5.7521,
      "mean_token_accuracy": 0.18460785895586013,
      "num_tokens": 67188515.0,
      "step": 31040
    },
    {
      "entropy": 6.488238954544068,
      "epoch": 3.322114612873883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039255565713364356,
      "loss": 5.7929,
      "mean_token_accuracy": 0.1770969182252884,
      "num_tokens": 67199139.0,
      "step": 31045
    },
    {
      "entropy": 6.4858222007751465,
      "epoch": 3.322649686981647,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003925230554599273,
      "loss": 5.8113,
      "mean_token_accuracy": 0.16955623775720596,
      "num_tokens": 67210335.0,
      "step": 31050
    },
    {
      "entropy": 6.5519640922546385,
      "epoch": 3.323184761089411,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039249045039274947,
      "loss": 5.7955,
      "mean_token_accuracy": 0.1745443433523178,
      "num_tokens": 67220128.0,
      "step": 31055
    },
    {
      "entropy": 6.506452322006226,
      "epoch": 3.3237198351971746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003924578419330513,
      "loss": 5.752,
      "mean_token_accuracy": 0.18343378007411956,
      "num_tokens": 67231251.0,
      "step": 31060
    },
    {
      "entropy": 6.511073923110962,
      "epoch": 3.3242549093049387,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003924252300817745,
      "loss": 5.7948,
      "mean_token_accuracy": 0.17337830662727355,
      "num_tokens": 67243336.0,
      "step": 31065
    },
    {
      "entropy": 6.560128784179687,
      "epoch": 3.324789983412703,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003923926148398606,
      "loss": 5.7824,
      "mean_token_accuracy": 0.17142828851938247,
      "num_tokens": 67256210.0,
      "step": 31070
    },
    {
      "entropy": 6.549068021774292,
      "epoch": 3.3253250575204665,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003923599962082516,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17599838823080063,
      "num_tokens": 67266271.0,
      "step": 31075
    },
    {
      "entropy": 6.526217603683472,
      "epoch": 3.3258601316282306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039232737418788905,
      "loss": 5.8034,
      "mean_token_accuracy": 0.1767657220363617,
      "num_tokens": 67278206.0,
      "step": 31080
    },
    {
      "entropy": 6.548417568206787,
      "epoch": 3.3263952057359942,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003922947487797151,
      "loss": 5.7861,
      "mean_token_accuracy": 0.17552710473537445,
      "num_tokens": 67289002.0,
      "step": 31085
    },
    {
      "entropy": 6.535490989685059,
      "epoch": 3.3269302798437583,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039226211998467177,
      "loss": 5.8394,
      "mean_token_accuracy": 0.17280467450618744,
      "num_tokens": 67300219.0,
      "step": 31090
    },
    {
      "entropy": 6.589786243438721,
      "epoch": 3.3274653539515224,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039222948780370126,
      "loss": 5.8739,
      "mean_token_accuracy": 0.1703687861561775,
      "num_tokens": 67311611.0,
      "step": 31095
    },
    {
      "entropy": 6.50086236000061,
      "epoch": 3.328000428059286,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003921968522377456,
      "loss": 5.828,
      "mean_token_accuracy": 0.17853648960590363,
      "num_tokens": 67322674.0,
      "step": 31100
    },
    {
      "entropy": 6.537430191040039,
      "epoch": 3.32853550216705,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003921642132877474,
      "loss": 5.9066,
      "mean_token_accuracy": 0.16887193620204927,
      "num_tokens": 67333574.0,
      "step": 31105
    },
    {
      "entropy": 6.593764781951904,
      "epoch": 3.3290705762748143,
      "grad_norm": 1.234375,
      "learning_rate": 0.000392131570954649,
      "loss": 5.7516,
      "mean_token_accuracy": 0.17375936657190322,
      "num_tokens": 67343500.0,
      "step": 31110
    },
    {
      "entropy": 6.562806797027588,
      "epoch": 3.329605650382578,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000392098925239393,
      "loss": 5.8343,
      "mean_token_accuracy": 0.17330351322889329,
      "num_tokens": 67354108.0,
      "step": 31115
    },
    {
      "entropy": 6.528130197525025,
      "epoch": 3.330140724490342,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039206627614292195,
      "loss": 5.8273,
      "mean_token_accuracy": 0.17021028846502304,
      "num_tokens": 67366177.0,
      "step": 31120
    },
    {
      "entropy": 6.53040657043457,
      "epoch": 3.3306757985981057,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003920336236661786,
      "loss": 5.8077,
      "mean_token_accuracy": 0.17919714301824569,
      "num_tokens": 67377000.0,
      "step": 31125
    },
    {
      "entropy": 6.485559558868408,
      "epoch": 3.33121087270587,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039200096781010586,
      "loss": 5.7473,
      "mean_token_accuracy": 0.1762051373720169,
      "num_tokens": 67388052.0,
      "step": 31130
    },
    {
      "entropy": 6.504048633575439,
      "epoch": 3.3317459468136335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003919683085756466,
      "loss": 5.7802,
      "mean_token_accuracy": 0.18248579800128936,
      "num_tokens": 67398701.0,
      "step": 31135
    },
    {
      "entropy": 6.5079937934875485,
      "epoch": 3.3322810209213976,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039193564596374386,
      "loss": 5.7885,
      "mean_token_accuracy": 0.17248773574829102,
      "num_tokens": 67410207.0,
      "step": 31140
    },
    {
      "entropy": 6.538727331161499,
      "epoch": 3.3328160950291617,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003919029799753408,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17321096509695053,
      "num_tokens": 67420883.0,
      "step": 31145
    },
    {
      "entropy": 6.494428396224976,
      "epoch": 3.3333511691369253,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003918703106113806,
      "loss": 5.757,
      "mean_token_accuracy": 0.17772935032844545,
      "num_tokens": 67430558.0,
      "step": 31150
    },
    {
      "entropy": 6.544112491607666,
      "epoch": 3.3338862432446894,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003918376378728066,
      "loss": 5.7913,
      "mean_token_accuracy": 0.17346956729888915,
      "num_tokens": 67441774.0,
      "step": 31155
    },
    {
      "entropy": 6.550991773605347,
      "epoch": 3.3344213173524535,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039180496176056224,
      "loss": 5.7873,
      "mean_token_accuracy": 0.17403036057949067,
      "num_tokens": 67452176.0,
      "step": 31160
    },
    {
      "entropy": 6.512957620620727,
      "epoch": 3.334956391460217,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039177228227559095,
      "loss": 5.6893,
      "mean_token_accuracy": 0.18752127438783645,
      "num_tokens": 67462427.0,
      "step": 31165
    },
    {
      "entropy": 6.524667024612427,
      "epoch": 3.3354914655679813,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039173959941883645,
      "loss": 5.8241,
      "mean_token_accuracy": 0.17459972947835922,
      "num_tokens": 67472611.0,
      "step": 31170
    },
    {
      "entropy": 6.513409423828125,
      "epoch": 3.336026539675745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039170691319124253,
      "loss": 5.8294,
      "mean_token_accuracy": 0.1618373617529869,
      "num_tokens": 67483138.0,
      "step": 31175
    },
    {
      "entropy": 6.520005559921264,
      "epoch": 3.336561613783509,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003916742235937528,
      "loss": 5.7665,
      "mean_token_accuracy": 0.1758747175335884,
      "num_tokens": 67493962.0,
      "step": 31180
    },
    {
      "entropy": 6.460419321060181,
      "epoch": 3.337096687891273,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003916415306273112,
      "loss": 5.759,
      "mean_token_accuracy": 0.17330896407365798,
      "num_tokens": 67506882.0,
      "step": 31185
    },
    {
      "entropy": 6.565799140930176,
      "epoch": 3.3376317619990368,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039160883429286195,
      "loss": 5.7922,
      "mean_token_accuracy": 0.1775461345911026,
      "num_tokens": 67517357.0,
      "step": 31190
    },
    {
      "entropy": 6.554059982299805,
      "epoch": 3.338166836106801,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039157613459134884,
      "loss": 5.8095,
      "mean_token_accuracy": 0.1702830746769905,
      "num_tokens": 67528480.0,
      "step": 31195
    },
    {
      "entropy": 6.547389030456543,
      "epoch": 3.3387019102145645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003915434315237163,
      "loss": 5.8722,
      "mean_token_accuracy": 0.17138558030128478,
      "num_tokens": 67539288.0,
      "step": 31200
    },
    {
      "entropy": 6.485790538787842,
      "epoch": 3.3392369843223286,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003915107250909085,
      "loss": 5.7125,
      "mean_token_accuracy": 0.17998767346143724,
      "num_tokens": 67551105.0,
      "step": 31205
    },
    {
      "entropy": 6.538157367706299,
      "epoch": 3.3397720584300927,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039147801529386997,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16643838733434677,
      "num_tokens": 67562421.0,
      "step": 31210
    },
    {
      "entropy": 6.519621276855469,
      "epoch": 3.3403071325378564,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039144530213354505,
      "loss": 5.8126,
      "mean_token_accuracy": 0.17140355408191682,
      "num_tokens": 67574134.0,
      "step": 31215
    },
    {
      "entropy": 6.511776971817016,
      "epoch": 3.3408422066456205,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003914125856108785,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17649825662374496,
      "num_tokens": 67585557.0,
      "step": 31220
    },
    {
      "entropy": 6.493662929534912,
      "epoch": 3.3413772807533846,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003913798657268148,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17185766845941544,
      "num_tokens": 67596854.0,
      "step": 31225
    },
    {
      "entropy": 6.609411954879761,
      "epoch": 3.3419123548611482,
      "grad_norm": 1.125,
      "learning_rate": 0.0003913471424822988,
      "loss": 5.7946,
      "mean_token_accuracy": 0.175542651116848,
      "num_tokens": 67607282.0,
      "step": 31230
    },
    {
      "entropy": 6.513317823410034,
      "epoch": 3.3424474289689123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003913144158782754,
      "loss": 5.7715,
      "mean_token_accuracy": 0.17507003098726273,
      "num_tokens": 67618228.0,
      "step": 31235
    },
    {
      "entropy": 6.420911550521851,
      "epoch": 3.342982503076676,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039128168591568975,
      "loss": 5.7663,
      "mean_token_accuracy": 0.1812153086066246,
      "num_tokens": 67630618.0,
      "step": 31240
    },
    {
      "entropy": 6.532337617874146,
      "epoch": 3.34351757718444,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003912489525954866,
      "loss": 5.8363,
      "mean_token_accuracy": 0.17622057795524598,
      "num_tokens": 67641052.0,
      "step": 31245
    },
    {
      "entropy": 6.531067895889282,
      "epoch": 3.3440526512922037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039121621591861134,
      "loss": 5.7833,
      "mean_token_accuracy": 0.17141319811344147,
      "num_tokens": 67651315.0,
      "step": 31250
    },
    {
      "entropy": 6.556587886810303,
      "epoch": 3.344587725399968,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039118347588600917,
      "loss": 5.8131,
      "mean_token_accuracy": 0.17195579260587693,
      "num_tokens": 67661637.0,
      "step": 31255
    },
    {
      "entropy": 6.445059204101563,
      "epoch": 3.345122799507732,
      "grad_norm": 1.125,
      "learning_rate": 0.0003911507324986254,
      "loss": 5.7029,
      "mean_token_accuracy": 0.18515167236328126,
      "num_tokens": 67672863.0,
      "step": 31260
    },
    {
      "entropy": 6.526938629150391,
      "epoch": 3.3456578736154956,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003911179857574056,
      "loss": 5.7912,
      "mean_token_accuracy": 0.17994380742311478,
      "num_tokens": 67683711.0,
      "step": 31265
    },
    {
      "entropy": 6.542290735244751,
      "epoch": 3.3461929477232597,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003910852356632953,
      "loss": 5.7319,
      "mean_token_accuracy": 0.18161020129919053,
      "num_tokens": 67694345.0,
      "step": 31270
    },
    {
      "entropy": 6.581700181961059,
      "epoch": 3.346728021831024,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003910524822172402,
      "loss": 5.8125,
      "mean_token_accuracy": 0.17515604943037033,
      "num_tokens": 67705339.0,
      "step": 31275
    },
    {
      "entropy": 6.489808559417725,
      "epoch": 3.3472630959387875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003910197254201858,
      "loss": 5.744,
      "mean_token_accuracy": 0.17294389754533768,
      "num_tokens": 67715800.0,
      "step": 31280
    },
    {
      "entropy": 6.56314263343811,
      "epoch": 3.3477981700465516,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003909869652730783,
      "loss": 5.9013,
      "mean_token_accuracy": 0.17451069504022598,
      "num_tokens": 67725733.0,
      "step": 31285
    },
    {
      "entropy": 6.525778675079346,
      "epoch": 3.348333244154315,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00039095420177686334,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17703335881233215,
      "num_tokens": 67737637.0,
      "step": 31290
    },
    {
      "entropy": 6.551197957992554,
      "epoch": 3.3488683182620793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003909214349324872,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17788169980049134,
      "num_tokens": 67748016.0,
      "step": 31295
    },
    {
      "entropy": 6.536272716522217,
      "epoch": 3.3494033923698434,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039088866474089583,
      "loss": 5.8185,
      "mean_token_accuracy": 0.1732855811715126,
      "num_tokens": 67758439.0,
      "step": 31300
    },
    {
      "entropy": 6.557870531082154,
      "epoch": 3.349938466477607,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039085589120303564,
      "loss": 5.7684,
      "mean_token_accuracy": 0.1744239568710327,
      "num_tokens": 67768978.0,
      "step": 31305
    },
    {
      "entropy": 6.435952997207641,
      "epoch": 3.350473540585371,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003908231143198529,
      "loss": 5.7795,
      "mean_token_accuracy": 0.18104429841041564,
      "num_tokens": 67779557.0,
      "step": 31310
    },
    {
      "entropy": 6.5819573402404785,
      "epoch": 3.351008614693135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039079033409229394,
      "loss": 5.8382,
      "mean_token_accuracy": 0.17001950591802598,
      "num_tokens": 67791561.0,
      "step": 31315
    },
    {
      "entropy": 6.551253318786621,
      "epoch": 3.351543688800899,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003907575505213053,
      "loss": 5.7524,
      "mean_token_accuracy": 0.1733856186270714,
      "num_tokens": 67801703.0,
      "step": 31320
    },
    {
      "entropy": 6.561055517196655,
      "epoch": 3.352078762908663,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003907247636078337,
      "loss": 5.8279,
      "mean_token_accuracy": 0.17553925067186354,
      "num_tokens": 67812413.0,
      "step": 31325
    },
    {
      "entropy": 6.568955564498902,
      "epoch": 3.3526138370164267,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039069197335282583,
      "loss": 5.7919,
      "mean_token_accuracy": 0.17597485333681107,
      "num_tokens": 67823184.0,
      "step": 31330
    },
    {
      "entropy": 6.587539768218994,
      "epoch": 3.353148911124191,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039065917975722854,
      "loss": 5.8185,
      "mean_token_accuracy": 0.17378318607807158,
      "num_tokens": 67834372.0,
      "step": 31335
    },
    {
      "entropy": 6.488184309005737,
      "epoch": 3.3536839852319544,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003906263828219886,
      "loss": 5.6953,
      "mean_token_accuracy": 0.17794273942708969,
      "num_tokens": 67845324.0,
      "step": 31340
    },
    {
      "entropy": 6.553366088867188,
      "epoch": 3.3542190593397185,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003905935825480531,
      "loss": 5.7373,
      "mean_token_accuracy": 0.17474237829446793,
      "num_tokens": 67856140.0,
      "step": 31345
    },
    {
      "entropy": 6.58642168045044,
      "epoch": 3.3547541334474826,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039056077893636916,
      "loss": 5.8655,
      "mean_token_accuracy": 0.17426013201475143,
      "num_tokens": 67867117.0,
      "step": 31350
    },
    {
      "entropy": 6.583284378051758,
      "epoch": 3.3552892075552463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039052797198788406,
      "loss": 5.8748,
      "mean_token_accuracy": 0.16928319185972213,
      "num_tokens": 67877006.0,
      "step": 31355
    },
    {
      "entropy": 6.471544170379639,
      "epoch": 3.3558242816630104,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003904951617035448,
      "loss": 5.6585,
      "mean_token_accuracy": 0.1905757188796997,
      "num_tokens": 67886873.0,
      "step": 31360
    },
    {
      "entropy": 6.569270133972168,
      "epoch": 3.356359355770774,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003904623480842992,
      "loss": 5.8375,
      "mean_token_accuracy": 0.17426177710294724,
      "num_tokens": 67897033.0,
      "step": 31365
    },
    {
      "entropy": 6.548561048507691,
      "epoch": 3.356894429878538,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003904295311310944,
      "loss": 5.7459,
      "mean_token_accuracy": 0.1787040948867798,
      "num_tokens": 67909030.0,
      "step": 31370
    },
    {
      "entropy": 6.575655126571656,
      "epoch": 3.3574295039863022,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003903967108448782,
      "loss": 5.8484,
      "mean_token_accuracy": 0.17500828057527543,
      "num_tokens": 67918736.0,
      "step": 31375
    },
    {
      "entropy": 6.457413101196289,
      "epoch": 3.357964578094066,
      "grad_norm": 1.3125,
      "learning_rate": 0.00039036388722659803,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1793823927640915,
      "num_tokens": 67929392.0,
      "step": 31380
    },
    {
      "entropy": 6.590271615982056,
      "epoch": 3.35849965220183,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000390331060277202,
      "loss": 5.8711,
      "mean_token_accuracy": 0.16455209702253343,
      "num_tokens": 67940634.0,
      "step": 31385
    },
    {
      "entropy": 6.5361988067626955,
      "epoch": 3.359034726309594,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003902982299976377,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17372827529907225,
      "num_tokens": 67951132.0,
      "step": 31390
    },
    {
      "entropy": 6.501071739196777,
      "epoch": 3.3595698004173578,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039026539638885326,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16242669373750687,
      "num_tokens": 67961651.0,
      "step": 31395
    },
    {
      "entropy": 6.4185630798339846,
      "epoch": 3.360104874525122,
      "grad_norm": 1.375,
      "learning_rate": 0.0003902325594517967,
      "loss": 5.6997,
      "mean_token_accuracy": 0.17974525243043898,
      "num_tokens": 67971906.0,
      "step": 31400
    },
    {
      "entropy": 6.551366853713989,
      "epoch": 3.3606399486328855,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039019971918741615,
      "loss": 5.8131,
      "mean_token_accuracy": 0.1779598042368889,
      "num_tokens": 67983059.0,
      "step": 31405
    },
    {
      "entropy": 6.5289918899536135,
      "epoch": 3.3611750227406496,
      "grad_norm": 1.375,
      "learning_rate": 0.00039016687559665985,
      "loss": 5.8008,
      "mean_token_accuracy": 0.17077480405569076,
      "num_tokens": 67993635.0,
      "step": 31410
    },
    {
      "entropy": 6.624131774902343,
      "epoch": 3.3617100968484133,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003901340286804763,
      "loss": 5.8095,
      "mean_token_accuracy": 0.1782313734292984,
      "num_tokens": 68003857.0,
      "step": 31415
    },
    {
      "entropy": 6.54319200515747,
      "epoch": 3.3622451709561774,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039010117843981377,
      "loss": 5.7804,
      "mean_token_accuracy": 0.1709308311343193,
      "num_tokens": 68014993.0,
      "step": 31420
    },
    {
      "entropy": 6.547025871276856,
      "epoch": 3.3627802450639415,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039006832487562096,
      "loss": 5.7988,
      "mean_token_accuracy": 0.17699144333600997,
      "num_tokens": 68026245.0,
      "step": 31425
    },
    {
      "entropy": 6.572801685333252,
      "epoch": 3.363315319171705,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003900354679888464,
      "loss": 5.8462,
      "mean_token_accuracy": 0.17546674609184265,
      "num_tokens": 68038023.0,
      "step": 31430
    },
    {
      "entropy": 6.57564115524292,
      "epoch": 3.363850393279469,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003900026077804389,
      "loss": 5.8113,
      "mean_token_accuracy": 0.17184728980064393,
      "num_tokens": 68048527.0,
      "step": 31435
    },
    {
      "entropy": 6.542205047607422,
      "epoch": 3.3643854673872333,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003899697442513472,
      "loss": 5.8,
      "mean_token_accuracy": 0.16523377299308778,
      "num_tokens": 68059345.0,
      "step": 31440
    },
    {
      "entropy": 6.476750040054322,
      "epoch": 3.364920541494997,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038993687740252034,
      "loss": 5.7923,
      "mean_token_accuracy": 0.16932451277971267,
      "num_tokens": 68070295.0,
      "step": 31445
    },
    {
      "entropy": 6.493661212921142,
      "epoch": 3.365455615602761,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003899040072349073,
      "loss": 5.7383,
      "mean_token_accuracy": 0.1768973723053932,
      "num_tokens": 68081665.0,
      "step": 31450
    },
    {
      "entropy": 6.507287693023682,
      "epoch": 3.3659906897105247,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003898711337494571,
      "loss": 5.7344,
      "mean_token_accuracy": 0.1807692155241966,
      "num_tokens": 68092474.0,
      "step": 31455
    },
    {
      "entropy": 6.582113742828369,
      "epoch": 3.366525763818289,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003898382569471191,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16545862704515457,
      "num_tokens": 68102981.0,
      "step": 31460
    },
    {
      "entropy": 6.537043237686158,
      "epoch": 3.367060837926053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038980537682884266,
      "loss": 5.7347,
      "mean_token_accuracy": 0.17701234966516494,
      "num_tokens": 68113785.0,
      "step": 31465
    },
    {
      "entropy": 6.51550703048706,
      "epoch": 3.3675959120338166,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038977249339557703,
      "loss": 5.8356,
      "mean_token_accuracy": 0.17468522042036055,
      "num_tokens": 68123419.0,
      "step": 31470
    },
    {
      "entropy": 6.464790534973145,
      "epoch": 3.3681309861415807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038973960664827177,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17545437663793564,
      "num_tokens": 68134816.0,
      "step": 31475
    },
    {
      "entropy": 6.577698707580566,
      "epoch": 3.3686660602493443,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038970671658787647,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17401240915060043,
      "num_tokens": 68146063.0,
      "step": 31480
    },
    {
      "entropy": 6.554414844512939,
      "epoch": 3.3692011343571084,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038967382321534094,
      "loss": 5.8452,
      "mean_token_accuracy": 0.17545505315065385,
      "num_tokens": 68156735.0,
      "step": 31485
    },
    {
      "entropy": 6.542793130874633,
      "epoch": 3.3697362084648725,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038964092653161474,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16973523497581483,
      "num_tokens": 68167768.0,
      "step": 31490
    },
    {
      "entropy": 6.5247596263885494,
      "epoch": 3.370271282572636,
      "grad_norm": 1.140625,
      "learning_rate": 0.000389608026537648,
      "loss": 5.6922,
      "mean_token_accuracy": 0.18696407973766327,
      "num_tokens": 68179325.0,
      "step": 31495
    },
    {
      "entropy": 6.539525175094605,
      "epoch": 3.3708063566804003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003895751232343906,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16998634934425355,
      "num_tokens": 68190615.0,
      "step": 31500
    },
    {
      "entropy": 6.543503093719482,
      "epoch": 3.3713414307881644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003895422166227926,
      "loss": 5.776,
      "mean_token_accuracy": 0.1729315012693405,
      "num_tokens": 68201323.0,
      "step": 31505
    },
    {
      "entropy": 6.576150894165039,
      "epoch": 3.371876504895928,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003895093067038042,
      "loss": 5.7968,
      "mean_token_accuracy": 0.17591637074947358,
      "num_tokens": 68213296.0,
      "step": 31510
    },
    {
      "entropy": 6.532728815078736,
      "epoch": 3.372411579003692,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003894763934783757,
      "loss": 5.8603,
      "mean_token_accuracy": 0.1646134912967682,
      "num_tokens": 68224707.0,
      "step": 31515
    },
    {
      "entropy": 6.57660665512085,
      "epoch": 3.372946653111456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003894434769474573,
      "loss": 5.8648,
      "mean_token_accuracy": 0.1688946008682251,
      "num_tokens": 68236651.0,
      "step": 31520
    },
    {
      "entropy": 6.606281995773315,
      "epoch": 3.37348172721922,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003894105571119996,
      "loss": 5.837,
      "mean_token_accuracy": 0.16779446303844453,
      "num_tokens": 68246933.0,
      "step": 31525
    },
    {
      "entropy": 6.5445271015167235,
      "epoch": 3.3740168013269836,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038937763397295323,
      "loss": 5.8066,
      "mean_token_accuracy": 0.1739519104361534,
      "num_tokens": 68257934.0,
      "step": 31530
    },
    {
      "entropy": 6.5331813335418705,
      "epoch": 3.3745518754347477,
      "grad_norm": 1.125,
      "learning_rate": 0.00038934470753126867,
      "loss": 5.7472,
      "mean_token_accuracy": 0.17695090174674988,
      "num_tokens": 68268841.0,
      "step": 31535
    },
    {
      "entropy": 6.564783811569214,
      "epoch": 3.3750869495425118,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038931177778789685,
      "loss": 5.79,
      "mean_token_accuracy": 0.17549800574779512,
      "num_tokens": 68280391.0,
      "step": 31540
    },
    {
      "entropy": 6.4929327964782715,
      "epoch": 3.3756220236502754,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003892788447437885,
      "loss": 5.7596,
      "mean_token_accuracy": 0.17777176648378373,
      "num_tokens": 68291886.0,
      "step": 31545
    },
    {
      "entropy": 6.500685214996338,
      "epoch": 3.3761570977580395,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038924590839989445,
      "loss": 5.7824,
      "mean_token_accuracy": 0.16870436519384385,
      "num_tokens": 68302909.0,
      "step": 31550
    },
    {
      "entropy": 6.516185665130616,
      "epoch": 3.3766921718658036,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038921296875716596,
      "loss": 5.8514,
      "mean_token_accuracy": 0.16688707917928697,
      "num_tokens": 68314204.0,
      "step": 31555
    },
    {
      "entropy": 6.507393407821655,
      "epoch": 3.3772272459735673,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038918002581655406,
      "loss": 5.7778,
      "mean_token_accuracy": 0.1754065528512001,
      "num_tokens": 68325861.0,
      "step": 31560
    },
    {
      "entropy": 6.632211971282959,
      "epoch": 3.3777623200813314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003891470795790098,
      "loss": 5.8287,
      "mean_token_accuracy": 0.1726941481232643,
      "num_tokens": 68336390.0,
      "step": 31565
    },
    {
      "entropy": 6.556076526641846,
      "epoch": 3.378297394189095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038911413004548477,
      "loss": 5.7793,
      "mean_token_accuracy": 0.17195264101028443,
      "num_tokens": 68347429.0,
      "step": 31570
    },
    {
      "entropy": 6.526967525482178,
      "epoch": 3.378832468296859,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003890811772169302,
      "loss": 5.8804,
      "mean_token_accuracy": 0.16870694160461425,
      "num_tokens": 68358417.0,
      "step": 31575
    },
    {
      "entropy": 6.537586069107055,
      "epoch": 3.3793675424046232,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038904822109429773,
      "loss": 5.7371,
      "mean_token_accuracy": 0.1786951467394829,
      "num_tokens": 68369495.0,
      "step": 31580
    },
    {
      "entropy": 6.504385280609131,
      "epoch": 3.379902616512387,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003890152616785389,
      "loss": 5.8113,
      "mean_token_accuracy": 0.1703680634498596,
      "num_tokens": 68380122.0,
      "step": 31585
    },
    {
      "entropy": 6.504606628417969,
      "epoch": 3.380437690620151,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038898229897060544,
      "loss": 5.7733,
      "mean_token_accuracy": 0.181170055270195,
      "num_tokens": 68389997.0,
      "step": 31590
    },
    {
      "entropy": 6.595870304107666,
      "epoch": 3.3809727647279146,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038894933297144906,
      "loss": 5.8166,
      "mean_token_accuracy": 0.1679777607321739,
      "num_tokens": 68399726.0,
      "step": 31595
    },
    {
      "entropy": 6.532872009277344,
      "epoch": 3.3815078388356787,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038891636368202173,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17596345245838166,
      "num_tokens": 68410022.0,
      "step": 31600
    },
    {
      "entropy": 6.329182529449463,
      "epoch": 3.382042912943443,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038888339110327544,
      "loss": 5.586,
      "mean_token_accuracy": 0.20846110880374907,
      "num_tokens": 68421269.0,
      "step": 31605
    },
    {
      "entropy": 6.394658327102661,
      "epoch": 3.3825779870512065,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038885041523616223,
      "loss": 5.73,
      "mean_token_accuracy": 0.18938157260417937,
      "num_tokens": 68432979.0,
      "step": 31610
    },
    {
      "entropy": 6.545127010345459,
      "epoch": 3.3831130611589706,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038881743608163414,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17627827674150467,
      "num_tokens": 68443379.0,
      "step": 31615
    },
    {
      "entropy": 6.5523864269256595,
      "epoch": 3.3836481352667347,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038878445364064375,
      "loss": 5.7848,
      "mean_token_accuracy": 0.17113473564386367,
      "num_tokens": 68454277.0,
      "step": 31620
    },
    {
      "entropy": 6.530547761917115,
      "epoch": 3.3841832093744983,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038875146791414315,
      "loss": 5.7862,
      "mean_token_accuracy": 0.1728412240743637,
      "num_tokens": 68465774.0,
      "step": 31625
    },
    {
      "entropy": 6.549650430679321,
      "epoch": 3.3847182834822624,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038871847890308497,
      "loss": 5.7952,
      "mean_token_accuracy": 0.175409497320652,
      "num_tokens": 68476083.0,
      "step": 31630
    },
    {
      "entropy": 6.542690229415894,
      "epoch": 3.385253357590026,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003886854866084216,
      "loss": 5.8321,
      "mean_token_accuracy": 0.16828635483980178,
      "num_tokens": 68486696.0,
      "step": 31635
    },
    {
      "entropy": 6.546208333969116,
      "epoch": 3.38578843169779,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003886524910311058,
      "loss": 5.8239,
      "mean_token_accuracy": 0.16811105012893676,
      "num_tokens": 68497458.0,
      "step": 31640
    },
    {
      "entropy": 6.53323712348938,
      "epoch": 3.386323505805554,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003886194921720904,
      "loss": 5.8033,
      "mean_token_accuracy": 0.17233773767948152,
      "num_tokens": 68508603.0,
      "step": 31645
    },
    {
      "entropy": 6.561120700836182,
      "epoch": 3.386858579913318,
      "grad_norm": 1.234375,
      "learning_rate": 0.000388586490032328,
      "loss": 5.777,
      "mean_token_accuracy": 0.16946887373924255,
      "num_tokens": 68519389.0,
      "step": 31650
    },
    {
      "entropy": 6.550912189483642,
      "epoch": 3.387393654021082,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003885534846127717,
      "loss": 5.6943,
      "mean_token_accuracy": 0.181453675031662,
      "num_tokens": 68529272.0,
      "step": 31655
    },
    {
      "entropy": 6.570882987976074,
      "epoch": 3.3879287281288457,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003885204759143746,
      "loss": 5.8028,
      "mean_token_accuracy": 0.1730898931622505,
      "num_tokens": 68539644.0,
      "step": 31660
    },
    {
      "entropy": 6.497036361694336,
      "epoch": 3.38846380223661,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003884874639380895,
      "loss": 5.7548,
      "mean_token_accuracy": 0.1786083459854126,
      "num_tokens": 68549601.0,
      "step": 31665
    },
    {
      "entropy": 6.4596545696258545,
      "epoch": 3.388998876344374,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038845444868486996,
      "loss": 5.817,
      "mean_token_accuracy": 0.18328086137771607,
      "num_tokens": 68561105.0,
      "step": 31670
    },
    {
      "entropy": 6.583676195144653,
      "epoch": 3.3895339504521376,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003884214301556692,
      "loss": 5.9009,
      "mean_token_accuracy": 0.16285618543624877,
      "num_tokens": 68572117.0,
      "step": 31675
    },
    {
      "entropy": 6.563605642318725,
      "epoch": 3.3900690245599017,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038838840835144053,
      "loss": 5.8226,
      "mean_token_accuracy": 0.17341247349977493,
      "num_tokens": 68582241.0,
      "step": 31680
    },
    {
      "entropy": 6.5013679504394535,
      "epoch": 3.3906040986676653,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038835538327313746,
      "loss": 5.7423,
      "mean_token_accuracy": 0.1722983479499817,
      "num_tokens": 68594977.0,
      "step": 31685
    },
    {
      "entropy": 6.534288740158081,
      "epoch": 3.3911391727754294,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038832235492171363,
      "loss": 5.7355,
      "mean_token_accuracy": 0.17684439420700074,
      "num_tokens": 68605483.0,
      "step": 31690
    },
    {
      "entropy": 6.445546340942383,
      "epoch": 3.391674246883193,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038828932329812275,
      "loss": 5.6715,
      "mean_token_accuracy": 0.19069867134094237,
      "num_tokens": 68615274.0,
      "step": 31695
    },
    {
      "entropy": 6.500322008132935,
      "epoch": 3.392209320990957,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003882562884033186,
      "loss": 5.7969,
      "mean_token_accuracy": 0.17306627631187438,
      "num_tokens": 68626512.0,
      "step": 31700
    },
    {
      "entropy": 6.507903957366944,
      "epoch": 3.3927443950987213,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038822325023825507,
      "loss": 5.7963,
      "mean_token_accuracy": 0.17093949466943742,
      "num_tokens": 68637846.0,
      "step": 31705
    },
    {
      "entropy": 6.575248336791992,
      "epoch": 3.393279469206485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038819020880388603,
      "loss": 5.8074,
      "mean_token_accuracy": 0.169352026283741,
      "num_tokens": 68650377.0,
      "step": 31710
    },
    {
      "entropy": 6.517714405059815,
      "epoch": 3.393814543314249,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003881571641011656,
      "loss": 5.7796,
      "mean_token_accuracy": 0.1739790216088295,
      "num_tokens": 68661358.0,
      "step": 31715
    },
    {
      "entropy": 6.524976682662964,
      "epoch": 3.394349617422013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000388124116131048,
      "loss": 5.7811,
      "mean_token_accuracy": 0.16810798794031143,
      "num_tokens": 68673592.0,
      "step": 31720
    },
    {
      "entropy": 6.538378572463989,
      "epoch": 3.394884691529777,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038809106489448743,
      "loss": 5.6958,
      "mean_token_accuracy": 0.1831065073609352,
      "num_tokens": 68684225.0,
      "step": 31725
    },
    {
      "entropy": 6.5340063095092775,
      "epoch": 3.395419765637541,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038805801039243826,
      "loss": 5.8531,
      "mean_token_accuracy": 0.171348737180233,
      "num_tokens": 68694243.0,
      "step": 31730
    },
    {
      "entropy": 6.53493857383728,
      "epoch": 3.3959548397453045,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000388024952625855,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1723739430308342,
      "num_tokens": 68704359.0,
      "step": 31735
    },
    {
      "entropy": 6.501222419738769,
      "epoch": 3.3964899138530686,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038799189159569194,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16959039270877838,
      "num_tokens": 68714317.0,
      "step": 31740
    },
    {
      "entropy": 6.537135982513428,
      "epoch": 3.3970249879608327,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000387958827302904,
      "loss": 5.8825,
      "mean_token_accuracy": 0.17036814689636232,
      "num_tokens": 68725601.0,
      "step": 31745
    },
    {
      "entropy": 6.564855146408081,
      "epoch": 3.3975600620685964,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038792575974844575,
      "loss": 5.8095,
      "mean_token_accuracy": 0.1726281002163887,
      "num_tokens": 68735692.0,
      "step": 31750
    },
    {
      "entropy": 6.49014801979065,
      "epoch": 3.3980951361763605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003878926889332721,
      "loss": 5.672,
      "mean_token_accuracy": 0.18582161366939545,
      "num_tokens": 68746693.0,
      "step": 31755
    },
    {
      "entropy": 6.50321216583252,
      "epoch": 3.398630210284124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003878596148583379,
      "loss": 5.7965,
      "mean_token_accuracy": 0.1724425360560417,
      "num_tokens": 68757769.0,
      "step": 31760
    },
    {
      "entropy": 6.472357273101807,
      "epoch": 3.3991652843918883,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003878265375245982,
      "loss": 5.7517,
      "mean_token_accuracy": 0.17330776751041413,
      "num_tokens": 68768066.0,
      "step": 31765
    },
    {
      "entropy": 6.480617904663086,
      "epoch": 3.3997003584996524,
      "grad_norm": 1.0625,
      "learning_rate": 0.000387793456933008,
      "loss": 5.7398,
      "mean_token_accuracy": 0.17099883109331132,
      "num_tokens": 68780093.0,
      "step": 31770
    },
    {
      "entropy": 6.513686609268189,
      "epoch": 3.400235432607416,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003877603730845227,
      "loss": 5.7617,
      "mean_token_accuracy": 0.173665551841259,
      "num_tokens": 68791113.0,
      "step": 31775
    },
    {
      "entropy": 6.5011321067810055,
      "epoch": 3.40077050671518,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003877272859800975,
      "loss": 5.8232,
      "mean_token_accuracy": 0.17451414763927459,
      "num_tokens": 68801723.0,
      "step": 31780
    },
    {
      "entropy": 6.5549431324005125,
      "epoch": 3.401305580822944,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003876941956206876,
      "loss": 5.8424,
      "mean_token_accuracy": 0.17405073642730712,
      "num_tokens": 68813130.0,
      "step": 31785
    },
    {
      "entropy": 6.590711975097657,
      "epoch": 3.401840654930708,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038766110200724876,
      "loss": 5.7757,
      "mean_token_accuracy": 0.17816961258649827,
      "num_tokens": 68823699.0,
      "step": 31790
    },
    {
      "entropy": 6.543046665191651,
      "epoch": 3.402375729038472,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038762800514073636,
      "loss": 5.7981,
      "mean_token_accuracy": 0.16459860950708388,
      "num_tokens": 68835295.0,
      "step": 31795
    },
    {
      "entropy": 6.513854503631592,
      "epoch": 3.4029108031462356,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003875949050221062,
      "loss": 5.7869,
      "mean_token_accuracy": 0.1785682827234268,
      "num_tokens": 68845618.0,
      "step": 31800
    },
    {
      "entropy": 6.528678607940674,
      "epoch": 3.4034458772539997,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003875618016523139,
      "loss": 5.8456,
      "mean_token_accuracy": 0.17445728182792664,
      "num_tokens": 68855829.0,
      "step": 31805
    },
    {
      "entropy": 6.476268339157104,
      "epoch": 3.4039809513617634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003875286950323155,
      "loss": 5.7425,
      "mean_token_accuracy": 0.1727634087204933,
      "num_tokens": 68866731.0,
      "step": 31810
    },
    {
      "entropy": 6.5979829788208,
      "epoch": 3.4045160254695275,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003874955851630668,
      "loss": 5.892,
      "mean_token_accuracy": 0.1680956155061722,
      "num_tokens": 68877795.0,
      "step": 31815
    },
    {
      "entropy": 6.595898962020874,
      "epoch": 3.4050510995772916,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038746247204552386,
      "loss": 5.7684,
      "mean_token_accuracy": 0.17451810389757155,
      "num_tokens": 68886993.0,
      "step": 31820
    },
    {
      "entropy": 6.511575841903687,
      "epoch": 3.4055861736850552,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003874293556806428,
      "loss": 5.8376,
      "mean_token_accuracy": 0.17191888391971588,
      "num_tokens": 68899022.0,
      "step": 31825
    },
    {
      "entropy": 6.551673793792725,
      "epoch": 3.4061212477928193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038739623606937995,
      "loss": 5.8109,
      "mean_token_accuracy": 0.17466481328010558,
      "num_tokens": 68911052.0,
      "step": 31830
    },
    {
      "entropy": 6.512486362457276,
      "epoch": 3.4066563219005834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038736311321269155,
      "loss": 5.6981,
      "mean_token_accuracy": 0.18165435642004013,
      "num_tokens": 68921093.0,
      "step": 31835
    },
    {
      "entropy": 6.533487844467163,
      "epoch": 3.407191396008347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003873299871115341,
      "loss": 5.8371,
      "mean_token_accuracy": 0.16833935678005219,
      "num_tokens": 68931664.0,
      "step": 31840
    },
    {
      "entropy": 6.4979393005371096,
      "epoch": 3.407726470116111,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003872968577668641,
      "loss": 5.8081,
      "mean_token_accuracy": 0.17401717305183412,
      "num_tokens": 68942758.0,
      "step": 31845
    },
    {
      "entropy": 6.468049049377441,
      "epoch": 3.408261544223875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038726372517963796,
      "loss": 5.7188,
      "mean_token_accuracy": 0.17449921518564224,
      "num_tokens": 68953408.0,
      "step": 31850
    },
    {
      "entropy": 6.5472174167633055,
      "epoch": 3.408796618331639,
      "grad_norm": 1.125,
      "learning_rate": 0.00038723058935081264,
      "loss": 5.7796,
      "mean_token_accuracy": 0.17836536169052125,
      "num_tokens": 68964077.0,
      "step": 31855
    },
    {
      "entropy": 6.523726272583008,
      "epoch": 3.409331692439403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003871974502813448,
      "loss": 5.7783,
      "mean_token_accuracy": 0.179816934466362,
      "num_tokens": 68975761.0,
      "step": 31860
    },
    {
      "entropy": 6.511775112152099,
      "epoch": 3.4098667665471667,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003871643079721914,
      "loss": 5.8286,
      "mean_token_accuracy": 0.1720130816102028,
      "num_tokens": 68986965.0,
      "step": 31865
    },
    {
      "entropy": 6.507981634140014,
      "epoch": 3.410401840654931,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003871311624243092,
      "loss": 5.7973,
      "mean_token_accuracy": 0.1665094643831253,
      "num_tokens": 68998102.0,
      "step": 31870
    },
    {
      "entropy": 6.513416719436646,
      "epoch": 3.4109369147626944,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003870980136386556,
      "loss": 5.8144,
      "mean_token_accuracy": 0.17153917998075485,
      "num_tokens": 69008530.0,
      "step": 31875
    },
    {
      "entropy": 6.571871566772461,
      "epoch": 3.4114719888704585,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038706486161618753,
      "loss": 5.8058,
      "mean_token_accuracy": 0.17167469710111619,
      "num_tokens": 69018866.0,
      "step": 31880
    },
    {
      "entropy": 6.544409561157226,
      "epoch": 3.4120070629782226,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00038703170635786237,
      "loss": 5.7847,
      "mean_token_accuracy": 0.17083981037139892,
      "num_tokens": 69029050.0,
      "step": 31885
    },
    {
      "entropy": 6.525383520126343,
      "epoch": 3.4125421370859863,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038699854786463744,
      "loss": 5.7669,
      "mean_token_accuracy": 0.1736610308289528,
      "num_tokens": 69040117.0,
      "step": 31890
    },
    {
      "entropy": 6.5115784168243405,
      "epoch": 3.4130772111937504,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038696538613747016,
      "loss": 5.7555,
      "mean_token_accuracy": 0.18180599212646484,
      "num_tokens": 69051704.0,
      "step": 31895
    },
    {
      "entropy": 6.568415403366089,
      "epoch": 3.4136122853015145,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038693222117731804,
      "loss": 5.7512,
      "mean_token_accuracy": 0.1686740204691887,
      "num_tokens": 69062359.0,
      "step": 31900
    },
    {
      "entropy": 6.463142919540405,
      "epoch": 3.414147359409278,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038689905298513886,
      "loss": 5.7382,
      "mean_token_accuracy": 0.1743228778243065,
      "num_tokens": 69072936.0,
      "step": 31905
    },
    {
      "entropy": 6.4844239234924315,
      "epoch": 3.4146824335170423,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038686588156189024,
      "loss": 5.8066,
      "mean_token_accuracy": 0.1754053369164467,
      "num_tokens": 69083468.0,
      "step": 31910
    },
    {
      "entropy": 6.5516094207763675,
      "epoch": 3.415217507624806,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003868327069085299,
      "loss": 5.8224,
      "mean_token_accuracy": 0.17038949728012084,
      "num_tokens": 69093565.0,
      "step": 31915
    },
    {
      "entropy": 6.516125917434692,
      "epoch": 3.41575258173257,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000386799529026016,
      "loss": 5.8064,
      "mean_token_accuracy": 0.17000507116317748,
      "num_tokens": 69103660.0,
      "step": 31920
    },
    {
      "entropy": 6.494402313232422,
      "epoch": 3.4162876558403337,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003867663479153063,
      "loss": 5.7737,
      "mean_token_accuracy": 0.1825764685869217,
      "num_tokens": 69114810.0,
      "step": 31925
    },
    {
      "entropy": 6.587028312683105,
      "epoch": 3.4168227299480978,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038673316357735907,
      "loss": 5.8106,
      "mean_token_accuracy": 0.17695737034082412,
      "num_tokens": 69127634.0,
      "step": 31930
    },
    {
      "entropy": 6.593101835250854,
      "epoch": 3.417357804055862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003866999760131324,
      "loss": 5.7789,
      "mean_token_accuracy": 0.17624811828136444,
      "num_tokens": 69136957.0,
      "step": 31935
    },
    {
      "entropy": 6.471784448623657,
      "epoch": 3.4178928781636255,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038666678522358464,
      "loss": 5.7656,
      "mean_token_accuracy": 0.1719100669026375,
      "num_tokens": 69148353.0,
      "step": 31940
    },
    {
      "entropy": 6.537476348876953,
      "epoch": 3.4184279522713896,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003866335912096741,
      "loss": 5.8242,
      "mean_token_accuracy": 0.17380507588386535,
      "num_tokens": 69158329.0,
      "step": 31945
    },
    {
      "entropy": 6.471709203720093,
      "epoch": 3.4189630263791537,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003866003939723595,
      "loss": 5.6768,
      "mean_token_accuracy": 0.18622092753648758,
      "num_tokens": 69168689.0,
      "step": 31950
    },
    {
      "entropy": 6.531798601150513,
      "epoch": 3.4194981004869174,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038656719351259894,
      "loss": 5.8275,
      "mean_token_accuracy": 0.1752835214138031,
      "num_tokens": 69181166.0,
      "step": 31955
    },
    {
      "entropy": 6.566580009460449,
      "epoch": 3.4200331745946815,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003865339898313515,
      "loss": 5.7677,
      "mean_token_accuracy": 0.1749481439590454,
      "num_tokens": 69192221.0,
      "step": 31960
    },
    {
      "entropy": 6.522357320785522,
      "epoch": 3.420568248702445,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038650078292957573,
      "loss": 5.8421,
      "mean_token_accuracy": 0.17076903879642485,
      "num_tokens": 69203225.0,
      "step": 31965
    },
    {
      "entropy": 6.549492883682251,
      "epoch": 3.4211033228102092,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038646757280823047,
      "loss": 5.8432,
      "mean_token_accuracy": 0.17104796469211578,
      "num_tokens": 69214349.0,
      "step": 31970
    },
    {
      "entropy": 6.565251207351684,
      "epoch": 3.421638396917973,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003864343594682748,
      "loss": 5.8428,
      "mean_token_accuracy": 0.16913563907146453,
      "num_tokens": 69225863.0,
      "step": 31975
    },
    {
      "entropy": 6.503674507141113,
      "epoch": 3.422173471025737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038640114291066765,
      "loss": 5.6983,
      "mean_token_accuracy": 0.18197257369756697,
      "num_tokens": 69236572.0,
      "step": 31980
    },
    {
      "entropy": 6.490173625946045,
      "epoch": 3.422708545133501,
      "grad_norm": 1.3125,
      "learning_rate": 0.00038636792313636816,
      "loss": 5.815,
      "mean_token_accuracy": 0.17230146527290344,
      "num_tokens": 69246034.0,
      "step": 31985
    },
    {
      "entropy": 6.489099788665771,
      "epoch": 3.4232436192412647,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003863347001463355,
      "loss": 5.7537,
      "mean_token_accuracy": 0.17430565804243087,
      "num_tokens": 69256471.0,
      "step": 31990
    },
    {
      "entropy": 6.492844963073731,
      "epoch": 3.423778693349029,
      "grad_norm": 1.125,
      "learning_rate": 0.000386301473941529,
      "loss": 5.7079,
      "mean_token_accuracy": 0.1794181674718857,
      "num_tokens": 69266601.0,
      "step": 31995
    },
    {
      "entropy": 6.527189874649048,
      "epoch": 3.424313767456793,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003862682445229081,
      "loss": 5.8196,
      "mean_token_accuracy": 0.169660022854805,
      "num_tokens": 69276929.0,
      "step": 32000
    },
    {
      "entropy": 6.507036399841309,
      "epoch": 3.4248488415645566,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038623501189143225,
      "loss": 5.8063,
      "mean_token_accuracy": 0.1717614471912384,
      "num_tokens": 69287654.0,
      "step": 32005
    },
    {
      "entropy": 6.497146224975586,
      "epoch": 3.4253839156723207,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003862017760480611,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1764647677540779,
      "num_tokens": 69297346.0,
      "step": 32010
    },
    {
      "entropy": 6.511147975921631,
      "epoch": 3.4259189897800844,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038616853699375425,
      "loss": 5.8138,
      "mean_token_accuracy": 0.1719893753528595,
      "num_tokens": 69307806.0,
      "step": 32015
    },
    {
      "entropy": 6.5184930801391605,
      "epoch": 3.4264540638878485,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003861352947294716,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16837963014841079,
      "num_tokens": 69317964.0,
      "step": 32020
    },
    {
      "entropy": 6.5005778789520265,
      "epoch": 3.4269891379956126,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038610204925617285,
      "loss": 5.7456,
      "mean_token_accuracy": 0.1804135411977768,
      "num_tokens": 69328051.0,
      "step": 32025
    },
    {
      "entropy": 6.514376163482666,
      "epoch": 3.427524212103376,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038606880057481807,
      "loss": 5.7602,
      "mean_token_accuracy": 0.17753705084323884,
      "num_tokens": 69339949.0,
      "step": 32030
    },
    {
      "entropy": 6.550789833068848,
      "epoch": 3.4280592862111403,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038603554868636735,
      "loss": 5.7337,
      "mean_token_accuracy": 0.17971349209547044,
      "num_tokens": 69350188.0,
      "step": 32035
    },
    {
      "entropy": 6.52921051979065,
      "epoch": 3.428594360318904,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003860022935917807,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16626688987016677,
      "num_tokens": 69360244.0,
      "step": 32040
    },
    {
      "entropy": 6.478483104705811,
      "epoch": 3.429129434426668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003859690352920184,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1770999550819397,
      "num_tokens": 69371060.0,
      "step": 32045
    },
    {
      "entropy": 6.521747016906739,
      "epoch": 3.429664508534432,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038593577378804087,
      "loss": 5.8096,
      "mean_token_accuracy": 0.17085815966129303,
      "num_tokens": 69382371.0,
      "step": 32050
    },
    {
      "entropy": 6.547934484481812,
      "epoch": 3.430199582642196,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003859025090808083,
      "loss": 5.8085,
      "mean_token_accuracy": 0.17460851669311522,
      "num_tokens": 69392747.0,
      "step": 32055
    },
    {
      "entropy": 6.522341728210449,
      "epoch": 3.43073465674996,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003858692411712815,
      "loss": 5.747,
      "mean_token_accuracy": 0.17882351726293563,
      "num_tokens": 69403403.0,
      "step": 32060
    },
    {
      "entropy": 6.485430955886841,
      "epoch": 3.431269730857724,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000385835970060421,
      "loss": 5.7457,
      "mean_token_accuracy": 0.1757091239094734,
      "num_tokens": 69414276.0,
      "step": 32065
    },
    {
      "entropy": 6.502235460281372,
      "epoch": 3.4318048049654877,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038580269574918735,
      "loss": 5.8077,
      "mean_token_accuracy": 0.17555781900882722,
      "num_tokens": 69424707.0,
      "step": 32070
    },
    {
      "entropy": 6.4781105518341064,
      "epoch": 3.4323398790732518,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038576941823854135,
      "loss": 5.7066,
      "mean_token_accuracy": 0.18917252868413925,
      "num_tokens": 69436638.0,
      "step": 32075
    },
    {
      "entropy": 6.479278993606568,
      "epoch": 3.4328749531810154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038573613752944407,
      "loss": 5.7286,
      "mean_token_accuracy": 0.1810982421040535,
      "num_tokens": 69448431.0,
      "step": 32080
    },
    {
      "entropy": 6.456598949432373,
      "epoch": 3.4334100272887795,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038570285362285635,
      "loss": 5.7503,
      "mean_token_accuracy": 0.18424051254987717,
      "num_tokens": 69458897.0,
      "step": 32085
    },
    {
      "entropy": 6.562639236450195,
      "epoch": 3.433945101396543,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038566956651973927,
      "loss": 5.8074,
      "mean_token_accuracy": 0.17458634227514266,
      "num_tokens": 69468517.0,
      "step": 32090
    },
    {
      "entropy": 6.493085050582886,
      "epoch": 3.4344801755043073,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000385636276221054,
      "loss": 5.8201,
      "mean_token_accuracy": 0.16984521895647048,
      "num_tokens": 69478634.0,
      "step": 32095
    },
    {
      "entropy": 6.5006739616394045,
      "epoch": 3.4350152496120714,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003856029827277619,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17568581253290178,
      "num_tokens": 69490235.0,
      "step": 32100
    },
    {
      "entropy": 6.556049442291259,
      "epoch": 3.435550323719835,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038556968604082413,
      "loss": 5.7995,
      "mean_token_accuracy": 0.16697111278772353,
      "num_tokens": 69501678.0,
      "step": 32105
    },
    {
      "entropy": 6.498092746734619,
      "epoch": 3.436085397827599,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038553638616120226,
      "loss": 5.7606,
      "mean_token_accuracy": 0.17682411670684814,
      "num_tokens": 69512214.0,
      "step": 32110
    },
    {
      "entropy": 6.522635650634766,
      "epoch": 3.4366204719353632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038550308308985773,
      "loss": 5.8253,
      "mean_token_accuracy": 0.17431548088788987,
      "num_tokens": 69522959.0,
      "step": 32115
    },
    {
      "entropy": 6.5511918544769285,
      "epoch": 3.437155546043127,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003854697768277522,
      "loss": 5.8369,
      "mean_token_accuracy": 0.1718675822019577,
      "num_tokens": 69534187.0,
      "step": 32120
    },
    {
      "entropy": 6.493026351928711,
      "epoch": 3.437690620150891,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003854364673758474,
      "loss": 5.8161,
      "mean_token_accuracy": 0.1751650467514992,
      "num_tokens": 69545240.0,
      "step": 32125
    },
    {
      "entropy": 6.4934593677520756,
      "epoch": 3.4382256942586547,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038540315473510514,
      "loss": 5.7392,
      "mean_token_accuracy": 0.1753550112247467,
      "num_tokens": 69555457.0,
      "step": 32130
    },
    {
      "entropy": 6.560977172851563,
      "epoch": 3.4387607683664188,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003853698389064873,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1734065219759941,
      "num_tokens": 69565604.0,
      "step": 32135
    },
    {
      "entropy": 6.532773017883301,
      "epoch": 3.439295842474183,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038533651989095587,
      "loss": 5.8284,
      "mean_token_accuracy": 0.175779090821743,
      "num_tokens": 69576265.0,
      "step": 32140
    },
    {
      "entropy": 6.566759490966797,
      "epoch": 3.4398309165819465,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003853031976894729,
      "loss": 5.8391,
      "mean_token_accuracy": 0.17292921096086503,
      "num_tokens": 69586983.0,
      "step": 32145
    },
    {
      "entropy": 6.476215171813965,
      "epoch": 3.4403659906897106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003852698723030006,
      "loss": 5.7421,
      "mean_token_accuracy": 0.18645557016134262,
      "num_tokens": 69598455.0,
      "step": 32150
    },
    {
      "entropy": 6.497090148925781,
      "epoch": 3.4409010647974743,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038523654373250125,
      "loss": 5.7509,
      "mean_token_accuracy": 0.1754988819360733,
      "num_tokens": 69609922.0,
      "step": 32155
    },
    {
      "entropy": 6.451273584365845,
      "epoch": 3.4414361389052384,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038520321197893715,
      "loss": 5.7045,
      "mean_token_accuracy": 0.1825033411383629,
      "num_tokens": 69620641.0,
      "step": 32160
    },
    {
      "entropy": 6.451138830184936,
      "epoch": 3.4419712130130025,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003851698770432709,
      "loss": 5.6627,
      "mean_token_accuracy": 0.18618523478507995,
      "num_tokens": 69631906.0,
      "step": 32165
    },
    {
      "entropy": 6.4755682945251465,
      "epoch": 3.442506287120766,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003851365389264649,
      "loss": 5.7981,
      "mean_token_accuracy": 0.17392975687980652,
      "num_tokens": 69642379.0,
      "step": 32170
    },
    {
      "entropy": 6.38673038482666,
      "epoch": 3.44304136122853,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038510319762948166,
      "loss": 5.6353,
      "mean_token_accuracy": 0.19125430136919022,
      "num_tokens": 69653363.0,
      "step": 32175
    },
    {
      "entropy": 6.514960050582886,
      "epoch": 3.4435764353362943,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003850698531532843,
      "loss": 5.7183,
      "mean_token_accuracy": 0.18035906702280044,
      "num_tokens": 69663693.0,
      "step": 32180
    },
    {
      "entropy": 6.5218418598175045,
      "epoch": 3.444111509444058,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038503650549883523,
      "loss": 5.8134,
      "mean_token_accuracy": 0.17171939313411713,
      "num_tokens": 69673935.0,
      "step": 32185
    },
    {
      "entropy": 6.449759006500244,
      "epoch": 3.444646583551822,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003850031546670975,
      "loss": 5.7291,
      "mean_token_accuracy": 0.17734221667051314,
      "num_tokens": 69684364.0,
      "step": 32190
    },
    {
      "entropy": 6.561565399169922,
      "epoch": 3.4451816576595857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038496980065903433,
      "loss": 5.8851,
      "mean_token_accuracy": 0.16574711203575135,
      "num_tokens": 69695034.0,
      "step": 32195
    },
    {
      "entropy": 6.553401374816895,
      "epoch": 3.44571673176735,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003849364434756084,
      "loss": 5.8355,
      "mean_token_accuracy": 0.17168413996696472,
      "num_tokens": 69705450.0,
      "step": 32200
    },
    {
      "entropy": 6.531741666793823,
      "epoch": 3.4462518058751135,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003849030831177832,
      "loss": 5.8494,
      "mean_token_accuracy": 0.17975082993507385,
      "num_tokens": 69716365.0,
      "step": 32205
    },
    {
      "entropy": 6.517589902877807,
      "epoch": 3.4467868799828776,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038486971958652196,
      "loss": 5.8421,
      "mean_token_accuracy": 0.17434273064136505,
      "num_tokens": 69726484.0,
      "step": 32210
    },
    {
      "entropy": 6.5368828773498535,
      "epoch": 3.4473219540906417,
      "grad_norm": 1.15625,
      "learning_rate": 0.000384836352882788,
      "loss": 5.7838,
      "mean_token_accuracy": 0.17197242826223375,
      "num_tokens": 69737498.0,
      "step": 32215
    },
    {
      "entropy": 6.524175310134888,
      "epoch": 3.4478570281984053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038480298300754473,
      "loss": 5.7521,
      "mean_token_accuracy": 0.1846578225493431,
      "num_tokens": 69748284.0,
      "step": 32220
    },
    {
      "entropy": 6.561102676391601,
      "epoch": 3.4483921023061694,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038476960996175583,
      "loss": 5.8055,
      "mean_token_accuracy": 0.17113113403320312,
      "num_tokens": 69759745.0,
      "step": 32225
    },
    {
      "entropy": 6.52344536781311,
      "epoch": 3.4489271764139335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003847362337463848,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17147112786769866,
      "num_tokens": 69770546.0,
      "step": 32230
    },
    {
      "entropy": 6.455391788482666,
      "epoch": 3.449462250521697,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038470285436239545,
      "loss": 5.7356,
      "mean_token_accuracy": 0.17960200905799867,
      "num_tokens": 69781154.0,
      "step": 32235
    },
    {
      "entropy": 6.479723644256592,
      "epoch": 3.4499973246294613,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003846694718107516,
      "loss": 5.795,
      "mean_token_accuracy": 0.17048170268535615,
      "num_tokens": 69792311.0,
      "step": 32240
    },
    {
      "entropy": 6.559449672698975,
      "epoch": 3.450532398737225,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003846360860924172,
      "loss": 5.8831,
      "mean_token_accuracy": 0.16872453838586807,
      "num_tokens": 69803181.0,
      "step": 32245
    },
    {
      "entropy": 6.5605193138122555,
      "epoch": 3.451067472844989,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038460269720835615,
      "loss": 5.6806,
      "mean_token_accuracy": 0.1909778669476509,
      "num_tokens": 69815177.0,
      "step": 32250
    },
    {
      "entropy": 6.572175455093384,
      "epoch": 3.451602546952753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038456930515953265,
      "loss": 5.8061,
      "mean_token_accuracy": 0.1698318138718605,
      "num_tokens": 69826730.0,
      "step": 32255
    },
    {
      "entropy": 6.586740016937256,
      "epoch": 3.452137621060517,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003845359099469109,
      "loss": 5.8756,
      "mean_token_accuracy": 0.16733684986829758,
      "num_tokens": 69837321.0,
      "step": 32260
    },
    {
      "entropy": 6.485474014282227,
      "epoch": 3.452672695168281,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003845025115714551,
      "loss": 5.715,
      "mean_token_accuracy": 0.17588987797498704,
      "num_tokens": 69849093.0,
      "step": 32265
    },
    {
      "entropy": 6.501943731307984,
      "epoch": 3.4532077692760446,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003844691100341297,
      "loss": 5.714,
      "mean_token_accuracy": 0.17841679900884627,
      "num_tokens": 69859131.0,
      "step": 32270
    },
    {
      "entropy": 6.519770717620849,
      "epoch": 3.4537428433838087,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003844357053358991,
      "loss": 5.823,
      "mean_token_accuracy": 0.17063815146684647,
      "num_tokens": 69870089.0,
      "step": 32275
    },
    {
      "entropy": 6.507721614837647,
      "epoch": 3.4542779174915728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003844022974777279,
      "loss": 5.7705,
      "mean_token_accuracy": 0.17558738440275193,
      "num_tokens": 69880761.0,
      "step": 32280
    },
    {
      "entropy": 6.565920686721801,
      "epoch": 3.4548129915993364,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003843688864605807,
      "loss": 5.7389,
      "mean_token_accuracy": 0.1777879923582077,
      "num_tokens": 69891205.0,
      "step": 32285
    },
    {
      "entropy": 6.5129533290863035,
      "epoch": 3.4553480657071005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038433547228542233,
      "loss": 5.7092,
      "mean_token_accuracy": 0.17957583963871002,
      "num_tokens": 69902038.0,
      "step": 32290
    },
    {
      "entropy": 6.601068449020386,
      "epoch": 3.455883139814864,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038430205495321753,
      "loss": 5.8691,
      "mean_token_accuracy": 0.17010733336210251,
      "num_tokens": 69914449.0,
      "step": 32295
    },
    {
      "entropy": 6.617779493331909,
      "epoch": 3.4564182139226283,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003842686344649312,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1713160365819931,
      "num_tokens": 69926096.0,
      "step": 32300
    },
    {
      "entropy": 6.518750238418579,
      "epoch": 3.4569532880303924,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038423521082152844,
      "loss": 5.7676,
      "mean_token_accuracy": 0.17468288093805312,
      "num_tokens": 69935947.0,
      "step": 32305
    },
    {
      "entropy": 6.46973180770874,
      "epoch": 3.457488362138156,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003842017840239744,
      "loss": 5.7192,
      "mean_token_accuracy": 0.17115322202444078,
      "num_tokens": 69946701.0,
      "step": 32310
    },
    {
      "entropy": 6.581831359863282,
      "epoch": 3.45802343624592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003841683540732341,
      "loss": 5.8738,
      "mean_token_accuracy": 0.1682122215628624,
      "num_tokens": 69957011.0,
      "step": 32315
    },
    {
      "entropy": 6.574256229400635,
      "epoch": 3.458558510353684,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003841349209702729,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17618471682071685,
      "num_tokens": 69968224.0,
      "step": 32320
    },
    {
      "entropy": 6.531284618377685,
      "epoch": 3.459093584461448,
      "grad_norm": 1.125,
      "learning_rate": 0.0003841014847160563,
      "loss": 5.8108,
      "mean_token_accuracy": 0.17096076607704164,
      "num_tokens": 69979171.0,
      "step": 32325
    },
    {
      "entropy": 6.532107639312744,
      "epoch": 3.459628658569212,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003840680453115496,
      "loss": 5.7887,
      "mean_token_accuracy": 0.17743092626333237,
      "num_tokens": 69990001.0,
      "step": 32330
    },
    {
      "entropy": 6.517334270477295,
      "epoch": 3.4601637326769756,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038403460275771836,
      "loss": 5.7297,
      "mean_token_accuracy": 0.17857916206121444,
      "num_tokens": 70001263.0,
      "step": 32335
    },
    {
      "entropy": 6.478394794464111,
      "epoch": 3.4606988067847397,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038400115705552833,
      "loss": 5.7695,
      "mean_token_accuracy": 0.17674271911382675,
      "num_tokens": 70011586.0,
      "step": 32340
    },
    {
      "entropy": 6.418226051330566,
      "epoch": 3.461233880892504,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003839677082059453,
      "loss": 5.7052,
      "mean_token_accuracy": 0.18129769414663316,
      "num_tokens": 70023607.0,
      "step": 32345
    },
    {
      "entropy": 6.539365005493164,
      "epoch": 3.4617689550002675,
      "grad_norm": 1.34375,
      "learning_rate": 0.00038393425620993483,
      "loss": 5.8443,
      "mean_token_accuracy": 0.16923927515745163,
      "num_tokens": 70033099.0,
      "step": 32350
    },
    {
      "entropy": 6.506365156173706,
      "epoch": 3.4623040291080316,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038390080106846314,
      "loss": 5.7656,
      "mean_token_accuracy": 0.18167202770709992,
      "num_tokens": 70043609.0,
      "step": 32355
    },
    {
      "entropy": 6.516669797897339,
      "epoch": 3.4628391032157952,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003838673427824962,
      "loss": 5.7811,
      "mean_token_accuracy": 0.1759556785225868,
      "num_tokens": 70054901.0,
      "step": 32360
    },
    {
      "entropy": 6.5647650241851805,
      "epoch": 3.4633741773235593,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038383388135299984,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16998098343610762,
      "num_tokens": 70065729.0,
      "step": 32365
    },
    {
      "entropy": 6.537253999710083,
      "epoch": 3.463909251431323,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003838004167809406,
      "loss": 5.775,
      "mean_token_accuracy": 0.1730008080601692,
      "num_tokens": 70077841.0,
      "step": 32370
    },
    {
      "entropy": 6.581489610671997,
      "epoch": 3.464444325539087,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003837669490672845,
      "loss": 5.8202,
      "mean_token_accuracy": 0.17084682136774063,
      "num_tokens": 70089205.0,
      "step": 32375
    },
    {
      "entropy": 6.514441108703613,
      "epoch": 3.464979399646851,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003837334782129981,
      "loss": 5.8271,
      "mean_token_accuracy": 0.17649988532066346,
      "num_tokens": 70100118.0,
      "step": 32380
    },
    {
      "entropy": 6.465589094161987,
      "epoch": 3.465514473754615,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038370000421904775,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17526978999376297,
      "num_tokens": 70110502.0,
      "step": 32385
    },
    {
      "entropy": 6.564020204544067,
      "epoch": 3.466049547862379,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003836665270864001,
      "loss": 5.8018,
      "mean_token_accuracy": 0.1701392337679863,
      "num_tokens": 70121040.0,
      "step": 32390
    },
    {
      "entropy": 6.556829214096069,
      "epoch": 3.466584621970143,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038363304681602174,
      "loss": 5.8417,
      "mean_token_accuracy": 0.17658375054597855,
      "num_tokens": 70132537.0,
      "step": 32395
    },
    {
      "entropy": 6.53735761642456,
      "epoch": 3.4671196960779067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003835995634088794,
      "loss": 5.7977,
      "mean_token_accuracy": 0.17992740571498872,
      "num_tokens": 70143449.0,
      "step": 32400
    },
    {
      "entropy": 6.47696361541748,
      "epoch": 3.467654770185671,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038356607686593994,
      "loss": 5.7004,
      "mean_token_accuracy": 0.18601094484329223,
      "num_tokens": 70154674.0,
      "step": 32405
    },
    {
      "entropy": 6.535722780227661,
      "epoch": 3.4681898442934345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003835325871881703,
      "loss": 5.7819,
      "mean_token_accuracy": 0.17512759417295456,
      "num_tokens": 70165342.0,
      "step": 32410
    },
    {
      "entropy": 6.529057359695434,
      "epoch": 3.4687249184011986,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003834990943765374,
      "loss": 5.7501,
      "mean_token_accuracy": 0.17949350029230118,
      "num_tokens": 70177028.0,
      "step": 32415
    },
    {
      "entropy": 6.503246021270752,
      "epoch": 3.4692599925089627,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003834655984320084,
      "loss": 5.836,
      "mean_token_accuracy": 0.1723273828625679,
      "num_tokens": 70188655.0,
      "step": 32420
    },
    {
      "entropy": 6.574833822250366,
      "epoch": 3.4697950666167263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003834320993555505,
      "loss": 5.8844,
      "mean_token_accuracy": 0.16770491600036622,
      "num_tokens": 70199288.0,
      "step": 32425
    },
    {
      "entropy": 6.5629185199737545,
      "epoch": 3.4703301407244904,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038339859714813104,
      "loss": 5.8082,
      "mean_token_accuracy": 0.17450107783079147,
      "num_tokens": 70210417.0,
      "step": 32430
    },
    {
      "entropy": 6.53646502494812,
      "epoch": 3.470865214832254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003833650918107172,
      "loss": 5.7864,
      "mean_token_accuracy": 0.17599458247423172,
      "num_tokens": 70220339.0,
      "step": 32435
    },
    {
      "entropy": 6.501313877105713,
      "epoch": 3.471400288940018,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003833315833442766,
      "loss": 5.8162,
      "mean_token_accuracy": 0.17372930496931077,
      "num_tokens": 70230699.0,
      "step": 32440
    },
    {
      "entropy": 6.543470430374145,
      "epoch": 3.4719353630477823,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038329807174977676,
      "loss": 5.7801,
      "mean_token_accuracy": 0.1762274131178856,
      "num_tokens": 70242464.0,
      "step": 32445
    },
    {
      "entropy": 6.551260137557984,
      "epoch": 3.472470437155546,
      "grad_norm": 1.25,
      "learning_rate": 0.00038326455702818537,
      "loss": 5.841,
      "mean_token_accuracy": 0.16957011222839355,
      "num_tokens": 70253631.0,
      "step": 32450
    },
    {
      "entropy": 6.5374963760375975,
      "epoch": 3.47300551126331,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038323103918047,
      "loss": 5.8252,
      "mean_token_accuracy": 0.16898305118083953,
      "num_tokens": 70266055.0,
      "step": 32455
    },
    {
      "entropy": 6.609314203262329,
      "epoch": 3.473540585371074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003831975182075987,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16578185111284255,
      "num_tokens": 70276680.0,
      "step": 32460
    },
    {
      "entropy": 6.608695363998413,
      "epoch": 3.474075659478838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038316399411053917,
      "loss": 5.7935,
      "mean_token_accuracy": 0.17382690459489822,
      "num_tokens": 70287424.0,
      "step": 32465
    },
    {
      "entropy": 6.531606817245484,
      "epoch": 3.474610733586602,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003831304668902595,
      "loss": 5.7564,
      "mean_token_accuracy": 0.17664196491241455,
      "num_tokens": 70298005.0,
      "step": 32470
    },
    {
      "entropy": 6.465831470489502,
      "epoch": 3.4751458076943655,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003830969365477279,
      "loss": 5.7402,
      "mean_token_accuracy": 0.17303248941898347,
      "num_tokens": 70309200.0,
      "step": 32475
    },
    {
      "entropy": 6.473586177825927,
      "epoch": 3.4756808818021296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038306340308391233,
      "loss": 5.7834,
      "mean_token_accuracy": 0.17868717461824418,
      "num_tokens": 70320957.0,
      "step": 32480
    },
    {
      "entropy": 6.564447736740112,
      "epoch": 3.4762159559098933,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003830298664997813,
      "loss": 5.788,
      "mean_token_accuracy": 0.17021074742078782,
      "num_tokens": 70331960.0,
      "step": 32485
    },
    {
      "entropy": 6.585651063919068,
      "epoch": 3.4767510300176574,
      "grad_norm": 1.25,
      "learning_rate": 0.000382996326796303,
      "loss": 5.7828,
      "mean_token_accuracy": 0.177692973613739,
      "num_tokens": 70342449.0,
      "step": 32490
    },
    {
      "entropy": 6.511133861541748,
      "epoch": 3.4772861041254215,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003829627839744459,
      "loss": 5.7552,
      "mean_token_accuracy": 0.17880776673555374,
      "num_tokens": 70353322.0,
      "step": 32495
    },
    {
      "entropy": 6.494898509979248,
      "epoch": 3.477821178233185,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003829292380351786,
      "loss": 5.7464,
      "mean_token_accuracy": 0.1930728882551193,
      "num_tokens": 70364824.0,
      "step": 32500
    },
    {
      "entropy": 6.516603231430054,
      "epoch": 3.4783562523409493,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003828956889794698,
      "loss": 5.7592,
      "mean_token_accuracy": 0.17874324023723603,
      "num_tokens": 70376811.0,
      "step": 32505
    },
    {
      "entropy": 6.550383901596069,
      "epoch": 3.4788913264487134,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038286213680828804,
      "loss": 5.8236,
      "mean_token_accuracy": 0.16950723081827163,
      "num_tokens": 70387735.0,
      "step": 32510
    },
    {
      "entropy": 6.531530666351318,
      "epoch": 3.479426400556477,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003828285815226023,
      "loss": 5.7367,
      "mean_token_accuracy": 0.17951344847679138,
      "num_tokens": 70398860.0,
      "step": 32515
    },
    {
      "entropy": 6.515701723098755,
      "epoch": 3.479961474664241,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038279502312338137,
      "loss": 5.756,
      "mean_token_accuracy": 0.17529616951942445,
      "num_tokens": 70409340.0,
      "step": 32520
    },
    {
      "entropy": 6.560732030868531,
      "epoch": 3.4804965487720048,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003827614616115943,
      "loss": 5.8624,
      "mean_token_accuracy": 0.1701030522584915,
      "num_tokens": 70419880.0,
      "step": 32525
    },
    {
      "entropy": 6.516998720169068,
      "epoch": 3.481031622879769,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003827278969882102,
      "loss": 5.7502,
      "mean_token_accuracy": 0.1831625834107399,
      "num_tokens": 70431033.0,
      "step": 32530
    },
    {
      "entropy": 6.544100999832153,
      "epoch": 3.481566696987533,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003826943292541982,
      "loss": 5.7758,
      "mean_token_accuracy": 0.1777363270521164,
      "num_tokens": 70442315.0,
      "step": 32535
    },
    {
      "entropy": 6.524233961105347,
      "epoch": 3.4821017710952966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038266075841052763,
      "loss": 5.8073,
      "mean_token_accuracy": 0.17717406153678894,
      "num_tokens": 70453195.0,
      "step": 32540
    },
    {
      "entropy": 6.5060523509979244,
      "epoch": 3.4826368452030607,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003826271844581677,
      "loss": 5.7711,
      "mean_token_accuracy": 0.17201472520828248,
      "num_tokens": 70463855.0,
      "step": 32545
    },
    {
      "entropy": 6.5082274913787845,
      "epoch": 3.4831719193108244,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000382593607398088,
      "loss": 5.8109,
      "mean_token_accuracy": 0.17004060447216035,
      "num_tokens": 70474517.0,
      "step": 32550
    },
    {
      "entropy": 6.572151517868042,
      "epoch": 3.4837069934185885,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000382560027231258,
      "loss": 5.85,
      "mean_token_accuracy": 0.16807512044906617,
      "num_tokens": 70486256.0,
      "step": 32555
    },
    {
      "entropy": 6.581733560562133,
      "epoch": 3.4842420675263526,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003825264439586473,
      "loss": 5.7813,
      "mean_token_accuracy": 0.17767055779695512,
      "num_tokens": 70496771.0,
      "step": 32560
    },
    {
      "entropy": 6.600895166397095,
      "epoch": 3.4847771416341162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038249285758122566,
      "loss": 5.9195,
      "mean_token_accuracy": 0.16923175156116485,
      "num_tokens": 70507994.0,
      "step": 32565
    },
    {
      "entropy": 6.537680006027221,
      "epoch": 3.4853122157418803,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038245926809996284,
      "loss": 5.8152,
      "mean_token_accuracy": 0.1733083188533783,
      "num_tokens": 70518544.0,
      "step": 32570
    },
    {
      "entropy": 6.515247058868408,
      "epoch": 3.485847289849644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003824256755158288,
      "loss": 5.8313,
      "mean_token_accuracy": 0.1774810492992401,
      "num_tokens": 70530075.0,
      "step": 32575
    },
    {
      "entropy": 6.487158632278442,
      "epoch": 3.486382363957408,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003823920798297934,
      "loss": 5.8144,
      "mean_token_accuracy": 0.1798624038696289,
      "num_tokens": 70542125.0,
      "step": 32580
    },
    {
      "entropy": 6.585095071792603,
      "epoch": 3.486917438065172,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038235848104282684,
      "loss": 5.7846,
      "mean_token_accuracy": 0.17543006539344788,
      "num_tokens": 70552710.0,
      "step": 32585
    },
    {
      "entropy": 6.562858819961548,
      "epoch": 3.487452512172936,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038232487915589925,
      "loss": 5.8114,
      "mean_token_accuracy": 0.17020724713802338,
      "num_tokens": 70564467.0,
      "step": 32590
    },
    {
      "entropy": 6.513343477249146,
      "epoch": 3.4879875862807,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038229127416998076,
      "loss": 5.7124,
      "mean_token_accuracy": 0.18566641807556153,
      "num_tokens": 70575164.0,
      "step": 32595
    },
    {
      "entropy": 6.527478170394898,
      "epoch": 3.4885226603884636,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038225766608604184,
      "loss": 5.8021,
      "mean_token_accuracy": 0.18069081604480744,
      "num_tokens": 70584841.0,
      "step": 32600
    },
    {
      "entropy": 6.494548320770264,
      "epoch": 3.4890577344962277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038222405490505296,
      "loss": 5.8325,
      "mean_token_accuracy": 0.17167699486017227,
      "num_tokens": 70595189.0,
      "step": 32605
    },
    {
      "entropy": 6.553327465057373,
      "epoch": 3.489592808603992,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038219044062798454,
      "loss": 5.8025,
      "mean_token_accuracy": 0.17526786476373674,
      "num_tokens": 70607858.0,
      "step": 32610
    },
    {
      "entropy": 6.555867767333984,
      "epoch": 3.4901278827117554,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038215682325580714,
      "loss": 5.7379,
      "mean_token_accuracy": 0.17721765786409377,
      "num_tokens": 70618281.0,
      "step": 32615
    },
    {
      "entropy": 6.510036325454712,
      "epoch": 3.4906629568195195,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003821232027894916,
      "loss": 5.7445,
      "mean_token_accuracy": 0.17590238451957702,
      "num_tokens": 70629822.0,
      "step": 32620
    },
    {
      "entropy": 6.4806968688964846,
      "epoch": 3.4911980309272836,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003820895792300086,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1770677924156189,
      "num_tokens": 70639981.0,
      "step": 32625
    },
    {
      "entropy": 6.615036678314209,
      "epoch": 3.4917331050350473,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038205595257832906,
      "loss": 5.8858,
      "mean_token_accuracy": 0.16516197025775908,
      "num_tokens": 70651196.0,
      "step": 32630
    },
    {
      "entropy": 6.531209945678711,
      "epoch": 3.4922681791428114,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038202232283542385,
      "loss": 5.7881,
      "mean_token_accuracy": 0.18201145231723787,
      "num_tokens": 70662930.0,
      "step": 32635
    },
    {
      "entropy": 6.495673990249633,
      "epoch": 3.492803253250575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003819886900022642,
      "loss": 5.8138,
      "mean_token_accuracy": 0.1709242880344391,
      "num_tokens": 70674276.0,
      "step": 32640
    },
    {
      "entropy": 6.534606838226319,
      "epoch": 3.493338327358339,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003819550540798212,
      "loss": 5.7652,
      "mean_token_accuracy": 0.17738081365823746,
      "num_tokens": 70684706.0,
      "step": 32645
    },
    {
      "entropy": 6.543149328231811,
      "epoch": 3.493873401466103,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000381921415069066,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17523010075092316,
      "num_tokens": 70695468.0,
      "step": 32650
    },
    {
      "entropy": 6.491376543045044,
      "epoch": 3.494408475573867,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038188777297096986,
      "loss": 5.7462,
      "mean_token_accuracy": 0.17795571833848953,
      "num_tokens": 70706141.0,
      "step": 32655
    },
    {
      "entropy": 6.5413402080535885,
      "epoch": 3.494943549681631,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00038185412778650435,
      "loss": 5.8609,
      "mean_token_accuracy": 0.16488420516252517,
      "num_tokens": 70716615.0,
      "step": 32660
    },
    {
      "entropy": 6.547104597091675,
      "epoch": 3.4954786237893947,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038182047951664097,
      "loss": 5.7578,
      "mean_token_accuracy": 0.17800500988960266,
      "num_tokens": 70727734.0,
      "step": 32665
    },
    {
      "entropy": 6.535657596588135,
      "epoch": 3.4960136978971588,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003817868281623512,
      "loss": 5.7128,
      "mean_token_accuracy": 0.18282048404216766,
      "num_tokens": 70738539.0,
      "step": 32670
    },
    {
      "entropy": 6.522606420516968,
      "epoch": 3.496548772004923,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038175317372460683,
      "loss": 5.7529,
      "mean_token_accuracy": 0.18146462738513947,
      "num_tokens": 70749043.0,
      "step": 32675
    },
    {
      "entropy": 6.505731916427612,
      "epoch": 3.4970838461126865,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038171951620437957,
      "loss": 5.7354,
      "mean_token_accuracy": 0.17125069946050644,
      "num_tokens": 70759154.0,
      "step": 32680
    },
    {
      "entropy": 6.484172201156616,
      "epoch": 3.4976189202204506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038168585560264124,
      "loss": 5.7897,
      "mean_token_accuracy": 0.17464125901460648,
      "num_tokens": 70770650.0,
      "step": 32685
    },
    {
      "entropy": 6.535211801528931,
      "epoch": 3.4981539943282143,
      "grad_norm": 1.125,
      "learning_rate": 0.0003816521919203639,
      "loss": 5.7947,
      "mean_token_accuracy": 0.17797617614269257,
      "num_tokens": 70781254.0,
      "step": 32690
    },
    {
      "entropy": 6.523111820220947,
      "epoch": 3.4986890684359784,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038161852515851934,
      "loss": 5.7278,
      "mean_token_accuracy": 0.17979024201631547,
      "num_tokens": 70791090.0,
      "step": 32695
    },
    {
      "entropy": 6.510346174240112,
      "epoch": 3.4992241425437425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003815848553180799,
      "loss": 5.7828,
      "mean_token_accuracy": 0.1756655380129814,
      "num_tokens": 70801020.0,
      "step": 32700
    },
    {
      "entropy": 6.4535881042480465,
      "epoch": 3.499759216651506,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003815511824000179,
      "loss": 5.6956,
      "mean_token_accuracy": 0.1885507360100746,
      "num_tokens": 70811308.0,
      "step": 32705
    },
    {
      "entropy": 6.533134937286377,
      "epoch": 3.5002942907592702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038151750640530533,
      "loss": 5.8038,
      "mean_token_accuracy": 0.17003858387470244,
      "num_tokens": 70822155.0,
      "step": 32710
    },
    {
      "entropy": 6.586384248733521,
      "epoch": 3.500829364867034,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003814838273349147,
      "loss": 5.8223,
      "mean_token_accuracy": 0.175823812186718,
      "num_tokens": 70834132.0,
      "step": 32715
    },
    {
      "entropy": 6.511475658416748,
      "epoch": 3.501364438974798,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038145014518981864,
      "loss": 5.8358,
      "mean_token_accuracy": 0.16849287152290343,
      "num_tokens": 70845756.0,
      "step": 32720
    },
    {
      "entropy": 6.512059545516967,
      "epoch": 3.501899513082562,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003814164599709895,
      "loss": 5.7292,
      "mean_token_accuracy": 0.17290823236107827,
      "num_tokens": 70856672.0,
      "step": 32725
    },
    {
      "entropy": 6.56741509437561,
      "epoch": 3.5024345871903257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038138277167940006,
      "loss": 5.8079,
      "mean_token_accuracy": 0.1712247297167778,
      "num_tokens": 70867993.0,
      "step": 32730
    },
    {
      "entropy": 6.528146076202392,
      "epoch": 3.50296966129809,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003813490803160231,
      "loss": 5.777,
      "mean_token_accuracy": 0.1743644133210182,
      "num_tokens": 70878061.0,
      "step": 32735
    },
    {
      "entropy": 6.436244010925293,
      "epoch": 3.503504735405854,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038131538588183133,
      "loss": 5.7299,
      "mean_token_accuracy": 0.18433666974306107,
      "num_tokens": 70889244.0,
      "step": 32740
    },
    {
      "entropy": 6.5530777931213375,
      "epoch": 3.5040398095136176,
      "grad_norm": 1.125,
      "learning_rate": 0.0003812816883777977,
      "loss": 5.7657,
      "mean_token_accuracy": 0.17324717342853546,
      "num_tokens": 70900391.0,
      "step": 32745
    },
    {
      "entropy": 6.559083032608032,
      "epoch": 3.5045748836213817,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003812479878048953,
      "loss": 5.6868,
      "mean_token_accuracy": 0.18228699415922164,
      "num_tokens": 70910743.0,
      "step": 32750
    },
    {
      "entropy": 6.5126344680786135,
      "epoch": 3.5051099577291454,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038121428416409725,
      "loss": 5.821,
      "mean_token_accuracy": 0.1796654537320137,
      "num_tokens": 70922126.0,
      "step": 32755
    },
    {
      "entropy": 6.5546306610107425,
      "epoch": 3.5056450318369095,
      "grad_norm": 1.125,
      "learning_rate": 0.0003811805774563765,
      "loss": 5.8176,
      "mean_token_accuracy": 0.17223797887563705,
      "num_tokens": 70932657.0,
      "step": 32760
    },
    {
      "entropy": 6.517896413803101,
      "epoch": 3.506180105944673,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003811468676827066,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17599011808633805,
      "num_tokens": 70943609.0,
      "step": 32765
    },
    {
      "entropy": 6.538372468948364,
      "epoch": 3.506715180052437,
      "grad_norm": 1.25,
      "learning_rate": 0.0003811131548440608,
      "loss": 5.819,
      "mean_token_accuracy": 0.1757177546620369,
      "num_tokens": 70954055.0,
      "step": 32770
    },
    {
      "entropy": 6.472061204910278,
      "epoch": 3.5072502541602013,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038107943894141256,
      "loss": 5.7301,
      "mean_token_accuracy": 0.17698537111282348,
      "num_tokens": 70965411.0,
      "step": 32775
    },
    {
      "entropy": 6.540686511993409,
      "epoch": 3.507785328267965,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038104571997573544,
      "loss": 5.8093,
      "mean_token_accuracy": 0.1768585279583931,
      "num_tokens": 70976310.0,
      "step": 32780
    },
    {
      "entropy": 6.56896390914917,
      "epoch": 3.508320402375729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003810119979480031,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16303891241550444,
      "num_tokens": 70986578.0,
      "step": 32785
    },
    {
      "entropy": 6.506561136245727,
      "epoch": 3.508855476483493,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003809782728591891,
      "loss": 5.7391,
      "mean_token_accuracy": 0.18655187487602234,
      "num_tokens": 70997360.0,
      "step": 32790
    },
    {
      "entropy": 6.521779108047485,
      "epoch": 3.509390550591257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003809445447102675,
      "loss": 5.7938,
      "mean_token_accuracy": 0.17266992181539537,
      "num_tokens": 71006754.0,
      "step": 32795
    },
    {
      "entropy": 6.530532741546631,
      "epoch": 3.509925624699021,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038091081350221203,
      "loss": 5.7455,
      "mean_token_accuracy": 0.17580094039440156,
      "num_tokens": 71017842.0,
      "step": 32800
    },
    {
      "entropy": 6.4968352794647215,
      "epoch": 3.5104606988067846,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038087707923599657,
      "loss": 5.7818,
      "mean_token_accuracy": 0.1768221452832222,
      "num_tokens": 71028622.0,
      "step": 32805
    },
    {
      "entropy": 6.429018163681031,
      "epoch": 3.5109957729145487,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038084334191259547,
      "loss": 5.6752,
      "mean_token_accuracy": 0.19451872557401656,
      "num_tokens": 71041212.0,
      "step": 32810
    },
    {
      "entropy": 6.548208904266358,
      "epoch": 3.5115308470223123,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003808096015329827,
      "loss": 5.8212,
      "mean_token_accuracy": 0.1733535870909691,
      "num_tokens": 71051394.0,
      "step": 32815
    },
    {
      "entropy": 6.512418270111084,
      "epoch": 3.5120659211300764,
      "grad_norm": 1.25,
      "learning_rate": 0.0003807758580981325,
      "loss": 5.7713,
      "mean_token_accuracy": 0.16816064715385437,
      "num_tokens": 71062099.0,
      "step": 32820
    },
    {
      "entropy": 6.60342230796814,
      "epoch": 3.5126009952378405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038074211160901933,
      "loss": 5.861,
      "mean_token_accuracy": 0.16894126385450364,
      "num_tokens": 71072647.0,
      "step": 32825
    },
    {
      "entropy": 6.475293445587158,
      "epoch": 3.513136069345604,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003807083620666175,
      "loss": 5.7436,
      "mean_token_accuracy": 0.1858312100172043,
      "num_tokens": 71083261.0,
      "step": 32830
    },
    {
      "entropy": 6.558720064163208,
      "epoch": 3.5136711434533683,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038067460947190156,
      "loss": 5.8334,
      "mean_token_accuracy": 0.1744430974125862,
      "num_tokens": 71094427.0,
      "step": 32835
    },
    {
      "entropy": 6.5860436916351315,
      "epoch": 3.5142062175611324,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003806408538258462,
      "loss": 5.7768,
      "mean_token_accuracy": 0.17570458799600602,
      "num_tokens": 71103985.0,
      "step": 32840
    },
    {
      "entropy": 6.561493682861328,
      "epoch": 3.514741291668896,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003806070951294259,
      "loss": 5.7676,
      "mean_token_accuracy": 0.1735914722084999,
      "num_tokens": 71114748.0,
      "step": 32845
    },
    {
      "entropy": 6.52962212562561,
      "epoch": 3.51527636577666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003805733333836158,
      "loss": 5.7698,
      "mean_token_accuracy": 0.17553718239068986,
      "num_tokens": 71126074.0,
      "step": 32850
    },
    {
      "entropy": 6.517365169525147,
      "epoch": 3.5158114398844242,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003805395685893903,
      "loss": 5.8401,
      "mean_token_accuracy": 0.17023670822381973,
      "num_tokens": 71136471.0,
      "step": 32855
    },
    {
      "entropy": 6.541059970855713,
      "epoch": 3.516346513992188,
      "grad_norm": 1.25,
      "learning_rate": 0.0003805058007477247,
      "loss": 5.865,
      "mean_token_accuracy": 0.17186566740274428,
      "num_tokens": 71147348.0,
      "step": 32860
    },
    {
      "entropy": 6.587239456176758,
      "epoch": 3.516881588099952,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038047202985959394,
      "loss": 5.8343,
      "mean_token_accuracy": 0.16868561506271362,
      "num_tokens": 71158015.0,
      "step": 32865
    },
    {
      "entropy": 6.573419713973999,
      "epoch": 3.5174166622077156,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003804382559259731,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1761852040886879,
      "num_tokens": 71169271.0,
      "step": 32870
    },
    {
      "entropy": 6.499637174606323,
      "epoch": 3.5179517363154797,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003804044789478374,
      "loss": 5.7523,
      "mean_token_accuracy": 0.17811030596494676,
      "num_tokens": 71179493.0,
      "step": 32875
    },
    {
      "entropy": 6.5366286277771,
      "epoch": 3.5184868104232434,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003803706989261622,
      "loss": 5.7827,
      "mean_token_accuracy": 0.18242678344249724,
      "num_tokens": 71190117.0,
      "step": 32880
    },
    {
      "entropy": 6.559920310974121,
      "epoch": 3.5190218845310075,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003803369158619229,
      "loss": 5.7793,
      "mean_token_accuracy": 0.16978861540555953,
      "num_tokens": 71200177.0,
      "step": 32885
    },
    {
      "entropy": 6.5351502895355225,
      "epoch": 3.5195569586387716,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038030312975609494,
      "loss": 5.811,
      "mean_token_accuracy": 0.17298413962125778,
      "num_tokens": 71211505.0,
      "step": 32890
    },
    {
      "entropy": 6.557123804092408,
      "epoch": 3.5200920327465353,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003802693406096539,
      "loss": 5.807,
      "mean_token_accuracy": 0.17865607291460037,
      "num_tokens": 71221055.0,
      "step": 32895
    },
    {
      "entropy": 6.583952856063843,
      "epoch": 3.5206271068542994,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003802355484235756,
      "loss": 5.8424,
      "mean_token_accuracy": 0.1652584493160248,
      "num_tokens": 71230702.0,
      "step": 32900
    },
    {
      "entropy": 6.5218480110168455,
      "epoch": 3.5211621809620635,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003802017531988354,
      "loss": 5.6997,
      "mean_token_accuracy": 0.17133854627609252,
      "num_tokens": 71241665.0,
      "step": 32905
    },
    {
      "entropy": 6.552606725692749,
      "epoch": 3.521697255069827,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038016795493640946,
      "loss": 5.8373,
      "mean_token_accuracy": 0.17065104991197586,
      "num_tokens": 71252662.0,
      "step": 32910
    },
    {
      "entropy": 6.550380754470825,
      "epoch": 3.522232329177591,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003801341536372736,
      "loss": 5.7587,
      "mean_token_accuracy": 0.17488544136285783,
      "num_tokens": 71263198.0,
      "step": 32915
    },
    {
      "entropy": 6.501708698272705,
      "epoch": 3.522767403285355,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038010034930240376,
      "loss": 5.7309,
      "mean_token_accuracy": 0.17825866788625716,
      "num_tokens": 71274464.0,
      "step": 32920
    },
    {
      "entropy": 6.493096208572387,
      "epoch": 3.523302477393119,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003800665419327762,
      "loss": 5.8059,
      "mean_token_accuracy": 0.17719385176897048,
      "num_tokens": 71285897.0,
      "step": 32925
    },
    {
      "entropy": 6.538716554641724,
      "epoch": 3.5238375515008826,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003800327315293669,
      "loss": 5.7416,
      "mean_token_accuracy": 0.17105747908353805,
      "num_tokens": 71296374.0,
      "step": 32930
    },
    {
      "entropy": 6.555496597290039,
      "epoch": 3.5243726256086467,
      "grad_norm": 1.25,
      "learning_rate": 0.00037999891809315233,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17886407673358917,
      "num_tokens": 71307401.0,
      "step": 32935
    },
    {
      "entropy": 6.506940078735352,
      "epoch": 3.524907699716411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037996510162510873,
      "loss": 5.7296,
      "mean_token_accuracy": 0.17698510140180587,
      "num_tokens": 71318261.0,
      "step": 32940
    },
    {
      "entropy": 6.455945014953613,
      "epoch": 3.5254427738241745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003799312821262126,
      "loss": 5.7131,
      "mean_token_accuracy": 0.18376672714948655,
      "num_tokens": 71328911.0,
      "step": 32945
    },
    {
      "entropy": 6.549301862716675,
      "epoch": 3.5259778479319386,
      "grad_norm": 1.375,
      "learning_rate": 0.00037989745959744037,
      "loss": 5.7846,
      "mean_token_accuracy": 0.17449821084737777,
      "num_tokens": 71339783.0,
      "step": 32950
    },
    {
      "entropy": 6.5763671875,
      "epoch": 3.5265129220397027,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037986363403976875,
      "loss": 5.7753,
      "mean_token_accuracy": 0.17007538229227065,
      "num_tokens": 71349561.0,
      "step": 32955
    },
    {
      "entropy": 6.507570838928222,
      "epoch": 3.5270479961474663,
      "grad_norm": 1.25,
      "learning_rate": 0.0003798298054541744,
      "loss": 5.7598,
      "mean_token_accuracy": 0.17853962630033493,
      "num_tokens": 71360037.0,
      "step": 32960
    },
    {
      "entropy": 6.41061110496521,
      "epoch": 3.5275830702552304,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003797959738416343,
      "loss": 5.6279,
      "mean_token_accuracy": 0.19612410962581633,
      "num_tokens": 71371484.0,
      "step": 32965
    },
    {
      "entropy": 6.546814966201782,
      "epoch": 3.5281181443629945,
      "grad_norm": 1.234375,
      "learning_rate": 0.000379762139203125,
      "loss": 5.8289,
      "mean_token_accuracy": 0.17612827718257903,
      "num_tokens": 71382818.0,
      "step": 32970
    },
    {
      "entropy": 6.533816289901734,
      "epoch": 3.528653218470758,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003797283015396238,
      "loss": 5.7938,
      "mean_token_accuracy": 0.17920994013547897,
      "num_tokens": 71393938.0,
      "step": 32975
    },
    {
      "entropy": 6.515238094329834,
      "epoch": 3.5291882925785223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037969446085210753,
      "loss": 5.7543,
      "mean_token_accuracy": 0.18086332678794861,
      "num_tokens": 71403678.0,
      "step": 32980
    },
    {
      "entropy": 6.492973327636719,
      "epoch": 3.529723366686286,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037966061714155344,
      "loss": 5.762,
      "mean_token_accuracy": 0.17325435876846312,
      "num_tokens": 71414877.0,
      "step": 32985
    },
    {
      "entropy": 6.503774738311767,
      "epoch": 3.53025844079405,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037962677040893867,
      "loss": 5.74,
      "mean_token_accuracy": 0.17412215024232863,
      "num_tokens": 71424823.0,
      "step": 32990
    },
    {
      "entropy": 6.527900505065918,
      "epoch": 3.5307935149018137,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003795929206552407,
      "loss": 5.7832,
      "mean_token_accuracy": 0.17445069849491118,
      "num_tokens": 71434664.0,
      "step": 32995
    },
    {
      "entropy": 6.509808826446533,
      "epoch": 3.531328589009578,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003795590678814368,
      "loss": 5.7516,
      "mean_token_accuracy": 0.1757136732339859,
      "num_tokens": 71444556.0,
      "step": 33000
    },
    {
      "epoch": 3.531328589009578,
      "eval_entropy": 6.272682781168071,
      "eval_loss": 5.892104148864746,
      "eval_mean_token_accuracy": 0.17736740394713824,
      "eval_num_tokens": 71444556.0,
      "eval_runtime": 17.4261,
      "eval_samples_per_second": 1703.137,
      "eval_steps_per_second": 212.899,
      "step": 33000
    }
  ],
  "logging_steps": 5,
  "max_steps": 93440,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.557171636406272e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}