{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.3978607487379415,
  "eval_steps": 3000,
  "global_step": 54000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 8.115966701507569,
      "epoch": 0.00049982506122857,
      "grad_norm": 1.265625,
      "learning_rate": 2e-06,
      "loss": 7.9779,
      "mean_token_accuracy": 0.09388983175158501,
      "num_tokens": 9036.0,
      "step": 5
    },
    {
      "entropy": 8.105617094039918,
      "epoch": 0.00099965012245714,
      "grad_norm": 1.1484375,
      "learning_rate": 4.5e-06,
      "loss": 8.1027,
      "mean_token_accuracy": 0.08293505646288395,
      "num_tokens": 19391.0,
      "step": 10
    },
    {
      "entropy": 8.137477827072143,
      "epoch": 0.00149947518368571,
      "grad_norm": 1.2734375,
      "learning_rate": 7e-06,
      "loss": 7.9326,
      "mean_token_accuracy": 0.08243411704897881,
      "num_tokens": 27896.0,
      "step": 15
    },
    {
      "entropy": 8.191983127593994,
      "epoch": 0.00199930024491428,
      "grad_norm": 1.09375,
      "learning_rate": 9.5e-06,
      "loss": 7.9817,
      "mean_token_accuracy": 0.08287211582064628,
      "num_tokens": 37280.0,
      "step": 20
    },
    {
      "entropy": 8.175366497039795,
      "epoch": 0.00249912530614285,
      "grad_norm": 1.046875,
      "learning_rate": 1.2e-05,
      "loss": 7.8896,
      "mean_token_accuracy": 0.0876060277223587,
      "num_tokens": 46010.0,
      "step": 25
    },
    {
      "entropy": 8.177831649780273,
      "epoch": 0.00299895036737142,
      "grad_norm": 1.0546875,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.9504,
      "mean_token_accuracy": 0.08859384059906006,
      "num_tokens": 55972.0,
      "step": 30
    },
    {
      "entropy": 8.151965236663818,
      "epoch": 0.00349877542859999,
      "grad_norm": 0.93359375,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.9227,
      "mean_token_accuracy": 0.08379799723625184,
      "num_tokens": 66142.0,
      "step": 35
    },
    {
      "entropy": 8.180653858184815,
      "epoch": 0.00399860048982856,
      "grad_norm": 1.1796875,
      "learning_rate": 1.95e-05,
      "loss": 7.8994,
      "mean_token_accuracy": 0.08740837424993515,
      "num_tokens": 74978.0,
      "step": 40
    },
    {
      "entropy": 8.107624816894532,
      "epoch": 0.00449842555105713,
      "grad_norm": 1.2109375,
      "learning_rate": 2.2e-05,
      "loss": 7.9414,
      "mean_token_accuracy": 0.09354466870427132,
      "num_tokens": 84153.0,
      "step": 45
    },
    {
      "entropy": 8.170318412780762,
      "epoch": 0.0049982506122857,
      "grad_norm": 1.0234375,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.9516,
      "mean_token_accuracy": 0.08247044086456298,
      "num_tokens": 94463.0,
      "step": 50
    },
    {
      "entropy": 8.15045108795166,
      "epoch": 0.00549807567351427,
      "grad_norm": 0.94140625,
      "learning_rate": 2.7e-05,
      "loss": 7.8657,
      "mean_token_accuracy": 0.08542923405766487,
      "num_tokens": 104182.0,
      "step": 55
    },
    {
      "entropy": 8.115217018127442,
      "epoch": 0.00599790073474284,
      "grad_norm": 1.1171875,
      "learning_rate": 2.95e-05,
      "loss": 7.8856,
      "mean_token_accuracy": 0.09453774318099022,
      "num_tokens": 114058.0,
      "step": 60
    },
    {
      "entropy": 8.119699668884277,
      "epoch": 0.00649772579597141,
      "grad_norm": 0.95703125,
      "learning_rate": 3.2e-05,
      "loss": 7.9511,
      "mean_token_accuracy": 0.08501083850860595,
      "num_tokens": 124154.0,
      "step": 65
    },
    {
      "entropy": 8.12986068725586,
      "epoch": 0.00699755085719998,
      "grad_norm": 1.171875,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.8663,
      "mean_token_accuracy": 0.09313991740345955,
      "num_tokens": 132840.0,
      "step": 70
    },
    {
      "entropy": 8.146929073333741,
      "epoch": 0.00749737591842855,
      "grad_norm": 1.109375,
      "learning_rate": 3.7e-05,
      "loss": 7.9197,
      "mean_token_accuracy": 0.08960949406027793,
      "num_tokens": 143070.0,
      "step": 75
    },
    {
      "entropy": 8.190567016601562,
      "epoch": 0.00799720097965712,
      "grad_norm": 0.98828125,
      "learning_rate": 3.95e-05,
      "loss": 7.9273,
      "mean_token_accuracy": 0.09039379954338074,
      "num_tokens": 153319.0,
      "step": 80
    },
    {
      "entropy": 8.190423202514648,
      "epoch": 0.00849702604088569,
      "grad_norm": 1.09375,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.9848,
      "mean_token_accuracy": 0.08413397669792175,
      "num_tokens": 163152.0,
      "step": 85
    },
    {
      "entropy": 8.127420330047608,
      "epoch": 0.00899685110211426,
      "grad_norm": 0.94921875,
      "learning_rate": 4.45e-05,
      "loss": 7.8803,
      "mean_token_accuracy": 0.09164257049560547,
      "num_tokens": 173321.0,
      "step": 90
    },
    {
      "entropy": 8.158605098724365,
      "epoch": 0.00949667616334283,
      "grad_norm": 1.0859375,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.9,
      "mean_token_accuracy": 0.0946331799030304,
      "num_tokens": 182827.0,
      "step": 95
    },
    {
      "entropy": 8.080010318756104,
      "epoch": 0.0099965012245714,
      "grad_norm": 0.95703125,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.9007,
      "mean_token_accuracy": 0.08509955257177353,
      "num_tokens": 191776.0,
      "step": 100
    },
    {
      "entropy": 8.090544700622559,
      "epoch": 0.01049632628579997,
      "grad_norm": 0.98828125,
      "learning_rate": 5.2e-05,
      "loss": 7.8874,
      "mean_token_accuracy": 0.0908079981803894,
      "num_tokens": 201891.0,
      "step": 105
    },
    {
      "entropy": 8.260288429260253,
      "epoch": 0.01099615134702854,
      "grad_norm": 1.1015625,
      "learning_rate": 5.45e-05,
      "loss": 7.9196,
      "mean_token_accuracy": 0.09156425222754479,
      "num_tokens": 210385.0,
      "step": 110
    },
    {
      "entropy": 8.154400157928468,
      "epoch": 0.01149597640825711,
      "grad_norm": 1.1171875,
      "learning_rate": 5.7e-05,
      "loss": 7.8798,
      "mean_token_accuracy": 0.09079133942723275,
      "num_tokens": 220098.0,
      "step": 115
    },
    {
      "entropy": 8.051148319244385,
      "epoch": 0.01199580146948568,
      "grad_norm": 1.1484375,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.9117,
      "mean_token_accuracy": 0.09130131006240845,
      "num_tokens": 230703.0,
      "step": 120
    },
    {
      "entropy": 8.107196235656739,
      "epoch": 0.01249562653071425,
      "grad_norm": 1.015625,
      "learning_rate": 6.2e-05,
      "loss": 7.8365,
      "mean_token_accuracy": 0.09351907223463059,
      "num_tokens": 239545.0,
      "step": 125
    },
    {
      "entropy": 8.116232061386109,
      "epoch": 0.01299545159194282,
      "grad_norm": 0.95703125,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.8594,
      "mean_token_accuracy": 0.08991236239671707,
      "num_tokens": 249726.0,
      "step": 130
    },
    {
      "entropy": 8.180912399291993,
      "epoch": 0.01349527665317139,
      "grad_norm": 0.9609375,
      "learning_rate": 6.7e-05,
      "loss": 8.0195,
      "mean_token_accuracy": 0.08512845635414124,
      "num_tokens": 259491.0,
      "step": 135
    },
    {
      "entropy": 8.232371520996093,
      "epoch": 0.01399510171439996,
      "grad_norm": 1.046875,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.9204,
      "mean_token_accuracy": 0.08833294585347176,
      "num_tokens": 268239.0,
      "step": 140
    },
    {
      "entropy": 8.114497804641724,
      "epoch": 0.01449492677562853,
      "grad_norm": 1.0859375,
      "learning_rate": 7.2e-05,
      "loss": 7.8927,
      "mean_token_accuracy": 0.08900318369269371,
      "num_tokens": 276767.0,
      "step": 145
    },
    {
      "entropy": 8.102681970596313,
      "epoch": 0.0149947518368571,
      "grad_norm": 1.0,
      "learning_rate": 7.45e-05,
      "loss": 7.8269,
      "mean_token_accuracy": 0.09594093039631843,
      "num_tokens": 286072.0,
      "step": 150
    },
    {
      "entropy": 8.218955421447754,
      "epoch": 0.01549457689808567,
      "grad_norm": 1.09375,
      "learning_rate": 7.7e-05,
      "loss": 7.8935,
      "mean_token_accuracy": 0.08453865349292755,
      "num_tokens": 295178.0,
      "step": 155
    },
    {
      "entropy": 8.177178192138673,
      "epoch": 0.01599440195931424,
      "grad_norm": 1.0625,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.9645,
      "mean_token_accuracy": 0.08729357868432999,
      "num_tokens": 305537.0,
      "step": 160
    },
    {
      "entropy": 8.17558240890503,
      "epoch": 0.01649422702054281,
      "grad_norm": 1.15625,
      "learning_rate": 8.2e-05,
      "loss": 7.8578,
      "mean_token_accuracy": 0.09048740342259406,
      "num_tokens": 315316.0,
      "step": 165
    },
    {
      "entropy": 8.087751197814942,
      "epoch": 0.01699405208177138,
      "grad_norm": 1.1015625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8939,
      "mean_token_accuracy": 0.09201501980423928,
      "num_tokens": 325089.0,
      "step": 170
    },
    {
      "entropy": 8.224938583374023,
      "epoch": 0.01749387714299995,
      "grad_norm": 1.5234375,
      "learning_rate": 8.7e-05,
      "loss": 7.8357,
      "mean_token_accuracy": 0.09008259326219559,
      "num_tokens": 334009.0,
      "step": 175
    },
    {
      "entropy": 8.072714281082153,
      "epoch": 0.01799370220422852,
      "grad_norm": 1.1953125,
      "learning_rate": 8.95e-05,
      "loss": 7.8322,
      "mean_token_accuracy": 0.09141119942069054,
      "num_tokens": 342535.0,
      "step": 180
    },
    {
      "entropy": 8.01605257987976,
      "epoch": 0.01849352726545709,
      "grad_norm": 1.1875,
      "learning_rate": 9.2e-05,
      "loss": 7.8298,
      "mean_token_accuracy": 0.09161713123321533,
      "num_tokens": 352324.0,
      "step": 185
    },
    {
      "entropy": 8.16107406616211,
      "epoch": 0.01899335232668566,
      "grad_norm": 1.1171875,
      "learning_rate": 9.45e-05,
      "loss": 7.8893,
      "mean_token_accuracy": 0.08870044574141503,
      "num_tokens": 362824.0,
      "step": 190
    },
    {
      "entropy": 8.062242889404297,
      "epoch": 0.01949317738791423,
      "grad_norm": 1.1015625,
      "learning_rate": 9.7e-05,
      "loss": 7.8395,
      "mean_token_accuracy": 0.08829269260168075,
      "num_tokens": 373056.0,
      "step": 195
    },
    {
      "entropy": 8.146231651306152,
      "epoch": 0.0199930024491428,
      "grad_norm": 0.98828125,
      "learning_rate": 9.95e-05,
      "loss": 7.8091,
      "mean_token_accuracy": 0.0935797117650509,
      "num_tokens": 382448.0,
      "step": 200
    },
    {
      "entropy": 8.083828449249268,
      "epoch": 0.02049282751037137,
      "grad_norm": 1.171875,
      "learning_rate": 0.000102,
      "loss": 7.894,
      "mean_token_accuracy": 0.09289678484201432,
      "num_tokens": 393640.0,
      "step": 205
    },
    {
      "entropy": 8.234550094604492,
      "epoch": 0.02099265257159994,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.9116,
      "mean_token_accuracy": 0.08593777976930142,
      "num_tokens": 403129.0,
      "step": 210
    },
    {
      "entropy": 8.09738302230835,
      "epoch": 0.02149247763282851,
      "grad_norm": 1.21875,
      "learning_rate": 0.000107,
      "loss": 7.8569,
      "mean_token_accuracy": 0.09456023722887039,
      "num_tokens": 412673.0,
      "step": 215
    },
    {
      "entropy": 8.173221969604493,
      "epoch": 0.02199230269405708,
      "grad_norm": 1.046875,
      "learning_rate": 0.0001095,
      "loss": 7.8759,
      "mean_token_accuracy": 0.09021985307335853,
      "num_tokens": 422960.0,
      "step": 220
    },
    {
      "entropy": 8.105310249328614,
      "epoch": 0.02249212775528565,
      "grad_norm": 1.171875,
      "learning_rate": 0.000112,
      "loss": 7.8843,
      "mean_token_accuracy": 0.0898541733622551,
      "num_tokens": 433248.0,
      "step": 225
    },
    {
      "entropy": 8.135746097564697,
      "epoch": 0.02299195281651422,
      "grad_norm": 1.234375,
      "learning_rate": 0.0001145,
      "loss": 7.8131,
      "mean_token_accuracy": 0.09401410818099976,
      "num_tokens": 441738.0,
      "step": 230
    },
    {
      "entropy": 8.084957170486451,
      "epoch": 0.02349177787774279,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.9135,
      "mean_token_accuracy": 0.09210107401013375,
      "num_tokens": 451894.0,
      "step": 235
    },
    {
      "entropy": 8.101123428344726,
      "epoch": 0.02399160293897136,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.8755,
      "mean_token_accuracy": 0.09426127821207046,
      "num_tokens": 462829.0,
      "step": 240
    },
    {
      "entropy": 8.216189861297607,
      "epoch": 0.02449142800019993,
      "grad_norm": 1.15625,
      "learning_rate": 0.000122,
      "loss": 7.8754,
      "mean_token_accuracy": 0.08819158300757408,
      "num_tokens": 472034.0,
      "step": 245
    },
    {
      "entropy": 8.028231763839722,
      "epoch": 0.0249912530614285,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0001245,
      "loss": 7.9268,
      "mean_token_accuracy": 0.08805325403809547,
      "num_tokens": 481963.0,
      "step": 250
    },
    {
      "entropy": 8.267205619812012,
      "epoch": 0.02549107812265707,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000127,
      "loss": 7.8241,
      "mean_token_accuracy": 0.09464174434542656,
      "num_tokens": 491562.0,
      "step": 255
    },
    {
      "entropy": 7.98403491973877,
      "epoch": 0.02599090318388564,
      "grad_norm": 1.71875,
      "learning_rate": 0.0001295,
      "loss": 7.7673,
      "mean_token_accuracy": 0.10128638967871666,
      "num_tokens": 500884.0,
      "step": 260
    },
    {
      "entropy": 8.152617645263671,
      "epoch": 0.026490728245114212,
      "grad_norm": 1.234375,
      "learning_rate": 0.000132,
      "loss": 7.8325,
      "mean_token_accuracy": 0.08933599889278412,
      "num_tokens": 510287.0,
      "step": 265
    },
    {
      "entropy": 7.9761005401611325,
      "epoch": 0.02699055330634278,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.7612,
      "mean_token_accuracy": 0.09821977689862252,
      "num_tokens": 519633.0,
      "step": 270
    },
    {
      "entropy": 8.178555679321288,
      "epoch": 0.02749037836757135,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.8645,
      "mean_token_accuracy": 0.08926928378641605,
      "num_tokens": 528999.0,
      "step": 275
    },
    {
      "entropy": 8.06614933013916,
      "epoch": 0.02799020342879992,
      "grad_norm": 1.203125,
      "learning_rate": 0.0001395,
      "loss": 7.8455,
      "mean_token_accuracy": 0.09183734282851219,
      "num_tokens": 538403.0,
      "step": 280
    },
    {
      "entropy": 8.094897556304932,
      "epoch": 0.02849002849002849,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.8199,
      "mean_token_accuracy": 0.09269031435251236,
      "num_tokens": 547716.0,
      "step": 285
    },
    {
      "entropy": 8.069192457199097,
      "epoch": 0.02898985355125706,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0001445,
      "loss": 7.8119,
      "mean_token_accuracy": 0.09283587783575058,
      "num_tokens": 557714.0,
      "step": 290
    },
    {
      "entropy": 8.185009765625,
      "epoch": 0.02948967861248563,
      "grad_norm": 1.28125,
      "learning_rate": 0.000147,
      "loss": 7.8673,
      "mean_token_accuracy": 0.09096960946917534,
      "num_tokens": 568591.0,
      "step": 295
    },
    {
      "entropy": 8.038140535354614,
      "epoch": 0.0299895036737142,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001495,
      "loss": 7.7626,
      "mean_token_accuracy": 0.09647799134254456,
      "num_tokens": 578435.0,
      "step": 300
    },
    {
      "entropy": 8.096207904815675,
      "epoch": 0.03048932873494277,
      "grad_norm": 1.234375,
      "learning_rate": 0.000152,
      "loss": 7.8044,
      "mean_token_accuracy": 0.09069146290421486,
      "num_tokens": 587461.0,
      "step": 305
    },
    {
      "entropy": 8.098102331161499,
      "epoch": 0.03098915379617134,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.8719,
      "mean_token_accuracy": 0.08908754512667656,
      "num_tokens": 597004.0,
      "step": 310
    },
    {
      "entropy": 8.09911379814148,
      "epoch": 0.03148897885739991,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000157,
      "loss": 7.8298,
      "mean_token_accuracy": 0.09009509831666947,
      "num_tokens": 607831.0,
      "step": 315
    },
    {
      "entropy": 8.104575824737548,
      "epoch": 0.03198880391862848,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0001595,
      "loss": 7.8515,
      "mean_token_accuracy": 0.09431877583265305,
      "num_tokens": 618008.0,
      "step": 320
    },
    {
      "entropy": 8.003451538085937,
      "epoch": 0.03248862897985705,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000162,
      "loss": 7.7285,
      "mean_token_accuracy": 0.09802025258541107,
      "num_tokens": 627970.0,
      "step": 325
    },
    {
      "entropy": 8.075142240524292,
      "epoch": 0.03298845404108562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.8052,
      "mean_token_accuracy": 0.09450135603547097,
      "num_tokens": 638059.0,
      "step": 330
    },
    {
      "entropy": 8.108322143554688,
      "epoch": 0.03348827910231419,
      "grad_norm": 1.28125,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.8892,
      "mean_token_accuracy": 0.08549424782395362,
      "num_tokens": 648169.0,
      "step": 335
    },
    {
      "entropy": 8.106694269180299,
      "epoch": 0.03398810416354276,
      "grad_norm": 0.9140625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.8806,
      "mean_token_accuracy": 0.08569936826825142,
      "num_tokens": 659351.0,
      "step": 340
    },
    {
      "entropy": 8.085410118103027,
      "epoch": 0.03448792922477133,
      "grad_norm": 1.125,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.7017,
      "mean_token_accuracy": 0.09293192699551582,
      "num_tokens": 669602.0,
      "step": 345
    },
    {
      "entropy": 8.03732442855835,
      "epoch": 0.0349877542859999,
      "grad_norm": 1.390625,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.8203,
      "mean_token_accuracy": 0.09097881987690926,
      "num_tokens": 679497.0,
      "step": 350
    },
    {
      "entropy": 8.093275213241578,
      "epoch": 0.03548757934722847,
      "grad_norm": 1.09375,
      "learning_rate": 0.000177,
      "loss": 7.7341,
      "mean_token_accuracy": 0.09658810496330261,
      "num_tokens": 689020.0,
      "step": 355
    },
    {
      "entropy": 7.989745807647705,
      "epoch": 0.03598740440845704,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0001795,
      "loss": 7.735,
      "mean_token_accuracy": 0.09265206158161163,
      "num_tokens": 697505.0,
      "step": 360
    },
    {
      "entropy": 8.095442819595338,
      "epoch": 0.03648722946968561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000182,
      "loss": 7.7433,
      "mean_token_accuracy": 0.0910780519247055,
      "num_tokens": 707646.0,
      "step": 365
    },
    {
      "entropy": 7.930270195007324,
      "epoch": 0.03698705453091418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0001845,
      "loss": 7.671,
      "mean_token_accuracy": 0.09586982429027557,
      "num_tokens": 717345.0,
      "step": 370
    },
    {
      "entropy": 8.013105392456055,
      "epoch": 0.03748687959214275,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000187,
      "loss": 7.7725,
      "mean_token_accuracy": 0.09069561287760734,
      "num_tokens": 727785.0,
      "step": 375
    },
    {
      "entropy": 8.108489894866944,
      "epoch": 0.03798670465337132,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001895,
      "loss": 7.7723,
      "mean_token_accuracy": 0.09322713911533356,
      "num_tokens": 737740.0,
      "step": 380
    },
    {
      "entropy": 7.981899881362915,
      "epoch": 0.03848652971459989,
      "grad_norm": 1.28125,
      "learning_rate": 0.000192,
      "loss": 7.7423,
      "mean_token_accuracy": 0.09115721732378006,
      "num_tokens": 747026.0,
      "step": 385
    },
    {
      "entropy": 7.915007829666138,
      "epoch": 0.03898635477582846,
      "grad_norm": 1.15625,
      "learning_rate": 0.0001945,
      "loss": 7.7139,
      "mean_token_accuracy": 0.09398729205131531,
      "num_tokens": 756798.0,
      "step": 390
    },
    {
      "entropy": 8.194568920135499,
      "epoch": 0.03948617983705703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.7745,
      "mean_token_accuracy": 0.08935758769512177,
      "num_tokens": 765730.0,
      "step": 395
    },
    {
      "entropy": 7.9383749008178714,
      "epoch": 0.0399860048982856,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.7555,
      "mean_token_accuracy": 0.0994470827281475,
      "num_tokens": 776330.0,
      "step": 400
    },
    {
      "entropy": 8.0953604221344,
      "epoch": 0.04048582995951417,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000202,
      "loss": 7.8216,
      "mean_token_accuracy": 0.08922872990369797,
      "num_tokens": 785840.0,
      "step": 405
    },
    {
      "entropy": 7.943257141113281,
      "epoch": 0.04098565502074274,
      "grad_norm": 1.1875,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.6662,
      "mean_token_accuracy": 0.0952746793627739,
      "num_tokens": 795126.0,
      "step": 410
    },
    {
      "entropy": 8.076482248306274,
      "epoch": 0.04148548008197131,
      "grad_norm": 1.140625,
      "learning_rate": 0.000207,
      "loss": 7.8195,
      "mean_token_accuracy": 0.08982234001159668,
      "num_tokens": 805766.0,
      "step": 415
    },
    {
      "entropy": 7.9645508289337155,
      "epoch": 0.04198530514319988,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0002095,
      "loss": 7.7635,
      "mean_token_accuracy": 0.09018896147608757,
      "num_tokens": 815202.0,
      "step": 420
    },
    {
      "entropy": 8.063913249969483,
      "epoch": 0.04248513020442845,
      "grad_norm": 1.265625,
      "learning_rate": 0.000212,
      "loss": 7.7291,
      "mean_token_accuracy": 0.0924349457025528,
      "num_tokens": 825065.0,
      "step": 425
    },
    {
      "entropy": 7.959709167480469,
      "epoch": 0.04298495526565702,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002145,
      "loss": 7.7472,
      "mean_token_accuracy": 0.09275062531232833,
      "num_tokens": 834088.0,
      "step": 430
    },
    {
      "entropy": 7.983949851989746,
      "epoch": 0.04348478032688559,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.7716,
      "mean_token_accuracy": 0.09366122782230377,
      "num_tokens": 843653.0,
      "step": 435
    },
    {
      "entropy": 7.977571678161621,
      "epoch": 0.04398460538811416,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002195,
      "loss": 7.6356,
      "mean_token_accuracy": 0.09651589393615723,
      "num_tokens": 853248.0,
      "step": 440
    },
    {
      "entropy": 8.041356611251832,
      "epoch": 0.04448443044934273,
      "grad_norm": 1.28125,
      "learning_rate": 0.000222,
      "loss": 7.8109,
      "mean_token_accuracy": 0.0912716869264841,
      "num_tokens": 863814.0,
      "step": 445
    },
    {
      "entropy": 7.979821443557739,
      "epoch": 0.0449842555105713,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002245,
      "loss": 7.7105,
      "mean_token_accuracy": 0.09547068625688553,
      "num_tokens": 874030.0,
      "step": 450
    },
    {
      "entropy": 7.984935569763183,
      "epoch": 0.04548408057179987,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.6921,
      "mean_token_accuracy": 0.09831217005848884,
      "num_tokens": 883006.0,
      "step": 455
    },
    {
      "entropy": 7.867571020126343,
      "epoch": 0.04598390563302844,
      "grad_norm": 1.03125,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.6585,
      "mean_token_accuracy": 0.10229817032814026,
      "num_tokens": 892953.0,
      "step": 460
    },
    {
      "entropy": 8.069862604141235,
      "epoch": 0.04648373069425701,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.7835,
      "mean_token_accuracy": 0.08745714724063873,
      "num_tokens": 904228.0,
      "step": 465
    },
    {
      "entropy": 7.877147006988525,
      "epoch": 0.04698355575548558,
      "grad_norm": 1.09375,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.6468,
      "mean_token_accuracy": 0.0926812507212162,
      "num_tokens": 915161.0,
      "step": 470
    },
    {
      "entropy": 8.05656394958496,
      "epoch": 0.04748338081671415,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000237,
      "loss": 7.7928,
      "mean_token_accuracy": 0.08807904869318009,
      "num_tokens": 925167.0,
      "step": 475
    },
    {
      "entropy": 7.888152980804444,
      "epoch": 0.04798320587794272,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002395,
      "loss": 7.666,
      "mean_token_accuracy": 0.09307035207748413,
      "num_tokens": 934742.0,
      "step": 480
    },
    {
      "entropy": 7.976586484909058,
      "epoch": 0.04848303093917129,
      "grad_norm": 1.265625,
      "learning_rate": 0.000242,
      "loss": 7.7478,
      "mean_token_accuracy": 0.0892964169383049,
      "num_tokens": 944549.0,
      "step": 485
    },
    {
      "entropy": 7.876341152191162,
      "epoch": 0.04898285600039986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002445,
      "loss": 7.629,
      "mean_token_accuracy": 0.0999104768037796,
      "num_tokens": 954668.0,
      "step": 490
    },
    {
      "entropy": 8.100045585632325,
      "epoch": 0.04948268106162843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000247,
      "loss": 7.7424,
      "mean_token_accuracy": 0.08877054788172245,
      "num_tokens": 964953.0,
      "step": 495
    },
    {
      "entropy": 7.773602247238159,
      "epoch": 0.049982506122857,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002495,
      "loss": 7.6328,
      "mean_token_accuracy": 0.09981884956359863,
      "num_tokens": 974820.0,
      "step": 500
    },
    {
      "entropy": 7.927160835266113,
      "epoch": 0.05048233118408557,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000252,
      "loss": 7.6404,
      "mean_token_accuracy": 0.09584514498710632,
      "num_tokens": 984888.0,
      "step": 505
    },
    {
      "entropy": 7.802907085418701,
      "epoch": 0.05098215624531414,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002545,
      "loss": 7.6596,
      "mean_token_accuracy": 0.0964170329272747,
      "num_tokens": 993731.0,
      "step": 510
    },
    {
      "entropy": 7.990255212783813,
      "epoch": 0.05148198130654271,
      "grad_norm": 1.21875,
      "learning_rate": 0.000257,
      "loss": 7.5766,
      "mean_token_accuracy": 0.09960651695728302,
      "num_tokens": 1003138.0,
      "step": 515
    },
    {
      "entropy": 7.7881842136383055,
      "epoch": 0.05198180636777128,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002595,
      "loss": 7.5953,
      "mean_token_accuracy": 0.09512121453881264,
      "num_tokens": 1012058.0,
      "step": 520
    },
    {
      "entropy": 7.89219651222229,
      "epoch": 0.05248163142899985,
      "grad_norm": 1.390625,
      "learning_rate": 0.000262,
      "loss": 7.614,
      "mean_token_accuracy": 0.10023076310753823,
      "num_tokens": 1020804.0,
      "step": 525
    },
    {
      "entropy": 7.951163721084595,
      "epoch": 0.052981456490228424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.6299,
      "mean_token_accuracy": 0.09837900400161743,
      "num_tokens": 1030432.0,
      "step": 530
    },
    {
      "entropy": 7.835769939422607,
      "epoch": 0.05348128155145699,
      "grad_norm": 1.125,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.5993,
      "mean_token_accuracy": 0.09920521229505538,
      "num_tokens": 1039336.0,
      "step": 535
    },
    {
      "entropy": 7.788568496704102,
      "epoch": 0.05398110661268556,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.5576,
      "mean_token_accuracy": 0.10085468515753745,
      "num_tokens": 1049314.0,
      "step": 540
    },
    {
      "entropy": 7.914288330078125,
      "epoch": 0.05448093167391413,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.6261,
      "mean_token_accuracy": 0.09385359957814217,
      "num_tokens": 1059248.0,
      "step": 545
    },
    {
      "entropy": 7.7171854972839355,
      "epoch": 0.0549807567351427,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002745,
      "loss": 7.5969,
      "mean_token_accuracy": 0.09892648234963416,
      "num_tokens": 1069307.0,
      "step": 550
    },
    {
      "entropy": 7.915244913101196,
      "epoch": 0.05548058179637127,
      "grad_norm": 1.125,
      "learning_rate": 0.000277,
      "loss": 7.5987,
      "mean_token_accuracy": 0.10670549720525742,
      "num_tokens": 1079603.0,
      "step": 555
    },
    {
      "entropy": 7.749522876739502,
      "epoch": 0.05598040685759984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002795,
      "loss": 7.5703,
      "mean_token_accuracy": 0.09797492027282714,
      "num_tokens": 1089386.0,
      "step": 560
    },
    {
      "entropy": 7.899583625793457,
      "epoch": 0.05648023191882841,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.6497,
      "mean_token_accuracy": 0.10379090011119843,
      "num_tokens": 1099501.0,
      "step": 565
    },
    {
      "entropy": 7.773384618759155,
      "epoch": 0.05698005698005698,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002845,
      "loss": 7.5484,
      "mean_token_accuracy": 0.09730408489704132,
      "num_tokens": 1108942.0,
      "step": 570
    },
    {
      "entropy": 7.850360488891601,
      "epoch": 0.05747988204128555,
      "grad_norm": 1.09375,
      "learning_rate": 0.000287,
      "loss": 7.6693,
      "mean_token_accuracy": 0.08937681168317795,
      "num_tokens": 1119484.0,
      "step": 575
    },
    {
      "entropy": 7.852486276626587,
      "epoch": 0.05797970710251412,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002895,
      "loss": 7.6211,
      "mean_token_accuracy": 0.0964928537607193,
      "num_tokens": 1130426.0,
      "step": 580
    },
    {
      "entropy": 7.850640439987183,
      "epoch": 0.05847953216374269,
      "grad_norm": 1.0625,
      "learning_rate": 0.000292,
      "loss": 7.5938,
      "mean_token_accuracy": 0.09419768303632736,
      "num_tokens": 1140917.0,
      "step": 585
    },
    {
      "entropy": 7.806917428970337,
      "epoch": 0.05897935722497126,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002945,
      "loss": 7.5431,
      "mean_token_accuracy": 0.09883978441357613,
      "num_tokens": 1149863.0,
      "step": 590
    },
    {
      "entropy": 7.7598425388336185,
      "epoch": 0.05947918228619983,
      "grad_norm": 1.109375,
      "learning_rate": 0.000297,
      "loss": 7.5827,
      "mean_token_accuracy": 0.09910288006067276,
      "num_tokens": 1158915.0,
      "step": 595
    },
    {
      "entropy": 7.7976579666137695,
      "epoch": 0.0599790073474284,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002995,
      "loss": 7.5754,
      "mean_token_accuracy": 0.09979746043682099,
      "num_tokens": 1167886.0,
      "step": 600
    },
    {
      "entropy": 7.82974419593811,
      "epoch": 0.06047883240865697,
      "grad_norm": 0.984375,
      "learning_rate": 0.000302,
      "loss": 7.6081,
      "mean_token_accuracy": 0.09990579262375832,
      "num_tokens": 1177878.0,
      "step": 605
    },
    {
      "entropy": 7.870800113677978,
      "epoch": 0.06097865746988554,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003045,
      "loss": 7.6324,
      "mean_token_accuracy": 0.0978995405137539,
      "num_tokens": 1189511.0,
      "step": 610
    },
    {
      "entropy": 7.803325939178467,
      "epoch": 0.06147848253111411,
      "grad_norm": 1.203125,
      "learning_rate": 0.000307,
      "loss": 7.6302,
      "mean_token_accuracy": 0.09554598703980446,
      "num_tokens": 1199350.0,
      "step": 615
    },
    {
      "entropy": 7.773017311096192,
      "epoch": 0.06197830759234268,
      "grad_norm": 1.46875,
      "learning_rate": 0.0003095,
      "loss": 7.5891,
      "mean_token_accuracy": 0.09778468236327172,
      "num_tokens": 1209036.0,
      "step": 620
    },
    {
      "entropy": 7.699917221069336,
      "epoch": 0.06247813265357125,
      "grad_norm": 1.1875,
      "learning_rate": 0.000312,
      "loss": 7.5951,
      "mean_token_accuracy": 0.09537834152579308,
      "num_tokens": 1218841.0,
      "step": 625
    },
    {
      "entropy": 7.926633691787719,
      "epoch": 0.06297795771479982,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003145,
      "loss": 7.6178,
      "mean_token_accuracy": 0.09458025097846985,
      "num_tokens": 1229700.0,
      "step": 630
    },
    {
      "entropy": 7.723057842254638,
      "epoch": 0.06347778277602839,
      "grad_norm": 1.234375,
      "learning_rate": 0.000317,
      "loss": 7.5944,
      "mean_token_accuracy": 0.09242647364735604,
      "num_tokens": 1239604.0,
      "step": 635
    },
    {
      "entropy": 7.595082759857178,
      "epoch": 0.06397760783725696,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003195,
      "loss": 7.4804,
      "mean_token_accuracy": 0.10766572058200836,
      "num_tokens": 1249305.0,
      "step": 640
    },
    {
      "entropy": 7.834075403213501,
      "epoch": 0.06447743289848552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000322,
      "loss": 7.6121,
      "mean_token_accuracy": 0.09711179062724114,
      "num_tokens": 1261289.0,
      "step": 645
    },
    {
      "entropy": 7.696876096725464,
      "epoch": 0.0649772579597141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.564,
      "mean_token_accuracy": 0.0964186169207096,
      "num_tokens": 1271062.0,
      "step": 650
    },
    {
      "entropy": 7.794726324081421,
      "epoch": 0.06547708302094267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.5588,
      "mean_token_accuracy": 0.09274660423398018,
      "num_tokens": 1280259.0,
      "step": 655
    },
    {
      "entropy": 7.854710912704467,
      "epoch": 0.06597690808217124,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.5464,
      "mean_token_accuracy": 0.10048911944031716,
      "num_tokens": 1289184.0,
      "step": 660
    },
    {
      "entropy": 7.689560794830323,
      "epoch": 0.06647673314339982,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.5339,
      "mean_token_accuracy": 0.10461926683783532,
      "num_tokens": 1297916.0,
      "step": 665
    },
    {
      "entropy": 7.747795343399048,
      "epoch": 0.06697655820462838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.5534,
      "mean_token_accuracy": 0.10210059732198715,
      "num_tokens": 1307786.0,
      "step": 670
    },
    {
      "entropy": 7.681216049194336,
      "epoch": 0.06747638326585695,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000337,
      "loss": 7.4656,
      "mean_token_accuracy": 0.09771529659628868,
      "num_tokens": 1317528.0,
      "step": 675
    },
    {
      "entropy": 7.736382722854614,
      "epoch": 0.06797620832708552,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003395,
      "loss": 7.5177,
      "mean_token_accuracy": 0.09926827922463417,
      "num_tokens": 1326170.0,
      "step": 680
    },
    {
      "entropy": 7.642406272888183,
      "epoch": 0.0684760333883141,
      "grad_norm": 1.21875,
      "learning_rate": 0.000342,
      "loss": 7.5243,
      "mean_token_accuracy": 0.09604377187788486,
      "num_tokens": 1336469.0,
      "step": 685
    },
    {
      "entropy": 7.716367149353028,
      "epoch": 0.06897585844954265,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.517,
      "mean_token_accuracy": 0.09137366265058518,
      "num_tokens": 1347449.0,
      "step": 690
    },
    {
      "entropy": 7.62940821647644,
      "epoch": 0.06947568351077123,
      "grad_norm": 1.15625,
      "learning_rate": 0.000347,
      "loss": 7.5357,
      "mean_token_accuracy": 0.10426082238554954,
      "num_tokens": 1357849.0,
      "step": 695
    },
    {
      "entropy": 7.692328882217407,
      "epoch": 0.0699755085719998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003495,
      "loss": 7.471,
      "mean_token_accuracy": 0.09552951604127884,
      "num_tokens": 1366784.0,
      "step": 700
    },
    {
      "entropy": 7.6839916706085205,
      "epoch": 0.07047533363322837,
      "grad_norm": 1.09375,
      "learning_rate": 0.000352,
      "loss": 7.4466,
      "mean_token_accuracy": 0.09935205802321434,
      "num_tokens": 1376059.0,
      "step": 705
    },
    {
      "entropy": 7.5484521865844725,
      "epoch": 0.07097515869445695,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003545,
      "loss": 7.4194,
      "mean_token_accuracy": 0.10240289196372032,
      "num_tokens": 1385505.0,
      "step": 710
    },
    {
      "entropy": 7.7426775932312015,
      "epoch": 0.0714749837556855,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000357,
      "loss": 7.4931,
      "mean_token_accuracy": 0.10483329370617867,
      "num_tokens": 1395200.0,
      "step": 715
    },
    {
      "entropy": 7.512200260162354,
      "epoch": 0.07197480881691408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003595,
      "loss": 7.4294,
      "mean_token_accuracy": 0.10536539554595947,
      "num_tokens": 1404599.0,
      "step": 720
    },
    {
      "entropy": 7.642330694198608,
      "epoch": 0.07247463387814265,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000362,
      "loss": 7.5033,
      "mean_token_accuracy": 0.10006041154265403,
      "num_tokens": 1414761.0,
      "step": 725
    },
    {
      "entropy": 7.612363624572754,
      "epoch": 0.07297445893937123,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003645,
      "loss": 7.4319,
      "mean_token_accuracy": 0.10430303364992141,
      "num_tokens": 1424296.0,
      "step": 730
    },
    {
      "entropy": 7.659904146194458,
      "epoch": 0.07347428400059978,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000367,
      "loss": 7.4992,
      "mean_token_accuracy": 0.10049317553639411,
      "num_tokens": 1433729.0,
      "step": 735
    },
    {
      "entropy": 7.663594388961792,
      "epoch": 0.07397410906182836,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003695,
      "loss": 7.57,
      "mean_token_accuracy": 0.09031622260808944,
      "num_tokens": 1444214.0,
      "step": 740
    },
    {
      "entropy": 7.683882474899292,
      "epoch": 0.07447393412305693,
      "grad_norm": 1.03125,
      "learning_rate": 0.000372,
      "loss": 7.5678,
      "mean_token_accuracy": 0.09561188071966172,
      "num_tokens": 1454364.0,
      "step": 745
    },
    {
      "entropy": 7.629066228866577,
      "epoch": 0.0749737591842855,
      "grad_norm": 1.125,
      "learning_rate": 0.0003745,
      "loss": 7.4202,
      "mean_token_accuracy": 0.10247516110539437,
      "num_tokens": 1465018.0,
      "step": 750
    },
    {
      "entropy": 7.573776149749756,
      "epoch": 0.07547358424551406,
      "grad_norm": 1.1875,
      "learning_rate": 0.000377,
      "loss": 7.4268,
      "mean_token_accuracy": 0.10416854470968247,
      "num_tokens": 1474316.0,
      "step": 755
    },
    {
      "entropy": 7.603347063064575,
      "epoch": 0.07597340930674264,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003795,
      "loss": 7.4846,
      "mean_token_accuracy": 0.09634485989809036,
      "num_tokens": 1483867.0,
      "step": 760
    },
    {
      "entropy": 7.588646364212036,
      "epoch": 0.07647323436797121,
      "grad_norm": 1.203125,
      "learning_rate": 0.000382,
      "loss": 7.5488,
      "mean_token_accuracy": 0.09254544489085674,
      "num_tokens": 1493760.0,
      "step": 765
    },
    {
      "entropy": 7.642456197738648,
      "epoch": 0.07697305942919978,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003845,
      "loss": 7.4797,
      "mean_token_accuracy": 0.10259066373109818,
      "num_tokens": 1503984.0,
      "step": 770
    },
    {
      "entropy": 7.631467628479004,
      "epoch": 0.07747288449042836,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.5315,
      "mean_token_accuracy": 0.09757167249917983,
      "num_tokens": 1513550.0,
      "step": 775
    },
    {
      "entropy": 7.658507061004639,
      "epoch": 0.07797270955165692,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.4792,
      "mean_token_accuracy": 0.09825642928481101,
      "num_tokens": 1523543.0,
      "step": 780
    },
    {
      "entropy": 7.600262832641602,
      "epoch": 0.07847253461288549,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.4436,
      "mean_token_accuracy": 0.09491991028189659,
      "num_tokens": 1533775.0,
      "step": 785
    },
    {
      "entropy": 7.519190692901612,
      "epoch": 0.07897235967411406,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.4623,
      "mean_token_accuracy": 0.10000543147325516,
      "num_tokens": 1545008.0,
      "step": 790
    },
    {
      "entropy": 7.6924022197723385,
      "epoch": 0.07947218473534264,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.4861,
      "mean_token_accuracy": 0.09440402016043663,
      "num_tokens": 1554458.0,
      "step": 795
    },
    {
      "entropy": 7.538026142120361,
      "epoch": 0.0799720097965712,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003995,
      "loss": 7.465,
      "mean_token_accuracy": 0.09284278675913811,
      "num_tokens": 1565827.0,
      "step": 800
    },
    {
      "entropy": 7.685927486419677,
      "epoch": 0.08047183485779977,
      "grad_norm": 1.078125,
      "learning_rate": 0.000402,
      "loss": 7.5537,
      "mean_token_accuracy": 0.0930624283850193,
      "num_tokens": 1577477.0,
      "step": 805
    },
    {
      "entropy": 7.504521799087525,
      "epoch": 0.08097165991902834,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004045,
      "loss": 7.3868,
      "mean_token_accuracy": 0.10175873637199402,
      "num_tokens": 1586622.0,
      "step": 810
    },
    {
      "entropy": 7.513061571121216,
      "epoch": 0.08147148498025691,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.4031,
      "mean_token_accuracy": 0.10046703219413758,
      "num_tokens": 1596038.0,
      "step": 815
    },
    {
      "entropy": 7.6118467330932615,
      "epoch": 0.08197131004148547,
      "grad_norm": 1.0,
      "learning_rate": 0.0004095,
      "loss": 7.5118,
      "mean_token_accuracy": 0.09201489984989167,
      "num_tokens": 1606419.0,
      "step": 820
    },
    {
      "entropy": 7.520540618896485,
      "epoch": 0.08247113510271405,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000412,
      "loss": 7.457,
      "mean_token_accuracy": 0.09784280806779862,
      "num_tokens": 1617178.0,
      "step": 825
    },
    {
      "entropy": 7.604987335205078,
      "epoch": 0.08297096016394262,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004145,
      "loss": 7.4982,
      "mean_token_accuracy": 0.09498569741845131,
      "num_tokens": 1628066.0,
      "step": 830
    },
    {
      "entropy": 7.5044020652771,
      "epoch": 0.08347078522517119,
      "grad_norm": 1.25,
      "learning_rate": 0.000417,
      "loss": 7.4454,
      "mean_token_accuracy": 0.09581194594502448,
      "num_tokens": 1637956.0,
      "step": 835
    },
    {
      "entropy": 7.500812101364136,
      "epoch": 0.08397061028639977,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004195,
      "loss": 7.4666,
      "mean_token_accuracy": 0.09536369666457176,
      "num_tokens": 1647388.0,
      "step": 840
    },
    {
      "entropy": 7.6309302806854244,
      "epoch": 0.08447043534762833,
      "grad_norm": 1.140625,
      "learning_rate": 0.000422,
      "loss": 7.4971,
      "mean_token_accuracy": 0.09469880238175392,
      "num_tokens": 1656983.0,
      "step": 845
    },
    {
      "entropy": 7.519527626037598,
      "epoch": 0.0849702604088569,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004245,
      "loss": 7.4669,
      "mean_token_accuracy": 0.09942631423473358,
      "num_tokens": 1666956.0,
      "step": 850
    },
    {
      "entropy": 7.404068660736084,
      "epoch": 0.08547008547008547,
      "grad_norm": 1.234375,
      "learning_rate": 0.000427,
      "loss": 7.3124,
      "mean_token_accuracy": 0.10641250684857369,
      "num_tokens": 1675249.0,
      "step": 855
    },
    {
      "entropy": 7.5542021751403805,
      "epoch": 0.08596991053131404,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004295,
      "loss": 7.4107,
      "mean_token_accuracy": 0.09953751862049103,
      "num_tokens": 1684142.0,
      "step": 860
    },
    {
      "entropy": 7.51811375617981,
      "epoch": 0.0864697355925426,
      "grad_norm": 1.125,
      "learning_rate": 0.000432,
      "loss": 7.4541,
      "mean_token_accuracy": 0.09805193170905113,
      "num_tokens": 1693512.0,
      "step": 865
    },
    {
      "entropy": 7.540082168579102,
      "epoch": 0.08696956065377118,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004345,
      "loss": 7.4372,
      "mean_token_accuracy": 0.09531431272625923,
      "num_tokens": 1703509.0,
      "step": 870
    },
    {
      "entropy": 7.518545818328858,
      "epoch": 0.08746938571499975,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000437,
      "loss": 7.3685,
      "mean_token_accuracy": 0.10407287031412124,
      "num_tokens": 1712158.0,
      "step": 875
    },
    {
      "entropy": 7.444411420822144,
      "epoch": 0.08796921077622832,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004395,
      "loss": 7.4137,
      "mean_token_accuracy": 0.10234727710485458,
      "num_tokens": 1721797.0,
      "step": 880
    },
    {
      "entropy": 7.505835008621216,
      "epoch": 0.0884690358374569,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000442,
      "loss": 7.4375,
      "mean_token_accuracy": 0.09196901768445968,
      "num_tokens": 1732139.0,
      "step": 885
    },
    {
      "entropy": 7.492173051834106,
      "epoch": 0.08896886089868546,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004445,
      "loss": 7.4836,
      "mean_token_accuracy": 0.09574326947331428,
      "num_tokens": 1742419.0,
      "step": 890
    },
    {
      "entropy": 7.495395612716675,
      "epoch": 0.08946868595991403,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447,
      "loss": 7.4579,
      "mean_token_accuracy": 0.09730211272835732,
      "num_tokens": 1752339.0,
      "step": 895
    },
    {
      "entropy": 7.465468120574951,
      "epoch": 0.0899685110211426,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.3246,
      "mean_token_accuracy": 0.10337135419249535,
      "num_tokens": 1763014.0,
      "step": 900
    },
    {
      "entropy": 7.376938724517823,
      "epoch": 0.09046833608237118,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.3169,
      "mean_token_accuracy": 0.10913435965776444,
      "num_tokens": 1772977.0,
      "step": 905
    },
    {
      "entropy": 7.61873426437378,
      "epoch": 0.09096816114359974,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.4175,
      "mean_token_accuracy": 0.10118973776698112,
      "num_tokens": 1782085.0,
      "step": 910
    },
    {
      "entropy": 7.334328937530517,
      "epoch": 0.09146798620482831,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.4619,
      "mean_token_accuracy": 0.10010594651103019,
      "num_tokens": 1791856.0,
      "step": 915
    },
    {
      "entropy": 7.482182407379151,
      "epoch": 0.09196781126605688,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.2896,
      "mean_token_accuracy": 0.10650820061564445,
      "num_tokens": 1800670.0,
      "step": 920
    },
    {
      "entropy": 7.426328849792481,
      "epoch": 0.09246763632728545,
      "grad_norm": 1.140625,
      "learning_rate": 0.000462,
      "loss": 7.4508,
      "mean_token_accuracy": 0.10260506570339203,
      "num_tokens": 1811138.0,
      "step": 925
    },
    {
      "entropy": 7.442462730407715,
      "epoch": 0.09296746138851401,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004645,
      "loss": 7.3307,
      "mean_token_accuracy": 0.09987644553184509,
      "num_tokens": 1819771.0,
      "step": 930
    },
    {
      "entropy": 7.374147653579712,
      "epoch": 0.09346728644974259,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000467,
      "loss": 7.3942,
      "mean_token_accuracy": 0.09991481155157089,
      "num_tokens": 1829075.0,
      "step": 935
    },
    {
      "entropy": 7.41774353981018,
      "epoch": 0.09396711151097116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695,
      "loss": 7.3552,
      "mean_token_accuracy": 0.10564769506454467,
      "num_tokens": 1837892.0,
      "step": 940
    },
    {
      "entropy": 7.433983230590821,
      "epoch": 0.09446693657219973,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000472,
      "loss": 7.4104,
      "mean_token_accuracy": 0.10003899708390236,
      "num_tokens": 1848359.0,
      "step": 945
    },
    {
      "entropy": 7.449016141891479,
      "epoch": 0.0949667616334283,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004745,
      "loss": 7.3765,
      "mean_token_accuracy": 0.09999907165765762,
      "num_tokens": 1857665.0,
      "step": 950
    },
    {
      "entropy": 7.458324146270752,
      "epoch": 0.09546658669465687,
      "grad_norm": 1.125,
      "learning_rate": 0.000477,
      "loss": 7.35,
      "mean_token_accuracy": 0.10186451897025109,
      "num_tokens": 1867313.0,
      "step": 955
    },
    {
      "entropy": 7.404160976409912,
      "epoch": 0.09596641175588544,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004795,
      "loss": 7.4371,
      "mean_token_accuracy": 0.0957965947687626,
      "num_tokens": 1877904.0,
      "step": 960
    },
    {
      "entropy": 7.462620973587036,
      "epoch": 0.09646623681711401,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000482,
      "loss": 7.3508,
      "mean_token_accuracy": 0.10784625113010407,
      "num_tokens": 1887173.0,
      "step": 965
    },
    {
      "entropy": 7.463750505447388,
      "epoch": 0.09696606187834259,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004845,
      "loss": 7.4364,
      "mean_token_accuracy": 0.09779566004872323,
      "num_tokens": 1896326.0,
      "step": 970
    },
    {
      "entropy": 7.386555910110474,
      "epoch": 0.09746588693957114,
      "grad_norm": 1.1875,
      "learning_rate": 0.000487,
      "loss": 7.3174,
      "mean_token_accuracy": 0.09860834032297135,
      "num_tokens": 1905403.0,
      "step": 975
    },
    {
      "entropy": 7.493797731399536,
      "epoch": 0.09796571200079972,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004895,
      "loss": 7.4104,
      "mean_token_accuracy": 0.09834235757589341,
      "num_tokens": 1915910.0,
      "step": 980
    },
    {
      "entropy": 7.351792240142823,
      "epoch": 0.09846553706202829,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 7.3515,
      "mean_token_accuracy": 0.10310368835926056,
      "num_tokens": 1926538.0,
      "step": 985
    },
    {
      "entropy": 7.331353616714478,
      "epoch": 0.09896536212325686,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004945,
      "loss": 7.4155,
      "mean_token_accuracy": 0.09870894104242325,
      "num_tokens": 1936604.0,
      "step": 990
    },
    {
      "entropy": 7.351730632781982,
      "epoch": 0.09946518718448542,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000497,
      "loss": 7.2687,
      "mean_token_accuracy": 0.10868923962116242,
      "num_tokens": 1946509.0,
      "step": 995
    },
    {
      "entropy": 7.449704694747925,
      "epoch": 0.099965012245714,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995,
      "loss": 7.4381,
      "mean_token_accuracy": 0.09592320919036865,
      "num_tokens": 1955562.0,
      "step": 1000
    },
    {
      "entropy": 7.3522120952606205,
      "epoch": 0.10046483730694257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999981884985,
      "loss": 7.2931,
      "mean_token_accuracy": 0.10394947901368141,
      "num_tokens": 1964805.0,
      "step": 1005
    },
    {
      "entropy": 7.453734540939331,
      "epoch": 0.10096466236817114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999999908292739,
      "loss": 7.3876,
      "mean_token_accuracy": 0.10147590786218644,
      "num_tokens": 1973735.0,
      "step": 1010
    },
    {
      "entropy": 7.296640205383301,
      "epoch": 0.10146448742939972,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999999778091072,
      "loss": 7.2858,
      "mean_token_accuracy": 0.09892769753932953,
      "num_tokens": 1983657.0,
      "step": 1015
    },
    {
      "entropy": 7.4748045921325685,
      "epoch": 0.10196431249062828,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999999591279991,
      "loss": 7.3675,
      "mean_token_accuracy": 0.10389427542686462,
      "num_tokens": 1994767.0,
      "step": 1020
    },
    {
      "entropy": 7.266379356384277,
      "epoch": 0.10246413755185685,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049999993478595,
      "loss": 7.3899,
      "mean_token_accuracy": 0.10590666383504868,
      "num_tokens": 2004634.0,
      "step": 1025
    },
    {
      "entropy": 7.3518266677856445,
      "epoch": 0.10296396261308542,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999999047829604,
      "loss": 7.2877,
      "mean_token_accuracy": 0.10322983711957931,
      "num_tokens": 2014346.0,
      "step": 1030
    },
    {
      "entropy": 7.398997354507446,
      "epoch": 0.103463787674314,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499999869119031,
      "loss": 7.3821,
      "mean_token_accuracy": 0.10241326168179513,
      "num_tokens": 2024132.0,
      "step": 1035
    },
    {
      "entropy": 7.32396936416626,
      "epoch": 0.10396361273554255,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999998277941629,
      "loss": 7.3135,
      "mean_token_accuracy": 0.09869552403688431,
      "num_tokens": 2034230.0,
      "step": 1040
    },
    {
      "entropy": 7.388594722747802,
      "epoch": 0.10446343779677113,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999997808083569,
      "loss": 7.3698,
      "mean_token_accuracy": 0.10165157392621041,
      "num_tokens": 2043936.0,
      "step": 1045
    },
    {
      "entropy": 7.276969957351684,
      "epoch": 0.1049632628579997,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004999997281616144,
      "loss": 7.3319,
      "mean_token_accuracy": 0.09708460494875908,
      "num_tokens": 2053386.0,
      "step": 1050
    },
    {
      "entropy": 7.300722360610962,
      "epoch": 0.10546308791922827,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999996698539366,
      "loss": 7.2492,
      "mean_token_accuracy": 0.1117588073015213,
      "num_tokens": 2063604.0,
      "step": 1055
    },
    {
      "entropy": 7.340676641464233,
      "epoch": 0.10596291298045685,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499999605885325,
      "loss": 7.3128,
      "mean_token_accuracy": 0.10286297127604485,
      "num_tokens": 2072795.0,
      "step": 1060
    },
    {
      "entropy": 7.354478883743286,
      "epoch": 0.1064627380416854,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004999995362557813,
      "loss": 7.2888,
      "mean_token_accuracy": 0.1015630230307579,
      "num_tokens": 2082113.0,
      "step": 1065
    },
    {
      "entropy": 7.253059196472168,
      "epoch": 0.10696256310291398,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000499999460965307,
      "loss": 7.1843,
      "mean_token_accuracy": 0.11151959002017975,
      "num_tokens": 2090484.0,
      "step": 1070
    },
    {
      "entropy": 7.203897428512573,
      "epoch": 0.10746238816414255,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999993800139043,
      "loss": 7.2397,
      "mean_token_accuracy": 0.10655209198594093,
      "num_tokens": 2100703.0,
      "step": 1075
    },
    {
      "entropy": 7.286228370666504,
      "epoch": 0.10796221322537113,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999992934015752,
      "loss": 7.3485,
      "mean_token_accuracy": 0.09915086925029755,
      "num_tokens": 2110861.0,
      "step": 1080
    },
    {
      "entropy": 7.353559732437134,
      "epoch": 0.10846203828659969,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999992011283215,
      "loss": 7.2853,
      "mean_token_accuracy": 0.10259805619716644,
      "num_tokens": 2121414.0,
      "step": 1085
    },
    {
      "entropy": 7.3230507373809814,
      "epoch": 0.10896186334782826,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999991031941459,
      "loss": 7.2654,
      "mean_token_accuracy": 0.10948954746127129,
      "num_tokens": 2131201.0,
      "step": 1090
    },
    {
      "entropy": 7.256815958023071,
      "epoch": 0.10946168840905683,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999989995990509,
      "loss": 7.2584,
      "mean_token_accuracy": 0.10254568606615067,
      "num_tokens": 2142071.0,
      "step": 1095
    },
    {
      "entropy": 7.249517440795898,
      "epoch": 0.1099615134702854,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999988903430389,
      "loss": 7.3155,
      "mean_token_accuracy": 0.10242817401885987,
      "num_tokens": 2151514.0,
      "step": 1100
    },
    {
      "entropy": 7.30422077178955,
      "epoch": 0.11046133853151396,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999987754261126,
      "loss": 7.2977,
      "mean_token_accuracy": 0.10412856489419937,
      "num_tokens": 2160828.0,
      "step": 1105
    },
    {
      "entropy": 7.329158115386963,
      "epoch": 0.11096116359274254,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999986548482751,
      "loss": 7.3347,
      "mean_token_accuracy": 0.10781540647149086,
      "num_tokens": 2169968.0,
      "step": 1110
    },
    {
      "entropy": 7.330237340927124,
      "epoch": 0.11146098865397111,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999985286095293,
      "loss": 7.2593,
      "mean_token_accuracy": 0.10637600645422936,
      "num_tokens": 2180434.0,
      "step": 1115
    },
    {
      "entropy": 7.239208841323853,
      "epoch": 0.11196081371519968,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999983967098785,
      "loss": 7.1956,
      "mean_token_accuracy": 0.10522906109690666,
      "num_tokens": 2190024.0,
      "step": 1120
    },
    {
      "entropy": 7.288405704498291,
      "epoch": 0.11246063877642826,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999982591493258,
      "loss": 7.2271,
      "mean_token_accuracy": 0.1099842593073845,
      "num_tokens": 2199897.0,
      "step": 1125
    },
    {
      "entropy": 7.168730354309082,
      "epoch": 0.11296046383765682,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999981159278749,
      "loss": 7.2887,
      "mean_token_accuracy": 0.10584141686558723,
      "num_tokens": 2209865.0,
      "step": 1130
    },
    {
      "entropy": 7.373020601272583,
      "epoch": 0.11346028889888539,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004999979670455292,
      "loss": 7.2426,
      "mean_token_accuracy": 0.1077901616692543,
      "num_tokens": 2219960.0,
      "step": 1135
    },
    {
      "entropy": 7.236533546447754,
      "epoch": 0.11396011396011396,
      "grad_norm": 0.91796875,
      "learning_rate": 0.0004999978125022926,
      "loss": 7.2442,
      "mean_token_accuracy": 0.10784188359975815,
      "num_tokens": 2230699.0,
      "step": 1140
    },
    {
      "entropy": 7.2002357006072994,
      "epoch": 0.11445993902134254,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499997652298169,
      "loss": 7.2507,
      "mean_token_accuracy": 0.10368520766496658,
      "num_tokens": 2239694.0,
      "step": 1145
    },
    {
      "entropy": 7.327735424041748,
      "epoch": 0.1149597640825711,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999974864331624,
      "loss": 7.2772,
      "mean_token_accuracy": 0.10986744239926338,
      "num_tokens": 2250624.0,
      "step": 1150
    },
    {
      "entropy": 7.30306601524353,
      "epoch": 0.11545958914379967,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999973149072768,
      "loss": 7.2842,
      "mean_token_accuracy": 0.10279571041464805,
      "num_tokens": 2260759.0,
      "step": 1155
    },
    {
      "entropy": 7.238110876083374,
      "epoch": 0.11595941420502824,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999971377205167,
      "loss": 7.3117,
      "mean_token_accuracy": 0.10007711425423622,
      "num_tokens": 2271151.0,
      "step": 1160
    },
    {
      "entropy": 7.237846374511719,
      "epoch": 0.11645923926625681,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999969548728864,
      "loss": 7.1965,
      "mean_token_accuracy": 0.11174095422029495,
      "num_tokens": 2280014.0,
      "step": 1165
    },
    {
      "entropy": 7.2414806365966795,
      "epoch": 0.11695906432748537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999967663643908,
      "loss": 7.1659,
      "mean_token_accuracy": 0.11187820881605148,
      "num_tokens": 2288550.0,
      "step": 1170
    },
    {
      "entropy": 7.226145553588867,
      "epoch": 0.11745888938871395,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999965721950345,
      "loss": 7.2567,
      "mean_token_accuracy": 0.10522151663899422,
      "num_tokens": 2297899.0,
      "step": 1175
    },
    {
      "entropy": 7.251861715316773,
      "epoch": 0.11795871444994252,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999963723648222,
      "loss": 7.2035,
      "mean_token_accuracy": 0.1062523826956749,
      "num_tokens": 2308021.0,
      "step": 1180
    },
    {
      "entropy": 7.288262224197387,
      "epoch": 0.1184585395111711,
      "grad_norm": 1.8203125,
      "learning_rate": 0.000499996166873759,
      "loss": 7.2486,
      "mean_token_accuracy": 0.10318779572844505,
      "num_tokens": 2319607.0,
      "step": 1185
    },
    {
      "entropy": 7.178311777114868,
      "epoch": 0.11895836457239967,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999959557218502,
      "loss": 7.2494,
      "mean_token_accuracy": 0.10606426224112511,
      "num_tokens": 2329293.0,
      "step": 1190
    },
    {
      "entropy": 7.335767650604248,
      "epoch": 0.11945818963362823,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999957389091012,
      "loss": 7.2712,
      "mean_token_accuracy": 0.10525668188929557,
      "num_tokens": 2339812.0,
      "step": 1195
    },
    {
      "entropy": 7.135515928268433,
      "epoch": 0.1199580146948568,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999955164355171,
      "loss": 7.2314,
      "mean_token_accuracy": 0.10235729217529296,
      "num_tokens": 2349770.0,
      "step": 1200
    },
    {
      "entropy": 7.282419157028198,
      "epoch": 0.12045783975608537,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004999952883011038,
      "loss": 7.1282,
      "mean_token_accuracy": 0.11181103363633156,
      "num_tokens": 2360355.0,
      "step": 1205
    },
    {
      "entropy": 7.135486125946045,
      "epoch": 0.12095766481731395,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499995054505867,
      "loss": 7.3192,
      "mean_token_accuracy": 0.10360833257436752,
      "num_tokens": 2369444.0,
      "step": 1210
    },
    {
      "entropy": 7.23374752998352,
      "epoch": 0.1214574898785425,
      "grad_norm": 0.87890625,
      "learning_rate": 0.0004999948150498124,
      "loss": 7.1863,
      "mean_token_accuracy": 0.10711781978607178,
      "num_tokens": 2380172.0,
      "step": 1215
    },
    {
      "entropy": 7.309315729141235,
      "epoch": 0.12195731493977108,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999945699329462,
      "loss": 7.3493,
      "mean_token_accuracy": 0.1012509100139141,
      "num_tokens": 2391520.0,
      "step": 1220
    },
    {
      "entropy": 7.295082521438599,
      "epoch": 0.12245714000099965,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999943191552745,
      "loss": 7.2548,
      "mean_token_accuracy": 0.1027175396680832,
      "num_tokens": 2400977.0,
      "step": 1225
    },
    {
      "entropy": 7.329327964782715,
      "epoch": 0.12295696506222822,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999940627168037,
      "loss": 7.2506,
      "mean_token_accuracy": 0.10020245537161827,
      "num_tokens": 2410220.0,
      "step": 1230
    },
    {
      "entropy": 7.180943441390991,
      "epoch": 0.12345679012345678,
      "grad_norm": 0.90234375,
      "learning_rate": 0.00049999380061754,
      "loss": 7.2642,
      "mean_token_accuracy": 0.10909186080098152,
      "num_tokens": 2420000.0,
      "step": 1235
    },
    {
      "entropy": 7.274267911911011,
      "epoch": 0.12395661518468536,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999935328574903,
      "loss": 7.2054,
      "mean_token_accuracy": 0.10308079868555069,
      "num_tokens": 2429236.0,
      "step": 1240
    },
    {
      "entropy": 7.123314380645752,
      "epoch": 0.12445644024591393,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999932594366613,
      "loss": 7.1065,
      "mean_token_accuracy": 0.1079021766781807,
      "num_tokens": 2438830.0,
      "step": 1245
    },
    {
      "entropy": 7.20734977722168,
      "epoch": 0.1249562653071425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999929803550597,
      "loss": 7.1211,
      "mean_token_accuracy": 0.11151803135871888,
      "num_tokens": 2448218.0,
      "step": 1250
    },
    {
      "entropy": 7.212431287765503,
      "epoch": 0.12545609036837108,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999926956126927,
      "loss": 7.1895,
      "mean_token_accuracy": 0.1051447942852974,
      "num_tokens": 2458016.0,
      "step": 1255
    },
    {
      "entropy": 7.124825382232666,
      "epoch": 0.12595591542959964,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999924052095672,
      "loss": 7.095,
      "mean_token_accuracy": 0.10699434652924537,
      "num_tokens": 2467973.0,
      "step": 1260
    },
    {
      "entropy": 7.153971910476685,
      "epoch": 0.12645574049082822,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999921091456907,
      "loss": 7.2014,
      "mean_token_accuracy": 0.10560873597860336,
      "num_tokens": 2477811.0,
      "step": 1265
    },
    {
      "entropy": 7.186617374420166,
      "epoch": 0.12695556555205678,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999918074210707,
      "loss": 7.1761,
      "mean_token_accuracy": 0.10755071565508842,
      "num_tokens": 2487477.0,
      "step": 1270
    },
    {
      "entropy": 7.109448385238648,
      "epoch": 0.12745539061328534,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999915000357147,
      "loss": 7.0675,
      "mean_token_accuracy": 0.11184290945529937,
      "num_tokens": 2497127.0,
      "step": 1275
    },
    {
      "entropy": 7.090885353088379,
      "epoch": 0.12795521567451393,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999911869896305,
      "loss": 7.1605,
      "mean_token_accuracy": 0.10718183293938637,
      "num_tokens": 2506340.0,
      "step": 1280
    },
    {
      "entropy": 7.297980165481567,
      "epoch": 0.1284550407357425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999908682828258,
      "loss": 7.1951,
      "mean_token_accuracy": 0.10084473490715026,
      "num_tokens": 2515210.0,
      "step": 1285
    },
    {
      "entropy": 7.0772905349731445,
      "epoch": 0.12895486579697105,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999905439153089,
      "loss": 7.0952,
      "mean_token_accuracy": 0.1110685035586357,
      "num_tokens": 2524653.0,
      "step": 1290
    },
    {
      "entropy": 7.150044536590576,
      "epoch": 0.12945469085819963,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999902138870878,
      "loss": 7.1397,
      "mean_token_accuracy": 0.10771988704800606,
      "num_tokens": 2534243.0,
      "step": 1295
    },
    {
      "entropy": 7.236660671234131,
      "epoch": 0.1299545159194282,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999898781981707,
      "loss": 7.1998,
      "mean_token_accuracy": 0.10407418459653854,
      "num_tokens": 2543758.0,
      "step": 1300
    },
    {
      "entropy": 7.205931663513184,
      "epoch": 0.13045434098065678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999895368485662,
      "loss": 7.2281,
      "mean_token_accuracy": 0.1009873628616333,
      "num_tokens": 2553537.0,
      "step": 1305
    },
    {
      "entropy": 7.116842460632324,
      "epoch": 0.13095416604188534,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499989189838283,
      "loss": 7.145,
      "mean_token_accuracy": 0.10926665514707565,
      "num_tokens": 2562812.0,
      "step": 1310
    },
    {
      "entropy": 7.184953212738037,
      "epoch": 0.1314539911031139,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999888371673295,
      "loss": 7.1309,
      "mean_token_accuracy": 0.11213063448667526,
      "num_tokens": 2571728.0,
      "step": 1315
    },
    {
      "entropy": 7.177727079391479,
      "epoch": 0.13195381616434249,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999884788357147,
      "loss": 7.2195,
      "mean_token_accuracy": 0.10633473843336105,
      "num_tokens": 2582309.0,
      "step": 1320
    },
    {
      "entropy": 7.1854188442230225,
      "epoch": 0.13245364122557104,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999881148434478,
      "loss": 7.128,
      "mean_token_accuracy": 0.11164407134056091,
      "num_tokens": 2591810.0,
      "step": 1325
    },
    {
      "entropy": 7.0932341575622555,
      "epoch": 0.13295346628679963,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999877451905378,
      "loss": 7.1083,
      "mean_token_accuracy": 0.11393467858433723,
      "num_tokens": 2602028.0,
      "step": 1330
    },
    {
      "entropy": 7.095207071304321,
      "epoch": 0.1334532913480282,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999873698769941,
      "loss": 7.108,
      "mean_token_accuracy": 0.11159966364502907,
      "num_tokens": 2610900.0,
      "step": 1335
    },
    {
      "entropy": 7.107607173919678,
      "epoch": 0.13395311640925675,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499986988902826,
      "loss": 7.1132,
      "mean_token_accuracy": 0.11127648279070854,
      "num_tokens": 2620918.0,
      "step": 1340
    },
    {
      "entropy": 7.139119338989258,
      "epoch": 0.13445294147048534,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999866022680431,
      "loss": 7.1284,
      "mean_token_accuracy": 0.11192728728055953,
      "num_tokens": 2631383.0,
      "step": 1345
    },
    {
      "entropy": 7.102079010009765,
      "epoch": 0.1349527665317139,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999862099726552,
      "loss": 7.0999,
      "mean_token_accuracy": 0.11033402979373932,
      "num_tokens": 2640824.0,
      "step": 1350
    },
    {
      "entropy": 7.178639459609985,
      "epoch": 0.13545259159294248,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999858120166721,
      "loss": 7.0784,
      "mean_token_accuracy": 0.1097441516816616,
      "num_tokens": 2649884.0,
      "step": 1355
    },
    {
      "entropy": 7.068162250518799,
      "epoch": 0.13595241665417104,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999854084001039,
      "loss": 7.1528,
      "mean_token_accuracy": 0.11337249279022217,
      "num_tokens": 2659464.0,
      "step": 1360
    },
    {
      "entropy": 7.113283252716064,
      "epoch": 0.1364522417153996,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999849991229608,
      "loss": 7.1487,
      "mean_token_accuracy": 0.10855353698134422,
      "num_tokens": 2668271.0,
      "step": 1365
    },
    {
      "entropy": 7.187230062484741,
      "epoch": 0.1369520667766282,
      "grad_norm": 1.0625,
      "learning_rate": 0.000499984584185253,
      "loss": 7.2098,
      "mean_token_accuracy": 0.10614682957530022,
      "num_tokens": 2678353.0,
      "step": 1370
    },
    {
      "entropy": 7.149409389495849,
      "epoch": 0.13745189183785675,
      "grad_norm": 0.9296875,
      "learning_rate": 0.0004999841635869909,
      "loss": 7.1448,
      "mean_token_accuracy": 0.10726541206240654,
      "num_tokens": 2688082.0,
      "step": 1375
    },
    {
      "entropy": 7.158280563354492,
      "epoch": 0.1379517168990853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999837373281852,
      "loss": 7.1357,
      "mean_token_accuracy": 0.1082352265715599,
      "num_tokens": 2697780.0,
      "step": 1380
    },
    {
      "entropy": 7.153148174285889,
      "epoch": 0.1384515419603139,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999833054088465,
      "loss": 7.1464,
      "mean_token_accuracy": 0.11163995936512947,
      "num_tokens": 2707851.0,
      "step": 1385
    },
    {
      "entropy": 7.146587085723877,
      "epoch": 0.13895136702154245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999828678289858,
      "loss": 7.2008,
      "mean_token_accuracy": 0.10463528037071228,
      "num_tokens": 2718041.0,
      "step": 1390
    },
    {
      "entropy": 7.154413461685181,
      "epoch": 0.13945119208277104,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499982424588614,
      "loss": 7.0933,
      "mean_token_accuracy": 0.1157912366092205,
      "num_tokens": 2726663.0,
      "step": 1395
    },
    {
      "entropy": 7.1343855381011965,
      "epoch": 0.1399510171439996,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999819756877422,
      "loss": 7.0596,
      "mean_token_accuracy": 0.11434894353151322,
      "num_tokens": 2734926.0,
      "step": 1400
    },
    {
      "entropy": 7.131910037994385,
      "epoch": 0.14045084220522816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999815211263819,
      "loss": 7.1364,
      "mean_token_accuracy": 0.10998270213603974,
      "num_tokens": 2744984.0,
      "step": 1405
    },
    {
      "entropy": 7.1278736114501955,
      "epoch": 0.14095066726645675,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999810609045444,
      "loss": 7.0561,
      "mean_token_accuracy": 0.11019435450434685,
      "num_tokens": 2754293.0,
      "step": 1410
    },
    {
      "entropy": 7.038761758804322,
      "epoch": 0.1414504923276853,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999805950222412,
      "loss": 7.1112,
      "mean_token_accuracy": 0.11249493733048439,
      "num_tokens": 2764602.0,
      "step": 1415
    },
    {
      "entropy": 7.136958646774292,
      "epoch": 0.1419503173889139,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999801234794843,
      "loss": 7.1579,
      "mean_token_accuracy": 0.10929414480924607,
      "num_tokens": 2775171.0,
      "step": 1420
    },
    {
      "entropy": 7.107983350753784,
      "epoch": 0.14245014245014245,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999796462762853,
      "loss": 7.052,
      "mean_token_accuracy": 0.1148157998919487,
      "num_tokens": 2784739.0,
      "step": 1425
    },
    {
      "entropy": 7.160243511199951,
      "epoch": 0.142949967511371,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004999791634126562,
      "loss": 7.1657,
      "mean_token_accuracy": 0.11298665180802345,
      "num_tokens": 2794549.0,
      "step": 1430
    },
    {
      "entropy": 7.166717195510865,
      "epoch": 0.1434497925725996,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999786748886094,
      "loss": 7.1116,
      "mean_token_accuracy": 0.11435734257102012,
      "num_tokens": 2803235.0,
      "step": 1435
    },
    {
      "entropy": 7.079634094238282,
      "epoch": 0.14394961763382816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999781807041569,
      "loss": 7.0513,
      "mean_token_accuracy": 0.10983316227793694,
      "num_tokens": 2812367.0,
      "step": 1440
    },
    {
      "entropy": 7.01149172782898,
      "epoch": 0.14444944269505672,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999776808593113,
      "loss": 7.0788,
      "mean_token_accuracy": 0.115450157225132,
      "num_tokens": 2822173.0,
      "step": 1445
    },
    {
      "entropy": 7.19095778465271,
      "epoch": 0.1449492677562853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999771753540852,
      "loss": 7.1296,
      "mean_token_accuracy": 0.10804187655448913,
      "num_tokens": 2831498.0,
      "step": 1450
    },
    {
      "entropy": 7.151987075805664,
      "epoch": 0.14544909281751386,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999766641884912,
      "loss": 7.2708,
      "mean_token_accuracy": 0.10761458948254585,
      "num_tokens": 2840883.0,
      "step": 1455
    },
    {
      "entropy": 7.108257818222046,
      "epoch": 0.14594891787874245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999761473625422,
      "loss": 7.0424,
      "mean_token_accuracy": 0.11248427405953407,
      "num_tokens": 2850694.0,
      "step": 1460
    },
    {
      "entropy": 6.983803033828735,
      "epoch": 0.146448742939971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999756248762513,
      "loss": 7.0643,
      "mean_token_accuracy": 0.11776667833328247,
      "num_tokens": 2861100.0,
      "step": 1465
    },
    {
      "entropy": 7.087214612960816,
      "epoch": 0.14694856800119957,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999750967296315,
      "loss": 7.0179,
      "mean_token_accuracy": 0.11294868513941765,
      "num_tokens": 2871054.0,
      "step": 1470
    },
    {
      "entropy": 7.02065601348877,
      "epoch": 0.14744839306242816,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999745629226962,
      "loss": 6.9776,
      "mean_token_accuracy": 0.11572617068886756,
      "num_tokens": 2880244.0,
      "step": 1475
    },
    {
      "entropy": 7.022220134735107,
      "epoch": 0.14794821812365672,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999740234554589,
      "loss": 7.0832,
      "mean_token_accuracy": 0.10742958262562752,
      "num_tokens": 2891441.0,
      "step": 1480
    },
    {
      "entropy": 7.101360940933228,
      "epoch": 0.1484480431848853,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999734783279327,
      "loss": 7.0758,
      "mean_token_accuracy": 0.10424959361553192,
      "num_tokens": 2901784.0,
      "step": 1485
    },
    {
      "entropy": 7.106187534332276,
      "epoch": 0.14894786824611386,
      "grad_norm": 1.125,
      "learning_rate": 0.000499972927540132,
      "loss": 7.0934,
      "mean_token_accuracy": 0.11211119145154953,
      "num_tokens": 2912294.0,
      "step": 1490
    },
    {
      "entropy": 7.037475490570069,
      "epoch": 0.14944769330734242,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999723710920703,
      "loss": 7.0035,
      "mean_token_accuracy": 0.10910386443138123,
      "num_tokens": 2921558.0,
      "step": 1495
    },
    {
      "entropy": 7.123606777191162,
      "epoch": 0.149947518368571,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999718089837616,
      "loss": 7.0266,
      "mean_token_accuracy": 0.10954056233167649,
      "num_tokens": 2930997.0,
      "step": 1500
    },
    {
      "entropy": 7.054976892471314,
      "epoch": 0.15044734342979957,
      "grad_norm": 1.265625,
      "learning_rate": 0.00049997124121522,
      "loss": 7.0144,
      "mean_token_accuracy": 0.11238231882452965,
      "num_tokens": 2939199.0,
      "step": 1505
    },
    {
      "entropy": 7.039057970046997,
      "epoch": 0.15094716849102813,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999706677864599,
      "loss": 7.0878,
      "mean_token_accuracy": 0.11365630775690079,
      "num_tokens": 2949466.0,
      "step": 1510
    },
    {
      "entropy": 7.079186916351318,
      "epoch": 0.15144699355225671,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999700886974958,
      "loss": 7.0284,
      "mean_token_accuracy": 0.11694424152374268,
      "num_tokens": 2959653.0,
      "step": 1515
    },
    {
      "entropy": 7.05449070930481,
      "epoch": 0.15194681861348527,
      "grad_norm": 0.8984375,
      "learning_rate": 0.000499969503948342,
      "loss": 7.0295,
      "mean_token_accuracy": 0.11299590468406677,
      "num_tokens": 2970331.0,
      "step": 1520
    },
    {
      "entropy": 6.995374393463135,
      "epoch": 0.15244664367471386,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999689135390134,
      "loss": 6.9897,
      "mean_token_accuracy": 0.11340121328830718,
      "num_tokens": 2980375.0,
      "step": 1525
    },
    {
      "entropy": 6.982559442520142,
      "epoch": 0.15294646873594242,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999683174695249,
      "loss": 7.0662,
      "mean_token_accuracy": 0.10985481515526771,
      "num_tokens": 2990562.0,
      "step": 1530
    },
    {
      "entropy": 7.104888010025024,
      "epoch": 0.15344629379717098,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999677157398914,
      "loss": 6.9735,
      "mean_token_accuracy": 0.12070909142494202,
      "num_tokens": 3000797.0,
      "step": 1535
    },
    {
      "entropy": 6.979942655563354,
      "epoch": 0.15394611885839957,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999671083501281,
      "loss": 7.0347,
      "mean_token_accuracy": 0.11492840200662613,
      "num_tokens": 3010520.0,
      "step": 1540
    },
    {
      "entropy": 7.0611762523651125,
      "epoch": 0.15444594391962813,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999664953002502,
      "loss": 6.9505,
      "mean_token_accuracy": 0.11264497265219689,
      "num_tokens": 3020249.0,
      "step": 1545
    },
    {
      "entropy": 7.055840873718262,
      "epoch": 0.1549457689808567,
      "grad_norm": 1.0,
      "learning_rate": 0.000499965876590273,
      "loss": 7.1282,
      "mean_token_accuracy": 0.11212135925889015,
      "num_tokens": 3031721.0,
      "step": 1550
    },
    {
      "entropy": 7.055178308486939,
      "epoch": 0.15544559404208527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999652522202125,
      "loss": 7.0479,
      "mean_token_accuracy": 0.11103066131472587,
      "num_tokens": 3041236.0,
      "step": 1555
    },
    {
      "entropy": 7.10380334854126,
      "epoch": 0.15594541910331383,
      "grad_norm": 0.87109375,
      "learning_rate": 0.0004999646221900839,
      "loss": 7.0251,
      "mean_token_accuracy": 0.11302450522780419,
      "num_tokens": 3051749.0,
      "step": 1560
    },
    {
      "entropy": 7.081645584106445,
      "epoch": 0.15644524416454242,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999639864999034,
      "loss": 7.1237,
      "mean_token_accuracy": 0.10856601297855377,
      "num_tokens": 3061928.0,
      "step": 1565
    },
    {
      "entropy": 6.9718320846557615,
      "epoch": 0.15694506922577098,
      "grad_norm": 1.0,
      "learning_rate": 0.000499963345149687,
      "loss": 6.9493,
      "mean_token_accuracy": 0.11934173181653022,
      "num_tokens": 3071598.0,
      "step": 1570
    },
    {
      "entropy": 6.99726939201355,
      "epoch": 0.15744489428699954,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999626981394506,
      "loss": 6.9957,
      "mean_token_accuracy": 0.11047958061099053,
      "num_tokens": 3081952.0,
      "step": 1575
    },
    {
      "entropy": 7.108104562759399,
      "epoch": 0.15794471934822812,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999620454692106,
      "loss": 7.0704,
      "mean_token_accuracy": 0.11186990886926651,
      "num_tokens": 3092047.0,
      "step": 1580
    },
    {
      "entropy": 7.076290416717529,
      "epoch": 0.15844454440945668,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999613871389836,
      "loss": 7.0374,
      "mean_token_accuracy": 0.10968389436602592,
      "num_tokens": 3101454.0,
      "step": 1585
    },
    {
      "entropy": 6.990331602096558,
      "epoch": 0.15894436947068527,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999607231487858,
      "loss": 6.997,
      "mean_token_accuracy": 0.11208927035331726,
      "num_tokens": 3111447.0,
      "step": 1590
    },
    {
      "entropy": 6.939959478378296,
      "epoch": 0.15944419453191383,
      "grad_norm": 1.0,
      "learning_rate": 0.000499960053498634,
      "loss": 6.9089,
      "mean_token_accuracy": 0.11952870041131973,
      "num_tokens": 3121085.0,
      "step": 1595
    },
    {
      "entropy": 6.9829981327056885,
      "epoch": 0.1599440195931424,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999593781885454,
      "loss": 6.9608,
      "mean_token_accuracy": 0.11509754583239555,
      "num_tokens": 3131233.0,
      "step": 1600
    },
    {
      "entropy": 7.03386549949646,
      "epoch": 0.16044384465437098,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999586972185366,
      "loss": 7.0519,
      "mean_token_accuracy": 0.11012849733233451,
      "num_tokens": 3140541.0,
      "step": 1605
    },
    {
      "entropy": 7.025159740447998,
      "epoch": 0.16094366971559954,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999580105886248,
      "loss": 7.0377,
      "mean_token_accuracy": 0.10797161757946014,
      "num_tokens": 3150166.0,
      "step": 1610
    },
    {
      "entropy": 7.028019046783447,
      "epoch": 0.16144349477682812,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999573182988275,
      "loss": 7.0528,
      "mean_token_accuracy": 0.11481094807386398,
      "num_tokens": 3160259.0,
      "step": 1615
    },
    {
      "entropy": 7.063313341140747,
      "epoch": 0.16194331983805668,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999566203491619,
      "loss": 6.9958,
      "mean_token_accuracy": 0.11163723319768906,
      "num_tokens": 3168801.0,
      "step": 1620
    },
    {
      "entropy": 7.082771110534668,
      "epoch": 0.16244314489928524,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999559167396455,
      "loss": 7.1206,
      "mean_token_accuracy": 0.10632694289088249,
      "num_tokens": 3181029.0,
      "step": 1625
    },
    {
      "entropy": 6.970886611938477,
      "epoch": 0.16294296996051383,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999552074702963,
      "loss": 6.9377,
      "mean_token_accuracy": 0.12233891040086746,
      "num_tokens": 3190963.0,
      "step": 1630
    },
    {
      "entropy": 6.8871989250183105,
      "epoch": 0.1634427950217424,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999544925411318,
      "loss": 6.9319,
      "mean_token_accuracy": 0.11681057810783387,
      "num_tokens": 3200714.0,
      "step": 1635
    },
    {
      "entropy": 7.0543567657470705,
      "epoch": 0.16394262008297095,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999537719521703,
      "loss": 6.9678,
      "mean_token_accuracy": 0.11916712298989296,
      "num_tokens": 3209965.0,
      "step": 1640
    },
    {
      "entropy": 6.976824760437012,
      "epoch": 0.16444244514419953,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999530457034297,
      "loss": 7.0004,
      "mean_token_accuracy": 0.10843002274632454,
      "num_tokens": 3218849.0,
      "step": 1645
    },
    {
      "entropy": 7.04572696685791,
      "epoch": 0.1649422702054281,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999523137949283,
      "loss": 7.0266,
      "mean_token_accuracy": 0.11135978549718857,
      "num_tokens": 3229483.0,
      "step": 1650
    },
    {
      "entropy": 7.008546638488769,
      "epoch": 0.16544209526665668,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999515762266846,
      "loss": 7.0348,
      "mean_token_accuracy": 0.11097366139292716,
      "num_tokens": 3239818.0,
      "step": 1655
    },
    {
      "entropy": 7.0645958423614506,
      "epoch": 0.16594192032788524,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999508329987173,
      "loss": 7.0354,
      "mean_token_accuracy": 0.1088833674788475,
      "num_tokens": 3249136.0,
      "step": 1660
    },
    {
      "entropy": 7.004113721847534,
      "epoch": 0.1664417453891138,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999500841110447,
      "loss": 6.9393,
      "mean_token_accuracy": 0.11334899812936783,
      "num_tokens": 3258624.0,
      "step": 1665
    },
    {
      "entropy": 7.035124921798706,
      "epoch": 0.16694157045034239,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000499949329563686,
      "loss": 7.0458,
      "mean_token_accuracy": 0.1125538133084774,
      "num_tokens": 3269079.0,
      "step": 1670
    },
    {
      "entropy": 6.938257694244385,
      "epoch": 0.16744139551157095,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00049994856935666,
      "loss": 6.9704,
      "mean_token_accuracy": 0.11796347573399543,
      "num_tokens": 3280069.0,
      "step": 1675
    },
    {
      "entropy": 7.02333631515503,
      "epoch": 0.16794122057279953,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499947803489986,
      "loss": 6.9613,
      "mean_token_accuracy": 0.11881004869937897,
      "num_tokens": 3289520.0,
      "step": 1680
    },
    {
      "entropy": 7.031835699081421,
      "epoch": 0.1684410456340281,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999470319636829,
      "loss": 6.9666,
      "mean_token_accuracy": 0.11380762606859207,
      "num_tokens": 3299162.0,
      "step": 1685
    },
    {
      "entropy": 6.830581951141357,
      "epoch": 0.16894087069525665,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999462547777705,
      "loss": 6.9635,
      "mean_token_accuracy": 0.1131022647023201,
      "num_tokens": 3308184.0,
      "step": 1690
    },
    {
      "entropy": 7.141867733001709,
      "epoch": 0.16944069575648524,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999454719322682,
      "loss": 6.9862,
      "mean_token_accuracy": 0.1092725582420826,
      "num_tokens": 3317647.0,
      "step": 1695
    },
    {
      "entropy": 6.935384607315063,
      "epoch": 0.1699405208177138,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999446834271958,
      "loss": 7.0218,
      "mean_token_accuracy": 0.11219843700528145,
      "num_tokens": 3329437.0,
      "step": 1700
    },
    {
      "entropy": 7.0339641094207765,
      "epoch": 0.17044034587894236,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999438892625728,
      "loss": 6.9907,
      "mean_token_accuracy": 0.11866170018911362,
      "num_tokens": 3340011.0,
      "step": 1705
    },
    {
      "entropy": 6.95525631904602,
      "epoch": 0.17094017094017094,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999430894384198,
      "loss": 6.9671,
      "mean_token_accuracy": 0.11526316031813622,
      "num_tokens": 3348979.0,
      "step": 1710
    },
    {
      "entropy": 7.004418611526489,
      "epoch": 0.1714399960013995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999422839547564,
      "loss": 7.0279,
      "mean_token_accuracy": 0.1140359953045845,
      "num_tokens": 3359055.0,
      "step": 1715
    },
    {
      "entropy": 7.011244440078736,
      "epoch": 0.1719398210626281,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004999414728116029,
      "loss": 6.9398,
      "mean_token_accuracy": 0.11288423910737037,
      "num_tokens": 3368206.0,
      "step": 1720
    },
    {
      "entropy": 6.919700384140015,
      "epoch": 0.17243964612385665,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999406560089798,
      "loss": 6.9835,
      "mean_token_accuracy": 0.11329871043562889,
      "num_tokens": 3378522.0,
      "step": 1725
    },
    {
      "entropy": 7.055575942993164,
      "epoch": 0.1729394711850852,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999398335469078,
      "loss": 6.9712,
      "mean_token_accuracy": 0.117507603764534,
      "num_tokens": 3389052.0,
      "step": 1730
    },
    {
      "entropy": 6.872907257080078,
      "epoch": 0.1734392962463138,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999390054254074,
      "loss": 6.8651,
      "mean_token_accuracy": 0.12086718901991844,
      "num_tokens": 3398269.0,
      "step": 1735
    },
    {
      "entropy": 7.002393960952759,
      "epoch": 0.17393912130754235,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999381716444996,
      "loss": 6.9319,
      "mean_token_accuracy": 0.11592117547988892,
      "num_tokens": 3408063.0,
      "step": 1740
    },
    {
      "entropy": 6.98644380569458,
      "epoch": 0.17443894636877094,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999373322042052,
      "loss": 6.9981,
      "mean_token_accuracy": 0.11340354084968567,
      "num_tokens": 3418903.0,
      "step": 1745
    },
    {
      "entropy": 6.914481353759766,
      "epoch": 0.1749387714299995,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999364871045455,
      "loss": 6.9073,
      "mean_token_accuracy": 0.11716172620654106,
      "num_tokens": 3428831.0,
      "step": 1750
    },
    {
      "entropy": 6.91617751121521,
      "epoch": 0.17543859649122806,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004999356363455415,
      "loss": 6.9844,
      "mean_token_accuracy": 0.11535566821694374,
      "num_tokens": 3438314.0,
      "step": 1755
    },
    {
      "entropy": 7.010489749908447,
      "epoch": 0.17593842155245665,
      "grad_norm": 0.95703125,
      "learning_rate": 0.000499934779927215,
      "loss": 6.9705,
      "mean_token_accuracy": 0.11254550069570542,
      "num_tokens": 3449248.0,
      "step": 1760
    },
    {
      "entropy": 6.90408673286438,
      "epoch": 0.1764382466136852,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999339178495872,
      "loss": 6.9531,
      "mean_token_accuracy": 0.11303596422076226,
      "num_tokens": 3460119.0,
      "step": 1765
    },
    {
      "entropy": 6.994387435913086,
      "epoch": 0.1769380716749138,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999330501126798,
      "loss": 6.884,
      "mean_token_accuracy": 0.11875045150518418,
      "num_tokens": 3469610.0,
      "step": 1770
    },
    {
      "entropy": 6.824668645858765,
      "epoch": 0.17743789673614235,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999321767165148,
      "loss": 6.9264,
      "mean_token_accuracy": 0.11300530284643173,
      "num_tokens": 3480219.0,
      "step": 1775
    },
    {
      "entropy": 7.031301259994507,
      "epoch": 0.1779377217973709,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999312976611142,
      "loss": 6.8611,
      "mean_token_accuracy": 0.11751789450645447,
      "num_tokens": 3489445.0,
      "step": 1780
    },
    {
      "entropy": 6.933603763580322,
      "epoch": 0.1784375468585995,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999304129465,
      "loss": 6.894,
      "mean_token_accuracy": 0.11873501911759377,
      "num_tokens": 3499164.0,
      "step": 1785
    },
    {
      "entropy": 6.993748378753662,
      "epoch": 0.17893737191982806,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999295225726945,
      "loss": 7.0016,
      "mean_token_accuracy": 0.10571552962064742,
      "num_tokens": 3510242.0,
      "step": 1790
    },
    {
      "entropy": 6.956596851348877,
      "epoch": 0.17943719698105662,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049992862653972,
      "loss": 6.9074,
      "mean_token_accuracy": 0.11266488656401634,
      "num_tokens": 3519481.0,
      "step": 1795
    },
    {
      "entropy": 6.899724102020263,
      "epoch": 0.1799370220422852,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999277248475992,
      "loss": 6.9373,
      "mean_token_accuracy": 0.11640474870800972,
      "num_tokens": 3529706.0,
      "step": 1800
    },
    {
      "entropy": 6.865712833404541,
      "epoch": 0.18043684710351376,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999268174963547,
      "loss": 6.8878,
      "mean_token_accuracy": 0.11141645014286042,
      "num_tokens": 3539239.0,
      "step": 1805
    },
    {
      "entropy": 6.976277828216553,
      "epoch": 0.18093667216474235,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999259044860093,
      "loss": 6.8532,
      "mean_token_accuracy": 0.12506242021918296,
      "num_tokens": 3548177.0,
      "step": 1810
    },
    {
      "entropy": 6.886385822296143,
      "epoch": 0.1814364972259709,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999249858165861,
      "loss": 6.927,
      "mean_token_accuracy": 0.11031646355986595,
      "num_tokens": 3557957.0,
      "step": 1815
    },
    {
      "entropy": 6.96489896774292,
      "epoch": 0.18193632228719947,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999240614881082,
      "loss": 6.9478,
      "mean_token_accuracy": 0.11222271621227264,
      "num_tokens": 3567701.0,
      "step": 1820
    },
    {
      "entropy": 6.956703519821167,
      "epoch": 0.18243614734842806,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999231315005987,
      "loss": 6.887,
      "mean_token_accuracy": 0.11699776723980904,
      "num_tokens": 3575830.0,
      "step": 1825
    },
    {
      "entropy": 6.923069858551026,
      "epoch": 0.18293597240965662,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999221958540811,
      "loss": 6.872,
      "mean_token_accuracy": 0.12267111539840699,
      "num_tokens": 3585453.0,
      "step": 1830
    },
    {
      "entropy": 6.869595146179199,
      "epoch": 0.1834357974708852,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999212545485789,
      "loss": 6.8817,
      "mean_token_accuracy": 0.11956133246421814,
      "num_tokens": 3595340.0,
      "step": 1835
    },
    {
      "entropy": 6.947260522842408,
      "epoch": 0.18393562253211376,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004999203075841159,
      "loss": 6.9164,
      "mean_token_accuracy": 0.11490913555026054,
      "num_tokens": 3605083.0,
      "step": 1840
    },
    {
      "entropy": 6.913357925415039,
      "epoch": 0.18443544759334232,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999193549607157,
      "loss": 6.8603,
      "mean_token_accuracy": 0.115931436419487,
      "num_tokens": 3614852.0,
      "step": 1845
    },
    {
      "entropy": 6.967413473129272,
      "epoch": 0.1849352726545709,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999183966784026,
      "loss": 6.869,
      "mean_token_accuracy": 0.11511830314993858,
      "num_tokens": 3624267.0,
      "step": 1850
    },
    {
      "entropy": 6.8707140445709225,
      "epoch": 0.18543509771579947,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004999174327372004,
      "loss": 6.8022,
      "mean_token_accuracy": 0.1178055003285408,
      "num_tokens": 3633826.0,
      "step": 1855
    },
    {
      "entropy": 6.904180192947388,
      "epoch": 0.18593492277702803,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999164631371335,
      "loss": 6.9176,
      "mean_token_accuracy": 0.11305896788835526,
      "num_tokens": 3643236.0,
      "step": 1860
    },
    {
      "entropy": 6.843501710891724,
      "epoch": 0.18643474783825661,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999154878782262,
      "loss": 6.7872,
      "mean_token_accuracy": 0.11652820855379105,
      "num_tokens": 3652859.0,
      "step": 1865
    },
    {
      "entropy": 6.843868541717529,
      "epoch": 0.18693457289948517,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499914506960503,
      "loss": 6.9538,
      "mean_token_accuracy": 0.11344718858599663,
      "num_tokens": 3663145.0,
      "step": 1870
    },
    {
      "entropy": 6.950706338882446,
      "epoch": 0.18743439796071376,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999135203839889,
      "loss": 6.9009,
      "mean_token_accuracy": 0.11639713943004608,
      "num_tokens": 3673492.0,
      "step": 1875
    },
    {
      "entropy": 6.898825740814209,
      "epoch": 0.18793422302194232,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999125281487084,
      "loss": 6.912,
      "mean_token_accuracy": 0.11780212670564652,
      "num_tokens": 3683582.0,
      "step": 1880
    },
    {
      "entropy": 6.920530509948731,
      "epoch": 0.18843404808317088,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999115302546866,
      "loss": 6.8498,
      "mean_token_accuracy": 0.11961384862661362,
      "num_tokens": 3693075.0,
      "step": 1885
    },
    {
      "entropy": 6.830911254882812,
      "epoch": 0.18893387314439947,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999105267019486,
      "loss": 6.8987,
      "mean_token_accuracy": 0.11175734475255013,
      "num_tokens": 3702290.0,
      "step": 1890
    },
    {
      "entropy": 6.879076433181763,
      "epoch": 0.18943369820562803,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999095174905195,
      "loss": 6.873,
      "mean_token_accuracy": 0.11547678783535957,
      "num_tokens": 3711811.0,
      "step": 1895
    },
    {
      "entropy": 6.915720272064209,
      "epoch": 0.1899335232668566,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999085026204249,
      "loss": 6.8422,
      "mean_token_accuracy": 0.1173863522708416,
      "num_tokens": 3720863.0,
      "step": 1900
    },
    {
      "entropy": 6.956280708312988,
      "epoch": 0.19043334832808517,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999074820916903,
      "loss": 7.0935,
      "mean_token_accuracy": 0.10641425475478172,
      "num_tokens": 3731797.0,
      "step": 1905
    },
    {
      "entropy": 6.9982743740081785,
      "epoch": 0.19093317338931373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999064559043412,
      "loss": 6.7111,
      "mean_token_accuracy": 0.12687933146953584,
      "num_tokens": 3741083.0,
      "step": 1910
    },
    {
      "entropy": 6.688722705841064,
      "epoch": 0.19143299845054232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004999054240584037,
      "loss": 6.838,
      "mean_token_accuracy": 0.11955974251031876,
      "num_tokens": 3750438.0,
      "step": 1915
    },
    {
      "entropy": 6.99309196472168,
      "epoch": 0.19193282351177088,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004999043865539035,
      "loss": 6.8709,
      "mean_token_accuracy": 0.11821675300598145,
      "num_tokens": 3759763.0,
      "step": 1920
    },
    {
      "entropy": 6.875771236419678,
      "epoch": 0.19243264857299944,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004999033433908667,
      "loss": 6.7974,
      "mean_token_accuracy": 0.12742093130946158,
      "num_tokens": 3769523.0,
      "step": 1925
    },
    {
      "entropy": 6.739752101898193,
      "epoch": 0.19293247363422802,
      "grad_norm": 0.8828125,
      "learning_rate": 0.0004999022945693198,
      "loss": 6.8347,
      "mean_token_accuracy": 0.1254890479147434,
      "num_tokens": 3779290.0,
      "step": 1930
    },
    {
      "entropy": 6.964044713973999,
      "epoch": 0.19343229869545658,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499901240089289,
      "loss": 6.8592,
      "mean_token_accuracy": 0.12065653949975967,
      "num_tokens": 3788582.0,
      "step": 1935
    },
    {
      "entropy": 6.872059679031372,
      "epoch": 0.19393212375668517,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004999001799508009,
      "loss": 6.85,
      "mean_token_accuracy": 0.11490488201379775,
      "num_tokens": 3798281.0,
      "step": 1940
    },
    {
      "entropy": 6.848344993591309,
      "epoch": 0.19443194881791373,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998991141538821,
      "loss": 6.823,
      "mean_token_accuracy": 0.11408965140581132,
      "num_tokens": 3807775.0,
      "step": 1945
    },
    {
      "entropy": 6.9306305885314945,
      "epoch": 0.1949317738791423,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998980426985596,
      "loss": 6.9891,
      "mean_token_accuracy": 0.11584151759743691,
      "num_tokens": 3818852.0,
      "step": 1950
    },
    {
      "entropy": 6.973212671279907,
      "epoch": 0.19543159894037088,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00049989696558486,
      "loss": 6.8774,
      "mean_token_accuracy": 0.11234008148312569,
      "num_tokens": 3826924.0,
      "step": 1955
    },
    {
      "entropy": 6.759911155700683,
      "epoch": 0.19593142400159944,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998958828128108,
      "loss": 6.7857,
      "mean_token_accuracy": 0.12838144823908806,
      "num_tokens": 3835098.0,
      "step": 1960
    },
    {
      "entropy": 6.982933855056762,
      "epoch": 0.19643124906282802,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004998947943824389,
      "loss": 6.9578,
      "mean_token_accuracy": 0.11286306157708167,
      "num_tokens": 3845914.0,
      "step": 1965
    },
    {
      "entropy": 6.852990055084229,
      "epoch": 0.19693107412405658,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998937002937719,
      "loss": 6.8304,
      "mean_token_accuracy": 0.12221820950508118,
      "num_tokens": 3855791.0,
      "step": 1970
    },
    {
      "entropy": 6.869904279708862,
      "epoch": 0.19743089918528514,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998926005468373,
      "loss": 6.8011,
      "mean_token_accuracy": 0.1274337239563465,
      "num_tokens": 3865970.0,
      "step": 1975
    },
    {
      "entropy": 6.778682661056519,
      "epoch": 0.19793072424651373,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998914951416627,
      "loss": 6.8579,
      "mean_token_accuracy": 0.12313915342092514,
      "num_tokens": 3875139.0,
      "step": 1980
    },
    {
      "entropy": 6.860070657730103,
      "epoch": 0.1984305493077423,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499890384078276,
      "loss": 6.8678,
      "mean_token_accuracy": 0.12257373183965684,
      "num_tokens": 3884075.0,
      "step": 1985
    },
    {
      "entropy": 6.945457220077515,
      "epoch": 0.19893037436897085,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998892673567051,
      "loss": 6.7782,
      "mean_token_accuracy": 0.11979793757200241,
      "num_tokens": 3894273.0,
      "step": 1990
    },
    {
      "entropy": 6.792132806777954,
      "epoch": 0.19943019943019943,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998881449769781,
      "loss": 6.8492,
      "mean_token_accuracy": 0.11639444679021835,
      "num_tokens": 3904940.0,
      "step": 1995
    },
    {
      "entropy": 7.058999633789062,
      "epoch": 0.199930024491428,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004998870169391232,
      "loss": 6.9935,
      "mean_token_accuracy": 0.10505019277334213,
      "num_tokens": 3916273.0,
      "step": 2000
    },
    {
      "entropy": 6.793080759048462,
      "epoch": 0.20042984955265658,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499885883243169,
      "loss": 6.7725,
      "mean_token_accuracy": 0.1189383864402771,
      "num_tokens": 3926495.0,
      "step": 2005
    },
    {
      "entropy": 6.804683971405029,
      "epoch": 0.20092967461388514,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998847438891437,
      "loss": 6.855,
      "mean_token_accuracy": 0.11757523566484451,
      "num_tokens": 3936286.0,
      "step": 2010
    },
    {
      "entropy": 6.880439329147339,
      "epoch": 0.2014294996751137,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998835988770761,
      "loss": 6.8972,
      "mean_token_accuracy": 0.11336359679698944,
      "num_tokens": 3946924.0,
      "step": 2015
    },
    {
      "entropy": 6.921899366378784,
      "epoch": 0.20192932473634229,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499882448206995,
      "loss": 6.8214,
      "mean_token_accuracy": 0.11907981783151626,
      "num_tokens": 3956815.0,
      "step": 2020
    },
    {
      "entropy": 6.759397315979004,
      "epoch": 0.20242914979757085,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998812918789293,
      "loss": 6.7921,
      "mean_token_accuracy": 0.12498126104474068,
      "num_tokens": 3967034.0,
      "step": 2025
    },
    {
      "entropy": 6.884777498245239,
      "epoch": 0.20292897485879943,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998801298929084,
      "loss": 6.8584,
      "mean_token_accuracy": 0.11634212955832482,
      "num_tokens": 3976800.0,
      "step": 2030
    },
    {
      "entropy": 6.952639722824097,
      "epoch": 0.203428799920028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998789622489611,
      "loss": 6.9162,
      "mean_token_accuracy": 0.11593235284090042,
      "num_tokens": 3987279.0,
      "step": 2035
    },
    {
      "entropy": 6.809817600250244,
      "epoch": 0.20392862498125655,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998777889471171,
      "loss": 6.8146,
      "mean_token_accuracy": 0.11954162642359734,
      "num_tokens": 3998414.0,
      "step": 2040
    },
    {
      "entropy": 6.752009677886963,
      "epoch": 0.20442845004248514,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998766099874056,
      "loss": 6.7814,
      "mean_token_accuracy": 0.12704155817627907,
      "num_tokens": 4007994.0,
      "step": 2045
    },
    {
      "entropy": 6.8828105449676515,
      "epoch": 0.2049282751037137,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998754253698566,
      "loss": 6.8057,
      "mean_token_accuracy": 0.11565538421273232,
      "num_tokens": 4018285.0,
      "step": 2050
    },
    {
      "entropy": 6.942424201965332,
      "epoch": 0.20542810016494226,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998742350944998,
      "loss": 6.8811,
      "mean_token_accuracy": 0.11090886443853379,
      "num_tokens": 4028689.0,
      "step": 2055
    },
    {
      "entropy": 6.7840094566345215,
      "epoch": 0.20592792522617084,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998730391613651,
      "loss": 6.7998,
      "mean_token_accuracy": 0.12368144318461419,
      "num_tokens": 4038944.0,
      "step": 2060
    },
    {
      "entropy": 6.943382596969604,
      "epoch": 0.2064277502873994,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998718375704825,
      "loss": 6.9236,
      "mean_token_accuracy": 0.11197515055537224,
      "num_tokens": 4049048.0,
      "step": 2065
    },
    {
      "entropy": 6.869805145263672,
      "epoch": 0.206927575348628,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998706303218825,
      "loss": 6.7558,
      "mean_token_accuracy": 0.12332033663988114,
      "num_tokens": 4058710.0,
      "step": 2070
    },
    {
      "entropy": 6.802251243591309,
      "epoch": 0.20742740040985655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998694174155952,
      "loss": 6.7397,
      "mean_token_accuracy": 0.1152039721608162,
      "num_tokens": 4069517.0,
      "step": 2075
    },
    {
      "entropy": 6.766680097579956,
      "epoch": 0.2079272254710851,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004998681988516512,
      "loss": 6.8445,
      "mean_token_accuracy": 0.1188851423561573,
      "num_tokens": 4080730.0,
      "step": 2080
    },
    {
      "entropy": 6.908146905899048,
      "epoch": 0.2084270505323137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998669746300811,
      "loss": 6.7327,
      "mean_token_accuracy": 0.12420074567198754,
      "num_tokens": 4090505.0,
      "step": 2085
    },
    {
      "entropy": 6.732927560806274,
      "epoch": 0.20892687559354226,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998657447509159,
      "loss": 6.7299,
      "mean_token_accuracy": 0.12293928116559982,
      "num_tokens": 4099963.0,
      "step": 2090
    },
    {
      "entropy": 6.810068845748901,
      "epoch": 0.20942670065477084,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998645092141864,
      "loss": 6.8291,
      "mean_token_accuracy": 0.12254174649715424,
      "num_tokens": 4109340.0,
      "step": 2095
    },
    {
      "entropy": 7.008067846298218,
      "epoch": 0.2099265257159994,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998632680199238,
      "loss": 6.8521,
      "mean_token_accuracy": 0.11651415005326271,
      "num_tokens": 4118476.0,
      "step": 2100
    },
    {
      "entropy": 6.791560316085816,
      "epoch": 0.21042635077722796,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998620211681591,
      "loss": 6.7869,
      "mean_token_accuracy": 0.12428712770342827,
      "num_tokens": 4127922.0,
      "step": 2105
    },
    {
      "entropy": 6.768087768554688,
      "epoch": 0.21092617583845655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998607686589239,
      "loss": 6.7947,
      "mean_token_accuracy": 0.11915560439229012,
      "num_tokens": 4137202.0,
      "step": 2110
    },
    {
      "entropy": 6.775341320037842,
      "epoch": 0.2114260008996851,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998595104922496,
      "loss": 6.7885,
      "mean_token_accuracy": 0.1231108471751213,
      "num_tokens": 4146789.0,
      "step": 2115
    },
    {
      "entropy": 6.813695859909058,
      "epoch": 0.2119258259609137,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998582466681678,
      "loss": 6.8566,
      "mean_token_accuracy": 0.11805249378085136,
      "num_tokens": 4157628.0,
      "step": 2120
    },
    {
      "entropy": 6.864239072799682,
      "epoch": 0.21242565102214225,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998569771867105,
      "loss": 6.8133,
      "mean_token_accuracy": 0.11805640012025834,
      "num_tokens": 4168522.0,
      "step": 2125
    },
    {
      "entropy": 6.8641472339630125,
      "epoch": 0.2129254760833708,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004998557020479095,
      "loss": 6.883,
      "mean_token_accuracy": 0.12175262719392776,
      "num_tokens": 4178590.0,
      "step": 2130
    },
    {
      "entropy": 6.829242372512818,
      "epoch": 0.2134253011445994,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998544212517968,
      "loss": 6.6766,
      "mean_token_accuracy": 0.13151731938123704,
      "num_tokens": 4188004.0,
      "step": 2135
    },
    {
      "entropy": 6.81755428314209,
      "epoch": 0.21392512620582796,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998531347984047,
      "loss": 6.7795,
      "mean_token_accuracy": 0.11880608722567558,
      "num_tokens": 4197481.0,
      "step": 2140
    },
    {
      "entropy": 6.842449045181274,
      "epoch": 0.21442495126705652,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998518426877656,
      "loss": 6.8284,
      "mean_token_accuracy": 0.12467390969395638,
      "num_tokens": 4207187.0,
      "step": 2145
    },
    {
      "entropy": 6.799112701416016,
      "epoch": 0.2149247763282851,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998505449199122,
      "loss": 6.8034,
      "mean_token_accuracy": 0.12435225024819374,
      "num_tokens": 4216485.0,
      "step": 2150
    },
    {
      "entropy": 6.803231239318848,
      "epoch": 0.21542460138951366,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998492414948768,
      "loss": 6.88,
      "mean_token_accuracy": 0.11924681589007377,
      "num_tokens": 4228268.0,
      "step": 2155
    },
    {
      "entropy": 6.819182586669922,
      "epoch": 0.21592442645074225,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998479324126924,
      "loss": 6.7148,
      "mean_token_accuracy": 0.12601147666573526,
      "num_tokens": 4237755.0,
      "step": 2160
    },
    {
      "entropy": 6.888819837570191,
      "epoch": 0.2164242515119708,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998466176733918,
      "loss": 6.8649,
      "mean_token_accuracy": 0.11659305468201638,
      "num_tokens": 4247571.0,
      "step": 2165
    },
    {
      "entropy": 6.720103931427002,
      "epoch": 0.21692407657319937,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998452972770082,
      "loss": 6.8112,
      "mean_token_accuracy": 0.12066830769181251,
      "num_tokens": 4257003.0,
      "step": 2170
    },
    {
      "entropy": 6.877090406417847,
      "epoch": 0.21742390163442796,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004998439712235747,
      "loss": 6.8425,
      "mean_token_accuracy": 0.11793331429362297,
      "num_tokens": 4266412.0,
      "step": 2175
    },
    {
      "entropy": 6.851199150085449,
      "epoch": 0.21792372669565652,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004998426395131248,
      "loss": 6.7926,
      "mean_token_accuracy": 0.11362633481621742,
      "num_tokens": 4276361.0,
      "step": 2180
    },
    {
      "entropy": 6.8144981384277346,
      "epoch": 0.2184235517568851,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998413021456919,
      "loss": 6.777,
      "mean_token_accuracy": 0.12194588780403137,
      "num_tokens": 4285815.0,
      "step": 2185
    },
    {
      "entropy": 6.815514469146729,
      "epoch": 0.21892337681811366,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998399591213098,
      "loss": 6.7683,
      "mean_token_accuracy": 0.12466176748275756,
      "num_tokens": 4296071.0,
      "step": 2190
    },
    {
      "entropy": 6.768910217285156,
      "epoch": 0.21942320187934222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998386104400122,
      "loss": 6.7675,
      "mean_token_accuracy": 0.12030394151806831,
      "num_tokens": 4306476.0,
      "step": 2195
    },
    {
      "entropy": 6.856772565841675,
      "epoch": 0.2199230269405708,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998372561018329,
      "loss": 6.8139,
      "mean_token_accuracy": 0.12123706415295601,
      "num_tokens": 4316312.0,
      "step": 2200
    },
    {
      "entropy": 6.765644359588623,
      "epoch": 0.22042285200179937,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998358961068062,
      "loss": 6.7023,
      "mean_token_accuracy": 0.13033238649368287,
      "num_tokens": 4325443.0,
      "step": 2205
    },
    {
      "entropy": 6.814023971557617,
      "epoch": 0.22092267706302793,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004998345304549661,
      "loss": 6.7587,
      "mean_token_accuracy": 0.11952677890658378,
      "num_tokens": 4334354.0,
      "step": 2210
    },
    {
      "entropy": 6.719250440597534,
      "epoch": 0.22142250212425652,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998331591463471,
      "loss": 6.665,
      "mean_token_accuracy": 0.12632832303643227,
      "num_tokens": 4344714.0,
      "step": 2215
    },
    {
      "entropy": 6.773249101638794,
      "epoch": 0.22192232718548507,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004998317821809837,
      "loss": 6.7179,
      "mean_token_accuracy": 0.12060520201921462,
      "num_tokens": 4354216.0,
      "step": 2220
    },
    {
      "entropy": 6.795089530944824,
      "epoch": 0.22242215224671366,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998303995589104,
      "loss": 6.7614,
      "mean_token_accuracy": 0.1252301424741745,
      "num_tokens": 4364764.0,
      "step": 2225
    },
    {
      "entropy": 6.863585424423218,
      "epoch": 0.22292197730794222,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998290112801623,
      "loss": 6.8389,
      "mean_token_accuracy": 0.11243418082594872,
      "num_tokens": 4375289.0,
      "step": 2230
    },
    {
      "entropy": 6.749932384490966,
      "epoch": 0.22342180236917078,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499827617344774,
      "loss": 6.7263,
      "mean_token_accuracy": 0.12714286744594575,
      "num_tokens": 4384984.0,
      "step": 2235
    },
    {
      "entropy": 6.787637710571289,
      "epoch": 0.22392162743039937,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998262177527807,
      "loss": 6.7057,
      "mean_token_accuracy": 0.12606788277626038,
      "num_tokens": 4393321.0,
      "step": 2240
    },
    {
      "entropy": 6.711228847503662,
      "epoch": 0.22442145249162793,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998248125042177,
      "loss": 6.7929,
      "mean_token_accuracy": 0.1157205656170845,
      "num_tokens": 4404908.0,
      "step": 2245
    },
    {
      "entropy": 6.944765281677246,
      "epoch": 0.2249212775528565,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998234015991202,
      "loss": 6.8806,
      "mean_token_accuracy": 0.11683219075202941,
      "num_tokens": 4415530.0,
      "step": 2250
    },
    {
      "entropy": 6.803278064727783,
      "epoch": 0.22542110261408507,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998219850375238,
      "loss": 6.7849,
      "mean_token_accuracy": 0.11715331003069877,
      "num_tokens": 4425678.0,
      "step": 2255
    },
    {
      "entropy": 6.70594801902771,
      "epoch": 0.22592092767531363,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998205628194641,
      "loss": 6.7662,
      "mean_token_accuracy": 0.12059013247489929,
      "num_tokens": 4436890.0,
      "step": 2260
    },
    {
      "entropy": 6.824479770660401,
      "epoch": 0.22642075273654222,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004998191349449768,
      "loss": 6.5938,
      "mean_token_accuracy": 0.12804949507117272,
      "num_tokens": 4446947.0,
      "step": 2265
    },
    {
      "entropy": 6.856104373931885,
      "epoch": 0.22692057779777078,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998177014140981,
      "loss": 6.7874,
      "mean_token_accuracy": 0.12311343476176262,
      "num_tokens": 4457052.0,
      "step": 2270
    },
    {
      "entropy": 6.644788122177124,
      "epoch": 0.22742040285899934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998162622268637,
      "loss": 6.6125,
      "mean_token_accuracy": 0.13069870918989182,
      "num_tokens": 4466171.0,
      "step": 2275
    },
    {
      "entropy": 6.739696311950683,
      "epoch": 0.22792022792022792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998148173833101,
      "loss": 6.8447,
      "mean_token_accuracy": 0.11991135403513908,
      "num_tokens": 4476147.0,
      "step": 2280
    },
    {
      "entropy": 6.839726161956787,
      "epoch": 0.22842005298145648,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004998133668834735,
      "loss": 6.7495,
      "mean_token_accuracy": 0.12500213459134102,
      "num_tokens": 4485786.0,
      "step": 2285
    },
    {
      "entropy": 6.724114465713501,
      "epoch": 0.22891987804268507,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004998119107273905,
      "loss": 6.8116,
      "mean_token_accuracy": 0.12492655441164971,
      "num_tokens": 4495376.0,
      "step": 2290
    },
    {
      "entropy": 6.879471206665039,
      "epoch": 0.22941970310391363,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004998104489150975,
      "loss": 6.7649,
      "mean_token_accuracy": 0.1197766162455082,
      "num_tokens": 4504500.0,
      "step": 2295
    },
    {
      "entropy": 6.749096775054932,
      "epoch": 0.2299195281651422,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998089814466318,
      "loss": 6.8406,
      "mean_token_accuracy": 0.11749223545193672,
      "num_tokens": 4514842.0,
      "step": 2300
    },
    {
      "entropy": 6.818306827545166,
      "epoch": 0.23041935322637078,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004998075083220295,
      "loss": 6.7133,
      "mean_token_accuracy": 0.12731910049915313,
      "num_tokens": 4523794.0,
      "step": 2305
    },
    {
      "entropy": 6.823913145065307,
      "epoch": 0.23091917828759934,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998060295413286,
      "loss": 6.7766,
      "mean_token_accuracy": 0.11975430250167847,
      "num_tokens": 4533315.0,
      "step": 2310
    },
    {
      "entropy": 6.660467863082886,
      "epoch": 0.23141900334882792,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998045451045656,
      "loss": 6.6327,
      "mean_token_accuracy": 0.1325990788638592,
      "num_tokens": 4542996.0,
      "step": 2315
    },
    {
      "entropy": 6.791189050674438,
      "epoch": 0.23191882841005648,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998030550117782,
      "loss": 6.7008,
      "mean_token_accuracy": 0.12028939574956894,
      "num_tokens": 4552719.0,
      "step": 2320
    },
    {
      "entropy": 6.704330396652222,
      "epoch": 0.23241865347128504,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004998015592630037,
      "loss": 6.648,
      "mean_token_accuracy": 0.12924702242016792,
      "num_tokens": 4562040.0,
      "step": 2325
    },
    {
      "entropy": 6.669022941589356,
      "epoch": 0.23291847853251363,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998000578582798,
      "loss": 6.6894,
      "mean_token_accuracy": 0.12677412778139113,
      "num_tokens": 4570855.0,
      "step": 2330
    },
    {
      "entropy": 6.832343626022339,
      "epoch": 0.2334183035937422,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997985507976443,
      "loss": 6.8056,
      "mean_token_accuracy": 0.11848708391189575,
      "num_tokens": 4581814.0,
      "step": 2335
    },
    {
      "entropy": 6.751095962524414,
      "epoch": 0.23391812865497075,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997970380811351,
      "loss": 6.7034,
      "mean_token_accuracy": 0.12684452831745147,
      "num_tokens": 4590731.0,
      "step": 2340
    },
    {
      "entropy": 6.754598665237427,
      "epoch": 0.23441795371619933,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997955197087904,
      "loss": 6.6995,
      "mean_token_accuracy": 0.11840205565094948,
      "num_tokens": 4600440.0,
      "step": 2345
    },
    {
      "entropy": 6.735138940811157,
      "epoch": 0.2349177787774279,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997939956806481,
      "loss": 6.6952,
      "mean_token_accuracy": 0.12488049641251564,
      "num_tokens": 4609510.0,
      "step": 2350
    },
    {
      "entropy": 6.743107795715332,
      "epoch": 0.23541760383865648,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004997924659967467,
      "loss": 6.7829,
      "mean_token_accuracy": 0.12009901478886605,
      "num_tokens": 4620481.0,
      "step": 2355
    },
    {
      "entropy": 6.784450817108154,
      "epoch": 0.23591742889988504,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997909306571248,
      "loss": 6.6512,
      "mean_token_accuracy": 0.1273428089916706,
      "num_tokens": 4629388.0,
      "step": 2360
    },
    {
      "entropy": 6.603338050842285,
      "epoch": 0.2364172539611136,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997893896618209,
      "loss": 6.6389,
      "mean_token_accuracy": 0.1293751984834671,
      "num_tokens": 4638429.0,
      "step": 2365
    },
    {
      "entropy": 6.7959905624389645,
      "epoch": 0.2369170790223422,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997878430108738,
      "loss": 6.7779,
      "mean_token_accuracy": 0.12506488636136054,
      "num_tokens": 4648995.0,
      "step": 2370
    },
    {
      "entropy": 6.788811445236206,
      "epoch": 0.23741690408357075,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997862907043224,
      "loss": 6.6646,
      "mean_token_accuracy": 0.1183782771229744,
      "num_tokens": 4658584.0,
      "step": 2375
    },
    {
      "entropy": 6.686025953292846,
      "epoch": 0.23791672914479933,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997847327422059,
      "loss": 6.7057,
      "mean_token_accuracy": 0.12550745904445648,
      "num_tokens": 4668859.0,
      "step": 2380
    },
    {
      "entropy": 6.788628959655762,
      "epoch": 0.2384165542060279,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0004997831691245632,
      "loss": 6.7445,
      "mean_token_accuracy": 0.1137440986931324,
      "num_tokens": 4678389.0,
      "step": 2385
    },
    {
      "entropy": 6.700611639022827,
      "epoch": 0.23891637926725645,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004997815998514337,
      "loss": 6.7501,
      "mean_token_accuracy": 0.12377952635288239,
      "num_tokens": 4688763.0,
      "step": 2390
    },
    {
      "entropy": 6.7670745849609375,
      "epoch": 0.23941620432848504,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997800249228572,
      "loss": 6.628,
      "mean_token_accuracy": 0.1264137051999569,
      "num_tokens": 4698180.0,
      "step": 2395
    },
    {
      "entropy": 6.683659410476684,
      "epoch": 0.2399160293897136,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004997784443388732,
      "loss": 6.6073,
      "mean_token_accuracy": 0.12491724789142608,
      "num_tokens": 4708111.0,
      "step": 2400
    },
    {
      "entropy": 6.736080121994019,
      "epoch": 0.24041585445094216,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997768580995212,
      "loss": 6.7446,
      "mean_token_accuracy": 0.12171546146273612,
      "num_tokens": 4717247.0,
      "step": 2405
    },
    {
      "entropy": 6.755503082275391,
      "epoch": 0.24091567951217074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997752662048413,
      "loss": 6.6726,
      "mean_token_accuracy": 0.12635646536946296,
      "num_tokens": 4727062.0,
      "step": 2410
    },
    {
      "entropy": 6.629852771759033,
      "epoch": 0.2414155045733993,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997736686548736,
      "loss": 6.6241,
      "mean_token_accuracy": 0.13165736421942711,
      "num_tokens": 4735806.0,
      "step": 2415
    },
    {
      "entropy": 6.795121574401856,
      "epoch": 0.2419153296346279,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997720654496582,
      "loss": 6.7557,
      "mean_token_accuracy": 0.12245218381285668,
      "num_tokens": 4745246.0,
      "step": 2420
    },
    {
      "entropy": 6.797226619720459,
      "epoch": 0.24241515469585645,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997704565892355,
      "loss": 6.6616,
      "mean_token_accuracy": 0.12756211906671525,
      "num_tokens": 4754867.0,
      "step": 2425
    },
    {
      "entropy": 6.677945375442505,
      "epoch": 0.242914979757085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997688420736459,
      "loss": 6.7266,
      "mean_token_accuracy": 0.12988825812935828,
      "num_tokens": 4764418.0,
      "step": 2430
    },
    {
      "entropy": 6.6626581192016605,
      "epoch": 0.2434148048183136,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997672219029301,
      "loss": 6.6897,
      "mean_token_accuracy": 0.12267833352088928,
      "num_tokens": 4774137.0,
      "step": 2435
    },
    {
      "entropy": 6.657618188858033,
      "epoch": 0.24391462987954216,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997655960771288,
      "loss": 6.7239,
      "mean_token_accuracy": 0.12646948918700218,
      "num_tokens": 4784503.0,
      "step": 2440
    },
    {
      "entropy": 6.762366390228271,
      "epoch": 0.24441445494077074,
      "grad_norm": 0.9609375,
      "learning_rate": 0.000499763964596283,
      "loss": 6.7776,
      "mean_token_accuracy": 0.118136465549469,
      "num_tokens": 4795847.0,
      "step": 2445
    },
    {
      "entropy": 6.702267074584961,
      "epoch": 0.2449142800019993,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997623274604336,
      "loss": 6.62,
      "mean_token_accuracy": 0.1314212679862976,
      "num_tokens": 4805230.0,
      "step": 2450
    },
    {
      "entropy": 6.675754165649414,
      "epoch": 0.24541410506322786,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499760684669622,
      "loss": 6.5836,
      "mean_token_accuracy": 0.12938425019383432,
      "num_tokens": 4815345.0,
      "step": 2455
    },
    {
      "entropy": 6.669795942306519,
      "epoch": 0.24591393012445645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997590362238894,
      "loss": 6.6613,
      "mean_token_accuracy": 0.12406706139445305,
      "num_tokens": 4824819.0,
      "step": 2460
    },
    {
      "entropy": 6.726718330383301,
      "epoch": 0.246413755185685,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004997573821232772,
      "loss": 6.6956,
      "mean_token_accuracy": 0.12626968026161195,
      "num_tokens": 4834339.0,
      "step": 2465
    },
    {
      "entropy": 6.672310066223145,
      "epoch": 0.24691358024691357,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997557223678272,
      "loss": 6.6875,
      "mean_token_accuracy": 0.1265610449016094,
      "num_tokens": 4844440.0,
      "step": 2470
    },
    {
      "entropy": 6.752259016036987,
      "epoch": 0.24741340530814215,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997540569575809,
      "loss": 6.6733,
      "mean_token_accuracy": 0.12238761708140374,
      "num_tokens": 4854688.0,
      "step": 2475
    },
    {
      "entropy": 6.686040306091309,
      "epoch": 0.2479132303693707,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004997523858925805,
      "loss": 6.6788,
      "mean_token_accuracy": 0.12459717616438866,
      "num_tokens": 4864629.0,
      "step": 2480
    },
    {
      "entropy": 6.6606574058532715,
      "epoch": 0.2484130554305993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997507091728678,
      "loss": 6.6391,
      "mean_token_accuracy": 0.12691431269049644,
      "num_tokens": 4873693.0,
      "step": 2485
    },
    {
      "entropy": 6.66929292678833,
      "epoch": 0.24891288049182786,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004997490267984853,
      "loss": 6.6117,
      "mean_token_accuracy": 0.1293844424188137,
      "num_tokens": 4882640.0,
      "step": 2490
    },
    {
      "entropy": 6.712693500518799,
      "epoch": 0.24941270555305642,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004997473387694749,
      "loss": 6.7247,
      "mean_token_accuracy": 0.12217676192522049,
      "num_tokens": 4893452.0,
      "step": 2495
    },
    {
      "entropy": 6.7423759460449215,
      "epoch": 0.249912530614285,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997456450858793,
      "loss": 6.744,
      "mean_token_accuracy": 0.11771068722009659,
      "num_tokens": 4902949.0,
      "step": 2500
    },
    {
      "entropy": 6.795651006698608,
      "epoch": 0.25041235567551356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997439457477412,
      "loss": 6.7037,
      "mean_token_accuracy": 0.12298239767551422,
      "num_tokens": 4912748.0,
      "step": 2505
    },
    {
      "entropy": 6.6983723640441895,
      "epoch": 0.25091218073674215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997422407551032,
      "loss": 6.6458,
      "mean_token_accuracy": 0.12610778957605362,
      "num_tokens": 4922405.0,
      "step": 2510
    },
    {
      "entropy": 6.636437129974365,
      "epoch": 0.2514120057979707,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997405301080082,
      "loss": 6.6314,
      "mean_token_accuracy": 0.12737202569842337,
      "num_tokens": 4931826.0,
      "step": 2515
    },
    {
      "entropy": 6.669645738601685,
      "epoch": 0.25191183085919927,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997388138064994,
      "loss": 6.6993,
      "mean_token_accuracy": 0.12631320729851722,
      "num_tokens": 4942165.0,
      "step": 2520
    },
    {
      "entropy": 6.79401969909668,
      "epoch": 0.25241165592042786,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997370918506198,
      "loss": 6.653,
      "mean_token_accuracy": 0.12517872527241708,
      "num_tokens": 4951287.0,
      "step": 2525
    },
    {
      "entropy": 6.670285224914551,
      "epoch": 0.25291148098165644,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997353642404129,
      "loss": 6.6299,
      "mean_token_accuracy": 0.12621796652674674,
      "num_tokens": 4960452.0,
      "step": 2530
    },
    {
      "entropy": 6.645723724365235,
      "epoch": 0.253411306042885,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997336309759219,
      "loss": 6.6873,
      "mean_token_accuracy": 0.12333321720361709,
      "num_tokens": 4971310.0,
      "step": 2535
    },
    {
      "entropy": 6.833752012252807,
      "epoch": 0.25391113110411356,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997318920571907,
      "loss": 6.6327,
      "mean_token_accuracy": 0.1271618574857712,
      "num_tokens": 4980609.0,
      "step": 2540
    },
    {
      "entropy": 6.708815956115723,
      "epoch": 0.25441095616534215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499730147484263,
      "loss": 6.6814,
      "mean_token_accuracy": 0.12484947890043259,
      "num_tokens": 4989991.0,
      "step": 2545
    },
    {
      "entropy": 6.629834270477295,
      "epoch": 0.2549107812265707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997283972571824,
      "loss": 6.6538,
      "mean_token_accuracy": 0.12454636245965958,
      "num_tokens": 5000493.0,
      "step": 2550
    },
    {
      "entropy": 6.724473524093628,
      "epoch": 0.25541060628779927,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997266413759934,
      "loss": 6.7358,
      "mean_token_accuracy": 0.1258261241018772,
      "num_tokens": 5010496.0,
      "step": 2555
    },
    {
      "entropy": 6.846870994567871,
      "epoch": 0.25591043134902786,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997248798407397,
      "loss": 6.7987,
      "mean_token_accuracy": 0.11873462423682213,
      "num_tokens": 5020783.0,
      "step": 2560
    },
    {
      "entropy": 6.65616626739502,
      "epoch": 0.2564102564102564,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997231126514659,
      "loss": 6.7041,
      "mean_token_accuracy": 0.12581986486911773,
      "num_tokens": 5029939.0,
      "step": 2565
    },
    {
      "entropy": 6.805445528030395,
      "epoch": 0.256910081471485,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997213398082165,
      "loss": 6.6965,
      "mean_token_accuracy": 0.12117815464735031,
      "num_tokens": 5039491.0,
      "step": 2570
    },
    {
      "entropy": 6.728080701828003,
      "epoch": 0.25740990653271356,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499719561311036,
      "loss": 6.703,
      "mean_token_accuracy": 0.12322363704442978,
      "num_tokens": 5049780.0,
      "step": 2575
    },
    {
      "entropy": 6.738219356536865,
      "epoch": 0.2579097315939421,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997177771599691,
      "loss": 6.6478,
      "mean_token_accuracy": 0.13262535333633424,
      "num_tokens": 5058986.0,
      "step": 2580
    },
    {
      "entropy": 6.644556093215942,
      "epoch": 0.2584095566551707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997159873550607,
      "loss": 6.6067,
      "mean_token_accuracy": 0.1257613942027092,
      "num_tokens": 5068269.0,
      "step": 2585
    },
    {
      "entropy": 6.6656200885772705,
      "epoch": 0.25890938171639927,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997141918963561,
      "loss": 6.7376,
      "mean_token_accuracy": 0.12358493357896805,
      "num_tokens": 5078674.0,
      "step": 2590
    },
    {
      "entropy": 6.735020542144776,
      "epoch": 0.25940920677762785,
      "grad_norm": 1.0,
      "learning_rate": 0.0004997123907839001,
      "loss": 6.6596,
      "mean_token_accuracy": 0.13049351200461387,
      "num_tokens": 5087599.0,
      "step": 2595
    },
    {
      "entropy": 6.6297863006591795,
      "epoch": 0.2599090318388564,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997105840177381,
      "loss": 6.573,
      "mean_token_accuracy": 0.13603698685765267,
      "num_tokens": 5097418.0,
      "step": 2600
    },
    {
      "entropy": 6.63222975730896,
      "epoch": 0.260408856900085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997087715979157,
      "loss": 6.6406,
      "mean_token_accuracy": 0.12458560839295388,
      "num_tokens": 5108248.0,
      "step": 2605
    },
    {
      "entropy": 6.792977571487427,
      "epoch": 0.26090868196131356,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997069535244785,
      "loss": 6.5949,
      "mean_token_accuracy": 0.13078215345740318,
      "num_tokens": 5118284.0,
      "step": 2610
    },
    {
      "entropy": 6.670963287353516,
      "epoch": 0.2614085070225421,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997051297974722,
      "loss": 6.7458,
      "mean_token_accuracy": 0.12364816442131996,
      "num_tokens": 5128485.0,
      "step": 2615
    },
    {
      "entropy": 6.724144172668457,
      "epoch": 0.2619083320837707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997033004169425,
      "loss": 6.6801,
      "mean_token_accuracy": 0.12441147118806839,
      "num_tokens": 5138058.0,
      "step": 2620
    },
    {
      "entropy": 6.594963788986206,
      "epoch": 0.26240815714499927,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997014653829357,
      "loss": 6.5718,
      "mean_token_accuracy": 0.12949975505471228,
      "num_tokens": 5148555.0,
      "step": 2625
    },
    {
      "entropy": 6.686042213439942,
      "epoch": 0.2629079822062278,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996996246954977,
      "loss": 6.7521,
      "mean_token_accuracy": 0.12199698463082313,
      "num_tokens": 5158729.0,
      "step": 2630
    },
    {
      "entropy": 6.8165223598480225,
      "epoch": 0.2634078072674564,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004996977783546752,
      "loss": 6.7889,
      "mean_token_accuracy": 0.12042256221175193,
      "num_tokens": 5168900.0,
      "step": 2635
    },
    {
      "entropy": 6.663141918182373,
      "epoch": 0.26390763232868497,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996959263605143,
      "loss": 6.6848,
      "mean_token_accuracy": 0.12062885612249374,
      "num_tokens": 5179742.0,
      "step": 2640
    },
    {
      "entropy": 6.765869045257569,
      "epoch": 0.2644074573899135,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996940687130617,
      "loss": 6.7068,
      "mean_token_accuracy": 0.12952368706464767,
      "num_tokens": 5190221.0,
      "step": 2645
    },
    {
      "entropy": 6.686702489852905,
      "epoch": 0.2649072824511421,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004996922054123641,
      "loss": 6.6781,
      "mean_token_accuracy": 0.1280037760734558,
      "num_tokens": 5200252.0,
      "step": 2650
    },
    {
      "entropy": 6.598478031158447,
      "epoch": 0.2654071075123707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996903364584685,
      "loss": 6.615,
      "mean_token_accuracy": 0.13113286942243577,
      "num_tokens": 5209388.0,
      "step": 2655
    },
    {
      "entropy": 6.72690258026123,
      "epoch": 0.26590693257359926,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996884618514219,
      "loss": 6.6068,
      "mean_token_accuracy": 0.12769930958747863,
      "num_tokens": 5219806.0,
      "step": 2660
    },
    {
      "entropy": 6.614221572875977,
      "epoch": 0.2664067576348278,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996865815912713,
      "loss": 6.6466,
      "mean_token_accuracy": 0.13306735530495645,
      "num_tokens": 5229485.0,
      "step": 2665
    },
    {
      "entropy": 6.752274036407471,
      "epoch": 0.2669065826960564,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996846956780642,
      "loss": 6.5997,
      "mean_token_accuracy": 0.12850728780031204,
      "num_tokens": 5238922.0,
      "step": 2670
    },
    {
      "entropy": 6.690626525878907,
      "epoch": 0.26740640775728497,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499682804111848,
      "loss": 6.6119,
      "mean_token_accuracy": 0.12641794681549073,
      "num_tokens": 5248391.0,
      "step": 2675
    },
    {
      "entropy": 6.60139684677124,
      "epoch": 0.2679062328185135,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996809068926702,
      "loss": 6.6419,
      "mean_token_accuracy": 0.123379535228014,
      "num_tokens": 5257605.0,
      "step": 2680
    },
    {
      "entropy": 6.855893611907959,
      "epoch": 0.2684060578797421,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004996790040205786,
      "loss": 6.733,
      "mean_token_accuracy": 0.12259264960885048,
      "num_tokens": 5266808.0,
      "step": 2685
    },
    {
      "entropy": 6.6668785572052,
      "epoch": 0.2689058829409707,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996770954956211,
      "loss": 6.588,
      "mean_token_accuracy": 0.12270834892988206,
      "num_tokens": 5276471.0,
      "step": 2690
    },
    {
      "entropy": 6.526516389846802,
      "epoch": 0.2694057080021992,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996751813178456,
      "loss": 6.5825,
      "mean_token_accuracy": 0.1287649765610695,
      "num_tokens": 5285312.0,
      "step": 2695
    },
    {
      "entropy": 6.743730878829956,
      "epoch": 0.2699055330634278,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996732614873005,
      "loss": 6.7142,
      "mean_token_accuracy": 0.1179872803390026,
      "num_tokens": 5295608.0,
      "step": 2700
    },
    {
      "entropy": 6.770171070098877,
      "epoch": 0.2704053581246564,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996713360040339,
      "loss": 6.6397,
      "mean_token_accuracy": 0.11945433616638183,
      "num_tokens": 5305181.0,
      "step": 2705
    },
    {
      "entropy": 6.633085536956787,
      "epoch": 0.27090518318588497,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996694048680942,
      "loss": 6.6363,
      "mean_token_accuracy": 0.1233569286763668,
      "num_tokens": 5314706.0,
      "step": 2710
    },
    {
      "entropy": 6.692262840270996,
      "epoch": 0.2714050082471135,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996674680795302,
      "loss": 6.5719,
      "mean_token_accuracy": 0.12792405486106873,
      "num_tokens": 5324207.0,
      "step": 2715
    },
    {
      "entropy": 6.684712648391724,
      "epoch": 0.2719048333083421,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996655256383905,
      "loss": 6.6373,
      "mean_token_accuracy": 0.13132793381810187,
      "num_tokens": 5333554.0,
      "step": 2720
    },
    {
      "entropy": 6.712253999710083,
      "epoch": 0.2724046583695707,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000499663577544724,
      "loss": 6.7384,
      "mean_token_accuracy": 0.12213098257780075,
      "num_tokens": 5343811.0,
      "step": 2725
    },
    {
      "entropy": 6.668386173248291,
      "epoch": 0.2729044834307992,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996616237985796,
      "loss": 6.6756,
      "mean_token_accuracy": 0.12649924755096437,
      "num_tokens": 5353980.0,
      "step": 2730
    },
    {
      "entropy": 6.679471397399903,
      "epoch": 0.2734043084920278,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996596644000066,
      "loss": 6.6248,
      "mean_token_accuracy": 0.12806927114725114,
      "num_tokens": 5363681.0,
      "step": 2735
    },
    {
      "entropy": 6.6617810249328615,
      "epoch": 0.2739041335532564,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996576993490543,
      "loss": 6.5761,
      "mean_token_accuracy": 0.131463872641325,
      "num_tokens": 5373855.0,
      "step": 2740
    },
    {
      "entropy": 6.67656135559082,
      "epoch": 0.2744039586144849,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004996557286457721,
      "loss": 6.6356,
      "mean_token_accuracy": 0.13073286339640616,
      "num_tokens": 5383831.0,
      "step": 2745
    },
    {
      "entropy": 6.643531417846679,
      "epoch": 0.2749037836757135,
      "grad_norm": 1.125,
      "learning_rate": 0.0004996537522902096,
      "loss": 6.6556,
      "mean_token_accuracy": 0.12728592455387117,
      "num_tokens": 5392247.0,
      "step": 2750
    },
    {
      "entropy": 6.7533129215240475,
      "epoch": 0.2754036087369421,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996517702824163,
      "loss": 6.5965,
      "mean_token_accuracy": 0.12399986758828163,
      "num_tokens": 5401091.0,
      "step": 2755
    },
    {
      "entropy": 6.579279232025146,
      "epoch": 0.2759034337981706,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996497826224425,
      "loss": 6.5962,
      "mean_token_accuracy": 0.12309449911117554,
      "num_tokens": 5409809.0,
      "step": 2760
    },
    {
      "entropy": 6.620116662979126,
      "epoch": 0.2764032588593992,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996477893103379,
      "loss": 6.5615,
      "mean_token_accuracy": 0.12355576753616333,
      "num_tokens": 5420104.0,
      "step": 2765
    },
    {
      "entropy": 6.729361152648925,
      "epoch": 0.2769030839206278,
      "grad_norm": 1.25,
      "learning_rate": 0.0004996457903461527,
      "loss": 6.6167,
      "mean_token_accuracy": 0.1286129795014858,
      "num_tokens": 5427905.0,
      "step": 2770
    },
    {
      "entropy": 6.547765827178955,
      "epoch": 0.2774029089818564,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996437857299373,
      "loss": 6.6143,
      "mean_token_accuracy": 0.12569204494357109,
      "num_tokens": 5438537.0,
      "step": 2775
    },
    {
      "entropy": 6.656113147735596,
      "epoch": 0.2779027340430849,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499641775461742,
      "loss": 6.6938,
      "mean_token_accuracy": 0.11961924210190773,
      "num_tokens": 5447622.0,
      "step": 2780
    },
    {
      "entropy": 6.70698127746582,
      "epoch": 0.2784025591043135,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004996397595416173,
      "loss": 6.5777,
      "mean_token_accuracy": 0.13341286554932594,
      "num_tokens": 5457407.0,
      "step": 2785
    },
    {
      "entropy": 6.622921562194824,
      "epoch": 0.2789023841655421,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996377379696143,
      "loss": 6.5933,
      "mean_token_accuracy": 0.13096230253577232,
      "num_tokens": 5467187.0,
      "step": 2790
    },
    {
      "entropy": 6.642458534240722,
      "epoch": 0.2794022092267706,
      "grad_norm": 1.0,
      "learning_rate": 0.0004996357107457834,
      "loss": 6.6399,
      "mean_token_accuracy": 0.1232068620622158,
      "num_tokens": 5477188.0,
      "step": 2795
    },
    {
      "entropy": 6.670599174499512,
      "epoch": 0.2799020342879992,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499633677870176,
      "loss": 6.6117,
      "mean_token_accuracy": 0.12734065651893617,
      "num_tokens": 5485851.0,
      "step": 2800
    },
    {
      "entropy": 6.606281375885009,
      "epoch": 0.2804018593492278,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996316393428429,
      "loss": 6.6693,
      "mean_token_accuracy": 0.12643833607435226,
      "num_tokens": 5496415.0,
      "step": 2805
    },
    {
      "entropy": 6.692524671554565,
      "epoch": 0.2809016844104563,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004996295951638357,
      "loss": 6.5686,
      "mean_token_accuracy": 0.12833034321665765,
      "num_tokens": 5505538.0,
      "step": 2810
    },
    {
      "entropy": 6.568633985519409,
      "epoch": 0.2814015094716849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996275453332057,
      "loss": 6.6514,
      "mean_token_accuracy": 0.12869037836790084,
      "num_tokens": 5514608.0,
      "step": 2815
    },
    {
      "entropy": 6.656951904296875,
      "epoch": 0.2819013345329135,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996254898510043,
      "loss": 6.6142,
      "mean_token_accuracy": 0.12892503589391707,
      "num_tokens": 5523935.0,
      "step": 2820
    },
    {
      "entropy": 6.599626064300537,
      "epoch": 0.282401159594142,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996234287172835,
      "loss": 6.5469,
      "mean_token_accuracy": 0.13197477608919145,
      "num_tokens": 5534393.0,
      "step": 2825
    },
    {
      "entropy": 6.674867105484009,
      "epoch": 0.2829009846553706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499621361932095,
      "loss": 6.554,
      "mean_token_accuracy": 0.13297315090894699,
      "num_tokens": 5543414.0,
      "step": 2830
    },
    {
      "entropy": 6.550549697875977,
      "epoch": 0.2834008097165992,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996192894954909,
      "loss": 6.5881,
      "mean_token_accuracy": 0.12630681097507476,
      "num_tokens": 5553742.0,
      "step": 2835
    },
    {
      "entropy": 6.552373313903809,
      "epoch": 0.2839006347778278,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996172114075232,
      "loss": 6.4778,
      "mean_token_accuracy": 0.14240361452102662,
      "num_tokens": 5562092.0,
      "step": 2840
    },
    {
      "entropy": 6.673096942901611,
      "epoch": 0.2844004598390563,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996151276682443,
      "loss": 6.7083,
      "mean_token_accuracy": 0.120253437012434,
      "num_tokens": 5571130.0,
      "step": 2845
    },
    {
      "entropy": 6.812530374526977,
      "epoch": 0.2849002849002849,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996130382777065,
      "loss": 6.8625,
      "mean_token_accuracy": 0.11538219377398491,
      "num_tokens": 5581237.0,
      "step": 2850
    },
    {
      "entropy": 6.786452722549439,
      "epoch": 0.2854001099615135,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996109432359625,
      "loss": 6.7088,
      "mean_token_accuracy": 0.11961062774062156,
      "num_tokens": 5592488.0,
      "step": 2855
    },
    {
      "entropy": 6.614932441711426,
      "epoch": 0.285899935022742,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004996088425430651,
      "loss": 6.6408,
      "mean_token_accuracy": 0.12676824256777763,
      "num_tokens": 5602958.0,
      "step": 2860
    },
    {
      "entropy": 6.676570177078247,
      "epoch": 0.2863997600839706,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996067361990669,
      "loss": 6.5944,
      "mean_token_accuracy": 0.12091765776276589,
      "num_tokens": 5613247.0,
      "step": 2865
    },
    {
      "entropy": 6.595865106582641,
      "epoch": 0.2868995851451992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996046242040209,
      "loss": 6.6159,
      "mean_token_accuracy": 0.12946290895342827,
      "num_tokens": 5624102.0,
      "step": 2870
    },
    {
      "entropy": 6.722023248672485,
      "epoch": 0.28739941020642773,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996025065579805,
      "loss": 6.5746,
      "mean_token_accuracy": 0.1296988345682621,
      "num_tokens": 5633147.0,
      "step": 2875
    },
    {
      "entropy": 6.571990823745727,
      "epoch": 0.2878992352676563,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004996003832609988,
      "loss": 6.52,
      "mean_token_accuracy": 0.12874272242188453,
      "num_tokens": 5643399.0,
      "step": 2880
    },
    {
      "entropy": 6.636166620254516,
      "epoch": 0.2883990603288849,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995982543131292,
      "loss": 6.6537,
      "mean_token_accuracy": 0.12803954407572746,
      "num_tokens": 5653129.0,
      "step": 2885
    },
    {
      "entropy": 6.659896039962769,
      "epoch": 0.28889888539011344,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995961197144254,
      "loss": 6.574,
      "mean_token_accuracy": 0.12690625563263894,
      "num_tokens": 5662834.0,
      "step": 2890
    },
    {
      "entropy": 6.62721266746521,
      "epoch": 0.289398710451342,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499593979464941,
      "loss": 6.5245,
      "mean_token_accuracy": 0.1367620535194874,
      "num_tokens": 5672393.0,
      "step": 2895
    },
    {
      "entropy": 6.575651407241821,
      "epoch": 0.2898985355125706,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995918335647298,
      "loss": 6.641,
      "mean_token_accuracy": 0.12922657430171966,
      "num_tokens": 5683189.0,
      "step": 2900
    },
    {
      "entropy": 6.698366451263428,
      "epoch": 0.2903983605737992,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995896820138459,
      "loss": 6.6047,
      "mean_token_accuracy": 0.12848019748926162,
      "num_tokens": 5692999.0,
      "step": 2905
    },
    {
      "entropy": 6.607630872726441,
      "epoch": 0.29089818563502773,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995875248123434,
      "loss": 6.5603,
      "mean_token_accuracy": 0.13163489624857902,
      "num_tokens": 5701707.0,
      "step": 2910
    },
    {
      "entropy": 6.550287914276123,
      "epoch": 0.2913980106962563,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004995853619602766,
      "loss": 6.5145,
      "mean_token_accuracy": 0.13139068633317946,
      "num_tokens": 5711824.0,
      "step": 2915
    },
    {
      "entropy": 6.6153514862060545,
      "epoch": 0.2918978357574849,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995831934576998,
      "loss": 6.6159,
      "mean_token_accuracy": 0.1276139497756958,
      "num_tokens": 5722681.0,
      "step": 2920
    },
    {
      "entropy": 6.689180564880371,
      "epoch": 0.29239766081871343,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995810193046677,
      "loss": 6.6413,
      "mean_token_accuracy": 0.12465909719467164,
      "num_tokens": 5731806.0,
      "step": 2925
    },
    {
      "entropy": 6.621496391296387,
      "epoch": 0.292897485879942,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499578839501235,
      "loss": 6.5787,
      "mean_token_accuracy": 0.13288640938699245,
      "num_tokens": 5742144.0,
      "step": 2930
    },
    {
      "entropy": 6.629134702682495,
      "epoch": 0.2933973109411706,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004995766540474564,
      "loss": 6.5861,
      "mean_token_accuracy": 0.12582591250538827,
      "num_tokens": 5752752.0,
      "step": 2935
    },
    {
      "entropy": 6.645046520233154,
      "epoch": 0.29389713600239914,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499574462943387,
      "loss": 6.5592,
      "mean_token_accuracy": 0.12969356775283813,
      "num_tokens": 5762031.0,
      "step": 2940
    },
    {
      "entropy": 6.555618953704834,
      "epoch": 0.2943969610636277,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499572266189082,
      "loss": 6.4955,
      "mean_token_accuracy": 0.13256025910377503,
      "num_tokens": 5772160.0,
      "step": 2945
    },
    {
      "entropy": 6.672516822814941,
      "epoch": 0.2948967861248563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995700637845964,
      "loss": 6.5956,
      "mean_token_accuracy": 0.1309208579361439,
      "num_tokens": 5782145.0,
      "step": 2950
    },
    {
      "entropy": 6.504296970367432,
      "epoch": 0.29539661118608485,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995678557299859,
      "loss": 6.4564,
      "mean_token_accuracy": 0.13597625270485877,
      "num_tokens": 5790889.0,
      "step": 2955
    },
    {
      "entropy": 6.603075456619263,
      "epoch": 0.29589643624731343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995656420253058,
      "loss": 6.5988,
      "mean_token_accuracy": 0.12828355878591538,
      "num_tokens": 5801239.0,
      "step": 2960
    },
    {
      "entropy": 6.740132427215576,
      "epoch": 0.296396261308542,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499563422670612,
      "loss": 6.7086,
      "mean_token_accuracy": 0.12531245574355127,
      "num_tokens": 5811390.0,
      "step": 2965
    },
    {
      "entropy": 6.689449691772461,
      "epoch": 0.2968960863697706,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004995611976659604,
      "loss": 6.66,
      "mean_token_accuracy": 0.11997519657015801,
      "num_tokens": 5821529.0,
      "step": 2970
    },
    {
      "entropy": 6.641141653060913,
      "epoch": 0.29739591143099914,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995589670114067,
      "loss": 6.5939,
      "mean_token_accuracy": 0.12763414531946182,
      "num_tokens": 5830551.0,
      "step": 2975
    },
    {
      "entropy": 6.6380698680877686,
      "epoch": 0.2978957364922277,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995567307070072,
      "loss": 6.6318,
      "mean_token_accuracy": 0.1293273761868477,
      "num_tokens": 5840768.0,
      "step": 2980
    },
    {
      "entropy": 6.71539797782898,
      "epoch": 0.2983955615534563,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004995544887528182,
      "loss": 6.6436,
      "mean_token_accuracy": 0.12486452683806419,
      "num_tokens": 5850515.0,
      "step": 2985
    },
    {
      "entropy": 6.6234863758087155,
      "epoch": 0.29889538661468484,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995522411488959,
      "loss": 6.5344,
      "mean_token_accuracy": 0.1385308474302292,
      "num_tokens": 5859480.0,
      "step": 2990
    },
    {
      "entropy": 6.690147447586059,
      "epoch": 0.29939521167591343,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995499878952971,
      "loss": 6.606,
      "mean_token_accuracy": 0.12124009132385254,
      "num_tokens": 5871255.0,
      "step": 2995
    },
    {
      "entropy": 6.553613376617432,
      "epoch": 0.299895036737142,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995477289920784,
      "loss": 6.5068,
      "mean_token_accuracy": 0.1348038263618946,
      "num_tokens": 5881306.0,
      "step": 3000
    },
    {
      "epoch": 0.299895036737142,
      "eval_entropy": 6.4320071945790005,
      "eval_loss": 6.631665229797363,
      "eval_mean_token_accuracy": 0.13145681092632214,
      "eval_num_tokens": 5881306.0,
      "eval_runtime": 23.9243,
      "eval_samples_per_second": 1297.635,
      "eval_steps_per_second": 162.22,
      "step": 3000
    },
    {
      "entropy": 6.600755977630615,
      "epoch": 0.30039486179837055,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995454644392967,
      "loss": 6.601,
      "mean_token_accuracy": 0.12665048539638518,
      "num_tokens": 5891209.0,
      "step": 3005
    },
    {
      "entropy": 6.689290380477905,
      "epoch": 0.30089468685959914,
      "grad_norm": 0.90625,
      "learning_rate": 0.0004995431942370087,
      "loss": 6.6651,
      "mean_token_accuracy": 0.125472728908062,
      "num_tokens": 5902003.0,
      "step": 3010
    },
    {
      "entropy": 6.66430549621582,
      "epoch": 0.3013945119208277,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995409183852718,
      "loss": 6.5784,
      "mean_token_accuracy": 0.12904052063822746,
      "num_tokens": 5911327.0,
      "step": 3015
    },
    {
      "entropy": 6.522570371627808,
      "epoch": 0.30189433698205626,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004995386368841432,
      "loss": 6.5223,
      "mean_token_accuracy": 0.13217691257596015,
      "num_tokens": 5920385.0,
      "step": 3020
    },
    {
      "entropy": 6.5316784381866455,
      "epoch": 0.30239416204328484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995363497336802,
      "loss": 6.484,
      "mean_token_accuracy": 0.1375424049794674,
      "num_tokens": 5929784.0,
      "step": 3025
    },
    {
      "entropy": 6.620974397659301,
      "epoch": 0.30289398710451343,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004995340569339405,
      "loss": 6.5671,
      "mean_token_accuracy": 0.1293088547885418,
      "num_tokens": 5938766.0,
      "step": 3030
    },
    {
      "entropy": 6.724143362045288,
      "epoch": 0.303393812165742,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995317584849817,
      "loss": 6.6737,
      "mean_token_accuracy": 0.12524416744709016,
      "num_tokens": 5948869.0,
      "step": 3035
    },
    {
      "entropy": 6.5645770072937015,
      "epoch": 0.30389363722697055,
      "grad_norm": 2.375,
      "learning_rate": 0.0004995294543868615,
      "loss": 6.4924,
      "mean_token_accuracy": 0.12778667509555816,
      "num_tokens": 5959074.0,
      "step": 3040
    },
    {
      "entropy": 6.522882747650146,
      "epoch": 0.30439346228819913,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995271446396382,
      "loss": 6.6146,
      "mean_token_accuracy": 0.1305883191525936,
      "num_tokens": 5968608.0,
      "step": 3045
    },
    {
      "entropy": 6.716164255142212,
      "epoch": 0.3048932873494277,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995248292433697,
      "loss": 6.635,
      "mean_token_accuracy": 0.12156384512782097,
      "num_tokens": 5979756.0,
      "step": 3050
    },
    {
      "entropy": 6.714064407348633,
      "epoch": 0.30539311241065625,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004995225081981142,
      "loss": 6.6951,
      "mean_token_accuracy": 0.11126383915543556,
      "num_tokens": 5990299.0,
      "step": 3055
    },
    {
      "entropy": 6.651313638687133,
      "epoch": 0.30589293747188484,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995201815039301,
      "loss": 6.6101,
      "mean_token_accuracy": 0.11899112910032272,
      "num_tokens": 6001721.0,
      "step": 3060
    },
    {
      "entropy": 6.571802711486816,
      "epoch": 0.3063927625331134,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995178491608762,
      "loss": 6.6714,
      "mean_token_accuracy": 0.1208619199693203,
      "num_tokens": 6012658.0,
      "step": 3065
    },
    {
      "entropy": 6.710856342315674,
      "epoch": 0.30689258759434196,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995155111690107,
      "loss": 6.6137,
      "mean_token_accuracy": 0.13041311129927635,
      "num_tokens": 6022470.0,
      "step": 3070
    },
    {
      "entropy": 6.576000165939331,
      "epoch": 0.30739241265557055,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000499513167528393,
      "loss": 6.4868,
      "mean_token_accuracy": 0.12843060344457627,
      "num_tokens": 6034227.0,
      "step": 3075
    },
    {
      "entropy": 6.577868127822876,
      "epoch": 0.30789223771679913,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004995108182390815,
      "loss": 6.5022,
      "mean_token_accuracy": 0.13164846673607827,
      "num_tokens": 6043122.0,
      "step": 3080
    },
    {
      "entropy": 6.622898721694947,
      "epoch": 0.30839206277802766,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995084633011359,
      "loss": 6.5123,
      "mean_token_accuracy": 0.12831469029188156,
      "num_tokens": 6052988.0,
      "step": 3085
    },
    {
      "entropy": 6.551307535171508,
      "epoch": 0.30889188783925625,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499506102714615,
      "loss": 6.5674,
      "mean_token_accuracy": 0.13058611005544662,
      "num_tokens": 6062800.0,
      "step": 3090
    },
    {
      "entropy": 6.713084983825683,
      "epoch": 0.30939171290048484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995037364795782,
      "loss": 6.6444,
      "mean_token_accuracy": 0.12327775061130523,
      "num_tokens": 6071327.0,
      "step": 3095
    },
    {
      "entropy": 6.651969480514526,
      "epoch": 0.3098915379617134,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004995013645960853,
      "loss": 6.6597,
      "mean_token_accuracy": 0.12309322506189346,
      "num_tokens": 6080574.0,
      "step": 3100
    },
    {
      "entropy": 6.622641134262085,
      "epoch": 0.31039136302294196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004994989870641958,
      "loss": 6.5323,
      "mean_token_accuracy": 0.13187582194805145,
      "num_tokens": 6088081.0,
      "step": 3105
    },
    {
      "entropy": 6.6546022415161135,
      "epoch": 0.31089118808417054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994966038839695,
      "loss": 6.6591,
      "mean_token_accuracy": 0.12742284461855888,
      "num_tokens": 6097785.0,
      "step": 3110
    },
    {
      "entropy": 6.614014625549316,
      "epoch": 0.31139101314539913,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994942150554666,
      "loss": 6.578,
      "mean_token_accuracy": 0.12638851180672644,
      "num_tokens": 6107402.0,
      "step": 3115
    },
    {
      "entropy": 6.604445695877075,
      "epoch": 0.31189083820662766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994918205787468,
      "loss": 6.5374,
      "mean_token_accuracy": 0.13036633431911468,
      "num_tokens": 6116439.0,
      "step": 3120
    },
    {
      "entropy": 6.566957235336304,
      "epoch": 0.31239066326785625,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994894204538706,
      "loss": 6.59,
      "mean_token_accuracy": 0.13152979761362077,
      "num_tokens": 6126854.0,
      "step": 3125
    },
    {
      "entropy": 6.613524532318115,
      "epoch": 0.31289048832908484,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994870146808984,
      "loss": 6.5429,
      "mean_token_accuracy": 0.13135527297854424,
      "num_tokens": 6136869.0,
      "step": 3130
    },
    {
      "entropy": 6.527092504501343,
      "epoch": 0.31339031339031337,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994846032598906,
      "loss": 6.5181,
      "mean_token_accuracy": 0.13491623252630233,
      "num_tokens": 6146772.0,
      "step": 3135
    },
    {
      "entropy": 6.628911352157592,
      "epoch": 0.31389013845154196,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994821861909082,
      "loss": 6.6088,
      "mean_token_accuracy": 0.1273271232843399,
      "num_tokens": 6156329.0,
      "step": 3140
    },
    {
      "entropy": 6.578382349014282,
      "epoch": 0.31438996351277054,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004994797634740114,
      "loss": 6.5051,
      "mean_token_accuracy": 0.13173323944211007,
      "num_tokens": 6166198.0,
      "step": 3145
    },
    {
      "entropy": 6.513795280456543,
      "epoch": 0.3148897885739991,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994773351092618,
      "loss": 6.5726,
      "mean_token_accuracy": 0.12777804657816888,
      "num_tokens": 6176180.0,
      "step": 3150
    },
    {
      "entropy": 6.634206628799438,
      "epoch": 0.31538961363522766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00049947490109672,
      "loss": 6.5288,
      "mean_token_accuracy": 0.13039560467004777,
      "num_tokens": 6185776.0,
      "step": 3155
    },
    {
      "entropy": 6.6662657260894775,
      "epoch": 0.31588943869645625,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994724614364476,
      "loss": 6.6597,
      "mean_token_accuracy": 0.1265087068080902,
      "num_tokens": 6196472.0,
      "step": 3160
    },
    {
      "entropy": 6.534713268280029,
      "epoch": 0.31638926375768484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994700161285058,
      "loss": 6.5747,
      "mean_token_accuracy": 0.12786666974425315,
      "num_tokens": 6206064.0,
      "step": 3165
    },
    {
      "entropy": 6.654185581207275,
      "epoch": 0.31688908881891337,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499467565172956,
      "loss": 6.5699,
      "mean_token_accuracy": 0.1291349373757839,
      "num_tokens": 6217044.0,
      "step": 3170
    },
    {
      "entropy": 6.609409284591675,
      "epoch": 0.31738891388014195,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004994651085698602,
      "loss": 6.5483,
      "mean_token_accuracy": 0.12804570198059081,
      "num_tokens": 6228294.0,
      "step": 3175
    },
    {
      "entropy": 6.578659629821777,
      "epoch": 0.31788873894137054,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049946264631928,
      "loss": 6.4947,
      "mean_token_accuracy": 0.13685906305909157,
      "num_tokens": 6238525.0,
      "step": 3180
    },
    {
      "entropy": 6.445104360580444,
      "epoch": 0.3183885640025991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994601784212774,
      "loss": 6.3748,
      "mean_token_accuracy": 0.14615772068500518,
      "num_tokens": 6247457.0,
      "step": 3185
    },
    {
      "entropy": 6.459098958969117,
      "epoch": 0.31888838906382766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994577048759144,
      "loss": 6.502,
      "mean_token_accuracy": 0.13235406056046486,
      "num_tokens": 6256414.0,
      "step": 3190
    },
    {
      "entropy": 6.555423212051392,
      "epoch": 0.31938821412505625,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994552256832533,
      "loss": 6.4339,
      "mean_token_accuracy": 0.13570524603128434,
      "num_tokens": 6266093.0,
      "step": 3195
    },
    {
      "entropy": 6.560033226013184,
      "epoch": 0.3198880391862848,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994527408433566,
      "loss": 6.4966,
      "mean_token_accuracy": 0.1354018658399582,
      "num_tokens": 6276340.0,
      "step": 3200
    },
    {
      "entropy": 6.495316362380981,
      "epoch": 0.32038786424751337,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994502503562865,
      "loss": 6.5232,
      "mean_token_accuracy": 0.13102975934743882,
      "num_tokens": 6285044.0,
      "step": 3205
    },
    {
      "entropy": 6.590799283981323,
      "epoch": 0.32088768930874195,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000499447754222106,
      "loss": 6.4512,
      "mean_token_accuracy": 0.13481636717915535,
      "num_tokens": 6294105.0,
      "step": 3210
    },
    {
      "entropy": 6.506353807449341,
      "epoch": 0.3213875143699705,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004994452524408777,
      "loss": 6.5033,
      "mean_token_accuracy": 0.129614145308733,
      "num_tokens": 6302503.0,
      "step": 3215
    },
    {
      "entropy": 6.593749284744263,
      "epoch": 0.32188733943119907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994427450126646,
      "loss": 6.5276,
      "mean_token_accuracy": 0.1262624278664589,
      "num_tokens": 6313068.0,
      "step": 3220
    },
    {
      "entropy": 6.668876218795776,
      "epoch": 0.32238716449242766,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004994402319375298,
      "loss": 6.6003,
      "mean_token_accuracy": 0.12393388748168946,
      "num_tokens": 6322770.0,
      "step": 3225
    },
    {
      "entropy": 6.52561993598938,
      "epoch": 0.32288698955365625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994377132155366,
      "loss": 6.4982,
      "mean_token_accuracy": 0.1299036055803299,
      "num_tokens": 6333055.0,
      "step": 3230
    },
    {
      "entropy": 6.4240704536437985,
      "epoch": 0.3233868146148848,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994351888467482,
      "loss": 6.5018,
      "mean_token_accuracy": 0.13404592424631118,
      "num_tokens": 6342780.0,
      "step": 3235
    },
    {
      "entropy": 6.614720582962036,
      "epoch": 0.32388663967611336,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004994326588312282,
      "loss": 6.4761,
      "mean_token_accuracy": 0.12766311541199685,
      "num_tokens": 6352021.0,
      "step": 3240
    },
    {
      "entropy": 6.545908498764038,
      "epoch": 0.32438646473734195,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004994301231690403,
      "loss": 6.5556,
      "mean_token_accuracy": 0.1301438555121422,
      "num_tokens": 6361062.0,
      "step": 3245
    },
    {
      "entropy": 6.564257335662842,
      "epoch": 0.3248862897985705,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994275818602482,
      "loss": 6.5841,
      "mean_token_accuracy": 0.12273027524352073,
      "num_tokens": 6371318.0,
      "step": 3250
    },
    {
      "entropy": 6.696441221237182,
      "epoch": 0.32538611485979907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994250349049159,
      "loss": 6.647,
      "mean_token_accuracy": 0.1229313313961029,
      "num_tokens": 6381078.0,
      "step": 3255
    },
    {
      "entropy": 6.655215072631836,
      "epoch": 0.32588593992102766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994224823031075,
      "loss": 6.4934,
      "mean_token_accuracy": 0.1329373300075531,
      "num_tokens": 6390382.0,
      "step": 3260
    },
    {
      "entropy": 6.507878828048706,
      "epoch": 0.3263857649822562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994199240548873,
      "loss": 6.5372,
      "mean_token_accuracy": 0.12851825654506682,
      "num_tokens": 6399849.0,
      "step": 3265
    },
    {
      "entropy": 6.581459188461304,
      "epoch": 0.3268855900434848,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004994173601603194,
      "loss": 6.5513,
      "mean_token_accuracy": 0.12433971241116523,
      "num_tokens": 6409150.0,
      "step": 3270
    },
    {
      "entropy": 6.622395944595337,
      "epoch": 0.32738541510471336,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994147906194684,
      "loss": 6.5666,
      "mean_token_accuracy": 0.1260290712118149,
      "num_tokens": 6419525.0,
      "step": 3275
    },
    {
      "entropy": 6.5685341358184814,
      "epoch": 0.3278852401659419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004994122154323992,
      "loss": 6.5131,
      "mean_token_accuracy": 0.12771548554301262,
      "num_tokens": 6430185.0,
      "step": 3280
    },
    {
      "entropy": 6.611548995971679,
      "epoch": 0.3283850652271705,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004994096345991763,
      "loss": 6.4981,
      "mean_token_accuracy": 0.13285920023918152,
      "num_tokens": 6440221.0,
      "step": 3285
    },
    {
      "entropy": 6.5308490753173825,
      "epoch": 0.32888489028839907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994070481198649,
      "loss": 6.5828,
      "mean_token_accuracy": 0.129660727083683,
      "num_tokens": 6448861.0,
      "step": 3290
    },
    {
      "entropy": 6.601012754440307,
      "epoch": 0.32938471534962765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994044559945296,
      "loss": 6.4577,
      "mean_token_accuracy": 0.13085916712880136,
      "num_tokens": 6458782.0,
      "step": 3295
    },
    {
      "entropy": 6.519365930557251,
      "epoch": 0.3298845404108562,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004994018582232361,
      "loss": 6.5129,
      "mean_token_accuracy": 0.12531593590974807,
      "num_tokens": 6470201.0,
      "step": 3300
    },
    {
      "entropy": 6.468400764465332,
      "epoch": 0.3303843654720848,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993992548060497,
      "loss": 6.5465,
      "mean_token_accuracy": 0.12943520471453668,
      "num_tokens": 6479507.0,
      "step": 3305
    },
    {
      "entropy": 6.613047409057617,
      "epoch": 0.33088419053331336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993966457430357,
      "loss": 6.53,
      "mean_token_accuracy": 0.1323752962052822,
      "num_tokens": 6488902.0,
      "step": 3310
    },
    {
      "entropy": 6.6111753463745115,
      "epoch": 0.3313840155945419,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993940310342598,
      "loss": 6.5284,
      "mean_token_accuracy": 0.1341463506221771,
      "num_tokens": 6498781.0,
      "step": 3315
    },
    {
      "entropy": 6.5431684970855715,
      "epoch": 0.3318838406557705,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993914106797879,
      "loss": 6.4361,
      "mean_token_accuracy": 0.1319537065923214,
      "num_tokens": 6507358.0,
      "step": 3320
    },
    {
      "entropy": 6.509786367416382,
      "epoch": 0.33238366571699907,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993887846796858,
      "loss": 6.4957,
      "mean_token_accuracy": 0.13879070654511452,
      "num_tokens": 6517976.0,
      "step": 3325
    },
    {
      "entropy": 6.6177679061889645,
      "epoch": 0.3328834907782276,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993861530340196,
      "loss": 6.6143,
      "mean_token_accuracy": 0.12213909402489662,
      "num_tokens": 6528407.0,
      "step": 3330
    },
    {
      "entropy": 6.737038612365723,
      "epoch": 0.3333833158394562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993835157428555,
      "loss": 6.5441,
      "mean_token_accuracy": 0.12688530683517457,
      "num_tokens": 6537619.0,
      "step": 3335
    },
    {
      "entropy": 6.442472696304321,
      "epoch": 0.33388314090068477,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004993808728062599,
      "loss": 6.4266,
      "mean_token_accuracy": 0.1311258815228939,
      "num_tokens": 6548393.0,
      "step": 3340
    },
    {
      "entropy": 6.520314359664917,
      "epoch": 0.3343829659619133,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993782242242994,
      "loss": 6.5057,
      "mean_token_accuracy": 0.1320674754679203,
      "num_tokens": 6558646.0,
      "step": 3345
    },
    {
      "entropy": 6.669322967529297,
      "epoch": 0.3348827910231419,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993755699970404,
      "loss": 6.5193,
      "mean_token_accuracy": 0.13431787863373756,
      "num_tokens": 6567534.0,
      "step": 3350
    },
    {
      "entropy": 6.54373140335083,
      "epoch": 0.3353826160843705,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004993729101245498,
      "loss": 6.5117,
      "mean_token_accuracy": 0.135849292576313,
      "num_tokens": 6577940.0,
      "step": 3355
    },
    {
      "entropy": 6.492143154144287,
      "epoch": 0.33588244114559906,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993702446068945,
      "loss": 6.5437,
      "mean_token_accuracy": 0.13114413619041443,
      "num_tokens": 6587421.0,
      "step": 3360
    },
    {
      "entropy": 6.591862964630127,
      "epoch": 0.3363822662068276,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993675734441416,
      "loss": 6.5199,
      "mean_token_accuracy": 0.12906850650906562,
      "num_tokens": 6597509.0,
      "step": 3365
    },
    {
      "entropy": 6.523139381408692,
      "epoch": 0.3368820912680562,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993648966363583,
      "loss": 6.4883,
      "mean_token_accuracy": 0.136234875023365,
      "num_tokens": 6607003.0,
      "step": 3370
    },
    {
      "entropy": 6.430811643600464,
      "epoch": 0.33738191632928477,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499362214183612,
      "loss": 6.5186,
      "mean_token_accuracy": 0.1348765529692173,
      "num_tokens": 6617454.0,
      "step": 3375
    },
    {
      "entropy": 6.555140542984009,
      "epoch": 0.3378817413905133,
      "grad_norm": 1.125,
      "learning_rate": 0.00049935952608597,
      "loss": 6.4927,
      "mean_token_accuracy": 0.1368827536702156,
      "num_tokens": 6628327.0,
      "step": 3380
    },
    {
      "entropy": 6.530352210998535,
      "epoch": 0.3383815664517419,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993568323435002,
      "loss": 6.4494,
      "mean_token_accuracy": 0.13094991967082023,
      "num_tokens": 6638282.0,
      "step": 3385
    },
    {
      "entropy": 6.468064785003662,
      "epoch": 0.3388813915129705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993541329562702,
      "loss": 6.4534,
      "mean_token_accuracy": 0.13519542589783667,
      "num_tokens": 6647138.0,
      "step": 3390
    },
    {
      "entropy": 6.578132963180542,
      "epoch": 0.339381216574199,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993514279243477,
      "loss": 6.5358,
      "mean_token_accuracy": 0.12947731614112853,
      "num_tokens": 6656129.0,
      "step": 3395
    },
    {
      "entropy": 6.566795492172242,
      "epoch": 0.3398810416354276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993487172478012,
      "loss": 6.5149,
      "mean_token_accuracy": 0.13720047250390052,
      "num_tokens": 6666485.0,
      "step": 3400
    },
    {
      "entropy": 6.613122987747192,
      "epoch": 0.3403808666966562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993460009266988,
      "loss": 6.6103,
      "mean_token_accuracy": 0.12625092417001724,
      "num_tokens": 6676885.0,
      "step": 3405
    },
    {
      "entropy": 6.563211297988891,
      "epoch": 0.3408806917578847,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993432789611086,
      "loss": 6.4762,
      "mean_token_accuracy": 0.1333943009376526,
      "num_tokens": 6687309.0,
      "step": 3410
    },
    {
      "entropy": 6.560053968429566,
      "epoch": 0.3413805168191133,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004993405513510992,
      "loss": 6.6194,
      "mean_token_accuracy": 0.13058197498321533,
      "num_tokens": 6696868.0,
      "step": 3415
    },
    {
      "entropy": 6.668599033355713,
      "epoch": 0.3418803418803419,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993378180967394,
      "loss": 6.5436,
      "mean_token_accuracy": 0.13377416431903838,
      "num_tokens": 6706526.0,
      "step": 3420
    },
    {
      "entropy": 6.586946868896485,
      "epoch": 0.3423801669415705,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993350791980978,
      "loss": 6.5802,
      "mean_token_accuracy": 0.1323694072663784,
      "num_tokens": 6716792.0,
      "step": 3425
    },
    {
      "entropy": 6.4420288562774655,
      "epoch": 0.342879992002799,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004993323346552433,
      "loss": 6.4162,
      "mean_token_accuracy": 0.13632412850856782,
      "num_tokens": 6726268.0,
      "step": 3430
    },
    {
      "entropy": 6.591781568527222,
      "epoch": 0.3433798170640276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993295844682449,
      "loss": 6.4962,
      "mean_token_accuracy": 0.13022203966975213,
      "num_tokens": 6735687.0,
      "step": 3435
    },
    {
      "entropy": 6.475664806365967,
      "epoch": 0.3438796421252562,
      "grad_norm": 0.97265625,
      "learning_rate": 0.000499326828637172,
      "loss": 6.3522,
      "mean_token_accuracy": 0.1417197085916996,
      "num_tokens": 6747055.0,
      "step": 3440
    },
    {
      "entropy": 6.402159023284912,
      "epoch": 0.3443794671864847,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993240671620938,
      "loss": 6.454,
      "mean_token_accuracy": 0.1360941469669342,
      "num_tokens": 6756190.0,
      "step": 3445
    },
    {
      "entropy": 6.599747228622436,
      "epoch": 0.3448792922477133,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993213000430797,
      "loss": 6.5588,
      "mean_token_accuracy": 0.12697630673646926,
      "num_tokens": 6766145.0,
      "step": 3450
    },
    {
      "entropy": 6.570261383056641,
      "epoch": 0.3453791173089419,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993185272801995,
      "loss": 6.4775,
      "mean_token_accuracy": 0.1353766769170761,
      "num_tokens": 6776294.0,
      "step": 3455
    },
    {
      "entropy": 6.516684436798096,
      "epoch": 0.3458789423701704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993157488735228,
      "loss": 6.5099,
      "mean_token_accuracy": 0.1365628369152546,
      "num_tokens": 6785349.0,
      "step": 3460
    },
    {
      "entropy": 6.545265340805054,
      "epoch": 0.346378767431399,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993129648231197,
      "loss": 6.3977,
      "mean_token_accuracy": 0.1309737704694271,
      "num_tokens": 6794733.0,
      "step": 3465
    },
    {
      "entropy": 6.581028318405151,
      "epoch": 0.3468785924926276,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049931017512906,
      "loss": 6.5085,
      "mean_token_accuracy": 0.12938275039196015,
      "num_tokens": 6803850.0,
      "step": 3470
    },
    {
      "entropy": 6.474708223342896,
      "epoch": 0.3473784175538562,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993073797914141,
      "loss": 6.4258,
      "mean_token_accuracy": 0.1346999816596508,
      "num_tokens": 6813201.0,
      "step": 3475
    },
    {
      "entropy": 6.515172481536865,
      "epoch": 0.3478782426150847,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993045788102522,
      "loss": 6.399,
      "mean_token_accuracy": 0.13415363654494286,
      "num_tokens": 6822238.0,
      "step": 3480
    },
    {
      "entropy": 6.577175855636597,
      "epoch": 0.3483780676763133,
      "grad_norm": 1.0,
      "learning_rate": 0.0004993017721856449,
      "loss": 6.525,
      "mean_token_accuracy": 0.12922550290822982,
      "num_tokens": 6832586.0,
      "step": 3485
    },
    {
      "entropy": 6.504951906204224,
      "epoch": 0.3488778927375419,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992989599176626,
      "loss": 6.5085,
      "mean_token_accuracy": 0.12983549758791924,
      "num_tokens": 6841699.0,
      "step": 3490
    },
    {
      "entropy": 6.5889708518981935,
      "epoch": 0.3493777177987704,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992961420063763,
      "loss": 6.5467,
      "mean_token_accuracy": 0.13208589106798171,
      "num_tokens": 6851455.0,
      "step": 3495
    },
    {
      "entropy": 6.55912971496582,
      "epoch": 0.349877542859999,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992933184518567,
      "loss": 6.4683,
      "mean_token_accuracy": 0.13468205258250238,
      "num_tokens": 6860646.0,
      "step": 3500
    },
    {
      "entropy": 6.477167224884033,
      "epoch": 0.3503773679212276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992904892541749,
      "loss": 6.5506,
      "mean_token_accuracy": 0.12879718095064163,
      "num_tokens": 6870641.0,
      "step": 3505
    },
    {
      "entropy": 6.599151563644409,
      "epoch": 0.3508771929824561,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004992876544134022,
      "loss": 6.565,
      "mean_token_accuracy": 0.12880437597632408,
      "num_tokens": 6880831.0,
      "step": 3510
    },
    {
      "entropy": 6.6267951965332035,
      "epoch": 0.3513770180436847,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004992848139296098,
      "loss": 6.5652,
      "mean_token_accuracy": 0.13073665574193,
      "num_tokens": 6890305.0,
      "step": 3515
    },
    {
      "entropy": 6.5009400844573975,
      "epoch": 0.3518768431049133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992819678028692,
      "loss": 6.4436,
      "mean_token_accuracy": 0.13384637907147406,
      "num_tokens": 6900147.0,
      "step": 3520
    },
    {
      "entropy": 6.5349523544311525,
      "epoch": 0.3523766681661418,
      "grad_norm": 1.171875,
      "learning_rate": 0.000499279116033252,
      "loss": 6.4357,
      "mean_token_accuracy": 0.13797437772154808,
      "num_tokens": 6910367.0,
      "step": 3525
    },
    {
      "entropy": 6.488935756683349,
      "epoch": 0.3528764932273704,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992762586208299,
      "loss": 6.418,
      "mean_token_accuracy": 0.13764555454254152,
      "num_tokens": 6919843.0,
      "step": 3530
    },
    {
      "entropy": 6.492925548553467,
      "epoch": 0.353376318288599,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992733955656749,
      "loss": 6.4652,
      "mean_token_accuracy": 0.13717108070850373,
      "num_tokens": 6929912.0,
      "step": 3535
    },
    {
      "entropy": 6.5468635082244875,
      "epoch": 0.3538761433498276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992705268678589,
      "loss": 6.5454,
      "mean_token_accuracy": 0.12351348623633385,
      "num_tokens": 6940836.0,
      "step": 3540
    },
    {
      "entropy": 6.559476566314697,
      "epoch": 0.3543759684110561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992676525274542,
      "loss": 6.5059,
      "mean_token_accuracy": 0.1319356568157673,
      "num_tokens": 6950313.0,
      "step": 3545
    },
    {
      "entropy": 6.533931303024292,
      "epoch": 0.3548757934722847,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992647725445331,
      "loss": 6.3724,
      "mean_token_accuracy": 0.13631908148527144,
      "num_tokens": 6958786.0,
      "step": 3550
    },
    {
      "entropy": 6.516522645950317,
      "epoch": 0.3553756185335133,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499261886919168,
      "loss": 6.5449,
      "mean_token_accuracy": 0.12808100506663322,
      "num_tokens": 6968453.0,
      "step": 3555
    },
    {
      "entropy": 6.525800895690918,
      "epoch": 0.3558754435947418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992589956514316,
      "loss": 6.3911,
      "mean_token_accuracy": 0.1394435554742813,
      "num_tokens": 6978143.0,
      "step": 3560
    },
    {
      "entropy": 6.437818622589111,
      "epoch": 0.3563752686559704,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004992560987413965,
      "loss": 6.4029,
      "mean_token_accuracy": 0.13268027678132058,
      "num_tokens": 6986711.0,
      "step": 3565
    },
    {
      "entropy": 6.424841594696045,
      "epoch": 0.356875093717199,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992531961891356,
      "loss": 6.4592,
      "mean_token_accuracy": 0.13201552405953407,
      "num_tokens": 6995904.0,
      "step": 3570
    },
    {
      "entropy": 6.542066812515259,
      "epoch": 0.35737491877842753,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992502879947222,
      "loss": 6.4104,
      "mean_token_accuracy": 0.13974365442991257,
      "num_tokens": 7005479.0,
      "step": 3575
    },
    {
      "entropy": 6.3777965068817135,
      "epoch": 0.3578747438396561,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992473741582291,
      "loss": 6.3169,
      "mean_token_accuracy": 0.14328965321183204,
      "num_tokens": 7015048.0,
      "step": 3580
    },
    {
      "entropy": 6.484763860702515,
      "epoch": 0.3583745689008847,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992444546797298,
      "loss": 6.5271,
      "mean_token_accuracy": 0.13071638196706772,
      "num_tokens": 7024738.0,
      "step": 3585
    },
    {
      "entropy": 6.54697551727295,
      "epoch": 0.35887439396211324,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992415295592977,
      "loss": 6.4431,
      "mean_token_accuracy": 0.1346895471215248,
      "num_tokens": 7033972.0,
      "step": 3590
    },
    {
      "entropy": 6.561047172546386,
      "epoch": 0.3593742190233418,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992385987970064,
      "loss": 6.4727,
      "mean_token_accuracy": 0.12772433459758759,
      "num_tokens": 7043713.0,
      "step": 3595
    },
    {
      "entropy": 6.533878374099731,
      "epoch": 0.3598740440845704,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992356623929297,
      "loss": 6.4653,
      "mean_token_accuracy": 0.13502703830599785,
      "num_tokens": 7053276.0,
      "step": 3600
    },
    {
      "entropy": 6.471671056747437,
      "epoch": 0.360373869145799,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992327203471414,
      "loss": 6.4704,
      "mean_token_accuracy": 0.13309010937809945,
      "num_tokens": 7063610.0,
      "step": 3605
    },
    {
      "entropy": 6.480327892303467,
      "epoch": 0.36087369420702753,
      "grad_norm": 1.0,
      "learning_rate": 0.0004992297726597156,
      "loss": 6.4317,
      "mean_token_accuracy": 0.13454311117529869,
      "num_tokens": 7073497.0,
      "step": 3610
    },
    {
      "entropy": 6.5450966358184814,
      "epoch": 0.3613735192682561,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992268193307264,
      "loss": 6.512,
      "mean_token_accuracy": 0.12889237701892853,
      "num_tokens": 7084387.0,
      "step": 3615
    },
    {
      "entropy": 6.464864444732666,
      "epoch": 0.3618733443294847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992238603602481,
      "loss": 6.3634,
      "mean_token_accuracy": 0.1431560583412647,
      "num_tokens": 7093774.0,
      "step": 3620
    },
    {
      "entropy": 6.544749546051025,
      "epoch": 0.36237316939071323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992208957483552,
      "loss": 6.5333,
      "mean_token_accuracy": 0.12903482094407082,
      "num_tokens": 7103367.0,
      "step": 3625
    },
    {
      "entropy": 6.504576063156128,
      "epoch": 0.3628729944519418,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004992179254951223,
      "loss": 6.4445,
      "mean_token_accuracy": 0.13610872253775597,
      "num_tokens": 7113261.0,
      "step": 3630
    },
    {
      "entropy": 6.5761260986328125,
      "epoch": 0.3633728195131704,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499214949600624,
      "loss": 6.4394,
      "mean_token_accuracy": 0.13936359509825708,
      "num_tokens": 7122775.0,
      "step": 3635
    },
    {
      "entropy": 6.465544176101685,
      "epoch": 0.36387264457439894,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004992119680649355,
      "loss": 6.498,
      "mean_token_accuracy": 0.13492224588990212,
      "num_tokens": 7133369.0,
      "step": 3640
    },
    {
      "entropy": 6.507680988311767,
      "epoch": 0.3643724696356275,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992089808881313,
      "loss": 6.371,
      "mean_token_accuracy": 0.13739476948976517,
      "num_tokens": 7142589.0,
      "step": 3645
    },
    {
      "entropy": 6.572415542602539,
      "epoch": 0.3648722946968561,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004992059880702869,
      "loss": 6.5535,
      "mean_token_accuracy": 0.12939954176545143,
      "num_tokens": 7152260.0,
      "step": 3650
    },
    {
      "entropy": 6.454878807067871,
      "epoch": 0.36537211975808465,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004992029896114775,
      "loss": 6.4848,
      "mean_token_accuracy": 0.13722681552171706,
      "num_tokens": 7161297.0,
      "step": 3655
    },
    {
      "entropy": 6.499073028564453,
      "epoch": 0.36587194481931323,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991999855117787,
      "loss": 6.5303,
      "mean_token_accuracy": 0.1349036253988743,
      "num_tokens": 7172835.0,
      "step": 3660
    },
    {
      "entropy": 6.494237041473388,
      "epoch": 0.3663717698805418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991969757712659,
      "loss": 6.4488,
      "mean_token_accuracy": 0.14180464968085288,
      "num_tokens": 7182647.0,
      "step": 3665
    },
    {
      "entropy": 6.479952478408814,
      "epoch": 0.3668715949417704,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004991939603900149,
      "loss": 6.5053,
      "mean_token_accuracy": 0.13072704821825026,
      "num_tokens": 7192048.0,
      "step": 3670
    },
    {
      "entropy": 6.490058422088623,
      "epoch": 0.36737142000299894,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991909393681015,
      "loss": 6.4041,
      "mean_token_accuracy": 0.13920209407806397,
      "num_tokens": 7201747.0,
      "step": 3675
    },
    {
      "entropy": 6.5024505138397215,
      "epoch": 0.3678712450642275,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991879127056017,
      "loss": 6.3565,
      "mean_token_accuracy": 0.13667716830968857,
      "num_tokens": 7210660.0,
      "step": 3680
    },
    {
      "entropy": 6.417922496795654,
      "epoch": 0.3683710701254561,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991848804025918,
      "loss": 6.4273,
      "mean_token_accuracy": 0.13781158849596978,
      "num_tokens": 7221474.0,
      "step": 3685
    },
    {
      "entropy": 6.467645978927612,
      "epoch": 0.36887089518668464,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499181842459148,
      "loss": 6.3858,
      "mean_token_accuracy": 0.13766940161585808,
      "num_tokens": 7231700.0,
      "step": 3690
    },
    {
      "entropy": 6.4335743427276615,
      "epoch": 0.36937072024791323,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004991787988753467,
      "loss": 6.4284,
      "mean_token_accuracy": 0.14367831647396087,
      "num_tokens": 7241887.0,
      "step": 3695
    },
    {
      "entropy": 6.565946340560913,
      "epoch": 0.3698705453091418,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991757496512645,
      "loss": 6.4796,
      "mean_token_accuracy": 0.13281952366232871,
      "num_tokens": 7251399.0,
      "step": 3700
    },
    {
      "entropy": 6.564613485336304,
      "epoch": 0.37037037037037035,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004991726947869782,
      "loss": 6.555,
      "mean_token_accuracy": 0.12984710559248924,
      "num_tokens": 7260555.0,
      "step": 3705
    },
    {
      "entropy": 6.477829217910767,
      "epoch": 0.37087019543159894,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991696342825645,
      "loss": 6.4365,
      "mean_token_accuracy": 0.14144129306077957,
      "num_tokens": 7270706.0,
      "step": 3710
    },
    {
      "entropy": 6.543673896789551,
      "epoch": 0.3713700204928275,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991665681381005,
      "loss": 6.4392,
      "mean_token_accuracy": 0.1305554449558258,
      "num_tokens": 7282491.0,
      "step": 3715
    },
    {
      "entropy": 6.4118969440460205,
      "epoch": 0.37186984555405606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991634963536633,
      "loss": 6.416,
      "mean_token_accuracy": 0.13627416118979455,
      "num_tokens": 7292620.0,
      "step": 3720
    },
    {
      "entropy": 6.503028106689453,
      "epoch": 0.37236967061528464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991604189293303,
      "loss": 6.4917,
      "mean_token_accuracy": 0.13409775272011756,
      "num_tokens": 7302383.0,
      "step": 3725
    },
    {
      "entropy": 6.506403636932373,
      "epoch": 0.37286949567651323,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991573358651787,
      "loss": 6.3779,
      "mean_token_accuracy": 0.1439211204648018,
      "num_tokens": 7312429.0,
      "step": 3730
    },
    {
      "entropy": 6.551886987686157,
      "epoch": 0.3733693207377418,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004991542471612864,
      "loss": 6.5527,
      "mean_token_accuracy": 0.12518264651298522,
      "num_tokens": 7322870.0,
      "step": 3735
    },
    {
      "entropy": 6.5870767593383786,
      "epoch": 0.37386914579897035,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004991511528177307,
      "loss": 6.49,
      "mean_token_accuracy": 0.13421414345502852,
      "num_tokens": 7333931.0,
      "step": 3740
    },
    {
      "entropy": 6.523426818847656,
      "epoch": 0.37436897086019894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991480528345898,
      "loss": 6.4986,
      "mean_token_accuracy": 0.1324862353503704,
      "num_tokens": 7344083.0,
      "step": 3745
    },
    {
      "entropy": 6.515659141540527,
      "epoch": 0.3748687959214275,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991449472119416,
      "loss": 6.4498,
      "mean_token_accuracy": 0.1334095098078251,
      "num_tokens": 7353544.0,
      "step": 3750
    },
    {
      "entropy": 6.502569532394409,
      "epoch": 0.37536862098265605,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991418359498642,
      "loss": 6.3899,
      "mean_token_accuracy": 0.13305365219712256,
      "num_tokens": 7362406.0,
      "step": 3755
    },
    {
      "entropy": 6.516208648681641,
      "epoch": 0.37586844604388464,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991387190484358,
      "loss": 6.4066,
      "mean_token_accuracy": 0.13303921073675157,
      "num_tokens": 7372056.0,
      "step": 3760
    },
    {
      "entropy": 6.49410400390625,
      "epoch": 0.37636827110511323,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499135596507735,
      "loss": 6.4253,
      "mean_token_accuracy": 0.13597688004374503,
      "num_tokens": 7381849.0,
      "step": 3765
    },
    {
      "entropy": 6.428682327270508,
      "epoch": 0.37686809616634176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991324683278402,
      "loss": 6.3933,
      "mean_token_accuracy": 0.138482216745615,
      "num_tokens": 7392648.0,
      "step": 3770
    },
    {
      "entropy": 6.400922012329102,
      "epoch": 0.37736792122757035,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991293345088301,
      "loss": 6.3272,
      "mean_token_accuracy": 0.13686150759458543,
      "num_tokens": 7402121.0,
      "step": 3775
    },
    {
      "entropy": 6.507281732559204,
      "epoch": 0.37786774628879893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991261950507837,
      "loss": 6.412,
      "mean_token_accuracy": 0.13798219338059425,
      "num_tokens": 7412097.0,
      "step": 3780
    },
    {
      "entropy": 6.571550798416138,
      "epoch": 0.37836757135002747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049912304995378,
      "loss": 6.5306,
      "mean_token_accuracy": 0.12735471054911612,
      "num_tokens": 7421945.0,
      "step": 3785
    },
    {
      "entropy": 6.567579460144043,
      "epoch": 0.37886739641125605,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004991198992178979,
      "loss": 6.4763,
      "mean_token_accuracy": 0.13734544292092324,
      "num_tokens": 7432245.0,
      "step": 3790
    },
    {
      "entropy": 6.4725042343139645,
      "epoch": 0.37936722147248464,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004991167428432168,
      "loss": 6.4371,
      "mean_token_accuracy": 0.13480219170451163,
      "num_tokens": 7441258.0,
      "step": 3795
    },
    {
      "entropy": 6.558250999450683,
      "epoch": 0.3798670465337132,
      "grad_norm": 1.125,
      "learning_rate": 0.0004991135808298162,
      "loss": 6.5256,
      "mean_token_accuracy": 0.12520443350076677,
      "num_tokens": 7451141.0,
      "step": 3800
    },
    {
      "entropy": 6.536304330825805,
      "epoch": 0.38036687159494176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004991104131777755,
      "loss": 6.4084,
      "mean_token_accuracy": 0.13391764536499978,
      "num_tokens": 7461146.0,
      "step": 3805
    },
    {
      "entropy": 6.4515245914459225,
      "epoch": 0.38086669665617034,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991072398871746,
      "loss": 6.412,
      "mean_token_accuracy": 0.13541153445839882,
      "num_tokens": 7470307.0,
      "step": 3810
    },
    {
      "entropy": 6.42217698097229,
      "epoch": 0.38136652171739893,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004991040609580932,
      "loss": 6.4991,
      "mean_token_accuracy": 0.1329572357237339,
      "num_tokens": 7478944.0,
      "step": 3815
    },
    {
      "entropy": 6.526308250427246,
      "epoch": 0.38186634677862746,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991008763906112,
      "loss": 6.4531,
      "mean_token_accuracy": 0.1284291811287403,
      "num_tokens": 7488143.0,
      "step": 3820
    },
    {
      "entropy": 6.528888988494873,
      "epoch": 0.38236617183985605,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990976861848087,
      "loss": 6.4455,
      "mean_token_accuracy": 0.12987478449940681,
      "num_tokens": 7497211.0,
      "step": 3825
    },
    {
      "entropy": 6.494016838073731,
      "epoch": 0.38286599690108464,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990944903407663,
      "loss": 6.4669,
      "mean_token_accuracy": 0.131079713255167,
      "num_tokens": 7506299.0,
      "step": 3830
    },
    {
      "entropy": 6.5576365947723385,
      "epoch": 0.38336582196231317,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990912888585641,
      "loss": 6.5064,
      "mean_token_accuracy": 0.13142666295170785,
      "num_tokens": 7516637.0,
      "step": 3835
    },
    {
      "entropy": 6.551287126541138,
      "epoch": 0.38386564702354176,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990880817382828,
      "loss": 6.409,
      "mean_token_accuracy": 0.14032236710190774,
      "num_tokens": 7526332.0,
      "step": 3840
    },
    {
      "entropy": 6.415332794189453,
      "epoch": 0.38436547208477034,
      "grad_norm": 1.15625,
      "learning_rate": 0.000499084868980003,
      "loss": 6.3618,
      "mean_token_accuracy": 0.14114759042859076,
      "num_tokens": 7535217.0,
      "step": 3845
    },
    {
      "entropy": 6.464153861999511,
      "epoch": 0.3848652971459989,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990816505838054,
      "loss": 6.468,
      "mean_token_accuracy": 0.1270267315208912,
      "num_tokens": 7544538.0,
      "step": 3850
    },
    {
      "entropy": 6.555702018737793,
      "epoch": 0.38536512220722746,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990784265497713,
      "loss": 6.444,
      "mean_token_accuracy": 0.13832010552287102,
      "num_tokens": 7555302.0,
      "step": 3855
    },
    {
      "entropy": 6.487966108322143,
      "epoch": 0.38586494726845605,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990751968779817,
      "loss": 6.4024,
      "mean_token_accuracy": 0.13191684782505037,
      "num_tokens": 7564543.0,
      "step": 3860
    },
    {
      "entropy": 6.411370420455933,
      "epoch": 0.38636477232968464,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004990719615685176,
      "loss": 6.5244,
      "mean_token_accuracy": 0.13178485706448556,
      "num_tokens": 7575042.0,
      "step": 3865
    },
    {
      "entropy": 6.533362817764282,
      "epoch": 0.38686459739091317,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004990687206214608,
      "loss": 6.5013,
      "mean_token_accuracy": 0.12548915296792984,
      "num_tokens": 7584467.0,
      "step": 3870
    },
    {
      "entropy": 6.492922067642212,
      "epoch": 0.38736442245214175,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990654740368925,
      "loss": 6.4605,
      "mean_token_accuracy": 0.1316137917339802,
      "num_tokens": 7594446.0,
      "step": 3875
    },
    {
      "entropy": 6.441819047927856,
      "epoch": 0.38786424751337034,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990622218148947,
      "loss": 6.3736,
      "mean_token_accuracy": 0.13815455064177512,
      "num_tokens": 7603200.0,
      "step": 3880
    },
    {
      "entropy": 6.529869604110718,
      "epoch": 0.3883640725745989,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990589639555489,
      "loss": 6.4235,
      "mean_token_accuracy": 0.14095360934734344,
      "num_tokens": 7613513.0,
      "step": 3885
    },
    {
      "entropy": 6.502209138870239,
      "epoch": 0.38886389763582746,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004990557004589373,
      "loss": 6.405,
      "mean_token_accuracy": 0.13607856631278992,
      "num_tokens": 7622913.0,
      "step": 3890
    },
    {
      "entropy": 6.461823844909668,
      "epoch": 0.38936372269705605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004990524313251419,
      "loss": 6.4402,
      "mean_token_accuracy": 0.13297955691814423,
      "num_tokens": 7632992.0,
      "step": 3895
    },
    {
      "entropy": 6.5716839790344235,
      "epoch": 0.3898635477582846,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004990491565542449,
      "loss": 6.4505,
      "mean_token_accuracy": 0.1292771354317665,
      "num_tokens": 7642620.0,
      "step": 3900
    },
    {
      "entropy": 6.393958282470703,
      "epoch": 0.39036337281951317,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990458761463289,
      "loss": 6.3498,
      "mean_token_accuracy": 0.14001418724656106,
      "num_tokens": 7651998.0,
      "step": 3905
    },
    {
      "entropy": 6.421148157119751,
      "epoch": 0.39086319788074175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004990425901014763,
      "loss": 6.447,
      "mean_token_accuracy": 0.12894242405891418,
      "num_tokens": 7660084.0,
      "step": 3910
    },
    {
      "entropy": 6.505327987670898,
      "epoch": 0.3913630229419703,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990392984197697,
      "loss": 6.3368,
      "mean_token_accuracy": 0.1454818181693554,
      "num_tokens": 7668833.0,
      "step": 3915
    },
    {
      "entropy": 6.481044960021973,
      "epoch": 0.39186284800319887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004990360011012921,
      "loss": 6.4152,
      "mean_token_accuracy": 0.13523860052227973,
      "num_tokens": 7677617.0,
      "step": 3920
    },
    {
      "entropy": 6.496452760696411,
      "epoch": 0.39236267306442746,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990326981461264,
      "loss": 6.5151,
      "mean_token_accuracy": 0.13482170030474663,
      "num_tokens": 7687794.0,
      "step": 3925
    },
    {
      "entropy": 6.515251302719117,
      "epoch": 0.39286249812565605,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004990293895543555,
      "loss": 6.4396,
      "mean_token_accuracy": 0.13241735622286796,
      "num_tokens": 7697608.0,
      "step": 3930
    },
    {
      "entropy": 6.452567052841187,
      "epoch": 0.3933623231868846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499026075326063,
      "loss": 6.3809,
      "mean_token_accuracy": 0.13826791048049927,
      "num_tokens": 7707391.0,
      "step": 3935
    },
    {
      "entropy": 6.465150833129883,
      "epoch": 0.39386214824811316,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499022755461332,
      "loss": 6.4368,
      "mean_token_accuracy": 0.12992112040519715,
      "num_tokens": 7718715.0,
      "step": 3940
    },
    {
      "entropy": 6.45237398147583,
      "epoch": 0.39436197330934175,
      "grad_norm": 0.9453125,
      "learning_rate": 0.000499019429960246,
      "loss": 6.4499,
      "mean_token_accuracy": 0.1323553889989853,
      "num_tokens": 7729481.0,
      "step": 3945
    },
    {
      "entropy": 6.359845161437988,
      "epoch": 0.3948617983705703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004990160988228888,
      "loss": 6.3715,
      "mean_token_accuracy": 0.13609442338347436,
      "num_tokens": 7739733.0,
      "step": 3950
    },
    {
      "entropy": 6.477339410781861,
      "epoch": 0.39536162343179887,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004990127620493444,
      "loss": 6.4121,
      "mean_token_accuracy": 0.14059725701808928,
      "num_tokens": 7749109.0,
      "step": 3955
    },
    {
      "entropy": 6.456976366043091,
      "epoch": 0.39586144849302746,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004990094196396964,
      "loss": 6.4823,
      "mean_token_accuracy": 0.1264084041118622,
      "num_tokens": 7758853.0,
      "step": 3960
    },
    {
      "entropy": 6.533520793914795,
      "epoch": 0.396361273554256,
      "grad_norm": 0.96875,
      "learning_rate": 0.000499006071594029,
      "loss": 6.4997,
      "mean_token_accuracy": 0.12903602197766303,
      "num_tokens": 7770831.0,
      "step": 3965
    },
    {
      "entropy": 6.5848551273345945,
      "epoch": 0.3968610986154846,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990027179124266,
      "loss": 6.5297,
      "mean_token_accuracy": 0.12930068746209145,
      "num_tokens": 7781094.0,
      "step": 3970
    },
    {
      "entropy": 6.603534841537476,
      "epoch": 0.39736092367671316,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989993585949733,
      "loss": 6.4915,
      "mean_token_accuracy": 0.1336494892835617,
      "num_tokens": 7790713.0,
      "step": 3975
    },
    {
      "entropy": 6.426853084564209,
      "epoch": 0.3978607487379417,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989959936417541,
      "loss": 6.4256,
      "mean_token_accuracy": 0.1350346639752388,
      "num_tokens": 7799425.0,
      "step": 3980
    },
    {
      "entropy": 6.470767307281494,
      "epoch": 0.3983605737991703,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000498992623052853,
      "loss": 6.4097,
      "mean_token_accuracy": 0.14077932760119438,
      "num_tokens": 7809451.0,
      "step": 3985
    },
    {
      "entropy": 6.517067861557007,
      "epoch": 0.39886039886039887,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989892468283553,
      "loss": 6.4471,
      "mean_token_accuracy": 0.13374447971582412,
      "num_tokens": 7819707.0,
      "step": 3990
    },
    {
      "entropy": 6.439810037612915,
      "epoch": 0.39936022392162746,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004989858649683457,
      "loss": 6.3512,
      "mean_token_accuracy": 0.13842166140675544,
      "num_tokens": 7828851.0,
      "step": 3995
    },
    {
      "entropy": 6.457694149017334,
      "epoch": 0.399860048982856,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989824774729094,
      "loss": 6.439,
      "mean_token_accuracy": 0.13478666245937349,
      "num_tokens": 7837999.0,
      "step": 4000
    },
    {
      "entropy": 6.546310710906982,
      "epoch": 0.4003598740440846,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989790843421317,
      "loss": 6.521,
      "mean_token_accuracy": 0.13211105540394782,
      "num_tokens": 7847227.0,
      "step": 4005
    },
    {
      "entropy": 6.565558958053589,
      "epoch": 0.40085969910531316,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989756855760979,
      "loss": 6.4825,
      "mean_token_accuracy": 0.1367221735417843,
      "num_tokens": 7857102.0,
      "step": 4010
    },
    {
      "entropy": 6.483669137954712,
      "epoch": 0.4013595241665417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004989722811748934,
      "loss": 6.5122,
      "mean_token_accuracy": 0.1291721984744072,
      "num_tokens": 7867395.0,
      "step": 4015
    },
    {
      "entropy": 6.518131256103516,
      "epoch": 0.4018593492277703,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004989688711386039,
      "loss": 6.3697,
      "mean_token_accuracy": 0.14073156714439392,
      "num_tokens": 7876461.0,
      "step": 4020
    },
    {
      "entropy": 6.4436448097229,
      "epoch": 0.40235917428899887,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989654554673152,
      "loss": 6.4458,
      "mean_token_accuracy": 0.1317490480840206,
      "num_tokens": 7886856.0,
      "step": 4025
    },
    {
      "entropy": 6.451264381408691,
      "epoch": 0.4028589993502274,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989620341611133,
      "loss": 6.301,
      "mean_token_accuracy": 0.1473988674581051,
      "num_tokens": 7896255.0,
      "step": 4030
    },
    {
      "entropy": 6.443401956558228,
      "epoch": 0.403358824411456,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989586072200843,
      "loss": 6.4495,
      "mean_token_accuracy": 0.13096531555056573,
      "num_tokens": 7905827.0,
      "step": 4035
    },
    {
      "entropy": 6.500145816802979,
      "epoch": 0.40385864947268457,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989551746443143,
      "loss": 6.4411,
      "mean_token_accuracy": 0.13025414422154427,
      "num_tokens": 7915610.0,
      "step": 4040
    },
    {
      "entropy": 6.513006258010864,
      "epoch": 0.4043584745339131,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989517364338899,
      "loss": 6.4719,
      "mean_token_accuracy": 0.1334102340042591,
      "num_tokens": 7926209.0,
      "step": 4045
    },
    {
      "entropy": 6.497201061248779,
      "epoch": 0.4048582995951417,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004989482925888974,
      "loss": 6.4493,
      "mean_token_accuracy": 0.13315200805664062,
      "num_tokens": 7936411.0,
      "step": 4050
    },
    {
      "entropy": 6.455505228042602,
      "epoch": 0.4053581246563703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004989448431094233,
      "loss": 6.3955,
      "mean_token_accuracy": 0.1399148739874363,
      "num_tokens": 7945874.0,
      "step": 4055
    },
    {
      "entropy": 6.492669630050659,
      "epoch": 0.40585794971759886,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989413879955548,
      "loss": 6.3911,
      "mean_token_accuracy": 0.13562870845198632,
      "num_tokens": 7955303.0,
      "step": 4060
    },
    {
      "entropy": 6.423491668701172,
      "epoch": 0.4063577747788274,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004989379272473785,
      "loss": 6.3813,
      "mean_token_accuracy": 0.13963470309972764,
      "num_tokens": 7964365.0,
      "step": 4065
    },
    {
      "entropy": 6.50611138343811,
      "epoch": 0.406857599840056,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004989344608649815,
      "loss": 6.3922,
      "mean_token_accuracy": 0.13037322759628295,
      "num_tokens": 7973692.0,
      "step": 4070
    },
    {
      "entropy": 6.476632690429687,
      "epoch": 0.40735742490128457,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989309888484511,
      "loss": 6.4088,
      "mean_token_accuracy": 0.13078400045633315,
      "num_tokens": 7984003.0,
      "step": 4075
    },
    {
      "entropy": 6.544618034362793,
      "epoch": 0.4078572499625131,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004989275111978748,
      "loss": 6.5038,
      "mean_token_accuracy": 0.13528040200471877,
      "num_tokens": 7994515.0,
      "step": 4080
    },
    {
      "entropy": 6.454639863967896,
      "epoch": 0.4083570750237417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989240279133398,
      "loss": 6.39,
      "mean_token_accuracy": 0.1299092449247837,
      "num_tokens": 8004000.0,
      "step": 4085
    },
    {
      "entropy": 6.433520269393921,
      "epoch": 0.4088569000849703,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004989205389949339,
      "loss": 6.4253,
      "mean_token_accuracy": 0.13190341517329215,
      "num_tokens": 8013880.0,
      "step": 4090
    },
    {
      "entropy": 6.513592958450317,
      "epoch": 0.4093567251461988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989170444427448,
      "loss": 6.3042,
      "mean_token_accuracy": 0.1402171105146408,
      "num_tokens": 8024395.0,
      "step": 4095
    },
    {
      "entropy": 6.468240737915039,
      "epoch": 0.4098565502074274,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004989135442568606,
      "loss": 6.3652,
      "mean_token_accuracy": 0.1370164304971695,
      "num_tokens": 8033198.0,
      "step": 4100
    },
    {
      "entropy": 6.3534058094024655,
      "epoch": 0.410356375268656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989100384373693,
      "loss": 6.32,
      "mean_token_accuracy": 0.1352598898112774,
      "num_tokens": 8042387.0,
      "step": 4105
    },
    {
      "entropy": 6.469276714324951,
      "epoch": 0.4108562003298845,
      "grad_norm": 1.25,
      "learning_rate": 0.0004989065269843588,
      "loss": 6.4567,
      "mean_token_accuracy": 0.1320345677435398,
      "num_tokens": 8051993.0,
      "step": 4110
    },
    {
      "entropy": 6.417930269241333,
      "epoch": 0.4113560253911131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498903009897918,
      "loss": 6.4308,
      "mean_token_accuracy": 0.1308579184114933,
      "num_tokens": 8062122.0,
      "step": 4115
    },
    {
      "entropy": 6.491260433197022,
      "epoch": 0.4118558504523417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988994871781349,
      "loss": 6.4327,
      "mean_token_accuracy": 0.1321652762591839,
      "num_tokens": 8071828.0,
      "step": 4120
    },
    {
      "entropy": 6.501120710372925,
      "epoch": 0.4123556755135703,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988959588250983,
      "loss": 6.3884,
      "mean_token_accuracy": 0.12932343259453774,
      "num_tokens": 8081017.0,
      "step": 4125
    },
    {
      "entropy": 6.476039409637451,
      "epoch": 0.4128555005747988,
      "grad_norm": 1.25,
      "learning_rate": 0.000498892424838897,
      "loss": 6.4357,
      "mean_token_accuracy": 0.1337422288954258,
      "num_tokens": 8091706.0,
      "step": 4130
    },
    {
      "entropy": 6.434182596206665,
      "epoch": 0.4133553256360274,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988888852196199,
      "loss": 6.4137,
      "mean_token_accuracy": 0.13369264751672744,
      "num_tokens": 8100098.0,
      "step": 4135
    },
    {
      "entropy": 6.483431243896485,
      "epoch": 0.413855150697256,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988853399673562,
      "loss": 6.4533,
      "mean_token_accuracy": 0.13221644908189772,
      "num_tokens": 8110003.0,
      "step": 4140
    },
    {
      "entropy": 6.487273025512695,
      "epoch": 0.4143549757584845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988817890821948,
      "loss": 6.3531,
      "mean_token_accuracy": 0.14411598891019822,
      "num_tokens": 8119440.0,
      "step": 4145
    },
    {
      "entropy": 6.463832950592041,
      "epoch": 0.4148548008197131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988782325642252,
      "loss": 6.4112,
      "mean_token_accuracy": 0.13876034691929817,
      "num_tokens": 8130347.0,
      "step": 4150
    },
    {
      "entropy": 6.488987922668457,
      "epoch": 0.4153546258809417,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498874670413537,
      "loss": 6.437,
      "mean_token_accuracy": 0.13633707016706467,
      "num_tokens": 8139252.0,
      "step": 4155
    },
    {
      "entropy": 6.531680297851563,
      "epoch": 0.4158544509421702,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004988711026302197,
      "loss": 6.5121,
      "mean_token_accuracy": 0.12838623225688933,
      "num_tokens": 8149130.0,
      "step": 4160
    },
    {
      "entropy": 6.4443779468536375,
      "epoch": 0.4163542760033988,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004988675292143629,
      "loss": 6.4153,
      "mean_token_accuracy": 0.13425033167004585,
      "num_tokens": 8158300.0,
      "step": 4165
    },
    {
      "entropy": 6.450055932998657,
      "epoch": 0.4168541010646274,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988639501660568,
      "loss": 6.3868,
      "mean_token_accuracy": 0.13577555269002914,
      "num_tokens": 8168631.0,
      "step": 4170
    },
    {
      "entropy": 6.517928552627564,
      "epoch": 0.4173539261258559,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988603654853913,
      "loss": 6.4999,
      "mean_token_accuracy": 0.1300140753388405,
      "num_tokens": 8178942.0,
      "step": 4175
    },
    {
      "entropy": 6.545652627944946,
      "epoch": 0.4178537511870845,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988567751724565,
      "loss": 6.402,
      "mean_token_accuracy": 0.13343972191214562,
      "num_tokens": 8188633.0,
      "step": 4180
    },
    {
      "entropy": 6.467659902572632,
      "epoch": 0.4183535762483131,
      "grad_norm": 1.0,
      "learning_rate": 0.000498853179227343,
      "loss": 6.3943,
      "mean_token_accuracy": 0.13899159505963327,
      "num_tokens": 8198486.0,
      "step": 4185
    },
    {
      "entropy": 6.492933893203736,
      "epoch": 0.4188534013095417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498849577650141,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13957841768860818,
      "num_tokens": 8208369.0,
      "step": 4190
    },
    {
      "entropy": 6.408676815032959,
      "epoch": 0.4193532263707702,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004988459704409413,
      "loss": 6.273,
      "mean_token_accuracy": 0.13954004943370818,
      "num_tokens": 8217968.0,
      "step": 4195
    },
    {
      "entropy": 6.450369024276734,
      "epoch": 0.4198530514319988,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004988423575998345,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13464478105306626,
      "num_tokens": 8228704.0,
      "step": 4200
    },
    {
      "entropy": 6.461948823928833,
      "epoch": 0.4203528764932274,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988387391269117,
      "loss": 6.331,
      "mean_token_accuracy": 0.13905152603983878,
      "num_tokens": 8239185.0,
      "step": 4205
    },
    {
      "entropy": 6.438327312469482,
      "epoch": 0.4208527015544559,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988351150222638,
      "loss": 6.4171,
      "mean_token_accuracy": 0.13897168040275573,
      "num_tokens": 8248644.0,
      "step": 4210
    },
    {
      "entropy": 6.460425949096679,
      "epoch": 0.4213525266156845,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004988314852859819,
      "loss": 6.4667,
      "mean_token_accuracy": 0.12945584654808046,
      "num_tokens": 8259875.0,
      "step": 4215
    },
    {
      "entropy": 6.394307851791382,
      "epoch": 0.4218523516769131,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004988278499181575,
      "loss": 6.3423,
      "mean_token_accuracy": 0.1456516355276108,
      "num_tokens": 8268894.0,
      "step": 4220
    },
    {
      "entropy": 6.4887917041778564,
      "epoch": 0.4223521767381416,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498824208918882,
      "loss": 6.4203,
      "mean_token_accuracy": 0.12808949053287505,
      "num_tokens": 8279236.0,
      "step": 4225
    },
    {
      "entropy": 6.395545101165771,
      "epoch": 0.4228520017993702,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498820562288247,
      "loss": 6.3225,
      "mean_token_accuracy": 0.139143730700016,
      "num_tokens": 8288464.0,
      "step": 4230
    },
    {
      "entropy": 6.328132152557373,
      "epoch": 0.4233518268605988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004988169100263442,
      "loss": 6.311,
      "mean_token_accuracy": 0.13848703801631929,
      "num_tokens": 8298291.0,
      "step": 4235
    },
    {
      "entropy": 6.534720182418823,
      "epoch": 0.4238516519218274,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988132521332656,
      "loss": 6.5146,
      "mean_token_accuracy": 0.12758759558200836,
      "num_tokens": 8308543.0,
      "step": 4240
    },
    {
      "entropy": 6.536624002456665,
      "epoch": 0.4243514769830559,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004988095886091031,
      "loss": 6.4931,
      "mean_token_accuracy": 0.12689031660556793,
      "num_tokens": 8318310.0,
      "step": 4245
    },
    {
      "entropy": 6.4499787330627445,
      "epoch": 0.4248513020442845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004988059194539489,
      "loss": 6.4006,
      "mean_token_accuracy": 0.1328957572579384,
      "num_tokens": 8327853.0,
      "step": 4250
    },
    {
      "entropy": 6.469836950302124,
      "epoch": 0.4253511271055131,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004988022446678954,
      "loss": 6.4892,
      "mean_token_accuracy": 0.1268715277314186,
      "num_tokens": 8337734.0,
      "step": 4255
    },
    {
      "entropy": 6.5203282833099365,
      "epoch": 0.4258509521667416,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004987985642510352,
      "loss": 6.4401,
      "mean_token_accuracy": 0.13205364644527434,
      "num_tokens": 8346681.0,
      "step": 4260
    },
    {
      "entropy": 6.51040678024292,
      "epoch": 0.4263507772279702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987948782034605,
      "loss": 6.2925,
      "mean_token_accuracy": 0.13818243965506555,
      "num_tokens": 8355597.0,
      "step": 4265
    },
    {
      "entropy": 6.510284423828125,
      "epoch": 0.4268506022891988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987911865252644,
      "loss": 6.436,
      "mean_token_accuracy": 0.13208968788385392,
      "num_tokens": 8365529.0,
      "step": 4270
    },
    {
      "entropy": 6.371225118637085,
      "epoch": 0.42735042735042733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987874892165395,
      "loss": 6.3501,
      "mean_token_accuracy": 0.1374231293797493,
      "num_tokens": 8374727.0,
      "step": 4275
    },
    {
      "entropy": 6.437613773345947,
      "epoch": 0.4278502524116559,
      "grad_norm": 1.0,
      "learning_rate": 0.000498783786277379,
      "loss": 6.3706,
      "mean_token_accuracy": 0.13507111370563507,
      "num_tokens": 8385624.0,
      "step": 4280
    },
    {
      "entropy": 6.496224737167358,
      "epoch": 0.4283500774728845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987800777078759,
      "loss": 6.3556,
      "mean_token_accuracy": 0.13502876758575438,
      "num_tokens": 8395242.0,
      "step": 4285
    },
    {
      "entropy": 6.33849139213562,
      "epoch": 0.42884990253411304,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987763635081238,
      "loss": 6.2662,
      "mean_token_accuracy": 0.14308104962110518,
      "num_tokens": 8404946.0,
      "step": 4290
    },
    {
      "entropy": 6.470030641555786,
      "epoch": 0.4293497275953416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498772643678216,
      "loss": 6.5357,
      "mean_token_accuracy": 0.13231894373893738,
      "num_tokens": 8413835.0,
      "step": 4295
    },
    {
      "entropy": 6.5561529159545895,
      "epoch": 0.4298495526565702,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004987689182182459,
      "loss": 6.3717,
      "mean_token_accuracy": 0.13529768362641334,
      "num_tokens": 8423338.0,
      "step": 4300
    },
    {
      "entropy": 6.5434637546539305,
      "epoch": 0.4303493777177988,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004987651871283075,
      "loss": 6.5096,
      "mean_token_accuracy": 0.126751921325922,
      "num_tokens": 8434370.0,
      "step": 4305
    },
    {
      "entropy": 6.438532543182373,
      "epoch": 0.43084920277902733,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987614504084946,
      "loss": 6.3438,
      "mean_token_accuracy": 0.13975988179445267,
      "num_tokens": 8444517.0,
      "step": 4310
    },
    {
      "entropy": 6.401457786560059,
      "epoch": 0.4313490278402559,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004987577080589012,
      "loss": 6.3868,
      "mean_token_accuracy": 0.13699072524905204,
      "num_tokens": 8453269.0,
      "step": 4315
    },
    {
      "entropy": 6.4171919345855715,
      "epoch": 0.4318488529014845,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987539600796213,
      "loss": 6.393,
      "mean_token_accuracy": 0.13605409860610962,
      "num_tokens": 8462142.0,
      "step": 4320
    },
    {
      "entropy": 6.567247486114502,
      "epoch": 0.43234867796271303,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004987502064707494,
      "loss": 6.4328,
      "mean_token_accuracy": 0.13245702013373375,
      "num_tokens": 8472020.0,
      "step": 4325
    },
    {
      "entropy": 6.462252378463745,
      "epoch": 0.4328485030239416,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987464472323799,
      "loss": 6.4331,
      "mean_token_accuracy": 0.13206341490149498,
      "num_tokens": 8481985.0,
      "step": 4330
    },
    {
      "entropy": 6.4948139667510985,
      "epoch": 0.4333483280851702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987426823646074,
      "loss": 6.4026,
      "mean_token_accuracy": 0.13848372176289558,
      "num_tokens": 8490859.0,
      "step": 4335
    },
    {
      "entropy": 6.452024841308594,
      "epoch": 0.43384815314639874,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004987389118675266,
      "loss": 6.3942,
      "mean_token_accuracy": 0.1345831722021103,
      "num_tokens": 8500714.0,
      "step": 4340
    },
    {
      "entropy": 6.424344491958618,
      "epoch": 0.4343479782076273,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004987351357412322,
      "loss": 6.2993,
      "mean_token_accuracy": 0.14222790896892548,
      "num_tokens": 8509832.0,
      "step": 4345
    },
    {
      "entropy": 6.325907468795776,
      "epoch": 0.4348478032688559,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004987313539858196,
      "loss": 6.2508,
      "mean_token_accuracy": 0.13821439296007157,
      "num_tokens": 8519455.0,
      "step": 4350
    },
    {
      "entropy": 6.501128530502319,
      "epoch": 0.43534762833008445,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004987275666013836,
      "loss": 6.4858,
      "mean_token_accuracy": 0.13098311722278594,
      "num_tokens": 8529015.0,
      "step": 4355
    },
    {
      "entropy": 6.47471809387207,
      "epoch": 0.43584745339131303,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987237735880195,
      "loss": 6.4051,
      "mean_token_accuracy": 0.1321643427014351,
      "num_tokens": 8539081.0,
      "step": 4360
    },
    {
      "entropy": 6.455320549011231,
      "epoch": 0.4363472784525416,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004987199749458229,
      "loss": 6.4125,
      "mean_token_accuracy": 0.13527554869651795,
      "num_tokens": 8549889.0,
      "step": 4365
    },
    {
      "entropy": 6.461476278305054,
      "epoch": 0.4368471035137702,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004987161706748893,
      "loss": 6.3904,
      "mean_token_accuracy": 0.13609796687960624,
      "num_tokens": 8560331.0,
      "step": 4370
    },
    {
      "entropy": 6.426600408554077,
      "epoch": 0.43734692857499874,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987123607753144,
      "loss": 6.3399,
      "mean_token_accuracy": 0.13802748620510102,
      "num_tokens": 8570035.0,
      "step": 4375
    },
    {
      "entropy": 6.419609642028808,
      "epoch": 0.4378467536362273,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498708545247194,
      "loss": 6.3681,
      "mean_token_accuracy": 0.1373009629547596,
      "num_tokens": 8579069.0,
      "step": 4380
    },
    {
      "entropy": 6.473028039932251,
      "epoch": 0.4383465786974559,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987047240906243,
      "loss": 6.4575,
      "mean_token_accuracy": 0.13494816347956656,
      "num_tokens": 8589542.0,
      "step": 4385
    },
    {
      "entropy": 6.50813364982605,
      "epoch": 0.43884640375868444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004987008973057012,
      "loss": 6.3705,
      "mean_token_accuracy": 0.13491581231355668,
      "num_tokens": 8600143.0,
      "step": 4390
    },
    {
      "entropy": 6.479402160644531,
      "epoch": 0.43934622881991303,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986970648925211,
      "loss": 6.3605,
      "mean_token_accuracy": 0.13793511241674422,
      "num_tokens": 8609062.0,
      "step": 4395
    },
    {
      "entropy": 6.423627805709839,
      "epoch": 0.4398460538811416,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986932268511804,
      "loss": 6.4203,
      "mean_token_accuracy": 0.1294531263411045,
      "num_tokens": 8618115.0,
      "step": 4400
    },
    {
      "entropy": 6.3790483474731445,
      "epoch": 0.44034587894237015,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986893831817758,
      "loss": 6.25,
      "mean_token_accuracy": 0.1501368187367916,
      "num_tokens": 8627091.0,
      "step": 4405
    },
    {
      "entropy": 6.4030008792877195,
      "epoch": 0.44084570400359874,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004986855338844038,
      "loss": 6.3224,
      "mean_token_accuracy": 0.14157467931509018,
      "num_tokens": 8637194.0,
      "step": 4410
    },
    {
      "entropy": 6.416748762130737,
      "epoch": 0.4413455290648273,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986816789591613,
      "loss": 6.3766,
      "mean_token_accuracy": 0.13759900704026223,
      "num_tokens": 8647281.0,
      "step": 4415
    },
    {
      "entropy": 6.520883798599243,
      "epoch": 0.44184535412605586,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986778184061453,
      "loss": 6.5028,
      "mean_token_accuracy": 0.12955915778875352,
      "num_tokens": 8658004.0,
      "step": 4420
    },
    {
      "entropy": 6.36964750289917,
      "epoch": 0.44234517918728444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986739522254529,
      "loss": 6.2533,
      "mean_token_accuracy": 0.14594935923814772,
      "num_tokens": 8666878.0,
      "step": 4425
    },
    {
      "entropy": 6.433926582336426,
      "epoch": 0.44284500424851303,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004986700804171816,
      "loss": 6.3598,
      "mean_token_accuracy": 0.1429945483803749,
      "num_tokens": 8677695.0,
      "step": 4430
    },
    {
      "entropy": 6.326544284820557,
      "epoch": 0.4433448293097416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986662029814285,
      "loss": 6.166,
      "mean_token_accuracy": 0.14916819408535958,
      "num_tokens": 8686767.0,
      "step": 4435
    },
    {
      "entropy": 6.3682225227355955,
      "epoch": 0.44384465437097015,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986623199182913,
      "loss": 6.4061,
      "mean_token_accuracy": 0.13438173830509187,
      "num_tokens": 8695804.0,
      "step": 4440
    },
    {
      "entropy": 6.371269845962525,
      "epoch": 0.44434447943219874,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986584312278678,
      "loss": 6.3766,
      "mean_token_accuracy": 0.1354999400675297,
      "num_tokens": 8706079.0,
      "step": 4445
    },
    {
      "entropy": 6.620289754867554,
      "epoch": 0.4448443044934273,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986545369102556,
      "loss": 6.4915,
      "mean_token_accuracy": 0.13768121525645255,
      "num_tokens": 8715754.0,
      "step": 4450
    },
    {
      "entropy": 6.516203212738037,
      "epoch": 0.44534412955465585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986506369655528,
      "loss": 6.3899,
      "mean_token_accuracy": 0.13472285643219947,
      "num_tokens": 8725745.0,
      "step": 4455
    },
    {
      "entropy": 6.386401653289795,
      "epoch": 0.44584395461588444,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004986467313938575,
      "loss": 6.3556,
      "mean_token_accuracy": 0.14282679110765456,
      "num_tokens": 8735043.0,
      "step": 4460
    },
    {
      "entropy": 6.366832685470581,
      "epoch": 0.44634377967711303,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498642820195268,
      "loss": 6.3194,
      "mean_token_accuracy": 0.1430388942360878,
      "num_tokens": 8744631.0,
      "step": 4465
    },
    {
      "entropy": 6.46613712310791,
      "epoch": 0.44684360473834156,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004986389033698827,
      "loss": 6.3923,
      "mean_token_accuracy": 0.13856377005577086,
      "num_tokens": 8754148.0,
      "step": 4470
    },
    {
      "entropy": 6.502831268310547,
      "epoch": 0.44734342979957015,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986349809178002,
      "loss": 6.4235,
      "mean_token_accuracy": 0.1359233595430851,
      "num_tokens": 8763230.0,
      "step": 4475
    },
    {
      "entropy": 6.422842645645142,
      "epoch": 0.44784325486079873,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004986310528391192,
      "loss": 6.4594,
      "mean_token_accuracy": 0.13673357516527176,
      "num_tokens": 8772547.0,
      "step": 4480
    },
    {
      "entropy": 6.466714572906494,
      "epoch": 0.44834307992202727,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004986271191339383,
      "loss": 6.3076,
      "mean_token_accuracy": 0.14119416773319243,
      "num_tokens": 8782766.0,
      "step": 4485
    },
    {
      "entropy": 6.451236438751221,
      "epoch": 0.44884290498325585,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004986231798023566,
      "loss": 6.3327,
      "mean_token_accuracy": 0.1371416814625263,
      "num_tokens": 8791793.0,
      "step": 4490
    },
    {
      "entropy": 6.4383210182189945,
      "epoch": 0.44934273004448444,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004986192348444734,
      "loss": 6.3957,
      "mean_token_accuracy": 0.13699855506420136,
      "num_tokens": 8801888.0,
      "step": 4495
    },
    {
      "entropy": 6.355791139602661,
      "epoch": 0.449842555105713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004986152842603877,
      "loss": 6.2381,
      "mean_token_accuracy": 0.1472399652004242,
      "num_tokens": 8810629.0,
      "step": 4500
    },
    {
      "entropy": 6.453516817092895,
      "epoch": 0.45034238016694156,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986113280501989,
      "loss": 6.4464,
      "mean_token_accuracy": 0.1258704885840416,
      "num_tokens": 8820598.0,
      "step": 4505
    },
    {
      "entropy": 6.522357416152954,
      "epoch": 0.45084220522817015,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986073662140068,
      "loss": 6.3824,
      "mean_token_accuracy": 0.1371746152639389,
      "num_tokens": 8829814.0,
      "step": 4510
    },
    {
      "entropy": 6.39769434928894,
      "epoch": 0.45134203028939873,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986033987519109,
      "loss": 6.3519,
      "mean_token_accuracy": 0.1288498856127262,
      "num_tokens": 8839899.0,
      "step": 4515
    },
    {
      "entropy": 6.5056960105896,
      "epoch": 0.45184185535062726,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004985994256640108,
      "loss": 6.3724,
      "mean_token_accuracy": 0.1361602909862995,
      "num_tokens": 8848318.0,
      "step": 4520
    },
    {
      "entropy": 6.48440113067627,
      "epoch": 0.45234168041185585,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004985954469504069,
      "loss": 6.3579,
      "mean_token_accuracy": 0.13688973411917688,
      "num_tokens": 8857965.0,
      "step": 4525
    },
    {
      "entropy": 6.371800327301026,
      "epoch": 0.45284150547308444,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004985914626111991,
      "loss": 6.3852,
      "mean_token_accuracy": 0.13857995495200157,
      "num_tokens": 8868324.0,
      "step": 4530
    },
    {
      "entropy": 6.44323410987854,
      "epoch": 0.45334133053431297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004985874726464875,
      "loss": 6.369,
      "mean_token_accuracy": 0.1334787741303444,
      "num_tokens": 8877110.0,
      "step": 4535
    },
    {
      "entropy": 6.420333528518677,
      "epoch": 0.45384115559554156,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004985834770563727,
      "loss": 6.3358,
      "mean_token_accuracy": 0.13506197556853294,
      "num_tokens": 8886635.0,
      "step": 4540
    },
    {
      "entropy": 6.437656354904175,
      "epoch": 0.45434098065677014,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004985794758409552,
      "loss": 6.3242,
      "mean_token_accuracy": 0.13961484134197236,
      "num_tokens": 8895184.0,
      "step": 4545
    },
    {
      "entropy": 6.378831434249878,
      "epoch": 0.4548408057179987,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004985754690003354,
      "loss": 6.2927,
      "mean_token_accuracy": 0.14994147494435311,
      "num_tokens": 8905638.0,
      "step": 4550
    },
    {
      "entropy": 6.4137040138244625,
      "epoch": 0.45534063077922726,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004985714565346145,
      "loss": 6.3917,
      "mean_token_accuracy": 0.1407515712082386,
      "num_tokens": 8915038.0,
      "step": 4555
    },
    {
      "entropy": 6.41608190536499,
      "epoch": 0.45584045584045585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004985674384438933,
      "loss": 6.3453,
      "mean_token_accuracy": 0.1369032919406891,
      "num_tokens": 8923745.0,
      "step": 4560
    },
    {
      "entropy": 6.364067602157593,
      "epoch": 0.45634028090168444,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985634147282729,
      "loss": 6.3464,
      "mean_token_accuracy": 0.1350460559129715,
      "num_tokens": 8934019.0,
      "step": 4565
    },
    {
      "entropy": 6.493156957626343,
      "epoch": 0.45684010596291297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985593853878544,
      "loss": 6.3951,
      "mean_token_accuracy": 0.13617311045527458,
      "num_tokens": 8943763.0,
      "step": 4570
    },
    {
      "entropy": 6.41776762008667,
      "epoch": 0.45733993102414155,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004985553504227394,
      "loss": 6.3156,
      "mean_token_accuracy": 0.14055898934602737,
      "num_tokens": 8953757.0,
      "step": 4575
    },
    {
      "entropy": 6.3229217529296875,
      "epoch": 0.45783975608537014,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985513098330293,
      "loss": 6.3313,
      "mean_token_accuracy": 0.1386001244187355,
      "num_tokens": 8963233.0,
      "step": 4580
    },
    {
      "entropy": 6.322848796844482,
      "epoch": 0.4583395811465987,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004985472636188257,
      "loss": 6.2801,
      "mean_token_accuracy": 0.14049771055579185,
      "num_tokens": 8972210.0,
      "step": 4585
    },
    {
      "entropy": 6.381848239898682,
      "epoch": 0.45883940620782726,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004985432117802305,
      "loss": 6.2408,
      "mean_token_accuracy": 0.14359138607978822,
      "num_tokens": 8982818.0,
      "step": 4590
    },
    {
      "entropy": 6.471052026748657,
      "epoch": 0.45933923126905585,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004985391543173456,
      "loss": 6.4111,
      "mean_token_accuracy": 0.1319870486855507,
      "num_tokens": 8992019.0,
      "step": 4595
    },
    {
      "entropy": 6.42074728012085,
      "epoch": 0.4598390563302844,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000498535091230273,
      "loss": 6.2938,
      "mean_token_accuracy": 0.14461407363414763,
      "num_tokens": 9000891.0,
      "step": 4600
    },
    {
      "entropy": 6.416874694824219,
      "epoch": 0.46033888139151297,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985310225191152,
      "loss": 6.3435,
      "mean_token_accuracy": 0.13044048994779586,
      "num_tokens": 9010660.0,
      "step": 4605
    },
    {
      "entropy": 6.432283926010132,
      "epoch": 0.46083870645274155,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985269481839744,
      "loss": 6.4227,
      "mean_token_accuracy": 0.1330641560256481,
      "num_tokens": 9020078.0,
      "step": 4610
    },
    {
      "entropy": 6.42190203666687,
      "epoch": 0.4613385315139701,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985228682249529,
      "loss": 6.3358,
      "mean_token_accuracy": 0.142258732765913,
      "num_tokens": 9030728.0,
      "step": 4615
    },
    {
      "entropy": 6.392468881607056,
      "epoch": 0.46183835657519867,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004985187826421538,
      "loss": 6.2669,
      "mean_token_accuracy": 0.14326582998037338,
      "num_tokens": 9041271.0,
      "step": 4620
    },
    {
      "entropy": 6.3167366027832035,
      "epoch": 0.46233818163642726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985146914356795,
      "loss": 6.3789,
      "mean_token_accuracy": 0.13923989087343216,
      "num_tokens": 9050135.0,
      "step": 4625
    },
    {
      "entropy": 6.399218511581421,
      "epoch": 0.46283800669765585,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498510594605633,
      "loss": 6.3561,
      "mean_token_accuracy": 0.13930042833089828,
      "num_tokens": 9059374.0,
      "step": 4630
    },
    {
      "entropy": 6.498583555221558,
      "epoch": 0.4633378317588844,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985064921521177,
      "loss": 6.3848,
      "mean_token_accuracy": 0.13779570236802102,
      "num_tokens": 9069568.0,
      "step": 4635
    },
    {
      "entropy": 6.437125587463379,
      "epoch": 0.46383765682011296,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004985023840752364,
      "loss": 6.319,
      "mean_token_accuracy": 0.13617421090602874,
      "num_tokens": 9079952.0,
      "step": 4640
    },
    {
      "entropy": 6.335849761962891,
      "epoch": 0.46433748188134155,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984982703750926,
      "loss": 6.2762,
      "mean_token_accuracy": 0.14604012295603752,
      "num_tokens": 9088968.0,
      "step": 4645
    },
    {
      "entropy": 6.408214282989502,
      "epoch": 0.4648373069425701,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984941510517899,
      "loss": 6.3846,
      "mean_token_accuracy": 0.13619247525930406,
      "num_tokens": 9099736.0,
      "step": 4650
    },
    {
      "entropy": 6.430678081512451,
      "epoch": 0.46533713200379867,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984900261054319,
      "loss": 6.3811,
      "mean_token_accuracy": 0.13281464502215384,
      "num_tokens": 9109200.0,
      "step": 4655
    },
    {
      "entropy": 6.432093381881714,
      "epoch": 0.46583695706502726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984858955361224,
      "loss": 6.4972,
      "mean_token_accuracy": 0.1263826735317707,
      "num_tokens": 9118283.0,
      "step": 4660
    },
    {
      "entropy": 6.491627931594849,
      "epoch": 0.4663367821262558,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984817593439651,
      "loss": 6.3284,
      "mean_token_accuracy": 0.14038715958595277,
      "num_tokens": 9126765.0,
      "step": 4665
    },
    {
      "entropy": 6.434687185287475,
      "epoch": 0.4668366071874844,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984776175290643,
      "loss": 6.3786,
      "mean_token_accuracy": 0.13448265716433525,
      "num_tokens": 9137210.0,
      "step": 4670
    },
    {
      "entropy": 6.342008113861084,
      "epoch": 0.46733643224871296,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984734700915242,
      "loss": 6.2895,
      "mean_token_accuracy": 0.14407761842012407,
      "num_tokens": 9147811.0,
      "step": 4675
    },
    {
      "entropy": 6.392057323455811,
      "epoch": 0.4678362573099415,
      "grad_norm": 1.140625,
      "learning_rate": 0.000498469317031449,
      "loss": 6.2955,
      "mean_token_accuracy": 0.14150434732437134,
      "num_tokens": 9159080.0,
      "step": 4680
    },
    {
      "entropy": 6.363104009628296,
      "epoch": 0.4683360823711701,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004984651583489434,
      "loss": 6.3116,
      "mean_token_accuracy": 0.14037407487630843,
      "num_tokens": 9168959.0,
      "step": 4685
    },
    {
      "entropy": 6.417545747756958,
      "epoch": 0.46883590743239867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984609940441119,
      "loss": 6.3564,
      "mean_token_accuracy": 0.13387664705514907,
      "num_tokens": 9178873.0,
      "step": 4690
    },
    {
      "entropy": 6.380366325378418,
      "epoch": 0.46933573249362726,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984568241170593,
      "loss": 6.2206,
      "mean_token_accuracy": 0.14628685638308525,
      "num_tokens": 9188056.0,
      "step": 4695
    },
    {
      "entropy": 6.329846715927124,
      "epoch": 0.4698355575548558,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984526485678905,
      "loss": 6.4526,
      "mean_token_accuracy": 0.1362912803888321,
      "num_tokens": 9197993.0,
      "step": 4700
    },
    {
      "entropy": 6.508334684371948,
      "epoch": 0.4703353826160844,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004984484673967105,
      "loss": 6.4467,
      "mean_token_accuracy": 0.13159896954894065,
      "num_tokens": 9207719.0,
      "step": 4705
    },
    {
      "entropy": 6.496994066238403,
      "epoch": 0.47083520767731296,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984442806036246,
      "loss": 6.3155,
      "mean_token_accuracy": 0.14291745722293853,
      "num_tokens": 9217036.0,
      "step": 4710
    },
    {
      "entropy": 6.3403778076171875,
      "epoch": 0.4713350327385415,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004984400881887381,
      "loss": 6.3212,
      "mean_token_accuracy": 0.13848632648587228,
      "num_tokens": 9226132.0,
      "step": 4715
    },
    {
      "entropy": 6.386812734603882,
      "epoch": 0.4718348577997701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984358901521565,
      "loss": 6.3428,
      "mean_token_accuracy": 0.1379224218428135,
      "num_tokens": 9235914.0,
      "step": 4720
    },
    {
      "entropy": 6.472326421737671,
      "epoch": 0.47233468286099867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984316864939853,
      "loss": 6.3721,
      "mean_token_accuracy": 0.13258715271949767,
      "num_tokens": 9245269.0,
      "step": 4725
    },
    {
      "entropy": 6.254420375823974,
      "epoch": 0.4728345079222272,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984274772143304,
      "loss": 6.245,
      "mean_token_accuracy": 0.1441103436052799,
      "num_tokens": 9254107.0,
      "step": 4730
    },
    {
      "entropy": 6.485814762115479,
      "epoch": 0.4733343329834558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984232623132977,
      "loss": 6.4476,
      "mean_token_accuracy": 0.12931791320443153,
      "num_tokens": 9265219.0,
      "step": 4735
    },
    {
      "entropy": 6.4238097190856935,
      "epoch": 0.4738341580446844,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004984190417909932,
      "loss": 6.3635,
      "mean_token_accuracy": 0.13524677157402037,
      "num_tokens": 9274610.0,
      "step": 4740
    },
    {
      "entropy": 6.376027536392212,
      "epoch": 0.4743339831059129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984148156475231,
      "loss": 6.3611,
      "mean_token_accuracy": 0.14140876084566117,
      "num_tokens": 9283576.0,
      "step": 4745
    },
    {
      "entropy": 6.442700386047363,
      "epoch": 0.4748338081671415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004984105838829937,
      "loss": 6.3807,
      "mean_token_accuracy": 0.13865118771791457,
      "num_tokens": 9293377.0,
      "step": 4750
    },
    {
      "entropy": 6.459874391555786,
      "epoch": 0.4753336332283701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004984063464975116,
      "loss": 6.4445,
      "mean_token_accuracy": 0.1293555311858654,
      "num_tokens": 9302647.0,
      "step": 4755
    },
    {
      "entropy": 6.4613429546356205,
      "epoch": 0.47583345828959867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984021034911833,
      "loss": 6.4217,
      "mean_token_accuracy": 0.13532638251781465,
      "num_tokens": 9312153.0,
      "step": 4760
    },
    {
      "entropy": 6.47735013961792,
      "epoch": 0.4763332833508272,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983978548641154,
      "loss": 6.3134,
      "mean_token_accuracy": 0.14263757467269897,
      "num_tokens": 9320537.0,
      "step": 4765
    },
    {
      "entropy": 6.332884979248047,
      "epoch": 0.4768331084120558,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004983936006164151,
      "loss": 6.2426,
      "mean_token_accuracy": 0.1459529921412468,
      "num_tokens": 9330193.0,
      "step": 4770
    },
    {
      "entropy": 6.449419450759888,
      "epoch": 0.47733293347328437,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983893407481892,
      "loss": 6.435,
      "mean_token_accuracy": 0.13397538512945176,
      "num_tokens": 9340377.0,
      "step": 4775
    },
    {
      "entropy": 6.510228490829467,
      "epoch": 0.4778327585345129,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004983850752595451,
      "loss": 6.3799,
      "mean_token_accuracy": 0.14320830255746841,
      "num_tokens": 9349773.0,
      "step": 4780
    },
    {
      "entropy": 6.359656047821045,
      "epoch": 0.4783325835957415,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004983808041505898,
      "loss": 6.2939,
      "mean_token_accuracy": 0.1384284809231758,
      "num_tokens": 9359844.0,
      "step": 4785
    },
    {
      "entropy": 6.439452171325684,
      "epoch": 0.4788324086569701,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983765274214311,
      "loss": 6.3622,
      "mean_token_accuracy": 0.1418504759669304,
      "num_tokens": 9369744.0,
      "step": 4790
    },
    {
      "entropy": 6.43484935760498,
      "epoch": 0.4793322337181986,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983722450721764,
      "loss": 6.3334,
      "mean_token_accuracy": 0.14106698855757713,
      "num_tokens": 9378489.0,
      "step": 4795
    },
    {
      "entropy": 6.484389543533325,
      "epoch": 0.4798320587794272,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004983679571029336,
      "loss": 6.414,
      "mean_token_accuracy": 0.13530160635709762,
      "num_tokens": 9388833.0,
      "step": 4800
    },
    {
      "entropy": 6.4622142791748045,
      "epoch": 0.4803318838406558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004983636635138103,
      "loss": 6.3119,
      "mean_token_accuracy": 0.13595396801829338,
      "num_tokens": 9398010.0,
      "step": 4805
    },
    {
      "entropy": 6.449434852600097,
      "epoch": 0.4808317089018843,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004983593643049148,
      "loss": 6.4273,
      "mean_token_accuracy": 0.13126106932759285,
      "num_tokens": 9407847.0,
      "step": 4810
    },
    {
      "entropy": 6.350287389755249,
      "epoch": 0.4813315339631129,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983550594763552,
      "loss": 6.3264,
      "mean_token_accuracy": 0.14140709340572358,
      "num_tokens": 9416965.0,
      "step": 4815
    },
    {
      "entropy": 6.443746519088745,
      "epoch": 0.4818313590243415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983507490282397,
      "loss": 6.3556,
      "mean_token_accuracy": 0.14113930761814117,
      "num_tokens": 9426646.0,
      "step": 4820
    },
    {
      "entropy": 6.38179612159729,
      "epoch": 0.4823311840855701,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004983464329606768,
      "loss": 6.3289,
      "mean_token_accuracy": 0.1313927136361599,
      "num_tokens": 9435961.0,
      "step": 4825
    },
    {
      "entropy": 6.337861633300781,
      "epoch": 0.4828310091467986,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983421112737752,
      "loss": 6.2879,
      "mean_token_accuracy": 0.13953153863549234,
      "num_tokens": 9446369.0,
      "step": 4830
    },
    {
      "entropy": 6.442879962921142,
      "epoch": 0.4833308342080272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983377839676435,
      "loss": 6.2985,
      "mean_token_accuracy": 0.14319235682487488,
      "num_tokens": 9456315.0,
      "step": 4835
    },
    {
      "entropy": 6.430643033981323,
      "epoch": 0.4838306592692558,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983334510423906,
      "loss": 6.4501,
      "mean_token_accuracy": 0.12999346926808358,
      "num_tokens": 9465500.0,
      "step": 4840
    },
    {
      "entropy": 6.400549650192261,
      "epoch": 0.4843304843304843,
      "grad_norm": 1.125,
      "learning_rate": 0.0004983291124981257,
      "loss": 6.3837,
      "mean_token_accuracy": 0.13492454662919046,
      "num_tokens": 9475437.0,
      "step": 4845
    },
    {
      "entropy": 6.4067586898803714,
      "epoch": 0.4848303093917129,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983247683349576,
      "loss": 6.3225,
      "mean_token_accuracy": 0.13367080986499785,
      "num_tokens": 9485178.0,
      "step": 4850
    },
    {
      "entropy": 6.431334733963013,
      "epoch": 0.4853301344529415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983204185529959,
      "loss": 6.3371,
      "mean_token_accuracy": 0.14342518001794816,
      "num_tokens": 9494666.0,
      "step": 4855
    },
    {
      "entropy": 6.348692464828491,
      "epoch": 0.48582995951417,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004983160631523499,
      "loss": 6.3897,
      "mean_token_accuracy": 0.14454202353954315,
      "num_tokens": 9504538.0,
      "step": 4860
    },
    {
      "entropy": 6.428907060623169,
      "epoch": 0.4863297845753986,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983117021331293,
      "loss": 6.3393,
      "mean_token_accuracy": 0.13695962801575662,
      "num_tokens": 9514660.0,
      "step": 4865
    },
    {
      "entropy": 6.366034317016601,
      "epoch": 0.4868296096366272,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004983073354954436,
      "loss": 6.3104,
      "mean_token_accuracy": 0.1393321968615055,
      "num_tokens": 9524734.0,
      "step": 4870
    },
    {
      "entropy": 6.414203691482544,
      "epoch": 0.4873294346978557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983029632394029,
      "loss": 6.2918,
      "mean_token_accuracy": 0.13608113825321197,
      "num_tokens": 9533469.0,
      "step": 4875
    },
    {
      "entropy": 6.409686326980591,
      "epoch": 0.4878292597590843,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498298585365117,
      "loss": 6.3447,
      "mean_token_accuracy": 0.13997287452220916,
      "num_tokens": 9543916.0,
      "step": 4880
    },
    {
      "entropy": 6.420151615142823,
      "epoch": 0.4883290848203129,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004982942018726963,
      "loss": 6.2519,
      "mean_token_accuracy": 0.1406138740479946,
      "num_tokens": 9553767.0,
      "step": 4885
    },
    {
      "entropy": 6.368744611740112,
      "epoch": 0.4888289098815415,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982898127622508,
      "loss": 6.363,
      "mean_token_accuracy": 0.13477477356791495,
      "num_tokens": 9564713.0,
      "step": 4890
    },
    {
      "entropy": 6.306026458740234,
      "epoch": 0.48932873494277,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982854180338911,
      "loss": 6.425,
      "mean_token_accuracy": 0.13270416334271432,
      "num_tokens": 9576216.0,
      "step": 4895
    },
    {
      "entropy": 6.51512393951416,
      "epoch": 0.4898285600039986,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004982810176877279,
      "loss": 6.3945,
      "mean_token_accuracy": 0.1362947069108486,
      "num_tokens": 9586428.0,
      "step": 4900
    },
    {
      "entropy": 6.5169775009155275,
      "epoch": 0.4903283850652272,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004982766117238716,
      "loss": 6.3987,
      "mean_token_accuracy": 0.1350581780076027,
      "num_tokens": 9596548.0,
      "step": 4905
    },
    {
      "entropy": 6.353964471817017,
      "epoch": 0.4908282101264557,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982722001424331,
      "loss": 6.2931,
      "mean_token_accuracy": 0.14037388265132905,
      "num_tokens": 9605190.0,
      "step": 4910
    },
    {
      "entropy": 6.390562725067139,
      "epoch": 0.4913280351876843,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982677829435236,
      "loss": 6.3129,
      "mean_token_accuracy": 0.13843245506286622,
      "num_tokens": 9614682.0,
      "step": 4915
    },
    {
      "entropy": 6.4096743106842045,
      "epoch": 0.4918278602489129,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982633601272543,
      "loss": 6.3433,
      "mean_token_accuracy": 0.1375841438770294,
      "num_tokens": 9625026.0,
      "step": 4920
    },
    {
      "entropy": 6.396756601333618,
      "epoch": 0.4923276853101414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982589316937361,
      "loss": 6.3168,
      "mean_token_accuracy": 0.13751595318317414,
      "num_tokens": 9634301.0,
      "step": 4925
    },
    {
      "entropy": 6.369342947006226,
      "epoch": 0.49282751037137,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982544976430808,
      "loss": 6.3356,
      "mean_token_accuracy": 0.14015720114111901,
      "num_tokens": 9645025.0,
      "step": 4930
    },
    {
      "entropy": 6.380672979354858,
      "epoch": 0.4933273354325986,
      "grad_norm": 1.125,
      "learning_rate": 0.0004982500579753995,
      "loss": 6.29,
      "mean_token_accuracy": 0.1400677368044853,
      "num_tokens": 9655595.0,
      "step": 4935
    },
    {
      "entropy": 6.404878568649292,
      "epoch": 0.49382716049382713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982456126908044,
      "loss": 6.324,
      "mean_token_accuracy": 0.1387333333492279,
      "num_tokens": 9664770.0,
      "step": 4940
    },
    {
      "entropy": 6.40715970993042,
      "epoch": 0.4943269855550557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004982411617894071,
      "loss": 6.3051,
      "mean_token_accuracy": 0.14008496552705765,
      "num_tokens": 9673820.0,
      "step": 4945
    },
    {
      "entropy": 6.374893140792847,
      "epoch": 0.4948268106162843,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004982367052713196,
      "loss": 6.3026,
      "mean_token_accuracy": 0.14693843573331833,
      "num_tokens": 9683087.0,
      "step": 4950
    },
    {
      "entropy": 6.4222311019897464,
      "epoch": 0.4953266356775129,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982322431366541,
      "loss": 6.4334,
      "mean_token_accuracy": 0.13418319970369338,
      "num_tokens": 9692798.0,
      "step": 4955
    },
    {
      "entropy": 6.406663036346435,
      "epoch": 0.4958264607387414,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004982277753855227,
      "loss": 6.2373,
      "mean_token_accuracy": 0.14051652699708939,
      "num_tokens": 9703062.0,
      "step": 4960
    },
    {
      "entropy": 6.375523376464844,
      "epoch": 0.49632628579997,
      "grad_norm": 1.203125,
      "learning_rate": 0.000498223302018038,
      "loss": 6.2379,
      "mean_token_accuracy": 0.1481524661183357,
      "num_tokens": 9713223.0,
      "step": 4965
    },
    {
      "entropy": 6.332384729385376,
      "epoch": 0.4968261108611986,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004982188230343124,
      "loss": 6.3424,
      "mean_token_accuracy": 0.13508185893297195,
      "num_tokens": 9722483.0,
      "step": 4970
    },
    {
      "entropy": 6.358487367630005,
      "epoch": 0.49732593592242713,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982143384344587,
      "loss": 6.3119,
      "mean_token_accuracy": 0.13788269087672234,
      "num_tokens": 9732350.0,
      "step": 4975
    },
    {
      "entropy": 6.395327758789063,
      "epoch": 0.4978257609836557,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982098482185897,
      "loss": 6.3528,
      "mean_token_accuracy": 0.14036380276083946,
      "num_tokens": 9741824.0,
      "step": 4980
    },
    {
      "entropy": 6.3652692317962645,
      "epoch": 0.4983255860448843,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004982053523868182,
      "loss": 6.2719,
      "mean_token_accuracy": 0.14404890984296798,
      "num_tokens": 9751428.0,
      "step": 4985
    },
    {
      "entropy": 6.310924386978149,
      "epoch": 0.49882541110611284,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004982008509392576,
      "loss": 6.2074,
      "mean_token_accuracy": 0.14892275482416154,
      "num_tokens": 9761430.0,
      "step": 4990
    },
    {
      "entropy": 6.452079343795776,
      "epoch": 0.4993252361673414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000498196343876021,
      "loss": 6.4267,
      "mean_token_accuracy": 0.13413360863924026,
      "num_tokens": 9771540.0,
      "step": 4995
    },
    {
      "entropy": 6.431906127929688,
      "epoch": 0.49982506122857,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981918311972219,
      "loss": 6.4045,
      "mean_token_accuracy": 0.13877223953604698,
      "num_tokens": 9781342.0,
      "step": 5000
    },
    {
      "entropy": 6.373316764831543,
      "epoch": 0.5003248862897985,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981873129029737,
      "loss": 6.3892,
      "mean_token_accuracy": 0.13581344187259675,
      "num_tokens": 9790313.0,
      "step": 5005
    },
    {
      "entropy": 6.478897619247436,
      "epoch": 0.5008247113510271,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049818278899339,
      "loss": 6.3285,
      "mean_token_accuracy": 0.1354820631444454,
      "num_tokens": 9799656.0,
      "step": 5010
    },
    {
      "entropy": 6.364614486694336,
      "epoch": 0.5013245364122557,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000498178259468585,
      "loss": 6.2763,
      "mean_token_accuracy": 0.14575451910495757,
      "num_tokens": 9808971.0,
      "step": 5015
    },
    {
      "entropy": 6.403961038589477,
      "epoch": 0.5018243614734843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004981737243286722,
      "loss": 6.3702,
      "mean_token_accuracy": 0.13478559702634813,
      "num_tokens": 9820397.0,
      "step": 5020
    },
    {
      "entropy": 6.418635606765747,
      "epoch": 0.5023241865347129,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981691835737661,
      "loss": 6.3514,
      "mean_token_accuracy": 0.1418923929333687,
      "num_tokens": 9829309.0,
      "step": 5025
    },
    {
      "entropy": 6.388303327560425,
      "epoch": 0.5028240115959414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981646372039807,
      "loss": 6.3412,
      "mean_token_accuracy": 0.13365231603384017,
      "num_tokens": 9838950.0,
      "step": 5030
    },
    {
      "entropy": 6.393351221084595,
      "epoch": 0.50332383665717,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004981600852194305,
      "loss": 6.2843,
      "mean_token_accuracy": 0.14295320436358452,
      "num_tokens": 9849613.0,
      "step": 5035
    },
    {
      "entropy": 6.47461256980896,
      "epoch": 0.5038236617183985,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00049815552762023,
      "loss": 6.4623,
      "mean_token_accuracy": 0.13026106879115104,
      "num_tokens": 9860842.0,
      "step": 5040
    },
    {
      "entropy": 6.38051700592041,
      "epoch": 0.5043234867796271,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004981509644064937,
      "loss": 6.3119,
      "mean_token_accuracy": 0.14273230060935022,
      "num_tokens": 9871936.0,
      "step": 5045
    },
    {
      "entropy": 6.317802238464355,
      "epoch": 0.5048233118408557,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981463955783367,
      "loss": 6.1781,
      "mean_token_accuracy": 0.14730376452207566,
      "num_tokens": 9882142.0,
      "step": 5050
    },
    {
      "entropy": 6.355389022827149,
      "epoch": 0.5053231369020843,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981418211358736,
      "loss": 6.3148,
      "mean_token_accuracy": 0.13774936497211457,
      "num_tokens": 9892713.0,
      "step": 5055
    },
    {
      "entropy": 6.464467573165893,
      "epoch": 0.5058229619633129,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981372410792199,
      "loss": 6.3614,
      "mean_token_accuracy": 0.13985327184200286,
      "num_tokens": 9901706.0,
      "step": 5060
    },
    {
      "entropy": 6.319796705245972,
      "epoch": 0.5063227870245414,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004981326554084906,
      "loss": 6.3158,
      "mean_token_accuracy": 0.13560432568192482,
      "num_tokens": 9912444.0,
      "step": 5065
    },
    {
      "entropy": 6.473723316192627,
      "epoch": 0.50682261208577,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981280641238011,
      "loss": 6.3041,
      "mean_token_accuracy": 0.1340654283761978,
      "num_tokens": 9921593.0,
      "step": 5070
    },
    {
      "entropy": 6.3504386901855465,
      "epoch": 0.5073224371469985,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981234672252669,
      "loss": 6.2619,
      "mean_token_accuracy": 0.14589202776551247,
      "num_tokens": 9931362.0,
      "step": 5075
    },
    {
      "entropy": 6.436716794967651,
      "epoch": 0.5078222622082271,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004981188647130037,
      "loss": 6.4004,
      "mean_token_accuracy": 0.13545019924640656,
      "num_tokens": 9941877.0,
      "step": 5080
    },
    {
      "entropy": 6.523155593872071,
      "epoch": 0.5083220872694557,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004981142565871272,
      "loss": 6.4583,
      "mean_token_accuracy": 0.1312290236353874,
      "num_tokens": 9952516.0,
      "step": 5085
    },
    {
      "entropy": 6.516927003860474,
      "epoch": 0.5088219123306843,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004981096428477535,
      "loss": 6.5212,
      "mean_token_accuracy": 0.12833143249154091,
      "num_tokens": 9963078.0,
      "step": 5090
    },
    {
      "entropy": 6.435417222976684,
      "epoch": 0.5093217373919128,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004981050234949986,
      "loss": 6.3992,
      "mean_token_accuracy": 0.13014468029141427,
      "num_tokens": 9972816.0,
      "step": 5095
    },
    {
      "entropy": 6.428530311584472,
      "epoch": 0.5098215624531414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981003985289786,
      "loss": 6.3401,
      "mean_token_accuracy": 0.1345435470342636,
      "num_tokens": 9982773.0,
      "step": 5100
    },
    {
      "entropy": 6.463855504989624,
      "epoch": 0.51032138751437,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980957679498101,
      "loss": 6.3402,
      "mean_token_accuracy": 0.14080399125814438,
      "num_tokens": 9992602.0,
      "step": 5105
    },
    {
      "entropy": 6.380956983566284,
      "epoch": 0.5108212125755985,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004980911317576095,
      "loss": 6.2909,
      "mean_token_accuracy": 0.13896886706352235,
      "num_tokens": 10002185.0,
      "step": 5110
    },
    {
      "entropy": 6.353172016143799,
      "epoch": 0.5113210376368271,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980864899524933,
      "loss": 6.3074,
      "mean_token_accuracy": 0.13692167326807975,
      "num_tokens": 10012028.0,
      "step": 5115
    },
    {
      "entropy": 6.386081981658935,
      "epoch": 0.5118208626980557,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004980818425345786,
      "loss": 6.3902,
      "mean_token_accuracy": 0.13124118521809577,
      "num_tokens": 10021914.0,
      "step": 5120
    },
    {
      "entropy": 6.461355495452881,
      "epoch": 0.5123206877592843,
      "grad_norm": 1.125,
      "learning_rate": 0.000498077189503982,
      "loss": 6.3361,
      "mean_token_accuracy": 0.13988266214728357,
      "num_tokens": 10031246.0,
      "step": 5125
    },
    {
      "entropy": 6.458877420425415,
      "epoch": 0.5128205128205128,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004980725308608208,
      "loss": 6.3474,
      "mean_token_accuracy": 0.13762072920799256,
      "num_tokens": 10042064.0,
      "step": 5130
    },
    {
      "entropy": 6.271750450134277,
      "epoch": 0.5133203378817414,
      "grad_norm": 2.421875,
      "learning_rate": 0.0004980678666052121,
      "loss": 6.1446,
      "mean_token_accuracy": 0.15244997069239616,
      "num_tokens": 10052943.0,
      "step": 5135
    },
    {
      "entropy": 6.392112350463867,
      "epoch": 0.51382016294297,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980631967372733,
      "loss": 6.3411,
      "mean_token_accuracy": 0.13862255215644836,
      "num_tokens": 10062497.0,
      "step": 5140
    },
    {
      "entropy": 6.418827819824219,
      "epoch": 0.5143199880041985,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004980585212571218,
      "loss": 6.3193,
      "mean_token_accuracy": 0.1380971699953079,
      "num_tokens": 10072819.0,
      "step": 5145
    },
    {
      "entropy": 6.294079065322876,
      "epoch": 0.5148198130654271,
      "grad_norm": 1.5,
      "learning_rate": 0.0004980538401648753,
      "loss": 6.1923,
      "mean_token_accuracy": 0.14632125124335288,
      "num_tokens": 10081834.0,
      "step": 5150
    },
    {
      "entropy": 6.391872978210449,
      "epoch": 0.5153196381266557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980491534606517,
      "loss": 6.3049,
      "mean_token_accuracy": 0.14436430484056473,
      "num_tokens": 10090746.0,
      "step": 5155
    },
    {
      "entropy": 6.419670915603637,
      "epoch": 0.5158194631878842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004980444611445686,
      "loss": 6.3202,
      "mean_token_accuracy": 0.13725758716464043,
      "num_tokens": 10101728.0,
      "step": 5160
    },
    {
      "entropy": 6.3820618152618405,
      "epoch": 0.5163192882491128,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004980397632167444,
      "loss": 6.295,
      "mean_token_accuracy": 0.14382118135690689,
      "num_tokens": 10111641.0,
      "step": 5165
    },
    {
      "entropy": 6.3883250713348385,
      "epoch": 0.5168191133103414,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004980350596772971,
      "loss": 6.2618,
      "mean_token_accuracy": 0.13175864443182944,
      "num_tokens": 10120708.0,
      "step": 5170
    },
    {
      "entropy": 6.331651115417481,
      "epoch": 0.51731893837157,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004980303505263451,
      "loss": 6.2804,
      "mean_token_accuracy": 0.14033057689666747,
      "num_tokens": 10131279.0,
      "step": 5175
    },
    {
      "entropy": 6.408781671524048,
      "epoch": 0.5178187634327985,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980256357640069,
      "loss": 6.1689,
      "mean_token_accuracy": 0.14727225303649902,
      "num_tokens": 10140329.0,
      "step": 5180
    },
    {
      "entropy": 6.4224334239959715,
      "epoch": 0.5183185884940271,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004980209153904011,
      "loss": 6.3868,
      "mean_token_accuracy": 0.13183563128113746,
      "num_tokens": 10149658.0,
      "step": 5185
    },
    {
      "entropy": 6.244342088699341,
      "epoch": 0.5188184135552557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980161894056464,
      "loss": 6.1579,
      "mean_token_accuracy": 0.14717303663492204,
      "num_tokens": 10159131.0,
      "step": 5190
    },
    {
      "entropy": 6.354598093032837,
      "epoch": 0.5193182386164842,
      "grad_norm": 1.125,
      "learning_rate": 0.0004980114578098618,
      "loss": 6.3208,
      "mean_token_accuracy": 0.1378541111946106,
      "num_tokens": 10169094.0,
      "step": 5195
    },
    {
      "entropy": 6.380809259414673,
      "epoch": 0.5198180636777128,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980067206031663,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1478392869234085,
      "num_tokens": 10177910.0,
      "step": 5200
    },
    {
      "entropy": 6.31223578453064,
      "epoch": 0.5203178887389414,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980019777856791,
      "loss": 6.2835,
      "mean_token_accuracy": 0.14610777646303177,
      "num_tokens": 10186801.0,
      "step": 5205
    },
    {
      "entropy": 6.4225239753723145,
      "epoch": 0.52081771380017,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004979972293575195,
      "loss": 6.3591,
      "mean_token_accuracy": 0.13425822481513022,
      "num_tokens": 10196841.0,
      "step": 5210
    },
    {
      "entropy": 6.481344985961914,
      "epoch": 0.5213175388613985,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979924753188069,
      "loss": 6.4116,
      "mean_token_accuracy": 0.12858324572443963,
      "num_tokens": 10207829.0,
      "step": 5215
    },
    {
      "entropy": 6.377887821197509,
      "epoch": 0.5218173639226271,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004979877156696612,
      "loss": 6.4059,
      "mean_token_accuracy": 0.13356024473905564,
      "num_tokens": 10217029.0,
      "step": 5220
    },
    {
      "entropy": 6.328592348098755,
      "epoch": 0.5223171889838556,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004979829504102018,
      "loss": 6.2687,
      "mean_token_accuracy": 0.1433059334754944,
      "num_tokens": 10227602.0,
      "step": 5225
    },
    {
      "entropy": 6.372079229354858,
      "epoch": 0.5228170140450842,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979781795405488,
      "loss": 6.275,
      "mean_token_accuracy": 0.1425568178296089,
      "num_tokens": 10237990.0,
      "step": 5230
    },
    {
      "entropy": 6.346520280838012,
      "epoch": 0.5233168391063128,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979734030608222,
      "loss": 6.3131,
      "mean_token_accuracy": 0.13744089677929877,
      "num_tokens": 10247381.0,
      "step": 5235
    },
    {
      "entropy": 6.333405017852783,
      "epoch": 0.5238166641675414,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979686209711422,
      "loss": 6.1594,
      "mean_token_accuracy": 0.15484359115362167,
      "num_tokens": 10256444.0,
      "step": 5240
    },
    {
      "entropy": 6.402522039413452,
      "epoch": 0.5243164892287699,
      "grad_norm": 1.3125,
      "learning_rate": 0.000497963833271629,
      "loss": 6.3168,
      "mean_token_accuracy": 0.13975608646869658,
      "num_tokens": 10266965.0,
      "step": 5245
    },
    {
      "entropy": 6.2881122589111325,
      "epoch": 0.5248163142899985,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004979590399624031,
      "loss": 6.2251,
      "mean_token_accuracy": 0.1437224604189396,
      "num_tokens": 10276520.0,
      "step": 5250
    },
    {
      "entropy": 6.347827291488647,
      "epoch": 0.5253161393512271,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004979542410435852,
      "loss": 6.3265,
      "mean_token_accuracy": 0.13439038172364234,
      "num_tokens": 10285676.0,
      "step": 5255
    },
    {
      "entropy": 6.307738018035889,
      "epoch": 0.5258159644124556,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004979494365152959,
      "loss": 6.2098,
      "mean_token_accuracy": 0.14656018018722533,
      "num_tokens": 10295643.0,
      "step": 5260
    },
    {
      "entropy": 6.382900142669678,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004979446263776562,
      "loss": 6.3212,
      "mean_token_accuracy": 0.13438362032175064,
      "num_tokens": 10305555.0,
      "step": 5265
    },
    {
      "entropy": 6.396904468536377,
      "epoch": 0.5268156145349128,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000497939810630787,
      "loss": 6.2857,
      "mean_token_accuracy": 0.14424057826399803,
      "num_tokens": 10315399.0,
      "step": 5270
    },
    {
      "entropy": 6.352623462677002,
      "epoch": 0.5273154395961414,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979349892748096,
      "loss": 6.3495,
      "mean_token_accuracy": 0.14324067085981368,
      "num_tokens": 10325414.0,
      "step": 5275
    },
    {
      "entropy": 6.366179132461548,
      "epoch": 0.5278152646573699,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004979301623098452,
      "loss": 6.2899,
      "mean_token_accuracy": 0.13897085860371589,
      "num_tokens": 10334701.0,
      "step": 5280
    },
    {
      "entropy": 6.351607131958008,
      "epoch": 0.5283150897185985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004979253297360153,
      "loss": 6.3915,
      "mean_token_accuracy": 0.13794035241007804,
      "num_tokens": 10343951.0,
      "step": 5285
    },
    {
      "entropy": 6.403583288192749,
      "epoch": 0.528814914779827,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004979204915534415,
      "loss": 6.2398,
      "mean_token_accuracy": 0.14738479256629944,
      "num_tokens": 10352990.0,
      "step": 5290
    },
    {
      "entropy": 6.409589195251465,
      "epoch": 0.5293147398410556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979156477622454,
      "loss": 6.3078,
      "mean_token_accuracy": 0.14044644087553024,
      "num_tokens": 10363059.0,
      "step": 5295
    },
    {
      "entropy": 6.379518127441406,
      "epoch": 0.5298145649022842,
      "grad_norm": 1.5,
      "learning_rate": 0.0004979107983625491,
      "loss": 6.2898,
      "mean_token_accuracy": 0.14207641482353212,
      "num_tokens": 10373237.0,
      "step": 5300
    },
    {
      "entropy": 6.4374383926391605,
      "epoch": 0.5303143899635128,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004979059433544744,
      "loss": 6.4078,
      "mean_token_accuracy": 0.12871226891875268,
      "num_tokens": 10383373.0,
      "step": 5305
    },
    {
      "entropy": 6.381015205383301,
      "epoch": 0.5308142150247414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979010827381435,
      "loss": 6.2786,
      "mean_token_accuracy": 0.1329106606543064,
      "num_tokens": 10392612.0,
      "step": 5310
    },
    {
      "entropy": 6.3076976299285885,
      "epoch": 0.5313140400859699,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978962165136787,
      "loss": 6.2015,
      "mean_token_accuracy": 0.14325112327933312,
      "num_tokens": 10402201.0,
      "step": 5315
    },
    {
      "entropy": 6.4017308235168455,
      "epoch": 0.5318138651471985,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004978913446812025,
      "loss": 6.25,
      "mean_token_accuracy": 0.14539417177438735,
      "num_tokens": 10411847.0,
      "step": 5320
    },
    {
      "entropy": 6.353742933273315,
      "epoch": 0.532313690208427,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978864672408374,
      "loss": 6.3578,
      "mean_token_accuracy": 0.13494831845164298,
      "num_tokens": 10422035.0,
      "step": 5325
    },
    {
      "entropy": 6.287677812576294,
      "epoch": 0.5328135152696556,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004978815841927061,
      "loss": 6.2718,
      "mean_token_accuracy": 0.14338641837239266,
      "num_tokens": 10432097.0,
      "step": 5330
    },
    {
      "entropy": 6.373471355438232,
      "epoch": 0.5333133403308842,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004978766955369315,
      "loss": 6.2775,
      "mean_token_accuracy": 0.1486929588019848,
      "num_tokens": 10442357.0,
      "step": 5335
    },
    {
      "entropy": 6.308533763885498,
      "epoch": 0.5338131653921128,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978718012736367,
      "loss": 6.2474,
      "mean_token_accuracy": 0.14917047917842866,
      "num_tokens": 10453491.0,
      "step": 5340
    },
    {
      "entropy": 6.325512981414795,
      "epoch": 0.5343129904533414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004978669014029446,
      "loss": 6.246,
      "mean_token_accuracy": 0.14361390322446824,
      "num_tokens": 10464016.0,
      "step": 5345
    },
    {
      "entropy": 6.327845525741577,
      "epoch": 0.5348128155145699,
      "grad_norm": 1.25,
      "learning_rate": 0.0004978619959249787,
      "loss": 6.2694,
      "mean_token_accuracy": 0.1354649394750595,
      "num_tokens": 10473471.0,
      "step": 5350
    },
    {
      "entropy": 6.323083877563477,
      "epoch": 0.5353126405757984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004978570848398623,
      "loss": 6.2602,
      "mean_token_accuracy": 0.1403007134795189,
      "num_tokens": 10482329.0,
      "step": 5355
    },
    {
      "entropy": 6.404816579818726,
      "epoch": 0.535812465637027,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000497852168147719,
      "loss": 6.3724,
      "mean_token_accuracy": 0.13451690077781678,
      "num_tokens": 10491731.0,
      "step": 5360
    },
    {
      "entropy": 6.4853349208831785,
      "epoch": 0.5363122906982556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004978472458486724,
      "loss": 6.3535,
      "mean_token_accuracy": 0.14545586705207825,
      "num_tokens": 10501317.0,
      "step": 5365
    },
    {
      "entropy": 6.416453123092651,
      "epoch": 0.5368121157594842,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004978423179428466,
      "loss": 6.251,
      "mean_token_accuracy": 0.13703420087695123,
      "num_tokens": 10509849.0,
      "step": 5370
    },
    {
      "entropy": 6.276719284057617,
      "epoch": 0.5373119408207128,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004978373844303653,
      "loss": 6.2815,
      "mean_token_accuracy": 0.13624661713838576,
      "num_tokens": 10518714.0,
      "step": 5375
    },
    {
      "entropy": 6.394107484817505,
      "epoch": 0.5378117658819414,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004978324453113528,
      "loss": 6.3032,
      "mean_token_accuracy": 0.13700484335422516,
      "num_tokens": 10529128.0,
      "step": 5380
    },
    {
      "entropy": 6.422897434234619,
      "epoch": 0.5383115909431699,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978275005859333,
      "loss": 6.2455,
      "mean_token_accuracy": 0.1425884783267975,
      "num_tokens": 10538748.0,
      "step": 5385
    },
    {
      "entropy": 6.396283054351807,
      "epoch": 0.5388114160043984,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978225502542313,
      "loss": 6.2887,
      "mean_token_accuracy": 0.1390642985701561,
      "num_tokens": 10549273.0,
      "step": 5390
    },
    {
      "entropy": 6.3315589904785154,
      "epoch": 0.539311241065627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978175943163712,
      "loss": 6.2396,
      "mean_token_accuracy": 0.1445097140967846,
      "num_tokens": 10559332.0,
      "step": 5395
    },
    {
      "entropy": 6.282561540603638,
      "epoch": 0.5398110661268556,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004978126327724777,
      "loss": 6.2356,
      "mean_token_accuracy": 0.14473070949316025,
      "num_tokens": 10568699.0,
      "step": 5400
    },
    {
      "entropy": 6.314286851882935,
      "epoch": 0.5403108911880842,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004978076656226758,
      "loss": 6.2113,
      "mean_token_accuracy": 0.14579913020133972,
      "num_tokens": 10579129.0,
      "step": 5405
    },
    {
      "entropy": 6.3720090866088865,
      "epoch": 0.5408107162493128,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978026928670902,
      "loss": 6.2979,
      "mean_token_accuracy": 0.13803005293011666,
      "num_tokens": 10589093.0,
      "step": 5410
    },
    {
      "entropy": 6.279857921600342,
      "epoch": 0.5413105413105413,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004977977145058463,
      "loss": 6.3073,
      "mean_token_accuracy": 0.1435902953147888,
      "num_tokens": 10598766.0,
      "step": 5415
    },
    {
      "entropy": 6.291266632080078,
      "epoch": 0.5418103663717699,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004977927305390693,
      "loss": 6.2365,
      "mean_token_accuracy": 0.1501827135682106,
      "num_tokens": 10609148.0,
      "step": 5420
    },
    {
      "entropy": 6.486551523208618,
      "epoch": 0.5423101914329984,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004977877409668845,
      "loss": 6.3957,
      "mean_token_accuracy": 0.13328467383980752,
      "num_tokens": 10618160.0,
      "step": 5425
    },
    {
      "entropy": 6.432691669464111,
      "epoch": 0.542810016494227,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004977827457894174,
      "loss": 6.3045,
      "mean_token_accuracy": 0.13389546573162078,
      "num_tokens": 10628839.0,
      "step": 5430
    },
    {
      "entropy": 6.345345592498779,
      "epoch": 0.5433098415554556,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977777450067937,
      "loss": 6.2925,
      "mean_token_accuracy": 0.14172665700316428,
      "num_tokens": 10638312.0,
      "step": 5435
    },
    {
      "entropy": 6.414798212051392,
      "epoch": 0.5438096666166842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977727386191393,
      "loss": 6.2874,
      "mean_token_accuracy": 0.14041476994752883,
      "num_tokens": 10647128.0,
      "step": 5440
    },
    {
      "entropy": 6.365611410140991,
      "epoch": 0.5443094916779128,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004977677266265802,
      "loss": 6.4301,
      "mean_token_accuracy": 0.13062752410769463,
      "num_tokens": 10657574.0,
      "step": 5445
    },
    {
      "entropy": 6.3441002368927,
      "epoch": 0.5448093167391413,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977627090292424,
      "loss": 6.2888,
      "mean_token_accuracy": 0.1356620594859123,
      "num_tokens": 10668497.0,
      "step": 5450
    },
    {
      "entropy": 6.395407247543335,
      "epoch": 0.5453091418003698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977576858272521,
      "loss": 6.3004,
      "mean_token_accuracy": 0.13830969408154486,
      "num_tokens": 10678728.0,
      "step": 5455
    },
    {
      "entropy": 6.3671125888824465,
      "epoch": 0.5458089668615984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977526570207358,
      "loss": 6.2994,
      "mean_token_accuracy": 0.14166254475712775,
      "num_tokens": 10688953.0,
      "step": 5460
    },
    {
      "entropy": 6.2921223640441895,
      "epoch": 0.546308791922827,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004977476226098199,
      "loss": 6.2684,
      "mean_token_accuracy": 0.1459941491484642,
      "num_tokens": 10698570.0,
      "step": 5465
    },
    {
      "entropy": 6.3392229080200195,
      "epoch": 0.5468086169840556,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004977425825946312,
      "loss": 6.1762,
      "mean_token_accuracy": 0.14860441982746125,
      "num_tokens": 10708438.0,
      "step": 5470
    },
    {
      "entropy": 6.4341628074646,
      "epoch": 0.5473084420452842,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977375369752964,
      "loss": 6.2609,
      "mean_token_accuracy": 0.14495741203427315,
      "num_tokens": 10718559.0,
      "step": 5475
    },
    {
      "entropy": 6.286415147781372,
      "epoch": 0.5478082671065128,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004977324857519425,
      "loss": 6.3602,
      "mean_token_accuracy": 0.13685523271560668,
      "num_tokens": 10729348.0,
      "step": 5480
    },
    {
      "entropy": 6.421134948730469,
      "epoch": 0.5483080921677413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977274289246967,
      "loss": 6.3455,
      "mean_token_accuracy": 0.13599513098597527,
      "num_tokens": 10739139.0,
      "step": 5485
    },
    {
      "entropy": 6.415587854385376,
      "epoch": 0.5488079172289698,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977223664936859,
      "loss": 6.3797,
      "mean_token_accuracy": 0.1356613539159298,
      "num_tokens": 10749451.0,
      "step": 5490
    },
    {
      "entropy": 6.410269927978516,
      "epoch": 0.5493077422901984,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004977172984590378,
      "loss": 6.2272,
      "mean_token_accuracy": 0.15208979845046997,
      "num_tokens": 10757990.0,
      "step": 5495
    },
    {
      "entropy": 6.378470230102539,
      "epoch": 0.549807567351427,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977122248208798,
      "loss": 6.1956,
      "mean_token_accuracy": 0.14191931933164598,
      "num_tokens": 10766825.0,
      "step": 5500
    },
    {
      "entropy": 6.288527774810791,
      "epoch": 0.5503073924126556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004977071455793394,
      "loss": 6.2482,
      "mean_token_accuracy": 0.14223040863871575,
      "num_tokens": 10776670.0,
      "step": 5505
    },
    {
      "entropy": 6.274076175689697,
      "epoch": 0.5508072174738842,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977020607345445,
      "loss": 6.2098,
      "mean_token_accuracy": 0.14522653967142105,
      "num_tokens": 10787152.0,
      "step": 5510
    },
    {
      "entropy": 6.348708152770996,
      "epoch": 0.5513070425351128,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004976969702866231,
      "loss": 6.3497,
      "mean_token_accuracy": 0.13757041245698928,
      "num_tokens": 10796903.0,
      "step": 5515
    },
    {
      "entropy": 6.38208498954773,
      "epoch": 0.5518068675963412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004976918742357033,
      "loss": 6.3107,
      "mean_token_accuracy": 0.14124908968806266,
      "num_tokens": 10806549.0,
      "step": 5520
    },
    {
      "entropy": 6.469176721572876,
      "epoch": 0.5523066926575698,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976867725819133,
      "loss": 6.4244,
      "mean_token_accuracy": 0.13139872923493384,
      "num_tokens": 10816755.0,
      "step": 5525
    },
    {
      "entropy": 6.373313665390015,
      "epoch": 0.5528065177187984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004976816653253813,
      "loss": 6.2561,
      "mean_token_accuracy": 0.13657409474253654,
      "num_tokens": 10827171.0,
      "step": 5530
    },
    {
      "entropy": 6.3197403907775875,
      "epoch": 0.553306342780027,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976765524662359,
      "loss": 6.2649,
      "mean_token_accuracy": 0.13727437779307367,
      "num_tokens": 10837048.0,
      "step": 5535
    },
    {
      "entropy": 6.410451650619507,
      "epoch": 0.5538061678412556,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004976714340046058,
      "loss": 6.279,
      "mean_token_accuracy": 0.14013988599181176,
      "num_tokens": 10846775.0,
      "step": 5540
    },
    {
      "entropy": 6.313901424407959,
      "epoch": 0.5543059929024842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004976663099406196,
      "loss": 6.2947,
      "mean_token_accuracy": 0.14201875254511834,
      "num_tokens": 10856234.0,
      "step": 5545
    },
    {
      "entropy": 6.303532695770263,
      "epoch": 0.5548058179637128,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004976611802744066,
      "loss": 6.3143,
      "mean_token_accuracy": 0.13508880361914635,
      "num_tokens": 10865943.0,
      "step": 5550
    },
    {
      "entropy": 6.40221095085144,
      "epoch": 0.5553056430249412,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004976560450060954,
      "loss": 6.2205,
      "mean_token_accuracy": 0.14128334447741508,
      "num_tokens": 10875286.0,
      "step": 5555
    },
    {
      "entropy": 6.408346271514892,
      "epoch": 0.5558054680861698,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976509041358154,
      "loss": 6.3258,
      "mean_token_accuracy": 0.13746749609708786,
      "num_tokens": 10884491.0,
      "step": 5560
    },
    {
      "entropy": 6.3140397548675535,
      "epoch": 0.5563052931473984,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497645757663696,
      "loss": 6.1862,
      "mean_token_accuracy": 0.1458214282989502,
      "num_tokens": 10892773.0,
      "step": 5565
    },
    {
      "entropy": 6.284867763519287,
      "epoch": 0.556805118208627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976406055898665,
      "loss": 6.3372,
      "mean_token_accuracy": 0.13722846955060958,
      "num_tokens": 10903517.0,
      "step": 5570
    },
    {
      "entropy": 6.359335470199585,
      "epoch": 0.5573049432698556,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004976354479144568,
      "loss": 6.3074,
      "mean_token_accuracy": 0.14492030292749405,
      "num_tokens": 10913075.0,
      "step": 5575
    },
    {
      "entropy": 6.410849142074585,
      "epoch": 0.5578047683310842,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976302846375965,
      "loss": 6.3111,
      "mean_token_accuracy": 0.14278343841433525,
      "num_tokens": 10922962.0,
      "step": 5580
    },
    {
      "entropy": 6.331055402755737,
      "epoch": 0.5583045933923126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976251157594155,
      "loss": 6.2352,
      "mean_token_accuracy": 0.13862547650933266,
      "num_tokens": 10932192.0,
      "step": 5585
    },
    {
      "entropy": 6.327994537353516,
      "epoch": 0.5588044184535412,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004976199412800437,
      "loss": 6.233,
      "mean_token_accuracy": 0.14954536631703377,
      "num_tokens": 10941748.0,
      "step": 5590
    },
    {
      "entropy": 6.407844018936157,
      "epoch": 0.5593042435147698,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976147611996118,
      "loss": 6.4087,
      "mean_token_accuracy": 0.13964784294366836,
      "num_tokens": 10951467.0,
      "step": 5595
    },
    {
      "entropy": 6.408154582977295,
      "epoch": 0.5598040685759984,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004976095755182496,
      "loss": 6.3147,
      "mean_token_accuracy": 0.1395675592124462,
      "num_tokens": 10962598.0,
      "step": 5600
    },
    {
      "entropy": 6.336530065536499,
      "epoch": 0.560303893637227,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004976043842360877,
      "loss": 6.2271,
      "mean_token_accuracy": 0.14004328548908235,
      "num_tokens": 10971206.0,
      "step": 5605
    },
    {
      "entropy": 6.3032941818237305,
      "epoch": 0.5608037186984556,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497599187353257,
      "loss": 6.2015,
      "mean_token_accuracy": 0.1449039548635483,
      "num_tokens": 10979922.0,
      "step": 5610
    },
    {
      "entropy": 6.278577184677124,
      "epoch": 0.5613035437596842,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004975939848698879,
      "loss": 6.2273,
      "mean_token_accuracy": 0.14664379358291627,
      "num_tokens": 10988886.0,
      "step": 5615
    },
    {
      "entropy": 6.238605070114136,
      "epoch": 0.5618033688209126,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004975887767861113,
      "loss": 6.2388,
      "mean_token_accuracy": 0.14862425327301027,
      "num_tokens": 10998570.0,
      "step": 5620
    },
    {
      "entropy": 6.3335308074951175,
      "epoch": 0.5623031938821412,
      "grad_norm": 1.125,
      "learning_rate": 0.0004975835631020585,
      "loss": 6.2014,
      "mean_token_accuracy": 0.1449519529938698,
      "num_tokens": 11009549.0,
      "step": 5625
    },
    {
      "entropy": 6.194306516647339,
      "epoch": 0.5628030189433698,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004975783438178605,
      "loss": 6.0879,
      "mean_token_accuracy": 0.16339403092861177,
      "num_tokens": 11019382.0,
      "step": 5630
    },
    {
      "entropy": 6.369560241699219,
      "epoch": 0.5633028440045984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004975731189336487,
      "loss": 6.3004,
      "mean_token_accuracy": 0.1338556043803692,
      "num_tokens": 11028205.0,
      "step": 5635
    },
    {
      "entropy": 6.33437032699585,
      "epoch": 0.563802669065827,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004975678884495544,
      "loss": 6.2262,
      "mean_token_accuracy": 0.14442106038331987,
      "num_tokens": 11037684.0,
      "step": 5640
    },
    {
      "entropy": 6.3031786441802975,
      "epoch": 0.5643024941270556,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004975626523657093,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14175864383578302,
      "num_tokens": 11047119.0,
      "step": 5645
    },
    {
      "entropy": 6.256673526763916,
      "epoch": 0.564802319188284,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004975574106822451,
      "loss": 6.2063,
      "mean_token_accuracy": 0.14650952890515329,
      "num_tokens": 11057362.0,
      "step": 5650
    },
    {
      "entropy": 6.294582557678223,
      "epoch": 0.5653021442495126,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004975521633992938,
      "loss": 6.1823,
      "mean_token_accuracy": 0.14804719835519792,
      "num_tokens": 11067685.0,
      "step": 5655
    },
    {
      "entropy": 6.336051607131958,
      "epoch": 0.5658019693107412,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004975469105169873,
      "loss": 6.1395,
      "mean_token_accuracy": 0.14505960121750833,
      "num_tokens": 11077468.0,
      "step": 5660
    },
    {
      "entropy": 6.357481622695923,
      "epoch": 0.5663017943719698,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004975416520354579,
      "loss": 6.3549,
      "mean_token_accuracy": 0.13785886764526367,
      "num_tokens": 11087266.0,
      "step": 5665
    },
    {
      "entropy": 6.328675174713135,
      "epoch": 0.5668016194331984,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975363879548377,
      "loss": 6.256,
      "mean_token_accuracy": 0.1395912855863571,
      "num_tokens": 11098053.0,
      "step": 5670
    },
    {
      "entropy": 6.477372121810913,
      "epoch": 0.567301444494427,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004975311182752592,
      "loss": 6.2949,
      "mean_token_accuracy": 0.14179019629955292,
      "num_tokens": 11107341.0,
      "step": 5675
    },
    {
      "entropy": 6.266103649139405,
      "epoch": 0.5678012695556556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004975258429968551,
      "loss": 6.243,
      "mean_token_accuracy": 0.13855696991086006,
      "num_tokens": 11116356.0,
      "step": 5680
    },
    {
      "entropy": 6.2798848152160645,
      "epoch": 0.568301094616884,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975205621197581,
      "loss": 6.2498,
      "mean_token_accuracy": 0.1494542546570301,
      "num_tokens": 11126504.0,
      "step": 5685
    },
    {
      "entropy": 6.308255290985107,
      "epoch": 0.5688009196781126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004975152756441009,
      "loss": 6.206,
      "mean_token_accuracy": 0.14534769654273988,
      "num_tokens": 11136473.0,
      "step": 5690
    },
    {
      "entropy": 6.347431612014771,
      "epoch": 0.5693007447393412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975099835700166,
      "loss": 6.2604,
      "mean_token_accuracy": 0.14021557867527007,
      "num_tokens": 11146798.0,
      "step": 5695
    },
    {
      "entropy": 6.342993307113647,
      "epoch": 0.5698005698005698,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004975046858976386,
      "loss": 6.2729,
      "mean_token_accuracy": 0.13971469178795815,
      "num_tokens": 11156833.0,
      "step": 5700
    },
    {
      "entropy": 6.3537538051605225,
      "epoch": 0.5703003948617984,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004974993826270997,
      "loss": 6.3185,
      "mean_token_accuracy": 0.13555603325366974,
      "num_tokens": 11165229.0,
      "step": 5705
    },
    {
      "entropy": 6.27754487991333,
      "epoch": 0.570800219923027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004974940737585337,
      "loss": 6.1545,
      "mean_token_accuracy": 0.14630194306373595,
      "num_tokens": 11174126.0,
      "step": 5710
    },
    {
      "entropy": 6.306702136993408,
      "epoch": 0.5713000449842555,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000497488759292074,
      "loss": 6.2216,
      "mean_token_accuracy": 0.14706498831510545,
      "num_tokens": 11183558.0,
      "step": 5715
    },
    {
      "entropy": 6.324210596084595,
      "epoch": 0.571799870045484,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004974834392278545,
      "loss": 6.313,
      "mean_token_accuracy": 0.134626804292202,
      "num_tokens": 11193449.0,
      "step": 5720
    },
    {
      "entropy": 6.4277262687683105,
      "epoch": 0.5722996951067126,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004974781135660088,
      "loss": 6.3173,
      "mean_token_accuracy": 0.13738598600029944,
      "num_tokens": 11203805.0,
      "step": 5725
    },
    {
      "entropy": 6.226198673248291,
      "epoch": 0.5727995201679412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004974727823066709,
      "loss": 6.1479,
      "mean_token_accuracy": 0.15619017332792282,
      "num_tokens": 11213578.0,
      "step": 5730
    },
    {
      "entropy": 6.3234227180480955,
      "epoch": 0.5732993452291698,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004974674454499753,
      "loss": 6.3744,
      "mean_token_accuracy": 0.13196191787719727,
      "num_tokens": 11223398.0,
      "step": 5735
    },
    {
      "entropy": 6.455641412734986,
      "epoch": 0.5737991702903984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004974621029960559,
      "loss": 6.4323,
      "mean_token_accuracy": 0.13141667544841767,
      "num_tokens": 11232947.0,
      "step": 5740
    },
    {
      "entropy": 6.352408027648925,
      "epoch": 0.574298995351627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004974567549450472,
      "loss": 6.1862,
      "mean_token_accuracy": 0.14535416513681412,
      "num_tokens": 11242992.0,
      "step": 5745
    },
    {
      "entropy": 6.357608509063721,
      "epoch": 0.5747988204128555,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004974514012970839,
      "loss": 6.3868,
      "mean_token_accuracy": 0.13135262876749038,
      "num_tokens": 11252825.0,
      "step": 5750
    },
    {
      "entropy": 6.399218988418579,
      "epoch": 0.575298645474084,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974460420523005,
      "loss": 6.2631,
      "mean_token_accuracy": 0.14056848585605622,
      "num_tokens": 11262777.0,
      "step": 5755
    },
    {
      "entropy": 6.305263328552246,
      "epoch": 0.5757984705353126,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004974406772108319,
      "loss": 6.2804,
      "mean_token_accuracy": 0.136069355905056,
      "num_tokens": 11272828.0,
      "step": 5760
    },
    {
      "entropy": 6.388300228118896,
      "epoch": 0.5762982955965412,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004974353067728132,
      "loss": 6.2878,
      "mean_token_accuracy": 0.14289739355444908,
      "num_tokens": 11282290.0,
      "step": 5765
    },
    {
      "entropy": 6.443116092681885,
      "epoch": 0.5767981206577698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004974299307383794,
      "loss": 6.3035,
      "mean_token_accuracy": 0.1328974463045597,
      "num_tokens": 11290844.0,
      "step": 5770
    },
    {
      "entropy": 6.287966108322143,
      "epoch": 0.5772979457189984,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004974245491076656,
      "loss": 6.2193,
      "mean_token_accuracy": 0.14826769977808,
      "num_tokens": 11299243.0,
      "step": 5775
    },
    {
      "entropy": 6.34992470741272,
      "epoch": 0.5777977707802269,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974191618808075,
      "loss": 6.274,
      "mean_token_accuracy": 0.140966959297657,
      "num_tokens": 11308556.0,
      "step": 5780
    },
    {
      "entropy": 6.429300212860108,
      "epoch": 0.5782975958414555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004974137690579405,
      "loss": 6.3619,
      "mean_token_accuracy": 0.13782633990049362,
      "num_tokens": 11318172.0,
      "step": 5785
    },
    {
      "entropy": 6.39644422531128,
      "epoch": 0.578797420902684,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004974083706392004,
      "loss": 6.3271,
      "mean_token_accuracy": 0.13633953854441644,
      "num_tokens": 11328967.0,
      "step": 5790
    },
    {
      "entropy": 6.35618028640747,
      "epoch": 0.5792972459639126,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004974029666247228,
      "loss": 6.3201,
      "mean_token_accuracy": 0.13923141434788705,
      "num_tokens": 11339234.0,
      "step": 5795
    },
    {
      "entropy": 6.510340976715088,
      "epoch": 0.5797970710251412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973975570146438,
      "loss": 6.2889,
      "mean_token_accuracy": 0.13487068116664885,
      "num_tokens": 11349934.0,
      "step": 5800
    },
    {
      "entropy": 6.45050745010376,
      "epoch": 0.5802968960863698,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004973921418090997,
      "loss": 6.4028,
      "mean_token_accuracy": 0.135823817551136,
      "num_tokens": 11359817.0,
      "step": 5805
    },
    {
      "entropy": 6.265371370315552,
      "epoch": 0.5807967211475984,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004973867210082263,
      "loss": 6.1712,
      "mean_token_accuracy": 0.14345002099871634,
      "num_tokens": 11369565.0,
      "step": 5810
    },
    {
      "entropy": 6.341412734985352,
      "epoch": 0.5812965462088269,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004973812946121603,
      "loss": 6.2948,
      "mean_token_accuracy": 0.14010498225688933,
      "num_tokens": 11379543.0,
      "step": 5815
    },
    {
      "entropy": 6.3904783725738525,
      "epoch": 0.5817963712700555,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004973758626210382,
      "loss": 6.3213,
      "mean_token_accuracy": 0.1388237662613392,
      "num_tokens": 11388909.0,
      "step": 5820
    },
    {
      "entropy": 6.347397804260254,
      "epoch": 0.582296196331284,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004973704250349965,
      "loss": 6.3736,
      "mean_token_accuracy": 0.13767568171024322,
      "num_tokens": 11399774.0,
      "step": 5825
    },
    {
      "entropy": 6.298463916778564,
      "epoch": 0.5827960213925126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004973649818541722,
      "loss": 6.2333,
      "mean_token_accuracy": 0.1425513356924057,
      "num_tokens": 11408701.0,
      "step": 5830
    },
    {
      "entropy": 6.417697381973267,
      "epoch": 0.5832958464537412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004973595330787022,
      "loss": 6.2461,
      "mean_token_accuracy": 0.138995760679245,
      "num_tokens": 11418582.0,
      "step": 5835
    },
    {
      "entropy": 6.347245883941651,
      "epoch": 0.5837956715149698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004973540787087235,
      "loss": 6.3009,
      "mean_token_accuracy": 0.1366384357213974,
      "num_tokens": 11429155.0,
      "step": 5840
    },
    {
      "entropy": 6.350349235534668,
      "epoch": 0.5842954965761983,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004973486187443734,
      "loss": 6.1686,
      "mean_token_accuracy": 0.1489595115184784,
      "num_tokens": 11438798.0,
      "step": 5845
    },
    {
      "entropy": 6.3874915599822994,
      "epoch": 0.5847953216374269,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004973431531857893,
      "loss": 6.3274,
      "mean_token_accuracy": 0.1386117771267891,
      "num_tokens": 11449862.0,
      "step": 5850
    },
    {
      "entropy": 6.306324768066406,
      "epoch": 0.5852951466986555,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004973376820331087,
      "loss": 6.2912,
      "mean_token_accuracy": 0.13946930021047593,
      "num_tokens": 11459697.0,
      "step": 5855
    },
    {
      "entropy": 6.4414434909820555,
      "epoch": 0.585794971759884,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004973322052864692,
      "loss": 6.3999,
      "mean_token_accuracy": 0.1409497581422329,
      "num_tokens": 11468973.0,
      "step": 5860
    },
    {
      "entropy": 6.2925050258636475,
      "epoch": 0.5862947968211126,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004973267229460087,
      "loss": 6.2273,
      "mean_token_accuracy": 0.142069161683321,
      "num_tokens": 11479453.0,
      "step": 5865
    },
    {
      "entropy": 6.311680889129638,
      "epoch": 0.5867946218823412,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497321235011865,
      "loss": 6.2928,
      "mean_token_accuracy": 0.14333089515566827,
      "num_tokens": 11488642.0,
      "step": 5870
    },
    {
      "entropy": 6.421400690078736,
      "epoch": 0.5872944469435698,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004973157414841763,
      "loss": 6.1854,
      "mean_token_accuracy": 0.14281654059886933,
      "num_tokens": 11497898.0,
      "step": 5875
    },
    {
      "entropy": 6.307196378707886,
      "epoch": 0.5877942720047983,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004973102423630806,
      "loss": 6.2604,
      "mean_token_accuracy": 0.1453785739839077,
      "num_tokens": 11508337.0,
      "step": 5880
    },
    {
      "entropy": 6.320552492141724,
      "epoch": 0.5882940970660269,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004973047376487166,
      "loss": 6.3266,
      "mean_token_accuracy": 0.14145833104848862,
      "num_tokens": 11518045.0,
      "step": 5885
    },
    {
      "entropy": 6.376059913635254,
      "epoch": 0.5887939221272555,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004972992273412224,
      "loss": 6.2193,
      "mean_token_accuracy": 0.13730005100369452,
      "num_tokens": 11527297.0,
      "step": 5890
    },
    {
      "entropy": 6.3773809432983395,
      "epoch": 0.589293747188484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972937114407369,
      "loss": 6.2353,
      "mean_token_accuracy": 0.14300990998744964,
      "num_tokens": 11537674.0,
      "step": 5895
    },
    {
      "entropy": 6.382708740234375,
      "epoch": 0.5897935722497126,
      "grad_norm": 1.15625,
      "learning_rate": 0.000497288189947399,
      "loss": 6.2803,
      "mean_token_accuracy": 0.1405448392033577,
      "num_tokens": 11548046.0,
      "step": 5900
    },
    {
      "entropy": 6.318603086471557,
      "epoch": 0.5902933973109412,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972826628613472,
      "loss": 6.2734,
      "mean_token_accuracy": 0.1428924657404423,
      "num_tokens": 11558075.0,
      "step": 5905
    },
    {
      "entropy": 6.339267158508301,
      "epoch": 0.5907932223721697,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004972771301827209,
      "loss": 6.2709,
      "mean_token_accuracy": 0.14029436483979224,
      "num_tokens": 11567983.0,
      "step": 5910
    },
    {
      "entropy": 6.304966974258423,
      "epoch": 0.5912930474333983,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004972715919116592,
      "loss": 6.2587,
      "mean_token_accuracy": 0.1437355786561966,
      "num_tokens": 11577438.0,
      "step": 5915
    },
    {
      "entropy": 6.336782312393188,
      "epoch": 0.5917928724946269,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004972660480483015,
      "loss": 6.1674,
      "mean_token_accuracy": 0.14765796214342117,
      "num_tokens": 11586579.0,
      "step": 5920
    },
    {
      "entropy": 6.222731637954712,
      "epoch": 0.5922926975558555,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972604985927871,
      "loss": 6.2782,
      "mean_token_accuracy": 0.1422460690140724,
      "num_tokens": 11595380.0,
      "step": 5925
    },
    {
      "entropy": 6.327992630004883,
      "epoch": 0.592792522617084,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004972549435452558,
      "loss": 6.2615,
      "mean_token_accuracy": 0.14188092723488807,
      "num_tokens": 11605383.0,
      "step": 5930
    },
    {
      "entropy": 6.41739411354065,
      "epoch": 0.5932923476783126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004972493829058474,
      "loss": 6.3191,
      "mean_token_accuracy": 0.1437163032591343,
      "num_tokens": 11614783.0,
      "step": 5935
    },
    {
      "entropy": 6.3635155200958256,
      "epoch": 0.5937921727395412,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004972438166747016,
      "loss": 6.2552,
      "mean_token_accuracy": 0.14206603318452835,
      "num_tokens": 11624524.0,
      "step": 5940
    },
    {
      "entropy": 6.2278985500335695,
      "epoch": 0.5942919978007697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972382448519586,
      "loss": 6.144,
      "mean_token_accuracy": 0.15115639418363572,
      "num_tokens": 11633908.0,
      "step": 5945
    },
    {
      "entropy": 6.3192017555236815,
      "epoch": 0.5947918228619983,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004972326674377586,
      "loss": 6.2006,
      "mean_token_accuracy": 0.14456763938069345,
      "num_tokens": 11644076.0,
      "step": 5950
    },
    {
      "entropy": 6.4500165462493895,
      "epoch": 0.5952916479232269,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004972270844322418,
      "loss": 6.2651,
      "mean_token_accuracy": 0.1369330771267414,
      "num_tokens": 11652754.0,
      "step": 5955
    },
    {
      "entropy": 6.320836782455444,
      "epoch": 0.5957914729844555,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004972214958355488,
      "loss": 6.2043,
      "mean_token_accuracy": 0.1442195162177086,
      "num_tokens": 11663468.0,
      "step": 5960
    },
    {
      "entropy": 6.275934028625488,
      "epoch": 0.596291298045684,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004972159016478201,
      "loss": 6.2774,
      "mean_token_accuracy": 0.13894289657473563,
      "num_tokens": 11672711.0,
      "step": 5965
    },
    {
      "entropy": 6.363453006744384,
      "epoch": 0.5967911231069126,
      "grad_norm": 1.125,
      "learning_rate": 0.0004972103018691965,
      "loss": 6.2103,
      "mean_token_accuracy": 0.14364408776164056,
      "num_tokens": 11681800.0,
      "step": 5970
    },
    {
      "entropy": 6.298955249786377,
      "epoch": 0.5972909481681411,
      "grad_norm": 1.25,
      "learning_rate": 0.0004972046964998189,
      "loss": 6.2324,
      "mean_token_accuracy": 0.15028997808694838,
      "num_tokens": 11691229.0,
      "step": 5975
    },
    {
      "entropy": 6.256961488723755,
      "epoch": 0.5977907732293697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004971990855398283,
      "loss": 6.2322,
      "mean_token_accuracy": 0.14177148491144181,
      "num_tokens": 11700027.0,
      "step": 5980
    },
    {
      "entropy": 6.3168580532073975,
      "epoch": 0.5982905982905983,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971934689893658,
      "loss": 6.2177,
      "mean_token_accuracy": 0.147872631251812,
      "num_tokens": 11708899.0,
      "step": 5985
    },
    {
      "entropy": 6.305580043792725,
      "epoch": 0.5987904233518269,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004971878468485729,
      "loss": 6.2068,
      "mean_token_accuracy": 0.14760489612817765,
      "num_tokens": 11719325.0,
      "step": 5990
    },
    {
      "entropy": 6.359370422363281,
      "epoch": 0.5992902484130554,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497182219117591,
      "loss": 6.2579,
      "mean_token_accuracy": 0.15097607672214508,
      "num_tokens": 11728835.0,
      "step": 5995
    },
    {
      "entropy": 6.3921833515167235,
      "epoch": 0.599790073474284,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004971765857965615,
      "loss": 6.321,
      "mean_token_accuracy": 0.1357443355023861,
      "num_tokens": 11738252.0,
      "step": 6000
    },
    {
      "epoch": 0.599790073474284,
      "eval_entropy": 6.171299409141432,
      "eval_loss": 6.291280746459961,
      "eval_mean_token_accuracy": 0.1456388643914217,
      "eval_num_tokens": 11738252.0,
      "eval_runtime": 24.1965,
      "eval_samples_per_second": 1283.039,
      "eval_steps_per_second": 160.395,
      "step": 6000
    },
    {
      "entropy": 6.274492406845093,
      "epoch": 0.6002898985355126,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004971709468856263,
      "loss": 6.1711,
      "mean_token_accuracy": 0.15069851726293565,
      "num_tokens": 11748448.0,
      "step": 6005
    },
    {
      "entropy": 6.368481111526489,
      "epoch": 0.6007897235967411,
      "grad_norm": 1.125,
      "learning_rate": 0.0004971653023849274,
      "loss": 6.3666,
      "mean_token_accuracy": 0.13822117671370507,
      "num_tokens": 11757835.0,
      "step": 6010
    },
    {
      "entropy": 6.399860191345215,
      "epoch": 0.6012895486579697,
      "grad_norm": 3.203125,
      "learning_rate": 0.0004971596522946065,
      "loss": 6.3469,
      "mean_token_accuracy": 0.13127551972866058,
      "num_tokens": 11767792.0,
      "step": 6015
    },
    {
      "entropy": 6.373989582061768,
      "epoch": 0.6017893737191983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004971539966148059,
      "loss": 6.2048,
      "mean_token_accuracy": 0.1530209571123123,
      "num_tokens": 11777329.0,
      "step": 6020
    },
    {
      "entropy": 6.355198621749878,
      "epoch": 0.6022891987804269,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004971483353456681,
      "loss": 6.2909,
      "mean_token_accuracy": 0.1368030898272991,
      "num_tokens": 11786776.0,
      "step": 6025
    },
    {
      "entropy": 6.292406320571899,
      "epoch": 0.6027890238416554,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004971426684873352,
      "loss": 6.2362,
      "mean_token_accuracy": 0.1482604995369911,
      "num_tokens": 11797669.0,
      "step": 6030
    },
    {
      "entropy": 6.306433391571045,
      "epoch": 0.603288848902884,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004971369960399499,
      "loss": 6.1906,
      "mean_token_accuracy": 0.14523768573999404,
      "num_tokens": 11807661.0,
      "step": 6035
    },
    {
      "entropy": 6.355361461639404,
      "epoch": 0.6037886739641125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497131318003655,
      "loss": 6.3082,
      "mean_token_accuracy": 0.1412230409681797,
      "num_tokens": 11817943.0,
      "step": 6040
    },
    {
      "entropy": 6.335895156860351,
      "epoch": 0.6042884990253411,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004971256343785932,
      "loss": 6.2234,
      "mean_token_accuracy": 0.14417879432439804,
      "num_tokens": 11826896.0,
      "step": 6045
    },
    {
      "entropy": 6.357933235168457,
      "epoch": 0.6047883240865697,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004971199451649078,
      "loss": 6.1759,
      "mean_token_accuracy": 0.14720185697078705,
      "num_tokens": 11836306.0,
      "step": 6050
    },
    {
      "entropy": 6.268744230270386,
      "epoch": 0.6052881491477983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004971142503627415,
      "loss": 6.2729,
      "mean_token_accuracy": 0.14198994114995003,
      "num_tokens": 11845998.0,
      "step": 6055
    },
    {
      "entropy": 6.389762926101684,
      "epoch": 0.6057879742090269,
      "grad_norm": 1.140625,
      "learning_rate": 0.000497108549972238,
      "loss": 6.2721,
      "mean_token_accuracy": 0.1422507084906101,
      "num_tokens": 11855009.0,
      "step": 6060
    },
    {
      "entropy": 6.363714361190796,
      "epoch": 0.6062877992702554,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004971028439935404,
      "loss": 6.1195,
      "mean_token_accuracy": 0.1464904725551605,
      "num_tokens": 11865163.0,
      "step": 6065
    },
    {
      "entropy": 6.2727690696716305,
      "epoch": 0.606787624331484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004970971324267925,
      "loss": 6.4078,
      "mean_token_accuracy": 0.13739993795752525,
      "num_tokens": 11875720.0,
      "step": 6070
    },
    {
      "entropy": 6.464136791229248,
      "epoch": 0.6072874493927125,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004970914152721378,
      "loss": 6.3065,
      "mean_token_accuracy": 0.1439572900533676,
      "num_tokens": 11884247.0,
      "step": 6075
    },
    {
      "entropy": 6.3112904071807865,
      "epoch": 0.6077872744539411,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004970856925297204,
      "loss": 6.1738,
      "mean_token_accuracy": 0.14496984332799911,
      "num_tokens": 11894978.0,
      "step": 6080
    },
    {
      "entropy": 6.330346918106079,
      "epoch": 0.6082870995151697,
      "grad_norm": 1.09375,
      "learning_rate": 0.000497079964199684,
      "loss": 6.2715,
      "mean_token_accuracy": 0.13917354494333267,
      "num_tokens": 11904904.0,
      "step": 6085
    },
    {
      "entropy": 6.371335029602051,
      "epoch": 0.6087869245763983,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004970742302821729,
      "loss": 6.3162,
      "mean_token_accuracy": 0.13768910467624665,
      "num_tokens": 11915339.0,
      "step": 6090
    },
    {
      "entropy": 6.309075784683228,
      "epoch": 0.6092867496376269,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970684907773312,
      "loss": 6.2403,
      "mean_token_accuracy": 0.14057565331459046,
      "num_tokens": 11925451.0,
      "step": 6095
    },
    {
      "entropy": 6.28705244064331,
      "epoch": 0.6097865746988554,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004970627456853036,
      "loss": 6.1991,
      "mean_token_accuracy": 0.13912660554051398,
      "num_tokens": 11934931.0,
      "step": 6100
    },
    {
      "entropy": 6.3549824237823485,
      "epoch": 0.6102863997600839,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970569950062343,
      "loss": 6.1411,
      "mean_token_accuracy": 0.15859687328338623,
      "num_tokens": 11943492.0,
      "step": 6105
    },
    {
      "entropy": 6.33065013885498,
      "epoch": 0.6107862248213125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004970512387402682,
      "loss": 6.2867,
      "mean_token_accuracy": 0.1350765645503998,
      "num_tokens": 11952880.0,
      "step": 6110
    },
    {
      "entropy": 6.411534166336059,
      "epoch": 0.6112860498825411,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004970454768875501,
      "loss": 6.234,
      "mean_token_accuracy": 0.1428788870573044,
      "num_tokens": 11961840.0,
      "step": 6115
    },
    {
      "entropy": 6.3422047138214115,
      "epoch": 0.6117858749437697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004970397094482249,
      "loss": 6.2988,
      "mean_token_accuracy": 0.14307170659303664,
      "num_tokens": 11970833.0,
      "step": 6120
    },
    {
      "entropy": 6.289044284820557,
      "epoch": 0.6122857000049983,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970339364224378,
      "loss": 6.2685,
      "mean_token_accuracy": 0.14215366765856743,
      "num_tokens": 11980497.0,
      "step": 6125
    },
    {
      "entropy": 6.396374034881592,
      "epoch": 0.6127855250662269,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004970281578103339,
      "loss": 6.1798,
      "mean_token_accuracy": 0.14526013880968094,
      "num_tokens": 11989505.0,
      "step": 6130
    },
    {
      "entropy": 6.33764762878418,
      "epoch": 0.6132853501274554,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004970223736120587,
      "loss": 6.2879,
      "mean_token_accuracy": 0.13932138979434966,
      "num_tokens": 11998952.0,
      "step": 6135
    },
    {
      "entropy": 6.2425133228302006,
      "epoch": 0.6137851751886839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004970165838277578,
      "loss": 6.1013,
      "mean_token_accuracy": 0.1466141700744629,
      "num_tokens": 12009249.0,
      "step": 6140
    },
    {
      "entropy": 6.32916145324707,
      "epoch": 0.6142850002499125,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004970107884575767,
      "loss": 6.1643,
      "mean_token_accuracy": 0.14610085487365723,
      "num_tokens": 12018345.0,
      "step": 6145
    },
    {
      "entropy": 6.388977527618408,
      "epoch": 0.6147848253111411,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004970049875016613,
      "loss": 6.2464,
      "mean_token_accuracy": 0.14082304686307906,
      "num_tokens": 12026530.0,
      "step": 6150
    },
    {
      "entropy": 6.217254877090454,
      "epoch": 0.6152846503723697,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004969991809601576,
      "loss": 6.1937,
      "mean_token_accuracy": 0.14848583266139032,
      "num_tokens": 12036232.0,
      "step": 6155
    },
    {
      "entropy": 6.18802490234375,
      "epoch": 0.6157844754335983,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969933688332115,
      "loss": 6.1082,
      "mean_token_accuracy": 0.15249510407447814,
      "num_tokens": 12045668.0,
      "step": 6160
    },
    {
      "entropy": 6.286074209213257,
      "epoch": 0.6162843004948269,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969875511209694,
      "loss": 6.1482,
      "mean_token_accuracy": 0.151823428273201,
      "num_tokens": 12055544.0,
      "step": 6165
    },
    {
      "entropy": 6.388773965835571,
      "epoch": 0.6167841255560553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969817278235777,
      "loss": 6.3467,
      "mean_token_accuracy": 0.13230136781930923,
      "num_tokens": 12065076.0,
      "step": 6170
    },
    {
      "entropy": 6.364807081222534,
      "epoch": 0.6172839506172839,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969758989411829,
      "loss": 6.3095,
      "mean_token_accuracy": 0.14353452250361443,
      "num_tokens": 12074623.0,
      "step": 6175
    },
    {
      "entropy": 6.386380529403686,
      "epoch": 0.6177837756785125,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004969700644739315,
      "loss": 6.1973,
      "mean_token_accuracy": 0.1431209549307823,
      "num_tokens": 12085040.0,
      "step": 6180
    },
    {
      "entropy": 6.290756607055664,
      "epoch": 0.6182836007397411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969642244219705,
      "loss": 6.2327,
      "mean_token_accuracy": 0.14316661432385444,
      "num_tokens": 12094699.0,
      "step": 6185
    },
    {
      "entropy": 6.3445250511169435,
      "epoch": 0.6187834258009697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004969583787854467,
      "loss": 6.2376,
      "mean_token_accuracy": 0.14207953587174416,
      "num_tokens": 12104256.0,
      "step": 6190
    },
    {
      "entropy": 6.280412244796753,
      "epoch": 0.6192832508621983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969525275645071,
      "loss": 6.2467,
      "mean_token_accuracy": 0.14455268755555153,
      "num_tokens": 12114058.0,
      "step": 6195
    },
    {
      "entropy": 6.263645267486572,
      "epoch": 0.6197830759234269,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004969466707592991,
      "loss": 6.264,
      "mean_token_accuracy": 0.1416216641664505,
      "num_tokens": 12124483.0,
      "step": 6200
    },
    {
      "entropy": 6.404595565795899,
      "epoch": 0.6202829009846553,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049694080836997,
      "loss": 6.2614,
      "mean_token_accuracy": 0.14459464699029922,
      "num_tokens": 12134700.0,
      "step": 6205
    },
    {
      "entropy": 6.411861276626587,
      "epoch": 0.6207827260458839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969349403966672,
      "loss": 6.2962,
      "mean_token_accuracy": 0.1364305056631565,
      "num_tokens": 12144688.0,
      "step": 6210
    },
    {
      "entropy": 6.315106201171875,
      "epoch": 0.6212825511071125,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004969290668395385,
      "loss": 6.155,
      "mean_token_accuracy": 0.15292517393827437,
      "num_tokens": 12154597.0,
      "step": 6215
    },
    {
      "entropy": 6.325231838226318,
      "epoch": 0.6217823761683411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969231876987315,
      "loss": 6.3214,
      "mean_token_accuracy": 0.1363390490412712,
      "num_tokens": 12164730.0,
      "step": 6220
    },
    {
      "entropy": 6.388926219940186,
      "epoch": 0.6222822012295697,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004969173029743942,
      "loss": 6.383,
      "mean_token_accuracy": 0.1325180247426033,
      "num_tokens": 12174238.0,
      "step": 6225
    },
    {
      "entropy": 6.323497200012207,
      "epoch": 0.6227820262907983,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004969114126666746,
      "loss": 6.2118,
      "mean_token_accuracy": 0.13866135030984877,
      "num_tokens": 12183124.0,
      "step": 6230
    },
    {
      "entropy": 6.392317581176758,
      "epoch": 0.6232818513520267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004969055167757211,
      "loss": 6.227,
      "mean_token_accuracy": 0.14396619349718093,
      "num_tokens": 12192076.0,
      "step": 6235
    },
    {
      "entropy": 6.333703231811524,
      "epoch": 0.6237816764132553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004968996153016819,
      "loss": 6.1994,
      "mean_token_accuracy": 0.15209077820181846,
      "num_tokens": 12201244.0,
      "step": 6240
    },
    {
      "entropy": 6.3309064388275145,
      "epoch": 0.6242815014744839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004968937082447054,
      "loss": 6.2683,
      "mean_token_accuracy": 0.1403204970061779,
      "num_tokens": 12211349.0,
      "step": 6245
    },
    {
      "entropy": 6.319456577301025,
      "epoch": 0.6247813265357125,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968877956049403,
      "loss": 6.2213,
      "mean_token_accuracy": 0.14028934240341187,
      "num_tokens": 12220880.0,
      "step": 6250
    },
    {
      "entropy": 6.328017139434815,
      "epoch": 0.6252811515969411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004968818773825355,
      "loss": 6.2958,
      "mean_token_accuracy": 0.13598917052149773,
      "num_tokens": 12231398.0,
      "step": 6255
    },
    {
      "entropy": 6.420849561691284,
      "epoch": 0.6257809766581697,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004968759535776396,
      "loss": 6.2738,
      "mean_token_accuracy": 0.13677876740694045,
      "num_tokens": 12239487.0,
      "step": 6260
    },
    {
      "entropy": 6.442674541473389,
      "epoch": 0.6262808017193983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004968700241904018,
      "loss": 6.4042,
      "mean_token_accuracy": 0.13405960202217101,
      "num_tokens": 12249984.0,
      "step": 6265
    },
    {
      "entropy": 6.364074611663819,
      "epoch": 0.6267806267806267,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004968640892209714,
      "loss": 6.1956,
      "mean_token_accuracy": 0.14548043608665467,
      "num_tokens": 12259876.0,
      "step": 6270
    },
    {
      "entropy": 6.295832824707031,
      "epoch": 0.6272804518418553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004968581486694975,
      "loss": 6.2063,
      "mean_token_accuracy": 0.145529080927372,
      "num_tokens": 12270262.0,
      "step": 6275
    },
    {
      "entropy": 6.305997657775879,
      "epoch": 0.6277802769030839,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004968522025361297,
      "loss": 6.2202,
      "mean_token_accuracy": 0.13966606184840202,
      "num_tokens": 12280432.0,
      "step": 6280
    },
    {
      "entropy": 6.36839280128479,
      "epoch": 0.6282801019643125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968462508210176,
      "loss": 6.2021,
      "mean_token_accuracy": 0.14929254874587058,
      "num_tokens": 12290518.0,
      "step": 6285
    },
    {
      "entropy": 6.397608089447021,
      "epoch": 0.6287799270255411,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004968402935243109,
      "loss": 6.2988,
      "mean_token_accuracy": 0.1373285174369812,
      "num_tokens": 12300305.0,
      "step": 6290
    },
    {
      "entropy": 6.223261308670044,
      "epoch": 0.6292797520867697,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004968343306461595,
      "loss": 6.2321,
      "mean_token_accuracy": 0.1458751752972603,
      "num_tokens": 12309079.0,
      "step": 6295
    },
    {
      "entropy": 6.398604059219361,
      "epoch": 0.6297795771479981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004968283621867134,
      "loss": 6.3142,
      "mean_token_accuracy": 0.13221606537699698,
      "num_tokens": 12319502.0,
      "step": 6300
    },
    {
      "entropy": 6.404465389251709,
      "epoch": 0.6302794022092267,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004968223881461227,
      "loss": 6.192,
      "mean_token_accuracy": 0.1482952281832695,
      "num_tokens": 12329025.0,
      "step": 6305
    },
    {
      "entropy": 6.253521680831909,
      "epoch": 0.6307792272704553,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000496816408524538,
      "loss": 6.2636,
      "mean_token_accuracy": 0.14268731996417044,
      "num_tokens": 12338822.0,
      "step": 6310
    },
    {
      "entropy": 6.331822156906128,
      "epoch": 0.6312790523316839,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968104233221094,
      "loss": 6.2981,
      "mean_token_accuracy": 0.14605509862303734,
      "num_tokens": 12348592.0,
      "step": 6315
    },
    {
      "entropy": 6.290080833435058,
      "epoch": 0.6317788773929125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004968044325389877,
      "loss": 6.1795,
      "mean_token_accuracy": 0.144684499502182,
      "num_tokens": 12357673.0,
      "step": 6320
    },
    {
      "entropy": 6.349246215820313,
      "epoch": 0.6322787024541411,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004967984361753235,
      "loss": 6.2357,
      "mean_token_accuracy": 0.1409025065600872,
      "num_tokens": 12367603.0,
      "step": 6325
    },
    {
      "entropy": 6.381235122680664,
      "epoch": 0.6327785275153697,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004967924342312677,
      "loss": 6.2592,
      "mean_token_accuracy": 0.1418330490589142,
      "num_tokens": 12376532.0,
      "step": 6330
    },
    {
      "entropy": 6.286419486999511,
      "epoch": 0.6332783525765981,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004967864267069713,
      "loss": 6.2178,
      "mean_token_accuracy": 0.14320805445313453,
      "num_tokens": 12386553.0,
      "step": 6335
    },
    {
      "entropy": 6.299349403381347,
      "epoch": 0.6337781776378267,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004967804136025856,
      "loss": 6.2621,
      "mean_token_accuracy": 0.1455235078930855,
      "num_tokens": 12396274.0,
      "step": 6340
    },
    {
      "entropy": 6.299895191192627,
      "epoch": 0.6342780026990553,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967743949182617,
      "loss": 6.1878,
      "mean_token_accuracy": 0.14820147305727005,
      "num_tokens": 12405993.0,
      "step": 6345
    },
    {
      "entropy": 6.268641567230224,
      "epoch": 0.6347778277602839,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967683706541511,
      "loss": 6.2266,
      "mean_token_accuracy": 0.1462365470826626,
      "num_tokens": 12417309.0,
      "step": 6350
    },
    {
      "entropy": 6.238659524917603,
      "epoch": 0.6352776528215125,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004967623408104054,
      "loss": 6.1101,
      "mean_token_accuracy": 0.14921127036213874,
      "num_tokens": 12425981.0,
      "step": 6355
    },
    {
      "entropy": 6.3483508110046385,
      "epoch": 0.6357774778827411,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004967563053871762,
      "loss": 6.2104,
      "mean_token_accuracy": 0.14522340297698974,
      "num_tokens": 12435197.0,
      "step": 6360
    },
    {
      "entropy": 6.357829570770264,
      "epoch": 0.6362773029439696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967502643846155,
      "loss": 6.1943,
      "mean_token_accuracy": 0.1444212503731251,
      "num_tokens": 12444697.0,
      "step": 6365
    },
    {
      "entropy": 6.285469055175781,
      "epoch": 0.6367771280051981,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000496744217802875,
      "loss": 6.1861,
      "mean_token_accuracy": 0.1461622893810272,
      "num_tokens": 12454490.0,
      "step": 6370
    },
    {
      "entropy": 6.362904691696167,
      "epoch": 0.6372769530664267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004967381656421073,
      "loss": 6.2176,
      "mean_token_accuracy": 0.13985132202506065,
      "num_tokens": 12463440.0,
      "step": 6375
    },
    {
      "entropy": 6.329201507568359,
      "epoch": 0.6377767781276553,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967321079024643,
      "loss": 6.3191,
      "mean_token_accuracy": 0.14112509712576865,
      "num_tokens": 12473285.0,
      "step": 6380
    },
    {
      "entropy": 6.37447600364685,
      "epoch": 0.6382766031888839,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004967260445840985,
      "loss": 6.171,
      "mean_token_accuracy": 0.14786145091056824,
      "num_tokens": 12481975.0,
      "step": 6385
    },
    {
      "entropy": 6.267659425735474,
      "epoch": 0.6387764282501125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004967199756871624,
      "loss": 6.2179,
      "mean_token_accuracy": 0.1422448791563511,
      "num_tokens": 12491736.0,
      "step": 6390
    },
    {
      "entropy": 6.2593241214752195,
      "epoch": 0.6392762533113411,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004967139012118088,
      "loss": 6.2785,
      "mean_token_accuracy": 0.13776970505714417,
      "num_tokens": 12502369.0,
      "step": 6395
    },
    {
      "entropy": 6.375499534606933,
      "epoch": 0.6397760783725696,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004967078211581907,
      "loss": 6.2218,
      "mean_token_accuracy": 0.14601902812719345,
      "num_tokens": 12512906.0,
      "step": 6400
    },
    {
      "entropy": 6.3216346263885494,
      "epoch": 0.6402759034337981,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004967017355264606,
      "loss": 6.1857,
      "mean_token_accuracy": 0.1441918671131134,
      "num_tokens": 12523454.0,
      "step": 6405
    },
    {
      "entropy": 6.361678838729858,
      "epoch": 0.6407757284950267,
      "grad_norm": 1.265625,
      "learning_rate": 0.000496695644316772,
      "loss": 6.355,
      "mean_token_accuracy": 0.13454906344413758,
      "num_tokens": 12533228.0,
      "step": 6410
    },
    {
      "entropy": 6.438969087600708,
      "epoch": 0.6412755535562553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966895475292781,
      "loss": 6.3414,
      "mean_token_accuracy": 0.13455577269196511,
      "num_tokens": 12543188.0,
      "step": 6415
    },
    {
      "entropy": 6.277749252319336,
      "epoch": 0.6417753786174839,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004966834451641322,
      "loss": 6.2644,
      "mean_token_accuracy": 0.13704225644469262,
      "num_tokens": 12553844.0,
      "step": 6420
    },
    {
      "entropy": 6.314139127731323,
      "epoch": 0.6422752036787125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966773372214878,
      "loss": 6.1432,
      "mean_token_accuracy": 0.14396685659885405,
      "num_tokens": 12563544.0,
      "step": 6425
    },
    {
      "entropy": 6.290168428421021,
      "epoch": 0.642775028739941,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004966712237014987,
      "loss": 6.1848,
      "mean_token_accuracy": 0.14483810365200042,
      "num_tokens": 12572894.0,
      "step": 6430
    },
    {
      "entropy": 6.354187631607056,
      "epoch": 0.6432748538011696,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004966651046043186,
      "loss": 6.3282,
      "mean_token_accuracy": 0.1313794285058975,
      "num_tokens": 12583533.0,
      "step": 6435
    },
    {
      "entropy": 6.380536222457886,
      "epoch": 0.6437746788623981,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004966589799301016,
      "loss": 6.2021,
      "mean_token_accuracy": 0.14208913892507552,
      "num_tokens": 12593160.0,
      "step": 6440
    },
    {
      "entropy": 6.397803258895874,
      "epoch": 0.6442745039236267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966528496790015,
      "loss": 6.3123,
      "mean_token_accuracy": 0.13903931006789208,
      "num_tokens": 12604010.0,
      "step": 6445
    },
    {
      "entropy": 6.252782583236694,
      "epoch": 0.6447743289848553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004966467138511729,
      "loss": 6.2253,
      "mean_token_accuracy": 0.14773206263780594,
      "num_tokens": 12614256.0,
      "step": 6450
    },
    {
      "entropy": 6.363953733444214,
      "epoch": 0.6452741540460839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00049664057244677,
      "loss": 6.3054,
      "mean_token_accuracy": 0.13916618898510932,
      "num_tokens": 12624228.0,
      "step": 6455
    },
    {
      "entropy": 6.392262363433838,
      "epoch": 0.6457739791073125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004966344254659473,
      "loss": 6.1667,
      "mean_token_accuracy": 0.14237217232584953,
      "num_tokens": 12634611.0,
      "step": 6460
    },
    {
      "entropy": 6.29950442314148,
      "epoch": 0.646273804168541,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004966282729088595,
      "loss": 6.1896,
      "mean_token_accuracy": 0.1421133279800415,
      "num_tokens": 12645309.0,
      "step": 6465
    },
    {
      "entropy": 6.3200812339782715,
      "epoch": 0.6467736292297696,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004966221147756615,
      "loss": 6.1801,
      "mean_token_accuracy": 0.14744684249162673,
      "num_tokens": 12655283.0,
      "step": 6470
    },
    {
      "entropy": 6.300661897659301,
      "epoch": 0.6472734542909981,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004966159510665079,
      "loss": 6.2131,
      "mean_token_accuracy": 0.14655552580952644,
      "num_tokens": 12664531.0,
      "step": 6475
    },
    {
      "entropy": 6.270278072357177,
      "epoch": 0.6477732793522267,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966097817815542,
      "loss": 6.2533,
      "mean_token_accuracy": 0.1458370253443718,
      "num_tokens": 12674791.0,
      "step": 6480
    },
    {
      "entropy": 6.293416118621826,
      "epoch": 0.6482731044134553,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004966036069209552,
      "loss": 6.2286,
      "mean_token_accuracy": 0.143087674677372,
      "num_tokens": 12685359.0,
      "step": 6485
    },
    {
      "entropy": 6.273835134506226,
      "epoch": 0.6487729294746839,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004965974264848666,
      "loss": 6.2206,
      "mean_token_accuracy": 0.14310649782419205,
      "num_tokens": 12694719.0,
      "step": 6490
    },
    {
      "entropy": 6.374215412139892,
      "epoch": 0.6492727545359124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965912404734438,
      "loss": 6.2441,
      "mean_token_accuracy": 0.14720723628997803,
      "num_tokens": 12704032.0,
      "step": 6495
    },
    {
      "entropy": 6.367592763900757,
      "epoch": 0.649772579597141,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004965850488868424,
      "loss": 6.2388,
      "mean_token_accuracy": 0.1453162431716919,
      "num_tokens": 12713785.0,
      "step": 6500
    },
    {
      "entropy": 6.160141468048096,
      "epoch": 0.6502724046583696,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965788517252181,
      "loss": 6.1582,
      "mean_token_accuracy": 0.14621974602341653,
      "num_tokens": 12724718.0,
      "step": 6505
    },
    {
      "entropy": 6.368258810043335,
      "epoch": 0.6507722297195981,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000496572648988727,
      "loss": 6.2652,
      "mean_token_accuracy": 0.14782266169786454,
      "num_tokens": 12733518.0,
      "step": 6510
    },
    {
      "entropy": 6.385526132583618,
      "epoch": 0.6512720547808267,
      "grad_norm": 1.3125,
      "learning_rate": 0.000496566440677525,
      "loss": 6.1919,
      "mean_token_accuracy": 0.14558335468173028,
      "num_tokens": 12743453.0,
      "step": 6515
    },
    {
      "entropy": 6.248903751373291,
      "epoch": 0.6517718798420553,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004965602267917685,
      "loss": 6.2701,
      "mean_token_accuracy": 0.14530854895710946,
      "num_tokens": 12753717.0,
      "step": 6520
    },
    {
      "entropy": 6.241318988800049,
      "epoch": 0.6522717049032839,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965540073316136,
      "loss": 6.1582,
      "mean_token_accuracy": 0.14887857139110566,
      "num_tokens": 12763240.0,
      "step": 6525
    },
    {
      "entropy": 6.376130199432373,
      "epoch": 0.6527715299645124,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004965477822972169,
      "loss": 6.2475,
      "mean_token_accuracy": 0.14736852347850798,
      "num_tokens": 12773284.0,
      "step": 6530
    },
    {
      "entropy": 6.352194786071777,
      "epoch": 0.653271355025741,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965415516887349,
      "loss": 6.2401,
      "mean_token_accuracy": 0.14244507774710655,
      "num_tokens": 12782189.0,
      "step": 6535
    },
    {
      "entropy": 6.249260711669922,
      "epoch": 0.6537711800869695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965353155063246,
      "loss": 6.1602,
      "mean_token_accuracy": 0.1446791686117649,
      "num_tokens": 12792790.0,
      "step": 6540
    },
    {
      "entropy": 6.3017161846160885,
      "epoch": 0.6542710051481981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004965290737501427,
      "loss": 6.1347,
      "mean_token_accuracy": 0.1483735680580139,
      "num_tokens": 12802368.0,
      "step": 6545
    },
    {
      "entropy": 6.256836366653443,
      "epoch": 0.6547708302094267,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004965228264203464,
      "loss": 6.1583,
      "mean_token_accuracy": 0.1409124404191971,
      "num_tokens": 12812226.0,
      "step": 6550
    },
    {
      "entropy": 6.283217096328736,
      "epoch": 0.6552706552706553,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004965165735170928,
      "loss": 6.1923,
      "mean_token_accuracy": 0.14620148241519929,
      "num_tokens": 12820986.0,
      "step": 6555
    },
    {
      "entropy": 6.244024801254272,
      "epoch": 0.6557704803318838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000496510315040539,
      "loss": 6.2309,
      "mean_token_accuracy": 0.14509713053703308,
      "num_tokens": 12830690.0,
      "step": 6560
    },
    {
      "entropy": 6.396744346618652,
      "epoch": 0.6562703053931124,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004965040509908429,
      "loss": 6.2685,
      "mean_token_accuracy": 0.14355552121996878,
      "num_tokens": 12841780.0,
      "step": 6565
    },
    {
      "entropy": 6.363442277908325,
      "epoch": 0.656770130454341,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004964977813681618,
      "loss": 6.1872,
      "mean_token_accuracy": 0.14511173814535142,
      "num_tokens": 12850713.0,
      "step": 6570
    },
    {
      "entropy": 6.19380555152893,
      "epoch": 0.6572699555155695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004964915061726534,
      "loss": 6.2447,
      "mean_token_accuracy": 0.14074426516890526,
      "num_tokens": 12862211.0,
      "step": 6575
    },
    {
      "entropy": 6.307309150695801,
      "epoch": 0.6577697805767981,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004964852254044759,
      "loss": 6.3074,
      "mean_token_accuracy": 0.1373126357793808,
      "num_tokens": 12871171.0,
      "step": 6580
    },
    {
      "entropy": 6.341297435760498,
      "epoch": 0.6582696056380267,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004964789390637871,
      "loss": 6.1616,
      "mean_token_accuracy": 0.1444268763065338,
      "num_tokens": 12880601.0,
      "step": 6585
    },
    {
      "entropy": 6.316075038909912,
      "epoch": 0.6587694306992553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004964726471507451,
      "loss": 6.1415,
      "mean_token_accuracy": 0.14696861207485198,
      "num_tokens": 12890702.0,
      "step": 6590
    },
    {
      "entropy": 6.294946527481079,
      "epoch": 0.6592692557604838,
      "grad_norm": 1.25,
      "learning_rate": 0.0004964663496655083,
      "loss": 6.2462,
      "mean_token_accuracy": 0.14137669578194617,
      "num_tokens": 12901156.0,
      "step": 6595
    },
    {
      "entropy": 6.389964962005616,
      "epoch": 0.6597690808217124,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004964600466082351,
      "loss": 6.2449,
      "mean_token_accuracy": 0.13817505463957785,
      "num_tokens": 12912183.0,
      "step": 6600
    },
    {
      "entropy": 6.2879331588745115,
      "epoch": 0.660268905882941,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004964537379790842,
      "loss": 6.221,
      "mean_token_accuracy": 0.14622295051813125,
      "num_tokens": 12922200.0,
      "step": 6605
    },
    {
      "entropy": 6.2114465713500975,
      "epoch": 0.6607687309441695,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004964474237782141,
      "loss": 6.1129,
      "mean_token_accuracy": 0.15032164603471757,
      "num_tokens": 12931201.0,
      "step": 6610
    },
    {
      "entropy": 6.268771314620972,
      "epoch": 0.6612685560053981,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004964411040057839,
      "loss": 6.116,
      "mean_token_accuracy": 0.15098268911242485,
      "num_tokens": 12941290.0,
      "step": 6615
    },
    {
      "entropy": 6.193240785598755,
      "epoch": 0.6617683810666267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004964347786619524,
      "loss": 6.1327,
      "mean_token_accuracy": 0.15584979429841042,
      "num_tokens": 12953458.0,
      "step": 6620
    },
    {
      "entropy": 6.136424922943116,
      "epoch": 0.6622682061278552,
      "grad_norm": 1.75,
      "learning_rate": 0.0004964284477468791,
      "loss": 5.9916,
      "mean_token_accuracy": 0.15900807455182076,
      "num_tokens": 12964072.0,
      "step": 6625
    },
    {
      "entropy": 6.396554088592529,
      "epoch": 0.6627680311890838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964221112607228,
      "loss": 6.2007,
      "mean_token_accuracy": 0.1418839856982231,
      "num_tokens": 12974092.0,
      "step": 6630
    },
    {
      "entropy": 6.216900396347046,
      "epoch": 0.6632678562503124,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004964157692036432,
      "loss": 6.1238,
      "mean_token_accuracy": 0.1494164824485779,
      "num_tokens": 12983546.0,
      "step": 6635
    },
    {
      "entropy": 6.295744895935059,
      "epoch": 0.663767681311541,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004964094215757998,
      "loss": 6.2074,
      "mean_token_accuracy": 0.14203025847673417,
      "num_tokens": 12993690.0,
      "step": 6640
    },
    {
      "entropy": 6.269428682327271,
      "epoch": 0.6642675063727695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004964030683773525,
      "loss": 6.1343,
      "mean_token_accuracy": 0.1497920498251915,
      "num_tokens": 13003248.0,
      "step": 6645
    },
    {
      "entropy": 6.280125379562378,
      "epoch": 0.6647673314339981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004963967096084608,
      "loss": 6.14,
      "mean_token_accuracy": 0.14949833750724792,
      "num_tokens": 13013182.0,
      "step": 6650
    },
    {
      "entropy": 6.300539493560791,
      "epoch": 0.6652671564952267,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004963903452692849,
      "loss": 6.1569,
      "mean_token_accuracy": 0.1493376262485981,
      "num_tokens": 13022726.0,
      "step": 6655
    },
    {
      "entropy": 6.21463451385498,
      "epoch": 0.6657669815564552,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004963839753599848,
      "loss": 6.116,
      "mean_token_accuracy": 0.14759032428264618,
      "num_tokens": 13031811.0,
      "step": 6660
    },
    {
      "entropy": 6.245493793487549,
      "epoch": 0.6662668066176838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000496377599880721,
      "loss": 6.1624,
      "mean_token_accuracy": 0.14878485426306726,
      "num_tokens": 13040884.0,
      "step": 6665
    },
    {
      "entropy": 6.34880223274231,
      "epoch": 0.6667666316789124,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004963712188316537,
      "loss": 6.2703,
      "mean_token_accuracy": 0.13781095147132874,
      "num_tokens": 13049937.0,
      "step": 6670
    },
    {
      "entropy": 6.282163000106811,
      "epoch": 0.667266456740141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004963648322129435,
      "loss": 6.1137,
      "mean_token_accuracy": 0.1517148993909359,
      "num_tokens": 13059307.0,
      "step": 6675
    },
    {
      "entropy": 6.276641321182251,
      "epoch": 0.6677662818013695,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000496358440024751,
      "loss": 6.2419,
      "mean_token_accuracy": 0.1388409234583378,
      "num_tokens": 13069400.0,
      "step": 6680
    },
    {
      "entropy": 6.315621709823608,
      "epoch": 0.6682661068625981,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004963520422672372,
      "loss": 6.1308,
      "mean_token_accuracy": 0.14684976562857627,
      "num_tokens": 13079965.0,
      "step": 6685
    },
    {
      "entropy": 6.385589170455932,
      "epoch": 0.6687659319238266,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496345638940563,
      "loss": 6.3153,
      "mean_token_accuracy": 0.13813058584928511,
      "num_tokens": 13090144.0,
      "step": 6690
    },
    {
      "entropy": 6.284987211227417,
      "epoch": 0.6692657569850552,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004963392300448894,
      "loss": 6.1161,
      "mean_token_accuracy": 0.1471381053328514,
      "num_tokens": 13100217.0,
      "step": 6695
    },
    {
      "entropy": 6.21724944114685,
      "epoch": 0.6697655820462838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963328155803778,
      "loss": 6.189,
      "mean_token_accuracy": 0.1490732327103615,
      "num_tokens": 13109031.0,
      "step": 6700
    },
    {
      "entropy": 6.371466398239136,
      "epoch": 0.6702654071075124,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963263955471895,
      "loss": 6.2883,
      "mean_token_accuracy": 0.13497610464692117,
      "num_tokens": 13119073.0,
      "step": 6705
    },
    {
      "entropy": 6.319394302368164,
      "epoch": 0.670765232168741,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496319969945486,
      "loss": 6.2655,
      "mean_token_accuracy": 0.1459631949663162,
      "num_tokens": 13129208.0,
      "step": 6710
    },
    {
      "entropy": 6.284710121154785,
      "epoch": 0.6712650572299695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000496313538775429,
      "loss": 6.1742,
      "mean_token_accuracy": 0.1504007652401924,
      "num_tokens": 13138559.0,
      "step": 6715
    },
    {
      "entropy": 6.331243562698364,
      "epoch": 0.6717648822911981,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004963071020371803,
      "loss": 6.2935,
      "mean_token_accuracy": 0.1420964501798153,
      "num_tokens": 13148456.0,
      "step": 6720
    },
    {
      "entropy": 6.3337866306304935,
      "epoch": 0.6722647073524266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004963006597309019,
      "loss": 6.2114,
      "mean_token_accuracy": 0.1429819457232952,
      "num_tokens": 13158245.0,
      "step": 6725
    },
    {
      "entropy": 6.36514048576355,
      "epoch": 0.6727645324136552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004962942118567559,
      "loss": 6.246,
      "mean_token_accuracy": 0.15058982968330384,
      "num_tokens": 13166269.0,
      "step": 6730
    },
    {
      "entropy": 6.321820068359375,
      "epoch": 0.6732643574748838,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004962877584149043,
      "loss": 6.2107,
      "mean_token_accuracy": 0.13691925257444382,
      "num_tokens": 13176763.0,
      "step": 6735
    },
    {
      "entropy": 6.301709985733032,
      "epoch": 0.6737641825361124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962812994055098,
      "loss": 6.1673,
      "mean_token_accuracy": 0.1468019053339958,
      "num_tokens": 13185945.0,
      "step": 6740
    },
    {
      "entropy": 6.350088691711425,
      "epoch": 0.674264007597341,
      "grad_norm": 1.375,
      "learning_rate": 0.0004962748348287347,
      "loss": 6.2539,
      "mean_token_accuracy": 0.14276561960577966,
      "num_tokens": 13197205.0,
      "step": 6745
    },
    {
      "entropy": 6.24924602508545,
      "epoch": 0.6747638326585695,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004962683646847416,
      "loss": 6.2246,
      "mean_token_accuracy": 0.1370372012257576,
      "num_tokens": 13206939.0,
      "step": 6750
    },
    {
      "entropy": 6.26561279296875,
      "epoch": 0.675263657719798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004962618889736936,
      "loss": 6.1563,
      "mean_token_accuracy": 0.14248717650771142,
      "num_tokens": 13216317.0,
      "step": 6755
    },
    {
      "entropy": 6.327649021148682,
      "epoch": 0.6757634827810266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962554076957533,
      "loss": 6.1696,
      "mean_token_accuracy": 0.15100847855210303,
      "num_tokens": 13226594.0,
      "step": 6760
    },
    {
      "entropy": 6.3978033542633055,
      "epoch": 0.6762633078422552,
      "grad_norm": 1.125,
      "learning_rate": 0.0004962489208510838,
      "loss": 6.3432,
      "mean_token_accuracy": 0.1349129468202591,
      "num_tokens": 13237477.0,
      "step": 6765
    },
    {
      "entropy": 6.335607051849365,
      "epoch": 0.6767631329034838,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004962424284398484,
      "loss": 6.1448,
      "mean_token_accuracy": 0.14610870331525802,
      "num_tokens": 13247041.0,
      "step": 6770
    },
    {
      "entropy": 6.248000860214233,
      "epoch": 0.6772629579647124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004962359304622105,
      "loss": 6.1268,
      "mean_token_accuracy": 0.15253702104091643,
      "num_tokens": 13257223.0,
      "step": 6775
    },
    {
      "entropy": 6.312354278564453,
      "epoch": 0.677762783025941,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962294269183335,
      "loss": 6.1855,
      "mean_token_accuracy": 0.14766823798418044,
      "num_tokens": 13267457.0,
      "step": 6780
    },
    {
      "entropy": 6.36751766204834,
      "epoch": 0.6782626080871695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000496222917808381,
      "loss": 6.2055,
      "mean_token_accuracy": 0.14150047898292542,
      "num_tokens": 13277656.0,
      "step": 6785
    },
    {
      "entropy": 6.301115703582764,
      "epoch": 0.678762433148398,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962164031325169,
      "loss": 6.1935,
      "mean_token_accuracy": 0.14375955387949943,
      "num_tokens": 13288483.0,
      "step": 6790
    },
    {
      "entropy": 6.280858945846558,
      "epoch": 0.6792622582096266,
      "grad_norm": 1.21875,
      "learning_rate": 0.000496209882890905,
      "loss": 6.1012,
      "mean_token_accuracy": 0.153845976293087,
      "num_tokens": 13298312.0,
      "step": 6795
    },
    {
      "entropy": 6.170512390136719,
      "epoch": 0.6797620832708552,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004962033570837094,
      "loss": 6.027,
      "mean_token_accuracy": 0.14800312519073486,
      "num_tokens": 13308038.0,
      "step": 6800
    },
    {
      "entropy": 6.20284628868103,
      "epoch": 0.6802619083320838,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961968257110941,
      "loss": 6.1231,
      "mean_token_accuracy": 0.14912261813879013,
      "num_tokens": 13317963.0,
      "step": 6805
    },
    {
      "entropy": 6.256144618988037,
      "epoch": 0.6807617333933124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004961902887732236,
      "loss": 6.2112,
      "mean_token_accuracy": 0.14450787454843522,
      "num_tokens": 13328210.0,
      "step": 6810
    },
    {
      "entropy": 6.291503047943115,
      "epoch": 0.681261558454541,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004961837462702626,
      "loss": 6.1605,
      "mean_token_accuracy": 0.14515035152435302,
      "num_tokens": 13337778.0,
      "step": 6815
    },
    {
      "entropy": 6.248762559890747,
      "epoch": 0.6817613835157694,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004961771982023752,
      "loss": 6.1431,
      "mean_token_accuracy": 0.14761996269226074,
      "num_tokens": 13348117.0,
      "step": 6820
    },
    {
      "entropy": 6.293111038208008,
      "epoch": 0.682261208576998,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004961706445697265,
      "loss": 6.1105,
      "mean_token_accuracy": 0.14693563356995581,
      "num_tokens": 13356379.0,
      "step": 6825
    },
    {
      "entropy": 6.249057245254517,
      "epoch": 0.6827610336382266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004961640853724813,
      "loss": 6.1236,
      "mean_token_accuracy": 0.14708415269851685,
      "num_tokens": 13366621.0,
      "step": 6830
    },
    {
      "entropy": 6.261090230941773,
      "epoch": 0.6832608586994552,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961575206108046,
      "loss": 6.267,
      "mean_token_accuracy": 0.14229123517870904,
      "num_tokens": 13375596.0,
      "step": 6835
    },
    {
      "entropy": 6.34532675743103,
      "epoch": 0.6837606837606838,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004961509502848616,
      "loss": 6.2277,
      "mean_token_accuracy": 0.1445683941245079,
      "num_tokens": 13385871.0,
      "step": 6840
    },
    {
      "entropy": 6.383232927322387,
      "epoch": 0.6842605088219124,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004961443743948177,
      "loss": 6.2398,
      "mean_token_accuracy": 0.1395413100719452,
      "num_tokens": 13395277.0,
      "step": 6845
    },
    {
      "entropy": 6.278546905517578,
      "epoch": 0.684760333883141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496137792940838,
      "loss": 6.2684,
      "mean_token_accuracy": 0.1410169631242752,
      "num_tokens": 13405064.0,
      "step": 6850
    },
    {
      "entropy": 6.305002641677857,
      "epoch": 0.6852601589443694,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004961312059230885,
      "loss": 6.214,
      "mean_token_accuracy": 0.14182735979557037,
      "num_tokens": 13414399.0,
      "step": 6855
    },
    {
      "entropy": 6.343762350082398,
      "epoch": 0.685759984005598,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004961246133417348,
      "loss": 6.3045,
      "mean_token_accuracy": 0.14280288964509963,
      "num_tokens": 13424343.0,
      "step": 6860
    },
    {
      "entropy": 6.320830869674682,
      "epoch": 0.6862598090668266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004961180151969427,
      "loss": 6.1837,
      "mean_token_accuracy": 0.14768005162477493,
      "num_tokens": 13434362.0,
      "step": 6865
    },
    {
      "entropy": 6.305441236495971,
      "epoch": 0.6867596341280552,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004961114114888782,
      "loss": 6.1641,
      "mean_token_accuracy": 0.14760831221938134,
      "num_tokens": 13444578.0,
      "step": 6870
    },
    {
      "entropy": 6.180065107345581,
      "epoch": 0.6872594591892838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961048022177074,
      "loss": 6.0746,
      "mean_token_accuracy": 0.14900707975029945,
      "num_tokens": 13454184.0,
      "step": 6875
    },
    {
      "entropy": 6.242380571365357,
      "epoch": 0.6877592842505124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004960981873835968,
      "loss": 6.1696,
      "mean_token_accuracy": 0.14485947638750077,
      "num_tokens": 13464269.0,
      "step": 6880
    },
    {
      "entropy": 6.419793224334716,
      "epoch": 0.6882591093117408,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004960915669867126,
      "loss": 6.2355,
      "mean_token_accuracy": 0.1395081490278244,
      "num_tokens": 13472747.0,
      "step": 6885
    },
    {
      "entropy": 6.283924102783203,
      "epoch": 0.6887589343729694,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004960849410272216,
      "loss": 6.1958,
      "mean_token_accuracy": 0.1465576171875,
      "num_tokens": 13481670.0,
      "step": 6890
    },
    {
      "entropy": 6.198819017410278,
      "epoch": 0.689258759434198,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004960783095052903,
      "loss": 6.0976,
      "mean_token_accuracy": 0.15426829755306243,
      "num_tokens": 13490428.0,
      "step": 6895
    },
    {
      "entropy": 6.215169191360474,
      "epoch": 0.6897585844954266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004960716724210856,
      "loss": 6.0435,
      "mean_token_accuracy": 0.14921848475933075,
      "num_tokens": 13499549.0,
      "step": 6900
    },
    {
      "entropy": 6.277715873718262,
      "epoch": 0.6902584095566552,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004960650297747746,
      "loss": 6.2332,
      "mean_token_accuracy": 0.14397919923067093,
      "num_tokens": 13508930.0,
      "step": 6905
    },
    {
      "entropy": 6.294648456573486,
      "epoch": 0.6907582346178838,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004960583815665242,
      "loss": 6.2406,
      "mean_token_accuracy": 0.14256019294261932,
      "num_tokens": 13518858.0,
      "step": 6910
    },
    {
      "entropy": 6.298892211914063,
      "epoch": 0.6912580596791124,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004960517277965018,
      "loss": 6.1825,
      "mean_token_accuracy": 0.1378312423825264,
      "num_tokens": 13529346.0,
      "step": 6915
    },
    {
      "entropy": 6.3459924221038815,
      "epoch": 0.6917578847403408,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004960450684648749,
      "loss": 6.1842,
      "mean_token_accuracy": 0.13952886387705804,
      "num_tokens": 13539096.0,
      "step": 6920
    },
    {
      "entropy": 6.2896405220031735,
      "epoch": 0.6922577098015694,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000496038403571811,
      "loss": 6.1979,
      "mean_token_accuracy": 0.14368968605995178,
      "num_tokens": 13548698.0,
      "step": 6925
    },
    {
      "entropy": 6.274778366088867,
      "epoch": 0.692757534862798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004960317331174777,
      "loss": 6.2011,
      "mean_token_accuracy": 0.14793051406741142,
      "num_tokens": 13559550.0,
      "step": 6930
    },
    {
      "entropy": 6.30463662147522,
      "epoch": 0.6932573599240266,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004960250571020428,
      "loss": 6.2687,
      "mean_token_accuracy": 0.1395127072930336,
      "num_tokens": 13569248.0,
      "step": 6935
    },
    {
      "entropy": 6.331027507781982,
      "epoch": 0.6937571849852552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004960183755256744,
      "loss": 6.1399,
      "mean_token_accuracy": 0.14608098790049553,
      "num_tokens": 13578957.0,
      "step": 6940
    },
    {
      "entropy": 6.323013687133789,
      "epoch": 0.6942570100464838,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004960116883885406,
      "loss": 6.2524,
      "mean_token_accuracy": 0.13893733620643617,
      "num_tokens": 13589630.0,
      "step": 6945
    },
    {
      "entropy": 6.25451283454895,
      "epoch": 0.6947568351077124,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004960049956908095,
      "loss": 6.1699,
      "mean_token_accuracy": 0.14550091028213502,
      "num_tokens": 13599067.0,
      "step": 6950
    },
    {
      "entropy": 6.318792629241943,
      "epoch": 0.6952566601689408,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004959982974326496,
      "loss": 6.2241,
      "mean_token_accuracy": 0.14055988267064096,
      "num_tokens": 13609068.0,
      "step": 6955
    },
    {
      "entropy": 6.359457159042359,
      "epoch": 0.6957564852301694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004959915936142294,
      "loss": 6.2748,
      "mean_token_accuracy": 0.14214019179344178,
      "num_tokens": 13619112.0,
      "step": 6960
    },
    {
      "entropy": 6.361245393753052,
      "epoch": 0.696256310291398,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004959848842357175,
      "loss": 6.2935,
      "mean_token_accuracy": 0.14663100615143776,
      "num_tokens": 13629518.0,
      "step": 6965
    },
    {
      "entropy": 6.399230909347534,
      "epoch": 0.6967561353526266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959781692972829,
      "loss": 6.3038,
      "mean_token_accuracy": 0.1396307483315468,
      "num_tokens": 13639796.0,
      "step": 6970
    },
    {
      "entropy": 6.35450291633606,
      "epoch": 0.6972559604138552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959714487990943,
      "loss": 6.2349,
      "mean_token_accuracy": 0.14618084356188774,
      "num_tokens": 13649876.0,
      "step": 6975
    },
    {
      "entropy": 6.302280473709106,
      "epoch": 0.6977557854750838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000495964722741321,
      "loss": 6.2086,
      "mean_token_accuracy": 0.14748310670256615,
      "num_tokens": 13660670.0,
      "step": 6980
    },
    {
      "entropy": 6.249253225326538,
      "epoch": 0.6982556105363122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959579911241322,
      "loss": 6.15,
      "mean_token_accuracy": 0.1474284589290619,
      "num_tokens": 13669958.0,
      "step": 6985
    },
    {
      "entropy": 6.240137100219727,
      "epoch": 0.6987554355975408,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004959512539476971,
      "loss": 6.0697,
      "mean_token_accuracy": 0.15434898436069489,
      "num_tokens": 13678414.0,
      "step": 6990
    },
    {
      "entropy": 6.218645906448364,
      "epoch": 0.6992552606587694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004959445112121853,
      "loss": 6.1325,
      "mean_token_accuracy": 0.14961965158581733,
      "num_tokens": 13687424.0,
      "step": 6995
    },
    {
      "entropy": 6.311236190795898,
      "epoch": 0.699755085719998,
      "grad_norm": 1.25,
      "learning_rate": 0.0004959377629177665,
      "loss": 6.2684,
      "mean_token_accuracy": 0.13884105309844016,
      "num_tokens": 13697249.0,
      "step": 7000
    },
    {
      "entropy": 6.310066986083984,
      "epoch": 0.7002549107812266,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004959310090646104,
      "loss": 6.1721,
      "mean_token_accuracy": 0.1475272737443447,
      "num_tokens": 13707059.0,
      "step": 7005
    },
    {
      "entropy": 6.322434949874878,
      "epoch": 0.7007547358424552,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959242496528869,
      "loss": 6.116,
      "mean_token_accuracy": 0.14815280884504317,
      "num_tokens": 13715899.0,
      "step": 7010
    },
    {
      "entropy": 6.270565557479858,
      "epoch": 0.7012545609036838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004959174846827663,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13940336182713509,
      "num_tokens": 13725318.0,
      "step": 7015
    },
    {
      "entropy": 6.304192447662354,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004959107141544186,
      "loss": 6.2139,
      "mean_token_accuracy": 0.14462756961584092,
      "num_tokens": 13734816.0,
      "step": 7020
    },
    {
      "entropy": 6.368301200866699,
      "epoch": 0.7022542110261408,
      "grad_norm": 1.21875,
      "learning_rate": 0.000495903938068014,
      "loss": 6.2381,
      "mean_token_accuracy": 0.14005376696586608,
      "num_tokens": 13744323.0,
      "step": 7025
    },
    {
      "entropy": 6.422999048233033,
      "epoch": 0.7027540360873694,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004958971564237233,
      "loss": 6.4322,
      "mean_token_accuracy": 0.12698919996619223,
      "num_tokens": 13754301.0,
      "step": 7030
    },
    {
      "entropy": 6.3674540519714355,
      "epoch": 0.703253861148598,
      "grad_norm": 1.125,
      "learning_rate": 0.000495890369221717,
      "loss": 6.2735,
      "mean_token_accuracy": 0.13697801530361176,
      "num_tokens": 13763561.0,
      "step": 7035
    },
    {
      "entropy": 6.29704179763794,
      "epoch": 0.7037536862098266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958835764621657,
      "loss": 6.1637,
      "mean_token_accuracy": 0.14473214745521545,
      "num_tokens": 13773500.0,
      "step": 7040
    },
    {
      "entropy": 6.3702723503112795,
      "epoch": 0.7042535112710552,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004958767781452406,
      "loss": 6.1574,
      "mean_token_accuracy": 0.1479409858584404,
      "num_tokens": 13782390.0,
      "step": 7045
    },
    {
      "entropy": 6.264047527313233,
      "epoch": 0.7047533363322837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004958699742711126,
      "loss": 6.1496,
      "mean_token_accuracy": 0.1466951124370098,
      "num_tokens": 13791790.0,
      "step": 7050
    },
    {
      "entropy": 6.299694347381592,
      "epoch": 0.7052531613935122,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004958631648399528,
      "loss": 6.2389,
      "mean_token_accuracy": 0.1398335501551628,
      "num_tokens": 13802352.0,
      "step": 7055
    },
    {
      "entropy": 6.274299812316895,
      "epoch": 0.7057529864547408,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004958563498519327,
      "loss": 6.1641,
      "mean_token_accuracy": 0.15290121510624885,
      "num_tokens": 13812905.0,
      "step": 7060
    },
    {
      "entropy": 6.303525018692016,
      "epoch": 0.7062528115159694,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004958495293072235,
      "loss": 6.2847,
      "mean_token_accuracy": 0.13876570016145706,
      "num_tokens": 13823456.0,
      "step": 7065
    },
    {
      "entropy": 6.311851072311401,
      "epoch": 0.706752636577198,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004958427032059971,
      "loss": 6.1348,
      "mean_token_accuracy": 0.1579107627272606,
      "num_tokens": 13832110.0,
      "step": 7070
    },
    {
      "entropy": 6.203722143173218,
      "epoch": 0.7072524616384266,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004958358715484251,
      "loss": 6.1431,
      "mean_token_accuracy": 0.15215402022004126,
      "num_tokens": 13841637.0,
      "step": 7075
    },
    {
      "entropy": 6.217518424987793,
      "epoch": 0.7077522866996552,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958290343346795,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14743370562791824,
      "num_tokens": 13852185.0,
      "step": 7080
    },
    {
      "entropy": 6.283063077926636,
      "epoch": 0.7082521117608837,
      "grad_norm": 1.25,
      "learning_rate": 0.000495822191564932,
      "loss": 6.0729,
      "mean_token_accuracy": 0.153565713763237,
      "num_tokens": 13862469.0,
      "step": 7085
    },
    {
      "entropy": 6.300814390182495,
      "epoch": 0.7087519368221122,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004958153432393553,
      "loss": 6.1602,
      "mean_token_accuracy": 0.1431555725634098,
      "num_tokens": 13872472.0,
      "step": 7090
    },
    {
      "entropy": 6.250600051879883,
      "epoch": 0.7092517618833408,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495808489358121,
      "loss": 6.1929,
      "mean_token_accuracy": 0.14235619083046913,
      "num_tokens": 13882622.0,
      "step": 7095
    },
    {
      "entropy": 6.2937623977661135,
      "epoch": 0.7097515869445694,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004958016299214022,
      "loss": 6.2032,
      "mean_token_accuracy": 0.14268952384591102,
      "num_tokens": 13891836.0,
      "step": 7100
    },
    {
      "entropy": 6.349701642990112,
      "epoch": 0.710251412005798,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957947649293712,
      "loss": 6.1016,
      "mean_token_accuracy": 0.14699932411313058,
      "num_tokens": 13901555.0,
      "step": 7105
    },
    {
      "entropy": 6.256446266174317,
      "epoch": 0.7107512370670266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004957878943822006,
      "loss": 6.1936,
      "mean_token_accuracy": 0.14552043229341508,
      "num_tokens": 13911145.0,
      "step": 7110
    },
    {
      "entropy": 6.1745147705078125,
      "epoch": 0.7112510621282551,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004957810182800634,
      "loss": 6.0952,
      "mean_token_accuracy": 0.14964592903852464,
      "num_tokens": 13920375.0,
      "step": 7115
    },
    {
      "entropy": 6.233187103271485,
      "epoch": 0.7117508871894836,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004957741366231327,
      "loss": 6.0953,
      "mean_token_accuracy": 0.15257853493094445,
      "num_tokens": 13930175.0,
      "step": 7120
    },
    {
      "entropy": 6.248271131515503,
      "epoch": 0.7122507122507122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957672494115815,
      "loss": 6.195,
      "mean_token_accuracy": 0.1488349109888077,
      "num_tokens": 13939935.0,
      "step": 7125
    },
    {
      "entropy": 6.28034143447876,
      "epoch": 0.7127505373119408,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000495760356645583,
      "loss": 6.1527,
      "mean_token_accuracy": 0.14131171256303787,
      "num_tokens": 13950356.0,
      "step": 7130
    },
    {
      "entropy": 6.324137735366821,
      "epoch": 0.7132503623731694,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004957534583253108,
      "loss": 6.1651,
      "mean_token_accuracy": 0.1476749062538147,
      "num_tokens": 13960661.0,
      "step": 7135
    },
    {
      "entropy": 6.295556163787841,
      "epoch": 0.713750187434398,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004957465544509384,
      "loss": 6.1833,
      "mean_token_accuracy": 0.14510277658700943,
      "num_tokens": 13970629.0,
      "step": 7140
    },
    {
      "entropy": 6.241059923171997,
      "epoch": 0.7142500124956266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004957396450226395,
      "loss": 6.0801,
      "mean_token_accuracy": 0.1525458239018917,
      "num_tokens": 13979801.0,
      "step": 7145
    },
    {
      "entropy": 6.330828714370727,
      "epoch": 0.7147498375568551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957327300405879,
      "loss": 6.3,
      "mean_token_accuracy": 0.15204864889383315,
      "num_tokens": 13989749.0,
      "step": 7150
    },
    {
      "entropy": 6.254339504241943,
      "epoch": 0.7152496626180836,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957258095049577,
      "loss": 6.1774,
      "mean_token_accuracy": 0.14385032206773757,
      "num_tokens": 13999483.0,
      "step": 7155
    },
    {
      "entropy": 6.217577409744263,
      "epoch": 0.7157494876793122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004957188834159229,
      "loss": 6.1431,
      "mean_token_accuracy": 0.1550522267818451,
      "num_tokens": 14009243.0,
      "step": 7160
    },
    {
      "entropy": 6.2134449005126955,
      "epoch": 0.7162493127405408,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004957119517736577,
      "loss": 6.1186,
      "mean_token_accuracy": 0.14946237355470657,
      "num_tokens": 14020452.0,
      "step": 7165
    },
    {
      "entropy": 6.250994253158569,
      "epoch": 0.7167491378017694,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004957050145783366,
      "loss": 6.1275,
      "mean_token_accuracy": 0.14473709985613822,
      "num_tokens": 14030446.0,
      "step": 7170
    },
    {
      "entropy": 6.305724859237671,
      "epoch": 0.717248962862998,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004956980718301342,
      "loss": 6.3376,
      "mean_token_accuracy": 0.1401042826473713,
      "num_tokens": 14040742.0,
      "step": 7175
    },
    {
      "entropy": 6.271581792831421,
      "epoch": 0.7177487879242265,
      "grad_norm": 1.9375,
      "learning_rate": 0.0004956911235292251,
      "loss": 6.1621,
      "mean_token_accuracy": 0.14784373790025712,
      "num_tokens": 14050255.0,
      "step": 7180
    },
    {
      "entropy": 6.274391222000122,
      "epoch": 0.7182486129854551,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004956841696757841,
      "loss": 6.0816,
      "mean_token_accuracy": 0.15628227069973946,
      "num_tokens": 14059745.0,
      "step": 7185
    },
    {
      "entropy": 6.157272386550903,
      "epoch": 0.7187484380466836,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004956772102699862,
      "loss": 6.1275,
      "mean_token_accuracy": 0.1462073117494583,
      "num_tokens": 14069741.0,
      "step": 7190
    },
    {
      "entropy": 6.234292507171631,
      "epoch": 0.7192482631079122,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956702453120064,
      "loss": 6.1853,
      "mean_token_accuracy": 0.14424651265144348,
      "num_tokens": 14078758.0,
      "step": 7195
    },
    {
      "entropy": 6.3222438335418705,
      "epoch": 0.7197480881691408,
      "grad_norm": 1.25,
      "learning_rate": 0.0004956632748020201,
      "loss": 6.2453,
      "mean_token_accuracy": 0.14846357256174086,
      "num_tokens": 14087770.0,
      "step": 7200
    },
    {
      "entropy": 6.275791835784912,
      "epoch": 0.7202479132303694,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004956562987402026,
      "loss": 6.2338,
      "mean_token_accuracy": 0.14000062867999077,
      "num_tokens": 14097139.0,
      "step": 7205
    },
    {
      "entropy": 6.192299461364746,
      "epoch": 0.720747738291598,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004956493171267293,
      "loss": 6.0733,
      "mean_token_accuracy": 0.14988180696964265,
      "num_tokens": 14107040.0,
      "step": 7210
    },
    {
      "entropy": 6.35655255317688,
      "epoch": 0.7212475633528265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000495642329961776,
      "loss": 6.2099,
      "mean_token_accuracy": 0.14649844840168952,
      "num_tokens": 14117103.0,
      "step": 7215
    },
    {
      "entropy": 6.348765802383423,
      "epoch": 0.7217473884140551,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004956353372455185,
      "loss": 6.2106,
      "mean_token_accuracy": 0.14905840307474136,
      "num_tokens": 14127608.0,
      "step": 7220
    },
    {
      "entropy": 6.208347415924072,
      "epoch": 0.7222472134752836,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004956283389781327,
      "loss": 6.1475,
      "mean_token_accuracy": 0.14699290841817855,
      "num_tokens": 14137618.0,
      "step": 7225
    },
    {
      "entropy": 6.224096822738647,
      "epoch": 0.7227470385365122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956213351597946,
      "loss": 6.1828,
      "mean_token_accuracy": 0.14870848506689072,
      "num_tokens": 14147285.0,
      "step": 7230
    },
    {
      "entropy": 6.277980661392212,
      "epoch": 0.7232468635977408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956143257906805,
      "loss": 6.1121,
      "mean_token_accuracy": 0.1521148107945919,
      "num_tokens": 14156398.0,
      "step": 7235
    },
    {
      "entropy": 6.171188735961914,
      "epoch": 0.7237466886589694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004956073108709667,
      "loss": 6.1939,
      "mean_token_accuracy": 0.14421972930431365,
      "num_tokens": 14166562.0,
      "step": 7240
    },
    {
      "entropy": 6.219372177124024,
      "epoch": 0.7242465137201979,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004956002904008298,
      "loss": 6.0409,
      "mean_token_accuracy": 0.1552780032157898,
      "num_tokens": 14175824.0,
      "step": 7245
    },
    {
      "entropy": 6.315196132659912,
      "epoch": 0.7247463387814265,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004955932643804463,
      "loss": 6.1125,
      "mean_token_accuracy": 0.14499305933713913,
      "num_tokens": 14185181.0,
      "step": 7250
    },
    {
      "entropy": 6.29328498840332,
      "epoch": 0.7252461638426551,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000495586232809993,
      "loss": 6.2337,
      "mean_token_accuracy": 0.1446804493665695,
      "num_tokens": 14194622.0,
      "step": 7255
    },
    {
      "entropy": 6.267118644714356,
      "epoch": 0.7257459889038836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955791956896469,
      "loss": 6.1858,
      "mean_token_accuracy": 0.14519859850406647,
      "num_tokens": 14204758.0,
      "step": 7260
    },
    {
      "entropy": 6.268581819534302,
      "epoch": 0.7262458139651122,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495572153019585,
      "loss": 6.1546,
      "mean_token_accuracy": 0.14576357752084732,
      "num_tokens": 14214926.0,
      "step": 7265
    },
    {
      "entropy": 6.283501768112183,
      "epoch": 0.7267456390263408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955651047999844,
      "loss": 6.1953,
      "mean_token_accuracy": 0.14461069256067277,
      "num_tokens": 14225372.0,
      "step": 7270
    },
    {
      "entropy": 6.28266487121582,
      "epoch": 0.7272454640875694,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004955580510310226,
      "loss": 6.1299,
      "mean_token_accuracy": 0.15068346112966538,
      "num_tokens": 14235676.0,
      "step": 7275
    },
    {
      "entropy": 6.273680973052978,
      "epoch": 0.7277452891487979,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004955509917128771,
      "loss": 6.2357,
      "mean_token_accuracy": 0.13851522281765938,
      "num_tokens": 14244196.0,
      "step": 7280
    },
    {
      "entropy": 6.262936115264893,
      "epoch": 0.7282451142100265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004955439268457252,
      "loss": 6.2105,
      "mean_token_accuracy": 0.1412004441022873,
      "num_tokens": 14253648.0,
      "step": 7285
    },
    {
      "entropy": 6.382965850830078,
      "epoch": 0.728744939271255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004955368564297449,
      "loss": 6.3083,
      "mean_token_accuracy": 0.13853543475270272,
      "num_tokens": 14264373.0,
      "step": 7290
    },
    {
      "entropy": 6.259505748748779,
      "epoch": 0.7292447643324836,
      "grad_norm": 1.375,
      "learning_rate": 0.0004955297804651142,
      "loss": 6.0897,
      "mean_token_accuracy": 0.15454344004392623,
      "num_tokens": 14274544.0,
      "step": 7295
    },
    {
      "entropy": 6.2876362800598145,
      "epoch": 0.7297445893937122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004955226989520107,
      "loss": 6.2146,
      "mean_token_accuracy": 0.13700388744473457,
      "num_tokens": 14284648.0,
      "step": 7300
    },
    {
      "entropy": 6.232805585861206,
      "epoch": 0.7302444144549408,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000495515611890613,
      "loss": 6.1373,
      "mean_token_accuracy": 0.14799095541238785,
      "num_tokens": 14294308.0,
      "step": 7305
    },
    {
      "entropy": 6.2365528583526615,
      "epoch": 0.7307442395161693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004955085192810993,
      "loss": 6.1854,
      "mean_token_accuracy": 0.14716747924685478,
      "num_tokens": 14304965.0,
      "step": 7310
    },
    {
      "entropy": 6.252486324310302,
      "epoch": 0.7312440645773979,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004955014211236479,
      "loss": 6.1624,
      "mean_token_accuracy": 0.1502046875655651,
      "num_tokens": 14315880.0,
      "step": 7315
    },
    {
      "entropy": 6.229627227783203,
      "epoch": 0.7317438896386265,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004954943174184375,
      "loss": 6.2201,
      "mean_token_accuracy": 0.14457692056894303,
      "num_tokens": 14326316.0,
      "step": 7320
    },
    {
      "entropy": 6.230609941482544,
      "epoch": 0.732243714699855,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004954872081656469,
      "loss": 6.1507,
      "mean_token_accuracy": 0.1498189941048622,
      "num_tokens": 14335802.0,
      "step": 7325
    },
    {
      "entropy": 6.2368857860565186,
      "epoch": 0.7327435397610836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004954800933654548,
      "loss": 6.1394,
      "mean_token_accuracy": 0.1550527609884739,
      "num_tokens": 14346050.0,
      "step": 7330
    },
    {
      "entropy": 6.287402200698852,
      "epoch": 0.7332433648223122,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004954729730180403,
      "loss": 6.0942,
      "mean_token_accuracy": 0.14972051680088044,
      "num_tokens": 14355483.0,
      "step": 7335
    },
    {
      "entropy": 6.263219690322876,
      "epoch": 0.7337431898835408,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004954658471235825,
      "loss": 6.0704,
      "mean_token_accuracy": 0.15057512819767,
      "num_tokens": 14364649.0,
      "step": 7340
    },
    {
      "entropy": 6.178263998031616,
      "epoch": 0.7342430149447693,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004954587156822607,
      "loss": 6.079,
      "mean_token_accuracy": 0.15476150885224343,
      "num_tokens": 14374495.0,
      "step": 7345
    },
    {
      "entropy": 6.247107601165771,
      "epoch": 0.7347428400059979,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004954515786942544,
      "loss": 6.1749,
      "mean_token_accuracy": 0.14233332872390747,
      "num_tokens": 14384247.0,
      "step": 7350
    },
    {
      "entropy": 6.27628173828125,
      "epoch": 0.7352426650672265,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495444436159743,
      "loss": 6.2306,
      "mean_token_accuracy": 0.135698539018631,
      "num_tokens": 14393675.0,
      "step": 7355
    },
    {
      "entropy": 6.25293550491333,
      "epoch": 0.735742490128455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004954372880789064,
      "loss": 6.1214,
      "mean_token_accuracy": 0.14554580748081208,
      "num_tokens": 14403579.0,
      "step": 7360
    },
    {
      "entropy": 6.34037594795227,
      "epoch": 0.7362423151896836,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004954301344519244,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14594241753220558,
      "num_tokens": 14413506.0,
      "step": 7365
    },
    {
      "entropy": 6.327856063842773,
      "epoch": 0.7367421402509122,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004954229752789769,
      "loss": 6.2728,
      "mean_token_accuracy": 0.14849380701780318,
      "num_tokens": 14422960.0,
      "step": 7370
    },
    {
      "entropy": 6.291201210021972,
      "epoch": 0.7372419653121407,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004954158105602441,
      "loss": 6.2646,
      "mean_token_accuracy": 0.14501268416643143,
      "num_tokens": 14432207.0,
      "step": 7375
    },
    {
      "entropy": 6.235497999191284,
      "epoch": 0.7377417903733693,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004954086402959061,
      "loss": 6.2419,
      "mean_token_accuracy": 0.1394680231809616,
      "num_tokens": 14441785.0,
      "step": 7380
    },
    {
      "entropy": 6.334630489349365,
      "epoch": 0.7382416154345979,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004954014644861435,
      "loss": 6.1719,
      "mean_token_accuracy": 0.14647593796253205,
      "num_tokens": 14451320.0,
      "step": 7385
    },
    {
      "entropy": 6.31368465423584,
      "epoch": 0.7387414404958265,
      "grad_norm": 1.25,
      "learning_rate": 0.0004953942831311367,
      "loss": 6.3051,
      "mean_token_accuracy": 0.14492149651050568,
      "num_tokens": 14460777.0,
      "step": 7390
    },
    {
      "entropy": 6.2608801364898685,
      "epoch": 0.739241265557055,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004953870962310666,
      "loss": 6.159,
      "mean_token_accuracy": 0.15158595144748688,
      "num_tokens": 14469729.0,
      "step": 7395
    },
    {
      "entropy": 6.1772789478302,
      "epoch": 0.7397410906182836,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953799037861138,
      "loss": 6.0997,
      "mean_token_accuracy": 0.1535406455397606,
      "num_tokens": 14480559.0,
      "step": 7400
    },
    {
      "entropy": 6.290306949615479,
      "epoch": 0.7402409156795122,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004953727057964594,
      "loss": 6.1251,
      "mean_token_accuracy": 0.15217110365629197,
      "num_tokens": 14489631.0,
      "step": 7405
    },
    {
      "entropy": 6.394688892364502,
      "epoch": 0.7407407407407407,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004953655022622843,
      "loss": 6.345,
      "mean_token_accuracy": 0.13441844061017036,
      "num_tokens": 14498994.0,
      "step": 7410
    },
    {
      "entropy": 6.331118488311768,
      "epoch": 0.7412405658019693,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004953582931837699,
      "loss": 6.1644,
      "mean_token_accuracy": 0.14437094256281852,
      "num_tokens": 14508456.0,
      "step": 7415
    },
    {
      "entropy": 6.1827723503112795,
      "epoch": 0.7417403908631979,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004953510785610976,
      "loss": 6.1573,
      "mean_token_accuracy": 0.14701317325234414,
      "num_tokens": 14517388.0,
      "step": 7420
    },
    {
      "entropy": 6.3219828605651855,
      "epoch": 0.7422402159244265,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953438583944487,
      "loss": 6.2402,
      "mean_token_accuracy": 0.1438342235982418,
      "num_tokens": 14526889.0,
      "step": 7425
    },
    {
      "entropy": 6.232723808288574,
      "epoch": 0.742740040985655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004953366326840052,
      "loss": 6.1857,
      "mean_token_accuracy": 0.14498441889882088,
      "num_tokens": 14537115.0,
      "step": 7430
    },
    {
      "entropy": 6.277800750732422,
      "epoch": 0.7432398660468836,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004953294014299486,
      "loss": 6.1497,
      "mean_token_accuracy": 0.1515778958797455,
      "num_tokens": 14547354.0,
      "step": 7435
    },
    {
      "entropy": 6.341536998748779,
      "epoch": 0.7437396911081121,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004953221646324611,
      "loss": 6.3355,
      "mean_token_accuracy": 0.13849153965711594,
      "num_tokens": 14557689.0,
      "step": 7440
    },
    {
      "entropy": 6.269274377822876,
      "epoch": 0.7442395161693407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004953149222917245,
      "loss": 6.1795,
      "mean_token_accuracy": 0.14602164998650552,
      "num_tokens": 14567687.0,
      "step": 7445
    },
    {
      "entropy": 6.316735458374024,
      "epoch": 0.7447393412305693,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004953076744079213,
      "loss": 6.2231,
      "mean_token_accuracy": 0.14437437653541565,
      "num_tokens": 14577800.0,
      "step": 7450
    },
    {
      "entropy": 6.364233303070068,
      "epoch": 0.7452391662917979,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004953004209812335,
      "loss": 6.1552,
      "mean_token_accuracy": 0.14624324440956116,
      "num_tokens": 14587467.0,
      "step": 7455
    },
    {
      "entropy": 6.243222761154175,
      "epoch": 0.7457389913530265,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952931620118439,
      "loss": 6.1339,
      "mean_token_accuracy": 0.15310154408216475,
      "num_tokens": 14597710.0,
      "step": 7460
    },
    {
      "entropy": 6.230521011352539,
      "epoch": 0.746238816414255,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004952858974999351,
      "loss": 6.0804,
      "mean_token_accuracy": 0.15012453943490983,
      "num_tokens": 14607212.0,
      "step": 7465
    },
    {
      "entropy": 6.306824970245361,
      "epoch": 0.7467386414754836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952786274456896,
      "loss": 6.1542,
      "mean_token_accuracy": 0.14083855897188186,
      "num_tokens": 14616237.0,
      "step": 7470
    },
    {
      "entropy": 6.178982543945312,
      "epoch": 0.7472384665367121,
      "grad_norm": 1.25,
      "learning_rate": 0.0004952713518492907,
      "loss": 6.0741,
      "mean_token_accuracy": 0.15001459047198296,
      "num_tokens": 14625645.0,
      "step": 7475
    },
    {
      "entropy": 6.209083366394043,
      "epoch": 0.7477382915979407,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004952640707109213,
      "loss": 6.0696,
      "mean_token_accuracy": 0.14975828379392625,
      "num_tokens": 14634897.0,
      "step": 7480
    },
    {
      "entropy": 6.186454582214355,
      "epoch": 0.7482381166591693,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004952567840307644,
      "loss": 6.1651,
      "mean_token_accuracy": 0.14596059471368789,
      "num_tokens": 14645310.0,
      "step": 7485
    },
    {
      "entropy": 6.310971736907959,
      "epoch": 0.7487379417203979,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004952494918090036,
      "loss": 6.2822,
      "mean_token_accuracy": 0.13946227803826333,
      "num_tokens": 14654983.0,
      "step": 7490
    },
    {
      "entropy": 6.318861770629883,
      "epoch": 0.7492377667816265,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004952421940458222,
      "loss": 6.1727,
      "mean_token_accuracy": 0.14883537366986274,
      "num_tokens": 14664384.0,
      "step": 7495
    },
    {
      "entropy": 6.299776649475097,
      "epoch": 0.749737591842855,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004952348907414039,
      "loss": 6.2019,
      "mean_token_accuracy": 0.1380913533270359,
      "num_tokens": 14674375.0,
      "step": 7500
    },
    {
      "entropy": 6.2658116817474365,
      "epoch": 0.7502374169040835,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004952275818959325,
      "loss": 6.1427,
      "mean_token_accuracy": 0.15061455667018891,
      "num_tokens": 14683212.0,
      "step": 7505
    },
    {
      "entropy": 6.2123131275177,
      "epoch": 0.7507372419653121,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004952202675095916,
      "loss": 6.1221,
      "mean_token_accuracy": 0.14865564331412315,
      "num_tokens": 14693267.0,
      "step": 7510
    },
    {
      "entropy": 6.203077793121338,
      "epoch": 0.7512370670265407,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004952129475825657,
      "loss": 6.1264,
      "mean_token_accuracy": 0.14782924801111222,
      "num_tokens": 14702835.0,
      "step": 7515
    },
    {
      "entropy": 6.2376631736755375,
      "epoch": 0.7517368920877693,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004952056221150385,
      "loss": 6.1523,
      "mean_token_accuracy": 0.14764661490917205,
      "num_tokens": 14713243.0,
      "step": 7520
    },
    {
      "entropy": 6.312833833694458,
      "epoch": 0.7522367171489979,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004951982911071945,
      "loss": 6.2385,
      "mean_token_accuracy": 0.1500071756541729,
      "num_tokens": 14724943.0,
      "step": 7525
    },
    {
      "entropy": 6.308235740661621,
      "epoch": 0.7527365422102265,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004951909545592183,
      "loss": 6.1236,
      "mean_token_accuracy": 0.15015341341495514,
      "num_tokens": 14734812.0,
      "step": 7530
    },
    {
      "entropy": 6.237270402908325,
      "epoch": 0.753236367271455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004951836124712942,
      "loss": 6.231,
      "mean_token_accuracy": 0.14692520275712012,
      "num_tokens": 14745465.0,
      "step": 7535
    },
    {
      "entropy": 6.242190980911255,
      "epoch": 0.7537361923326835,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004951762648436073,
      "loss": 6.1838,
      "mean_token_accuracy": 0.1425321213901043,
      "num_tokens": 14755088.0,
      "step": 7540
    },
    {
      "entropy": 6.386645317077637,
      "epoch": 0.7542360173939121,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004951689116763419,
      "loss": 6.2503,
      "mean_token_accuracy": 0.14557131975889206,
      "num_tokens": 14764866.0,
      "step": 7545
    },
    {
      "entropy": 6.304442691802978,
      "epoch": 0.7547358424551407,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004951615529696835,
      "loss": 6.143,
      "mean_token_accuracy": 0.14637164771556854,
      "num_tokens": 14775528.0,
      "step": 7550
    },
    {
      "entropy": 6.232445240020752,
      "epoch": 0.7552356675163693,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004951541887238172,
      "loss": 6.1363,
      "mean_token_accuracy": 0.1502537950873375,
      "num_tokens": 14785698.0,
      "step": 7555
    },
    {
      "entropy": 6.213312578201294,
      "epoch": 0.7557354925775979,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004951468189389281,
      "loss": 6.1299,
      "mean_token_accuracy": 0.149754199385643,
      "num_tokens": 14794765.0,
      "step": 7560
    },
    {
      "entropy": 6.4043238162994385,
      "epoch": 0.7562353176388265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951394436152016,
      "loss": 6.2909,
      "mean_token_accuracy": 0.1413774698972702,
      "num_tokens": 14804185.0,
      "step": 7565
    },
    {
      "entropy": 6.281268072128296,
      "epoch": 0.7567351427000549,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004951320627528234,
      "loss": 6.0713,
      "mean_token_accuracy": 0.1501740463078022,
      "num_tokens": 14814270.0,
      "step": 7570
    },
    {
      "entropy": 6.122609186172485,
      "epoch": 0.7572349677612835,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004951246763519791,
      "loss": 6.1147,
      "mean_token_accuracy": 0.15486182495951653,
      "num_tokens": 14824879.0,
      "step": 7575
    },
    {
      "entropy": 6.309225749969483,
      "epoch": 0.7577347928225121,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004951172844128546,
      "loss": 6.1801,
      "mean_token_accuracy": 0.1444603182375431,
      "num_tokens": 14834882.0,
      "step": 7580
    },
    {
      "entropy": 6.2771814346313475,
      "epoch": 0.7582346178837407,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004951098869356358,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14502521753311157,
      "num_tokens": 14845518.0,
      "step": 7585
    },
    {
      "entropy": 6.2875982284545895,
      "epoch": 0.7587344429449693,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000495102483920509,
      "loss": 6.2324,
      "mean_token_accuracy": 0.140113715082407,
      "num_tokens": 14856046.0,
      "step": 7590
    },
    {
      "entropy": 6.394089603424073,
      "epoch": 0.7592342680061979,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004950950753676603,
      "loss": 6.3484,
      "mean_token_accuracy": 0.13357651680707933,
      "num_tokens": 14865214.0,
      "step": 7595
    },
    {
      "entropy": 6.282820510864258,
      "epoch": 0.7597340930674265,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004950876612772763,
      "loss": 6.1319,
      "mean_token_accuracy": 0.1496227778494358,
      "num_tokens": 14875708.0,
      "step": 7600
    },
    {
      "entropy": 6.187641334533692,
      "epoch": 0.7602339181286549,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004950802416495431,
      "loss": 6.1799,
      "mean_token_accuracy": 0.14271902963519095,
      "num_tokens": 14886041.0,
      "step": 7605
    },
    {
      "entropy": 6.189352798461914,
      "epoch": 0.7607337431898835,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004950728164846478,
      "loss": 6.0414,
      "mean_token_accuracy": 0.15440895408391953,
      "num_tokens": 14895824.0,
      "step": 7610
    },
    {
      "entropy": 6.217437076568603,
      "epoch": 0.7612335682511121,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950653857827771,
      "loss": 6.0872,
      "mean_token_accuracy": 0.15182681530714034,
      "num_tokens": 14904946.0,
      "step": 7615
    },
    {
      "entropy": 6.276913452148437,
      "epoch": 0.7617333933123407,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950579495441177,
      "loss": 6.0619,
      "mean_token_accuracy": 0.1454996719956398,
      "num_tokens": 14914088.0,
      "step": 7620
    },
    {
      "entropy": 6.251385498046875,
      "epoch": 0.7622332183735693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004950505077688572,
      "loss": 6.2515,
      "mean_token_accuracy": 0.1402482844889164,
      "num_tokens": 14924780.0,
      "step": 7625
    },
    {
      "entropy": 6.189641332626342,
      "epoch": 0.7627330434347979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004950430604571824,
      "loss": 6.0812,
      "mean_token_accuracy": 0.14813830256462096,
      "num_tokens": 14933857.0,
      "step": 7630
    },
    {
      "entropy": 6.2296754837036135,
      "epoch": 0.7632328684960263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004950356076092808,
      "loss": 6.1229,
      "mean_token_accuracy": 0.15082956999540328,
      "num_tokens": 14944223.0,
      "step": 7635
    },
    {
      "entropy": 6.3322978019714355,
      "epoch": 0.7637326935572549,
      "grad_norm": 1.515625,
      "learning_rate": 0.00049502814922534,
      "loss": 6.3396,
      "mean_token_accuracy": 0.13234080672264098,
      "num_tokens": 14953677.0,
      "step": 7640
    },
    {
      "entropy": 6.228654909133911,
      "epoch": 0.7642325186184835,
      "grad_norm": 1.125,
      "learning_rate": 0.0004950206853055475,
      "loss": 6.1564,
      "mean_token_accuracy": 0.14697617292404175,
      "num_tokens": 14963883.0,
      "step": 7645
    },
    {
      "entropy": 6.2224109172821045,
      "epoch": 0.7647323436797121,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950132158500913,
      "loss": 6.1264,
      "mean_token_accuracy": 0.14641739577054977,
      "num_tokens": 14973680.0,
      "step": 7650
    },
    {
      "entropy": 6.285792827606201,
      "epoch": 0.7652321687409407,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004950057408591591,
      "loss": 6.0651,
      "mean_token_accuracy": 0.1546121656894684,
      "num_tokens": 14982492.0,
      "step": 7655
    },
    {
      "entropy": 6.2460472106933596,
      "epoch": 0.7657319938021693,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004949982603329391,
      "loss": 6.0194,
      "mean_token_accuracy": 0.15341078341007233,
      "num_tokens": 14991555.0,
      "step": 7660
    },
    {
      "entropy": 6.3247962474823,
      "epoch": 0.7662318188633979,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004949907742716195,
      "loss": 6.266,
      "mean_token_accuracy": 0.13541306406259537,
      "num_tokens": 15000678.0,
      "step": 7665
    },
    {
      "entropy": 6.223152685165405,
      "epoch": 0.7667316439246263,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004949832826753886,
      "loss": 6.1273,
      "mean_token_accuracy": 0.1462394155561924,
      "num_tokens": 15010418.0,
      "step": 7670
    },
    {
      "entropy": 6.27063307762146,
      "epoch": 0.7672314689858549,
      "grad_norm": 1.140625,
      "learning_rate": 0.000494975785544435,
      "loss": 6.2474,
      "mean_token_accuracy": 0.13888389691710473,
      "num_tokens": 15020829.0,
      "step": 7675
    },
    {
      "entropy": 6.264683103561401,
      "epoch": 0.7677312940470835,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004949682828789472,
      "loss": 6.1368,
      "mean_token_accuracy": 0.1456165038049221,
      "num_tokens": 15030055.0,
      "step": 7680
    },
    {
      "entropy": 6.247903203964233,
      "epoch": 0.7682311191083121,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494960774679114,
      "loss": 6.1002,
      "mean_token_accuracy": 0.14857350364327432,
      "num_tokens": 15039620.0,
      "step": 7685
    },
    {
      "entropy": 6.256641912460327,
      "epoch": 0.7687309441695407,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004949532609451242,
      "loss": 6.186,
      "mean_token_accuracy": 0.14775176122784614,
      "num_tokens": 15049299.0,
      "step": 7690
    },
    {
      "entropy": 6.208614873886108,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.25,
      "learning_rate": 0.0004949457416771671,
      "loss": 6.1209,
      "mean_token_accuracy": 0.14751720875501634,
      "num_tokens": 15059554.0,
      "step": 7695
    },
    {
      "entropy": 6.091894435882568,
      "epoch": 0.7697305942919977,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004949382168754316,
      "loss": 6.0403,
      "mean_token_accuracy": 0.1542319692671299,
      "num_tokens": 15069651.0,
      "step": 7700
    },
    {
      "entropy": 6.314790678024292,
      "epoch": 0.7702304193532263,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004949306865401073,
      "loss": 6.1645,
      "mean_token_accuracy": 0.1441565252840519,
      "num_tokens": 15080000.0,
      "step": 7705
    },
    {
      "entropy": 6.260981798171997,
      "epoch": 0.7707302444144549,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004949231506713833,
      "loss": 6.1535,
      "mean_token_accuracy": 0.1464677169919014,
      "num_tokens": 15089237.0,
      "step": 7710
    },
    {
      "entropy": 6.116284322738648,
      "epoch": 0.7712300694756835,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004949156092694496,
      "loss": 6.0173,
      "mean_token_accuracy": 0.1555821865797043,
      "num_tokens": 15097829.0,
      "step": 7715
    },
    {
      "entropy": 6.227576351165771,
      "epoch": 0.7717298945369121,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004949080623344958,
      "loss": 6.1703,
      "mean_token_accuracy": 0.146850772947073,
      "num_tokens": 15106697.0,
      "step": 7720
    },
    {
      "entropy": 6.199356746673584,
      "epoch": 0.7722297195981407,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949005098667117,
      "loss": 5.9889,
      "mean_token_accuracy": 0.15549849420785905,
      "num_tokens": 15115003.0,
      "step": 7725
    },
    {
      "entropy": 6.293274450302124,
      "epoch": 0.7727295446593693,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004948929518662873,
      "loss": 6.2308,
      "mean_token_accuracy": 0.1477031335234642,
      "num_tokens": 15125281.0,
      "step": 7730
    },
    {
      "entropy": 6.200959300994873,
      "epoch": 0.7732293697205977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948853883334129,
      "loss": 6.0855,
      "mean_token_accuracy": 0.1566949427127838,
      "num_tokens": 15134609.0,
      "step": 7735
    },
    {
      "entropy": 6.310623025894165,
      "epoch": 0.7737291947818263,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948778192682787,
      "loss": 6.2442,
      "mean_token_accuracy": 0.14099528640508652,
      "num_tokens": 15143922.0,
      "step": 7740
    },
    {
      "entropy": 6.2802393436431885,
      "epoch": 0.7742290198430549,
      "grad_norm": 1.28125,
      "learning_rate": 0.000494870244671075,
      "loss": 6.1644,
      "mean_token_accuracy": 0.1503096915781498,
      "num_tokens": 15153490.0,
      "step": 7745
    },
    {
      "entropy": 6.276309823989868,
      "epoch": 0.7747288449042835,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004948626645419926,
      "loss": 6.1653,
      "mean_token_accuracy": 0.14215842336416246,
      "num_tokens": 15162422.0,
      "step": 7750
    },
    {
      "entropy": 6.253599500656128,
      "epoch": 0.7752286699655121,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004948550788812222,
      "loss": 6.1044,
      "mean_token_accuracy": 0.14733337759971618,
      "num_tokens": 15171016.0,
      "step": 7755
    },
    {
      "entropy": 6.250903224945068,
      "epoch": 0.7757284950267407,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004948474876889545,
      "loss": 6.065,
      "mean_token_accuracy": 0.1572606235742569,
      "num_tokens": 15179880.0,
      "step": 7760
    },
    {
      "entropy": 6.208278179168701,
      "epoch": 0.7762283200879692,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004948398909653806,
      "loss": 6.1996,
      "mean_token_accuracy": 0.14867712706327438,
      "num_tokens": 15190412.0,
      "step": 7765
    },
    {
      "entropy": 6.301343250274658,
      "epoch": 0.7767281451491977,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948322887106917,
      "loss": 6.3124,
      "mean_token_accuracy": 0.13784149289131165,
      "num_tokens": 15201311.0,
      "step": 7770
    },
    {
      "entropy": 6.3385007858276365,
      "epoch": 0.7772279702104263,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004948246809250788,
      "loss": 6.1799,
      "mean_token_accuracy": 0.14733841866254807,
      "num_tokens": 15210058.0,
      "step": 7775
    },
    {
      "entropy": 6.3492076873779295,
      "epoch": 0.7777277952716549,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004948170676087338,
      "loss": 6.2143,
      "mean_token_accuracy": 0.14083099141716957,
      "num_tokens": 15220023.0,
      "step": 7780
    },
    {
      "entropy": 6.1954363822937015,
      "epoch": 0.7782276203328835,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004948094487618476,
      "loss": 6.0929,
      "mean_token_accuracy": 0.1523064285516739,
      "num_tokens": 15229757.0,
      "step": 7785
    },
    {
      "entropy": 6.264597034454345,
      "epoch": 0.7787274453941121,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948018243846126,
      "loss": 6.1734,
      "mean_token_accuracy": 0.14349630177021028,
      "num_tokens": 15238928.0,
      "step": 7790
    },
    {
      "entropy": 6.167093515396118,
      "epoch": 0.7792272704553407,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004947941944772199,
      "loss": 6.1521,
      "mean_token_accuracy": 0.15023285523056984,
      "num_tokens": 15248864.0,
      "step": 7795
    },
    {
      "entropy": 6.309709453582764,
      "epoch": 0.7797270955165692,
      "grad_norm": 1.171875,
      "learning_rate": 0.000494786559039862,
      "loss": 6.2512,
      "mean_token_accuracy": 0.1476570710539818,
      "num_tokens": 15259135.0,
      "step": 7800
    },
    {
      "entropy": 6.262609767913818,
      "epoch": 0.7802269205777977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004947789180727308,
      "loss": 6.0479,
      "mean_token_accuracy": 0.15698638558387756,
      "num_tokens": 15268410.0,
      "step": 7805
    },
    {
      "entropy": 6.231771993637085,
      "epoch": 0.7807267456390263,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004947712715760186,
      "loss": 6.0896,
      "mean_token_accuracy": 0.14790061265230178,
      "num_tokens": 15277863.0,
      "step": 7810
    },
    {
      "entropy": 6.25667176246643,
      "epoch": 0.7812265707002549,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004947636195499177,
      "loss": 6.1512,
      "mean_token_accuracy": 0.14559535309672356,
      "num_tokens": 15286638.0,
      "step": 7815
    },
    {
      "entropy": 6.249594116210938,
      "epoch": 0.7817263957614835,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004947559619946207,
      "loss": 6.0943,
      "mean_token_accuracy": 0.14469757080078124,
      "num_tokens": 15296395.0,
      "step": 7820
    },
    {
      "entropy": 6.205122089385986,
      "epoch": 0.7822262208227121,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004947482989103203,
      "loss": 6.1698,
      "mean_token_accuracy": 0.15072274953126907,
      "num_tokens": 15305940.0,
      "step": 7825
    },
    {
      "entropy": 6.228153085708618,
      "epoch": 0.7827260458839406,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004947406302972092,
      "loss": 6.1632,
      "mean_token_accuracy": 0.14219047501683235,
      "num_tokens": 15315803.0,
      "step": 7830
    },
    {
      "entropy": 6.280290031433106,
      "epoch": 0.7832258709451692,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004947329561554804,
      "loss": 6.2002,
      "mean_token_accuracy": 0.1436838574707508,
      "num_tokens": 15325742.0,
      "step": 7835
    },
    {
      "entropy": 6.2278694152832035,
      "epoch": 0.7837256960063977,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004947252764853269,
      "loss": 6.0965,
      "mean_token_accuracy": 0.14611087813973428,
      "num_tokens": 15336366.0,
      "step": 7840
    },
    {
      "entropy": 6.284488773345947,
      "epoch": 0.7842255210676263,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004947175912869421,
      "loss": 6.1606,
      "mean_token_accuracy": 0.14678517505526542,
      "num_tokens": 15347044.0,
      "step": 7845
    },
    {
      "entropy": 6.240821409225464,
      "epoch": 0.7847253461288549,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004947099005605191,
      "loss": 6.0529,
      "mean_token_accuracy": 0.15114965066313743,
      "num_tokens": 15356397.0,
      "step": 7850
    },
    {
      "entropy": 6.21326904296875,
      "epoch": 0.7852251711900835,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004947022043062517,
      "loss": 6.1341,
      "mean_token_accuracy": 0.14893074035644532,
      "num_tokens": 15366105.0,
      "step": 7855
    },
    {
      "entropy": 6.288484334945679,
      "epoch": 0.7857249962513121,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004946945025243333,
      "loss": 6.2266,
      "mean_token_accuracy": 0.14281736463308334,
      "num_tokens": 15375377.0,
      "step": 7860
    },
    {
      "entropy": 6.313417768478393,
      "epoch": 0.7862248213125406,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004946867952149577,
      "loss": 6.2099,
      "mean_token_accuracy": 0.14447736591100693,
      "num_tokens": 15385718.0,
      "step": 7865
    },
    {
      "entropy": 6.348818683624268,
      "epoch": 0.7867246463737692,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004946790823783189,
      "loss": 6.2251,
      "mean_token_accuracy": 0.1428031288087368,
      "num_tokens": 15395032.0,
      "step": 7870
    },
    {
      "entropy": 6.249912548065185,
      "epoch": 0.7872244714349977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000494671364014611,
      "loss": 6.1955,
      "mean_token_accuracy": 0.14394095316529273,
      "num_tokens": 15405023.0,
      "step": 7875
    },
    {
      "entropy": 6.202657413482666,
      "epoch": 0.7877242964962263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000494663640124028,
      "loss": 6.0866,
      "mean_token_accuracy": 0.15288192108273507,
      "num_tokens": 15414174.0,
      "step": 7880
    },
    {
      "entropy": 6.2145037174224855,
      "epoch": 0.7882241215574549,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004946559107067643,
      "loss": 6.1226,
      "mean_token_accuracy": 0.14434168487787247,
      "num_tokens": 15423622.0,
      "step": 7885
    },
    {
      "entropy": 6.280716896057129,
      "epoch": 0.7887239466186835,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004946481757630146,
      "loss": 6.171,
      "mean_token_accuracy": 0.1485359862446785,
      "num_tokens": 15433873.0,
      "step": 7890
    },
    {
      "entropy": 6.247428560256958,
      "epoch": 0.789223771679912,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004946404352929732,
      "loss": 6.0438,
      "mean_token_accuracy": 0.1546467885375023,
      "num_tokens": 15443600.0,
      "step": 7895
    },
    {
      "entropy": 6.213374900817871,
      "epoch": 0.7897235967411406,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494632689296835,
      "loss": 6.1601,
      "mean_token_accuracy": 0.15238333344459534,
      "num_tokens": 15454531.0,
      "step": 7900
    },
    {
      "entropy": 6.1597654819488525,
      "epoch": 0.7902234218023692,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004946249377747947,
      "loss": 6.0507,
      "mean_token_accuracy": 0.15995033979415893,
      "num_tokens": 15464313.0,
      "step": 7905
    },
    {
      "entropy": 6.253023529052735,
      "epoch": 0.7907232468635977,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004946171807270476,
      "loss": 6.1471,
      "mean_token_accuracy": 0.1469114378094673,
      "num_tokens": 15473363.0,
      "step": 7910
    },
    {
      "entropy": 6.28628396987915,
      "epoch": 0.7912230719248263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004946094181537887,
      "loss": 6.1814,
      "mean_token_accuracy": 0.14534091129899024,
      "num_tokens": 15483296.0,
      "step": 7915
    },
    {
      "entropy": 6.205390405654907,
      "epoch": 0.7917228969860549,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004946016500552134,
      "loss": 6.1962,
      "mean_token_accuracy": 0.14217067286372184,
      "num_tokens": 15493579.0,
      "step": 7920
    },
    {
      "entropy": 6.291490793228149,
      "epoch": 0.7922227220472835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945938764315171,
      "loss": 6.0933,
      "mean_token_accuracy": 0.14788947030901908,
      "num_tokens": 15502105.0,
      "step": 7925
    },
    {
      "entropy": 6.2864326477050785,
      "epoch": 0.792722547108512,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004945860972828954,
      "loss": 6.0929,
      "mean_token_accuracy": 0.14404995292425155,
      "num_tokens": 15511888.0,
      "step": 7930
    },
    {
      "entropy": 6.209086656570435,
      "epoch": 0.7932223721697406,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004945783126095439,
      "loss": 6.1436,
      "mean_token_accuracy": 0.14489920288324357,
      "num_tokens": 15522469.0,
      "step": 7935
    },
    {
      "entropy": 6.19474983215332,
      "epoch": 0.7937221972309692,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945705224116586,
      "loss": 6.2191,
      "mean_token_accuracy": 0.14202917367219925,
      "num_tokens": 15533332.0,
      "step": 7940
    },
    {
      "entropy": 6.291547632217407,
      "epoch": 0.7942220222921977,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945627266894354,
      "loss": 6.1212,
      "mean_token_accuracy": 0.14555009603500366,
      "num_tokens": 15542861.0,
      "step": 7945
    },
    {
      "entropy": 6.246603965759277,
      "epoch": 0.7947218473534263,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004945549254430707,
      "loss": 6.1071,
      "mean_token_accuracy": 0.14058320075273514,
      "num_tokens": 15552503.0,
      "step": 7950
    },
    {
      "entropy": 6.146363067626953,
      "epoch": 0.7952216724146549,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004945471186727603,
      "loss": 6.0959,
      "mean_token_accuracy": 0.15288271605968476,
      "num_tokens": 15562397.0,
      "step": 7955
    },
    {
      "entropy": 6.3110949993133545,
      "epoch": 0.7957214974758834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945393063787011,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14611591398715973,
      "num_tokens": 15570956.0,
      "step": 7960
    },
    {
      "entropy": 6.275159025192261,
      "epoch": 0.796221322537112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004945314885610894,
      "loss": 6.0463,
      "mean_token_accuracy": 0.15439539402723312,
      "num_tokens": 15579938.0,
      "step": 7965
    },
    {
      "entropy": 6.22276120185852,
      "epoch": 0.7967211475983406,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004945236652201219,
      "loss": 6.1274,
      "mean_token_accuracy": 0.1521826207637787,
      "num_tokens": 15589906.0,
      "step": 7970
    },
    {
      "entropy": 6.261705112457276,
      "epoch": 0.7972209726595691,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004945158363559955,
      "loss": 6.202,
      "mean_token_accuracy": 0.1415676474571228,
      "num_tokens": 15599891.0,
      "step": 7975
    },
    {
      "entropy": 6.295880270004273,
      "epoch": 0.7977207977207977,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004945080019689071,
      "loss": 6.3027,
      "mean_token_accuracy": 0.135616684705019,
      "num_tokens": 15609669.0,
      "step": 7980
    },
    {
      "entropy": 6.344135999679565,
      "epoch": 0.7982206227820263,
      "grad_norm": 1.125,
      "learning_rate": 0.0004945001620590538,
      "loss": 6.219,
      "mean_token_accuracy": 0.14308143854141236,
      "num_tokens": 15619899.0,
      "step": 7985
    },
    {
      "entropy": 6.300057220458984,
      "epoch": 0.7987204478432549,
      "grad_norm": 1.1875,
      "learning_rate": 0.000494492316626633,
      "loss": 6.1962,
      "mean_token_accuracy": 0.14479330629110337,
      "num_tokens": 15629822.0,
      "step": 7990
    },
    {
      "entropy": 6.227403736114502,
      "epoch": 0.7992202729044834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004944844656718419,
      "loss": 6.1323,
      "mean_token_accuracy": 0.1442696876823902,
      "num_tokens": 15639810.0,
      "step": 7995
    },
    {
      "entropy": 6.2585186004638675,
      "epoch": 0.799720097965712,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004944766091948783,
      "loss": 6.1348,
      "mean_token_accuracy": 0.14523012042045594,
      "num_tokens": 15649701.0,
      "step": 8000
    },
    {
      "entropy": 6.307030487060547,
      "epoch": 0.8002199230269406,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004944687471959395,
      "loss": 6.1858,
      "mean_token_accuracy": 0.1499359354376793,
      "num_tokens": 15659969.0,
      "step": 8005
    },
    {
      "entropy": 6.251560497283935,
      "epoch": 0.8007197480881691,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004944608796752237,
      "loss": 6.1445,
      "mean_token_accuracy": 0.14524403363466262,
      "num_tokens": 15669947.0,
      "step": 8010
    },
    {
      "entropy": 6.247959280014038,
      "epoch": 0.8012195731493977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944530066329284,
      "loss": 6.2166,
      "mean_token_accuracy": 0.14227289631962775,
      "num_tokens": 15678799.0,
      "step": 8015
    },
    {
      "entropy": 6.31643443107605,
      "epoch": 0.8017193982106263,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004944451280692521,
      "loss": 6.1353,
      "mean_token_accuracy": 0.145330198854208,
      "num_tokens": 15688859.0,
      "step": 8020
    },
    {
      "entropy": 6.22497444152832,
      "epoch": 0.8022192232718548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004944372439843927,
      "loss": 6.0699,
      "mean_token_accuracy": 0.15040146559476852,
      "num_tokens": 15697121.0,
      "step": 8025
    },
    {
      "entropy": 6.216375207901001,
      "epoch": 0.8027190483330834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004944293543785489,
      "loss": 6.234,
      "mean_token_accuracy": 0.1470854915678501,
      "num_tokens": 15707918.0,
      "step": 8030
    },
    {
      "entropy": 6.308790826797486,
      "epoch": 0.803218873394312,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004944214592519188,
      "loss": 6.1235,
      "mean_token_accuracy": 0.147299225628376,
      "num_tokens": 15717342.0,
      "step": 8035
    },
    {
      "entropy": 6.2364935874938965,
      "epoch": 0.8037186984555406,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004944135586047014,
      "loss": 6.1271,
      "mean_token_accuracy": 0.15071971565485,
      "num_tokens": 15727767.0,
      "step": 8040
    },
    {
      "entropy": 6.234315872192383,
      "epoch": 0.8042185235167691,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004944056524370953,
      "loss": 6.1587,
      "mean_token_accuracy": 0.1487676367163658,
      "num_tokens": 15737049.0,
      "step": 8045
    },
    {
      "entropy": 6.212104558944702,
      "epoch": 0.8047183485779977,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004943977407492994,
      "loss": 6.0738,
      "mean_token_accuracy": 0.14936081022024156,
      "num_tokens": 15746640.0,
      "step": 8050
    },
    {
      "entropy": 6.2690435409545895,
      "epoch": 0.8052181736392263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004943898235415128,
      "loss": 6.1001,
      "mean_token_accuracy": 0.1538916528224945,
      "num_tokens": 15756151.0,
      "step": 8055
    },
    {
      "entropy": 6.118404531478882,
      "epoch": 0.8057179987004548,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004943819008139347,
      "loss": 6.0332,
      "mean_token_accuracy": 0.15337835252285004,
      "num_tokens": 15765226.0,
      "step": 8060
    },
    {
      "entropy": 6.181343841552734,
      "epoch": 0.8062178237616834,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004943739725667644,
      "loss": 6.0783,
      "mean_token_accuracy": 0.1571650892496109,
      "num_tokens": 15775916.0,
      "step": 8065
    },
    {
      "entropy": 6.328836393356323,
      "epoch": 0.806717648822912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004943660388002014,
      "loss": 6.2064,
      "mean_token_accuracy": 0.14186447486281395,
      "num_tokens": 15785660.0,
      "step": 8070
    },
    {
      "entropy": 6.259009885787964,
      "epoch": 0.8072174738841406,
      "grad_norm": 1.25,
      "learning_rate": 0.0004943580995144454,
      "loss": 6.1652,
      "mean_token_accuracy": 0.14962283596396447,
      "num_tokens": 15794995.0,
      "step": 8075
    },
    {
      "entropy": 6.272754669189453,
      "epoch": 0.8077172989453691,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004943501547096961,
      "loss": 6.1448,
      "mean_token_accuracy": 0.1426381826400757,
      "num_tokens": 15804370.0,
      "step": 8080
    },
    {
      "entropy": 6.219016695022583,
      "epoch": 0.8082171240065977,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004943422043861531,
      "loss": 6.0797,
      "mean_token_accuracy": 0.14968337789177893,
      "num_tokens": 15814759.0,
      "step": 8085
    },
    {
      "entropy": 6.2805791854858395,
      "epoch": 0.8087169490678262,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004943342485440169,
      "loss": 6.1442,
      "mean_token_accuracy": 0.15440394282341002,
      "num_tokens": 15823385.0,
      "step": 8090
    },
    {
      "entropy": 6.2116412162780765,
      "epoch": 0.8092167741290548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004943262871834874,
      "loss": 6.1286,
      "mean_token_accuracy": 0.14776497483253478,
      "num_tokens": 15833226.0,
      "step": 8095
    },
    {
      "entropy": 6.3029687881469725,
      "epoch": 0.8097165991902834,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004943183203047649,
      "loss": 6.257,
      "mean_token_accuracy": 0.14333696737885476,
      "num_tokens": 15842565.0,
      "step": 8100
    },
    {
      "entropy": 6.251518869400025,
      "epoch": 0.810216424251512,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004943103479080499,
      "loss": 6.1107,
      "mean_token_accuracy": 0.15320774912834167,
      "num_tokens": 15851816.0,
      "step": 8105
    },
    {
      "entropy": 6.171735429763794,
      "epoch": 0.8107162493127406,
      "grad_norm": 1.5703125,
      "learning_rate": 0.000494302369993543,
      "loss": 6.0538,
      "mean_token_accuracy": 0.1450734816491604,
      "num_tokens": 15861830.0,
      "step": 8110
    },
    {
      "entropy": 6.233936929702759,
      "epoch": 0.8112160743739691,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004942943865614449,
      "loss": 6.1802,
      "mean_token_accuracy": 0.13488271906971933,
      "num_tokens": 15871118.0,
      "step": 8115
    },
    {
      "entropy": 6.277757024765014,
      "epoch": 0.8117158994351977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004942863976119565,
      "loss": 6.1861,
      "mean_token_accuracy": 0.1451020672917366,
      "num_tokens": 15880935.0,
      "step": 8120
    },
    {
      "entropy": 6.28985915184021,
      "epoch": 0.8122157244964262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004942784031452788,
      "loss": 6.1149,
      "mean_token_accuracy": 0.14731285721063614,
      "num_tokens": 15890458.0,
      "step": 8125
    },
    {
      "entropy": 6.165459632873535,
      "epoch": 0.8127155495576548,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004942704031616127,
      "loss": 6.0112,
      "mean_token_accuracy": 0.1544952243566513,
      "num_tokens": 15899194.0,
      "step": 8130
    },
    {
      "entropy": 6.201440048217774,
      "epoch": 0.8132153746188834,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004942623976611598,
      "loss": 6.1518,
      "mean_token_accuracy": 0.15082441344857217,
      "num_tokens": 15909529.0,
      "step": 8135
    },
    {
      "entropy": 6.249154376983642,
      "epoch": 0.813715199680112,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004942543866441215,
      "loss": 6.2129,
      "mean_token_accuracy": 0.1500617727637291,
      "num_tokens": 15919370.0,
      "step": 8140
    },
    {
      "entropy": 6.2765878200531,
      "epoch": 0.8142150247413406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004942463701106992,
      "loss": 6.0726,
      "mean_token_accuracy": 0.1520904041826725,
      "num_tokens": 15928651.0,
      "step": 8145
    },
    {
      "entropy": 6.296664619445801,
      "epoch": 0.8147148498025691,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004942383480610946,
      "loss": 6.0986,
      "mean_token_accuracy": 0.15131353884935378,
      "num_tokens": 15937738.0,
      "step": 8150
    },
    {
      "entropy": 6.333215570449829,
      "epoch": 0.8152146748637976,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004942303204955096,
      "loss": 6.248,
      "mean_token_accuracy": 0.1416212096810341,
      "num_tokens": 15947167.0,
      "step": 8155
    },
    {
      "entropy": 6.245408201217652,
      "epoch": 0.8157144999250262,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004942222874141461,
      "loss": 6.1913,
      "mean_token_accuracy": 0.14824823141098023,
      "num_tokens": 15956611.0,
      "step": 8160
    },
    {
      "entropy": 6.300709295272827,
      "epoch": 0.8162143249862548,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004942142488172063,
      "loss": 6.1559,
      "mean_token_accuracy": 0.14230960607528687,
      "num_tokens": 15967085.0,
      "step": 8165
    },
    {
      "entropy": 6.236518621444702,
      "epoch": 0.8167141500474834,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004942062047048924,
      "loss": 6.1617,
      "mean_token_accuracy": 0.14845765382051468,
      "num_tokens": 15978008.0,
      "step": 8170
    },
    {
      "entropy": 6.293435668945312,
      "epoch": 0.817213975108712,
      "grad_norm": 2.703125,
      "learning_rate": 0.0004941981550774068,
      "loss": 6.3412,
      "mean_token_accuracy": 0.1444306753575802,
      "num_tokens": 15986975.0,
      "step": 8175
    },
    {
      "entropy": 6.151211166381836,
      "epoch": 0.8177138001699406,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004941900999349521,
      "loss": 6.056,
      "mean_token_accuracy": 0.14957771301269532,
      "num_tokens": 15997334.0,
      "step": 8180
    },
    {
      "entropy": 6.177789783477783,
      "epoch": 0.8182136252311691,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004941820392777307,
      "loss": 6.028,
      "mean_token_accuracy": 0.15541132688522338,
      "num_tokens": 16007485.0,
      "step": 8185
    },
    {
      "entropy": 6.273186683654785,
      "epoch": 0.8187134502923976,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004941739731059457,
      "loss": 6.1878,
      "mean_token_accuracy": 0.14398434087634088,
      "num_tokens": 16017874.0,
      "step": 8190
    },
    {
      "entropy": 6.26248025894165,
      "epoch": 0.8192132753536262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004941659014198,
      "loss": 6.2404,
      "mean_token_accuracy": 0.14299032539129258,
      "num_tokens": 16027778.0,
      "step": 8195
    },
    {
      "entropy": 6.28945837020874,
      "epoch": 0.8197131004148548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004941578242194965,
      "loss": 6.1236,
      "mean_token_accuracy": 0.14991429448127747,
      "num_tokens": 16037116.0,
      "step": 8200
    },
    {
      "entropy": 6.360007190704346,
      "epoch": 0.8202129254760834,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004941497415052387,
      "loss": 6.2585,
      "mean_token_accuracy": 0.14210824742913247,
      "num_tokens": 16046890.0,
      "step": 8205
    },
    {
      "entropy": 6.236407661437989,
      "epoch": 0.820712750537312,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004941416532772296,
      "loss": 6.1372,
      "mean_token_accuracy": 0.15049769580364228,
      "num_tokens": 16057870.0,
      "step": 8210
    },
    {
      "entropy": 6.333528518676758,
      "epoch": 0.8212125755985406,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494133559535673,
      "loss": 6.1894,
      "mean_token_accuracy": 0.14113891646265983,
      "num_tokens": 16067578.0,
      "step": 8215
    },
    {
      "entropy": 6.265466642379761,
      "epoch": 0.821712400659769,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004941254602807724,
      "loss": 6.1417,
      "mean_token_accuracy": 0.14570298194885253,
      "num_tokens": 16077108.0,
      "step": 8220
    },
    {
      "entropy": 6.279190969467163,
      "epoch": 0.8222122257209976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004941173555127315,
      "loss": 6.0891,
      "mean_token_accuracy": 0.1519876480102539,
      "num_tokens": 16087864.0,
      "step": 8225
    },
    {
      "entropy": 6.139479637145996,
      "epoch": 0.8227120507822262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004941092452317546,
      "loss": 6.0249,
      "mean_token_accuracy": 0.15852599889039992,
      "num_tokens": 16096853.0,
      "step": 8230
    },
    {
      "entropy": 6.20801272392273,
      "epoch": 0.8232118758434548,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941011294380452,
      "loss": 6.0602,
      "mean_token_accuracy": 0.15360495299100876,
      "num_tokens": 16106012.0,
      "step": 8235
    },
    {
      "entropy": 6.215699481964111,
      "epoch": 0.8237117009046834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004940930081318079,
      "loss": 6.097,
      "mean_token_accuracy": 0.15093111991882324,
      "num_tokens": 16115764.0,
      "step": 8240
    },
    {
      "entropy": 6.240601062774658,
      "epoch": 0.824211525965912,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004940848813132469,
      "loss": 6.1659,
      "mean_token_accuracy": 0.1450254090130329,
      "num_tokens": 16127211.0,
      "step": 8245
    },
    {
      "entropy": 6.262681913375855,
      "epoch": 0.8247113510271405,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004940767489825666,
      "loss": 6.1219,
      "mean_token_accuracy": 0.14916089028120041,
      "num_tokens": 16136204.0,
      "step": 8250
    },
    {
      "entropy": 6.197828865051269,
      "epoch": 0.825211176088369,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004940686111399716,
      "loss": 6.0611,
      "mean_token_accuracy": 0.1544598326086998,
      "num_tokens": 16145411.0,
      "step": 8255
    },
    {
      "entropy": 6.194559097290039,
      "epoch": 0.8257110011495976,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004940604677856669,
      "loss": 6.1911,
      "mean_token_accuracy": 0.14180412963032724,
      "num_tokens": 16156575.0,
      "step": 8260
    },
    {
      "entropy": 6.27162652015686,
      "epoch": 0.8262108262108262,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494052318919857,
      "loss": 6.0981,
      "mean_token_accuracy": 0.147871433198452,
      "num_tokens": 16167387.0,
      "step": 8265
    },
    {
      "entropy": 6.264508533477783,
      "epoch": 0.8267106512720548,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004940441645427473,
      "loss": 6.1034,
      "mean_token_accuracy": 0.1383362501859665,
      "num_tokens": 16177302.0,
      "step": 8270
    },
    {
      "entropy": 6.261044645309449,
      "epoch": 0.8272104763332834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004940360046545427,
      "loss": 6.2132,
      "mean_token_accuracy": 0.14373020455241203,
      "num_tokens": 16186830.0,
      "step": 8275
    },
    {
      "entropy": 6.317985916137696,
      "epoch": 0.827710301394512,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004940278392554486,
      "loss": 6.1628,
      "mean_token_accuracy": 0.14414869099855424,
      "num_tokens": 16197424.0,
      "step": 8280
    },
    {
      "entropy": 6.228089618682861,
      "epoch": 0.8282101264557404,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004940196683456704,
      "loss": 6.1154,
      "mean_token_accuracy": 0.14977436363697053,
      "num_tokens": 16207209.0,
      "step": 8285
    },
    {
      "entropy": 6.2766077518463135,
      "epoch": 0.828709951516969,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004940114919254137,
      "loss": 6.1237,
      "mean_token_accuracy": 0.14867519289255143,
      "num_tokens": 16217952.0,
      "step": 8290
    },
    {
      "entropy": 6.168486833572388,
      "epoch": 0.8292097765781976,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004940033099948843,
      "loss": 6.1746,
      "mean_token_accuracy": 0.144381994754076,
      "num_tokens": 16227749.0,
      "step": 8295
    },
    {
      "entropy": 6.201283311843872,
      "epoch": 0.8297096016394262,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004939951225542879,
      "loss": 6.124,
      "mean_token_accuracy": 0.1489291861653328,
      "num_tokens": 16236716.0,
      "step": 8300
    },
    {
      "entropy": 6.233332729339599,
      "epoch": 0.8302094267006548,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004939869296038306,
      "loss": 6.0943,
      "mean_token_accuracy": 0.15528877973556518,
      "num_tokens": 16246471.0,
      "step": 8305
    },
    {
      "entropy": 6.259131669998169,
      "epoch": 0.8307092517618834,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004939787311437185,
      "loss": 6.2136,
      "mean_token_accuracy": 0.14248901158571242,
      "num_tokens": 16256694.0,
      "step": 8310
    },
    {
      "entropy": 6.184653425216675,
      "epoch": 0.831209076823112,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004939705271741579,
      "loss": 6.0665,
      "mean_token_accuracy": 0.15338120087981225,
      "num_tokens": 16266073.0,
      "step": 8315
    },
    {
      "entropy": 6.279538965225219,
      "epoch": 0.8317089018843404,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004939623176953551,
      "loss": 6.0978,
      "mean_token_accuracy": 0.14992624521255493,
      "num_tokens": 16274608.0,
      "step": 8320
    },
    {
      "entropy": 6.214292860031128,
      "epoch": 0.832208726945569,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004939541027075168,
      "loss": 6.0627,
      "mean_token_accuracy": 0.14932852387428283,
      "num_tokens": 16284311.0,
      "step": 8325
    },
    {
      "entropy": 6.188909387588501,
      "epoch": 0.8327085520067976,
      "grad_norm": 1.375,
      "learning_rate": 0.0004939458822108496,
      "loss": 6.1286,
      "mean_token_accuracy": 0.1526828497648239,
      "num_tokens": 16293829.0,
      "step": 8330
    },
    {
      "entropy": 6.257095861434936,
      "epoch": 0.8332083770680262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004939376562055602,
      "loss": 6.2122,
      "mean_token_accuracy": 0.14815813153982163,
      "num_tokens": 16303876.0,
      "step": 8335
    },
    {
      "entropy": 6.270600509643555,
      "epoch": 0.8337082021292548,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004939294246918558,
      "loss": 6.0865,
      "mean_token_accuracy": 0.152264703810215,
      "num_tokens": 16313781.0,
      "step": 8340
    },
    {
      "entropy": 6.173567008972168,
      "epoch": 0.8342080271904834,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004939211876699435,
      "loss": 6.0286,
      "mean_token_accuracy": 0.1585654392838478,
      "num_tokens": 16325214.0,
      "step": 8345
    },
    {
      "entropy": 6.307163333892822,
      "epoch": 0.8347078522517118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004939129451400305,
      "loss": 6.0442,
      "mean_token_accuracy": 0.1515487775206566,
      "num_tokens": 16335635.0,
      "step": 8350
    },
    {
      "entropy": 6.265123128890991,
      "epoch": 0.8352076773129404,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000493904697102324,
      "loss": 6.159,
      "mean_token_accuracy": 0.15151993483304976,
      "num_tokens": 16345459.0,
      "step": 8355
    },
    {
      "entropy": 6.1508848667144775,
      "epoch": 0.835707502374169,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004938964435570317,
      "loss": 6.0113,
      "mean_token_accuracy": 0.15290557742118835,
      "num_tokens": 16354972.0,
      "step": 8360
    },
    {
      "entropy": 6.160230350494385,
      "epoch": 0.8362073274353976,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004938881845043612,
      "loss": 6.0374,
      "mean_token_accuracy": 0.1551657125353813,
      "num_tokens": 16363257.0,
      "step": 8365
    },
    {
      "entropy": 6.100004243850708,
      "epoch": 0.8367071524966262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004938799199445202,
      "loss": 6.0869,
      "mean_token_accuracy": 0.16130529418587686,
      "num_tokens": 16374102.0,
      "step": 8370
    },
    {
      "entropy": 6.245328569412232,
      "epoch": 0.8372069775578548,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004938716498777169,
      "loss": 6.0847,
      "mean_token_accuracy": 0.15187839716672896,
      "num_tokens": 16383809.0,
      "step": 8375
    },
    {
      "entropy": 6.237498378753662,
      "epoch": 0.8377068026190834,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004938633743041592,
      "loss": 6.1714,
      "mean_token_accuracy": 0.1458502247929573,
      "num_tokens": 16393743.0,
      "step": 8380
    },
    {
      "entropy": 6.228604793548584,
      "epoch": 0.8382066276803118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004938550932240552,
      "loss": 6.0568,
      "mean_token_accuracy": 0.15981181412935258,
      "num_tokens": 16403225.0,
      "step": 8385
    },
    {
      "entropy": 6.299680900573731,
      "epoch": 0.8387064527415404,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938468066376134,
      "loss": 6.0949,
      "mean_token_accuracy": 0.15609302446246148,
      "num_tokens": 16413104.0,
      "step": 8390
    },
    {
      "entropy": 6.22253384590149,
      "epoch": 0.839206277802769,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938385145450423,
      "loss": 6.147,
      "mean_token_accuracy": 0.1449175976216793,
      "num_tokens": 16422831.0,
      "step": 8395
    },
    {
      "entropy": 6.26677393913269,
      "epoch": 0.8397061028639976,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004938302169465505,
      "loss": 6.1117,
      "mean_token_accuracy": 0.1512998253107071,
      "num_tokens": 16431955.0,
      "step": 8400
    },
    {
      "entropy": 6.352572107315064,
      "epoch": 0.8402059279252262,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004938219138423467,
      "loss": 6.2278,
      "mean_token_accuracy": 0.13825422450900077,
      "num_tokens": 16443266.0,
      "step": 8405
    },
    {
      "entropy": 6.2466073513031,
      "epoch": 0.8407057529864548,
      "grad_norm": 1.25,
      "learning_rate": 0.0004938136052326399,
      "loss": 6.1185,
      "mean_token_accuracy": 0.1468016266822815,
      "num_tokens": 16454264.0,
      "step": 8410
    },
    {
      "entropy": 6.090001153945923,
      "epoch": 0.8412055780476833,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004938052911176392,
      "loss": 5.9993,
      "mean_token_accuracy": 0.15437717139720916,
      "num_tokens": 16463613.0,
      "step": 8415
    },
    {
      "entropy": 6.184136962890625,
      "epoch": 0.8417054031089118,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004937969714975534,
      "loss": 6.0648,
      "mean_token_accuracy": 0.14991025179624556,
      "num_tokens": 16472517.0,
      "step": 8420
    },
    {
      "entropy": 6.278586339950562,
      "epoch": 0.8422052281701404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937886463725923,
      "loss": 6.1492,
      "mean_token_accuracy": 0.14934741258621215,
      "num_tokens": 16482328.0,
      "step": 8425
    },
    {
      "entropy": 6.17987289428711,
      "epoch": 0.842705053231369,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004937803157429651,
      "loss": 6.1488,
      "mean_token_accuracy": 0.14923462495207787,
      "num_tokens": 16492152.0,
      "step": 8430
    },
    {
      "entropy": 6.206251573562622,
      "epoch": 0.8432048782925976,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004937719796088815,
      "loss": 6.1094,
      "mean_token_accuracy": 0.14845191389322282,
      "num_tokens": 16501740.0,
      "step": 8435
    },
    {
      "entropy": 6.262111663818359,
      "epoch": 0.8437047033538262,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004937636379705511,
      "loss": 6.1248,
      "mean_token_accuracy": 0.1462855666875839,
      "num_tokens": 16512048.0,
      "step": 8440
    },
    {
      "entropy": 6.235871887207031,
      "epoch": 0.8442045284150548,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937552908281839,
      "loss": 6.1112,
      "mean_token_accuracy": 0.14632420092821122,
      "num_tokens": 16521935.0,
      "step": 8445
    },
    {
      "entropy": 6.225899505615234,
      "epoch": 0.8447043534762833,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004937469381819899,
      "loss": 6.1424,
      "mean_token_accuracy": 0.14733776152133943,
      "num_tokens": 16531953.0,
      "step": 8450
    },
    {
      "entropy": 6.251934909820557,
      "epoch": 0.8452041785375118,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004937385800321792,
      "loss": 6.0514,
      "mean_token_accuracy": 0.14804559275507928,
      "num_tokens": 16541196.0,
      "step": 8455
    },
    {
      "entropy": 6.141452503204346,
      "epoch": 0.8457040035987404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004937302163789621,
      "loss": 5.972,
      "mean_token_accuracy": 0.15394088849425316,
      "num_tokens": 16550453.0,
      "step": 8460
    },
    {
      "entropy": 6.231950092315674,
      "epoch": 0.846203828659969,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004937218472225491,
      "loss": 6.1376,
      "mean_token_accuracy": 0.14471847862005233,
      "num_tokens": 16560125.0,
      "step": 8465
    },
    {
      "entropy": 6.226166868209839,
      "epoch": 0.8467036537211976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937134725631506,
      "loss": 6.0635,
      "mean_token_accuracy": 0.15243830382823945,
      "num_tokens": 16569097.0,
      "step": 8470
    },
    {
      "entropy": 6.262660932540894,
      "epoch": 0.8472034787824262,
      "grad_norm": 1.375,
      "learning_rate": 0.0004937050924009775,
      "loss": 6.2221,
      "mean_token_accuracy": 0.14258782118558883,
      "num_tokens": 16579032.0,
      "step": 8475
    },
    {
      "entropy": 6.202022457122803,
      "epoch": 0.8477033038436548,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004936967067362406,
      "loss": 6.1253,
      "mean_token_accuracy": 0.14877662509679795,
      "num_tokens": 16588868.0,
      "step": 8480
    },
    {
      "entropy": 6.189423418045044,
      "epoch": 0.8482031289048833,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004936883155691507,
      "loss": 6.0584,
      "mean_token_accuracy": 0.15391803681850433,
      "num_tokens": 16598043.0,
      "step": 8485
    },
    {
      "entropy": 6.1896501064300535,
      "epoch": 0.8487029539661118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004936799188999192,
      "loss": 6.0663,
      "mean_token_accuracy": 0.15145521312952043,
      "num_tokens": 16607145.0,
      "step": 8490
    },
    {
      "entropy": 6.223206996917725,
      "epoch": 0.8492027790273404,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004936715167287571,
      "loss": 6.1084,
      "mean_token_accuracy": 0.14796570241451262,
      "num_tokens": 16616930.0,
      "step": 8495
    },
    {
      "entropy": 6.180533981323242,
      "epoch": 0.849702604088569,
      "grad_norm": 1.171875,
      "learning_rate": 0.000493663109055876,
      "loss": 6.0886,
      "mean_token_accuracy": 0.1554548293352127,
      "num_tokens": 16628175.0,
      "step": 8500
    },
    {
      "entropy": 6.264107418060303,
      "epoch": 0.8502024291497976,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004936546958814873,
      "loss": 6.1505,
      "mean_token_accuracy": 0.14727152287960052,
      "num_tokens": 16638914.0,
      "step": 8505
    },
    {
      "entropy": 6.293072175979614,
      "epoch": 0.8507022542110262,
      "grad_norm": 2.171875,
      "learning_rate": 0.0004936462772058026,
      "loss": 6.1117,
      "mean_token_accuracy": 0.14431769996881486,
      "num_tokens": 16648299.0,
      "step": 8510
    },
    {
      "entropy": 6.219326066970825,
      "epoch": 0.8512020792722547,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004936378530290339,
      "loss": 6.1195,
      "mean_token_accuracy": 0.14665945768356323,
      "num_tokens": 16658088.0,
      "step": 8515
    },
    {
      "entropy": 6.214383602142334,
      "epoch": 0.8517019043334833,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004936294233513931,
      "loss": 6.1889,
      "mean_token_accuracy": 0.14050958156585694,
      "num_tokens": 16669324.0,
      "step": 8520
    },
    {
      "entropy": 6.2402863025665285,
      "epoch": 0.8522017293947118,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004936209881730922,
      "loss": 6.0698,
      "mean_token_accuracy": 0.15220809653401374,
      "num_tokens": 16679118.0,
      "step": 8525
    },
    {
      "entropy": 6.278643703460693,
      "epoch": 0.8527015544559404,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936125474943436,
      "loss": 6.138,
      "mean_token_accuracy": 0.14224927127361298,
      "num_tokens": 16689245.0,
      "step": 8530
    },
    {
      "entropy": 6.242617511749268,
      "epoch": 0.853201379517169,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004936041013153596,
      "loss": 6.1435,
      "mean_token_accuracy": 0.15116460174322127,
      "num_tokens": 16698364.0,
      "step": 8535
    },
    {
      "entropy": 6.1793372631073,
      "epoch": 0.8537012045783976,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004935956496363525,
      "loss": 6.0789,
      "mean_token_accuracy": 0.1525582790374756,
      "num_tokens": 16708059.0,
      "step": 8540
    },
    {
      "entropy": 6.234001302719117,
      "epoch": 0.8542010296396262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935871924575352,
      "loss": 6.0775,
      "mean_token_accuracy": 0.14982742220163345,
      "num_tokens": 16717206.0,
      "step": 8545
    },
    {
      "entropy": 6.192938661575317,
      "epoch": 0.8547008547008547,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004935787297791204,
      "loss": 6.0269,
      "mean_token_accuracy": 0.14812390208244325,
      "num_tokens": 16726889.0,
      "step": 8550
    },
    {
      "entropy": 6.20928807258606,
      "epoch": 0.8552006797620832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004935702616013209,
      "loss": 6.1688,
      "mean_token_accuracy": 0.14763138592243194,
      "num_tokens": 16737232.0,
      "step": 8555
    },
    {
      "entropy": 6.165909624099731,
      "epoch": 0.8557005048233118,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00049356178792435,
      "loss": 6.0429,
      "mean_token_accuracy": 0.16092220991849898,
      "num_tokens": 16747213.0,
      "step": 8560
    },
    {
      "entropy": 6.207089948654175,
      "epoch": 0.8562003298845404,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004935533087484207,
      "loss": 6.0568,
      "mean_token_accuracy": 0.14912815913558006,
      "num_tokens": 16756412.0,
      "step": 8565
    },
    {
      "entropy": 6.205042362213135,
      "epoch": 0.856700154945769,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935448240737464,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14908301904797555,
      "num_tokens": 16766475.0,
      "step": 8570
    },
    {
      "entropy": 6.216852283477783,
      "epoch": 0.8571999800069976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004935363339005405,
      "loss": 6.0976,
      "mean_token_accuracy": 0.15041948705911637,
      "num_tokens": 16775903.0,
      "step": 8575
    },
    {
      "entropy": 6.289575624465942,
      "epoch": 0.8576998050682261,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004935278382290168,
      "loss": 6.2006,
      "mean_token_accuracy": 0.1445719912648201,
      "num_tokens": 16785347.0,
      "step": 8580
    },
    {
      "entropy": 6.233403062820434,
      "epoch": 0.8581996301294547,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004935193370593887,
      "loss": 6.1068,
      "mean_token_accuracy": 0.15213785022497178,
      "num_tokens": 16794704.0,
      "step": 8585
    },
    {
      "entropy": 6.234503507614136,
      "epoch": 0.8586994551906832,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004935108303918705,
      "loss": 6.0704,
      "mean_token_accuracy": 0.15252133458852768,
      "num_tokens": 16804711.0,
      "step": 8590
    },
    {
      "entropy": 6.18677864074707,
      "epoch": 0.8591992802519118,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004935023182266761,
      "loss": 6.0994,
      "mean_token_accuracy": 0.1476341649889946,
      "num_tokens": 16814217.0,
      "step": 8595
    },
    {
      "entropy": 6.248226833343506,
      "epoch": 0.8596991053131404,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004934938005640194,
      "loss": 6.0846,
      "mean_token_accuracy": 0.15099609196186065,
      "num_tokens": 16823652.0,
      "step": 8600
    },
    {
      "entropy": 6.26075587272644,
      "epoch": 0.860198930374369,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000493485277404115,
      "loss": 6.0918,
      "mean_token_accuracy": 0.15145450979471206,
      "num_tokens": 16832293.0,
      "step": 8605
    },
    {
      "entropy": 6.240613317489624,
      "epoch": 0.8606987554355976,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004934767487471772,
      "loss": 6.1493,
      "mean_token_accuracy": 0.1473841294646263,
      "num_tokens": 16840877.0,
      "step": 8610
    },
    {
      "entropy": 6.184689235687256,
      "epoch": 0.8611985804968261,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934682145934207,
      "loss": 6.1477,
      "mean_token_accuracy": 0.14328195974230767,
      "num_tokens": 16851245.0,
      "step": 8615
    },
    {
      "entropy": 6.31551423072815,
      "epoch": 0.8616984055580547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00049345967494306,
      "loss": 6.1254,
      "mean_token_accuracy": 0.14677524492144584,
      "num_tokens": 16862023.0,
      "step": 8620
    },
    {
      "entropy": 6.227668762207031,
      "epoch": 0.8621982306192832,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004934511297963102,
      "loss": 6.1018,
      "mean_token_accuracy": 0.15159463733434678,
      "num_tokens": 16871262.0,
      "step": 8625
    },
    {
      "entropy": 6.3318336486816404,
      "epoch": 0.8626980556805118,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004934425791533861,
      "loss": 6.2254,
      "mean_token_accuracy": 0.1389852873980999,
      "num_tokens": 16881513.0,
      "step": 8630
    },
    {
      "entropy": 6.1456138610839846,
      "epoch": 0.8631978807417404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004934340230145031,
      "loss": 5.9989,
      "mean_token_accuracy": 0.1533973515033722,
      "num_tokens": 16891516.0,
      "step": 8635
    },
    {
      "entropy": 6.260723495483399,
      "epoch": 0.863697705802969,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004934254613798761,
      "loss": 6.1382,
      "mean_token_accuracy": 0.14526411145925522,
      "num_tokens": 16901405.0,
      "step": 8640
    },
    {
      "entropy": 6.229184007644653,
      "epoch": 0.8641975308641975,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004934168942497207,
      "loss": 6.0596,
      "mean_token_accuracy": 0.1561588615179062,
      "num_tokens": 16911404.0,
      "step": 8645
    },
    {
      "entropy": 6.236042737960815,
      "epoch": 0.8646973559254261,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004934083216242526,
      "loss": 6.0529,
      "mean_token_accuracy": 0.14699126332998275,
      "num_tokens": 16920837.0,
      "step": 8650
    },
    {
      "entropy": 6.259059762954712,
      "epoch": 0.8651971809866547,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004933997435036873,
      "loss": 6.1903,
      "mean_token_accuracy": 0.141011643409729,
      "num_tokens": 16930255.0,
      "step": 8655
    },
    {
      "entropy": 6.193679571151733,
      "epoch": 0.8656970060478832,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004933911598882405,
      "loss": 6.1045,
      "mean_token_accuracy": 0.15010160803794861,
      "num_tokens": 16940188.0,
      "step": 8660
    },
    {
      "entropy": 6.258629369735718,
      "epoch": 0.8661968311091118,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004933825707781284,
      "loss": 6.2199,
      "mean_token_accuracy": 0.14718852415680886,
      "num_tokens": 16950260.0,
      "step": 8665
    },
    {
      "entropy": 6.269115591049195,
      "epoch": 0.8666966561703404,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004933739761735671,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14714471623301506,
      "num_tokens": 16959299.0,
      "step": 8670
    },
    {
      "entropy": 6.2602075099945065,
      "epoch": 0.867196481231569,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004933653760747726,
      "loss": 6.2697,
      "mean_token_accuracy": 0.13971225395798684,
      "num_tokens": 16969541.0,
      "step": 8675
    },
    {
      "entropy": 6.21373233795166,
      "epoch": 0.8676963062927975,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004933567704819615,
      "loss": 6.0891,
      "mean_token_accuracy": 0.1499331533908844,
      "num_tokens": 16978754.0,
      "step": 8680
    },
    {
      "entropy": 6.278979301452637,
      "epoch": 0.8681961313540261,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004933481593953502,
      "loss": 6.1656,
      "mean_token_accuracy": 0.14540855064988137,
      "num_tokens": 16989188.0,
      "step": 8685
    },
    {
      "entropy": 6.211424493789673,
      "epoch": 0.8686959564152547,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004933395428151554,
      "loss": 6.0874,
      "mean_token_accuracy": 0.14567918404936792,
      "num_tokens": 16999295.0,
      "step": 8690
    },
    {
      "entropy": 6.279827451705932,
      "epoch": 0.8691957814764832,
      "grad_norm": 1.546875,
      "learning_rate": 0.000493330920741594,
      "loss": 6.1623,
      "mean_token_accuracy": 0.14284103959798813,
      "num_tokens": 17009080.0,
      "step": 8695
    },
    {
      "entropy": 6.248566341400147,
      "epoch": 0.8696956065377118,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004933222931748827,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14833930358290673,
      "num_tokens": 17018201.0,
      "step": 8700
    },
    {
      "entropy": 6.26016321182251,
      "epoch": 0.8701954315989404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004933136601152387,
      "loss": 6.1935,
      "mean_token_accuracy": 0.1432765930891037,
      "num_tokens": 17027896.0,
      "step": 8705
    },
    {
      "entropy": 6.148264598846436,
      "epoch": 0.8706952566601689,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004933050215628793,
      "loss": 6.0236,
      "mean_token_accuracy": 0.15624282211065293,
      "num_tokens": 17038846.0,
      "step": 8710
    },
    {
      "entropy": 6.204269695281982,
      "epoch": 0.8711950817213975,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004932963775180216,
      "loss": 6.0657,
      "mean_token_accuracy": 0.15615527033805848,
      "num_tokens": 17048021.0,
      "step": 8715
    },
    {
      "entropy": 6.254320287704468,
      "epoch": 0.8716949067826261,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004932877279808833,
      "loss": 6.1702,
      "mean_token_accuracy": 0.1452657774090767,
      "num_tokens": 17058076.0,
      "step": 8720
    },
    {
      "entropy": 6.2646490097045895,
      "epoch": 0.8721947318438547,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004932790729516818,
      "loss": 6.1238,
      "mean_token_accuracy": 0.1446006029844284,
      "num_tokens": 17068988.0,
      "step": 8725
    },
    {
      "entropy": 6.241519212722778,
      "epoch": 0.8726945569050832,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004932704124306352,
      "loss": 6.2305,
      "mean_token_accuracy": 0.14832010716199875,
      "num_tokens": 17079423.0,
      "step": 8730
    },
    {
      "entropy": 6.316327667236328,
      "epoch": 0.8731943819663118,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000493261746417961,
      "loss": 6.1407,
      "mean_token_accuracy": 0.15326604545116423,
      "num_tokens": 17089675.0,
      "step": 8735
    },
    {
      "entropy": 6.176316261291504,
      "epoch": 0.8736942070275404,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004932530749138775,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15550363659858704,
      "num_tokens": 17099234.0,
      "step": 8740
    },
    {
      "entropy": 6.229638671875,
      "epoch": 0.8741940320887689,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932443979186028,
      "loss": 6.0768,
      "mean_token_accuracy": 0.1509983316063881,
      "num_tokens": 17108966.0,
      "step": 8745
    },
    {
      "entropy": 6.219983530044556,
      "epoch": 0.8746938571499975,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004932357154323552,
      "loss": 6.1306,
      "mean_token_accuracy": 0.1460599720478058,
      "num_tokens": 17117941.0,
      "step": 8750
    },
    {
      "entropy": 6.158332586288452,
      "epoch": 0.8751936822112261,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004932270274553531,
      "loss": 6.0212,
      "mean_token_accuracy": 0.1550137385725975,
      "num_tokens": 17127561.0,
      "step": 8755
    },
    {
      "entropy": 6.179317855834961,
      "epoch": 0.8756935072724547,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004932183339878153,
      "loss": 6.1072,
      "mean_token_accuracy": 0.14765550121665,
      "num_tokens": 17137645.0,
      "step": 8760
    },
    {
      "entropy": 6.256644344329834,
      "epoch": 0.8761933323336832,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004932096350299602,
      "loss": 6.1331,
      "mean_token_accuracy": 0.14467127323150636,
      "num_tokens": 17146955.0,
      "step": 8765
    },
    {
      "entropy": 6.277185344696045,
      "epoch": 0.8766931573949118,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004932009305820069,
      "loss": 6.1559,
      "mean_token_accuracy": 0.1432305671274662,
      "num_tokens": 17158287.0,
      "step": 8770
    },
    {
      "entropy": 6.355773401260376,
      "epoch": 0.8771929824561403,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004931922206441743,
      "loss": 6.1652,
      "mean_token_accuracy": 0.14233695194125176,
      "num_tokens": 17168646.0,
      "step": 8775
    },
    {
      "entropy": 6.256005668640137,
      "epoch": 0.8776928075173689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004931835052166817,
      "loss": 6.1311,
      "mean_token_accuracy": 0.14276591464877128,
      "num_tokens": 17178533.0,
      "step": 8780
    },
    {
      "entropy": 6.232534408569336,
      "epoch": 0.8781926325785975,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004931747842997481,
      "loss": 6.1765,
      "mean_token_accuracy": 0.14459428638219834,
      "num_tokens": 17189584.0,
      "step": 8785
    },
    {
      "entropy": 6.207731103897094,
      "epoch": 0.8786924576398261,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004931660578935931,
      "loss": 6.0694,
      "mean_token_accuracy": 0.15274597555398942,
      "num_tokens": 17199521.0,
      "step": 8790
    },
    {
      "entropy": 6.276787757873535,
      "epoch": 0.8791922827010547,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004931573259984363,
      "loss": 6.112,
      "mean_token_accuracy": 0.15395358949899673,
      "num_tokens": 17209812.0,
      "step": 8795
    },
    {
      "entropy": 6.338973903656006,
      "epoch": 0.8796921077622832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004931485886144972,
      "loss": 6.1723,
      "mean_token_accuracy": 0.14504109248518943,
      "num_tokens": 17220119.0,
      "step": 8800
    },
    {
      "entropy": 6.189262104034424,
      "epoch": 0.8801919328235118,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004931398457419959,
      "loss": 6.0848,
      "mean_token_accuracy": 0.15483706444501877,
      "num_tokens": 17229777.0,
      "step": 8805
    },
    {
      "entropy": 6.19629316329956,
      "epoch": 0.8806917578847403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004931310973811521,
      "loss": 6.0705,
      "mean_token_accuracy": 0.15523007065057753,
      "num_tokens": 17239039.0,
      "step": 8810
    },
    {
      "entropy": 6.205324268341064,
      "epoch": 0.8811915829459689,
      "grad_norm": 1.234375,
      "learning_rate": 0.000493122343532186,
      "loss": 6.029,
      "mean_token_accuracy": 0.14334864169359207,
      "num_tokens": 17248202.0,
      "step": 8815
    },
    {
      "entropy": 6.198616361618042,
      "epoch": 0.8816914080071975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000493113584195318,
      "loss": 6.1232,
      "mean_token_accuracy": 0.15694979280233384,
      "num_tokens": 17258253.0,
      "step": 8820
    },
    {
      "entropy": 6.180670833587646,
      "epoch": 0.8821912330684261,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004931048193707681,
      "loss": 6.0912,
      "mean_token_accuracy": 0.15559390485286712,
      "num_tokens": 17268381.0,
      "step": 8825
    },
    {
      "entropy": 6.2818155765533445,
      "epoch": 0.8826910581296546,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004930960490587573,
      "loss": 6.1735,
      "mean_token_accuracy": 0.14679909721016884,
      "num_tokens": 17278220.0,
      "step": 8830
    },
    {
      "entropy": 6.334381151199341,
      "epoch": 0.8831908831908832,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000493087273259506,
      "loss": 6.2383,
      "mean_token_accuracy": 0.14389952272176743,
      "num_tokens": 17288388.0,
      "step": 8835
    },
    {
      "entropy": 6.3111402034759525,
      "epoch": 0.8836907082521117,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004930784919732349,
      "loss": 6.2627,
      "mean_token_accuracy": 0.14755900651216508,
      "num_tokens": 17297308.0,
      "step": 8840
    },
    {
      "entropy": 6.108510303497314,
      "epoch": 0.8841905333133403,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004930697052001651,
      "loss": 6.0202,
      "mean_token_accuracy": 0.15119539201259613,
      "num_tokens": 17307586.0,
      "step": 8845
    },
    {
      "entropy": 6.21013503074646,
      "epoch": 0.8846903583745689,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004930609129405177,
      "loss": 6.1537,
      "mean_token_accuracy": 0.14370280578732492,
      "num_tokens": 17317342.0,
      "step": 8850
    },
    {
      "entropy": 6.2945633888244625,
      "epoch": 0.8851901834357975,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004930521151945138,
      "loss": 6.079,
      "mean_token_accuracy": 0.14940526932477952,
      "num_tokens": 17326586.0,
      "step": 8855
    },
    {
      "entropy": 6.2995665073394775,
      "epoch": 0.8856900084970261,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004930433119623749,
      "loss": 6.1966,
      "mean_token_accuracy": 0.13972629830241204,
      "num_tokens": 17337551.0,
      "step": 8860
    },
    {
      "entropy": 6.24941840171814,
      "epoch": 0.8861898335582546,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004930345032443221,
      "loss": 6.1371,
      "mean_token_accuracy": 0.14059261456131936,
      "num_tokens": 17346702.0,
      "step": 8865
    },
    {
      "entropy": 6.16758542060852,
      "epoch": 0.8866896586194832,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004930256890405777,
      "loss": 6.0987,
      "mean_token_accuracy": 0.15280970707535743,
      "num_tokens": 17356595.0,
      "step": 8870
    },
    {
      "entropy": 6.2948156833648685,
      "epoch": 0.8871894836807117,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004930168693513627,
      "loss": 6.1659,
      "mean_token_accuracy": 0.1484626516699791,
      "num_tokens": 17366264.0,
      "step": 8875
    },
    {
      "entropy": 6.3261146068573,
      "epoch": 0.8876893087419403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004930080441768995,
      "loss": 6.1478,
      "mean_token_accuracy": 0.14828914031386375,
      "num_tokens": 17376324.0,
      "step": 8880
    },
    {
      "entropy": 6.211109733581543,
      "epoch": 0.8881891338031689,
      "grad_norm": 2.046875,
      "learning_rate": 0.00049299921351741,
      "loss": 6.0868,
      "mean_token_accuracy": 0.14687345400452614,
      "num_tokens": 17386436.0,
      "step": 8885
    },
    {
      "entropy": 6.218961143493653,
      "epoch": 0.8886889588643975,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004929903773731164,
      "loss": 6.0666,
      "mean_token_accuracy": 0.1524926722049713,
      "num_tokens": 17395511.0,
      "step": 8890
    },
    {
      "entropy": 6.227129650115967,
      "epoch": 0.8891887839256261,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004929815357442409,
      "loss": 6.079,
      "mean_token_accuracy": 0.15277189761400223,
      "num_tokens": 17404887.0,
      "step": 8895
    },
    {
      "entropy": 6.195880031585693,
      "epoch": 0.8896886089868546,
      "grad_norm": 1.375,
      "learning_rate": 0.0004929726886310062,
      "loss": 6.0658,
      "mean_token_accuracy": 0.15348223000764846,
      "num_tokens": 17413936.0,
      "step": 8900
    },
    {
      "entropy": 6.219245862960816,
      "epoch": 0.8901884340480831,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004929638360336348,
      "loss": 6.1415,
      "mean_token_accuracy": 0.143099145591259,
      "num_tokens": 17425011.0,
      "step": 8905
    },
    {
      "entropy": 6.247063827514649,
      "epoch": 0.8906882591093117,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004929549779523492,
      "loss": 6.073,
      "mean_token_accuracy": 0.15219858437776565,
      "num_tokens": 17434508.0,
      "step": 8910
    },
    {
      "entropy": 6.267627859115601,
      "epoch": 0.8911880841705403,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004929461143873725,
      "loss": 6.0861,
      "mean_token_accuracy": 0.1511171728372574,
      "num_tokens": 17444155.0,
      "step": 8915
    },
    {
      "entropy": 6.310231399536133,
      "epoch": 0.8916879092317689,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004929372453389277,
      "loss": 6.1974,
      "mean_token_accuracy": 0.14196919277310371,
      "num_tokens": 17453834.0,
      "step": 8920
    },
    {
      "entropy": 6.205585575103759,
      "epoch": 0.8921877342929975,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004929283708072378,
      "loss": 6.0704,
      "mean_token_accuracy": 0.15700343251228333,
      "num_tokens": 17463491.0,
      "step": 8925
    },
    {
      "entropy": 6.141632175445556,
      "epoch": 0.8926875593542261,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004929194907925262,
      "loss": 6.0173,
      "mean_token_accuracy": 0.15369464755058287,
      "num_tokens": 17472882.0,
      "step": 8930
    },
    {
      "entropy": 6.257689809799194,
      "epoch": 0.8931873844154546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004929106052950163,
      "loss": 6.1239,
      "mean_token_accuracy": 0.15023526847362517,
      "num_tokens": 17483318.0,
      "step": 8935
    },
    {
      "entropy": 6.19890341758728,
      "epoch": 0.8936872094766831,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004929017143149317,
      "loss": 6.0571,
      "mean_token_accuracy": 0.15334602296352387,
      "num_tokens": 17493212.0,
      "step": 8940
    },
    {
      "entropy": 6.209878730773926,
      "epoch": 0.8941870345379117,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000492892817852496,
      "loss": 6.0662,
      "mean_token_accuracy": 0.14698252230882644,
      "num_tokens": 17502782.0,
      "step": 8945
    },
    {
      "entropy": 6.188276100158691,
      "epoch": 0.8946868595991403,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492883915907933,
      "loss": 6.2281,
      "mean_token_accuracy": 0.14035460650920867,
      "num_tokens": 17514321.0,
      "step": 8950
    },
    {
      "entropy": 6.221416568756103,
      "epoch": 0.8951866846603689,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004928750084814668,
      "loss": 6.0881,
      "mean_token_accuracy": 0.15225918591022491,
      "num_tokens": 17522910.0,
      "step": 8955
    },
    {
      "entropy": 6.229080486297607,
      "epoch": 0.8956865097215975,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004928660955733214,
      "loss": 6.0928,
      "mean_token_accuracy": 0.14701801985502244,
      "num_tokens": 17532747.0,
      "step": 8960
    },
    {
      "entropy": 6.211772871017456,
      "epoch": 0.896186334782826,
      "grad_norm": 1.5,
      "learning_rate": 0.0004928571771837211,
      "loss": 6.1053,
      "mean_token_accuracy": 0.14951809197664262,
      "num_tokens": 17542956.0,
      "step": 8965
    },
    {
      "entropy": 6.26898045539856,
      "epoch": 0.8966861598440545,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928482533128902,
      "loss": 6.1735,
      "mean_token_accuracy": 0.145123091340065,
      "num_tokens": 17552462.0,
      "step": 8970
    },
    {
      "entropy": 6.305005788803101,
      "epoch": 0.8971859849052831,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004928393239610535,
      "loss": 6.1209,
      "mean_token_accuracy": 0.1499975249171257,
      "num_tokens": 17561413.0,
      "step": 8975
    },
    {
      "entropy": 6.235485649108886,
      "epoch": 0.8976858099665117,
      "grad_norm": 1.25,
      "learning_rate": 0.0004928303891284354,
      "loss": 6.0876,
      "mean_token_accuracy": 0.15316251218318938,
      "num_tokens": 17571157.0,
      "step": 8980
    },
    {
      "entropy": 6.298145818710327,
      "epoch": 0.8981856350277403,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004928214488152607,
      "loss": 6.1413,
      "mean_token_accuracy": 0.1458333596587181,
      "num_tokens": 17581645.0,
      "step": 8985
    },
    {
      "entropy": 6.269522714614868,
      "epoch": 0.8986854600889689,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004928125030217545,
      "loss": 6.1134,
      "mean_token_accuracy": 0.14980985820293427,
      "num_tokens": 17591990.0,
      "step": 8990
    },
    {
      "entropy": 6.277906084060669,
      "epoch": 0.8991852851501975,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004928035517481416,
      "loss": 6.1181,
      "mean_token_accuracy": 0.153779099881649,
      "num_tokens": 17602063.0,
      "step": 8995
    },
    {
      "entropy": 6.199600028991699,
      "epoch": 0.899685110211426,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004927945949946476,
      "loss": 6.1606,
      "mean_token_accuracy": 0.14495958834886552,
      "num_tokens": 17612258.0,
      "step": 9000
    },
    {
      "epoch": 0.899685110211426,
      "eval_entropy": 6.117284948633554,
      "eval_loss": 6.155108451843262,
      "eval_mean_token_accuracy": 0.15283872304002868,
      "eval_num_tokens": 17612258.0,
      "eval_runtime": 24.0947,
      "eval_samples_per_second": 1288.455,
      "eval_steps_per_second": 161.072,
      "step": 9000
    },
    {
      "entropy": 6.278792476654052,
      "epoch": 0.9001849352726545,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004927856327614975,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15170859843492507,
      "num_tokens": 17621901.0,
      "step": 9005
    },
    {
      "entropy": 6.240556764602661,
      "epoch": 0.9006847603338831,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000492776665048917,
      "loss": 6.0592,
      "mean_token_accuracy": 0.1467784136533737,
      "num_tokens": 17630854.0,
      "step": 9010
    },
    {
      "entropy": 6.268830871582031,
      "epoch": 0.9011845853951117,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004927676918571316,
      "loss": 6.2073,
      "mean_token_accuracy": 0.14159762263298034,
      "num_tokens": 17640758.0,
      "step": 9015
    },
    {
      "entropy": 6.228778982162476,
      "epoch": 0.9016844104563403,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004927587131863671,
      "loss": 6.0622,
      "mean_token_accuracy": 0.15188493058085442,
      "num_tokens": 17650171.0,
      "step": 9020
    },
    {
      "entropy": 6.246705055236816,
      "epoch": 0.9021842355175689,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004927497290368493,
      "loss": 6.2037,
      "mean_token_accuracy": 0.14825357347726822,
      "num_tokens": 17660403.0,
      "step": 9025
    },
    {
      "entropy": 6.370050764083862,
      "epoch": 0.9026840605787975,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004927407394088046,
      "loss": 6.212,
      "mean_token_accuracy": 0.14207073748111726,
      "num_tokens": 17670764.0,
      "step": 9030
    },
    {
      "entropy": 6.228684329986573,
      "epoch": 0.9031838856400259,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004927317443024587,
      "loss": 6.1177,
      "mean_token_accuracy": 0.15171560347080232,
      "num_tokens": 17679804.0,
      "step": 9035
    },
    {
      "entropy": 6.2582700729370115,
      "epoch": 0.9036837107012545,
      "grad_norm": 1.25,
      "learning_rate": 0.0004927227437180382,
      "loss": 6.1617,
      "mean_token_accuracy": 0.13992909640073775,
      "num_tokens": 17689819.0,
      "step": 9040
    },
    {
      "entropy": 6.330976581573486,
      "epoch": 0.9041835357624831,
      "grad_norm": 1.375,
      "learning_rate": 0.0004927137376557696,
      "loss": 6.2057,
      "mean_token_accuracy": 0.14406838044524192,
      "num_tokens": 17699716.0,
      "step": 9045
    },
    {
      "entropy": 6.292686510086059,
      "epoch": 0.9046833608237117,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004927047261158793,
      "loss": 6.0762,
      "mean_token_accuracy": 0.15342100262641906,
      "num_tokens": 17709988.0,
      "step": 9050
    },
    {
      "entropy": 6.182733631134033,
      "epoch": 0.9051831858849403,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004926957090985942,
      "loss": 6.0328,
      "mean_token_accuracy": 0.15095359981060028,
      "num_tokens": 17719821.0,
      "step": 9055
    },
    {
      "entropy": 6.354136276245117,
      "epoch": 0.9056830109461689,
      "grad_norm": 1.25,
      "learning_rate": 0.0004926866866041409,
      "loss": 6.1994,
      "mean_token_accuracy": 0.14148515239357948,
      "num_tokens": 17728813.0,
      "step": 9060
    },
    {
      "entropy": 6.226703834533692,
      "epoch": 0.9061828360073975,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004926776586327467,
      "loss": 6.0662,
      "mean_token_accuracy": 0.14839670956134796,
      "num_tokens": 17738906.0,
      "step": 9065
    },
    {
      "entropy": 6.188530731201172,
      "epoch": 0.9066826610686259,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004926686251846386,
      "loss": 6.0487,
      "mean_token_accuracy": 0.15427898615598679,
      "num_tokens": 17748869.0,
      "step": 9070
    },
    {
      "entropy": 6.305594158172608,
      "epoch": 0.9071824861298545,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004926595862600439,
      "loss": 6.1951,
      "mean_token_accuracy": 0.14372513815760612,
      "num_tokens": 17758772.0,
      "step": 9075
    },
    {
      "entropy": 6.308678245544433,
      "epoch": 0.9076823111910831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004926505418591901,
      "loss": 6.1822,
      "mean_token_accuracy": 0.1418959878385067,
      "num_tokens": 17768575.0,
      "step": 9080
    },
    {
      "entropy": 6.303319025039673,
      "epoch": 0.9081821362523117,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004926414919823045,
      "loss": 6.2345,
      "mean_token_accuracy": 0.14073817431926727,
      "num_tokens": 17777445.0,
      "step": 9085
    },
    {
      "entropy": 6.228063726425171,
      "epoch": 0.9086819613135403,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004926324366296151,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15475739911198616,
      "num_tokens": 17787995.0,
      "step": 9090
    },
    {
      "entropy": 6.215692043304443,
      "epoch": 0.9091817863747689,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004926233758013494,
      "loss": 6.0282,
      "mean_token_accuracy": 0.15679309144616127,
      "num_tokens": 17797899.0,
      "step": 9095
    },
    {
      "entropy": 6.286261177062988,
      "epoch": 0.9096816114359974,
      "grad_norm": 1.375,
      "learning_rate": 0.0004926143094977359,
      "loss": 6.0917,
      "mean_token_accuracy": 0.1526429273188114,
      "num_tokens": 17807640.0,
      "step": 9100
    },
    {
      "entropy": 6.204017877578735,
      "epoch": 0.9101814364972259,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004926052377190022,
      "loss": 6.1081,
      "mean_token_accuracy": 0.15165030658245088,
      "num_tokens": 17817997.0,
      "step": 9105
    },
    {
      "entropy": 6.252846956253052,
      "epoch": 0.9106812615584545,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004925961604653765,
      "loss": 6.1377,
      "mean_token_accuracy": 0.1446901634335518,
      "num_tokens": 17828145.0,
      "step": 9110
    },
    {
      "entropy": 6.31077299118042,
      "epoch": 0.9111810866196831,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004925870777370877,
      "loss": 6.0322,
      "mean_token_accuracy": 0.15969755202531816,
      "num_tokens": 17837403.0,
      "step": 9115
    },
    {
      "entropy": 6.251461219787598,
      "epoch": 0.9116809116809117,
      "grad_norm": 1.5,
      "learning_rate": 0.0004925779895343639,
      "loss": 6.0972,
      "mean_token_accuracy": 0.15471886098384857,
      "num_tokens": 17846745.0,
      "step": 9120
    },
    {
      "entropy": 6.225797462463379,
      "epoch": 0.9121807367421403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004925688958574339,
      "loss": 6.1319,
      "mean_token_accuracy": 0.1483653098344803,
      "num_tokens": 17857422.0,
      "step": 9125
    },
    {
      "entropy": 6.196518468856811,
      "epoch": 0.9126805618033689,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004925597967065264,
      "loss": 6.0443,
      "mean_token_accuracy": 0.15260802507400512,
      "num_tokens": 17867004.0,
      "step": 9130
    },
    {
      "entropy": 6.233826684951782,
      "epoch": 0.9131803868645973,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004925506920818705,
      "loss": 6.1584,
      "mean_token_accuracy": 0.1427806869149208,
      "num_tokens": 17876933.0,
      "step": 9135
    },
    {
      "entropy": 6.263123989105225,
      "epoch": 0.9136802119258259,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004925415819836951,
      "loss": 6.1322,
      "mean_token_accuracy": 0.14953113198280335,
      "num_tokens": 17887639.0,
      "step": 9140
    },
    {
      "entropy": 6.245287084579468,
      "epoch": 0.9141800369870545,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004925324664122294,
      "loss": 6.005,
      "mean_token_accuracy": 0.1594272755086422,
      "num_tokens": 17897125.0,
      "step": 9145
    },
    {
      "entropy": 6.202782535552979,
      "epoch": 0.9146798620482831,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004925233453677029,
      "loss": 6.1586,
      "mean_token_accuracy": 0.14720847010612487,
      "num_tokens": 17907580.0,
      "step": 9150
    },
    {
      "entropy": 6.1721839427948,
      "epoch": 0.9151796871095117,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004925142188503451,
      "loss": 6.0075,
      "mean_token_accuracy": 0.15930448323488236,
      "num_tokens": 17915937.0,
      "step": 9155
    },
    {
      "entropy": 6.165093326568604,
      "epoch": 0.9156795121707403,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004925050868603854,
      "loss": 5.9947,
      "mean_token_accuracy": 0.15624236166477204,
      "num_tokens": 17926337.0,
      "step": 9160
    },
    {
      "entropy": 6.279834604263305,
      "epoch": 0.9161793372319688,
      "grad_norm": 1.765625,
      "learning_rate": 0.0004924959493980537,
      "loss": 6.1865,
      "mean_token_accuracy": 0.13933350667357444,
      "num_tokens": 17936123.0,
      "step": 9165
    },
    {
      "entropy": 6.190799856185913,
      "epoch": 0.9166791622931973,
      "grad_norm": 1.375,
      "learning_rate": 0.0004924868064635798,
      "loss": 6.0529,
      "mean_token_accuracy": 0.1513209544122219,
      "num_tokens": 17946143.0,
      "step": 9170
    },
    {
      "entropy": 6.342693185806274,
      "epoch": 0.9171789873544259,
      "grad_norm": 1.46875,
      "learning_rate": 0.000492477658057194,
      "loss": 6.1995,
      "mean_token_accuracy": 0.13836629539728165,
      "num_tokens": 17956034.0,
      "step": 9175
    },
    {
      "entropy": 6.259401082992554,
      "epoch": 0.9176788124156545,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004924685041791262,
      "loss": 6.0594,
      "mean_token_accuracy": 0.14909986704587935,
      "num_tokens": 17965455.0,
      "step": 9180
    },
    {
      "entropy": 6.218113231658935,
      "epoch": 0.9181786374768831,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004924593448296068,
      "loss": 6.0901,
      "mean_token_accuracy": 0.14673807024955748,
      "num_tokens": 17974506.0,
      "step": 9185
    },
    {
      "entropy": 6.170910501480103,
      "epoch": 0.9186784625381117,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004924501800088662,
      "loss": 6.1269,
      "mean_token_accuracy": 0.1422795407474041,
      "num_tokens": 17984750.0,
      "step": 9190
    },
    {
      "entropy": 6.234971332550049,
      "epoch": 0.9191782875993403,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004924410097171351,
      "loss": 6.1185,
      "mean_token_accuracy": 0.15175771713256836,
      "num_tokens": 17993871.0,
      "step": 9195
    },
    {
      "entropy": 6.239521646499634,
      "epoch": 0.9196781126605688,
      "grad_norm": 1.34375,
      "learning_rate": 0.000492431833954644,
      "loss": 6.076,
      "mean_token_accuracy": 0.15147107541561128,
      "num_tokens": 18003686.0,
      "step": 9200
    },
    {
      "entropy": 6.245208835601806,
      "epoch": 0.9201779377217973,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004924226527216241,
      "loss": 6.0982,
      "mean_token_accuracy": 0.15223220735788345,
      "num_tokens": 18013024.0,
      "step": 9205
    },
    {
      "entropy": 6.235642910003662,
      "epoch": 0.9206777627830259,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004924134660183061,
      "loss": 6.0748,
      "mean_token_accuracy": 0.15241352915763856,
      "num_tokens": 18022718.0,
      "step": 9210
    },
    {
      "entropy": 6.093504858016968,
      "epoch": 0.9211775878442545,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004924042738449212,
      "loss": 5.9206,
      "mean_token_accuracy": 0.15399529933929443,
      "num_tokens": 18032185.0,
      "step": 9215
    },
    {
      "entropy": 6.261243629455566,
      "epoch": 0.9216774129054831,
      "grad_norm": 1.5,
      "learning_rate": 0.0004923950762017008,
      "loss": 6.1395,
      "mean_token_accuracy": 0.14699485898017883,
      "num_tokens": 18041864.0,
      "step": 9220
    },
    {
      "entropy": 6.142456579208374,
      "epoch": 0.9221772379667117,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004923858730888762,
      "loss": 5.9605,
      "mean_token_accuracy": 0.16439410895109177,
      "num_tokens": 18050797.0,
      "step": 9225
    },
    {
      "entropy": 6.170009660720825,
      "epoch": 0.9226770630279402,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000492376664506679,
      "loss": 6.0105,
      "mean_token_accuracy": 0.1508944496512413,
      "num_tokens": 18060414.0,
      "step": 9230
    },
    {
      "entropy": 6.195092964172363,
      "epoch": 0.9231768880891688,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004923674504553408,
      "loss": 6.1223,
      "mean_token_accuracy": 0.14924529939889908,
      "num_tokens": 18070832.0,
      "step": 9235
    },
    {
      "entropy": 6.247225284576416,
      "epoch": 0.9236767131503973,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004923582309350935,
      "loss": 6.0723,
      "mean_token_accuracy": 0.14971695244312286,
      "num_tokens": 18078918.0,
      "step": 9240
    },
    {
      "entropy": 6.211710691452026,
      "epoch": 0.9241765382116259,
      "grad_norm": 1.421875,
      "learning_rate": 0.000492349005946169,
      "loss": 6.1075,
      "mean_token_accuracy": 0.15068391114473342,
      "num_tokens": 18087896.0,
      "step": 9245
    },
    {
      "entropy": 6.258350229263305,
      "epoch": 0.9246763632728545,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004923397754887994,
      "loss": 6.1876,
      "mean_token_accuracy": 0.14481544494628906,
      "num_tokens": 18097535.0,
      "step": 9250
    },
    {
      "entropy": 6.200371360778808,
      "epoch": 0.9251761883340831,
      "grad_norm": 1.375,
      "learning_rate": 0.0004923305395632172,
      "loss": 6.0632,
      "mean_token_accuracy": 0.15488176196813583,
      "num_tokens": 18107503.0,
      "step": 9255
    },
    {
      "entropy": 6.241928386688232,
      "epoch": 0.9256760133953117,
      "grad_norm": 1.25,
      "learning_rate": 0.0004923212981696543,
      "loss": 6.0706,
      "mean_token_accuracy": 0.14943889304995536,
      "num_tokens": 18117188.0,
      "step": 9260
    },
    {
      "entropy": 6.33441915512085,
      "epoch": 0.9261758384565402,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004923120513083436,
      "loss": 6.2309,
      "mean_token_accuracy": 0.14044254571199416,
      "num_tokens": 18127965.0,
      "step": 9265
    },
    {
      "entropy": 6.241382694244384,
      "epoch": 0.9266756635177688,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004923027989795175,
      "loss": 6.0833,
      "mean_token_accuracy": 0.15091975778341293,
      "num_tokens": 18136861.0,
      "step": 9270
    },
    {
      "entropy": 6.155618906021118,
      "epoch": 0.9271754885789973,
      "grad_norm": 1.46875,
      "learning_rate": 0.000492293541183409,
      "loss": 6.0587,
      "mean_token_accuracy": 0.15113596469163895,
      "num_tokens": 18145702.0,
      "step": 9275
    },
    {
      "entropy": 6.240806436538696,
      "epoch": 0.9276753136402259,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004922842779202509,
      "loss": 6.1587,
      "mean_token_accuracy": 0.14900632873177527,
      "num_tokens": 18155177.0,
      "step": 9280
    },
    {
      "entropy": 6.2426543712615965,
      "epoch": 0.9281751387014545,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004922750091902763,
      "loss": 6.0321,
      "mean_token_accuracy": 0.15409672409296035,
      "num_tokens": 18165361.0,
      "step": 9285
    },
    {
      "entropy": 6.236721754074097,
      "epoch": 0.9286749637626831,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004922657349937183,
      "loss": 6.088,
      "mean_token_accuracy": 0.1463526926934719,
      "num_tokens": 18175300.0,
      "step": 9290
    },
    {
      "entropy": 6.246470165252686,
      "epoch": 0.9291747888239116,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004922564553308103,
      "loss": 6.0721,
      "mean_token_accuracy": 0.15268152728676795,
      "num_tokens": 18183985.0,
      "step": 9295
    },
    {
      "entropy": 6.286758232116699,
      "epoch": 0.9296746138851402,
      "grad_norm": 1.328125,
      "learning_rate": 0.000492247170201786,
      "loss": 6.1934,
      "mean_token_accuracy": 0.1485764890909195,
      "num_tokens": 18193251.0,
      "step": 9300
    },
    {
      "entropy": 6.276211309432983,
      "epoch": 0.9301744389463688,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004922378796068785,
      "loss": 6.0325,
      "mean_token_accuracy": 0.15912700369954108,
      "num_tokens": 18202081.0,
      "step": 9305
    },
    {
      "entropy": 6.182103824615479,
      "epoch": 0.9306742640075973,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000492228583546322,
      "loss": 6.0446,
      "mean_token_accuracy": 0.14939359873533248,
      "num_tokens": 18211499.0,
      "step": 9310
    },
    {
      "entropy": 6.152326202392578,
      "epoch": 0.9311740890688259,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004922192820203502,
      "loss": 6.0473,
      "mean_token_accuracy": 0.1521271914243698,
      "num_tokens": 18220764.0,
      "step": 9315
    },
    {
      "entropy": 6.16752290725708,
      "epoch": 0.9316739141300545,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004922099750291972,
      "loss": 6.0445,
      "mean_token_accuracy": 0.15772737860679625,
      "num_tokens": 18230126.0,
      "step": 9320
    },
    {
      "entropy": 6.112668657302857,
      "epoch": 0.9321737391912831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004922006625730971,
      "loss": 6.0539,
      "mean_token_accuracy": 0.14889685809612274,
      "num_tokens": 18238426.0,
      "step": 9325
    },
    {
      "entropy": 6.201256370544433,
      "epoch": 0.9326735642525116,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004921913446522843,
      "loss": 6.1391,
      "mean_token_accuracy": 0.15078041404485704,
      "num_tokens": 18248215.0,
      "step": 9330
    },
    {
      "entropy": 6.314979553222656,
      "epoch": 0.9331733893137402,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000492182021266993,
      "loss": 6.1007,
      "mean_token_accuracy": 0.1482422836124897,
      "num_tokens": 18258967.0,
      "step": 9335
    },
    {
      "entropy": 6.3010539531707765,
      "epoch": 0.9336732143749688,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000492172692417458,
      "loss": 6.1543,
      "mean_token_accuracy": 0.14212929382920264,
      "num_tokens": 18268901.0,
      "step": 9340
    },
    {
      "entropy": 6.318927955627442,
      "epoch": 0.9341730394361973,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000492163358103914,
      "loss": 6.1757,
      "mean_token_accuracy": 0.14328315407037734,
      "num_tokens": 18279615.0,
      "step": 9345
    },
    {
      "entropy": 6.175833225250244,
      "epoch": 0.9346728644974259,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004921540183265957,
      "loss": 6.0735,
      "mean_token_accuracy": 0.15296206921339034,
      "num_tokens": 18289630.0,
      "step": 9350
    },
    {
      "entropy": 6.2162850856781,
      "epoch": 0.9351726895586545,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004921446730857383,
      "loss": 6.1319,
      "mean_token_accuracy": 0.1450078748166561,
      "num_tokens": 18299764.0,
      "step": 9355
    },
    {
      "entropy": 6.173314332962036,
      "epoch": 0.935672514619883,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004921353223815767,
      "loss": 6.0047,
      "mean_token_accuracy": 0.16180949360132219,
      "num_tokens": 18310252.0,
      "step": 9360
    },
    {
      "entropy": 6.262190294265747,
      "epoch": 0.9361723396811116,
      "grad_norm": 1.25,
      "learning_rate": 0.0004921259662143463,
      "loss": 6.0638,
      "mean_token_accuracy": 0.15603608787059783,
      "num_tokens": 18319753.0,
      "step": 9365
    },
    {
      "entropy": 6.224160432815552,
      "epoch": 0.9366721647423402,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004921166045842825,
      "loss": 6.1126,
      "mean_token_accuracy": 0.15129012614488602,
      "num_tokens": 18330307.0,
      "step": 9370
    },
    {
      "entropy": 6.225104331970215,
      "epoch": 0.9371719898035688,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004921072374916208,
      "loss": 6.1619,
      "mean_token_accuracy": 0.14460510462522508,
      "num_tokens": 18340867.0,
      "step": 9375
    },
    {
      "entropy": 6.219675159454345,
      "epoch": 0.9376718148647973,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004920978649365968,
      "loss": 5.9767,
      "mean_token_accuracy": 0.16016425639390947,
      "num_tokens": 18350852.0,
      "step": 9380
    },
    {
      "entropy": 6.123375749588012,
      "epoch": 0.9381716399260259,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004920884869194463,
      "loss": 5.9818,
      "mean_token_accuracy": 0.1563676565885544,
      "num_tokens": 18361821.0,
      "step": 9385
    },
    {
      "entropy": 6.204911899566651,
      "epoch": 0.9386714649872545,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004920791034404054,
      "loss": 6.1581,
      "mean_token_accuracy": 0.1447551116347313,
      "num_tokens": 18371391.0,
      "step": 9390
    },
    {
      "entropy": 6.252169847488403,
      "epoch": 0.939171290048483,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004920697144997102,
      "loss": 6.1413,
      "mean_token_accuracy": 0.15190670415759086,
      "num_tokens": 18380661.0,
      "step": 9395
    },
    {
      "entropy": 6.233347797393799,
      "epoch": 0.9396711151097116,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004920603200975968,
      "loss": 6.0698,
      "mean_token_accuracy": 0.15179983749985695,
      "num_tokens": 18390365.0,
      "step": 9400
    },
    {
      "entropy": 6.170382070541382,
      "epoch": 0.9401709401709402,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004920509202343016,
      "loss": 6.0337,
      "mean_token_accuracy": 0.15368024185299872,
      "num_tokens": 18399536.0,
      "step": 9405
    },
    {
      "entropy": 6.211355209350586,
      "epoch": 0.9406707652321687,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492041514910061,
      "loss": 6.0917,
      "mean_token_accuracy": 0.15307086259126662,
      "num_tokens": 18408769.0,
      "step": 9410
    },
    {
      "entropy": 6.197820711135864,
      "epoch": 0.9411705902933973,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004920321041251118,
      "loss": 6.1133,
      "mean_token_accuracy": 0.14328097850084304,
      "num_tokens": 18418554.0,
      "step": 9415
    },
    {
      "entropy": 6.16051836013794,
      "epoch": 0.9416704153546259,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004920226878796908,
      "loss": 6.0891,
      "mean_token_accuracy": 0.14609383866190911,
      "num_tokens": 18429670.0,
      "step": 9420
    },
    {
      "entropy": 6.3255408763885494,
      "epoch": 0.9421702404158544,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004920132661740348,
      "loss": 6.1654,
      "mean_token_accuracy": 0.15098945796489716,
      "num_tokens": 18438747.0,
      "step": 9425
    },
    {
      "entropy": 6.319615507125855,
      "epoch": 0.942670065477083,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004920038390083808,
      "loss": 6.1307,
      "mean_token_accuracy": 0.1464590087532997,
      "num_tokens": 18449350.0,
      "step": 9430
    },
    {
      "entropy": 6.196408605575561,
      "epoch": 0.9431698905383116,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000491994406382966,
      "loss": 6.0532,
      "mean_token_accuracy": 0.15202998369932175,
      "num_tokens": 18459052.0,
      "step": 9435
    },
    {
      "entropy": 6.2291124820709225,
      "epoch": 0.9436697155995402,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004919849682980278,
      "loss": 6.1578,
      "mean_token_accuracy": 0.15054467245936393,
      "num_tokens": 18469266.0,
      "step": 9440
    },
    {
      "entropy": 6.2922014713287355,
      "epoch": 0.9441695406607687,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004919755247538037,
      "loss": 6.1399,
      "mean_token_accuracy": 0.14657674878835678,
      "num_tokens": 18479381.0,
      "step": 9445
    },
    {
      "entropy": 6.207374954223633,
      "epoch": 0.9446693657219973,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004919660757505312,
      "loss": 5.9956,
      "mean_token_accuracy": 0.15650558322668076,
      "num_tokens": 18488973.0,
      "step": 9450
    },
    {
      "entropy": 6.064509868621826,
      "epoch": 0.9451691907832259,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004919566212884481,
      "loss": 6.0016,
      "mean_token_accuracy": 0.15545373857021333,
      "num_tokens": 18497907.0,
      "step": 9455
    },
    {
      "entropy": 6.160729932785034,
      "epoch": 0.9456690158444544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004919471613677921,
      "loss": 6.0567,
      "mean_token_accuracy": 0.15381805151700972,
      "num_tokens": 18507628.0,
      "step": 9460
    },
    {
      "entropy": 6.265036296844483,
      "epoch": 0.946168840905683,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004919376959888015,
      "loss": 6.1867,
      "mean_token_accuracy": 0.14997860565781593,
      "num_tokens": 18516346.0,
      "step": 9465
    },
    {
      "entropy": 6.281708478927612,
      "epoch": 0.9466686659669116,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004919282251517142,
      "loss": 5.9942,
      "mean_token_accuracy": 0.15608662068843843,
      "num_tokens": 18525568.0,
      "step": 9470
    },
    {
      "entropy": 6.231860589981079,
      "epoch": 0.9471684910281402,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004919187488567686,
      "loss": 6.1524,
      "mean_token_accuracy": 0.14924800246953965,
      "num_tokens": 18536715.0,
      "step": 9475
    },
    {
      "entropy": 6.299409103393555,
      "epoch": 0.9476683160893687,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004919092671042031,
      "loss": 6.1404,
      "mean_token_accuracy": 0.1481575511395931,
      "num_tokens": 18546405.0,
      "step": 9480
    },
    {
      "entropy": 6.286548233032226,
      "epoch": 0.9481681411505973,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004918997798942562,
      "loss": 6.1875,
      "mean_token_accuracy": 0.1450359307229519,
      "num_tokens": 18556721.0,
      "step": 9485
    },
    {
      "entropy": 6.303821849822998,
      "epoch": 0.9486679662118258,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004918902872271668,
      "loss": 6.217,
      "mean_token_accuracy": 0.14441457614302636,
      "num_tokens": 18567875.0,
      "step": 9490
    },
    {
      "entropy": 6.211621332168579,
      "epoch": 0.9491677912730544,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004918807891031734,
      "loss": 6.0986,
      "mean_token_accuracy": 0.14712619334459304,
      "num_tokens": 18578185.0,
      "step": 9495
    },
    {
      "entropy": 6.205819702148437,
      "epoch": 0.949667616334283,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004918712855225154,
      "loss": 6.067,
      "mean_token_accuracy": 0.15275223106145858,
      "num_tokens": 18587503.0,
      "step": 9500
    },
    {
      "entropy": 6.220349645614624,
      "epoch": 0.9501674413955116,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004918617764854314,
      "loss": 6.1182,
      "mean_token_accuracy": 0.14702729433774947,
      "num_tokens": 18597018.0,
      "step": 9505
    },
    {
      "entropy": 6.263982105255127,
      "epoch": 0.9506672664567402,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000491852261992161,
      "loss": 6.0756,
      "mean_token_accuracy": 0.1505356714129448,
      "num_tokens": 18606262.0,
      "step": 9510
    },
    {
      "entropy": 6.319805002212524,
      "epoch": 0.9511670915179687,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004918427420429435,
      "loss": 6.1587,
      "mean_token_accuracy": 0.14641866981983184,
      "num_tokens": 18616070.0,
      "step": 9515
    },
    {
      "entropy": 6.247110843658447,
      "epoch": 0.9516669165791973,
      "grad_norm": 1.8828125,
      "learning_rate": 0.0004918332166380185,
      "loss": 6.1215,
      "mean_token_accuracy": 0.1476816341280937,
      "num_tokens": 18625882.0,
      "step": 9520
    },
    {
      "entropy": 6.238479328155518,
      "epoch": 0.9521667416404258,
      "grad_norm": 1.25,
      "learning_rate": 0.0004918236857776254,
      "loss": 6.1747,
      "mean_token_accuracy": 0.1430346630513668,
      "num_tokens": 18635997.0,
      "step": 9525
    },
    {
      "entropy": 6.1794861316680905,
      "epoch": 0.9526665667016544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004918141494620042,
      "loss": 6.0288,
      "mean_token_accuracy": 0.14789827838540076,
      "num_tokens": 18646234.0,
      "step": 9530
    },
    {
      "entropy": 6.212112903594971,
      "epoch": 0.953166391762883,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004918046076913949,
      "loss": 6.0595,
      "mean_token_accuracy": 0.15233168751001358,
      "num_tokens": 18655407.0,
      "step": 9535
    },
    {
      "entropy": 6.181972932815552,
      "epoch": 0.9536662168241116,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004917950604660374,
      "loss": 6.027,
      "mean_token_accuracy": 0.1586668461561203,
      "num_tokens": 18664695.0,
      "step": 9540
    },
    {
      "entropy": 6.226296043395996,
      "epoch": 0.9541660418853402,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000491785507786172,
      "loss": 5.9811,
      "mean_token_accuracy": 0.16150170266628266,
      "num_tokens": 18675170.0,
      "step": 9545
    },
    {
      "entropy": 6.144332408905029,
      "epoch": 0.9546658669465687,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004917759496520388,
      "loss": 5.9884,
      "mean_token_accuracy": 0.15726979374885558,
      "num_tokens": 18684619.0,
      "step": 9550
    },
    {
      "entropy": 6.150395250320434,
      "epoch": 0.9551656920077972,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004917663860638787,
      "loss": 6.1488,
      "mean_token_accuracy": 0.14865331947803498,
      "num_tokens": 18694133.0,
      "step": 9555
    },
    {
      "entropy": 6.213038492202759,
      "epoch": 0.9556655170690258,
      "grad_norm": 1.5,
      "learning_rate": 0.0004917568170219321,
      "loss": 6.1047,
      "mean_token_accuracy": 0.14572429805994033,
      "num_tokens": 18705084.0,
      "step": 9560
    },
    {
      "entropy": 6.198029661178589,
      "epoch": 0.9561653421302544,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004917472425264397,
      "loss": 6.0106,
      "mean_token_accuracy": 0.15527254864573478,
      "num_tokens": 18714822.0,
      "step": 9565
    },
    {
      "entropy": 6.250082778930664,
      "epoch": 0.956665167191483,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004917376625776425,
      "loss": 6.1566,
      "mean_token_accuracy": 0.1495956838130951,
      "num_tokens": 18725471.0,
      "step": 9570
    },
    {
      "entropy": 6.240475416183472,
      "epoch": 0.9571649922527116,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004917280771757814,
      "loss": 6.1247,
      "mean_token_accuracy": 0.1448077939450741,
      "num_tokens": 18734884.0,
      "step": 9575
    },
    {
      "entropy": 6.027051830291748,
      "epoch": 0.9576648173139402,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004917184863210977,
      "loss": 5.8465,
      "mean_token_accuracy": 0.16873854100704194,
      "num_tokens": 18744720.0,
      "step": 9580
    },
    {
      "entropy": 6.149065780639648,
      "epoch": 0.9581646423751687,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004917088900138326,
      "loss": 6.0891,
      "mean_token_accuracy": 0.15208886861801146,
      "num_tokens": 18753636.0,
      "step": 9585
    },
    {
      "entropy": 6.281219911575318,
      "epoch": 0.9586644674363972,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004916992882542276,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14902181178331375,
      "num_tokens": 18763750.0,
      "step": 9590
    },
    {
      "entropy": 6.195375919342041,
      "epoch": 0.9591642924976258,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004916896810425244,
      "loss": 6.038,
      "mean_token_accuracy": 0.15284986943006515,
      "num_tokens": 18773626.0,
      "step": 9595
    },
    {
      "entropy": 6.172411394119263,
      "epoch": 0.9596641175588544,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004916800683789645,
      "loss": 6.0397,
      "mean_token_accuracy": 0.15392668843269347,
      "num_tokens": 18782898.0,
      "step": 9600
    },
    {
      "entropy": 6.259211254119873,
      "epoch": 0.960163942620083,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004916704502637897,
      "loss": 6.1675,
      "mean_token_accuracy": 0.14454340934753418,
      "num_tokens": 18792894.0,
      "step": 9605
    },
    {
      "entropy": 6.222532320022583,
      "epoch": 0.9606637676813116,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004916608266972422,
      "loss": 6.1045,
      "mean_token_accuracy": 0.15336891114711762,
      "num_tokens": 18802030.0,
      "step": 9610
    },
    {
      "entropy": 6.264726400375366,
      "epoch": 0.9611635927425402,
      "grad_norm": 1.625,
      "learning_rate": 0.0004916511976795642,
      "loss": 6.1779,
      "mean_token_accuracy": 0.14640460461378096,
      "num_tokens": 18812011.0,
      "step": 9615
    },
    {
      "entropy": 6.246786403656006,
      "epoch": 0.9616634178037686,
      "grad_norm": 1.375,
      "learning_rate": 0.0004916415632109977,
      "loss": 6.0717,
      "mean_token_accuracy": 0.14942505955696106,
      "num_tokens": 18821217.0,
      "step": 9620
    },
    {
      "entropy": 6.264371013641357,
      "epoch": 0.9621632428649972,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004916319232917852,
      "loss": 6.1417,
      "mean_token_accuracy": 0.14131503477692603,
      "num_tokens": 18831597.0,
      "step": 9625
    },
    {
      "entropy": 6.286407995223999,
      "epoch": 0.9626630679262258,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004916222779221694,
      "loss": 6.0919,
      "mean_token_accuracy": 0.1549926981329918,
      "num_tokens": 18840925.0,
      "step": 9630
    },
    {
      "entropy": 6.216050720214843,
      "epoch": 0.9631628929874544,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004916126271023926,
      "loss": 6.1215,
      "mean_token_accuracy": 0.1503046229481697,
      "num_tokens": 18850838.0,
      "step": 9635
    },
    {
      "entropy": 6.361771821975708,
      "epoch": 0.963662718048683,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000491602970832698,
      "loss": 6.1351,
      "mean_token_accuracy": 0.14643780440092086,
      "num_tokens": 18860527.0,
      "step": 9640
    },
    {
      "entropy": 6.225776195526123,
      "epoch": 0.9641625431099116,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004915933091133284,
      "loss": 6.0575,
      "mean_token_accuracy": 0.15646929144859315,
      "num_tokens": 18870971.0,
      "step": 9645
    },
    {
      "entropy": 6.144989633560181,
      "epoch": 0.9646623681711401,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004915836419445268,
      "loss": 6.0391,
      "mean_token_accuracy": 0.15933233201503755,
      "num_tokens": 18879985.0,
      "step": 9650
    },
    {
      "entropy": 6.254337215423584,
      "epoch": 0.9651621932323686,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004915739693265365,
      "loss": 6.0708,
      "mean_token_accuracy": 0.14735745638608932,
      "num_tokens": 18889583.0,
      "step": 9655
    },
    {
      "entropy": 6.18007173538208,
      "epoch": 0.9656620182935972,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004915642912596009,
      "loss": 6.0709,
      "mean_token_accuracy": 0.15414795875549317,
      "num_tokens": 18900520.0,
      "step": 9660
    },
    {
      "entropy": 6.256510400772095,
      "epoch": 0.9661618433548258,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004915546077439634,
      "loss": 6.1167,
      "mean_token_accuracy": 0.14692374169826508,
      "num_tokens": 18910924.0,
      "step": 9665
    },
    {
      "entropy": 6.292695713043213,
      "epoch": 0.9666616684160544,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915449187798677,
      "loss": 6.0331,
      "mean_token_accuracy": 0.16000572890043258,
      "num_tokens": 18919628.0,
      "step": 9670
    },
    {
      "entropy": 6.2442426681518555,
      "epoch": 0.967161493477283,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004915352243675576,
      "loss": 6.1459,
      "mean_token_accuracy": 0.1535830833017826,
      "num_tokens": 18929779.0,
      "step": 9675
    },
    {
      "entropy": 6.231713676452637,
      "epoch": 0.9676613185385116,
      "grad_norm": 1.59375,
      "learning_rate": 0.000491525524507277,
      "loss": 6.0657,
      "mean_token_accuracy": 0.1500541165471077,
      "num_tokens": 18939791.0,
      "step": 9680
    },
    {
      "entropy": 6.177619600296021,
      "epoch": 0.96816114359974,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004915158191992699,
      "loss": 5.9966,
      "mean_token_accuracy": 0.1484245926141739,
      "num_tokens": 18949611.0,
      "step": 9685
    },
    {
      "entropy": 6.231144666671753,
      "epoch": 0.9686609686609686,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004915061084437805,
      "loss": 6.0672,
      "mean_token_accuracy": 0.1534964770078659,
      "num_tokens": 18959018.0,
      "step": 9690
    },
    {
      "entropy": 6.257300567626953,
      "epoch": 0.9691607937221972,
      "grad_norm": 1.375,
      "learning_rate": 0.0004914963922410531,
      "loss": 6.0041,
      "mean_token_accuracy": 0.15159140527248383,
      "num_tokens": 18967460.0,
      "step": 9695
    },
    {
      "entropy": 6.222787094116211,
      "epoch": 0.9696606187834258,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004914866705913322,
      "loss": 6.1252,
      "mean_token_accuracy": 0.14473089277744294,
      "num_tokens": 18977545.0,
      "step": 9700
    },
    {
      "entropy": 6.180872583389283,
      "epoch": 0.9701604438446544,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004914769434948623,
      "loss": 6.0891,
      "mean_token_accuracy": 0.15525092631578447,
      "num_tokens": 18987304.0,
      "step": 9705
    },
    {
      "entropy": 6.206975793838501,
      "epoch": 0.970660268905883,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004914672109518883,
      "loss": 6.1587,
      "mean_token_accuracy": 0.1489308997988701,
      "num_tokens": 18996571.0,
      "step": 9710
    },
    {
      "entropy": 6.301168060302734,
      "epoch": 0.9711600939671116,
      "grad_norm": 1.328125,
      "learning_rate": 0.000491457472962655,
      "loss": 6.0877,
      "mean_token_accuracy": 0.14536493718624116,
      "num_tokens": 19007345.0,
      "step": 9715
    },
    {
      "entropy": 6.28914966583252,
      "epoch": 0.97165991902834,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004914477295274073,
      "loss": 6.1015,
      "mean_token_accuracy": 0.14678324162960052,
      "num_tokens": 19018202.0,
      "step": 9720
    },
    {
      "entropy": 6.248431015014648,
      "epoch": 0.9721597440895686,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004914379806463905,
      "loss": 6.0721,
      "mean_token_accuracy": 0.1530905023217201,
      "num_tokens": 19027642.0,
      "step": 9725
    },
    {
      "entropy": 6.242633533477783,
      "epoch": 0.9726595691507972,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004914282263198498,
      "loss": 6.084,
      "mean_token_accuracy": 0.15435991436243057,
      "num_tokens": 19037433.0,
      "step": 9730
    },
    {
      "entropy": 6.235606050491333,
      "epoch": 0.9731593942120258,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004914184665480305,
      "loss": 6.1247,
      "mean_token_accuracy": 0.15331421345472335,
      "num_tokens": 19046835.0,
      "step": 9735
    },
    {
      "entropy": 6.244014549255371,
      "epoch": 0.9736592192732544,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004914087013311784,
      "loss": 6.1754,
      "mean_token_accuracy": 0.14731296375393868,
      "num_tokens": 19057109.0,
      "step": 9740
    },
    {
      "entropy": 6.170975303649902,
      "epoch": 0.974159044334483,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004913989306695389,
      "loss": 6.0832,
      "mean_token_accuracy": 0.14898450225591658,
      "num_tokens": 19067688.0,
      "step": 9745
    },
    {
      "entropy": 6.184198427200317,
      "epoch": 0.9746588693957114,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004913891545633582,
      "loss": 6.0473,
      "mean_token_accuracy": 0.1572623372077942,
      "num_tokens": 19077127.0,
      "step": 9750
    },
    {
      "entropy": 6.207472276687622,
      "epoch": 0.97515869445694,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000491379373012882,
      "loss": 6.1234,
      "mean_token_accuracy": 0.15586711913347245,
      "num_tokens": 19088797.0,
      "step": 9755
    },
    {
      "entropy": 6.17994122505188,
      "epoch": 0.9756585195181686,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004913695860183563,
      "loss": 6.0499,
      "mean_token_accuracy": 0.15262471735477448,
      "num_tokens": 19098647.0,
      "step": 9760
    },
    {
      "entropy": 6.154853343963623,
      "epoch": 0.9761583445793972,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004913597935800276,
      "loss": 6.0126,
      "mean_token_accuracy": 0.15679847598075866,
      "num_tokens": 19108385.0,
      "step": 9765
    },
    {
      "entropy": 6.287287664413452,
      "epoch": 0.9766581696406258,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004913499956981422,
      "loss": 6.1214,
      "mean_token_accuracy": 0.14269307106733323,
      "num_tokens": 19117593.0,
      "step": 9770
    },
    {
      "entropy": 6.2425737380981445,
      "epoch": 0.9771579947018544,
      "grad_norm": 1.5,
      "learning_rate": 0.0004913401923729465,
      "loss": 6.0476,
      "mean_token_accuracy": 0.1474355824291706,
      "num_tokens": 19127440.0,
      "step": 9775
    },
    {
      "entropy": 6.245705461502075,
      "epoch": 0.977657819763083,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004913303836046873,
      "loss": 6.0649,
      "mean_token_accuracy": 0.14675208777189255,
      "num_tokens": 19136649.0,
      "step": 9780
    },
    {
      "entropy": 6.185900449752808,
      "epoch": 0.9781576448243114,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004913205693936112,
      "loss": 6.118,
      "mean_token_accuracy": 0.14976860582828522,
      "num_tokens": 19147702.0,
      "step": 9785
    },
    {
      "entropy": 6.183582830429077,
      "epoch": 0.97865746988554,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004913107497399653,
      "loss": 6.0731,
      "mean_token_accuracy": 0.15234436914324762,
      "num_tokens": 19157146.0,
      "step": 9790
    },
    {
      "entropy": 6.254842948913574,
      "epoch": 0.9791572949467686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004913009246439966,
      "loss": 6.1693,
      "mean_token_accuracy": 0.1422388143837452,
      "num_tokens": 19167157.0,
      "step": 9795
    },
    {
      "entropy": 6.347534275054931,
      "epoch": 0.9796571200079972,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004912910941059525,
      "loss": 6.1307,
      "mean_token_accuracy": 0.1388924703001976,
      "num_tokens": 19177188.0,
      "step": 9800
    },
    {
      "entropy": 6.13766622543335,
      "epoch": 0.9801569450692258,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004912812581260799,
      "loss": 5.9403,
      "mean_token_accuracy": 0.1578400894999504,
      "num_tokens": 19185940.0,
      "step": 9805
    },
    {
      "entropy": 6.15360746383667,
      "epoch": 0.9806567701304544,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004912714167046265,
      "loss": 6.0404,
      "mean_token_accuracy": 0.14680216759443282,
      "num_tokens": 19194634.0,
      "step": 9810
    },
    {
      "entropy": 6.2024157524108885,
      "epoch": 0.9811565951916829,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004912615698418399,
      "loss": 6.0305,
      "mean_token_accuracy": 0.15555737987160684,
      "num_tokens": 19203835.0,
      "step": 9815
    },
    {
      "entropy": 6.206526374816894,
      "epoch": 0.9816564202529114,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004912517175379679,
      "loss": 6.0573,
      "mean_token_accuracy": 0.15136691480875014,
      "num_tokens": 19213458.0,
      "step": 9820
    },
    {
      "entropy": 6.305220270156861,
      "epoch": 0.98215624531414,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004912418597932584,
      "loss": 6.1254,
      "mean_token_accuracy": 0.1457577221095562,
      "num_tokens": 19222494.0,
      "step": 9825
    },
    {
      "entropy": 6.201668357849121,
      "epoch": 0.9826560703753686,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004912319966079592,
      "loss": 6.0478,
      "mean_token_accuracy": 0.15180102586746216,
      "num_tokens": 19232153.0,
      "step": 9830
    },
    {
      "entropy": 6.146945476531982,
      "epoch": 0.9831558954365972,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912221279823187,
      "loss": 6.0721,
      "mean_token_accuracy": 0.14704425856471062,
      "num_tokens": 19241673.0,
      "step": 9835
    },
    {
      "entropy": 6.327292728424072,
      "epoch": 0.9836557204978258,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004912122539165851,
      "loss": 6.0767,
      "mean_token_accuracy": 0.1497425191104412,
      "num_tokens": 19250146.0,
      "step": 9840
    },
    {
      "entropy": 6.329761695861817,
      "epoch": 0.9841555455590544,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004912023744110068,
      "loss": 6.1815,
      "mean_token_accuracy": 0.1504547968506813,
      "num_tokens": 19259969.0,
      "step": 9845
    },
    {
      "entropy": 6.149716329574585,
      "epoch": 0.9846553706202829,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004911924894658324,
      "loss": 6.089,
      "mean_token_accuracy": 0.14434171542525293,
      "num_tokens": 19269502.0,
      "step": 9850
    },
    {
      "entropy": 6.120806550979614,
      "epoch": 0.9851551956815114,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004911825990813107,
      "loss": 5.9868,
      "mean_token_accuracy": 0.158699232339859,
      "num_tokens": 19277965.0,
      "step": 9855
    },
    {
      "entropy": 6.187296581268311,
      "epoch": 0.98565502074274,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004911727032576903,
      "loss": 6.0346,
      "mean_token_accuracy": 0.15413086265325546,
      "num_tokens": 19287618.0,
      "step": 9860
    },
    {
      "entropy": 6.289133501052857,
      "epoch": 0.9861548458039686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004911628019952203,
      "loss": 6.0576,
      "mean_token_accuracy": 0.15355061814188958,
      "num_tokens": 19297865.0,
      "step": 9865
    },
    {
      "entropy": 6.214590930938721,
      "epoch": 0.9866546708651972,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004911528952941499,
      "loss": 6.0344,
      "mean_token_accuracy": 0.15351168662309647,
      "num_tokens": 19308957.0,
      "step": 9870
    },
    {
      "entropy": 6.206920337677002,
      "epoch": 0.9871544959264258,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004911429831547283,
      "loss": 6.1226,
      "mean_token_accuracy": 0.14904448688030242,
      "num_tokens": 19319310.0,
      "step": 9875
    },
    {
      "entropy": 6.200511407852173,
      "epoch": 0.9876543209876543,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004911330655772047,
      "loss": 6.0997,
      "mean_token_accuracy": 0.15194911360740662,
      "num_tokens": 19329854.0,
      "step": 9880
    },
    {
      "entropy": 6.271060800552368,
      "epoch": 0.9881541460488829,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000491123142561829,
      "loss": 6.0709,
      "mean_token_accuracy": 0.14249143451452256,
      "num_tokens": 19340327.0,
      "step": 9885
    },
    {
      "entropy": 6.200604200363159,
      "epoch": 0.9886539711101114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004911132141088505,
      "loss": 5.9787,
      "mean_token_accuracy": 0.1530202992260456,
      "num_tokens": 19349577.0,
      "step": 9890
    },
    {
      "entropy": 6.188788938522339,
      "epoch": 0.98915379617134,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004911032802185192,
      "loss": 6.1061,
      "mean_token_accuracy": 0.14456491619348527,
      "num_tokens": 19359601.0,
      "step": 9895
    },
    {
      "entropy": 6.2374790668487545,
      "epoch": 0.9896536212325686,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000491093340891085,
      "loss": 6.1038,
      "mean_token_accuracy": 0.14851144924759865,
      "num_tokens": 19368592.0,
      "step": 9900
    },
    {
      "entropy": 6.263872718811035,
      "epoch": 0.9901534462937972,
      "grad_norm": 1.375,
      "learning_rate": 0.0004910833961267979,
      "loss": 6.1023,
      "mean_token_accuracy": 0.14427311047911645,
      "num_tokens": 19379220.0,
      "step": 9905
    },
    {
      "entropy": 6.140177774429321,
      "epoch": 0.9906532713550258,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004910734459259081,
      "loss": 5.9914,
      "mean_token_accuracy": 0.1525796577334404,
      "num_tokens": 19388339.0,
      "step": 9910
    },
    {
      "entropy": 6.1730693817138675,
      "epoch": 0.9911530964162543,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004910634902886662,
      "loss": 6.0546,
      "mean_token_accuracy": 0.1553937539458275,
      "num_tokens": 19396873.0,
      "step": 9915
    },
    {
      "entropy": 6.239046335220337,
      "epoch": 0.9916529214774829,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004910535292153224,
      "loss": 6.0321,
      "mean_token_accuracy": 0.15082822665572165,
      "num_tokens": 19406268.0,
      "step": 9920
    },
    {
      "entropy": 6.319290399551392,
      "epoch": 0.9921527465387114,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004910435627061273,
      "loss": 6.1226,
      "mean_token_accuracy": 0.15568157881498337,
      "num_tokens": 19415478.0,
      "step": 9925
    },
    {
      "entropy": 6.1647711277008055,
      "epoch": 0.99265257159994,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004910335907613318,
      "loss": 6.0333,
      "mean_token_accuracy": 0.1527284175157547,
      "num_tokens": 19424835.0,
      "step": 9930
    },
    {
      "entropy": 6.185374593734741,
      "epoch": 0.9931523966611686,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004910236133811867,
      "loss": 6.0568,
      "mean_token_accuracy": 0.1525961771607399,
      "num_tokens": 19434147.0,
      "step": 9935
    },
    {
      "entropy": 6.297340631484985,
      "epoch": 0.9936522217223972,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004910136305659432,
      "loss": 6.1254,
      "mean_token_accuracy": 0.15179211944341658,
      "num_tokens": 19444157.0,
      "step": 9940
    },
    {
      "entropy": 6.230135440826416,
      "epoch": 0.9941520467836257,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004910036423158523,
      "loss": 6.0546,
      "mean_token_accuracy": 0.15903421193361283,
      "num_tokens": 19453859.0,
      "step": 9945
    },
    {
      "entropy": 6.215265846252441,
      "epoch": 0.9946518718448543,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004909936486311653,
      "loss": 6.1013,
      "mean_token_accuracy": 0.1508074626326561,
      "num_tokens": 19464191.0,
      "step": 9950
    },
    {
      "entropy": 6.231240510940552,
      "epoch": 0.9951516969060828,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004909836495121336,
      "loss": 6.0657,
      "mean_token_accuracy": 0.15187131762504577,
      "num_tokens": 19473025.0,
      "step": 9955
    },
    {
      "entropy": 6.278972148895264,
      "epoch": 0.9956515219673114,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004909736449590091,
      "loss": 6.1833,
      "mean_token_accuracy": 0.14728196263313292,
      "num_tokens": 19483329.0,
      "step": 9960
    },
    {
      "entropy": 6.299745559692383,
      "epoch": 0.99615134702854,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000490963634972043,
      "loss": 6.0889,
      "mean_token_accuracy": 0.14778028577566146,
      "num_tokens": 19492756.0,
      "step": 9965
    },
    {
      "entropy": 6.279085779190064,
      "epoch": 0.9966511720897686,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004909536195514875,
      "loss": 6.132,
      "mean_token_accuracy": 0.1445053368806839,
      "num_tokens": 19502363.0,
      "step": 9970
    },
    {
      "entropy": 6.269314527511597,
      "epoch": 0.9971509971509972,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004909435986975946,
      "loss": 6.1647,
      "mean_token_accuracy": 0.1455976828932762,
      "num_tokens": 19512180.0,
      "step": 9975
    },
    {
      "entropy": 6.286091423034668,
      "epoch": 0.9976508222122257,
      "grad_norm": 1.375,
      "learning_rate": 0.0004909335724106162,
      "loss": 6.1562,
      "mean_token_accuracy": 0.14785073772072793,
      "num_tokens": 19521819.0,
      "step": 9980
    },
    {
      "entropy": 6.232504177093506,
      "epoch": 0.9981506472734543,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004909235406908047,
      "loss": 6.099,
      "mean_token_accuracy": 0.14757360368967057,
      "num_tokens": 19531960.0,
      "step": 9985
    },
    {
      "entropy": 6.19018063545227,
      "epoch": 0.9986504723346828,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004909135035384126,
      "loss": 6.0574,
      "mean_token_accuracy": 0.15128915756940842,
      "num_tokens": 19542278.0,
      "step": 9990
    },
    {
      "entropy": 6.26682596206665,
      "epoch": 0.9991502973959114,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004909034609536922,
      "loss": 6.114,
      "mean_token_accuracy": 0.14700626730918884,
      "num_tokens": 19552740.0,
      "step": 9995
    },
    {
      "entropy": 6.278173446655273,
      "epoch": 0.99965012245714,
      "grad_norm": 2.65625,
      "learning_rate": 0.0004908934129368963,
      "loss": 6.0442,
      "mean_token_accuracy": 0.14751166850328445,
      "num_tokens": 19563174.0,
      "step": 10000
    },
    {
      "entropy": 6.215727170308431,
      "epoch": 1.0000999650122457,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004908833594882777,
      "loss": 6.0534,
      "mean_token_accuracy": 0.14916856421364677,
      "num_tokens": 19571268.0,
      "step": 10005
    },
    {
      "entropy": 6.2368239402771,
      "epoch": 1.0005997900734742,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004908733006080893,
      "loss": 5.9327,
      "mean_token_accuracy": 0.15712171345949172,
      "num_tokens": 19580615.0,
      "step": 10010
    },
    {
      "entropy": 6.20062289237976,
      "epoch": 1.001099615134703,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004908632362965842,
      "loss": 5.988,
      "mean_token_accuracy": 0.15762433260679246,
      "num_tokens": 19590983.0,
      "step": 10015
    },
    {
      "entropy": 6.205209684371948,
      "epoch": 1.0015994401959314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004908531665540158,
      "loss": 5.9794,
      "mean_token_accuracy": 0.15464912354946136,
      "num_tokens": 19600440.0,
      "step": 10020
    },
    {
      "entropy": 6.190546369552612,
      "epoch": 1.00209926525716,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000490843091380637,
      "loss": 5.9958,
      "mean_token_accuracy": 0.1520358145236969,
      "num_tokens": 19610779.0,
      "step": 10025
    },
    {
      "entropy": 6.187043952941894,
      "epoch": 1.0025990903183886,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004908330107767018,
      "loss": 6.0247,
      "mean_token_accuracy": 0.14727614521980287,
      "num_tokens": 19620729.0,
      "step": 10030
    },
    {
      "entropy": 6.2144287586212155,
      "epoch": 1.003098915379617,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004908229247424633,
      "loss": 6.0069,
      "mean_token_accuracy": 0.15553976744413375,
      "num_tokens": 19630493.0,
      "step": 10035
    },
    {
      "entropy": 6.191425228118897,
      "epoch": 1.0035987404408457,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004908128332781758,
      "loss": 5.9779,
      "mean_token_accuracy": 0.15235588774085046,
      "num_tokens": 19640672.0,
      "step": 10040
    },
    {
      "entropy": 6.191244840621948,
      "epoch": 1.0040985655020742,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004908027363840928,
      "loss": 6.0196,
      "mean_token_accuracy": 0.1485897608101368,
      "num_tokens": 19651009.0,
      "step": 10045
    },
    {
      "entropy": 6.262838840484619,
      "epoch": 1.004598390563303,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004907926340604686,
      "loss": 6.0454,
      "mean_token_accuracy": 0.15007798597216607,
      "num_tokens": 19661422.0,
      "step": 10050
    },
    {
      "entropy": 6.23834981918335,
      "epoch": 1.0050982156245314,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004907825263075573,
      "loss": 5.9655,
      "mean_token_accuracy": 0.14877621456980705,
      "num_tokens": 19671039.0,
      "step": 10055
    },
    {
      "entropy": 6.146915912628174,
      "epoch": 1.00559804068576,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004907724131256129,
      "loss": 5.9821,
      "mean_token_accuracy": 0.15393402501940728,
      "num_tokens": 19680385.0,
      "step": 10060
    },
    {
      "entropy": 6.209895324707031,
      "epoch": 1.0060978657469886,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004907622945148903,
      "loss": 6.0474,
      "mean_token_accuracy": 0.15334030836820603,
      "num_tokens": 19690626.0,
      "step": 10065
    },
    {
      "entropy": 6.257166147232056,
      "epoch": 1.006597690808217,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004907521704756438,
      "loss": 5.9879,
      "mean_token_accuracy": 0.14535645842552186,
      "num_tokens": 19700234.0,
      "step": 10070
    },
    {
      "entropy": 6.207798051834106,
      "epoch": 1.0070975158694457,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004907420410081284,
      "loss": 5.9935,
      "mean_token_accuracy": 0.15128348022699356,
      "num_tokens": 19708986.0,
      "step": 10075
    },
    {
      "entropy": 6.16580638885498,
      "epoch": 1.0075973409306742,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004907319061125986,
      "loss": 6.0278,
      "mean_token_accuracy": 0.14916041493415833,
      "num_tokens": 19718292.0,
      "step": 10080
    },
    {
      "entropy": 6.157388639450073,
      "epoch": 1.008097165991903,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004907217657893096,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15840791165828705,
      "num_tokens": 19727563.0,
      "step": 10085
    },
    {
      "entropy": 6.221927833557129,
      "epoch": 1.0085969910531314,
      "grad_norm": 1.5,
      "learning_rate": 0.0004907116200385165,
      "loss": 6.0021,
      "mean_token_accuracy": 0.14704521968960763,
      "num_tokens": 19737649.0,
      "step": 10090
    },
    {
      "entropy": 6.246343278884888,
      "epoch": 1.00909681611436,
      "grad_norm": 2.453125,
      "learning_rate": 0.0004907014688604745,
      "loss": 6.0027,
      "mean_token_accuracy": 0.15232981592416764,
      "num_tokens": 19746911.0,
      "step": 10095
    },
    {
      "entropy": 6.110485601425171,
      "epoch": 1.0095966411755886,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004906913122554391,
      "loss": 6.0028,
      "mean_token_accuracy": 0.15218098014593123,
      "num_tokens": 19757144.0,
      "step": 10100
    },
    {
      "entropy": 6.154960870742798,
      "epoch": 1.010096466236817,
      "grad_norm": 1.375,
      "learning_rate": 0.0004906811502236657,
      "loss": 5.93,
      "mean_token_accuracy": 0.15480373650789261,
      "num_tokens": 19766422.0,
      "step": 10105
    },
    {
      "entropy": 6.26474084854126,
      "epoch": 1.0105962912980457,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004906709827654101,
      "loss": 5.9835,
      "mean_token_accuracy": 0.1561916321516037,
      "num_tokens": 19774711.0,
      "step": 10110
    },
    {
      "entropy": 6.164414691925049,
      "epoch": 1.0110961163592742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004906608098809281,
      "loss": 6.0001,
      "mean_token_accuracy": 0.15171277672052383,
      "num_tokens": 19784347.0,
      "step": 10115
    },
    {
      "entropy": 6.239012908935547,
      "epoch": 1.011595941420503,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004906506315704756,
      "loss": 6.1191,
      "mean_token_accuracy": 0.14562618136405944,
      "num_tokens": 19796100.0,
      "step": 10120
    },
    {
      "entropy": 6.226727867126465,
      "epoch": 1.0120957664817314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004906404478343087,
      "loss": 5.9424,
      "mean_token_accuracy": 0.15364138036966324,
      "num_tokens": 19805942.0,
      "step": 10125
    },
    {
      "entropy": 6.164436483383179,
      "epoch": 1.0125955915429599,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004906302586726837,
      "loss": 5.9163,
      "mean_token_accuracy": 0.16098824441432952,
      "num_tokens": 19814707.0,
      "step": 10130
    },
    {
      "entropy": 6.165250301361084,
      "epoch": 1.0130954166041886,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004906200640858567,
      "loss": 6.0422,
      "mean_token_accuracy": 0.1479193851351738,
      "num_tokens": 19825061.0,
      "step": 10135
    },
    {
      "entropy": 6.234692239761353,
      "epoch": 1.013595241665417,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004906098640740846,
      "loss": 5.9866,
      "mean_token_accuracy": 0.15638822168111802,
      "num_tokens": 19834572.0,
      "step": 10140
    },
    {
      "entropy": 6.251728487014771,
      "epoch": 1.0140950667266457,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004905996586376236,
      "loss": 6.0396,
      "mean_token_accuracy": 0.1497843861579895,
      "num_tokens": 19844691.0,
      "step": 10145
    },
    {
      "entropy": 6.113300371170044,
      "epoch": 1.0145948917878742,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004905894477767308,
      "loss": 5.8998,
      "mean_token_accuracy": 0.16126980185508727,
      "num_tokens": 19854806.0,
      "step": 10150
    },
    {
      "entropy": 6.181596136093139,
      "epoch": 1.015094716849103,
      "grad_norm": 1.375,
      "learning_rate": 0.0004905792314916629,
      "loss": 6.0188,
      "mean_token_accuracy": 0.1518865332007408,
      "num_tokens": 19863943.0,
      "step": 10155
    },
    {
      "entropy": 6.252336072921753,
      "epoch": 1.0155945419103314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004905690097826769,
      "loss": 6.0063,
      "mean_token_accuracy": 0.1458735793828964,
      "num_tokens": 19873417.0,
      "step": 10160
    },
    {
      "entropy": 6.198622369766236,
      "epoch": 1.0160943669715599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004905587826500303,
      "loss": 5.9441,
      "mean_token_accuracy": 0.14900024384260177,
      "num_tokens": 19883088.0,
      "step": 10165
    },
    {
      "entropy": 6.223319864273071,
      "epoch": 1.0165941920327886,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004905485500939801,
      "loss": 5.9791,
      "mean_token_accuracy": 0.15268445983529091,
      "num_tokens": 19893851.0,
      "step": 10170
    },
    {
      "entropy": 6.123204135894776,
      "epoch": 1.017094017094017,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004905383121147838,
      "loss": 5.9449,
      "mean_token_accuracy": 0.14863955080509186,
      "num_tokens": 19904243.0,
      "step": 10175
    },
    {
      "entropy": 6.188581657409668,
      "epoch": 1.0175938421552457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905280687126991,
      "loss": 6.109,
      "mean_token_accuracy": 0.15098904073238373,
      "num_tokens": 19914825.0,
      "step": 10180
    },
    {
      "entropy": 6.213014888763428,
      "epoch": 1.0180936672164742,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004905178198879837,
      "loss": 6.0037,
      "mean_token_accuracy": 0.1505586475133896,
      "num_tokens": 19925455.0,
      "step": 10185
    },
    {
      "entropy": 6.193584346771241,
      "epoch": 1.018593492277703,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004905075656408953,
      "loss": 6.0134,
      "mean_token_accuracy": 0.15090175122022628,
      "num_tokens": 19935810.0,
      "step": 10190
    },
    {
      "entropy": 6.209959506988525,
      "epoch": 1.0190933173389314,
      "grad_norm": 1.28125,
      "learning_rate": 0.000490497305971692,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1518132969737053,
      "num_tokens": 19945508.0,
      "step": 10195
    },
    {
      "entropy": 6.1608971118927,
      "epoch": 1.0195931424001599,
      "grad_norm": 1.40625,
      "learning_rate": 0.000490487040880632,
      "loss": 5.938,
      "mean_token_accuracy": 0.15057673752307893,
      "num_tokens": 19955148.0,
      "step": 10200
    },
    {
      "entropy": 6.096030282974243,
      "epoch": 1.0200929674613886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004904767703679735,
      "loss": 5.9416,
      "mean_token_accuracy": 0.15454292446374893,
      "num_tokens": 19965607.0,
      "step": 10205
    },
    {
      "entropy": 6.1834112167358395,
      "epoch": 1.020592792522617,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004904664944339748,
      "loss": 5.9853,
      "mean_token_accuracy": 0.15563032627105713,
      "num_tokens": 19976598.0,
      "step": 10210
    },
    {
      "entropy": 6.192605447769165,
      "epoch": 1.0210926175838457,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004904562130788947,
      "loss": 5.9052,
      "mean_token_accuracy": 0.1585889995098114,
      "num_tokens": 19986183.0,
      "step": 10215
    },
    {
      "entropy": 6.214938449859619,
      "epoch": 1.0215924426450742,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004904459263029916,
      "loss": 6.0029,
      "mean_token_accuracy": 0.15410468429327012,
      "num_tokens": 19995930.0,
      "step": 10220
    },
    {
      "entropy": 6.208783435821533,
      "epoch": 1.0220922677063027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004904356341065245,
      "loss": 6.083,
      "mean_token_accuracy": 0.14757224470376967,
      "num_tokens": 20005826.0,
      "step": 10225
    },
    {
      "entropy": 6.20314245223999,
      "epoch": 1.0225920927675314,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004904253364897521,
      "loss": 6.0337,
      "mean_token_accuracy": 0.15435700714588166,
      "num_tokens": 20015057.0,
      "step": 10230
    },
    {
      "entropy": 6.259528112411499,
      "epoch": 1.0230919178287599,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004904150334529339,
      "loss": 5.9793,
      "mean_token_accuracy": 0.14383609890937804,
      "num_tokens": 20024181.0,
      "step": 10235
    },
    {
      "entropy": 6.195091295242309,
      "epoch": 1.0235917428899886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004904047249963289,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15413728654384612,
      "num_tokens": 20033453.0,
      "step": 10240
    },
    {
      "entropy": 6.16310567855835,
      "epoch": 1.024091567951217,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004903944111201964,
      "loss": 6.025,
      "mean_token_accuracy": 0.15242297649383546,
      "num_tokens": 20043324.0,
      "step": 10245
    },
    {
      "entropy": 6.14918212890625,
      "epoch": 1.0245913930124457,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004903840918247958,
      "loss": 6.0005,
      "mean_token_accuracy": 0.15244850963354112,
      "num_tokens": 20053010.0,
      "step": 10250
    },
    {
      "entropy": 6.263448572158813,
      "epoch": 1.0250912180736742,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000490373767110387,
      "loss": 5.972,
      "mean_token_accuracy": 0.1525738060474396,
      "num_tokens": 20063457.0,
      "step": 10255
    },
    {
      "entropy": 6.212802934646606,
      "epoch": 1.0255910431349027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004903634369772296,
      "loss": 5.9919,
      "mean_token_accuracy": 0.15551566779613496,
      "num_tokens": 20073479.0,
      "step": 10260
    },
    {
      "entropy": 6.1847161769866945,
      "epoch": 1.0260908681961314,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004903531014255835,
      "loss": 5.9986,
      "mean_token_accuracy": 0.15058821439743042,
      "num_tokens": 20082711.0,
      "step": 10265
    },
    {
      "entropy": 6.17112021446228,
      "epoch": 1.0265906932573599,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004903427604557089,
      "loss": 5.9608,
      "mean_token_accuracy": 0.15690279752016068,
      "num_tokens": 20092535.0,
      "step": 10270
    },
    {
      "entropy": 6.209600782394409,
      "epoch": 1.0270905183185886,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004903324140678658,
      "loss": 5.9444,
      "mean_token_accuracy": 0.160567906498909,
      "num_tokens": 20101564.0,
      "step": 10275
    },
    {
      "entropy": 6.028101444244385,
      "epoch": 1.027590343379817,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004903220622623146,
      "loss": 5.8611,
      "mean_token_accuracy": 0.16170192658901214,
      "num_tokens": 20110361.0,
      "step": 10280
    },
    {
      "entropy": 6.143051528930664,
      "epoch": 1.0280901684410457,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004903117050393157,
      "loss": 5.951,
      "mean_token_accuracy": 0.1578829377889633,
      "num_tokens": 20119816.0,
      "step": 10285
    },
    {
      "entropy": 6.206754207611084,
      "epoch": 1.0285899935022742,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004903013423991296,
      "loss": 6.0096,
      "mean_token_accuracy": 0.15609005987644195,
      "num_tokens": 20129246.0,
      "step": 10290
    },
    {
      "entropy": 6.2067467212677006,
      "epoch": 1.0290898185635027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004902909743420173,
      "loss": 5.9905,
      "mean_token_accuracy": 0.15093543902039527,
      "num_tokens": 20140029.0,
      "step": 10295
    },
    {
      "entropy": 6.123083400726318,
      "epoch": 1.0295896436247314,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004902806008682394,
      "loss": 5.9095,
      "mean_token_accuracy": 0.1591138571500778,
      "num_tokens": 20150684.0,
      "step": 10300
    },
    {
      "entropy": 6.148619747161865,
      "epoch": 1.0300894686859599,
      "grad_norm": 1.25,
      "learning_rate": 0.000490270221978057,
      "loss": 5.9789,
      "mean_token_accuracy": 0.15402569323778154,
      "num_tokens": 20160007.0,
      "step": 10305
    },
    {
      "entropy": 6.2723650455474855,
      "epoch": 1.0305892937471886,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004902598376717312,
      "loss": 6.0852,
      "mean_token_accuracy": 0.14653107672929763,
      "num_tokens": 20170512.0,
      "step": 10310
    },
    {
      "entropy": 6.137090158462525,
      "epoch": 1.031089118808417,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004902494479495233,
      "loss": 5.9771,
      "mean_token_accuracy": 0.15818046256899834,
      "num_tokens": 20181954.0,
      "step": 10315
    },
    {
      "entropy": 6.193036031723023,
      "epoch": 1.0315889438696457,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004902390528116946,
      "loss": 5.972,
      "mean_token_accuracy": 0.15062784850597383,
      "num_tokens": 20191780.0,
      "step": 10320
    },
    {
      "entropy": 6.193860244750977,
      "epoch": 1.0320887689308742,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004902286522585068,
      "loss": 6.037,
      "mean_token_accuracy": 0.14930700212717057,
      "num_tokens": 20201646.0,
      "step": 10325
    },
    {
      "entropy": 6.152127456665039,
      "epoch": 1.0325885939921027,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004902182462902215,
      "loss": 6.0034,
      "mean_token_accuracy": 0.1526464432477951,
      "num_tokens": 20210943.0,
      "step": 10330
    },
    {
      "entropy": 6.185603857040405,
      "epoch": 1.0330884190533314,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004902078349071004,
      "loss": 5.9851,
      "mean_token_accuracy": 0.14743729680776596,
      "num_tokens": 20219901.0,
      "step": 10335
    },
    {
      "entropy": 6.192271423339844,
      "epoch": 1.0335882441145599,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004901974181094057,
      "loss": 5.9862,
      "mean_token_accuracy": 0.15042221397161484,
      "num_tokens": 20229630.0,
      "step": 10340
    },
    {
      "entropy": 6.256238222122192,
      "epoch": 1.0340880691757885,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004901869958973991,
      "loss": 6.0319,
      "mean_token_accuracy": 0.14707885533571244,
      "num_tokens": 20239932.0,
      "step": 10345
    },
    {
      "entropy": 6.2349104404449465,
      "epoch": 1.034587894237017,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004901765682713432,
      "loss": 5.9631,
      "mean_token_accuracy": 0.1530364841222763,
      "num_tokens": 20248863.0,
      "step": 10350
    },
    {
      "entropy": 6.094595336914063,
      "epoch": 1.0350877192982457,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004901661352315003,
      "loss": 5.8512,
      "mean_token_accuracy": 0.1574080094695091,
      "num_tokens": 20257916.0,
      "step": 10355
    },
    {
      "entropy": 6.1496976852417,
      "epoch": 1.0355875443594742,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004901556967781327,
      "loss": 5.985,
      "mean_token_accuracy": 0.15070026367902756,
      "num_tokens": 20267579.0,
      "step": 10360
    },
    {
      "entropy": 6.205862045288086,
      "epoch": 1.0360873694207027,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004901452529115031,
      "loss": 6.0197,
      "mean_token_accuracy": 0.14466966539621354,
      "num_tokens": 20278909.0,
      "step": 10365
    },
    {
      "entropy": 6.214945316314697,
      "epoch": 1.0365871944819314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004901348036318743,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15217370837926864,
      "num_tokens": 20288933.0,
      "step": 10370
    },
    {
      "entropy": 6.176287508010864,
      "epoch": 1.0370870195431598,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004901243489395093,
      "loss": 5.991,
      "mean_token_accuracy": 0.1489490918815136,
      "num_tokens": 20298329.0,
      "step": 10375
    },
    {
      "entropy": 6.149645137786865,
      "epoch": 1.0375868446043885,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000490113888834671,
      "loss": 6.0245,
      "mean_token_accuracy": 0.15549336820840837,
      "num_tokens": 20308742.0,
      "step": 10380
    },
    {
      "entropy": 6.092371702194214,
      "epoch": 1.038086669665617,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004901034233176227,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16808654516935348,
      "num_tokens": 20318379.0,
      "step": 10385
    },
    {
      "entropy": 6.163353443145752,
      "epoch": 1.0385864947268455,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004900929523886274,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15306385606527328,
      "num_tokens": 20328127.0,
      "step": 10390
    },
    {
      "entropy": 6.14065990447998,
      "epoch": 1.0390863197880742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000490082476047949,
      "loss": 5.9863,
      "mean_token_accuracy": 0.15795013457536697,
      "num_tokens": 20338562.0,
      "step": 10395
    },
    {
      "entropy": 6.212368822097778,
      "epoch": 1.0395861448493027,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004900719942958506,
      "loss": 5.9542,
      "mean_token_accuracy": 0.1611110895872116,
      "num_tokens": 20346984.0,
      "step": 10400
    },
    {
      "entropy": 6.159471082687378,
      "epoch": 1.0400859699105314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004900615071325963,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15324174016714096,
      "num_tokens": 20356666.0,
      "step": 10405
    },
    {
      "entropy": 6.172993850708008,
      "epoch": 1.0405857949717598,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00049005101455845,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15331986397504807,
      "num_tokens": 20365834.0,
      "step": 10410
    },
    {
      "entropy": 6.201247453689575,
      "epoch": 1.0410856200329885,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004900405165736753,
      "loss": 6.0249,
      "mean_token_accuracy": 0.1481031969189644,
      "num_tokens": 20377184.0,
      "step": 10415
    },
    {
      "entropy": 6.143035411834717,
      "epoch": 1.041585445094217,
      "grad_norm": 1.625,
      "learning_rate": 0.0004900300131785365,
      "loss": 5.9468,
      "mean_token_accuracy": 0.15708318948745728,
      "num_tokens": 20387059.0,
      "step": 10420
    },
    {
      "entropy": 6.213377332687378,
      "epoch": 1.0420852701554455,
      "grad_norm": 1.265625,
      "learning_rate": 0.000490019504373298,
      "loss": 6.0628,
      "mean_token_accuracy": 0.1523689314723015,
      "num_tokens": 20397162.0,
      "step": 10425
    },
    {
      "entropy": 6.110128116607666,
      "epoch": 1.0425850952166742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004900089901582241,
      "loss": 5.8713,
      "mean_token_accuracy": 0.15967605784535407,
      "num_tokens": 20407755.0,
      "step": 10430
    },
    {
      "entropy": 6.176903963088989,
      "epoch": 1.0430849202779027,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004899984705335792,
      "loss": 5.9783,
      "mean_token_accuracy": 0.15300601869821548,
      "num_tokens": 20417729.0,
      "step": 10435
    },
    {
      "entropy": 6.20291805267334,
      "epoch": 1.0435847453391314,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004899879454996282,
      "loss": 5.9962,
      "mean_token_accuracy": 0.1497774600982666,
      "num_tokens": 20428523.0,
      "step": 10440
    },
    {
      "entropy": 6.073294687271118,
      "epoch": 1.0440845704003598,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004899774150566358,
      "loss": 5.941,
      "mean_token_accuracy": 0.16532975137233735,
      "num_tokens": 20438667.0,
      "step": 10445
    },
    {
      "entropy": 6.150847434997559,
      "epoch": 1.0445843954615885,
      "grad_norm": 1.8984375,
      "learning_rate": 0.000489966879204867,
      "loss": 5.9844,
      "mean_token_accuracy": 0.1478474259376526,
      "num_tokens": 20449568.0,
      "step": 10450
    },
    {
      "entropy": 6.16922173500061,
      "epoch": 1.045084220522817,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004899563379445869,
      "loss": 5.969,
      "mean_token_accuracy": 0.1540171794593334,
      "num_tokens": 20458665.0,
      "step": 10455
    },
    {
      "entropy": 6.15083270072937,
      "epoch": 1.0455840455840455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004899457912760605,
      "loss": 5.9322,
      "mean_token_accuracy": 0.16014875173568727,
      "num_tokens": 20468587.0,
      "step": 10460
    },
    {
      "entropy": 6.1498230457305905,
      "epoch": 1.0460838706452742,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004899352391995533,
      "loss": 5.93,
      "mean_token_accuracy": 0.15350079238414766,
      "num_tokens": 20479186.0,
      "step": 10465
    },
    {
      "entropy": 6.2104816913604735,
      "epoch": 1.0465836957065027,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004899246817153309,
      "loss": 6.0359,
      "mean_token_accuracy": 0.14881793037056923,
      "num_tokens": 20488035.0,
      "step": 10470
    },
    {
      "entropy": 6.212858724594116,
      "epoch": 1.0470835207677314,
      "grad_norm": 1.375,
      "learning_rate": 0.0004899141188236587,
      "loss": 5.9851,
      "mean_token_accuracy": 0.14767462611198426,
      "num_tokens": 20498189.0,
      "step": 10475
    },
    {
      "entropy": 6.193490362167358,
      "epoch": 1.0475833458289598,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004899035505248027,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15572832077741622,
      "num_tokens": 20508199.0,
      "step": 10480
    },
    {
      "entropy": 6.182927465438842,
      "epoch": 1.0480831708901883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004898929768190286,
      "loss": 6.0734,
      "mean_token_accuracy": 0.14621295630931855,
      "num_tokens": 20517155.0,
      "step": 10485
    },
    {
      "entropy": 6.143451976776123,
      "epoch": 1.048582995951417,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004898823977066024,
      "loss": 5.9604,
      "mean_token_accuracy": 0.15642309337854385,
      "num_tokens": 20526559.0,
      "step": 10490
    },
    {
      "entropy": 6.154499626159668,
      "epoch": 1.0490828210126455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004898718131877906,
      "loss": 6.0186,
      "mean_token_accuracy": 0.15283958464860917,
      "num_tokens": 20536299.0,
      "step": 10495
    },
    {
      "entropy": 6.115988254547119,
      "epoch": 1.0495826460738742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004898612232628591,
      "loss": 5.8515,
      "mean_token_accuracy": 0.1695870116353035,
      "num_tokens": 20545290.0,
      "step": 10500
    },
    {
      "entropy": 6.194677209854126,
      "epoch": 1.0500824711351027,
      "grad_norm": 1.5,
      "learning_rate": 0.0004898506279320745,
      "loss": 5.9584,
      "mean_token_accuracy": 0.1560017205774784,
      "num_tokens": 20554691.0,
      "step": 10505
    },
    {
      "entropy": 6.115988445281983,
      "epoch": 1.0505822961963314,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004898400271957034,
      "loss": 5.9716,
      "mean_token_accuracy": 0.14933924823999406,
      "num_tokens": 20564595.0,
      "step": 10510
    },
    {
      "entropy": 6.099642992019653,
      "epoch": 1.0510821212575598,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898294210540126,
      "loss": 5.9168,
      "mean_token_accuracy": 0.15362110435962678,
      "num_tokens": 20574174.0,
      "step": 10515
    },
    {
      "entropy": 6.153954362869262,
      "epoch": 1.0515819463187883,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004898188095072688,
      "loss": 5.9317,
      "mean_token_accuracy": 0.16048941016197205,
      "num_tokens": 20583666.0,
      "step": 10520
    },
    {
      "entropy": 6.273473072052002,
      "epoch": 1.052081771380017,
      "grad_norm": 1.578125,
      "learning_rate": 0.000489808192555739,
      "loss": 6.0436,
      "mean_token_accuracy": 0.14504244551062584,
      "num_tokens": 20594040.0,
      "step": 10525
    },
    {
      "entropy": 6.212101459503174,
      "epoch": 1.0525815964412455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004897975701996903,
      "loss": 6.0537,
      "mean_token_accuracy": 0.1437474213540554,
      "num_tokens": 20605329.0,
      "step": 10530
    },
    {
      "entropy": 6.086110591888428,
      "epoch": 1.0530814215024742,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004897869424393902,
      "loss": 5.8786,
      "mean_token_accuracy": 0.16070354133844375,
      "num_tokens": 20615107.0,
      "step": 10535
    },
    {
      "entropy": 6.103875398635864,
      "epoch": 1.0535812465637027,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004897763092751057,
      "loss": 5.9151,
      "mean_token_accuracy": 0.1589644059538841,
      "num_tokens": 20624636.0,
      "step": 10540
    },
    {
      "entropy": 6.187276363372803,
      "epoch": 1.0540810716249314,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004897656707071045,
      "loss": 5.8998,
      "mean_token_accuracy": 0.15897430926561357,
      "num_tokens": 20634541.0,
      "step": 10545
    },
    {
      "entropy": 6.1509631156921385,
      "epoch": 1.0545808966861598,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004897550267356545,
      "loss": 5.9818,
      "mean_token_accuracy": 0.1588425800204277,
      "num_tokens": 20644941.0,
      "step": 10550
    },
    {
      "entropy": 6.178316926956176,
      "epoch": 1.0550807217473883,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004897443773610231,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16408261358737947,
      "num_tokens": 20653609.0,
      "step": 10555
    },
    {
      "entropy": 6.158814334869385,
      "epoch": 1.055580546808617,
      "grad_norm": 1.375,
      "learning_rate": 0.0004897337225834785,
      "loss": 5.907,
      "mean_token_accuracy": 0.16265378296375274,
      "num_tokens": 20663302.0,
      "step": 10560
    },
    {
      "entropy": 6.110200214385986,
      "epoch": 1.0560803718698455,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004897230624032888,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16467681229114534,
      "num_tokens": 20673785.0,
      "step": 10565
    },
    {
      "entropy": 6.102143144607544,
      "epoch": 1.0565801969310742,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004897123968207219,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15707050412893295,
      "num_tokens": 20683156.0,
      "step": 10570
    },
    {
      "entropy": 6.1429486751556395,
      "epoch": 1.0570800219923027,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004897017258360465,
      "loss": 5.9769,
      "mean_token_accuracy": 0.15604089945554733,
      "num_tokens": 20692153.0,
      "step": 10575
    },
    {
      "entropy": 6.193581771850586,
      "epoch": 1.0575798470535314,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004896910494495308,
      "loss": 5.9916,
      "mean_token_accuracy": 0.1569104701280594,
      "num_tokens": 20702343.0,
      "step": 10580
    },
    {
      "entropy": 6.0397984981536865,
      "epoch": 1.0580796721147598,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004896803676614436,
      "loss": 5.7977,
      "mean_token_accuracy": 0.17679254710674286,
      "num_tokens": 20712866.0,
      "step": 10585
    },
    {
      "entropy": 6.1743958473205565,
      "epoch": 1.0585794971759883,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004896696804720536,
      "loss": 6.0742,
      "mean_token_accuracy": 0.14831232130527497,
      "num_tokens": 20722679.0,
      "step": 10590
    },
    {
      "entropy": 6.236320734024048,
      "epoch": 1.059079322237217,
      "grad_norm": 1.5,
      "learning_rate": 0.0004896589878816297,
      "loss": 6.0395,
      "mean_token_accuracy": 0.15379220098257065,
      "num_tokens": 20732177.0,
      "step": 10595
    },
    {
      "entropy": 6.20904049873352,
      "epoch": 1.0595791472984455,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004896482898904409,
      "loss": 6.0499,
      "mean_token_accuracy": 0.15035803914070128,
      "num_tokens": 20742900.0,
      "step": 10600
    },
    {
      "entropy": 6.151769351959229,
      "epoch": 1.0600789723596742,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004896375864987563,
      "loss": 5.8601,
      "mean_token_accuracy": 0.15999141633510588,
      "num_tokens": 20752039.0,
      "step": 10605
    },
    {
      "entropy": 6.113235187530518,
      "epoch": 1.0605787974209027,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004896268777068453,
      "loss": 5.9089,
      "mean_token_accuracy": 0.16152562573552132,
      "num_tokens": 20760352.0,
      "step": 10610
    },
    {
      "entropy": 6.255431079864502,
      "epoch": 1.0610786224821314,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004896161635149773,
      "loss": 6.0917,
      "mean_token_accuracy": 0.14888167306780814,
      "num_tokens": 20770038.0,
      "step": 10615
    },
    {
      "entropy": 6.148290061950684,
      "epoch": 1.0615784475433598,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004896054439234217,
      "loss": 5.8755,
      "mean_token_accuracy": 0.16387573033571243,
      "num_tokens": 20779076.0,
      "step": 10620
    },
    {
      "entropy": 6.169048881530761,
      "epoch": 1.0620782726045883,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004895947189324486,
      "loss": 5.986,
      "mean_token_accuracy": 0.1495654746890068,
      "num_tokens": 20789272.0,
      "step": 10625
    },
    {
      "entropy": 6.15388970375061,
      "epoch": 1.062578097665817,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004895839885423273,
      "loss": 5.9756,
      "mean_token_accuracy": 0.15637347623705863,
      "num_tokens": 20799381.0,
      "step": 10630
    },
    {
      "entropy": 6.147679853439331,
      "epoch": 1.0630779227270455,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004895732527533283,
      "loss": 5.8733,
      "mean_token_accuracy": 0.16132555156946182,
      "num_tokens": 20809886.0,
      "step": 10635
    },
    {
      "entropy": 6.198393630981445,
      "epoch": 1.0635777477882742,
      "grad_norm": 1.375,
      "learning_rate": 0.0004895625115657215,
      "loss": 5.9998,
      "mean_token_accuracy": 0.15395113378763198,
      "num_tokens": 20819625.0,
      "step": 10640
    },
    {
      "entropy": 6.126437711715698,
      "epoch": 1.0640775728495027,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000489551764979777,
      "loss": 5.945,
      "mean_token_accuracy": 0.15016387403011322,
      "num_tokens": 20829481.0,
      "step": 10645
    },
    {
      "entropy": 6.1206090450286865,
      "epoch": 1.0645773979107314,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004895410129957653,
      "loss": 5.9779,
      "mean_token_accuracy": 0.15460684448480605,
      "num_tokens": 20840715.0,
      "step": 10650
    },
    {
      "entropy": 6.147835779190063,
      "epoch": 1.0650772229719598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004895302556139569,
      "loss": 5.9754,
      "mean_token_accuracy": 0.1584075063467026,
      "num_tokens": 20851334.0,
      "step": 10655
    },
    {
      "entropy": 6.195162630081176,
      "epoch": 1.0655770480331883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004895194928346226,
      "loss": 5.92,
      "mean_token_accuracy": 0.15483710765838624,
      "num_tokens": 20860357.0,
      "step": 10660
    },
    {
      "entropy": 6.14746470451355,
      "epoch": 1.066076873094417,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000489508724658033,
      "loss": 6.0633,
      "mean_token_accuracy": 0.1499548688530922,
      "num_tokens": 20870360.0,
      "step": 10665
    },
    {
      "entropy": 6.260917472839355,
      "epoch": 1.0665766981556455,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000489497951084459,
      "loss": 5.9824,
      "mean_token_accuracy": 0.1578219309449196,
      "num_tokens": 20880640.0,
      "step": 10670
    },
    {
      "entropy": 6.170552492141724,
      "epoch": 1.0670765232168742,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004894871721141718,
      "loss": 5.972,
      "mean_token_accuracy": 0.15283366292715073,
      "num_tokens": 20890524.0,
      "step": 10675
    },
    {
      "entropy": 6.1482892513275145,
      "epoch": 1.0675763482781027,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004894763877474426,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1547914445400238,
      "num_tokens": 20901469.0,
      "step": 10680
    },
    {
      "entropy": 6.156713390350342,
      "epoch": 1.0680761733393311,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004894655979845427,
      "loss": 6.0089,
      "mean_token_accuracy": 0.1489558055996895,
      "num_tokens": 20910703.0,
      "step": 10685
    },
    {
      "entropy": 6.20417423248291,
      "epoch": 1.0685759984005598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004894548028257434,
      "loss": 5.9498,
      "mean_token_accuracy": 0.15425267964601516,
      "num_tokens": 20920241.0,
      "step": 10690
    },
    {
      "entropy": 6.130399608612061,
      "epoch": 1.0690758234617883,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004894440022713165,
      "loss": 5.9786,
      "mean_token_accuracy": 0.15912894010543824,
      "num_tokens": 20930138.0,
      "step": 10695
    },
    {
      "entropy": 6.1744224548339846,
      "epoch": 1.069575648523017,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004894331963215337,
      "loss": 5.9325,
      "mean_token_accuracy": 0.15975070297718047,
      "num_tokens": 20939592.0,
      "step": 10700
    },
    {
      "entropy": 6.137640428543091,
      "epoch": 1.0700754735842455,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000489422384976667,
      "loss": 5.9429,
      "mean_token_accuracy": 0.1629180505871773,
      "num_tokens": 20948208.0,
      "step": 10705
    },
    {
      "entropy": 6.147330570220947,
      "epoch": 1.0705752986454742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004894115682369881,
      "loss": 5.9431,
      "mean_token_accuracy": 0.15658419877290725,
      "num_tokens": 20958323.0,
      "step": 10710
    },
    {
      "entropy": 6.146972942352295,
      "epoch": 1.0710751237067027,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004894007461027693,
      "loss": 5.917,
      "mean_token_accuracy": 0.15976602435112,
      "num_tokens": 20967593.0,
      "step": 10715
    },
    {
      "entropy": 6.1270524024963375,
      "epoch": 1.0715749487679311,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004893899185742828,
      "loss": 5.92,
      "mean_token_accuracy": 0.15857637226581572,
      "num_tokens": 20977236.0,
      "step": 10720
    },
    {
      "entropy": 6.111745738983155,
      "epoch": 1.0720747738291598,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004893790856518014,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15700402855873108,
      "num_tokens": 20986346.0,
      "step": 10725
    },
    {
      "entropy": 6.096800708770752,
      "epoch": 1.0725745988903883,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004893682473355972,
      "loss": 5.8498,
      "mean_token_accuracy": 0.16415757238864898,
      "num_tokens": 20995794.0,
      "step": 10730
    },
    {
      "entropy": 6.081625032424927,
      "epoch": 1.073074423951617,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004893574036259432,
      "loss": 5.8587,
      "mean_token_accuracy": 0.16565004736185074,
      "num_tokens": 21005133.0,
      "step": 10735
    },
    {
      "entropy": 6.157091236114502,
      "epoch": 1.0735742490128455,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004893465545231119,
      "loss": 6.0029,
      "mean_token_accuracy": 0.15228886902332306,
      "num_tokens": 21014445.0,
      "step": 10740
    },
    {
      "entropy": 6.097654056549072,
      "epoch": 1.074074074074074,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004893357000273765,
      "loss": 5.8611,
      "mean_token_accuracy": 0.15485004484653472,
      "num_tokens": 21023093.0,
      "step": 10745
    },
    {
      "entropy": 6.139210844039917,
      "epoch": 1.0745738991353027,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048932484013901,
      "loss": 5.9517,
      "mean_token_accuracy": 0.15576918572187423,
      "num_tokens": 21031445.0,
      "step": 10750
    },
    {
      "entropy": 6.161428356170655,
      "epoch": 1.0750737241965311,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004893139748582857,
      "loss": 5.9328,
      "mean_token_accuracy": 0.1552219122648239,
      "num_tokens": 21041026.0,
      "step": 10755
    },
    {
      "entropy": 6.160657596588135,
      "epoch": 1.0755735492577598,
      "grad_norm": 1.578125,
      "learning_rate": 0.000489303104185477,
      "loss": 5.9822,
      "mean_token_accuracy": 0.1538809798657894,
      "num_tokens": 21050665.0,
      "step": 10760
    },
    {
      "entropy": 6.106903982162476,
      "epoch": 1.0760733743189883,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004892922281208573,
      "loss": 5.923,
      "mean_token_accuracy": 0.15252332091331483,
      "num_tokens": 21060470.0,
      "step": 10765
    },
    {
      "entropy": 6.252746963500977,
      "epoch": 1.076573199380217,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004892813466647002,
      "loss": 6.0646,
      "mean_token_accuracy": 0.14947535172104837,
      "num_tokens": 21070174.0,
      "step": 10770
    },
    {
      "entropy": 6.255621767044067,
      "epoch": 1.0770730244414455,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004892704598172797,
      "loss": 6.0577,
      "mean_token_accuracy": 0.1452260285615921,
      "num_tokens": 21079686.0,
      "step": 10775
    },
    {
      "entropy": 6.224305820465088,
      "epoch": 1.077572849502674,
      "grad_norm": 1.25,
      "learning_rate": 0.0004892595675788696,
      "loss": 6.0067,
      "mean_token_accuracy": 0.15130189284682274,
      "num_tokens": 21089509.0,
      "step": 10780
    },
    {
      "entropy": 6.155075454711914,
      "epoch": 1.0780726745639027,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004892486699497439,
      "loss": 6.0434,
      "mean_token_accuracy": 0.14843635782599449,
      "num_tokens": 21099992.0,
      "step": 10785
    },
    {
      "entropy": 6.178429698944091,
      "epoch": 1.0785724996251311,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004892377669301766,
      "loss": 5.9625,
      "mean_token_accuracy": 0.15360143408179283,
      "num_tokens": 21109593.0,
      "step": 10790
    },
    {
      "entropy": 6.220020961761475,
      "epoch": 1.0790723246863598,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004892268585204423,
      "loss": 5.9312,
      "mean_token_accuracy": 0.15521704256534577,
      "num_tokens": 21119837.0,
      "step": 10795
    },
    {
      "entropy": 6.207529783248901,
      "epoch": 1.0795721497475883,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004892159447208154,
      "loss": 6.0223,
      "mean_token_accuracy": 0.150117227435112,
      "num_tokens": 21129908.0,
      "step": 10800
    },
    {
      "entropy": 6.183628845214844,
      "epoch": 1.080071974808817,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004892050255315705,
      "loss": 6.0275,
      "mean_token_accuracy": 0.15520211905241013,
      "num_tokens": 21140362.0,
      "step": 10805
    },
    {
      "entropy": 6.173284673690796,
      "epoch": 1.0805717998700455,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004891941009529822,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15586920976638793,
      "num_tokens": 21151617.0,
      "step": 10810
    },
    {
      "entropy": 6.159666299819946,
      "epoch": 1.081071624931274,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004891831709853255,
      "loss": 5.9835,
      "mean_token_accuracy": 0.1505484662950039,
      "num_tokens": 21160736.0,
      "step": 10815
    },
    {
      "entropy": 6.005563116073608,
      "epoch": 1.0815714499925027,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004891722356288752,
      "loss": 5.9138,
      "mean_token_accuracy": 0.16409018486738206,
      "num_tokens": 21170311.0,
      "step": 10820
    },
    {
      "entropy": 6.09224853515625,
      "epoch": 1.0820712750537311,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004891612948839068,
      "loss": 5.8853,
      "mean_token_accuracy": 0.1556367263197899,
      "num_tokens": 21179732.0,
      "step": 10825
    },
    {
      "entropy": 6.151172018051147,
      "epoch": 1.0825711001149598,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004891503487506951,
      "loss": 5.9036,
      "mean_token_accuracy": 0.15249819308519363,
      "num_tokens": 21189600.0,
      "step": 10830
    },
    {
      "entropy": 6.202696847915649,
      "epoch": 1.0830709251761883,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004891393972295159,
      "loss": 5.9524,
      "mean_token_accuracy": 0.15775438845157624,
      "num_tokens": 21199016.0,
      "step": 10835
    },
    {
      "entropy": 6.09175181388855,
      "epoch": 1.083570750237417,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004891284403206444,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15898170322179794,
      "num_tokens": 21208176.0,
      "step": 10840
    },
    {
      "entropy": 6.079210090637207,
      "epoch": 1.0840705752986455,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004891174780243566,
      "loss": 5.8803,
      "mean_token_accuracy": 0.16342149823904037,
      "num_tokens": 21217948.0,
      "step": 10845
    },
    {
      "entropy": 6.111714553833008,
      "epoch": 1.084570400359874,
      "grad_norm": 1.46875,
      "learning_rate": 0.000489106510340928,
      "loss": 5.9009,
      "mean_token_accuracy": 0.166984686255455,
      "num_tokens": 21227994.0,
      "step": 10850
    },
    {
      "entropy": 6.148243093490601,
      "epoch": 1.0850702254211027,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004890955372706348,
      "loss": 5.9027,
      "mean_token_accuracy": 0.16607349514961242,
      "num_tokens": 21237676.0,
      "step": 10855
    },
    {
      "entropy": 6.116034126281738,
      "epoch": 1.0855700504823311,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004890845588137531,
      "loss": 5.8774,
      "mean_token_accuracy": 0.1596287563443184,
      "num_tokens": 21248992.0,
      "step": 10860
    },
    {
      "entropy": 6.125313472747803,
      "epoch": 1.0860698755435598,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004890735749705587,
      "loss": 5.9495,
      "mean_token_accuracy": 0.1583699345588684,
      "num_tokens": 21258414.0,
      "step": 10865
    },
    {
      "entropy": 6.149082660675049,
      "epoch": 1.0865697006047883,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004890625857413283,
      "loss": 5.9325,
      "mean_token_accuracy": 0.1562967836856842,
      "num_tokens": 21267687.0,
      "step": 10870
    },
    {
      "entropy": 6.172228670120239,
      "epoch": 1.087069525666017,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004890515911263385,
      "loss": 5.9785,
      "mean_token_accuracy": 0.15632716417312623,
      "num_tokens": 21277623.0,
      "step": 10875
    },
    {
      "entropy": 6.1687828540802006,
      "epoch": 1.0875693507272455,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004890405911258656,
      "loss": 6.0299,
      "mean_token_accuracy": 0.1474304936826229,
      "num_tokens": 21288178.0,
      "step": 10880
    },
    {
      "entropy": 6.21861834526062,
      "epoch": 1.088069175788474,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004890295857401866,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15387789011001587,
      "num_tokens": 21298367.0,
      "step": 10885
    },
    {
      "entropy": 6.185503816604614,
      "epoch": 1.0885690008497027,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004890185749695783,
      "loss": 5.9832,
      "mean_token_accuracy": 0.15526565164327621,
      "num_tokens": 21307475.0,
      "step": 10890
    },
    {
      "entropy": 6.190867614746094,
      "epoch": 1.0890688259109311,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004890075588143176,
      "loss": 5.9778,
      "mean_token_accuracy": 0.14986664354801177,
      "num_tokens": 21317715.0,
      "step": 10895
    },
    {
      "entropy": 6.200929069519043,
      "epoch": 1.0895686509721598,
      "grad_norm": 1.25,
      "learning_rate": 0.0004889965372746819,
      "loss": 6.0205,
      "mean_token_accuracy": 0.14441814348101617,
      "num_tokens": 21327487.0,
      "step": 10900
    },
    {
      "entropy": 6.158633518218994,
      "epoch": 1.0900684760333883,
      "grad_norm": 1.375,
      "learning_rate": 0.0004889855103509484,
      "loss": 5.923,
      "mean_token_accuracy": 0.15800025016069413,
      "num_tokens": 21337588.0,
      "step": 10905
    },
    {
      "entropy": 6.1208963871002195,
      "epoch": 1.090568301094617,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004889744780433944,
      "loss": 5.7998,
      "mean_token_accuracy": 0.1666943535208702,
      "num_tokens": 21347037.0,
      "step": 10910
    },
    {
      "entropy": 6.101598167419434,
      "epoch": 1.0910681261558455,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004889634403522976,
      "loss": 5.9988,
      "mean_token_accuracy": 0.14901192486286163,
      "num_tokens": 21358784.0,
      "step": 10915
    },
    {
      "entropy": 6.119239044189453,
      "epoch": 1.091567951217074,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004889523972779359,
      "loss": 5.9772,
      "mean_token_accuracy": 0.1528187796473503,
      "num_tokens": 21368245.0,
      "step": 10920
    },
    {
      "entropy": 6.178964757919312,
      "epoch": 1.0920677762783026,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004889413488205868,
      "loss": 5.9218,
      "mean_token_accuracy": 0.16007325947284698,
      "num_tokens": 21377409.0,
      "step": 10925
    },
    {
      "entropy": 6.1624198913574215,
      "epoch": 1.0925676013395311,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004889302949805283,
      "loss": 5.9134,
      "mean_token_accuracy": 0.1607838898897171,
      "num_tokens": 21388109.0,
      "step": 10930
    },
    {
      "entropy": 6.073905563354492,
      "epoch": 1.0930674264007598,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004889192357580388,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16193442195653915,
      "num_tokens": 21396645.0,
      "step": 10935
    },
    {
      "entropy": 6.180797433853149,
      "epoch": 1.0935672514619883,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004889081711533964,
      "loss": 6.0247,
      "mean_token_accuracy": 0.1479413092136383,
      "num_tokens": 21407398.0,
      "step": 10940
    },
    {
      "entropy": 6.148538494110108,
      "epoch": 1.0940670765232168,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004888971011668793,
      "loss": 5.8556,
      "mean_token_accuracy": 0.1640845537185669,
      "num_tokens": 21417185.0,
      "step": 10945
    },
    {
      "entropy": 6.216218137741089,
      "epoch": 1.0945669015844455,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004888860257987662,
      "loss": 6.0475,
      "mean_token_accuracy": 0.14773295298218728,
      "num_tokens": 21426743.0,
      "step": 10950
    },
    {
      "entropy": 6.131222438812256,
      "epoch": 1.095066726645674,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004888749450493358,
      "loss": 5.9628,
      "mean_token_accuracy": 0.15966471955180167,
      "num_tokens": 21436528.0,
      "step": 10955
    },
    {
      "entropy": 6.1315834522247314,
      "epoch": 1.0955665517069026,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004888638589188669,
      "loss": 5.9586,
      "mean_token_accuracy": 0.1620188757777214,
      "num_tokens": 21445286.0,
      "step": 10960
    },
    {
      "entropy": 6.25554928779602,
      "epoch": 1.0960663767681311,
      "grad_norm": 1.75,
      "learning_rate": 0.0004888527674076382,
      "loss": 6.0834,
      "mean_token_accuracy": 0.15158738046884537,
      "num_tokens": 21455695.0,
      "step": 10965
    },
    {
      "entropy": 6.176150560379028,
      "epoch": 1.0965662018293598,
      "grad_norm": 1.59375,
      "learning_rate": 0.000488841670515929,
      "loss": 5.9498,
      "mean_token_accuracy": 0.15517933219671248,
      "num_tokens": 21465305.0,
      "step": 10970
    },
    {
      "entropy": 6.071147060394287,
      "epoch": 1.0970660268905883,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004888305682440183,
      "loss": 5.8685,
      "mean_token_accuracy": 0.1680384635925293,
      "num_tokens": 21475101.0,
      "step": 10975
    },
    {
      "entropy": 6.11292290687561,
      "epoch": 1.0975658519518168,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004888194605921855,
      "loss": 5.9074,
      "mean_token_accuracy": 0.16730916053056716,
      "num_tokens": 21484240.0,
      "step": 10980
    },
    {
      "entropy": 6.228977155685425,
      "epoch": 1.0980656770130455,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004888083475607102,
      "loss": 6.0206,
      "mean_token_accuracy": 0.1488730400800705,
      "num_tokens": 21494700.0,
      "step": 10985
    },
    {
      "entropy": 6.116166734695435,
      "epoch": 1.098565502074274,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004887972291498718,
      "loss": 5.9101,
      "mean_token_accuracy": 0.15668945237994195,
      "num_tokens": 21504418.0,
      "step": 10990
    },
    {
      "entropy": 6.185805702209473,
      "epoch": 1.0990653271355026,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004887861053599503,
      "loss": 5.8933,
      "mean_token_accuracy": 0.16097963154315947,
      "num_tokens": 21514019.0,
      "step": 10995
    },
    {
      "entropy": 6.142436122894287,
      "epoch": 1.0995651521967311,
      "grad_norm": 1.375,
      "learning_rate": 0.0004887749761912252,
      "loss": 5.9216,
      "mean_token_accuracy": 0.15695045590400697,
      "num_tokens": 21524279.0,
      "step": 11000
    },
    {
      "entropy": 6.1633013725280765,
      "epoch": 1.1000649772579596,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004887638416439768,
      "loss": 5.9377,
      "mean_token_accuracy": 0.15189138501882554,
      "num_tokens": 21533851.0,
      "step": 11005
    },
    {
      "entropy": 6.11244387626648,
      "epoch": 1.1005648023191883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004887527017184851,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15764300972223283,
      "num_tokens": 21542589.0,
      "step": 11010
    },
    {
      "entropy": 6.161046934127808,
      "epoch": 1.1010646273804168,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0004887415564150305,
      "loss": 5.9716,
      "mean_token_accuracy": 0.15137407705187797,
      "num_tokens": 21553018.0,
      "step": 11015
    },
    {
      "entropy": 6.308100891113281,
      "epoch": 1.1015644524416455,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004887304057338934,
      "loss": 6.0725,
      "mean_token_accuracy": 0.1491433709859848,
      "num_tokens": 21561830.0,
      "step": 11020
    },
    {
      "entropy": 6.200220108032227,
      "epoch": 1.102064277502874,
      "grad_norm": 1.453125,
      "learning_rate": 0.000488719249675354,
      "loss": 5.9272,
      "mean_token_accuracy": 0.15634792000055314,
      "num_tokens": 21571126.0,
      "step": 11025
    },
    {
      "entropy": 6.0580237865447994,
      "epoch": 1.1025641025641026,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004887080882396936,
      "loss": 5.9534,
      "mean_token_accuracy": 0.16289110034704207,
      "num_tokens": 21580771.0,
      "step": 11030
    },
    {
      "entropy": 6.186287593841553,
      "epoch": 1.1030639276253311,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004886969214271925,
      "loss": 6.0628,
      "mean_token_accuracy": 0.1516507513821125,
      "num_tokens": 21591687.0,
      "step": 11035
    },
    {
      "entropy": 6.104765462875366,
      "epoch": 1.1035637526865596,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004886857492381319,
      "loss": 5.9378,
      "mean_token_accuracy": 0.1597822442650795,
      "num_tokens": 21601056.0,
      "step": 11040
    },
    {
      "entropy": 6.157061290740967,
      "epoch": 1.1040635777477883,
      "grad_norm": 1.5,
      "learning_rate": 0.0004886745716727928,
      "loss": 5.9357,
      "mean_token_accuracy": 0.15853065103292466,
      "num_tokens": 21610846.0,
      "step": 11045
    },
    {
      "entropy": 6.155249214172363,
      "epoch": 1.1045634028090168,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004886633887314565,
      "loss": 5.9123,
      "mean_token_accuracy": 0.15513363033533095,
      "num_tokens": 21621504.0,
      "step": 11050
    },
    {
      "entropy": 6.098737478256226,
      "epoch": 1.1050632278702455,
      "grad_norm": 1.5,
      "learning_rate": 0.0004886522004144044,
      "loss": 5.8691,
      "mean_token_accuracy": 0.1679858535528183,
      "num_tokens": 21631566.0,
      "step": 11055
    },
    {
      "entropy": 6.191392993927002,
      "epoch": 1.105563052931474,
      "grad_norm": 1.5,
      "learning_rate": 0.0004886410067219178,
      "loss": 5.9294,
      "mean_token_accuracy": 0.158718304336071,
      "num_tokens": 21641392.0,
      "step": 11060
    },
    {
      "entropy": 6.148838186264038,
      "epoch": 1.1060628779927026,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004886298076542785,
      "loss": 5.8698,
      "mean_token_accuracy": 0.16629353761672974,
      "num_tokens": 21650356.0,
      "step": 11065
    },
    {
      "entropy": 6.181856775283814,
      "epoch": 1.1065627030539311,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004886186032117682,
      "loss": 6.0055,
      "mean_token_accuracy": 0.15484943091869355,
      "num_tokens": 21660704.0,
      "step": 11070
    },
    {
      "entropy": 6.137094974517822,
      "epoch": 1.1070625281151596,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004886073933946688,
      "loss": 5.9418,
      "mean_token_accuracy": 0.1511370614171028,
      "num_tokens": 21671002.0,
      "step": 11075
    },
    {
      "entropy": 6.094307088851929,
      "epoch": 1.1075623531763883,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004885961782032624,
      "loss": 5.9054,
      "mean_token_accuracy": 0.16321012377738953,
      "num_tokens": 21680802.0,
      "step": 11080
    },
    {
      "entropy": 6.125639820098877,
      "epoch": 1.1080621782376168,
      "grad_norm": 1.6328125,
      "learning_rate": 0.000488584957637831,
      "loss": 5.9052,
      "mean_token_accuracy": 0.16174551993608474,
      "num_tokens": 21690535.0,
      "step": 11085
    },
    {
      "entropy": 6.164075660705566,
      "epoch": 1.1085620032988455,
      "grad_norm": 1.625,
      "learning_rate": 0.0004885737316986571,
      "loss": 6.0181,
      "mean_token_accuracy": 0.1486960008740425,
      "num_tokens": 21700283.0,
      "step": 11090
    },
    {
      "entropy": 6.207850408554077,
      "epoch": 1.109061828360074,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000488562500386023,
      "loss": 5.9413,
      "mean_token_accuracy": 0.1550242632627487,
      "num_tokens": 21709927.0,
      "step": 11095
    },
    {
      "entropy": 6.129647541046142,
      "epoch": 1.1095616534213026,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004885512637002115,
      "loss": 5.8763,
      "mean_token_accuracy": 0.1647261068224907,
      "num_tokens": 21720011.0,
      "step": 11100
    },
    {
      "entropy": 6.143113231658935,
      "epoch": 1.1100614784825311,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004885400216415051,
      "loss": 5.9811,
      "mean_token_accuracy": 0.15524290651082992,
      "num_tokens": 21729456.0,
      "step": 11105
    },
    {
      "entropy": 6.240492582321167,
      "epoch": 1.1105613035437596,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004885287742101865,
      "loss": 6.035,
      "mean_token_accuracy": 0.15184471011161804,
      "num_tokens": 21739341.0,
      "step": 11110
    },
    {
      "entropy": 6.224813652038574,
      "epoch": 1.1110611286049883,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004885175214065392,
      "loss": 6.0383,
      "mean_token_accuracy": 0.14809160754084588,
      "num_tokens": 21749549.0,
      "step": 11115
    },
    {
      "entropy": 6.145921611785889,
      "epoch": 1.1115609536662168,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004885062632308457,
      "loss": 5.9093,
      "mean_token_accuracy": 0.15170687139034272,
      "num_tokens": 21760030.0,
      "step": 11120
    },
    {
      "entropy": 6.187044382095337,
      "epoch": 1.1120607787274455,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004884949996833898,
      "loss": 6.0067,
      "mean_token_accuracy": 0.15139136761426925,
      "num_tokens": 21769754.0,
      "step": 11125
    },
    {
      "entropy": 6.1913032054901125,
      "epoch": 1.112560603788674,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004884837307644545,
      "loss": 5.9882,
      "mean_token_accuracy": 0.15175608098506926,
      "num_tokens": 21780248.0,
      "step": 11130
    },
    {
      "entropy": 6.146751499176025,
      "epoch": 1.1130604288499026,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004884724564743236,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16302238553762435,
      "num_tokens": 21789014.0,
      "step": 11135
    },
    {
      "entropy": 6.11230788230896,
      "epoch": 1.1135602539111311,
      "grad_norm": 1.25,
      "learning_rate": 0.0004884611768132805,
      "loss": 5.9101,
      "mean_token_accuracy": 0.1565244510769844,
      "num_tokens": 21798210.0,
      "step": 11140
    },
    {
      "entropy": 6.127377939224243,
      "epoch": 1.1140600789723596,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004884498917816092,
      "loss": 5.929,
      "mean_token_accuracy": 0.15679699778556824,
      "num_tokens": 21807947.0,
      "step": 11145
    },
    {
      "entropy": 6.1256935596466064,
      "epoch": 1.1145599040335883,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004884386013795935,
      "loss": 5.9679,
      "mean_token_accuracy": 0.15247399508953094,
      "num_tokens": 21816685.0,
      "step": 11150
    },
    {
      "entropy": 6.238236331939698,
      "epoch": 1.1150597290948168,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004884273056075175,
      "loss": 5.9352,
      "mean_token_accuracy": 0.15477223992347716,
      "num_tokens": 21826479.0,
      "step": 11155
    },
    {
      "entropy": 6.1257719039917,
      "epoch": 1.1155595541560455,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004884160044656654,
      "loss": 5.9586,
      "mean_token_accuracy": 0.15268905758857726,
      "num_tokens": 21837518.0,
      "step": 11160
    },
    {
      "entropy": 6.089190578460693,
      "epoch": 1.116059379217274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004884046979543217,
      "loss": 5.8734,
      "mean_token_accuracy": 0.15650648176670073,
      "num_tokens": 21846069.0,
      "step": 11165
    },
    {
      "entropy": 6.194562721252441,
      "epoch": 1.1165592042785026,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004883933860737706,
      "loss": 6.0634,
      "mean_token_accuracy": 0.15407290160655976,
      "num_tokens": 21855594.0,
      "step": 11170
    },
    {
      "entropy": 6.205375289916992,
      "epoch": 1.1170590293397311,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004883820688242968,
      "loss": 5.9492,
      "mean_token_accuracy": 0.15076741874217986,
      "num_tokens": 21864648.0,
      "step": 11175
    },
    {
      "entropy": 6.189843606948853,
      "epoch": 1.1175588544009596,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004883707462061851,
      "loss": 5.9632,
      "mean_token_accuracy": 0.14979364946484566,
      "num_tokens": 21875516.0,
      "step": 11180
    },
    {
      "entropy": 6.118408966064453,
      "epoch": 1.1180586794621883,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004883594182197203,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15247273147106172,
      "num_tokens": 21885232.0,
      "step": 11185
    },
    {
      "entropy": 6.124889755249024,
      "epoch": 1.1185585045234168,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004883480848651875,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1589203357696533,
      "num_tokens": 21895913.0,
      "step": 11190
    },
    {
      "entropy": 6.204986810684204,
      "epoch": 1.1190583295846455,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004883367461428716,
      "loss": 6.0187,
      "mean_token_accuracy": 0.15594663172960282,
      "num_tokens": 21906160.0,
      "step": 11195
    },
    {
      "entropy": 6.1398121356964115,
      "epoch": 1.119558154645874,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004883254020530582,
      "loss": 5.889,
      "mean_token_accuracy": 0.16003624349832535,
      "num_tokens": 21915326.0,
      "step": 11200
    },
    {
      "entropy": 6.1442748546600345,
      "epoch": 1.1200579797071024,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004883140525960327,
      "loss": 5.958,
      "mean_token_accuracy": 0.15367740243673325,
      "num_tokens": 21925804.0,
      "step": 11205
    },
    {
      "entropy": 6.1771252155303955,
      "epoch": 1.1205578047683311,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004883026977720803,
      "loss": 5.8852,
      "mean_token_accuracy": 0.1577083170413971,
      "num_tokens": 21934226.0,
      "step": 11210
    },
    {
      "entropy": 6.057019662857056,
      "epoch": 1.1210576298295596,
      "grad_norm": 1.6796875,
      "learning_rate": 0.000488291337581487,
      "loss": 5.7742,
      "mean_token_accuracy": 0.17952765375375748,
      "num_tokens": 21944068.0,
      "step": 11215
    },
    {
      "entropy": 6.1522480010986325,
      "epoch": 1.1215574548907883,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004882799720245385,
      "loss": 6.0185,
      "mean_token_accuracy": 0.15693057477474212,
      "num_tokens": 21954509.0,
      "step": 11220
    },
    {
      "entropy": 6.139623069763184,
      "epoch": 1.1220572799520168,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004882686011015208,
      "loss": 6.0129,
      "mean_token_accuracy": 0.15139480531215668,
      "num_tokens": 21963753.0,
      "step": 11225
    },
    {
      "entropy": 6.134170007705689,
      "epoch": 1.1225571050132455,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048825722481272006,
      "loss": 5.9449,
      "mean_token_accuracy": 0.15493905767798424,
      "num_tokens": 21973634.0,
      "step": 11230
    },
    {
      "entropy": 6.1677783012390135,
      "epoch": 1.123056930074474,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004882458431584223,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1583867371082306,
      "num_tokens": 21984029.0,
      "step": 11235
    },
    {
      "entropy": 6.124644041061401,
      "epoch": 1.1235567551357024,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004882344561389141,
      "loss": 6.0195,
      "mean_token_accuracy": 0.1517020508646965,
      "num_tokens": 21994270.0,
      "step": 11240
    },
    {
      "entropy": 6.273230314254761,
      "epoch": 1.124056580196931,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048822306375448176,
      "loss": 6.0398,
      "mean_token_accuracy": 0.1463012471795082,
      "num_tokens": 22002911.0,
      "step": 11245
    },
    {
      "entropy": 6.126321411132812,
      "epoch": 1.1245564052581596,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004882116660054121,
      "loss": 6.016,
      "mean_token_accuracy": 0.1513655811548233,
      "num_tokens": 22012615.0,
      "step": 11250
    },
    {
      "entropy": 6.158256721496582,
      "epoch": 1.1250562303193883,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004882002628919917,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15256650745868683,
      "num_tokens": 22022274.0,
      "step": 11255
    },
    {
      "entropy": 6.136745119094849,
      "epoch": 1.1255560553806168,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004881888544145076,
      "loss": 5.8458,
      "mean_token_accuracy": 0.15463765114545822,
      "num_tokens": 22031238.0,
      "step": 11260
    },
    {
      "entropy": 6.132595491409302,
      "epoch": 1.1260558804418452,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048817744057324673,
      "loss": 5.9229,
      "mean_token_accuracy": 0.1540322870016098,
      "num_tokens": 22041003.0,
      "step": 11265
    },
    {
      "entropy": 6.0907604694366455,
      "epoch": 1.126555705503074,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048816602136849644,
      "loss": 5.9212,
      "mean_token_accuracy": 0.1601533830165863,
      "num_tokens": 22050621.0,
      "step": 11270
    },
    {
      "entropy": 6.1333081245422365,
      "epoch": 1.1270555305643024,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00048815459680054375,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15122018307447432,
      "num_tokens": 22060382.0,
      "step": 11275
    },
    {
      "entropy": 6.15812497138977,
      "epoch": 1.127555355625531,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048814316686967635,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15779192969202996,
      "num_tokens": 22069896.0,
      "step": 11280
    },
    {
      "entropy": 6.209627628326416,
      "epoch": 1.1280551806867596,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004881317315761817,
      "loss": 5.8687,
      "mean_token_accuracy": 0.1675721988081932,
      "num_tokens": 22080234.0,
      "step": 11285
    },
    {
      "entropy": 6.100448131561279,
      "epoch": 1.1285550057479883,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048812029092034747,
      "loss": 5.986,
      "mean_token_accuracy": 0.14991760849952698,
      "num_tokens": 22089866.0,
      "step": 11290
    },
    {
      "entropy": 6.049789524078369,
      "epoch": 1.1290548308092168,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048810884490246164,
      "loss": 5.9169,
      "mean_token_accuracy": 0.16051019877195358,
      "num_tokens": 22100740.0,
      "step": 11295
    },
    {
      "entropy": 6.228305149078369,
      "epoch": 1.1295546558704452,
      "grad_norm": 1.390625,
      "learning_rate": 0.000488097393522812,
      "loss": 5.944,
      "mean_token_accuracy": 0.15547244101762772,
      "num_tokens": 22110519.0,
      "step": 11300
    },
    {
      "entropy": 6.022213077545166,
      "epoch": 1.130054480931674,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004880859367816868,
      "loss": 5.873,
      "mean_token_accuracy": 0.16479989290237426,
      "num_tokens": 22119671.0,
      "step": 11305
    },
    {
      "entropy": 6.1264959335327145,
      "epoch": 1.1305543059929024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004880744746793744,
      "loss": 5.9348,
      "mean_token_accuracy": 0.14781904667615892,
      "num_tokens": 22130022.0,
      "step": 11310
    },
    {
      "entropy": 6.236459732055664,
      "epoch": 1.131054131054131,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048806300721616295,
      "loss": 6.0057,
      "mean_token_accuracy": 0.15469181090593337,
      "num_tokens": 22140875.0,
      "step": 11315
    },
    {
      "entropy": 6.188546752929687,
      "epoch": 1.1315539561153596,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048805153439234107,
      "loss": 5.9489,
      "mean_token_accuracy": 0.15679532885551453,
      "num_tokens": 22150144.0,
      "step": 11320
    },
    {
      "entropy": 6.079636573791504,
      "epoch": 1.1320537811765883,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004880400562081974,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15974163115024567,
      "num_tokens": 22158951.0,
      "step": 11325
    },
    {
      "entropy": 6.021935844421387,
      "epoch": 1.1325536062378168,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048802857266402075,
      "loss": 5.9182,
      "mean_token_accuracy": 0.15889034122228624,
      "num_tokens": 22167458.0,
      "step": 11330
    },
    {
      "entropy": 6.114293479919434,
      "epoch": 1.1330534312990452,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004880170837601,
      "loss": 5.8851,
      "mean_token_accuracy": 0.16265192329883577,
      "num_tokens": 22176760.0,
      "step": 11335
    },
    {
      "entropy": 6.234913635253906,
      "epoch": 1.133553256360274,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004880055894967242,
      "loss": 5.931,
      "mean_token_accuracy": 0.15519655048847197,
      "num_tokens": 22186476.0,
      "step": 11340
    },
    {
      "entropy": 6.04418454170227,
      "epoch": 1.1340530814215024,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048799408987418264,
      "loss": 5.9367,
      "mean_token_accuracy": 0.1631470128893852,
      "num_tokens": 22196420.0,
      "step": 11345
    },
    {
      "entropy": 6.117028617858887,
      "epoch": 1.134552906482731,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048798258489276457,
      "loss": 5.9602,
      "mean_token_accuracy": 0.15774163156747817,
      "num_tokens": 22204989.0,
      "step": 11350
    },
    {
      "entropy": 6.121002197265625,
      "epoch": 1.1350527315439596,
      "grad_norm": 1.375,
      "learning_rate": 0.0004879710745527595,
      "loss": 5.9167,
      "mean_token_accuracy": 0.16072121784090995,
      "num_tokens": 22214831.0,
      "step": 11355
    },
    {
      "entropy": 6.260396718978882,
      "epoch": 1.1355525566051883,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048795955885445696,
      "loss": 5.98,
      "mean_token_accuracy": 0.16021077930927277,
      "num_tokens": 22223772.0,
      "step": 11360
    },
    {
      "entropy": 6.087626028060913,
      "epoch": 1.1360523816664168,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004879480377981467,
      "loss": 5.8729,
      "mean_token_accuracy": 0.16185031682252884,
      "num_tokens": 22231989.0,
      "step": 11365
    },
    {
      "entropy": 6.1144266605377195,
      "epoch": 1.1365522067276452,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048793651138411866,
      "loss": 5.892,
      "mean_token_accuracy": 0.15948569923639297,
      "num_tokens": 22241018.0,
      "step": 11370
    },
    {
      "entropy": 6.07927713394165,
      "epoch": 1.137052031788874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048792497961266273,
      "loss": 5.864,
      "mean_token_accuracy": 0.15769780576229095,
      "num_tokens": 22250575.0,
      "step": 11375
    },
    {
      "entropy": 6.089027547836304,
      "epoch": 1.1375518568501024,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048791344248406916,
      "loss": 5.9749,
      "mean_token_accuracy": 0.15610423386096955,
      "num_tokens": 22259471.0,
      "step": 11380
    },
    {
      "entropy": 6.173133420944214,
      "epoch": 1.138051681911331,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048790189999862814,
      "loss": 5.9511,
      "mean_token_accuracy": 0.15375120490789412,
      "num_tokens": 22269402.0,
      "step": 11385
    },
    {
      "entropy": 6.12853684425354,
      "epoch": 1.1385515069725596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048789035215663013,
      "loss": 5.9922,
      "mean_token_accuracy": 0.152625123411417,
      "num_tokens": 22279685.0,
      "step": 11390
    },
    {
      "entropy": 6.173277091979981,
      "epoch": 1.1390513320337883,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048787879895836563,
      "loss": 5.9065,
      "mean_token_accuracy": 0.1541547805070877,
      "num_tokens": 22288628.0,
      "step": 11395
    },
    {
      "entropy": 6.183324909210205,
      "epoch": 1.1395511570950168,
      "grad_norm": 1.375,
      "learning_rate": 0.0004878672404041253,
      "loss": 5.9368,
      "mean_token_accuracy": 0.15636537671089173,
      "num_tokens": 22297762.0,
      "step": 11400
    },
    {
      "entropy": 6.148720741271973,
      "epoch": 1.1400509821562452,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048785567649419994,
      "loss": 6.1025,
      "mean_token_accuracy": 0.15237845182418824,
      "num_tokens": 22307838.0,
      "step": 11405
    },
    {
      "entropy": 6.102895736694336,
      "epoch": 1.140550807217474,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004878441072288806,
      "loss": 5.8729,
      "mean_token_accuracy": 0.1631200760602951,
      "num_tokens": 22317553.0,
      "step": 11410
    },
    {
      "entropy": 6.075419616699219,
      "epoch": 1.1410506322787024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048783253260845826,
      "loss": 5.8228,
      "mean_token_accuracy": 0.16844069510698317,
      "num_tokens": 22327892.0,
      "step": 11415
    },
    {
      "entropy": 6.117490291595459,
      "epoch": 1.141550457339931,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004878209526332242,
      "loss": 5.8866,
      "mean_token_accuracy": 0.16349421069025993,
      "num_tokens": 22338363.0,
      "step": 11420
    },
    {
      "entropy": 6.185057353973389,
      "epoch": 1.1420502824011596,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048780936730346967,
      "loss": 5.8904,
      "mean_token_accuracy": 0.14991442039608954,
      "num_tokens": 22347645.0,
      "step": 11425
    },
    {
      "entropy": 6.1305585384368895,
      "epoch": 1.1425501074623883,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048779777661948627,
      "loss": 5.9672,
      "mean_token_accuracy": 0.15605722069740297,
      "num_tokens": 22356906.0,
      "step": 11430
    },
    {
      "entropy": 6.103601360321045,
      "epoch": 1.1430499325236168,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004877861805815655,
      "loss": 5.9119,
      "mean_token_accuracy": 0.15328551307320595,
      "num_tokens": 22365806.0,
      "step": 11435
    },
    {
      "entropy": 6.052314758300781,
      "epoch": 1.1435497575848452,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004877745791899993,
      "loss": 5.8072,
      "mean_token_accuracy": 0.16704245507717133,
      "num_tokens": 22374752.0,
      "step": 11440
    },
    {
      "entropy": 6.101722383499146,
      "epoch": 1.144049582646074,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048776297244507936,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16174139827489853,
      "num_tokens": 22383280.0,
      "step": 11445
    },
    {
      "entropy": 6.13017258644104,
      "epoch": 1.1445494077073024,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048775136034709784,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15550433695316315,
      "num_tokens": 22393873.0,
      "step": 11450
    },
    {
      "entropy": 6.140684795379639,
      "epoch": 1.145049232768531,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00048773974289634676,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15891901552677154,
      "num_tokens": 22404880.0,
      "step": 11455
    },
    {
      "entropy": 6.205452346801758,
      "epoch": 1.1455490578297596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004877281200931185,
      "loss": 5.9933,
      "mean_token_accuracy": 0.14621582627296448,
      "num_tokens": 22414848.0,
      "step": 11460
    },
    {
      "entropy": 6.147256183624267,
      "epoch": 1.1460488828909883,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048771649193770554,
      "loss": 5.9517,
      "mean_token_accuracy": 0.1535073548555374,
      "num_tokens": 22425377.0,
      "step": 11465
    },
    {
      "entropy": 6.092804479598999,
      "epoch": 1.1465487079522168,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004877048584304004,
      "loss": 5.936,
      "mean_token_accuracy": 0.15238030031323432,
      "num_tokens": 22435787.0,
      "step": 11470
    },
    {
      "entropy": 6.187614583969117,
      "epoch": 1.1470485330134452,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048769321957149557,
      "loss": 6.0876,
      "mean_token_accuracy": 0.14758014678955078,
      "num_tokens": 22446540.0,
      "step": 11475
    },
    {
      "entropy": 6.142482852935791,
      "epoch": 1.147548358074674,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004876815753612843,
      "loss": 5.9196,
      "mean_token_accuracy": 0.16081787794828414,
      "num_tokens": 22456170.0,
      "step": 11480
    },
    {
      "entropy": 6.140836191177368,
      "epoch": 1.1480481831359024,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004876699258000592,
      "loss": 5.941,
      "mean_token_accuracy": 0.15535363033413888,
      "num_tokens": 22464502.0,
      "step": 11485
    },
    {
      "entropy": 6.162165784835816,
      "epoch": 1.148548008197131,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004876582708881135,
      "loss": 5.9758,
      "mean_token_accuracy": 0.1513952985405922,
      "num_tokens": 22475093.0,
      "step": 11490
    },
    {
      "entropy": 6.129208612442016,
      "epoch": 1.1490478332583596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004876466106257404,
      "loss": 6.0133,
      "mean_token_accuracy": 0.15047112554311753,
      "num_tokens": 22484229.0,
      "step": 11495
    },
    {
      "entropy": 6.228025579452515,
      "epoch": 1.149547658319588,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048763494501323333,
      "loss": 6.0213,
      "mean_token_accuracy": 0.15408311337232589,
      "num_tokens": 22493979.0,
      "step": 11500
    },
    {
      "entropy": 6.264937114715576,
      "epoch": 1.1500474833808167,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048762327405088573,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15505050495266914,
      "num_tokens": 22504675.0,
      "step": 11505
    },
    {
      "entropy": 6.139344501495361,
      "epoch": 1.1505473084420452,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004876115977389913,
      "loss": 5.916,
      "mean_token_accuracy": 0.16023336797952653,
      "num_tokens": 22514407.0,
      "step": 11510
    },
    {
      "entropy": 6.137159061431885,
      "epoch": 1.151047133503274,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048759991607784375,
      "loss": 6.0088,
      "mean_token_accuracy": 0.15007738322019576,
      "num_tokens": 22524234.0,
      "step": 11515
    },
    {
      "entropy": 6.062471151351929,
      "epoch": 1.1515469585645024,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048758822906773706,
      "loss": 5.8516,
      "mean_token_accuracy": 0.16017749160528183,
      "num_tokens": 22534004.0,
      "step": 11520
    },
    {
      "entropy": 6.122036457061768,
      "epoch": 1.1520467836257309,
      "grad_norm": 1.25,
      "learning_rate": 0.00048757653670896523,
      "loss": 6.0286,
      "mean_token_accuracy": 0.16197733730077743,
      "num_tokens": 22543926.0,
      "step": 11525
    },
    {
      "entropy": 6.159911680221557,
      "epoch": 1.1525466086869596,
      "grad_norm": 1.375,
      "learning_rate": 0.0004875648390018224,
      "loss": 5.9404,
      "mean_token_accuracy": 0.1591268926858902,
      "num_tokens": 22553671.0,
      "step": 11530
    },
    {
      "entropy": 6.102838850021362,
      "epoch": 1.153046433748188,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000487553135946603,
      "loss": 5.9032,
      "mean_token_accuracy": 0.15892914682626724,
      "num_tokens": 22562503.0,
      "step": 11535
    },
    {
      "entropy": 6.118688488006592,
      "epoch": 1.1535462588094167,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048754142754360135,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15045636743307114,
      "num_tokens": 22572555.0,
      "step": 11540
    },
    {
      "entropy": 6.206082391738891,
      "epoch": 1.1540460838706452,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004875297137931121,
      "loss": 6.0176,
      "mean_token_accuracy": 0.15411990880966187,
      "num_tokens": 22583720.0,
      "step": 11545
    },
    {
      "entropy": 6.14354944229126,
      "epoch": 1.154545908931874,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004875179946954298,
      "loss": 5.9542,
      "mean_token_accuracy": 0.15763311535120011,
      "num_tokens": 22593373.0,
      "step": 11550
    },
    {
      "entropy": 6.156551551818848,
      "epoch": 1.1550457339931024,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048750627025084955,
      "loss": 5.9091,
      "mean_token_accuracy": 0.15514979287981986,
      "num_tokens": 22604229.0,
      "step": 11555
    },
    {
      "entropy": 6.122233772277832,
      "epoch": 1.1555455590543309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004874945404596662,
      "loss": 5.8493,
      "mean_token_accuracy": 0.1641916260123253,
      "num_tokens": 22613451.0,
      "step": 11560
    },
    {
      "entropy": 6.061130714416504,
      "epoch": 1.1560453841155596,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048748280532217485,
      "loss": 6.0675,
      "mean_token_accuracy": 0.14884755611419678,
      "num_tokens": 22622809.0,
      "step": 11565
    },
    {
      "entropy": 6.1993413925170895,
      "epoch": 1.156545209176788,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004874710648386709,
      "loss": 6.0009,
      "mean_token_accuracy": 0.14978320449590682,
      "num_tokens": 22633121.0,
      "step": 11570
    },
    {
      "entropy": 6.207958650588989,
      "epoch": 1.1570450342380167,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004874593190094496,
      "loss": 5.948,
      "mean_token_accuracy": 0.15517426729202272,
      "num_tokens": 22642346.0,
      "step": 11575
    },
    {
      "entropy": 6.104215288162232,
      "epoch": 1.1575448592992452,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004874475678348065,
      "loss": 5.906,
      "mean_token_accuracy": 0.1592264786362648,
      "num_tokens": 22653157.0,
      "step": 11580
    },
    {
      "entropy": 6.088120031356811,
      "epoch": 1.158044684360474,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048743581131503716,
      "loss": 5.9522,
      "mean_token_accuracy": 0.16002236157655716,
      "num_tokens": 22663841.0,
      "step": 11585
    },
    {
      "entropy": 6.275557374954223,
      "epoch": 1.1585445094217024,
      "grad_norm": 1.6953125,
      "learning_rate": 0.0004874240494504376,
      "loss": 6.0337,
      "mean_token_accuracy": 0.1566216006875038,
      "num_tokens": 22674242.0,
      "step": 11590
    },
    {
      "entropy": 6.119044589996338,
      "epoch": 1.1590443344829309,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048741228224130357,
      "loss": 5.8634,
      "mean_token_accuracy": 0.1718853384256363,
      "num_tokens": 22683371.0,
      "step": 11595
    },
    {
      "entropy": 6.090816688537598,
      "epoch": 1.1595441595441596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048740050968793116,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15999417304992675,
      "num_tokens": 22693767.0,
      "step": 11600
    },
    {
      "entropy": 6.163295030593872,
      "epoch": 1.160043984605388,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004873887317906166,
      "loss": 5.9612,
      "mean_token_accuracy": 0.15297043398022653,
      "num_tokens": 22705108.0,
      "step": 11605
    },
    {
      "entropy": 6.195741271972656,
      "epoch": 1.1605438096666167,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048737694854965617,
      "loss": 5.8803,
      "mean_token_accuracy": 0.1559995234012604,
      "num_tokens": 22715244.0,
      "step": 11610
    },
    {
      "entropy": 6.096596097946167,
      "epoch": 1.1610436347278452,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048736515996534644,
      "loss": 5.8499,
      "mean_token_accuracy": 0.1678282305598259,
      "num_tokens": 22724519.0,
      "step": 11615
    },
    {
      "entropy": 5.9922895431518555,
      "epoch": 1.161543459789074,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004873533660379839,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16628708690404892,
      "num_tokens": 22733847.0,
      "step": 11620
    },
    {
      "entropy": 6.1042012691497805,
      "epoch": 1.1620432848503024,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048734156676786525,
      "loss": 5.8699,
      "mean_token_accuracy": 0.16311614364385604,
      "num_tokens": 22743023.0,
      "step": 11625
    },
    {
      "entropy": 6.116984176635742,
      "epoch": 1.1625431099115309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004873297621552875,
      "loss": 5.9417,
      "mean_token_accuracy": 0.15949445515871047,
      "num_tokens": 22754101.0,
      "step": 11630
    },
    {
      "entropy": 6.074648714065551,
      "epoch": 1.1630429349727596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048731795220054755,
      "loss": 5.8693,
      "mean_token_accuracy": 0.15466681718826295,
      "num_tokens": 22763815.0,
      "step": 11635
    },
    {
      "entropy": 6.090009880065918,
      "epoch": 1.163542760033988,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004873061369039425,
      "loss": 6.015,
      "mean_token_accuracy": 0.15377447679638861,
      "num_tokens": 22774041.0,
      "step": 11640
    },
    {
      "entropy": 6.168779611587524,
      "epoch": 1.1640425850952167,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004872943162657697,
      "loss": 5.985,
      "mean_token_accuracy": 0.1590366169810295,
      "num_tokens": 22784046.0,
      "step": 11645
    },
    {
      "entropy": 6.21264796257019,
      "epoch": 1.1645424101564452,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004872824902863265,
      "loss": 6.0083,
      "mean_token_accuracy": 0.1491926297545433,
      "num_tokens": 22794316.0,
      "step": 11650
    },
    {
      "entropy": 6.135050535202026,
      "epoch": 1.165042235217674,
      "grad_norm": 1.25,
      "learning_rate": 0.0004872706589659106,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16417643874883653,
      "num_tokens": 22805185.0,
      "step": 11655
    },
    {
      "entropy": 6.115337276458741,
      "epoch": 1.1655420602789024,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048725882230481947,
      "loss": 6.0034,
      "mean_token_accuracy": 0.15407361686229706,
      "num_tokens": 22815336.0,
      "step": 11660
    },
    {
      "entropy": 6.1821112632751465,
      "epoch": 1.1660418853401309,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048724698030335096,
      "loss": 6.0285,
      "mean_token_accuracy": 0.15356458798050882,
      "num_tokens": 22825530.0,
      "step": 11665
    },
    {
      "entropy": 6.120978116989136,
      "epoch": 1.1665417104013596,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004872351329618031,
      "loss": 5.8429,
      "mean_token_accuracy": 0.17565896660089492,
      "num_tokens": 22834194.0,
      "step": 11670
    },
    {
      "entropy": 6.122122478485108,
      "epoch": 1.167041535462588,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048722328028047386,
      "loss": 6.0162,
      "mean_token_accuracy": 0.15129246488213538,
      "num_tokens": 22844688.0,
      "step": 11675
    },
    {
      "entropy": 6.20037841796875,
      "epoch": 1.1675413605238167,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048721142225966146,
      "loss": 6.0174,
      "mean_token_accuracy": 0.1524915352463722,
      "num_tokens": 22855239.0,
      "step": 11680
    },
    {
      "entropy": 6.051656723022461,
      "epoch": 1.1680411855850452,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004871995588996644,
      "loss": 5.8048,
      "mean_token_accuracy": 0.1658563569188118,
      "num_tokens": 22866248.0,
      "step": 11685
    },
    {
      "entropy": 6.1115974426269535,
      "epoch": 1.168541010646274,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004871876902007809,
      "loss": 5.9498,
      "mean_token_accuracy": 0.15423931032419205,
      "num_tokens": 22876119.0,
      "step": 11690
    },
    {
      "entropy": 6.11442928314209,
      "epoch": 1.1690408357075024,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048717581616330985,
      "loss": 5.8807,
      "mean_token_accuracy": 0.162296886742115,
      "num_tokens": 22885854.0,
      "step": 11695
    },
    {
      "entropy": 6.1583396911621096,
      "epoch": 1.1695406607687309,
      "grad_norm": 1.25,
      "learning_rate": 0.00048716393678754975,
      "loss": 5.9962,
      "mean_token_accuracy": 0.15022187009453775,
      "num_tokens": 22895544.0,
      "step": 11700
    },
    {
      "entropy": 6.170709180831909,
      "epoch": 1.1700404858299596,
      "grad_norm": 1.25,
      "learning_rate": 0.0004871520520737996,
      "loss": 5.9929,
      "mean_token_accuracy": 0.1504383683204651,
      "num_tokens": 22904608.0,
      "step": 11705
    },
    {
      "entropy": 6.14715085029602,
      "epoch": 1.170540310891188,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004871401620223585,
      "loss": 5.9662,
      "mean_token_accuracy": 0.16029154062271117,
      "num_tokens": 22914716.0,
      "step": 11710
    },
    {
      "entropy": 6.212332248687744,
      "epoch": 1.1710401359524167,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004871282666335255,
      "loss": 6.0447,
      "mean_token_accuracy": 0.14545145258307457,
      "num_tokens": 22925466.0,
      "step": 11715
    },
    {
      "entropy": 6.085535621643066,
      "epoch": 1.1715399610136452,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048711636590759985,
      "loss": 5.9329,
      "mean_token_accuracy": 0.1545786067843437,
      "num_tokens": 22934499.0,
      "step": 11720
    },
    {
      "entropy": 6.114617967605591,
      "epoch": 1.172039786074874,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048710445984488106,
      "loss": 5.9263,
      "mean_token_accuracy": 0.15594572275877,
      "num_tokens": 22943822.0,
      "step": 11725
    },
    {
      "entropy": 6.056703662872314,
      "epoch": 1.1725396111361024,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004870925484456686,
      "loss": 5.9381,
      "mean_token_accuracy": 0.16310598105192184,
      "num_tokens": 22952375.0,
      "step": 11730
    },
    {
      "entropy": 6.114869117736816,
      "epoch": 1.1730394361973309,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004870806317102622,
      "loss": 5.9409,
      "mean_token_accuracy": 0.15606504082679748,
      "num_tokens": 22961945.0,
      "step": 11735
    },
    {
      "entropy": 6.223062562942505,
      "epoch": 1.1735392612585596,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004870687096389617,
      "loss": 6.0589,
      "mean_token_accuracy": 0.1501568228006363,
      "num_tokens": 22972029.0,
      "step": 11740
    },
    {
      "entropy": 6.238933134078979,
      "epoch": 1.174039086319788,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000487056782232067,
      "loss": 5.9633,
      "mean_token_accuracy": 0.15245493054389953,
      "num_tokens": 22980954.0,
      "step": 11745
    },
    {
      "entropy": 6.105295705795288,
      "epoch": 1.1745389113810167,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048704484948987824,
      "loss": 5.8813,
      "mean_token_accuracy": 0.15601636171340943,
      "num_tokens": 22991213.0,
      "step": 11750
    },
    {
      "entropy": 6.154436922073364,
      "epoch": 1.1750387364422452,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004870329114126956,
      "loss": 5.9107,
      "mean_token_accuracy": 0.15792324542999267,
      "num_tokens": 23000833.0,
      "step": 11755
    },
    {
      "entropy": 6.220010662078858,
      "epoch": 1.1755385615034737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004870209680008196,
      "loss": 5.9922,
      "mean_token_accuracy": 0.1581309825181961,
      "num_tokens": 23010598.0,
      "step": 11760
    },
    {
      "entropy": 6.1391808032989506,
      "epoch": 1.1760383865647024,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004870090192545505,
      "loss": 5.9682,
      "mean_token_accuracy": 0.15206198245286942,
      "num_tokens": 23019929.0,
      "step": 11765
    },
    {
      "entropy": 6.120934104919433,
      "epoch": 1.1765382116259309,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000486997065174189,
      "loss": 5.889,
      "mean_token_accuracy": 0.1597314417362213,
      "num_tokens": 23029880.0,
      "step": 11770
    },
    {
      "entropy": 6.171854305267334,
      "epoch": 1.1770380366871596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048698510576003593,
      "loss": 5.9937,
      "mean_token_accuracy": 0.1530468210577965,
      "num_tokens": 23039153.0,
      "step": 11775
    },
    {
      "entropy": 6.214794683456421,
      "epoch": 1.177537861748388,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048697314101239217,
      "loss": 5.9767,
      "mean_token_accuracy": 0.15725300461053848,
      "num_tokens": 23049578.0,
      "step": 11780
    },
    {
      "entropy": 6.096735048294067,
      "epoch": 1.1780376868096165,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004869611709315587,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16283528804779052,
      "num_tokens": 23058882.0,
      "step": 11785
    },
    {
      "entropy": 6.135355758666992,
      "epoch": 1.1785375118708452,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048694919551783665,
      "loss": 5.949,
      "mean_token_accuracy": 0.15908836647868158,
      "num_tokens": 23069056.0,
      "step": 11790
    },
    {
      "entropy": 6.169654655456543,
      "epoch": 1.1790373369320737,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048693721477152755,
      "loss": 5.9631,
      "mean_token_accuracy": 0.15236636251211166,
      "num_tokens": 23079155.0,
      "step": 11795
    },
    {
      "entropy": 6.295092439651489,
      "epoch": 1.1795371619933024,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048692522869293246,
      "loss": 6.0895,
      "mean_token_accuracy": 0.14466610699892044,
      "num_tokens": 23089001.0,
      "step": 11800
    },
    {
      "entropy": 6.191784381866455,
      "epoch": 1.1800369870545309,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004869132372823533,
      "loss": 5.8935,
      "mean_token_accuracy": 0.15965962707996367,
      "num_tokens": 23098824.0,
      "step": 11805
    },
    {
      "entropy": 6.135591602325439,
      "epoch": 1.1805368121157596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048690124054009157,
      "loss": 5.9113,
      "mean_token_accuracy": 0.15112822353839875,
      "num_tokens": 23107752.0,
      "step": 11810
    },
    {
      "entropy": 6.001223134994507,
      "epoch": 1.181036637176988,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004868892384664491,
      "loss": 5.8641,
      "mean_token_accuracy": 0.1568370684981346,
      "num_tokens": 23116815.0,
      "step": 11815
    },
    {
      "entropy": 6.107484626770019,
      "epoch": 1.1815364622382165,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048687723106172805,
      "loss": 5.8555,
      "mean_token_accuracy": 0.15521827191114426,
      "num_tokens": 23126093.0,
      "step": 11820
    },
    {
      "entropy": 6.193655967712402,
      "epoch": 1.1820362872994452,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048686521832623026,
      "loss": 5.9488,
      "mean_token_accuracy": 0.15967611372470855,
      "num_tokens": 23135928.0,
      "step": 11825
    },
    {
      "entropy": 6.216633462905884,
      "epoch": 1.1825361123606737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004868532002602581,
      "loss": 6.0652,
      "mean_token_accuracy": 0.1421751230955124,
      "num_tokens": 23145345.0,
      "step": 11830
    },
    {
      "entropy": 6.172040319442749,
      "epoch": 1.1830359374219024,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000486841176864114,
      "loss": 5.9601,
      "mean_token_accuracy": 0.1529302790760994,
      "num_tokens": 23155184.0,
      "step": 11835
    },
    {
      "entropy": 6.122590589523315,
      "epoch": 1.1835357624831309,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048682914813810035,
      "loss": 6.0159,
      "mean_token_accuracy": 0.14796580150723457,
      "num_tokens": 23166297.0,
      "step": 11840
    },
    {
      "entropy": 6.100281620025635,
      "epoch": 1.1840355875443596,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048681711408251986,
      "loss": 5.8906,
      "mean_token_accuracy": 0.15775033086538315,
      "num_tokens": 23175268.0,
      "step": 11845
    },
    {
      "entropy": 6.230629873275757,
      "epoch": 1.184535412605588,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048680507469767533,
      "loss": 5.9144,
      "mean_token_accuracy": 0.15430782735347748,
      "num_tokens": 23185400.0,
      "step": 11850
    },
    {
      "entropy": 6.196260166168213,
      "epoch": 1.1850352376668165,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004867930299838696,
      "loss": 5.991,
      "mean_token_accuracy": 0.15497647374868392,
      "num_tokens": 23194540.0,
      "step": 11855
    },
    {
      "entropy": 6.119010353088379,
      "epoch": 1.1855350627280452,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048678097994140566,
      "loss": 6.0719,
      "mean_token_accuracy": 0.15496022552251815,
      "num_tokens": 23205049.0,
      "step": 11860
    },
    {
      "entropy": 6.162032175064087,
      "epoch": 1.1860348877892737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048676892457058687,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15531864017248154,
      "num_tokens": 23214369.0,
      "step": 11865
    },
    {
      "entropy": 6.104197263717651,
      "epoch": 1.1865347128505024,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004867568638717164,
      "loss": 5.9034,
      "mean_token_accuracy": 0.15877461880445481,
      "num_tokens": 23223806.0,
      "step": 11870
    },
    {
      "entropy": 6.190630483627319,
      "epoch": 1.1870345379117309,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048674479784509766,
      "loss": 5.9852,
      "mean_token_accuracy": 0.15658920556306838,
      "num_tokens": 23233086.0,
      "step": 11875
    },
    {
      "entropy": 6.121031808853149,
      "epoch": 1.1875343629729596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048673272649103435,
      "loss": 5.9535,
      "mean_token_accuracy": 0.16053103655576706,
      "num_tokens": 23242564.0,
      "step": 11880
    },
    {
      "entropy": 6.103305625915527,
      "epoch": 1.188034188034188,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048672064980983013,
      "loss": 5.8572,
      "mean_token_accuracy": 0.16485465914011002,
      "num_tokens": 23252396.0,
      "step": 11885
    },
    {
      "entropy": 6.141440725326538,
      "epoch": 1.1885340130954165,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048670856780178874,
      "loss": 5.9543,
      "mean_token_accuracy": 0.15211909934878348,
      "num_tokens": 23261528.0,
      "step": 11890
    },
    {
      "entropy": 6.15968074798584,
      "epoch": 1.1890338381566452,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048669648046721436,
      "loss": 6.0258,
      "mean_token_accuracy": 0.15618912428617476,
      "num_tokens": 23271794.0,
      "step": 11895
    },
    {
      "entropy": 6.200293111801147,
      "epoch": 1.1895336632178737,
      "grad_norm": 1.21875,
      "learning_rate": 0.000486684387806411,
      "loss": 5.9522,
      "mean_token_accuracy": 0.1543474942445755,
      "num_tokens": 23281340.0,
      "step": 11900
    },
    {
      "entropy": 6.179886150360107,
      "epoch": 1.1900334882791024,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048667228981968285,
      "loss": 5.999,
      "mean_token_accuracy": 0.15246480256319045,
      "num_tokens": 23290579.0,
      "step": 11905
    },
    {
      "entropy": 6.0783895492553714,
      "epoch": 1.1905333133403309,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048666018650733446,
      "loss": 5.8899,
      "mean_token_accuracy": 0.16014678478240968,
      "num_tokens": 23300762.0,
      "step": 11910
    },
    {
      "entropy": 6.142749643325805,
      "epoch": 1.1910331384015596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004866480778696702,
      "loss": 5.9339,
      "mean_token_accuracy": 0.1566346064209938,
      "num_tokens": 23310793.0,
      "step": 11915
    },
    {
      "entropy": 6.104770803451538,
      "epoch": 1.191532963462788,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048663596390699477,
      "loss": 5.8374,
      "mean_token_accuracy": 0.1586197406053543,
      "num_tokens": 23321289.0,
      "step": 11920
    },
    {
      "entropy": 6.132608318328858,
      "epoch": 1.1920327885240165,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004866238446196129,
      "loss": 6.0197,
      "mean_token_accuracy": 0.1529877632856369,
      "num_tokens": 23331173.0,
      "step": 11925
    },
    {
      "entropy": 6.217101049423218,
      "epoch": 1.1925326135852452,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004866117200078296,
      "loss": 6.0977,
      "mean_token_accuracy": 0.14580259174108506,
      "num_tokens": 23341905.0,
      "step": 11930
    },
    {
      "entropy": 6.188102722167969,
      "epoch": 1.1930324386464737,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048659959007194985,
      "loss": 5.9436,
      "mean_token_accuracy": 0.15946682840585708,
      "num_tokens": 23351840.0,
      "step": 11935
    },
    {
      "entropy": 6.224713706970215,
      "epoch": 1.1935322637077024,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048658745481227884,
      "loss": 6.0298,
      "mean_token_accuracy": 0.14707545340061187,
      "num_tokens": 23362269.0,
      "step": 11940
    },
    {
      "entropy": 6.265580368041992,
      "epoch": 1.1940320887689309,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000486575314229122,
      "loss": 6.1131,
      "mean_token_accuracy": 0.147088935226202,
      "num_tokens": 23372431.0,
      "step": 11945
    },
    {
      "entropy": 6.171709394454956,
      "epoch": 1.1945319138301596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004865631683227846,
      "loss": 5.8595,
      "mean_token_accuracy": 0.1586290180683136,
      "num_tokens": 23381650.0,
      "step": 11950
    },
    {
      "entropy": 6.011830234527588,
      "epoch": 1.195031738891388,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048655101709357243,
      "loss": 5.8715,
      "mean_token_accuracy": 0.16672436594963075,
      "num_tokens": 23390842.0,
      "step": 11955
    },
    {
      "entropy": 6.069748640060425,
      "epoch": 1.1955315639526165,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000486538860541791,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16740934401750565,
      "num_tokens": 23400109.0,
      "step": 11960
    },
    {
      "entropy": 6.087883758544922,
      "epoch": 1.1960313890138452,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048652669866774636,
      "loss": 5.7891,
      "mean_token_accuracy": 0.1583450347185135,
      "num_tokens": 23409234.0,
      "step": 11965
    },
    {
      "entropy": 6.087847709655762,
      "epoch": 1.1965312140750737,
      "grad_norm": 1.375,
      "learning_rate": 0.0004865145314717443,
      "loss": 5.9351,
      "mean_token_accuracy": 0.1552174374461174,
      "num_tokens": 23419094.0,
      "step": 11970
    },
    {
      "entropy": 6.20829210281372,
      "epoch": 1.1970310391363024,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048650235895409115,
      "loss": 5.9188,
      "mean_token_accuracy": 0.15147635638713836,
      "num_tokens": 23428185.0,
      "step": 11975
    },
    {
      "entropy": 6.162901449203491,
      "epoch": 1.1975308641975309,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004864901811150931,
      "loss": 5.9513,
      "mean_token_accuracy": 0.1511921301484108,
      "num_tokens": 23438224.0,
      "step": 11980
    },
    {
      "entropy": 6.087011384963989,
      "epoch": 1.1980306892587596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004864779979550564,
      "loss": 5.8823,
      "mean_token_accuracy": 0.16098508536815642,
      "num_tokens": 23447982.0,
      "step": 11985
    },
    {
      "entropy": 6.109376430511475,
      "epoch": 1.198530514319988,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004864658094742878,
      "loss": 5.9308,
      "mean_token_accuracy": 0.15717553794384004,
      "num_tokens": 23458237.0,
      "step": 11990
    },
    {
      "entropy": 6.200475454330444,
      "epoch": 1.1990303393812165,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048645361567309375,
      "loss": 5.9877,
      "mean_token_accuracy": 0.15413230583071708,
      "num_tokens": 23467570.0,
      "step": 11995
    },
    {
      "entropy": 6.207857656478882,
      "epoch": 1.1995301644424452,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004864414165517812,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15324960350990297,
      "num_tokens": 23478192.0,
      "step": 12000
    },
    {
      "epoch": 1.1995301644424452,
      "eval_entropy": 6.01082161191013,
      "eval_loss": 6.058248043060303,
      "eval_mean_token_accuracy": 0.15896150581221918,
      "eval_num_tokens": 23478192.0,
      "eval_runtime": 23.9713,
      "eval_samples_per_second": 1295.088,
      "eval_steps_per_second": 161.902,
      "step": 12000
    },
    {
      "entropy": 6.180596351623535,
      "epoch": 1.2000299895036737,
      "grad_norm": 1.265625,
      "learning_rate": 0.000486429212110657,
      "loss": 6.0035,
      "mean_token_accuracy": 0.15233340337872506,
      "num_tokens": 23488128.0,
      "step": 12005
    },
    {
      "entropy": 6.189651870727539,
      "epoch": 1.2005298145649024,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004864170023500282,
      "loss": 6.0235,
      "mean_token_accuracy": 0.1500731736421585,
      "num_tokens": 23496740.0,
      "step": 12010
    },
    {
      "entropy": 6.148842668533325,
      "epoch": 1.2010296396261309,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004864047872702021,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1641639679670334,
      "num_tokens": 23505606.0,
      "step": 12015
    },
    {
      "entropy": 6.133725261688232,
      "epoch": 1.2015294646873593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004863925668714859,
      "loss": 5.8777,
      "mean_token_accuracy": 0.15371025949716569,
      "num_tokens": 23515001.0,
      "step": 12020
    },
    {
      "entropy": 6.130723857879639,
      "epoch": 1.202029289748588,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004863803411541871,
      "loss": 6.0145,
      "mean_token_accuracy": 0.14644960388541223,
      "num_tokens": 23526353.0,
      "step": 12025
    },
    {
      "entropy": 6.107809400558471,
      "epoch": 1.2025291148098165,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004863681101186134,
      "loss": 5.826,
      "mean_token_accuracy": 0.16862476766109466,
      "num_tokens": 23535571.0,
      "step": 12030
    },
    {
      "entropy": 6.0320535659790036,
      "epoch": 1.2030289398710452,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004863558737650723,
      "loss": 5.9085,
      "mean_token_accuracy": 0.16674088686704636,
      "num_tokens": 23545495.0,
      "step": 12035
    },
    {
      "entropy": 6.080349063873291,
      "epoch": 1.2035287649322737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004863436320938719,
      "loss": 5.8943,
      "mean_token_accuracy": 0.1566479340195656,
      "num_tokens": 23554942.0,
      "step": 12040
    },
    {
      "entropy": 6.183640003204346,
      "epoch": 1.2040285899935022,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048633138510532003,
      "loss": 5.9712,
      "mean_token_accuracy": 0.15449244976043702,
      "num_tokens": 23564297.0,
      "step": 12045
    },
    {
      "entropy": 6.192479801177979,
      "epoch": 1.2045284150547308,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048631913279972497,
      "loss": 5.9874,
      "mean_token_accuracy": 0.15561846047639846,
      "num_tokens": 23574899.0,
      "step": 12050
    },
    {
      "entropy": 6.196197080612182,
      "epoch": 1.2050282401159593,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048630687517739484,
      "loss": 6.0412,
      "mean_token_accuracy": 0.15124500393867493,
      "num_tokens": 23585309.0,
      "step": 12055
    },
    {
      "entropy": 6.075184011459351,
      "epoch": 1.205528065177188,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004862946122386381,
      "loss": 5.8813,
      "mean_token_accuracy": 0.16596224159002304,
      "num_tokens": 23595270.0,
      "step": 12060
    },
    {
      "entropy": 6.147265434265137,
      "epoch": 1.2060278902384165,
      "grad_norm": 1.25,
      "learning_rate": 0.00048628234398376334,
      "loss": 5.9547,
      "mean_token_accuracy": 0.15813295394182206,
      "num_tokens": 23604698.0,
      "step": 12065
    },
    {
      "entropy": 6.1474110126495365,
      "epoch": 1.2065277152996452,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048627007041307914,
      "loss": 5.9828,
      "mean_token_accuracy": 0.15180117636919022,
      "num_tokens": 23614588.0,
      "step": 12070
    },
    {
      "entropy": 6.096248197555542,
      "epoch": 1.2070275403608737,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048625779152689435,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16313449591398238,
      "num_tokens": 23624394.0,
      "step": 12075
    },
    {
      "entropy": 6.193012237548828,
      "epoch": 1.2075273654221022,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004862455073255179,
      "loss": 5.9059,
      "mean_token_accuracy": 0.16055546700954437,
      "num_tokens": 23633558.0,
      "step": 12080
    },
    {
      "entropy": 6.15625114440918,
      "epoch": 1.2080271904833308,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048623321780925883,
      "loss": 5.9439,
      "mean_token_accuracy": 0.15447635650634767,
      "num_tokens": 23643320.0,
      "step": 12085
    },
    {
      "entropy": 6.169744300842285,
      "epoch": 1.2085270155445593,
      "grad_norm": 1.25,
      "learning_rate": 0.00048622092297842637,
      "loss": 6.0468,
      "mean_token_accuracy": 0.15066591799259185,
      "num_tokens": 23653660.0,
      "step": 12090
    },
    {
      "entropy": 6.147512722015381,
      "epoch": 1.209026840605788,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004862086228333298,
      "loss": 5.9195,
      "mean_token_accuracy": 0.15709773898124696,
      "num_tokens": 23662556.0,
      "step": 12095
    },
    {
      "entropy": 6.094486427307129,
      "epoch": 1.2095266656670165,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048619631737427865,
      "loss": 5.9292,
      "mean_token_accuracy": 0.15432622730731965,
      "num_tokens": 23672628.0,
      "step": 12100
    },
    {
      "entropy": 6.069489622116089,
      "epoch": 1.2100264907282452,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004861840066015825,
      "loss": 5.828,
      "mean_token_accuracy": 0.17151496708393096,
      "num_tokens": 23682574.0,
      "step": 12105
    },
    {
      "entropy": 6.136559104919433,
      "epoch": 1.2105263157894737,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048617169051555113,
      "loss": 5.9793,
      "mean_token_accuracy": 0.1525229975581169,
      "num_tokens": 23691870.0,
      "step": 12110
    },
    {
      "entropy": 6.115087175369263,
      "epoch": 1.2110261408507021,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004861593691164942,
      "loss": 5.8617,
      "mean_token_accuracy": 0.15609296038746834,
      "num_tokens": 23701896.0,
      "step": 12115
    },
    {
      "entropy": 6.118126583099365,
      "epoch": 1.2115259659119308,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004861470424047221,
      "loss": 5.8842,
      "mean_token_accuracy": 0.15831695944070817,
      "num_tokens": 23711960.0,
      "step": 12120
    },
    {
      "entropy": 6.121831893920898,
      "epoch": 1.2120257909731593,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004861347103805446,
      "loss": 6.0079,
      "mean_token_accuracy": 0.15315590426325798,
      "num_tokens": 23721090.0,
      "step": 12125
    },
    {
      "entropy": 6.222769069671631,
      "epoch": 1.212525616034388,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004861223730442722,
      "loss": 6.1242,
      "mean_token_accuracy": 0.1528979256749153,
      "num_tokens": 23731245.0,
      "step": 12130
    },
    {
      "entropy": 6.175526189804077,
      "epoch": 1.2130254410956165,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048611003039621516,
      "loss": 5.9568,
      "mean_token_accuracy": 0.15517653077840804,
      "num_tokens": 23740891.0,
      "step": 12135
    },
    {
      "entropy": 6.180291938781738,
      "epoch": 1.2135252661568452,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004860976824366841,
      "loss": 6.0324,
      "mean_token_accuracy": 0.15124095156788825,
      "num_tokens": 23751275.0,
      "step": 12140
    },
    {
      "entropy": 6.211285066604614,
      "epoch": 1.2140250912180737,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048608532916598964,
      "loss": 6.0868,
      "mean_token_accuracy": 0.1467791751027107,
      "num_tokens": 23760994.0,
      "step": 12145
    },
    {
      "entropy": 6.1316328048706055,
      "epoch": 1.2145249162793021,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048607297058444267,
      "loss": 5.8526,
      "mean_token_accuracy": 0.15721215605735778,
      "num_tokens": 23770084.0,
      "step": 12150
    },
    {
      "entropy": 6.132530307769775,
      "epoch": 1.2150247413405308,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004860606066923541,
      "loss": 5.8972,
      "mean_token_accuracy": 0.15950650572776795,
      "num_tokens": 23780241.0,
      "step": 12155
    },
    {
      "entropy": 6.051456499099731,
      "epoch": 1.2155245664017593,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048604823749003485,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16618875712156295,
      "num_tokens": 23789970.0,
      "step": 12160
    },
    {
      "entropy": 6.193726825714111,
      "epoch": 1.216024391462988,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004860358629777963,
      "loss": 5.9912,
      "mean_token_accuracy": 0.15462590008974075,
      "num_tokens": 23799924.0,
      "step": 12165
    },
    {
      "entropy": 6.190695285797119,
      "epoch": 1.2165242165242165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004860234831559498,
      "loss": 5.8771,
      "mean_token_accuracy": 0.1680048570036888,
      "num_tokens": 23809258.0,
      "step": 12170
    },
    {
      "entropy": 5.98964409828186,
      "epoch": 1.2170240415854452,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048601109802480676,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16293641477823256,
      "num_tokens": 23817979.0,
      "step": 12175
    },
    {
      "entropy": 6.075962829589844,
      "epoch": 1.2175238666466737,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048599870758467875,
      "loss": 5.8524,
      "mean_token_accuracy": 0.16405078768730164,
      "num_tokens": 23828182.0,
      "step": 12180
    },
    {
      "entropy": 6.152235603332519,
      "epoch": 1.2180236917079021,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048598631183587753,
      "loss": 5.8631,
      "mean_token_accuracy": 0.16311971396207808,
      "num_tokens": 23838135.0,
      "step": 12185
    },
    {
      "entropy": 6.055262613296509,
      "epoch": 1.2185235167691308,
      "grad_norm": 1.25,
      "learning_rate": 0.00048597391077871504,
      "loss": 5.8312,
      "mean_token_accuracy": 0.16072398871183396,
      "num_tokens": 23847586.0,
      "step": 12190
    },
    {
      "entropy": 6.029722404479981,
      "epoch": 1.2190233418303593,
      "grad_norm": 1.375,
      "learning_rate": 0.00048596150441350324,
      "loss": 5.874,
      "mean_token_accuracy": 0.15591765493154525,
      "num_tokens": 23857112.0,
      "step": 12195
    },
    {
      "entropy": 6.1827309131622314,
      "epoch": 1.219523166891588,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004859490927405543,
      "loss": 6.0811,
      "mean_token_accuracy": 0.149028692394495,
      "num_tokens": 23866990.0,
      "step": 12200
    },
    {
      "entropy": 6.294626712799072,
      "epoch": 1.2200229919528165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004859366757601804,
      "loss": 6.1089,
      "mean_token_accuracy": 0.14310978651046752,
      "num_tokens": 23877037.0,
      "step": 12205
    },
    {
      "entropy": 6.196153974533081,
      "epoch": 1.2205228170140452,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048592425347269405,
      "loss": 6.0207,
      "mean_token_accuracy": 0.14905758202075958,
      "num_tokens": 23886341.0,
      "step": 12210
    },
    {
      "entropy": 6.211529684066773,
      "epoch": 1.2210226420752737,
      "grad_norm": 2.671875,
      "learning_rate": 0.00048591182587840777,
      "loss": 5.9744,
      "mean_token_accuracy": 0.1541582465171814,
      "num_tokens": 23896143.0,
      "step": 12215
    },
    {
      "entropy": 6.207743215560913,
      "epoch": 1.2215224671365021,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004858993929776342,
      "loss": 6.0035,
      "mean_token_accuracy": 0.14831817969679834,
      "num_tokens": 23906086.0,
      "step": 12220
    },
    {
      "entropy": 6.138950967788697,
      "epoch": 1.2220222921977308,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048588695477068617,
      "loss": 5.9068,
      "mean_token_accuracy": 0.15571671724319458,
      "num_tokens": 23914669.0,
      "step": 12225
    },
    {
      "entropy": 6.143393421173096,
      "epoch": 1.2225221172589593,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048587451125787665,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1570882812142372,
      "num_tokens": 23925461.0,
      "step": 12230
    },
    {
      "entropy": 6.105398464202881,
      "epoch": 1.223021942320188,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004858620624395187,
      "loss": 5.9591,
      "mean_token_accuracy": 0.1608840972185135,
      "num_tokens": 23934436.0,
      "step": 12235
    },
    {
      "entropy": 6.136993885040283,
      "epoch": 1.2235217673814165,
      "grad_norm": 1.375,
      "learning_rate": 0.00048584960831592547,
      "loss": 5.9848,
      "mean_token_accuracy": 0.15421614944934844,
      "num_tokens": 23944894.0,
      "step": 12240
    },
    {
      "entropy": 6.192833948135376,
      "epoch": 1.2240215924426452,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048583714888741035,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16422077715396882,
      "num_tokens": 23954621.0,
      "step": 12245
    },
    {
      "entropy": 6.130822277069091,
      "epoch": 1.2245214175038737,
      "grad_norm": 1.25,
      "learning_rate": 0.00048582468415428687,
      "loss": 5.8796,
      "mean_token_accuracy": 0.15927744135260583,
      "num_tokens": 23964379.0,
      "step": 12250
    },
    {
      "entropy": 6.086642694473267,
      "epoch": 1.2250212425651021,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048581221411686857,
      "loss": 5.8615,
      "mean_token_accuracy": 0.15981559306383133,
      "num_tokens": 23973414.0,
      "step": 12255
    },
    {
      "entropy": 6.124850797653198,
      "epoch": 1.2255210676263308,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048579973877546916,
      "loss": 5.9176,
      "mean_token_accuracy": 0.15582998245954513,
      "num_tokens": 23982912.0,
      "step": 12260
    },
    {
      "entropy": 6.2598395347595215,
      "epoch": 1.2260208926875593,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048578725813040263,
      "loss": 6.0456,
      "mean_token_accuracy": 0.14233586713671684,
      "num_tokens": 23993927.0,
      "step": 12265
    },
    {
      "entropy": 6.157826328277588,
      "epoch": 1.226520717748788,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004857747721819829,
      "loss": 5.86,
      "mean_token_accuracy": 0.15139659792184829,
      "num_tokens": 24003518.0,
      "step": 12270
    },
    {
      "entropy": 6.101655530929565,
      "epoch": 1.2270205428100165,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004857622809305241,
      "loss": 5.9393,
      "mean_token_accuracy": 0.1609980970621109,
      "num_tokens": 24013013.0,
      "step": 12275
    },
    {
      "entropy": 6.208060455322266,
      "epoch": 1.2275203678712452,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048574978437634057,
      "loss": 6.045,
      "mean_token_accuracy": 0.15419430881738663,
      "num_tokens": 24022276.0,
      "step": 12280
    },
    {
      "entropy": 6.083693027496338,
      "epoch": 1.2280201929324737,
      "grad_norm": 1.375,
      "learning_rate": 0.0004857372825197468,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16444548666477204,
      "num_tokens": 24032165.0,
      "step": 12285
    },
    {
      "entropy": 6.172849512100219,
      "epoch": 1.2285200179937021,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004857247753610571,
      "loss": 5.9153,
      "mean_token_accuracy": 0.16230880171060563,
      "num_tokens": 24041313.0,
      "step": 12290
    },
    {
      "entropy": 6.186947441101074,
      "epoch": 1.2290198430549308,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048571226290058635,
      "loss": 5.9763,
      "mean_token_accuracy": 0.15651389583945274,
      "num_tokens": 24050810.0,
      "step": 12295
    },
    {
      "entropy": 6.11482424736023,
      "epoch": 1.2295196681161593,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004856997451386492,
      "loss": 5.9022,
      "mean_token_accuracy": 0.16044087558984757,
      "num_tokens": 24060582.0,
      "step": 12300
    },
    {
      "entropy": 6.122136735916138,
      "epoch": 1.2300194931773878,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004856872220755607,
      "loss": 6.0002,
      "mean_token_accuracy": 0.15123990550637245,
      "num_tokens": 24070208.0,
      "step": 12305
    },
    {
      "entropy": 6.094393682479859,
      "epoch": 1.2305193182386165,
      "grad_norm": 1.5546875,
      "learning_rate": 0.000485674693711636,
      "loss": 5.9307,
      "mean_token_accuracy": 0.15833743363618852,
      "num_tokens": 24080837.0,
      "step": 12310
    },
    {
      "entropy": 6.147878932952881,
      "epoch": 1.231019143299845,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004856621600471901,
      "loss": 5.9429,
      "mean_token_accuracy": 0.15806545615196227,
      "num_tokens": 24090696.0,
      "step": 12315
    },
    {
      "entropy": 6.222177600860595,
      "epoch": 1.2315189683610737,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048564962108253853,
      "loss": 6.0146,
      "mean_token_accuracy": 0.1541634276509285,
      "num_tokens": 24100957.0,
      "step": 12320
    },
    {
      "entropy": 6.209986829757691,
      "epoch": 1.2320187934223021,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004856370768179967,
      "loss": 5.9239,
      "mean_token_accuracy": 0.1646161898970604,
      "num_tokens": 24109177.0,
      "step": 12325
    },
    {
      "entropy": 6.069714307785034,
      "epoch": 1.2325186184835308,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048562452725388024,
      "loss": 5.9536,
      "mean_token_accuracy": 0.1579720839858055,
      "num_tokens": 24118918.0,
      "step": 12330
    },
    {
      "entropy": 6.093706178665161,
      "epoch": 1.2330184435447593,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004856119723905048,
      "loss": 5.8669,
      "mean_token_accuracy": 0.1543697491288185,
      "num_tokens": 24128255.0,
      "step": 12335
    },
    {
      "entropy": 6.161298847198486,
      "epoch": 1.2335182686059878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004855994122281864,
      "loss": 5.9174,
      "mean_token_accuracy": 0.15708560347557068,
      "num_tokens": 24137670.0,
      "step": 12340
    },
    {
      "entropy": 6.100286388397217,
      "epoch": 1.2340180936672165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048558684676724083,
      "loss": 5.9217,
      "mean_token_accuracy": 0.16080491244792938,
      "num_tokens": 24147417.0,
      "step": 12345
    },
    {
      "entropy": 6.140875816345215,
      "epoch": 1.234517918728445,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048557427600798454,
      "loss": 5.9525,
      "mean_token_accuracy": 0.15836237818002702,
      "num_tokens": 24158019.0,
      "step": 12350
    },
    {
      "entropy": 6.199957752227784,
      "epoch": 1.2350177437896737,
      "grad_norm": 1.375,
      "learning_rate": 0.0004855616999507336,
      "loss": 5.9676,
      "mean_token_accuracy": 0.15126212164759636,
      "num_tokens": 24168118.0,
      "step": 12355
    },
    {
      "entropy": 6.18003625869751,
      "epoch": 1.2355175688509021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048554911859580445,
      "loss": 6.0224,
      "mean_token_accuracy": 0.1502956509590149,
      "num_tokens": 24177700.0,
      "step": 12360
    },
    {
      "entropy": 6.14128680229187,
      "epoch": 1.2360173939121308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004855365319435137,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1638193815946579,
      "num_tokens": 24187701.0,
      "step": 12365
    },
    {
      "entropy": 6.191411018371582,
      "epoch": 1.2365172189733593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000485523939994178,
      "loss": 5.907,
      "mean_token_accuracy": 0.1560323044657707,
      "num_tokens": 24197364.0,
      "step": 12370
    },
    {
      "entropy": 6.156819438934326,
      "epoch": 1.2370170440345878,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004855113427481141,
      "loss": 5.9939,
      "mean_token_accuracy": 0.1583786629140377,
      "num_tokens": 24208015.0,
      "step": 12375
    },
    {
      "entropy": 6.124359130859375,
      "epoch": 1.2375168690958165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000485498740205639,
      "loss": 5.9548,
      "mean_token_accuracy": 0.15511894822120667,
      "num_tokens": 24218576.0,
      "step": 12380
    },
    {
      "entropy": 6.1889136791229244,
      "epoch": 1.238016694157045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048548613236706976,
      "loss": 5.9856,
      "mean_token_accuracy": 0.15507199689745904,
      "num_tokens": 24228727.0,
      "step": 12385
    },
    {
      "entropy": 6.123276090621948,
      "epoch": 1.2385165192182737,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004854735192327236,
      "loss": 5.9339,
      "mean_token_accuracy": 0.16070912629365922,
      "num_tokens": 24238580.0,
      "step": 12390
    },
    {
      "entropy": 6.154482650756836,
      "epoch": 1.2390163442795021,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048546090080291783,
      "loss": 5.8971,
      "mean_token_accuracy": 0.15372254997491835,
      "num_tokens": 24249543.0,
      "step": 12395
    },
    {
      "entropy": 6.129493141174317,
      "epoch": 1.2395161693407308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048544827707797,
      "loss": 5.9183,
      "mean_token_accuracy": 0.15647836923599243,
      "num_tokens": 24259346.0,
      "step": 12400
    },
    {
      "entropy": 6.142840242385864,
      "epoch": 1.2400159944019593,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048543564805819764,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16449064761400223,
      "num_tokens": 24267987.0,
      "step": 12405
    },
    {
      "entropy": 6.0619124412536625,
      "epoch": 1.2405158194631878,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004854230137439185,
      "loss": 5.8191,
      "mean_token_accuracy": 0.1636724963784218,
      "num_tokens": 24278398.0,
      "step": 12410
    },
    {
      "entropy": 6.239473819732666,
      "epoch": 1.2410156445244165,
      "grad_norm": 2.375,
      "learning_rate": 0.00048541037413545046,
      "loss": 6.161,
      "mean_token_accuracy": 0.145528407394886,
      "num_tokens": 24289425.0,
      "step": 12415
    },
    {
      "entropy": 6.213787651062011,
      "epoch": 1.241515469585645,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004853977292331116,
      "loss": 5.9837,
      "mean_token_accuracy": 0.1525195226073265,
      "num_tokens": 24298866.0,
      "step": 12420
    },
    {
      "entropy": 6.213447713851929,
      "epoch": 1.2420152946468737,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004853850790372201,
      "loss": 5.8973,
      "mean_token_accuracy": 0.16932913511991501,
      "num_tokens": 24309237.0,
      "step": 12425
    },
    {
      "entropy": 6.1542774677276615,
      "epoch": 1.2425151197081021,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048537242354809403,
      "loss": 5.8882,
      "mean_token_accuracy": 0.15374805629253388,
      "num_tokens": 24320175.0,
      "step": 12430
    },
    {
      "entropy": 6.057026386260986,
      "epoch": 1.2430149447693308,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048535976276605186,
      "loss": 5.9255,
      "mean_token_accuracy": 0.15403307527303695,
      "num_tokens": 24328896.0,
      "step": 12435
    },
    {
      "entropy": 6.160553455352783,
      "epoch": 1.2435147698305593,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004853470966914123,
      "loss": 5.9715,
      "mean_token_accuracy": 0.15797725170850754,
      "num_tokens": 24337988.0,
      "step": 12440
    },
    {
      "entropy": 6.190245294570923,
      "epoch": 1.2440145948917878,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004853344253244939,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16221871078014374,
      "num_tokens": 24347154.0,
      "step": 12445
    },
    {
      "entropy": 6.186558723449707,
      "epoch": 1.2445144199530165,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00048532174866561543,
      "loss": 5.9944,
      "mean_token_accuracy": 0.14735816270112992,
      "num_tokens": 24357564.0,
      "step": 12450
    },
    {
      "entropy": 6.099512195587158,
      "epoch": 1.245014245014245,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048530906671509597,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16346918046474457,
      "num_tokens": 24366413.0,
      "step": 12455
    },
    {
      "entropy": 6.101252603530884,
      "epoch": 1.2455140700754737,
      "grad_norm": 1.375,
      "learning_rate": 0.00048529637947325444,
      "loss": 5.8674,
      "mean_token_accuracy": 0.1621327057480812,
      "num_tokens": 24374950.0,
      "step": 12460
    },
    {
      "entropy": 5.994110488891602,
      "epoch": 1.2460138951367021,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004852836869404101,
      "loss": 5.6753,
      "mean_token_accuracy": 0.17131358832120896,
      "num_tokens": 24384354.0,
      "step": 12465
    },
    {
      "entropy": 6.151537370681763,
      "epoch": 1.2465137201979308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004852709891168823,
      "loss": 5.8982,
      "mean_token_accuracy": 0.1560425043106079,
      "num_tokens": 24395671.0,
      "step": 12470
    },
    {
      "entropy": 6.185190725326538,
      "epoch": 1.2470135452591593,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004852582860029906,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15676460564136505,
      "num_tokens": 24404497.0,
      "step": 12475
    },
    {
      "entropy": 6.046434450149536,
      "epoch": 1.2475133703203878,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048524557759905445,
      "loss": 5.9049,
      "mean_token_accuracy": 0.15762938112020491,
      "num_tokens": 24413651.0,
      "step": 12480
    },
    {
      "entropy": 6.103031301498413,
      "epoch": 1.2480131953816165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004852328639053937,
      "loss": 5.952,
      "mean_token_accuracy": 0.15817653387784958,
      "num_tokens": 24424037.0,
      "step": 12485
    },
    {
      "entropy": 6.17068452835083,
      "epoch": 1.248513020442845,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004852201449223282,
      "loss": 5.8577,
      "mean_token_accuracy": 0.16303713023662567,
      "num_tokens": 24433005.0,
      "step": 12490
    },
    {
      "entropy": 6.12118558883667,
      "epoch": 1.2490128455040737,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000485207420650178,
      "loss": 6.019,
      "mean_token_accuracy": 0.14964741021394729,
      "num_tokens": 24443241.0,
      "step": 12495
    },
    {
      "entropy": 6.135250854492187,
      "epoch": 1.2495126705653021,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004851946910892631,
      "loss": 5.8735,
      "mean_token_accuracy": 0.15894473642110823,
      "num_tokens": 24452676.0,
      "step": 12500
    },
    {
      "entropy": 6.1575624465942385,
      "epoch": 1.2500124956265308,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004851819562399039,
      "loss": 5.9619,
      "mean_token_accuracy": 0.1505800910294056,
      "num_tokens": 24462287.0,
      "step": 12505
    },
    {
      "entropy": 6.210162591934204,
      "epoch": 1.2505123206877593,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004851692161024207,
      "loss": 6.0264,
      "mean_token_accuracy": 0.1506255954504013,
      "num_tokens": 24471707.0,
      "step": 12510
    },
    {
      "entropy": 6.161726427078247,
      "epoch": 1.2510121457489878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048515647067713425,
      "loss": 5.8923,
      "mean_token_accuracy": 0.15752028673887253,
      "num_tokens": 24481920.0,
      "step": 12515
    },
    {
      "entropy": 6.109935998916626,
      "epoch": 1.2515119708102165,
      "grad_norm": 1.25,
      "learning_rate": 0.000485143719964365,
      "loss": 5.8302,
      "mean_token_accuracy": 0.16883427351713182,
      "num_tokens": 24491890.0,
      "step": 12520
    },
    {
      "entropy": 6.0941685199737545,
      "epoch": 1.252011795871445,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004851309639644338,
      "loss": 5.9176,
      "mean_token_accuracy": 0.15839326232671738,
      "num_tokens": 24501563.0,
      "step": 12525
    },
    {
      "entropy": 6.0821606636047365,
      "epoch": 1.2525116209326734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004851182026776616,
      "loss": 5.8635,
      "mean_token_accuracy": 0.16440026462078094,
      "num_tokens": 24511828.0,
      "step": 12530
    },
    {
      "entropy": 6.174101877212524,
      "epoch": 1.2530114459939021,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004851054361043696,
      "loss": 5.9122,
      "mean_token_accuracy": 0.1581023119390011,
      "num_tokens": 24521695.0,
      "step": 12535
    },
    {
      "entropy": 6.185136127471924,
      "epoch": 1.2535112710551308,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048509266424487873,
      "loss": 5.9828,
      "mean_token_accuracy": 0.14992508962750434,
      "num_tokens": 24531755.0,
      "step": 12540
    },
    {
      "entropy": 6.12101788520813,
      "epoch": 1.2540110961163593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004850798870995106,
      "loss": 5.8855,
      "mean_token_accuracy": 0.16575640738010405,
      "num_tokens": 24540221.0,
      "step": 12545
    },
    {
      "entropy": 6.235535383224487,
      "epoch": 1.2545109211775878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048506710466858657,
      "loss": 5.9234,
      "mean_token_accuracy": 0.15413894057273864,
      "num_tokens": 24549749.0,
      "step": 12550
    },
    {
      "entropy": 6.16100697517395,
      "epoch": 1.2550107462388165,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048505431695242823,
      "loss": 5.9481,
      "mean_token_accuracy": 0.15367883145809175,
      "num_tokens": 24559318.0,
      "step": 12555
    },
    {
      "entropy": 6.12654480934143,
      "epoch": 1.255510571300045,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048504152395135726,
      "loss": 5.9784,
      "mean_token_accuracy": 0.15792434215545653,
      "num_tokens": 24568779.0,
      "step": 12560
    },
    {
      "entropy": 6.125205421447754,
      "epoch": 1.2560103963612734,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048502872566569564,
      "loss": 5.9503,
      "mean_token_accuracy": 0.15098431557416916,
      "num_tokens": 24578258.0,
      "step": 12565
    },
    {
      "entropy": 6.140887880325318,
      "epoch": 1.2565102214225021,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048501592209576535,
      "loss": 5.9535,
      "mean_token_accuracy": 0.1561593919992447,
      "num_tokens": 24587485.0,
      "step": 12570
    },
    {
      "entropy": 6.113513469696045,
      "epoch": 1.2570100464837308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004850031132418885,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15562413185834884,
      "num_tokens": 24597656.0,
      "step": 12575
    },
    {
      "entropy": 6.114263391494751,
      "epoch": 1.2575098715449593,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048499029910438734,
      "loss": 5.8421,
      "mean_token_accuracy": 0.1591611161828041,
      "num_tokens": 24606986.0,
      "step": 12580
    },
    {
      "entropy": 6.1446521282196045,
      "epoch": 1.2580096966061878,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048497747968358424,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1605367347598076,
      "num_tokens": 24617091.0,
      "step": 12585
    },
    {
      "entropy": 6.11664400100708,
      "epoch": 1.2585095216674165,
      "grad_norm": 1.375,
      "learning_rate": 0.0004849646549798019,
      "loss": 5.9226,
      "mean_token_accuracy": 0.1554800972342491,
      "num_tokens": 24627139.0,
      "step": 12590
    },
    {
      "entropy": 6.243822145462036,
      "epoch": 1.259009346728645,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048495182499336275,
      "loss": 5.9874,
      "mean_token_accuracy": 0.15350937992334365,
      "num_tokens": 24635632.0,
      "step": 12595
    },
    {
      "entropy": 6.167522239685058,
      "epoch": 1.2595091717898734,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048493898972458985,
      "loss": 5.9353,
      "mean_token_accuracy": 0.15566777735948562,
      "num_tokens": 24646590.0,
      "step": 12600
    },
    {
      "entropy": 6.155635023117066,
      "epoch": 1.2600089968511021,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004849261491738059,
      "loss": 6.0287,
      "mean_token_accuracy": 0.14981914386153222,
      "num_tokens": 24656865.0,
      "step": 12605
    },
    {
      "entropy": 6.172837114334106,
      "epoch": 1.2605088219123306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048491330334133405,
      "loss": 6.0184,
      "mean_token_accuracy": 0.1499500185251236,
      "num_tokens": 24666438.0,
      "step": 12610
    },
    {
      "entropy": 6.207957029342651,
      "epoch": 1.2610086469735593,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048490045222749744,
      "loss": 5.9622,
      "mean_token_accuracy": 0.15867784321308137,
      "num_tokens": 24675935.0,
      "step": 12615
    },
    {
      "entropy": 6.132725524902344,
      "epoch": 1.2615084720347878,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048488759583261955,
      "loss": 5.8625,
      "mean_token_accuracy": 0.16611118763685226,
      "num_tokens": 24686165.0,
      "step": 12620
    },
    {
      "entropy": 6.202854347229004,
      "epoch": 1.2620082970960165,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004848747341570238,
      "loss": 5.9444,
      "mean_token_accuracy": 0.1541835680603981,
      "num_tokens": 24696564.0,
      "step": 12625
    },
    {
      "entropy": 6.118795919418335,
      "epoch": 1.262508122157245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048486186720103364,
      "loss": 5.8939,
      "mean_token_accuracy": 0.158332984149456,
      "num_tokens": 24706462.0,
      "step": 12630
    },
    {
      "entropy": 6.100149822235108,
      "epoch": 1.2630079472184734,
      "grad_norm": 1.390625,
      "learning_rate": 0.000484848994964973,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15603938922286034,
      "num_tokens": 24715153.0,
      "step": 12635
    },
    {
      "entropy": 6.137001323699951,
      "epoch": 1.2635077722797021,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048483611744916556,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15871885269880295,
      "num_tokens": 24725215.0,
      "step": 12640
    },
    {
      "entropy": 6.083922386169434,
      "epoch": 1.2640075973409306,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048482323465393545,
      "loss": 5.9104,
      "mean_token_accuracy": 0.16072021871805192,
      "num_tokens": 24735081.0,
      "step": 12645
    },
    {
      "entropy": 6.093176412582397,
      "epoch": 1.2645074224021593,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004848103465796067,
      "loss": 5.874,
      "mean_token_accuracy": 0.1625130593776703,
      "num_tokens": 24744948.0,
      "step": 12650
    },
    {
      "entropy": 6.075563573837281,
      "epoch": 1.2650072474633878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004847974532265037,
      "loss": 5.9113,
      "mean_token_accuracy": 0.15873206555843353,
      "num_tokens": 24754123.0,
      "step": 12655
    },
    {
      "entropy": 6.1799976348876955,
      "epoch": 1.2655070725246165,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004847845545949507,
      "loss": 5.9963,
      "mean_token_accuracy": 0.14864538460969925,
      "num_tokens": 24763864.0,
      "step": 12660
    },
    {
      "entropy": 6.189121866226197,
      "epoch": 1.266006897585845,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048477165068527237,
      "loss": 5.9765,
      "mean_token_accuracy": 0.15525337755680085,
      "num_tokens": 24774088.0,
      "step": 12665
    },
    {
      "entropy": 6.114309215545655,
      "epoch": 1.2665067226470734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048475874149779313,
      "loss": 5.9168,
      "mean_token_accuracy": 0.15420229285955428,
      "num_tokens": 24782485.0,
      "step": 12670
    },
    {
      "entropy": 6.160350656509399,
      "epoch": 1.2670065477083021,
      "grad_norm": 1.3125,
      "learning_rate": 0.000484745827032838,
      "loss": 5.9308,
      "mean_token_accuracy": 0.1574045166373253,
      "num_tokens": 24791776.0,
      "step": 12675
    },
    {
      "entropy": 6.208744382858276,
      "epoch": 1.2675063727695306,
      "grad_norm": 1.25,
      "learning_rate": 0.00048473290729073194,
      "loss": 5.9629,
      "mean_token_accuracy": 0.1527264580130577,
      "num_tokens": 24802053.0,
      "step": 12680
    },
    {
      "entropy": 6.131521272659302,
      "epoch": 1.2680061978307593,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004847199822717998,
      "loss": 5.897,
      "mean_token_accuracy": 0.15579057410359382,
      "num_tokens": 24811235.0,
      "step": 12685
    },
    {
      "entropy": 6.1487141132354735,
      "epoch": 1.2685060228919878,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004847070519763668,
      "loss": 6.0087,
      "mean_token_accuracy": 0.14812286347150802,
      "num_tokens": 24821904.0,
      "step": 12690
    },
    {
      "entropy": 6.145418643951416,
      "epoch": 1.2690058479532165,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004846941164047585,
      "loss": 5.8991,
      "mean_token_accuracy": 0.15500079989433288,
      "num_tokens": 24831555.0,
      "step": 12695
    },
    {
      "entropy": 6.245338392257691,
      "epoch": 1.269505673014445,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048468117555730013,
      "loss": 6.0121,
      "mean_token_accuracy": 0.15079040229320526,
      "num_tokens": 24841840.0,
      "step": 12700
    },
    {
      "entropy": 6.083285713195801,
      "epoch": 1.2700054980756734,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004846682294343173,
      "loss": 5.9015,
      "mean_token_accuracy": 0.15662896260619164,
      "num_tokens": 24850172.0,
      "step": 12705
    },
    {
      "entropy": 6.164756393432617,
      "epoch": 1.2705053231369021,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004846552780361358,
      "loss": 5.9018,
      "mean_token_accuracy": 0.1535533994436264,
      "num_tokens": 24860433.0,
      "step": 12710
    },
    {
      "entropy": 6.169408226013184,
      "epoch": 1.2710051481981306,
      "grad_norm": 1.25,
      "learning_rate": 0.00048464232136308146,
      "loss": 5.9668,
      "mean_token_accuracy": 0.15852106735110283,
      "num_tokens": 24870677.0,
      "step": 12715
    },
    {
      "entropy": 6.20074577331543,
      "epoch": 1.2715049732593593,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004846293594154802,
      "loss": 5.9651,
      "mean_token_accuracy": 0.15370364487171173,
      "num_tokens": 24881804.0,
      "step": 12720
    },
    {
      "entropy": 6.154753684997559,
      "epoch": 1.2720047983205878,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004846163921936583,
      "loss": 5.9081,
      "mean_token_accuracy": 0.16247408241033554,
      "num_tokens": 24892632.0,
      "step": 12725
    },
    {
      "entropy": 6.149202823638916,
      "epoch": 1.2725046233818165,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048460341969794183,
      "loss": 5.9399,
      "mean_token_accuracy": 0.1517065055668354,
      "num_tokens": 24902055.0,
      "step": 12730
    },
    {
      "entropy": 6.1957234859466555,
      "epoch": 1.273004448443045,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048459044192865726,
      "loss": 6.0565,
      "mean_token_accuracy": 0.14943547323346137,
      "num_tokens": 24912450.0,
      "step": 12735
    },
    {
      "entropy": 6.173855018615723,
      "epoch": 1.2735042735042734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004845774588861311,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15688476115465164,
      "num_tokens": 24922206.0,
      "step": 12740
    },
    {
      "entropy": 6.104503345489502,
      "epoch": 1.2740040985655021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048456447057069004,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1675427883863449,
      "num_tokens": 24931295.0,
      "step": 12745
    },
    {
      "entropy": 6.0886742115020756,
      "epoch": 1.2745039236267306,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048455147698266084,
      "loss": 5.8665,
      "mean_token_accuracy": 0.16386393308639527,
      "num_tokens": 24940979.0,
      "step": 12750
    },
    {
      "entropy": 6.063467884063721,
      "epoch": 1.2750037486879593,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004845384781223703,
      "loss": 5.9297,
      "mean_token_accuracy": 0.1529921531677246,
      "num_tokens": 24950704.0,
      "step": 12755
    },
    {
      "entropy": 6.177727365493775,
      "epoch": 1.2755035737491878,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048452547399014566,
      "loss": 6.0075,
      "mean_token_accuracy": 0.15610215961933135,
      "num_tokens": 24961667.0,
      "step": 12760
    },
    {
      "entropy": 6.145414018630982,
      "epoch": 1.2760033988104165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000484512464586314,
      "loss": 5.9484,
      "mean_token_accuracy": 0.15527768731117247,
      "num_tokens": 24971531.0,
      "step": 12765
    },
    {
      "entropy": 6.098261785507202,
      "epoch": 1.276503223871645,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048449944991120264,
      "loss": 5.8291,
      "mean_token_accuracy": 0.15930492728948592,
      "num_tokens": 24981002.0,
      "step": 12770
    },
    {
      "entropy": 6.116815662384033,
      "epoch": 1.2770030489328734,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004844864299651391,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15677466243505478,
      "num_tokens": 24990402.0,
      "step": 12775
    },
    {
      "entropy": 6.056716394424439,
      "epoch": 1.2775028739941021,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048447340474845076,
      "loss": 5.8794,
      "mean_token_accuracy": 0.1652596116065979,
      "num_tokens": 24999399.0,
      "step": 12780
    },
    {
      "entropy": 5.9514140605926515,
      "epoch": 1.2780026990553306,
      "grad_norm": 1.5,
      "learning_rate": 0.00048446037426146555,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17866737842559816,
      "num_tokens": 25010411.0,
      "step": 12785
    },
    {
      "entropy": 6.189856004714966,
      "epoch": 1.278502524116559,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004844473385045112,
      "loss": 6.0076,
      "mean_token_accuracy": 0.15052021890878678,
      "num_tokens": 25019834.0,
      "step": 12790
    },
    {
      "entropy": 6.189896488189698,
      "epoch": 1.2790023491777878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004844342974779157,
      "loss": 6.0048,
      "mean_token_accuracy": 0.15291009247303008,
      "num_tokens": 25030331.0,
      "step": 12795
    },
    {
      "entropy": 6.172100925445557,
      "epoch": 1.2795021742390165,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048442125118200707,
      "loss": 5.904,
      "mean_token_accuracy": 0.15975322127342223,
      "num_tokens": 25040369.0,
      "step": 12800
    },
    {
      "entropy": 6.085309171676636,
      "epoch": 1.280001999300245,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004844081996171137,
      "loss": 5.8833,
      "mean_token_accuracy": 0.16446298658847808,
      "num_tokens": 25049798.0,
      "step": 12805
    },
    {
      "entropy": 6.19209303855896,
      "epoch": 1.2805018243614734,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000484395142783564,
      "loss": 5.937,
      "mean_token_accuracy": 0.1537824258208275,
      "num_tokens": 25060982.0,
      "step": 12810
    },
    {
      "entropy": 6.15929946899414,
      "epoch": 1.2810016494227021,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004843820806816863,
      "loss": 6.0359,
      "mean_token_accuracy": 0.1448635198175907,
      "num_tokens": 25070532.0,
      "step": 12815
    },
    {
      "entropy": 6.156884479522705,
      "epoch": 1.2815014744839306,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004843690133118094,
      "loss": 5.9313,
      "mean_token_accuracy": 0.1576389729976654,
      "num_tokens": 25080146.0,
      "step": 12820
    },
    {
      "entropy": 6.094932222366333,
      "epoch": 1.282001299545159,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000484355940674262,
      "loss": 5.9122,
      "mean_token_accuracy": 0.15721622854471207,
      "num_tokens": 25090758.0,
      "step": 12825
    },
    {
      "entropy": 6.067590570449829,
      "epoch": 1.2825011246063878,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048434286276937294,
      "loss": 5.9521,
      "mean_token_accuracy": 0.16110042929649354,
      "num_tokens": 25100735.0,
      "step": 12830
    },
    {
      "entropy": 6.131159496307373,
      "epoch": 1.2830009496676165,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004843297795974713,
      "loss": 5.9081,
      "mean_token_accuracy": 0.1634131044149399,
      "num_tokens": 25110996.0,
      "step": 12835
    },
    {
      "entropy": 6.23368616104126,
      "epoch": 1.283500774728845,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004843166911588864,
      "loss": 5.8672,
      "mean_token_accuracy": 0.15603855922818183,
      "num_tokens": 25119720.0,
      "step": 12840
    },
    {
      "entropy": 6.108460092544556,
      "epoch": 1.2840005997900734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004843035974539473,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15453407168388367,
      "num_tokens": 25129916.0,
      "step": 12845
    },
    {
      "entropy": 6.186034059524536,
      "epoch": 1.2845004248513021,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048429049848298364,
      "loss": 5.9888,
      "mean_token_accuracy": 0.1550639882683754,
      "num_tokens": 25138647.0,
      "step": 12850
    },
    {
      "entropy": 6.181397867202759,
      "epoch": 1.2850002499125306,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048427739424632485,
      "loss": 5.8569,
      "mean_token_accuracy": 0.1590894378721714,
      "num_tokens": 25146621.0,
      "step": 12855
    },
    {
      "entropy": 6.114289426803589,
      "epoch": 1.285500074973759,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048426428474430055,
      "loss": 5.9945,
      "mean_token_accuracy": 0.1516559213399887,
      "num_tokens": 25158600.0,
      "step": 12860
    },
    {
      "entropy": 6.123188304901123,
      "epoch": 1.2859999000349878,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048425116997724083,
      "loss": 5.9069,
      "mean_token_accuracy": 0.15513090342283248,
      "num_tokens": 25169051.0,
      "step": 12865
    },
    {
      "entropy": 6.1581357479095455,
      "epoch": 1.2864997250962165,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048423804994547555,
      "loss": 5.8216,
      "mean_token_accuracy": 0.1701357841491699,
      "num_tokens": 25177866.0,
      "step": 12870
    },
    {
      "entropy": 6.134276056289673,
      "epoch": 1.286999550157445,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048422492464933464,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15638816207647324,
      "num_tokens": 25186868.0,
      "step": 12875
    },
    {
      "entropy": 6.100353097915649,
      "epoch": 1.2874993752186734,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048421179408914857,
      "loss": 6.0121,
      "mean_token_accuracy": 0.15230512320995332,
      "num_tokens": 25196682.0,
      "step": 12880
    },
    {
      "entropy": 6.183548450469971,
      "epoch": 1.2879992002799021,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048419865826524757,
      "loss": 5.9934,
      "mean_token_accuracy": 0.15629564523696898,
      "num_tokens": 25206326.0,
      "step": 12885
    },
    {
      "entropy": 6.175526332855225,
      "epoch": 1.2884990253411306,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048418551717796217,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15119197964668274,
      "num_tokens": 25216499.0,
      "step": 12890
    },
    {
      "entropy": 6.121239233016968,
      "epoch": 1.288998850402359,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048417237082762295,
      "loss": 5.8756,
      "mean_token_accuracy": 0.16114613711833953,
      "num_tokens": 25226854.0,
      "step": 12895
    },
    {
      "entropy": 6.250212144851685,
      "epoch": 1.2894986754635878,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048415921921456074,
      "loss": 6.0999,
      "mean_token_accuracy": 0.14770000874996186,
      "num_tokens": 25237536.0,
      "step": 12900
    },
    {
      "entropy": 6.1297276496887205,
      "epoch": 1.2899985005248162,
      "grad_norm": 1.734375,
      "learning_rate": 0.00048414606233910634,
      "loss": 5.8592,
      "mean_token_accuracy": 0.1612721025943756,
      "num_tokens": 25246536.0,
      "step": 12905
    },
    {
      "entropy": 6.09002914428711,
      "epoch": 1.290498325586045,
      "grad_norm": 1.375,
      "learning_rate": 0.0004841329002015909,
      "loss": 5.9543,
      "mean_token_accuracy": 0.15666455179452896,
      "num_tokens": 25256305.0,
      "step": 12910
    },
    {
      "entropy": 6.122639942169189,
      "epoch": 1.2909981506472734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048411973280234553,
      "loss": 5.8685,
      "mean_token_accuracy": 0.15847896486520768,
      "num_tokens": 25265691.0,
      "step": 12915
    },
    {
      "entropy": 6.2116837978363035,
      "epoch": 1.291497975708502,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048410656014170135,
      "loss": 5.9948,
      "mean_token_accuracy": 0.15522881597280502,
      "num_tokens": 25275963.0,
      "step": 12920
    },
    {
      "entropy": 6.1699652671813965,
      "epoch": 1.2919978007697306,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048409338221999,
      "loss": 5.9251,
      "mean_token_accuracy": 0.15512767136096955,
      "num_tokens": 25286646.0,
      "step": 12925
    },
    {
      "entropy": 6.089895820617675,
      "epoch": 1.292497625830959,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048408019903754297,
      "loss": 5.9619,
      "mean_token_accuracy": 0.15888506472110747,
      "num_tokens": 25296144.0,
      "step": 12930
    },
    {
      "entropy": 6.1756562232971195,
      "epoch": 1.2929974508921878,
      "grad_norm": 1.390625,
      "learning_rate": 0.000484067010594692,
      "loss": 5.9681,
      "mean_token_accuracy": 0.14814321100711822,
      "num_tokens": 25306802.0,
      "step": 12935
    },
    {
      "entropy": 6.203471946716308,
      "epoch": 1.2934972759534162,
      "grad_norm": 1.578125,
      "learning_rate": 0.00048405381689176877,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15865119993686677,
      "num_tokens": 25316021.0,
      "step": 12940
    },
    {
      "entropy": 6.143183994293213,
      "epoch": 1.293997101014645,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048404061792910527,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15674288123846053,
      "num_tokens": 25325583.0,
      "step": 12945
    },
    {
      "entropy": 6.069062280654907,
      "epoch": 1.2944969260758734,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048402741370703363,
      "loss": 5.912,
      "mean_token_accuracy": 0.15656570494174957,
      "num_tokens": 25335046.0,
      "step": 12950
    },
    {
      "entropy": 6.191985177993774,
      "epoch": 1.294996751137102,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004840142042258861,
      "loss": 5.9697,
      "mean_token_accuracy": 0.15093196779489518,
      "num_tokens": 25344180.0,
      "step": 12955
    },
    {
      "entropy": 6.13280086517334,
      "epoch": 1.2954965761983306,
      "grad_norm": 1.46875,
      "learning_rate": 0.000484000989485995,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15331289917230606,
      "num_tokens": 25353131.0,
      "step": 12960
    },
    {
      "entropy": 6.143387651443481,
      "epoch": 1.295996401259559,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048398776948769265,
      "loss": 5.9369,
      "mean_token_accuracy": 0.15384265780448914,
      "num_tokens": 25362546.0,
      "step": 12965
    },
    {
      "entropy": 6.0823853492736815,
      "epoch": 1.2964962263207878,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004839745442313119,
      "loss": 5.8099,
      "mean_token_accuracy": 0.16645328104496002,
      "num_tokens": 25372773.0,
      "step": 12970
    },
    {
      "entropy": 6.187387704849243,
      "epoch": 1.2969960513820162,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048396131371718535,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16201002448797225,
      "num_tokens": 25382459.0,
      "step": 12975
    },
    {
      "entropy": 6.162664365768433,
      "epoch": 1.297495876443245,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048394807794564586,
      "loss": 5.9055,
      "mean_token_accuracy": 0.1562524124979973,
      "num_tokens": 25392561.0,
      "step": 12980
    },
    {
      "entropy": 6.0856410503387455,
      "epoch": 1.2979957015044734,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004839348369170266,
      "loss": 5.8312,
      "mean_token_accuracy": 0.1739437997341156,
      "num_tokens": 25402250.0,
      "step": 12985
    },
    {
      "entropy": 6.097572946548462,
      "epoch": 1.298495526565702,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048392159063166055,
      "loss": 5.8778,
      "mean_token_accuracy": 0.1568425066769123,
      "num_tokens": 25411813.0,
      "step": 12990
    },
    {
      "entropy": 6.0746817111969,
      "epoch": 1.2989953516269306,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004839083390898811,
      "loss": 5.8795,
      "mean_token_accuracy": 0.16151158213615419,
      "num_tokens": 25421232.0,
      "step": 12995
    },
    {
      "entropy": 6.0864492893219,
      "epoch": 1.299495176688159,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004838950822920215,
      "loss": 5.9333,
      "mean_token_accuracy": 0.1535163089632988,
      "num_tokens": 25430617.0,
      "step": 13000
    },
    {
      "entropy": 6.08137583732605,
      "epoch": 1.2999950017493878,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004838818202384155,
      "loss": 5.848,
      "mean_token_accuracy": 0.16881164014339448,
      "num_tokens": 25440514.0,
      "step": 13005
    },
    {
      "entropy": 6.236407232284546,
      "epoch": 1.3004948268106162,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048386855292939664,
      "loss": 6.0009,
      "mean_token_accuracy": 0.15506416261196138,
      "num_tokens": 25450058.0,
      "step": 13010
    },
    {
      "entropy": 6.133755397796631,
      "epoch": 1.300994651871845,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004838552803652986,
      "loss": 5.8969,
      "mean_token_accuracy": 0.15956542789936065,
      "num_tokens": 25459418.0,
      "step": 13015
    },
    {
      "entropy": 6.143233919143677,
      "epoch": 1.3014944769330734,
      "grad_norm": 1.65625,
      "learning_rate": 0.00048384200254645557,
      "loss": 5.9872,
      "mean_token_accuracy": 0.1539509564638138,
      "num_tokens": 25468944.0,
      "step": 13020
    },
    {
      "entropy": 6.1728332996368405,
      "epoch": 1.301994301994302,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048382871947320145,
      "loss": 5.8575,
      "mean_token_accuracy": 0.16375951915979386,
      "num_tokens": 25478072.0,
      "step": 13025
    },
    {
      "entropy": 6.1166160106658936,
      "epoch": 1.3024941270555306,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004838154311458705,
      "loss": 5.9797,
      "mean_token_accuracy": 0.15346761047840118,
      "num_tokens": 25488301.0,
      "step": 13030
    },
    {
      "entropy": 6.13940691947937,
      "epoch": 1.302993952116759,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048380213756479704,
      "loss": 5.9491,
      "mean_token_accuracy": 0.15289733111858367,
      "num_tokens": 25498343.0,
      "step": 13035
    },
    {
      "entropy": 6.22944974899292,
      "epoch": 1.3034937771779878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048378883873031553,
      "loss": 6.038,
      "mean_token_accuracy": 0.1556611105799675,
      "num_tokens": 25509302.0,
      "step": 13040
    },
    {
      "entropy": 6.190519666671753,
      "epoch": 1.3039936022392162,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004837755346427605,
      "loss": 5.9454,
      "mean_token_accuracy": 0.15846358835697175,
      "num_tokens": 25520246.0,
      "step": 13045
    },
    {
      "entropy": 6.153158712387085,
      "epoch": 1.3044934273004447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048376222530246684,
      "loss": 5.9171,
      "mean_token_accuracy": 0.15632445216178895,
      "num_tokens": 25528879.0,
      "step": 13050
    },
    {
      "entropy": 6.153446006774902,
      "epoch": 1.3049932523616734,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048374891070976925,
      "loss": 5.9157,
      "mean_token_accuracy": 0.16499528288841248,
      "num_tokens": 25538029.0,
      "step": 13055
    },
    {
      "entropy": 6.150085926055908,
      "epoch": 1.305493077422902,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004837355908650028,
      "loss": 5.9409,
      "mean_token_accuracy": 0.15373849868774414,
      "num_tokens": 25547752.0,
      "step": 13060
    },
    {
      "entropy": 6.108612537384033,
      "epoch": 1.3059929024841306,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004837222657685026,
      "loss": 5.9241,
      "mean_token_accuracy": 0.15909179896116257,
      "num_tokens": 25558197.0,
      "step": 13065
    },
    {
      "entropy": 6.197899580001831,
      "epoch": 1.306492727545359,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048370893542060383,
      "loss": 5.9616,
      "mean_token_accuracy": 0.15223454535007477,
      "num_tokens": 25569910.0,
      "step": 13070
    },
    {
      "entropy": 6.199585580825806,
      "epoch": 1.3069925526065878,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004836955998216421,
      "loss": 5.9719,
      "mean_token_accuracy": 0.1505519285798073,
      "num_tokens": 25579069.0,
      "step": 13075
    },
    {
      "entropy": 6.195717811584473,
      "epoch": 1.3074923776678162,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048368225897195264,
      "loss": 5.9384,
      "mean_token_accuracy": 0.15709191858768462,
      "num_tokens": 25588164.0,
      "step": 13080
    },
    {
      "entropy": 6.1322712898254395,
      "epoch": 1.3079922027290447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048366891287187133,
      "loss": 5.9949,
      "mean_token_accuracy": 0.15012910217046738,
      "num_tokens": 25598741.0,
      "step": 13085
    },
    {
      "entropy": 6.057543659210205,
      "epoch": 1.3084920277902734,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004836555615217337,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16177406907081604,
      "num_tokens": 25608572.0,
      "step": 13090
    },
    {
      "entropy": 6.1613095760345455,
      "epoch": 1.308991852851502,
      "grad_norm": 1.375,
      "learning_rate": 0.0004836422049218761,
      "loss": 5.8474,
      "mean_token_accuracy": 0.16407003551721572,
      "num_tokens": 25617686.0,
      "step": 13095
    },
    {
      "entropy": 6.05910177230835,
      "epoch": 1.3094916779127306,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048362884307263416,
      "loss": 5.8526,
      "mean_token_accuracy": 0.15647654831409455,
      "num_tokens": 25627268.0,
      "step": 13100
    },
    {
      "entropy": 6.086300086975098,
      "epoch": 1.309991502973959,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048361547597434426,
      "loss": 5.8424,
      "mean_token_accuracy": 0.16003944873809814,
      "num_tokens": 25636341.0,
      "step": 13105
    },
    {
      "entropy": 6.204550170898438,
      "epoch": 1.3104913280351878,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00048360210362734257,
      "loss": 5.9692,
      "mean_token_accuracy": 0.15372108072042465,
      "num_tokens": 25645403.0,
      "step": 13110
    },
    {
      "entropy": 6.176354122161865,
      "epoch": 1.3109911530964162,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004835887260319657,
      "loss": 5.9348,
      "mean_token_accuracy": 0.16364216804504395,
      "num_tokens": 25656062.0,
      "step": 13115
    },
    {
      "entropy": 6.132684993743896,
      "epoch": 1.3114909781576447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004835753431885502,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16075434386730195,
      "num_tokens": 25665449.0,
      "step": 13120
    },
    {
      "entropy": 6.038437938690185,
      "epoch": 1.3119908032188734,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048356195509743265,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15974109172821044,
      "num_tokens": 25674844.0,
      "step": 13125
    },
    {
      "entropy": 6.141631746292115,
      "epoch": 1.312490628280102,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048354856175895,
      "loss": 6.0228,
      "mean_token_accuracy": 0.15184858292341233,
      "num_tokens": 25686058.0,
      "step": 13130
    },
    {
      "entropy": 6.187440204620361,
      "epoch": 1.3129904533413306,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004835351631734392,
      "loss": 5.9287,
      "mean_token_accuracy": 0.1556500166654587,
      "num_tokens": 25696000.0,
      "step": 13135
    },
    {
      "entropy": 6.111225128173828,
      "epoch": 1.313490278402559,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004835217593412374,
      "loss": 5.8444,
      "mean_token_accuracy": 0.16574382781982422,
      "num_tokens": 25705912.0,
      "step": 13140
    },
    {
      "entropy": 6.057226371765137,
      "epoch": 1.3139901034637878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004835083502626818,
      "loss": 5.8612,
      "mean_token_accuracy": 0.15915179550647734,
      "num_tokens": 25716288.0,
      "step": 13145
    },
    {
      "entropy": 6.183387708663941,
      "epoch": 1.3144899285250162,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048349493593810973,
      "loss": 5.9697,
      "mean_token_accuracy": 0.14963506013154984,
      "num_tokens": 25726235.0,
      "step": 13150
    },
    {
      "entropy": 6.172686386108398,
      "epoch": 1.3149897535862447,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004834815163678587,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15596357733011246,
      "num_tokens": 25735740.0,
      "step": 13155
    },
    {
      "entropy": 6.051524019241333,
      "epoch": 1.3154895786474734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004834680915522664,
      "loss": 5.8465,
      "mean_token_accuracy": 0.16457830965518952,
      "num_tokens": 25744500.0,
      "step": 13160
    },
    {
      "entropy": 6.163321685791016,
      "epoch": 1.3159894037087019,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004834546614916705,
      "loss": 5.9146,
      "mean_token_accuracy": 0.1633441135287285,
      "num_tokens": 25754798.0,
      "step": 13165
    },
    {
      "entropy": 6.14723048210144,
      "epoch": 1.3164892287699306,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048344122618640905,
      "loss": 5.8931,
      "mean_token_accuracy": 0.15949040353298188,
      "num_tokens": 25764985.0,
      "step": 13170
    },
    {
      "entropy": 6.160395622253418,
      "epoch": 1.316989053831159,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004834277856368199,
      "loss": 5.9064,
      "mean_token_accuracy": 0.16058584898710251,
      "num_tokens": 25775929.0,
      "step": 13175
    },
    {
      "entropy": 6.217636632919311,
      "epoch": 1.3174888788923877,
      "grad_norm": 1.625,
      "learning_rate": 0.00048341433984324127,
      "loss": 6.1028,
      "mean_token_accuracy": 0.14271200224757194,
      "num_tokens": 25786319.0,
      "step": 13180
    },
    {
      "entropy": 6.09163727760315,
      "epoch": 1.3179887039536162,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048340088880601163,
      "loss": 5.9148,
      "mean_token_accuracy": 0.15661486983299255,
      "num_tokens": 25796394.0,
      "step": 13185
    },
    {
      "entropy": 6.119255065917969,
      "epoch": 1.3184885290148447,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004833874325254691,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1609025001525879,
      "num_tokens": 25805965.0,
      "step": 13190
    },
    {
      "entropy": 6.15711669921875,
      "epoch": 1.3189883540760734,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048337397100195245,
      "loss": 5.921,
      "mean_token_accuracy": 0.15351503640413283,
      "num_tokens": 25816219.0,
      "step": 13195
    },
    {
      "entropy": 6.1882867336273195,
      "epoch": 1.3194881791373019,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004833605042358003,
      "loss": 5.9699,
      "mean_token_accuracy": 0.1611633226275444,
      "num_tokens": 25826263.0,
      "step": 13200
    },
    {
      "entropy": 6.139550018310547,
      "epoch": 1.3199880041985306,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004833470322273515,
      "loss": 5.9508,
      "mean_token_accuracy": 0.15692346394062043,
      "num_tokens": 25836262.0,
      "step": 13205
    },
    {
      "entropy": 6.163528394699097,
      "epoch": 1.320487829259759,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048333355497694506,
      "loss": 5.858,
      "mean_token_accuracy": 0.16287244707345963,
      "num_tokens": 25845906.0,
      "step": 13210
    },
    {
      "entropy": 6.037047481536865,
      "epoch": 1.3209876543209877,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004833200724849199,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16854290664196014,
      "num_tokens": 25856251.0,
      "step": 13215
    },
    {
      "entropy": 6.060172367095947,
      "epoch": 1.3214874793822162,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048330658475161526,
      "loss": 5.9148,
      "mean_token_accuracy": 0.1643696941435337,
      "num_tokens": 25866695.0,
      "step": 13220
    },
    {
      "entropy": 6.131005430221558,
      "epoch": 1.3219873044434447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004832930917773706,
      "loss": 5.9796,
      "mean_token_accuracy": 0.1501013807952404,
      "num_tokens": 25877208.0,
      "step": 13225
    },
    {
      "entropy": 6.169745826721192,
      "epoch": 1.3224871295046734,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048327959356252544,
      "loss": 5.8693,
      "mean_token_accuracy": 0.16273389607667924,
      "num_tokens": 25887062.0,
      "step": 13230
    },
    {
      "entropy": 6.164175701141358,
      "epoch": 1.3229869545659019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004832660901074192,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15556773096323012,
      "num_tokens": 25896410.0,
      "step": 13235
    },
    {
      "entropy": 6.069396877288819,
      "epoch": 1.3234867796271306,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048325258141239174,
      "loss": 5.8348,
      "mean_token_accuracy": 0.16509501338005067,
      "num_tokens": 25905085.0,
      "step": 13240
    },
    {
      "entropy": 6.1346070766448975,
      "epoch": 1.323986604688359,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004832390674777829,
      "loss": 5.9129,
      "mean_token_accuracy": 0.1581639364361763,
      "num_tokens": 25914334.0,
      "step": 13245
    },
    {
      "entropy": 6.1127067565917965,
      "epoch": 1.3244864297495877,
      "grad_norm": 2.0625,
      "learning_rate": 0.00048322554830393274,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15506815910339355,
      "num_tokens": 25922676.0,
      "step": 13250
    },
    {
      "entropy": 6.1611357688903805,
      "epoch": 1.3249862548108162,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004832120238911813,
      "loss": 5.9583,
      "mean_token_accuracy": 0.1608611211180687,
      "num_tokens": 25931910.0,
      "step": 13255
    },
    {
      "entropy": 6.086991643905639,
      "epoch": 1.3254860798720447,
      "grad_norm": 1.6015625,
      "learning_rate": 0.000483198494239869,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16960249990224838,
      "num_tokens": 25942713.0,
      "step": 13260
    },
    {
      "entropy": 6.073838615417481,
      "epoch": 1.3259859049332734,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048318495935033615,
      "loss": 5.8181,
      "mean_token_accuracy": 0.16953064352273942,
      "num_tokens": 25951462.0,
      "step": 13265
    },
    {
      "entropy": 6.1616658687591555,
      "epoch": 1.3264857299945019,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048317141922292326,
      "loss": 5.8704,
      "mean_token_accuracy": 0.15909883603453637,
      "num_tokens": 25961172.0,
      "step": 13270
    },
    {
      "entropy": 6.131941556930542,
      "epoch": 1.3269855550557306,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004831578738579711,
      "loss": 5.8634,
      "mean_token_accuracy": 0.15851347744464875,
      "num_tokens": 25971119.0,
      "step": 13275
    },
    {
      "entropy": 6.178161096572876,
      "epoch": 1.327485380116959,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004831443232558203,
      "loss": 6.0491,
      "mean_token_accuracy": 0.15256685242056847,
      "num_tokens": 25981758.0,
      "step": 13280
    },
    {
      "entropy": 6.202067041397095,
      "epoch": 1.3279852051781877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048313076741681186,
      "loss": 6.002,
      "mean_token_accuracy": 0.15383404716849328,
      "num_tokens": 25991033.0,
      "step": 13285
    },
    {
      "entropy": 6.122411394119263,
      "epoch": 1.3284850302394162,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00048311720634128697,
      "loss": 5.9949,
      "mean_token_accuracy": 0.15210942924022675,
      "num_tokens": 26001547.0,
      "step": 13290
    },
    {
      "entropy": 6.198547506332398,
      "epoch": 1.3289848553006447,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004831036400295867,
      "loss": 6.0427,
      "mean_token_accuracy": 0.14736489132046698,
      "num_tokens": 26012140.0,
      "step": 13295
    },
    {
      "entropy": 6.2174742221832275,
      "epoch": 1.3294846803618734,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048309006848205237,
      "loss": 5.9385,
      "mean_token_accuracy": 0.15778605192899703,
      "num_tokens": 26022068.0,
      "step": 13300
    },
    {
      "entropy": 6.159972047805786,
      "epoch": 1.3299845054231019,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004830764916990255,
      "loss": 5.8856,
      "mean_token_accuracy": 0.15862743854522704,
      "num_tokens": 26031313.0,
      "step": 13305
    },
    {
      "entropy": 6.122745132446289,
      "epoch": 1.3304843304843303,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004830629096808476,
      "loss": 5.8526,
      "mean_token_accuracy": 0.16505161225795745,
      "num_tokens": 26040937.0,
      "step": 13310
    },
    {
      "entropy": 6.052115821838379,
      "epoch": 1.330984155545559,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004830493224278605,
      "loss": 5.7411,
      "mean_token_accuracy": 0.17441050708293915,
      "num_tokens": 26049719.0,
      "step": 13315
    },
    {
      "entropy": 6.2025166034698485,
      "epoch": 1.3314839806067877,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004830357299404059,
      "loss": 5.979,
      "mean_token_accuracy": 0.14776917546987534,
      "num_tokens": 26060064.0,
      "step": 13320
    },
    {
      "entropy": 6.110042715072632,
      "epoch": 1.3319838056680162,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004830221322188259,
      "loss": 5.8694,
      "mean_token_accuracy": 0.17106518447399138,
      "num_tokens": 26069605.0,
      "step": 13325
    },
    {
      "entropy": 6.227109098434449,
      "epoch": 1.3324836307292447,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004830085292634626,
      "loss": 5.9647,
      "mean_token_accuracy": 0.15052176117897034,
      "num_tokens": 26079075.0,
      "step": 13330
    },
    {
      "entropy": 6.078362894058228,
      "epoch": 1.3329834557904734,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048299492107465816,
      "loss": 5.9084,
      "mean_token_accuracy": 0.15480217337608337,
      "num_tokens": 26089003.0,
      "step": 13335
    },
    {
      "entropy": 6.149612140655518,
      "epoch": 1.3334832808517019,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004829813076527551,
      "loss": 5.9332,
      "mean_token_accuracy": 0.15504590123891832,
      "num_tokens": 26099044.0,
      "step": 13340
    },
    {
      "entropy": 6.087321376800537,
      "epoch": 1.3339831059129303,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048296768899809577,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15348293334245683,
      "num_tokens": 26109044.0,
      "step": 13345
    },
    {
      "entropy": 6.067951107025147,
      "epoch": 1.334482930974159,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048295406511102303,
      "loss": 5.9644,
      "mean_token_accuracy": 0.15014076605439186,
      "num_tokens": 26118593.0,
      "step": 13350
    },
    {
      "entropy": 6.228148174285889,
      "epoch": 1.3349827560353877,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004829404359918795,
      "loss": 5.9747,
      "mean_token_accuracy": 0.1605292797088623,
      "num_tokens": 26128885.0,
      "step": 13355
    },
    {
      "entropy": 6.280249977111817,
      "epoch": 1.3354825810966162,
      "grad_norm": 1.34375,
      "learning_rate": 0.000482926801641008,
      "loss": 5.9545,
      "mean_token_accuracy": 0.15650294050574304,
      "num_tokens": 26138173.0,
      "step": 13360
    },
    {
      "entropy": 6.170852708816528,
      "epoch": 1.3359824061578447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048291316205875174,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15836167633533477,
      "num_tokens": 26147724.0,
      "step": 13365
    },
    {
      "entropy": 6.083120250701905,
      "epoch": 1.3364822312190734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048289951724545384,
      "loss": 5.8724,
      "mean_token_accuracy": 0.16120509803295135,
      "num_tokens": 26156970.0,
      "step": 13370
    },
    {
      "entropy": 6.057164287567138,
      "epoch": 1.3369820562803019,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048288586720145757,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16292880922555925,
      "num_tokens": 26166757.0,
      "step": 13375
    },
    {
      "entropy": 6.155002117156982,
      "epoch": 1.3374818813415303,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004828722119271064,
      "loss": 5.9588,
      "mean_token_accuracy": 0.15484961867332458,
      "num_tokens": 26176944.0,
      "step": 13380
    },
    {
      "entropy": 6.110808372497559,
      "epoch": 1.337981706402759,
      "grad_norm": 1.25,
      "learning_rate": 0.00048285855142274375,
      "loss": 5.8103,
      "mean_token_accuracy": 0.1687612995505333,
      "num_tokens": 26185380.0,
      "step": 13385
    },
    {
      "entropy": 6.166572380065918,
      "epoch": 1.3384815314639877,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004828448856887135,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15843659192323684,
      "num_tokens": 26194936.0,
      "step": 13390
    },
    {
      "entropy": 6.047982549667358,
      "epoch": 1.3389813565252162,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004828312147253595,
      "loss": 5.8266,
      "mean_token_accuracy": 0.16571958363056183,
      "num_tokens": 26205825.0,
      "step": 13395
    },
    {
      "entropy": 6.189558362960815,
      "epoch": 1.3394811815864447,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004828175385330255,
      "loss": 5.9797,
      "mean_token_accuracy": 0.1481102451682091,
      "num_tokens": 26215977.0,
      "step": 13400
    },
    {
      "entropy": 6.194770288467407,
      "epoch": 1.3399810066476734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004828038571120558,
      "loss": 5.9349,
      "mean_token_accuracy": 0.15716130286455154,
      "num_tokens": 26225424.0,
      "step": 13405
    },
    {
      "entropy": 6.1802901268005375,
      "epoch": 1.3404808317089019,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048279017046279446,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15692477077245712,
      "num_tokens": 26235491.0,
      "step": 13410
    },
    {
      "entropy": 6.161337423324585,
      "epoch": 1.3409806567701303,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048277647858558596,
      "loss": 6.0087,
      "mean_token_accuracy": 0.15507736206054687,
      "num_tokens": 26246030.0,
      "step": 13415
    },
    {
      "entropy": 6.22878794670105,
      "epoch": 1.341480481831359,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048276278148077465,
      "loss": 6.0332,
      "mean_token_accuracy": 0.14370579421520233,
      "num_tokens": 26255008.0,
      "step": 13420
    },
    {
      "entropy": 6.090753555297852,
      "epoch": 1.3419803068925875,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048274907914870526,
      "loss": 5.7004,
      "mean_token_accuracy": 0.18034014999866485,
      "num_tokens": 26263319.0,
      "step": 13425
    },
    {
      "entropy": 6.08425087928772,
      "epoch": 1.3424801319538162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048273537158972246,
      "loss": 5.8522,
      "mean_token_accuracy": 0.1581566572189331,
      "num_tokens": 26272555.0,
      "step": 13430
    },
    {
      "entropy": 6.1538777351379395,
      "epoch": 1.3429799570150447,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048272165880417124,
      "loss": 5.9134,
      "mean_token_accuracy": 0.15936014503240586,
      "num_tokens": 26282915.0,
      "step": 13435
    },
    {
      "entropy": 6.144521760940552,
      "epoch": 1.3434797820762734,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048270794079239643,
      "loss": 5.8706,
      "mean_token_accuracy": 0.1634807601571083,
      "num_tokens": 26292705.0,
      "step": 13440
    },
    {
      "entropy": 6.236977624893188,
      "epoch": 1.3439796071375019,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048269421755474336,
      "loss": 5.9845,
      "mean_token_accuracy": 0.14948619902133942,
      "num_tokens": 26303678.0,
      "step": 13445
    },
    {
      "entropy": 6.1484180927276615,
      "epoch": 1.3444794321987303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004826804890915572,
      "loss": 5.9059,
      "mean_token_accuracy": 0.15340238958597183,
      "num_tokens": 26312844.0,
      "step": 13450
    },
    {
      "entropy": 6.18296709060669,
      "epoch": 1.344979257259959,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004826667554031834,
      "loss": 5.9756,
      "mean_token_accuracy": 0.14909876957535745,
      "num_tokens": 26322763.0,
      "step": 13455
    },
    {
      "entropy": 6.081702136993409,
      "epoch": 1.3454790823211875,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004826530164899674,
      "loss": 5.8957,
      "mean_token_accuracy": 0.15366754159331322,
      "num_tokens": 26331381.0,
      "step": 13460
    },
    {
      "entropy": 5.980902814865113,
      "epoch": 1.3459789073824162,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048263927235225505,
      "loss": 5.6593,
      "mean_token_accuracy": 0.18007916510105132,
      "num_tokens": 26340637.0,
      "step": 13465
    },
    {
      "entropy": 6.106406831741333,
      "epoch": 1.3464787324436447,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004826255229903919,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1597217619419098,
      "num_tokens": 26350508.0,
      "step": 13470
    },
    {
      "entropy": 6.116240072250366,
      "epoch": 1.3469785575048734,
      "grad_norm": 1.96875,
      "learning_rate": 0.00048261176840472416,
      "loss": 6.069,
      "mean_token_accuracy": 0.1550678178668022,
      "num_tokens": 26360419.0,
      "step": 13475
    },
    {
      "entropy": 6.1709277629852295,
      "epoch": 1.3474783825661019,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048259800859559764,
      "loss": 5.9124,
      "mean_token_accuracy": 0.15869924575090408,
      "num_tokens": 26370355.0,
      "step": 13480
    },
    {
      "entropy": 6.083157587051391,
      "epoch": 1.3479782076273303,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004825842435633587,
      "loss": 5.8536,
      "mean_token_accuracy": 0.16081081032752992,
      "num_tokens": 26380085.0,
      "step": 13485
    },
    {
      "entropy": 6.123891687393188,
      "epoch": 1.348478032688559,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048257047330835354,
      "loss": 5.9029,
      "mean_token_accuracy": 0.16023032516241073,
      "num_tokens": 26389571.0,
      "step": 13490
    },
    {
      "entropy": 6.107320737838745,
      "epoch": 1.3489778577497875,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048255669783092876,
      "loss": 5.8892,
      "mean_token_accuracy": 0.16531556099653244,
      "num_tokens": 26399560.0,
      "step": 13495
    },
    {
      "entropy": 6.097224998474121,
      "epoch": 1.3494776828110162,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048254291713143083,
      "loss": 5.9002,
      "mean_token_accuracy": 0.16168479472398758,
      "num_tokens": 26409290.0,
      "step": 13500
    },
    {
      "entropy": 6.191761207580567,
      "epoch": 1.3499775078722447,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004825291312102066,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15715757012367249,
      "num_tokens": 26419458.0,
      "step": 13505
    },
    {
      "entropy": 6.172069072723389,
      "epoch": 1.3504773329334734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004825153400676027,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16611333042383195,
      "num_tokens": 26428947.0,
      "step": 13510
    },
    {
      "entropy": 5.952283430099487,
      "epoch": 1.3509771579947019,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004825015437039663,
      "loss": 5.7527,
      "mean_token_accuracy": 0.16579325646162033,
      "num_tokens": 26438874.0,
      "step": 13515
    },
    {
      "entropy": 6.110401010513305,
      "epoch": 1.3514769830559303,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004824877421196444,
      "loss": 5.8505,
      "mean_token_accuracy": 0.1588907554745674,
      "num_tokens": 26448510.0,
      "step": 13520
    },
    {
      "entropy": 5.996318387985229,
      "epoch": 1.351976808117159,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004824739353149844,
      "loss": 5.876,
      "mean_token_accuracy": 0.1532279998064041,
      "num_tokens": 26458552.0,
      "step": 13525
    },
    {
      "entropy": 6.078644132614135,
      "epoch": 1.3524766331783875,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048246012329033355,
      "loss": 5.9148,
      "mean_token_accuracy": 0.15787014067173005,
      "num_tokens": 26467900.0,
      "step": 13530
    },
    {
      "entropy": 6.140193367004395,
      "epoch": 1.3529764582396162,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048244630604603935,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16075575053691865,
      "num_tokens": 26477229.0,
      "step": 13535
    },
    {
      "entropy": 6.1568327903747555,
      "epoch": 1.3534762833008447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048243248358244945,
      "loss": 5.9397,
      "mean_token_accuracy": 0.1578460529446602,
      "num_tokens": 26487358.0,
      "step": 13540
    },
    {
      "entropy": 6.08443284034729,
      "epoch": 1.3539761083620734,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004824186558999117,
      "loss": 5.7943,
      "mean_token_accuracy": 0.16813967674970626,
      "num_tokens": 26496399.0,
      "step": 13545
    },
    {
      "entropy": 6.049694442749024,
      "epoch": 1.3544759334233019,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048240482299877386,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16037366837263106,
      "num_tokens": 26506170.0,
      "step": 13550
    },
    {
      "entropy": 6.174593734741211,
      "epoch": 1.3549757584845303,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004823909848793841,
      "loss": 5.9391,
      "mean_token_accuracy": 0.15266840755939484,
      "num_tokens": 26516414.0,
      "step": 13555
    },
    {
      "entropy": 6.079722738265991,
      "epoch": 1.355475583545759,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048237714154209053,
      "loss": 5.8381,
      "mean_token_accuracy": 0.16278315037488938,
      "num_tokens": 26526345.0,
      "step": 13560
    },
    {
      "entropy": 6.185301113128662,
      "epoch": 1.3559754086069875,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004823632929872415,
      "loss": 5.9302,
      "mean_token_accuracy": 0.1555140197277069,
      "num_tokens": 26535768.0,
      "step": 13565
    },
    {
      "entropy": 6.132519721984863,
      "epoch": 1.3564752336682162,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048234943921518524,
      "loss": 5.9261,
      "mean_token_accuracy": 0.15608844459056853,
      "num_tokens": 26546181.0,
      "step": 13570
    },
    {
      "entropy": 6.090471172332764,
      "epoch": 1.3569750587294447,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004823355802262705,
      "loss": 5.8964,
      "mean_token_accuracy": 0.16082009077072143,
      "num_tokens": 26555664.0,
      "step": 13575
    },
    {
      "entropy": 6.174325704574585,
      "epoch": 1.3574748837906734,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004823217160208459,
      "loss": 5.9566,
      "mean_token_accuracy": 0.15544834062457086,
      "num_tokens": 26564801.0,
      "step": 13580
    },
    {
      "entropy": 6.241074323654175,
      "epoch": 1.3579747088519019,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004823078465992604,
      "loss": 6.0554,
      "mean_token_accuracy": 0.15187243819236756,
      "num_tokens": 26575397.0,
      "step": 13585
    },
    {
      "entropy": 6.122386169433594,
      "epoch": 1.3584745339131303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004822939719618627,
      "loss": 5.8929,
      "mean_token_accuracy": 0.1644446596503258,
      "num_tokens": 26584533.0,
      "step": 13590
    },
    {
      "entropy": 6.176441431045532,
      "epoch": 1.358974358974359,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000482280092109002,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16560427397489547,
      "num_tokens": 26594176.0,
      "step": 13595
    },
    {
      "entropy": 6.123038816452026,
      "epoch": 1.3594741840355875,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048226620704102757,
      "loss": 5.9091,
      "mean_token_accuracy": 0.1552777484059334,
      "num_tokens": 26603033.0,
      "step": 13600
    },
    {
      "entropy": 6.16979489326477,
      "epoch": 1.359974009096816,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004822523167582886,
      "loss": 5.8807,
      "mean_token_accuracy": 0.16403100937604903,
      "num_tokens": 26612327.0,
      "step": 13605
    },
    {
      "entropy": 6.116194581985473,
      "epoch": 1.3604738341580447,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004822384212611348,
      "loss": 5.8675,
      "mean_token_accuracy": 0.15489978045225145,
      "num_tokens": 26622180.0,
      "step": 13610
    },
    {
      "entropy": 6.146957540512085,
      "epoch": 1.3609736592192734,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048222452054991563,
      "loss": 6.0014,
      "mean_token_accuracy": 0.14460513442754747,
      "num_tokens": 26633137.0,
      "step": 13615
    },
    {
      "entropy": 6.131438255310059,
      "epoch": 1.3614734842805019,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004822106146249808,
      "loss": 5.7959,
      "mean_token_accuracy": 0.16340333446860314,
      "num_tokens": 26642642.0,
      "step": 13620
    },
    {
      "entropy": 6.198835611343384,
      "epoch": 1.3619733093417303,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048219670348668026,
      "loss": 6.003,
      "mean_token_accuracy": 0.1553576596081257,
      "num_tokens": 26652465.0,
      "step": 13625
    },
    {
      "entropy": 6.167665815353393,
      "epoch": 1.362473134402959,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048218278713536394,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15892619043588638,
      "num_tokens": 26662971.0,
      "step": 13630
    },
    {
      "entropy": 6.024131345748901,
      "epoch": 1.3629729594641875,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004821688655713821,
      "loss": 5.8346,
      "mean_token_accuracy": 0.15902095586061477,
      "num_tokens": 26672373.0,
      "step": 13635
    },
    {
      "entropy": 6.060509109497071,
      "epoch": 1.363472784525416,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048215493879508474,
      "loss": 5.8399,
      "mean_token_accuracy": 0.1661190167069435,
      "num_tokens": 26681491.0,
      "step": 13640
    },
    {
      "entropy": 6.174294137954712,
      "epoch": 1.3639726095866447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004821410068068226,
      "loss": 5.9273,
      "mean_token_accuracy": 0.15761102885007858,
      "num_tokens": 26690094.0,
      "step": 13645
    },
    {
      "entropy": 6.170170545578003,
      "epoch": 1.3644724346478734,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004821270696069459,
      "loss": 5.9214,
      "mean_token_accuracy": 0.15198128968477248,
      "num_tokens": 26699525.0,
      "step": 13650
    },
    {
      "entropy": 6.089055395126342,
      "epoch": 1.3649722597091019,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048211312719580553,
      "loss": 5.8278,
      "mean_token_accuracy": 0.1570058971643448,
      "num_tokens": 26709943.0,
      "step": 13655
    },
    {
      "entropy": 6.073455476760865,
      "epoch": 1.3654720847703303,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004820991795737521,
      "loss": 5.8931,
      "mean_token_accuracy": 0.15664578676223756,
      "num_tokens": 26719517.0,
      "step": 13660
    },
    {
      "entropy": 6.164521503448486,
      "epoch": 1.365971909831559,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048208522674113665,
      "loss": 5.9416,
      "mean_token_accuracy": 0.16593834087252618,
      "num_tokens": 26728753.0,
      "step": 13665
    },
    {
      "entropy": 6.130479860305786,
      "epoch": 1.3664717348927875,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004820712686983101,
      "loss": 5.9018,
      "mean_token_accuracy": 0.15508417338132857,
      "num_tokens": 26737662.0,
      "step": 13670
    },
    {
      "entropy": 6.0040360450744625,
      "epoch": 1.366971559954016,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048205730544562385,
      "loss": 5.8644,
      "mean_token_accuracy": 0.16468303203582763,
      "num_tokens": 26748167.0,
      "step": 13675
    },
    {
      "entropy": 6.1554969310760494,
      "epoch": 1.3674713850152447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000482043336983429,
      "loss": 5.9465,
      "mean_token_accuracy": 0.156490296125412,
      "num_tokens": 26756692.0,
      "step": 13680
    },
    {
      "entropy": 6.109146499633789,
      "epoch": 1.3679712100764732,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048202936331207707,
      "loss": 5.8753,
      "mean_token_accuracy": 0.16479385197162627,
      "num_tokens": 26766201.0,
      "step": 13685
    },
    {
      "entropy": 6.090929269790649,
      "epoch": 1.3684710351377019,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004820153844319196,
      "loss": 5.883,
      "mean_token_accuracy": 0.1596833199262619,
      "num_tokens": 26774911.0,
      "step": 13690
    },
    {
      "entropy": 6.130761241912841,
      "epoch": 1.3689708601989303,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048200140034330844,
      "loss": 5.8594,
      "mean_token_accuracy": 0.16206910461187363,
      "num_tokens": 26784271.0,
      "step": 13695
    },
    {
      "entropy": 6.18708872795105,
      "epoch": 1.369470685260159,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048198741104659516,
      "loss": 5.9514,
      "mean_token_accuracy": 0.15701534748077392,
      "num_tokens": 26794975.0,
      "step": 13700
    },
    {
      "entropy": 6.205992269515991,
      "epoch": 1.3699705103213875,
      "grad_norm": 1.171875,
      "learning_rate": 0.000481973416542132,
      "loss": 6.0088,
      "mean_token_accuracy": 0.15671658664941787,
      "num_tokens": 26805989.0,
      "step": 13705
    },
    {
      "entropy": 6.128444671630859,
      "epoch": 1.370470335382616,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004819594168302709,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15357046127319335,
      "num_tokens": 26815961.0,
      "step": 13710
    },
    {
      "entropy": 6.050806713104248,
      "epoch": 1.3709701604438447,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004819454119113641,
      "loss": 5.797,
      "mean_token_accuracy": 0.16666629463434218,
      "num_tokens": 26825697.0,
      "step": 13715
    },
    {
      "entropy": 6.176212358474731,
      "epoch": 1.3714699855050732,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000481931401785764,
      "loss": 5.8502,
      "mean_token_accuracy": 0.16377743929624558,
      "num_tokens": 26834965.0,
      "step": 13720
    },
    {
      "entropy": 6.124526071548462,
      "epoch": 1.3719698105663019,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004819173864538231,
      "loss": 5.8847,
      "mean_token_accuracy": 0.16484444141387938,
      "num_tokens": 26845621.0,
      "step": 13725
    },
    {
      "entropy": 6.195318174362183,
      "epoch": 1.3724696356275303,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048190336591589394,
      "loss": 5.9773,
      "mean_token_accuracy": 0.15513915419578553,
      "num_tokens": 26855176.0,
      "step": 13730
    },
    {
      "entropy": 6.199703216552734,
      "epoch": 1.372969460688759,
      "grad_norm": 1.625,
      "learning_rate": 0.0004818893401723294,
      "loss": 5.9339,
      "mean_token_accuracy": 0.15509801656007766,
      "num_tokens": 26864197.0,
      "step": 13735
    },
    {
      "entropy": 6.157736873626709,
      "epoch": 1.3734692857499875,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048187530922348226,
      "loss": 5.9006,
      "mean_token_accuracy": 0.1579801246523857,
      "num_tokens": 26873895.0,
      "step": 13740
    },
    {
      "entropy": 6.10373363494873,
      "epoch": 1.373969110811216,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004818612730697056,
      "loss": 6.0244,
      "mean_token_accuracy": 0.151571524143219,
      "num_tokens": 26883685.0,
      "step": 13745
    },
    {
      "entropy": 6.102708721160889,
      "epoch": 1.3744689358724447,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048184723171135256,
      "loss": 5.7828,
      "mean_token_accuracy": 0.16517499834299088,
      "num_tokens": 26893939.0,
      "step": 13750
    },
    {
      "entropy": 6.061699533462525,
      "epoch": 1.3749687609336732,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048183318514877634,
      "loss": 5.8718,
      "mean_token_accuracy": 0.1628713995218277,
      "num_tokens": 26903588.0,
      "step": 13755
    },
    {
      "entropy": 6.182736730575561,
      "epoch": 1.3754685859949018,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004818191333823305,
      "loss": 6.019,
      "mean_token_accuracy": 0.1535254068672657,
      "num_tokens": 26914068.0,
      "step": 13760
    },
    {
      "entropy": 6.114836311340332,
      "epoch": 1.3759684110561303,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004818050764123684,
      "loss": 5.9061,
      "mean_token_accuracy": 0.15789632946252824,
      "num_tokens": 26923928.0,
      "step": 13765
    },
    {
      "entropy": 6.077105188369751,
      "epoch": 1.376468236117359,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004817910142392438,
      "loss": 5.8365,
      "mean_token_accuracy": 0.16467103064060212,
      "num_tokens": 26933202.0,
      "step": 13770
    },
    {
      "entropy": 5.947284984588623,
      "epoch": 1.3769680611785875,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048177694686331047,
      "loss": 5.727,
      "mean_token_accuracy": 0.1658219039440155,
      "num_tokens": 26943466.0,
      "step": 13775
    },
    {
      "entropy": 6.0689859867095945,
      "epoch": 1.377467886239816,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004817628742849224,
      "loss": 5.8346,
      "mean_token_accuracy": 0.15967981219291688,
      "num_tokens": 26953996.0,
      "step": 13780
    },
    {
      "entropy": 6.146295118331909,
      "epoch": 1.3779677113010447,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048174879650443364,
      "loss": 5.8994,
      "mean_token_accuracy": 0.16113250702619553,
      "num_tokens": 26963238.0,
      "step": 13785
    },
    {
      "entropy": 6.161510705947876,
      "epoch": 1.3784675363622731,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048173471352219835,
      "loss": 5.8906,
      "mean_token_accuracy": 0.1501619264483452,
      "num_tokens": 26973124.0,
      "step": 13790
    },
    {
      "entropy": 6.134519433975219,
      "epoch": 1.3789673614235018,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004817206253385708,
      "loss": 5.8273,
      "mean_token_accuracy": 0.1644789382815361,
      "num_tokens": 26981698.0,
      "step": 13795
    },
    {
      "entropy": 6.150260066986084,
      "epoch": 1.3794671864847303,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00048170653195390555,
      "loss": 5.9482,
      "mean_token_accuracy": 0.1558808885514736,
      "num_tokens": 26991263.0,
      "step": 13800
    },
    {
      "entropy": 6.018583106994629,
      "epoch": 1.379967011545959,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004816924333685572,
      "loss": 5.8256,
      "mean_token_accuracy": 0.17047991082072259,
      "num_tokens": 27001835.0,
      "step": 13805
    },
    {
      "entropy": 6.0284007549285885,
      "epoch": 1.3804668366071875,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004816783295828804,
      "loss": 5.8834,
      "mean_token_accuracy": 0.15807196050882338,
      "num_tokens": 27011306.0,
      "step": 13810
    },
    {
      "entropy": 6.239481592178345,
      "epoch": 1.380966661668416,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048166422059722996,
      "loss": 6.0645,
      "mean_token_accuracy": 0.14694201350212097,
      "num_tokens": 27022038.0,
      "step": 13815
    },
    {
      "entropy": 6.174250793457031,
      "epoch": 1.3814664867296447,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004816501064119609,
      "loss": 5.9429,
      "mean_token_accuracy": 0.1534029796719551,
      "num_tokens": 27032225.0,
      "step": 13820
    },
    {
      "entropy": 6.173323917388916,
      "epoch": 1.3819663117908731,
      "grad_norm": 1.25,
      "learning_rate": 0.0004816359870274284,
      "loss": 5.9027,
      "mean_token_accuracy": 0.1597764328122139,
      "num_tokens": 27042770.0,
      "step": 13825
    },
    {
      "entropy": 6.156355953216552,
      "epoch": 1.3824661368521018,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004816218624439876,
      "loss": 5.9304,
      "mean_token_accuracy": 0.16274769008159637,
      "num_tokens": 27052543.0,
      "step": 13830
    },
    {
      "entropy": 6.203956651687622,
      "epoch": 1.3829659619133303,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048160773266199384,
      "loss": 5.9231,
      "mean_token_accuracy": 0.15539293140172958,
      "num_tokens": 27061955.0,
      "step": 13835
    },
    {
      "entropy": 6.1275670528411865,
      "epoch": 1.383465786974559,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004815935976818028,
      "loss": 5.8418,
      "mean_token_accuracy": 0.16450047641992568,
      "num_tokens": 27071458.0,
      "step": 13840
    },
    {
      "entropy": 6.105495548248291,
      "epoch": 1.3839656120357875,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048157945750377005,
      "loss": 5.864,
      "mean_token_accuracy": 0.16048668324947357,
      "num_tokens": 27081331.0,
      "step": 13845
    },
    {
      "entropy": 6.040698766708374,
      "epoch": 1.384465437097016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004815653121282512,
      "loss": 5.8297,
      "mean_token_accuracy": 0.1665486916899681,
      "num_tokens": 27090159.0,
      "step": 13850
    },
    {
      "entropy": 6.083501815795898,
      "epoch": 1.3849652621582447,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048155116155560227,
      "loss": 5.9098,
      "mean_token_accuracy": 0.16268494725227356,
      "num_tokens": 27099389.0,
      "step": 13855
    },
    {
      "entropy": 6.103564500808716,
      "epoch": 1.3854650872194731,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048153700578617924,
      "loss": 5.9808,
      "mean_token_accuracy": 0.15656136050820352,
      "num_tokens": 27109878.0,
      "step": 13860
    },
    {
      "entropy": 6.170550537109375,
      "epoch": 1.3859649122807016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004815228448203384,
      "loss": 5.9501,
      "mean_token_accuracy": 0.15168124660849572,
      "num_tokens": 27119556.0,
      "step": 13865
    },
    {
      "entropy": 6.153482866287232,
      "epoch": 1.3864647373419303,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004815086786584358,
      "loss": 5.8905,
      "mean_token_accuracy": 0.16171278133988382,
      "num_tokens": 27129086.0,
      "step": 13870
    },
    {
      "entropy": 6.169485807418823,
      "epoch": 1.386964562403159,
      "grad_norm": 1.25,
      "learning_rate": 0.000481494507300828,
      "loss": 5.9653,
      "mean_token_accuracy": 0.16194775104522705,
      "num_tokens": 27138908.0,
      "step": 13875
    },
    {
      "entropy": 6.104734134674072,
      "epoch": 1.3874643874643875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004814803307478716,
      "loss": 5.8856,
      "mean_token_accuracy": 0.16163360625505446,
      "num_tokens": 27149306.0,
      "step": 13880
    },
    {
      "entropy": 6.017145442962646,
      "epoch": 1.387964212525616,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048146614899992313,
      "loss": 5.7393,
      "mean_token_accuracy": 0.17187042236328126,
      "num_tokens": 27159293.0,
      "step": 13885
    },
    {
      "entropy": 6.124736642837524,
      "epoch": 1.3884640375868447,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048145196205733954,
      "loss": 5.8436,
      "mean_token_accuracy": 0.16460649222135543,
      "num_tokens": 27168450.0,
      "step": 13890
    },
    {
      "entropy": 6.100143241882324,
      "epoch": 1.3889638626480731,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048143776992047767,
      "loss": 5.8493,
      "mean_token_accuracy": 0.1661736100912094,
      "num_tokens": 27177548.0,
      "step": 13895
    },
    {
      "entropy": 6.111844539642334,
      "epoch": 1.3894636877093016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048142357258969465,
      "loss": 5.9105,
      "mean_token_accuracy": 0.1620581939816475,
      "num_tokens": 27186826.0,
      "step": 13900
    },
    {
      "entropy": 6.1437780380249025,
      "epoch": 1.3899635127705303,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004814093700653477,
      "loss": 5.868,
      "mean_token_accuracy": 0.15649183243513107,
      "num_tokens": 27196363.0,
      "step": 13905
    },
    {
      "entropy": 6.192549514770508,
      "epoch": 1.390463337831759,
      "grad_norm": 1.25,
      "learning_rate": 0.00048139516234779403,
      "loss": 5.9462,
      "mean_token_accuracy": 0.15697361677885055,
      "num_tokens": 27205461.0,
      "step": 13910
    },
    {
      "entropy": 6.25398736000061,
      "epoch": 1.3909631628929875,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048138094943739123,
      "loss": 6.0663,
      "mean_token_accuracy": 0.14634933322668076,
      "num_tokens": 27214490.0,
      "step": 13915
    },
    {
      "entropy": 6.126421976089477,
      "epoch": 1.391462987954216,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048136673133449687,
      "loss": 5.8512,
      "mean_token_accuracy": 0.1629421517252922,
      "num_tokens": 27224020.0,
      "step": 13920
    },
    {
      "entropy": 5.980487728118897,
      "epoch": 1.3919628130154447,
      "grad_norm": 1.9296875,
      "learning_rate": 0.0004813525080394686,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16845010593533516,
      "num_tokens": 27234591.0,
      "step": 13925
    },
    {
      "entropy": 6.091335010528565,
      "epoch": 1.3924626380766731,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004813382795526644,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1623811185359955,
      "num_tokens": 27244991.0,
      "step": 13930
    },
    {
      "entropy": 6.042789649963379,
      "epoch": 1.3929624631379016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004813240458744421,
      "loss": 5.8657,
      "mean_token_accuracy": 0.15885193794965743,
      "num_tokens": 27254663.0,
      "step": 13935
    },
    {
      "entropy": 6.205064678192139,
      "epoch": 1.3934622881991303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048130980700516,
      "loss": 5.9665,
      "mean_token_accuracy": 0.15413131564855576,
      "num_tokens": 27265238.0,
      "step": 13940
    },
    {
      "entropy": 6.2062047004699705,
      "epoch": 1.3939621132603588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004812955629451761,
      "loss": 5.8944,
      "mean_token_accuracy": 0.16959484964609145,
      "num_tokens": 27273844.0,
      "step": 13945
    },
    {
      "entropy": 6.093850755691529,
      "epoch": 1.3944619383215875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004812813136948491,
      "loss": 5.844,
      "mean_token_accuracy": 0.15456968247890474,
      "num_tokens": 27283356.0,
      "step": 13950
    },
    {
      "entropy": 5.8942337989807125,
      "epoch": 1.394961763382816,
      "grad_norm": 1.25,
      "learning_rate": 0.00048126705925453716,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17931944578886033,
      "num_tokens": 27292685.0,
      "step": 13955
    },
    {
      "entropy": 6.086632299423218,
      "epoch": 1.3954615884440447,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004812527996245992,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16086450964212418,
      "num_tokens": 27302347.0,
      "step": 13960
    },
    {
      "entropy": 6.169581413269043,
      "epoch": 1.3959614135052731,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004812385348053938,
      "loss": 5.974,
      "mean_token_accuracy": 0.1595055177807808,
      "num_tokens": 27311807.0,
      "step": 13965
    },
    {
      "entropy": 6.140850448608399,
      "epoch": 1.3964612385665016,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048122426479728,
      "loss": 5.9601,
      "mean_token_accuracy": 0.1551966704428196,
      "num_tokens": 27322561.0,
      "step": 13970
    },
    {
      "entropy": 6.141512012481689,
      "epoch": 1.3969610636277303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004812099896006167,
      "loss": 5.8519,
      "mean_token_accuracy": 0.1656806915998459,
      "num_tokens": 27332105.0,
      "step": 13975
    },
    {
      "entropy": 6.068198251724243,
      "epoch": 1.3974608886889588,
      "grad_norm": 1.375,
      "learning_rate": 0.0004811957092157631,
      "loss": 5.8649,
      "mean_token_accuracy": 0.16362714469432832,
      "num_tokens": 27341900.0,
      "step": 13980
    },
    {
      "entropy": 6.0383471012115475,
      "epoch": 1.3979607137501875,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048118142364307864,
      "loss": 5.7931,
      "mean_token_accuracy": 0.16795430779457093,
      "num_tokens": 27351991.0,
      "step": 13985
    },
    {
      "entropy": 6.137063217163086,
      "epoch": 1.398460538811416,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004811671328829225,
      "loss": 5.8648,
      "mean_token_accuracy": 0.1637093484401703,
      "num_tokens": 27361634.0,
      "step": 13990
    },
    {
      "entropy": 6.163334512710572,
      "epoch": 1.3989603638726447,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048115283693565445,
      "loss": 5.8853,
      "mean_token_accuracy": 0.1610477328300476,
      "num_tokens": 27373528.0,
      "step": 13995
    },
    {
      "entropy": 6.12969012260437,
      "epoch": 1.3994601889338731,
      "grad_norm": 1.28125,
      "learning_rate": 0.000481138535801634,
      "loss": 5.8917,
      "mean_token_accuracy": 0.16088420152664185,
      "num_tokens": 27382123.0,
      "step": 14000
    },
    {
      "entropy": 6.149842977523804,
      "epoch": 1.3999600139951016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000481124229481221,
      "loss": 5.86,
      "mean_token_accuracy": 0.16204839199781418,
      "num_tokens": 27392486.0,
      "step": 14005
    },
    {
      "entropy": 6.162126255035401,
      "epoch": 1.4004598390563303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004811099179747755,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1572623446583748,
      "num_tokens": 27402018.0,
      "step": 14010
    },
    {
      "entropy": 6.127962493896485,
      "epoch": 1.4009596641175588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048109560128265747,
      "loss": 5.8596,
      "mean_token_accuracy": 0.16236720532178878,
      "num_tokens": 27411404.0,
      "step": 14015
    },
    {
      "entropy": 6.1767247200012205,
      "epoch": 1.4014594891787875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004810812794052271,
      "loss": 6.0351,
      "mean_token_accuracy": 0.1528657242655754,
      "num_tokens": 27420968.0,
      "step": 14020
    },
    {
      "entropy": 6.156317234039307,
      "epoch": 1.401959314240016,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004810669523428449,
      "loss": 5.8621,
      "mean_token_accuracy": 0.15596481263637543,
      "num_tokens": 27430964.0,
      "step": 14025
    },
    {
      "entropy": 6.112464332580567,
      "epoch": 1.4024591393012447,
      "grad_norm": 1.28125,
      "learning_rate": 0.000481052620095871,
      "loss": 5.8682,
      "mean_token_accuracy": 0.15811689645051957,
      "num_tokens": 27440517.0,
      "step": 14030
    },
    {
      "entropy": 6.062809753417969,
      "epoch": 1.4029589643624731,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004810382826646664,
      "loss": 5.9121,
      "mean_token_accuracy": 0.1564284011721611,
      "num_tokens": 27449085.0,
      "step": 14035
    },
    {
      "entropy": 6.070509672164917,
      "epoch": 1.4034587894237016,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004810239400495915,
      "loss": 5.804,
      "mean_token_accuracy": 0.16930284351110458,
      "num_tokens": 27457637.0,
      "step": 14040
    },
    {
      "entropy": 6.165882205963134,
      "epoch": 1.4039586144849303,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004810095922510073,
      "loss": 5.9233,
      "mean_token_accuracy": 0.15596123337745665,
      "num_tokens": 27467967.0,
      "step": 14045
    },
    {
      "entropy": 6.134672784805298,
      "epoch": 1.4044584395461588,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048099523926927487,
      "loss": 5.8781,
      "mean_token_accuracy": 0.1616300716996193,
      "num_tokens": 27477386.0,
      "step": 14050
    },
    {
      "entropy": 6.0718554019927975,
      "epoch": 1.4049582646073875,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004809808811047552,
      "loss": 5.8005,
      "mean_token_accuracy": 0.1618144541978836,
      "num_tokens": 27487289.0,
      "step": 14055
    },
    {
      "entropy": 6.112859582901001,
      "epoch": 1.405458089668616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048096651775780943,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16705528497695923,
      "num_tokens": 27496687.0,
      "step": 14060
    },
    {
      "entropy": 6.063718032836914,
      "epoch": 1.4059579147298447,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004809521492287992,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16294287890195847,
      "num_tokens": 27507040.0,
      "step": 14065
    },
    {
      "entropy": 6.13452000617981,
      "epoch": 1.4064577397910731,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004809377755180858,
      "loss": 5.9439,
      "mean_token_accuracy": 0.1507239207625389,
      "num_tokens": 27516861.0,
      "step": 14070
    },
    {
      "entropy": 6.131076526641846,
      "epoch": 1.4069575648523016,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004809233966260311,
      "loss": 5.8215,
      "mean_token_accuracy": 0.17124989926815032,
      "num_tokens": 27526816.0,
      "step": 14075
    },
    {
      "entropy": 6.061521339416504,
      "epoch": 1.4074573899135303,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004809090125529966,
      "loss": 5.8964,
      "mean_token_accuracy": 0.16105823069810868,
      "num_tokens": 27535984.0,
      "step": 14080
    },
    {
      "entropy": 6.118185091018677,
      "epoch": 1.4079572149747588,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048089462329934437,
      "loss": 5.891,
      "mean_token_accuracy": 0.1569160960614681,
      "num_tokens": 27545515.0,
      "step": 14085
    },
    {
      "entropy": 6.11595606803894,
      "epoch": 1.4084570400359875,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048088022886543636,
      "loss": 5.8288,
      "mean_token_accuracy": 0.16541706919670104,
      "num_tokens": 27554673.0,
      "step": 14090
    },
    {
      "entropy": 6.126456212997437,
      "epoch": 1.408956865097216,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004808658292516348,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15936289727687836,
      "num_tokens": 27564282.0,
      "step": 14095
    },
    {
      "entropy": 6.0902503490447994,
      "epoch": 1.4094566901584447,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048085142445830197,
      "loss": 5.9467,
      "mean_token_accuracy": 0.14794485941529273,
      "num_tokens": 27574205.0,
      "step": 14100
    },
    {
      "entropy": 6.1554711818695065,
      "epoch": 1.4099565152196731,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048083701448580027,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15965884625911714,
      "num_tokens": 27584283.0,
      "step": 14105
    },
    {
      "entropy": 6.162276554107666,
      "epoch": 1.4104563402809016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048082259933449224,
      "loss": 5.8458,
      "mean_token_accuracy": 0.15597057044506074,
      "num_tokens": 27592828.0,
      "step": 14110
    },
    {
      "entropy": 6.058228588104248,
      "epoch": 1.4109561653421303,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004808081790047406,
      "loss": 5.8752,
      "mean_token_accuracy": 0.16809998899698259,
      "num_tokens": 27602412.0,
      "step": 14115
    },
    {
      "entropy": 6.090178298950195,
      "epoch": 1.4114559904033588,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048079375349690807,
      "loss": 5.8666,
      "mean_token_accuracy": 0.1604020744562149,
      "num_tokens": 27611685.0,
      "step": 14120
    },
    {
      "entropy": 6.16297550201416,
      "epoch": 1.4119558154645873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048077932281135775,
      "loss": 5.9028,
      "mean_token_accuracy": 0.15416688323020936,
      "num_tokens": 27621108.0,
      "step": 14125
    },
    {
      "entropy": 6.213150310516357,
      "epoch": 1.412455640525816,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048076488694845257,
      "loss": 5.8289,
      "mean_token_accuracy": 0.16369918435811998,
      "num_tokens": 27631212.0,
      "step": 14130
    },
    {
      "entropy": 6.1659749984741214,
      "epoch": 1.4129554655870447,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048075044590855574,
      "loss": 5.9825,
      "mean_token_accuracy": 0.1577750787138939,
      "num_tokens": 27641071.0,
      "step": 14135
    },
    {
      "entropy": 6.121898078918457,
      "epoch": 1.4134552906482731,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048073599969203064,
      "loss": 5.9818,
      "mean_token_accuracy": 0.1521691471338272,
      "num_tokens": 27650903.0,
      "step": 14140
    },
    {
      "entropy": 6.04588680267334,
      "epoch": 1.4139551157095016,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004807215482992408,
      "loss": 5.8849,
      "mean_token_accuracy": 0.16351685523986817,
      "num_tokens": 27661272.0,
      "step": 14145
    },
    {
      "entropy": 6.114631938934326,
      "epoch": 1.4144549407707303,
      "grad_norm": 2.84375,
      "learning_rate": 0.0004807070917305497,
      "loss": 5.8886,
      "mean_token_accuracy": 0.15800164192914962,
      "num_tokens": 27671042.0,
      "step": 14150
    },
    {
      "entropy": 6.16136212348938,
      "epoch": 1.4149547658319588,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048069262998632115,
      "loss": 5.8976,
      "mean_token_accuracy": 0.1551163136959076,
      "num_tokens": 27680876.0,
      "step": 14155
    },
    {
      "entropy": 6.136555767059326,
      "epoch": 1.4154545908931873,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004806781630669189,
      "loss": 5.8494,
      "mean_token_accuracy": 0.1600144922733307,
      "num_tokens": 27690193.0,
      "step": 14160
    },
    {
      "entropy": 6.103119659423828,
      "epoch": 1.415954415954416,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048066369097270694,
      "loss": 5.8546,
      "mean_token_accuracy": 0.1648570254445076,
      "num_tokens": 27700095.0,
      "step": 14165
    },
    {
      "entropy": 6.066010522842407,
      "epoch": 1.4164542410156447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004806492137040495,
      "loss": 5.9045,
      "mean_token_accuracy": 0.1574258416891098,
      "num_tokens": 27710172.0,
      "step": 14170
    },
    {
      "entropy": 6.123028182983399,
      "epoch": 1.4169540660768731,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048063473126131086,
      "loss": 5.8449,
      "mean_token_accuracy": 0.1633906602859497,
      "num_tokens": 27720241.0,
      "step": 14175
    },
    {
      "entropy": 6.114968347549438,
      "epoch": 1.4174538911381016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004806202436448551,
      "loss": 5.9239,
      "mean_token_accuracy": 0.16329588145017623,
      "num_tokens": 27730959.0,
      "step": 14180
    },
    {
      "entropy": 6.054730319976807,
      "epoch": 1.4179537161993303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004806057508550471,
      "loss": 5.9087,
      "mean_token_accuracy": 0.1636994495987892,
      "num_tokens": 27740690.0,
      "step": 14185
    },
    {
      "entropy": 6.088113307952881,
      "epoch": 1.4184535412605588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004805912528922512,
      "loss": 5.7741,
      "mean_token_accuracy": 0.16418852508068085,
      "num_tokens": 27749538.0,
      "step": 14190
    },
    {
      "entropy": 6.141475582122803,
      "epoch": 1.4189533663217873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004805767497568323,
      "loss": 5.8792,
      "mean_token_accuracy": 0.16102715134620665,
      "num_tokens": 27759150.0,
      "step": 14195
    },
    {
      "entropy": 6.114573431015015,
      "epoch": 1.419453191383016,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048056224144915534,
      "loss": 5.8524,
      "mean_token_accuracy": 0.16008137613534928,
      "num_tokens": 27767150.0,
      "step": 14200
    },
    {
      "entropy": 6.123822212219238,
      "epoch": 1.4199530164442444,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.9192,
      "mean_token_accuracy": 0.15818282216787338,
      "num_tokens": 27776866.0,
      "step": 14205
    },
    {
      "entropy": 6.040418672561645,
      "epoch": 1.4204528415054731,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048053320931848713,
      "loss": 5.7566,
      "mean_token_accuracy": 0.1675146460533142,
      "num_tokens": 27786941.0,
      "step": 14210
    },
    {
      "entropy": 6.164069700241089,
      "epoch": 1.4209526665667016,
      "grad_norm": 1.375,
      "learning_rate": 0.00048051868549622644,
      "loss": 5.9906,
      "mean_token_accuracy": 0.15533746182918548,
      "num_tokens": 27796962.0,
      "step": 14215
    },
    {
      "entropy": 6.0720874786376955,
      "epoch": 1.4214524916279303,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048050415650316853,
      "loss": 5.874,
      "mean_token_accuracy": 0.15893545746803284,
      "num_tokens": 27806609.0,
      "step": 14220
    },
    {
      "entropy": 6.1439048767089846,
      "epoch": 1.4219523166891588,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004804896223396789,
      "loss": 5.9009,
      "mean_token_accuracy": 0.1598852187395096,
      "num_tokens": 27816734.0,
      "step": 14225
    },
    {
      "entropy": 6.1008155822753904,
      "epoch": 1.4224521417503873,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004804750830061232,
      "loss": 5.9537,
      "mean_token_accuracy": 0.15372129231691362,
      "num_tokens": 27826686.0,
      "step": 14230
    },
    {
      "entropy": 6.086089992523194,
      "epoch": 1.422951966811616,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004804605385028673,
      "loss": 5.872,
      "mean_token_accuracy": 0.15534260720014573,
      "num_tokens": 27836193.0,
      "step": 14235
    },
    {
      "entropy": 6.078110647201538,
      "epoch": 1.4234517918728444,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004804459888302772,
      "loss": 5.8257,
      "mean_token_accuracy": 0.16841569989919664,
      "num_tokens": 27845175.0,
      "step": 14240
    },
    {
      "entropy": 6.154541492462158,
      "epoch": 1.4239516169340731,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048043143398871874,
      "loss": 5.873,
      "mean_token_accuracy": 0.16193023547530175,
      "num_tokens": 27853829.0,
      "step": 14245
    },
    {
      "entropy": 6.108955764770508,
      "epoch": 1.4244514419953016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004804168739785583,
      "loss": 5.917,
      "mean_token_accuracy": 0.15816067904233932,
      "num_tokens": 27863468.0,
      "step": 14250
    },
    {
      "entropy": 6.1031813621521,
      "epoch": 1.4249512670565303,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004804023088001623,
      "loss": 5.9157,
      "mean_token_accuracy": 0.15397956296801568,
      "num_tokens": 27873565.0,
      "step": 14255
    },
    {
      "entropy": 6.197043704986572,
      "epoch": 1.4254510921177588,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004803877384538969,
      "loss": 6.0222,
      "mean_token_accuracy": 0.1524301737546921,
      "num_tokens": 27883687.0,
      "step": 14260
    },
    {
      "entropy": 6.163683080673218,
      "epoch": 1.4259509171789873,
      "grad_norm": 1.40625,
      "learning_rate": 0.000480373162940129,
      "loss": 6.0099,
      "mean_token_accuracy": 0.15164605155587196,
      "num_tokens": 27893702.0,
      "step": 14265
    },
    {
      "entropy": 6.169226217269897,
      "epoch": 1.426450742240216,
      "grad_norm": 1.515625,
      "learning_rate": 0.00048035858225922506,
      "loss": 5.8636,
      "mean_token_accuracy": 0.16417737379670144,
      "num_tokens": 27902809.0,
      "step": 14270
    },
    {
      "entropy": 6.2560203075408936,
      "epoch": 1.4269505673014444,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048034399641155206,
      "loss": 5.9527,
      "mean_token_accuracy": 0.15579653903841972,
      "num_tokens": 27912810.0,
      "step": 14275
    },
    {
      "entropy": 6.141145038604736,
      "epoch": 1.4274503923626731,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048032940539747697,
      "loss": 5.9258,
      "mean_token_accuracy": 0.1533887580037117,
      "num_tokens": 27922402.0,
      "step": 14280
    },
    {
      "entropy": 6.125055885314941,
      "epoch": 1.4279502174239016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048031480921736686,
      "loss": 5.9309,
      "mean_token_accuracy": 0.1620784044265747,
      "num_tokens": 27932379.0,
      "step": 14285
    },
    {
      "entropy": 6.204127550125122,
      "epoch": 1.4284500424851303,
      "grad_norm": 1.34375,
      "learning_rate": 0.000480300207871589,
      "loss": 6.1004,
      "mean_token_accuracy": 0.149103607237339,
      "num_tokens": 27942994.0,
      "step": 14290
    },
    {
      "entropy": 6.21116828918457,
      "epoch": 1.4289498675463588,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048028560136051076,
      "loss": 5.906,
      "mean_token_accuracy": 0.16192933470010756,
      "num_tokens": 27952860.0,
      "step": 14295
    },
    {
      "entropy": 6.133937358856201,
      "epoch": 1.4294496926075873,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048027098968449954,
      "loss": 5.9187,
      "mean_token_accuracy": 0.15826760977506638,
      "num_tokens": 27962435.0,
      "step": 14300
    },
    {
      "entropy": 6.136843585968018,
      "epoch": 1.429949517668816,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004802563728439231,
      "loss": 5.89,
      "mean_token_accuracy": 0.156623674929142,
      "num_tokens": 27971889.0,
      "step": 14305
    },
    {
      "entropy": 6.181384181976318,
      "epoch": 1.4304493427300444,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004802417508391491,
      "loss": 5.9487,
      "mean_token_accuracy": 0.1524609625339508,
      "num_tokens": 27982683.0,
      "step": 14310
    },
    {
      "entropy": 6.091646146774292,
      "epoch": 1.4309491677912731,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004802271236705455,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16464757025241852,
      "num_tokens": 27992408.0,
      "step": 14315
    },
    {
      "entropy": 6.248935651779175,
      "epoch": 1.4314489928525016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048021249133848025,
      "loss": 5.9516,
      "mean_token_accuracy": 0.1558517411351204,
      "num_tokens": 28002592.0,
      "step": 14320
    },
    {
      "entropy": 6.182414722442627,
      "epoch": 1.4319488179137303,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048019785384332154,
      "loss": 5.9342,
      "mean_token_accuracy": 0.15491797626018525,
      "num_tokens": 28012148.0,
      "step": 14325
    },
    {
      "entropy": 6.135596418380738,
      "epoch": 1.4324486429749588,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004801832111854376,
      "loss": 5.9387,
      "mean_token_accuracy": 0.15597166270017623,
      "num_tokens": 28021341.0,
      "step": 14330
    },
    {
      "entropy": 6.1069506168365475,
      "epoch": 1.4329484680361873,
      "grad_norm": 1.25,
      "learning_rate": 0.0004801685633651969,
      "loss": 5.8669,
      "mean_token_accuracy": 0.1626984417438507,
      "num_tokens": 28030721.0,
      "step": 14335
    },
    {
      "entropy": 6.071425867080689,
      "epoch": 1.433448293097416,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004801539103829679,
      "loss": 5.7878,
      "mean_token_accuracy": 0.16495172679424286,
      "num_tokens": 28040665.0,
      "step": 14340
    },
    {
      "entropy": 6.261182117462158,
      "epoch": 1.4339481181586444,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004801392522391193,
      "loss": 6.0081,
      "mean_token_accuracy": 0.14548941031098367,
      "num_tokens": 28052307.0,
      "step": 14345
    },
    {
      "entropy": 6.13683705329895,
      "epoch": 1.4344479432198731,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048012458893401995,
      "loss": 5.8715,
      "mean_token_accuracy": 0.16539184451103212,
      "num_tokens": 28061347.0,
      "step": 14350
    },
    {
      "entropy": 6.102767276763916,
      "epoch": 1.4349477682811016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004801099204680387,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15780195146799086,
      "num_tokens": 28072088.0,
      "step": 14355
    },
    {
      "entropy": 6.16151270866394,
      "epoch": 1.4354475933423303,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048009524684154464,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15768671035766602,
      "num_tokens": 28081674.0,
      "step": 14360
    },
    {
      "entropy": 6.093067741394043,
      "epoch": 1.4359474184035588,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004800805680549069,
      "loss": 5.8505,
      "mean_token_accuracy": 0.17006828263401985,
      "num_tokens": 28091530.0,
      "step": 14365
    },
    {
      "entropy": 6.050295686721801,
      "epoch": 1.4364472434647872,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048006588410849497,
      "loss": 5.8663,
      "mean_token_accuracy": 0.15799159109592437,
      "num_tokens": 28102322.0,
      "step": 14370
    },
    {
      "entropy": 6.1337967872619625,
      "epoch": 1.436947068526016,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000480051195002678,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16182428151369094,
      "num_tokens": 28112041.0,
      "step": 14375
    },
    {
      "entropy": 6.195472574234008,
      "epoch": 1.4374468935872444,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048003650073782586,
      "loss": 5.9334,
      "mean_token_accuracy": 0.14942919090390205,
      "num_tokens": 28122451.0,
      "step": 14380
    },
    {
      "entropy": 6.160152578353882,
      "epoch": 1.437946718648473,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000480021801314308,
      "loss": 5.8666,
      "mean_token_accuracy": 0.16077100485563278,
      "num_tokens": 28131949.0,
      "step": 14385
    },
    {
      "entropy": 6.091319894790649,
      "epoch": 1.4384465437097016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004800070967324944,
      "loss": 5.9089,
      "mean_token_accuracy": 0.15270036458969116,
      "num_tokens": 28141905.0,
      "step": 14390
    },
    {
      "entropy": 6.042965030670166,
      "epoch": 1.4389463687709303,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047999238699275506,
      "loss": 5.8768,
      "mean_token_accuracy": 0.15576011687517166,
      "num_tokens": 28150735.0,
      "step": 14395
    },
    {
      "entropy": 6.117394685745239,
      "epoch": 1.4394461938321588,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047997767209546,
      "loss": 5.8765,
      "mean_token_accuracy": 0.1626116469502449,
      "num_tokens": 28160872.0,
      "step": 14400
    },
    {
      "entropy": 6.100717544555664,
      "epoch": 1.4399460188933872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004799629520409795,
      "loss": 5.9259,
      "mean_token_accuracy": 0.1554840847849846,
      "num_tokens": 28170873.0,
      "step": 14405
    },
    {
      "entropy": 6.190504884719848,
      "epoch": 1.440445843954616,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047994822682968373,
      "loss": 5.8919,
      "mean_token_accuracy": 0.15975164324045182,
      "num_tokens": 28180637.0,
      "step": 14410
    },
    {
      "entropy": 6.168624019622802,
      "epoch": 1.4409456690158444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047993349646194343,
      "loss": 5.9248,
      "mean_token_accuracy": 0.15888790041208267,
      "num_tokens": 28189326.0,
      "step": 14415
    },
    {
      "entropy": 6.076530647277832,
      "epoch": 1.441445494077073,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047991876093812904,
      "loss": 5.8342,
      "mean_token_accuracy": 0.16267921775579453,
      "num_tokens": 28198729.0,
      "step": 14420
    },
    {
      "entropy": 6.063698673248291,
      "epoch": 1.4419453191383016,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047990402025861143,
      "loss": 5.807,
      "mean_token_accuracy": 0.16690940409898758,
      "num_tokens": 28207525.0,
      "step": 14425
    },
    {
      "entropy": 6.141130495071411,
      "epoch": 1.4424451441995303,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004798892744237613,
      "loss": 6.0127,
      "mean_token_accuracy": 0.1538618892431259,
      "num_tokens": 28218176.0,
      "step": 14430
    },
    {
      "entropy": 6.265495729446411,
      "epoch": 1.4429449692607588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047987452343394976,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15929054766893386,
      "num_tokens": 28229207.0,
      "step": 14435
    },
    {
      "entropy": 6.186412906646728,
      "epoch": 1.4434447943219872,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047985976728954794,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16019906401634215,
      "num_tokens": 28238124.0,
      "step": 14440
    },
    {
      "entropy": 6.052690029144287,
      "epoch": 1.443944619383216,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004798450059909271,
      "loss": 5.9488,
      "mean_token_accuracy": 0.14833700954914092,
      "num_tokens": 28248359.0,
      "step": 14445
    },
    {
      "entropy": 6.141966485977173,
      "epoch": 1.4444444444444444,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004798302395384586,
      "loss": 5.9327,
      "mean_token_accuracy": 0.15207404568791388,
      "num_tokens": 28258200.0,
      "step": 14450
    },
    {
      "entropy": 6.189845085144043,
      "epoch": 1.444944269505673,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047981546793251405,
      "loss": 5.9257,
      "mean_token_accuracy": 0.15451301634311676,
      "num_tokens": 28268512.0,
      "step": 14455
    },
    {
      "entropy": 6.122415351867676,
      "epoch": 1.4454440945669016,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000479800691173465,
      "loss": 5.8756,
      "mean_token_accuracy": 0.16211687475442887,
      "num_tokens": 28278657.0,
      "step": 14460
    },
    {
      "entropy": 6.0825906753540036,
      "epoch": 1.4459439196281303,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004797859092616832,
      "loss": 5.9478,
      "mean_token_accuracy": 0.1584588184952736,
      "num_tokens": 28289104.0,
      "step": 14465
    },
    {
      "entropy": 6.115813255310059,
      "epoch": 1.4464437446893588,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047977112219754064,
      "loss": 5.8026,
      "mean_token_accuracy": 0.1676919713616371,
      "num_tokens": 28299110.0,
      "step": 14470
    },
    {
      "entropy": 6.100197172164917,
      "epoch": 1.4469435697505872,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004797563299814094,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16809366196393966,
      "num_tokens": 28309350.0,
      "step": 14475
    },
    {
      "entropy": 6.018046951293945,
      "epoch": 1.447443394811816,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047974153261366156,
      "loss": 5.7332,
      "mean_token_accuracy": 0.1696474939584732,
      "num_tokens": 28319080.0,
      "step": 14480
    },
    {
      "entropy": 6.187850093841552,
      "epoch": 1.4479432198730444,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004797267300946694,
      "loss": 5.894,
      "mean_token_accuracy": 0.15671452209353448,
      "num_tokens": 28328853.0,
      "step": 14485
    },
    {
      "entropy": 6.162902355194092,
      "epoch": 1.448443044934273,
      "grad_norm": 1.375,
      "learning_rate": 0.00047971192242480544,
      "loss": 5.8905,
      "mean_token_accuracy": 0.15962706357240677,
      "num_tokens": 28339499.0,
      "step": 14490
    },
    {
      "entropy": 6.119480991363526,
      "epoch": 1.4489428699955016,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004796971096044422,
      "loss": 5.9047,
      "mean_token_accuracy": 0.16110700815916063,
      "num_tokens": 28350068.0,
      "step": 14495
    },
    {
      "entropy": 6.133473014831543,
      "epoch": 1.44944269505673,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004796822916339523,
      "loss": 5.9553,
      "mean_token_accuracy": 0.15094479024410248,
      "num_tokens": 28360912.0,
      "step": 14500
    },
    {
      "entropy": 6.185051965713501,
      "epoch": 1.4499425201179588,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047966746851370865,
      "loss": 5.9286,
      "mean_token_accuracy": 0.15161790400743486,
      "num_tokens": 28370094.0,
      "step": 14505
    },
    {
      "entropy": 6.108567333221435,
      "epoch": 1.4504423451791872,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004796526402440841,
      "loss": 5.8905,
      "mean_token_accuracy": 0.1640714354813099,
      "num_tokens": 28380520.0,
      "step": 14510
    },
    {
      "entropy": 6.1056968688964846,
      "epoch": 1.450942170240416,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004796378068254518,
      "loss": 5.9497,
      "mean_token_accuracy": 0.1638961911201477,
      "num_tokens": 28390406.0,
      "step": 14515
    },
    {
      "entropy": 6.125859928131104,
      "epoch": 1.4514419953016444,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000479622968258185,
      "loss": 5.88,
      "mean_token_accuracy": 0.16428112536668776,
      "num_tokens": 28399935.0,
      "step": 14520
    },
    {
      "entropy": 6.123615646362305,
      "epoch": 1.451941820362873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004796081245426569,
      "loss": 5.8577,
      "mean_token_accuracy": 0.162043759226799,
      "num_tokens": 28409167.0,
      "step": 14525
    },
    {
      "entropy": 6.156009531021118,
      "epoch": 1.4524416454241016,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047959327567924115,
      "loss": 5.9504,
      "mean_token_accuracy": 0.15257749408483506,
      "num_tokens": 28419093.0,
      "step": 14530
    },
    {
      "entropy": 6.100450801849365,
      "epoch": 1.45294147048533,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047957842166831103,
      "loss": 5.8667,
      "mean_token_accuracy": 0.16456527709960939,
      "num_tokens": 28428803.0,
      "step": 14535
    },
    {
      "entropy": 6.126444959640503,
      "epoch": 1.4534412955465588,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047956356251024066,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16712221652269363,
      "num_tokens": 28439724.0,
      "step": 14540
    },
    {
      "entropy": 6.131212949752808,
      "epoch": 1.4539411206077872,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004795486982054036,
      "loss": 5.8315,
      "mean_token_accuracy": 0.1698771074414253,
      "num_tokens": 28448450.0,
      "step": 14545
    },
    {
      "entropy": 6.097634601593017,
      "epoch": 1.454440945669016,
      "grad_norm": 1.328125,
      "learning_rate": 0.000479533828754174,
      "loss": 5.9528,
      "mean_token_accuracy": 0.15638633519411088,
      "num_tokens": 28458862.0,
      "step": 14550
    },
    {
      "entropy": 6.167285346984864,
      "epoch": 1.4549407707302444,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004795189541569258,
      "loss": 6.0285,
      "mean_token_accuracy": 0.145337051153183,
      "num_tokens": 28469602.0,
      "step": 14555
    },
    {
      "entropy": 6.153655481338501,
      "epoch": 1.455440595791473,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047950407441403343,
      "loss": 5.861,
      "mean_token_accuracy": 0.1644306242465973,
      "num_tokens": 28480635.0,
      "step": 14560
    },
    {
      "entropy": 6.178288269042969,
      "epoch": 1.4559404208527016,
      "grad_norm": 1.25,
      "learning_rate": 0.0004794891895258711,
      "loss": 5.8784,
      "mean_token_accuracy": 0.16245242357254028,
      "num_tokens": 28490471.0,
      "step": 14565
    },
    {
      "entropy": 6.037000226974487,
      "epoch": 1.45644024591393,
      "grad_norm": 2.078125,
      "learning_rate": 0.0004794742994928134,
      "loss": 5.7544,
      "mean_token_accuracy": 0.17279160767793655,
      "num_tokens": 28499595.0,
      "step": 14570
    },
    {
      "entropy": 6.106502628326416,
      "epoch": 1.4569400709751588,
      "grad_norm": 1.34375,
      "learning_rate": 0.000479459404315235,
      "loss": 5.8754,
      "mean_token_accuracy": 0.15996685326099397,
      "num_tokens": 28508278.0,
      "step": 14575
    },
    {
      "entropy": 6.09531536102295,
      "epoch": 1.4574398960363872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047944450399351054,
      "loss": 5.921,
      "mean_token_accuracy": 0.15202573984861373,
      "num_tokens": 28518286.0,
      "step": 14580
    },
    {
      "entropy": 6.202413272857666,
      "epoch": 1.457939721097616,
      "grad_norm": 1.359375,
      "learning_rate": 0.000479429598528015,
      "loss": 5.9358,
      "mean_token_accuracy": 0.15157345831394195,
      "num_tokens": 28527517.0,
      "step": 14585
    },
    {
      "entropy": 6.320925617218018,
      "epoch": 1.4584395461588444,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004794146879191233,
      "loss": 6.1341,
      "mean_token_accuracy": 0.14534393027424813,
      "num_tokens": 28538521.0,
      "step": 14590
    },
    {
      "entropy": 6.143481159210205,
      "epoch": 1.458939371220073,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004793997721672107,
      "loss": 5.8866,
      "mean_token_accuracy": 0.15985874235630035,
      "num_tokens": 28547696.0,
      "step": 14595
    },
    {
      "entropy": 6.076105260848999,
      "epoch": 1.4594391962813016,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047938485127265243,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1639146849513054,
      "num_tokens": 28558083.0,
      "step": 14600
    },
    {
      "entropy": 6.13893494606018,
      "epoch": 1.45993902134253,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004793699252358239,
      "loss": 5.8516,
      "mean_token_accuracy": 0.16429237872362137,
      "num_tokens": 28568453.0,
      "step": 14605
    },
    {
      "entropy": 6.099099826812744,
      "epoch": 1.4604388464037588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004793549940571006,
      "loss": 5.8797,
      "mean_token_accuracy": 0.15845344662666322,
      "num_tokens": 28578401.0,
      "step": 14610
    },
    {
      "entropy": 6.168532943725586,
      "epoch": 1.4609386714649872,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00047934005773685824,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1603742927312851,
      "num_tokens": 28587717.0,
      "step": 14615
    },
    {
      "entropy": 6.152024602890014,
      "epoch": 1.461438496526216,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047932511627547266,
      "loss": 5.9056,
      "mean_token_accuracy": 0.15554160475730897,
      "num_tokens": 28597609.0,
      "step": 14620
    },
    {
      "entropy": 6.091575574874878,
      "epoch": 1.4619383215874444,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004793101696733197,
      "loss": 5.836,
      "mean_token_accuracy": 0.15883092060685158,
      "num_tokens": 28606742.0,
      "step": 14625
    },
    {
      "entropy": 6.069307565689087,
      "epoch": 1.4624381466486729,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047929521793077545,
      "loss": 5.847,
      "mean_token_accuracy": 0.1595710977911949,
      "num_tokens": 28615822.0,
      "step": 14630
    },
    {
      "entropy": 6.141051483154297,
      "epoch": 1.4629379717099016,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004792802610482161,
      "loss": 5.9507,
      "mean_token_accuracy": 0.15661249309778214,
      "num_tokens": 28625372.0,
      "step": 14635
    },
    {
      "entropy": 6.160237836837768,
      "epoch": 1.46343779677113,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004792652990260179,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15853878557682038,
      "num_tokens": 28635329.0,
      "step": 14640
    },
    {
      "entropy": 6.156243562698364,
      "epoch": 1.4639376218323585,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004792503318645574,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1580753430724144,
      "num_tokens": 28645534.0,
      "step": 14645
    },
    {
      "entropy": 6.183218097686767,
      "epoch": 1.4644374468935872,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047923535956421104,
      "loss": 5.9757,
      "mean_token_accuracy": 0.15771511644124986,
      "num_tokens": 28657207.0,
      "step": 14650
    },
    {
      "entropy": 6.152878522872925,
      "epoch": 1.464937271954816,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047922038212535565,
      "loss": 5.9162,
      "mean_token_accuracy": 0.16059714332222938,
      "num_tokens": 28667260.0,
      "step": 14655
    },
    {
      "entropy": 6.104574346542359,
      "epoch": 1.4654370970160444,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047920539954836793,
      "loss": 5.802,
      "mean_token_accuracy": 0.1603698745369911,
      "num_tokens": 28677255.0,
      "step": 14660
    },
    {
      "entropy": 6.128570699691773,
      "epoch": 1.4659369220772729,
      "grad_norm": 1.28125,
      "learning_rate": 0.000479190411833625,
      "loss": 5.8641,
      "mean_token_accuracy": 0.16825679391622544,
      "num_tokens": 28687136.0,
      "step": 14665
    },
    {
      "entropy": 6.177998495101929,
      "epoch": 1.4664367471385016,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047917541898150377,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15842775702476503,
      "num_tokens": 28697499.0,
      "step": 14670
    },
    {
      "entropy": 6.186580085754395,
      "epoch": 1.46693657219973,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047916042099238154,
      "loss": 6.1089,
      "mean_token_accuracy": 0.14769342616200448,
      "num_tokens": 28706843.0,
      "step": 14675
    },
    {
      "entropy": 6.1012156963348385,
      "epoch": 1.4674363972609585,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047914541786663566,
      "loss": 5.8704,
      "mean_token_accuracy": 0.1633995294570923,
      "num_tokens": 28715676.0,
      "step": 14680
    },
    {
      "entropy": 6.072559833526611,
      "epoch": 1.4679362223221872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047913040960464363,
      "loss": 5.8346,
      "mean_token_accuracy": 0.15809477791190146,
      "num_tokens": 28725714.0,
      "step": 14685
    },
    {
      "entropy": 6.0916643142700195,
      "epoch": 1.468436047383416,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047911539620678305,
      "loss": 5.875,
      "mean_token_accuracy": 0.1569311425089836,
      "num_tokens": 28734928.0,
      "step": 14690
    },
    {
      "entropy": 6.171659898757935,
      "epoch": 1.4689358724446444,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004791003776734316,
      "loss": 5.8656,
      "mean_token_accuracy": 0.15570638626813887,
      "num_tokens": 28743999.0,
      "step": 14695
    },
    {
      "entropy": 6.134366369247436,
      "epoch": 1.4694356975058729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047908535400496713,
      "loss": 5.8483,
      "mean_token_accuracy": 0.16599692404270172,
      "num_tokens": 28753220.0,
      "step": 14700
    },
    {
      "entropy": 6.090525579452515,
      "epoch": 1.4699355225671016,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047907032520176765,
      "loss": 5.8864,
      "mean_token_accuracy": 0.15720940977334977,
      "num_tokens": 28762525.0,
      "step": 14705
    },
    {
      "entropy": 6.059252452850342,
      "epoch": 1.47043534762833,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004790552912642113,
      "loss": 5.8353,
      "mean_token_accuracy": 0.1640519082546234,
      "num_tokens": 28772749.0,
      "step": 14710
    },
    {
      "entropy": 6.050037050247193,
      "epoch": 1.4709351726895585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047904025219267643,
      "loss": 5.8332,
      "mean_token_accuracy": 0.16434992104768753,
      "num_tokens": 28782685.0,
      "step": 14715
    },
    {
      "entropy": 6.166154241561889,
      "epoch": 1.4714349977507872,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047902520798754125,
      "loss": 5.9816,
      "mean_token_accuracy": 0.15362559258937836,
      "num_tokens": 28794172.0,
      "step": 14720
    },
    {
      "entropy": 6.131908226013183,
      "epoch": 1.471934822812016,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004790101586491843,
      "loss": 5.9475,
      "mean_token_accuracy": 0.1498093254864216,
      "num_tokens": 28805513.0,
      "step": 14725
    },
    {
      "entropy": 6.154462909698486,
      "epoch": 1.4724346478732444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004789951041779843,
      "loss": 5.9088,
      "mean_token_accuracy": 0.16052107661962509,
      "num_tokens": 28815737.0,
      "step": 14730
    },
    {
      "entropy": 6.107979440689087,
      "epoch": 1.4729344729344729,
      "grad_norm": 1.375,
      "learning_rate": 0.00047898004457432,
      "loss": 5.8922,
      "mean_token_accuracy": 0.15878304541110994,
      "num_tokens": 28824901.0,
      "step": 14735
    },
    {
      "entropy": 6.10954098701477,
      "epoch": 1.4734342979957016,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004789649798385702,
      "loss": 5.8347,
      "mean_token_accuracy": 0.16197631508111954,
      "num_tokens": 28833625.0,
      "step": 14740
    },
    {
      "entropy": 5.996745204925537,
      "epoch": 1.47393412305693,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047894990997111403,
      "loss": 5.7791,
      "mean_token_accuracy": 0.17287053912878036,
      "num_tokens": 28843084.0,
      "step": 14745
    },
    {
      "entropy": 6.07992000579834,
      "epoch": 1.4744339481181585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047893483497233057,
      "loss": 5.7964,
      "mean_token_accuracy": 0.16760654896497726,
      "num_tokens": 28852320.0,
      "step": 14750
    },
    {
      "entropy": 6.110156774520874,
      "epoch": 1.4749337731793872,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004789197548425992,
      "loss": 5.8456,
      "mean_token_accuracy": 0.16678748577833175,
      "num_tokens": 28861379.0,
      "step": 14755
    },
    {
      "entropy": 6.0724061012268065,
      "epoch": 1.4754335982406157,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004789046695822992,
      "loss": 5.8585,
      "mean_token_accuracy": 0.16366672813892363,
      "num_tokens": 28871181.0,
      "step": 14760
    },
    {
      "entropy": 6.158563184738159,
      "epoch": 1.4759334233018444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047888957919181026,
      "loss": 5.902,
      "mean_token_accuracy": 0.15973577797412872,
      "num_tokens": 28880763.0,
      "step": 14765
    },
    {
      "entropy": 6.062250709533691,
      "epoch": 1.4764332483630729,
      "grad_norm": 1.375,
      "learning_rate": 0.0004788744836715119,
      "loss": 5.8151,
      "mean_token_accuracy": 0.17094155251979828,
      "num_tokens": 28889876.0,
      "step": 14770
    },
    {
      "entropy": 6.151125764846801,
      "epoch": 1.4769330734243016,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004788593830217841,
      "loss": 5.8797,
      "mean_token_accuracy": 0.1579310953617096,
      "num_tokens": 28899765.0,
      "step": 14775
    },
    {
      "entropy": 6.148186206817627,
      "epoch": 1.47743289848553,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047884427724300657,
      "loss": 5.9158,
      "mean_token_accuracy": 0.151596137881279,
      "num_tokens": 28909755.0,
      "step": 14780
    },
    {
      "entropy": 6.095546054840088,
      "epoch": 1.4779327235467585,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004788291663355595,
      "loss": 5.871,
      "mean_token_accuracy": 0.1662254348397255,
      "num_tokens": 28919368.0,
      "step": 14785
    },
    {
      "entropy": 6.12842869758606,
      "epoch": 1.4784325486079872,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004788140502998231,
      "loss": 5.904,
      "mean_token_accuracy": 0.1611200749874115,
      "num_tokens": 28930076.0,
      "step": 14790
    },
    {
      "entropy": 6.177410507202149,
      "epoch": 1.4789323736692157,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004787989291361776,
      "loss": 5.9049,
      "mean_token_accuracy": 0.16091834157705306,
      "num_tokens": 28939555.0,
      "step": 14795
    },
    {
      "entropy": 6.1310876369476315,
      "epoch": 1.4794321987304444,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004787838028450035,
      "loss": 5.9557,
      "mean_token_accuracy": 0.1496977798640728,
      "num_tokens": 28951084.0,
      "step": 14800
    },
    {
      "entropy": 6.164492082595825,
      "epoch": 1.4799320237916729,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047876867142668136,
      "loss": 5.9217,
      "mean_token_accuracy": 0.15498317629098893,
      "num_tokens": 28960798.0,
      "step": 14805
    },
    {
      "entropy": 6.141341876983643,
      "epoch": 1.4804318488529016,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004787535348815918,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16298890560865403,
      "num_tokens": 28970543.0,
      "step": 14810
    },
    {
      "entropy": 6.185896587371826,
      "epoch": 1.48093167391413,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047873839321011587,
      "loss": 5.9908,
      "mean_token_accuracy": 0.15114522054791452,
      "num_tokens": 28979863.0,
      "step": 14815
    },
    {
      "entropy": 6.105471897125244,
      "epoch": 1.4814314989753585,
      "grad_norm": 1.75,
      "learning_rate": 0.0004787232464126343,
      "loss": 5.8445,
      "mean_token_accuracy": 0.16736557930707932,
      "num_tokens": 28989774.0,
      "step": 14820
    },
    {
      "entropy": 6.152853298187256,
      "epoch": 1.4819313240365872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004787080944895283,
      "loss": 5.9844,
      "mean_token_accuracy": 0.14713040739297867,
      "num_tokens": 29000608.0,
      "step": 14825
    },
    {
      "entropy": 6.165623378753662,
      "epoch": 1.4824311490978157,
      "grad_norm": 1.375,
      "learning_rate": 0.00047869293744117906,
      "loss": 5.9879,
      "mean_token_accuracy": 0.15227052494883536,
      "num_tokens": 29010977.0,
      "step": 14830
    },
    {
      "entropy": 6.211656808853149,
      "epoch": 1.4829309741590444,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047867777526796793,
      "loss": 5.8582,
      "mean_token_accuracy": 0.15856895893812178,
      "num_tokens": 29021868.0,
      "step": 14835
    },
    {
      "entropy": 6.146788454055786,
      "epoch": 1.4834307992202729,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004786626079702764,
      "loss": 5.893,
      "mean_token_accuracy": 0.160254168510437,
      "num_tokens": 29031878.0,
      "step": 14840
    },
    {
      "entropy": 6.1923610210418705,
      "epoch": 1.4839306242815016,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004786474355484861,
      "loss": 6.0051,
      "mean_token_accuracy": 0.15587825253605841,
      "num_tokens": 29041862.0,
      "step": 14845
    },
    {
      "entropy": 6.077951288223266,
      "epoch": 1.48443044934273,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004786322580029786,
      "loss": 5.9209,
      "mean_token_accuracy": 0.16161490827798844,
      "num_tokens": 29052534.0,
      "step": 14850
    },
    {
      "entropy": 6.08109483718872,
      "epoch": 1.4849302744039585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047861707533413593,
      "loss": 5.838,
      "mean_token_accuracy": 0.16527173668146133,
      "num_tokens": 29061958.0,
      "step": 14855
    },
    {
      "entropy": 6.054335260391236,
      "epoch": 1.4854300994651872,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004786018875423401,
      "loss": 5.7241,
      "mean_token_accuracy": 0.1656826339662075,
      "num_tokens": 29071816.0,
      "step": 14860
    },
    {
      "entropy": 6.073173093795776,
      "epoch": 1.4859299245264157,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047858669462797317,
      "loss": 5.8301,
      "mean_token_accuracy": 0.17052276134490968,
      "num_tokens": 29081198.0,
      "step": 14865
    },
    {
      "entropy": 6.126138162612915,
      "epoch": 1.4864297495876444,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047857149659141736,
      "loss": 5.9726,
      "mean_token_accuracy": 0.15358896553516388,
      "num_tokens": 29091197.0,
      "step": 14870
    },
    {
      "entropy": 6.173553800582885,
      "epoch": 1.4869295746488729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004785562934330551,
      "loss": 5.9189,
      "mean_token_accuracy": 0.1611097350716591,
      "num_tokens": 29100386.0,
      "step": 14875
    },
    {
      "entropy": 6.180949783325195,
      "epoch": 1.4874293997101016,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047854108515326885,
      "loss": 5.8788,
      "mean_token_accuracy": 0.15916969627141953,
      "num_tokens": 29110056.0,
      "step": 14880
    },
    {
      "entropy": 6.086675214767456,
      "epoch": 1.48792922477133,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004785258717524412,
      "loss": 5.7272,
      "mean_token_accuracy": 0.1751652330160141,
      "num_tokens": 29119157.0,
      "step": 14885
    },
    {
      "entropy": 6.157320880889893,
      "epoch": 1.4884290498325585,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047851065323095514,
      "loss": 5.8675,
      "mean_token_accuracy": 0.15782440900802613,
      "num_tokens": 29129075.0,
      "step": 14890
    },
    {
      "entropy": 6.153360223770141,
      "epoch": 1.4889288748937872,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004784954295891933,
      "loss": 5.9187,
      "mean_token_accuracy": 0.16031213849782944,
      "num_tokens": 29139026.0,
      "step": 14895
    },
    {
      "entropy": 6.07058424949646,
      "epoch": 1.4894286999550157,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047848020082753883,
      "loss": 5.8006,
      "mean_token_accuracy": 0.16586316227912903,
      "num_tokens": 29149315.0,
      "step": 14900
    },
    {
      "entropy": 6.060660696029663,
      "epoch": 1.4899285250162442,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004784649669463749,
      "loss": 5.8814,
      "mean_token_accuracy": 0.15292759537696837,
      "num_tokens": 29158916.0,
      "step": 14905
    },
    {
      "entropy": 6.1569701671600345,
      "epoch": 1.4904283500774729,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047844972794608474,
      "loss": 5.866,
      "mean_token_accuracy": 0.16034657657146453,
      "num_tokens": 29166844.0,
      "step": 14910
    },
    {
      "entropy": 6.139120054244995,
      "epoch": 1.4909281751387016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047843448382705185,
      "loss": 5.8412,
      "mean_token_accuracy": 0.17186597883701324,
      "num_tokens": 29176057.0,
      "step": 14915
    },
    {
      "entropy": 6.081025743484497,
      "epoch": 1.49142800019993,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047841923458965966,
      "loss": 5.8835,
      "mean_token_accuracy": 0.1620857909321785,
      "num_tokens": 29186212.0,
      "step": 14920
    },
    {
      "entropy": 6.0858300685882565,
      "epoch": 1.4919278252611585,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047840398023429176,
      "loss": 5.8267,
      "mean_token_accuracy": 0.1655636504292488,
      "num_tokens": 29196138.0,
      "step": 14925
    },
    {
      "entropy": 6.222097778320313,
      "epoch": 1.4924276503223872,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047838872076133213,
      "loss": 5.9188,
      "mean_token_accuracy": 0.1556027799844742,
      "num_tokens": 29205979.0,
      "step": 14930
    },
    {
      "entropy": 6.172806978225708,
      "epoch": 1.4929274753836157,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004783734561711647,
      "loss": 5.9175,
      "mean_token_accuracy": 0.15185590386390685,
      "num_tokens": 29214776.0,
      "step": 14935
    },
    {
      "entropy": 6.0403636455535885,
      "epoch": 1.4934273004448442,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047835818646417333,
      "loss": 5.7957,
      "mean_token_accuracy": 0.16428822427988052,
      "num_tokens": 29223823.0,
      "step": 14940
    },
    {
      "entropy": 6.085786485671997,
      "epoch": 1.4939271255060729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004783429116407423,
      "loss": 5.855,
      "mean_token_accuracy": 0.16285622119903564,
      "num_tokens": 29232974.0,
      "step": 14945
    },
    {
      "entropy": 6.148810863494873,
      "epoch": 1.4944269505673016,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047832763170125605,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16319494247436522,
      "num_tokens": 29241773.0,
      "step": 14950
    },
    {
      "entropy": 6.124076700210571,
      "epoch": 1.49492677562853,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047831234664609887,
      "loss": 5.809,
      "mean_token_accuracy": 0.15870502442121506,
      "num_tokens": 29251490.0,
      "step": 14955
    },
    {
      "entropy": 6.009474468231201,
      "epoch": 1.4954266006897585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047829705647565535,
      "loss": 5.8062,
      "mean_token_accuracy": 0.171224544942379,
      "num_tokens": 29260338.0,
      "step": 14960
    },
    {
      "entropy": 6.074115943908692,
      "epoch": 1.4959264257509872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004782817611903102,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16277654618024825,
      "num_tokens": 29270192.0,
      "step": 14965
    },
    {
      "entropy": 6.192416572570801,
      "epoch": 1.4964262508122157,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047826646079044826,
      "loss": 5.9724,
      "mean_token_accuracy": 0.15793358758091927,
      "num_tokens": 29281413.0,
      "step": 14970
    },
    {
      "entropy": 6.2376172065734865,
      "epoch": 1.4969260758734442,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047825115527645447,
      "loss": 5.9259,
      "mean_token_accuracy": 0.1583219900727272,
      "num_tokens": 29291959.0,
      "step": 14975
    },
    {
      "entropy": 6.147307872772217,
      "epoch": 1.4974259009346729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004782358446487139,
      "loss": 5.9117,
      "mean_token_accuracy": 0.16535613387823106,
      "num_tokens": 29302577.0,
      "step": 14980
    },
    {
      "entropy": 6.059080123901367,
      "epoch": 1.4979257259959016,
      "grad_norm": 1.375,
      "learning_rate": 0.00047822052890761173,
      "loss": 5.8355,
      "mean_token_accuracy": 0.1717567652463913,
      "num_tokens": 29312437.0,
      "step": 14985
    },
    {
      "entropy": 6.045353031158447,
      "epoch": 1.49842555105713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004782052080535334,
      "loss": 5.7788,
      "mean_token_accuracy": 0.16749671548604966,
      "num_tokens": 29321777.0,
      "step": 14990
    },
    {
      "entropy": 6.12557544708252,
      "epoch": 1.4989253761183585,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004781898820868644,
      "loss": 5.8443,
      "mean_token_accuracy": 0.15956485867500306,
      "num_tokens": 29330551.0,
      "step": 14995
    },
    {
      "entropy": 6.105945682525634,
      "epoch": 1.4994252011795872,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004781745510079901,
      "loss": 5.8524,
      "mean_token_accuracy": 0.16126408576965331,
      "num_tokens": 29340110.0,
      "step": 15000
    },
    {
      "epoch": 1.4994252011795872,
      "eval_entropy": 5.938331310972053,
      "eval_loss": 5.974091053009033,
      "eval_mean_token_accuracy": 0.16375261917518327,
      "eval_num_tokens": 29340110.0,
      "eval_runtime": 24.0646,
      "eval_samples_per_second": 1290.069,
      "eval_steps_per_second": 161.274,
      "step": 15000
    },
    {
      "entropy": 6.129257011413574,
      "epoch": 1.4999250262408157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047815921481729647,
      "loss": 5.9411,
      "mean_token_accuracy": 0.158613021671772,
      "num_tokens": 29348935.0,
      "step": 15005
    },
    {
      "entropy": 6.186389446258545,
      "epoch": 1.5004248513020442,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004781438735151693,
      "loss": 5.9561,
      "mean_token_accuracy": 0.15055255740880966,
      "num_tokens": 29359108.0,
      "step": 15010
    },
    {
      "entropy": 6.111485385894776,
      "epoch": 1.5009246763632729,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047812852710199446,
      "loss": 5.8361,
      "mean_token_accuracy": 0.1730700448155403,
      "num_tokens": 29369074.0,
      "step": 15015
    },
    {
      "entropy": 6.1240287780761715,
      "epoch": 1.5014245014245016,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004781131755781582,
      "loss": 5.8923,
      "mean_token_accuracy": 0.15005211532115936,
      "num_tokens": 29379138.0,
      "step": 15020
    },
    {
      "entropy": 6.139255428314209,
      "epoch": 1.50192432648573,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004780978189440467,
      "loss": 5.8987,
      "mean_token_accuracy": 0.15596793293952943,
      "num_tokens": 29389124.0,
      "step": 15025
    },
    {
      "entropy": 6.056244897842407,
      "epoch": 1.5024241515469585,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047808245720004633,
      "loss": 5.7705,
      "mean_token_accuracy": 0.1694566413760185,
      "num_tokens": 29398179.0,
      "step": 15030
    },
    {
      "entropy": 6.090901613235474,
      "epoch": 1.5029239766081872,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047806709034654355,
      "loss": 5.8459,
      "mean_token_accuracy": 0.1612010195851326,
      "num_tokens": 29408177.0,
      "step": 15035
    },
    {
      "entropy": 6.140955781936645,
      "epoch": 1.5034238016694157,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004780517183839251,
      "loss": 5.8908,
      "mean_token_accuracy": 0.1615663528442383,
      "num_tokens": 29418699.0,
      "step": 15040
    },
    {
      "entropy": 6.128840017318725,
      "epoch": 1.5039236267306442,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004780363413125777,
      "loss": 5.9303,
      "mean_token_accuracy": 0.1672878608107567,
      "num_tokens": 29429271.0,
      "step": 15045
    },
    {
      "entropy": 6.225508022308349,
      "epoch": 1.5044234517918729,
      "grad_norm": 1.9453125,
      "learning_rate": 0.00047802095913288814,
      "loss": 6.102,
      "mean_token_accuracy": 0.1550429031252861,
      "num_tokens": 29439004.0,
      "step": 15050
    },
    {
      "entropy": 6.206991958618164,
      "epoch": 1.5049232768531016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047800557184524345,
      "loss": 5.9839,
      "mean_token_accuracy": 0.15507105812430383,
      "num_tokens": 29449359.0,
      "step": 15055
    },
    {
      "entropy": 6.167322015762329,
      "epoch": 1.5054231019143298,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004779901794500308,
      "loss": 5.9634,
      "mean_token_accuracy": 0.1549869582056999,
      "num_tokens": 29459277.0,
      "step": 15060
    },
    {
      "entropy": 6.1343762397766115,
      "epoch": 1.5059229269755585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047797478194763754,
      "loss": 5.8897,
      "mean_token_accuracy": 0.15956515520811082,
      "num_tokens": 29468676.0,
      "step": 15065
    },
    {
      "entropy": 6.165472364425659,
      "epoch": 1.5064227520367872,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047795937933845103,
      "loss": 5.9517,
      "mean_token_accuracy": 0.1539572685956955,
      "num_tokens": 29478343.0,
      "step": 15070
    },
    {
      "entropy": 6.203839683532715,
      "epoch": 1.5069225770980157,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004779439716228587,
      "loss": 5.8542,
      "mean_token_accuracy": 0.15676685497164727,
      "num_tokens": 29488434.0,
      "step": 15075
    },
    {
      "entropy": 6.136231327056885,
      "epoch": 1.5074224021592442,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004779285588012482,
      "loss": 5.826,
      "mean_token_accuracy": 0.1644906520843506,
      "num_tokens": 29498303.0,
      "step": 15080
    },
    {
      "entropy": 6.091013860702515,
      "epoch": 1.5079222272204729,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047791314087400745,
      "loss": 5.9553,
      "mean_token_accuracy": 0.15362822860479355,
      "num_tokens": 29508597.0,
      "step": 15085
    },
    {
      "entropy": 6.052167749404907,
      "epoch": 1.5084220522817016,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004778977178415243,
      "loss": 5.8537,
      "mean_token_accuracy": 0.17050477713346482,
      "num_tokens": 29517640.0,
      "step": 15090
    },
    {
      "entropy": 6.1827721118927,
      "epoch": 1.5089218773429298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004778822897041868,
      "loss": 5.8995,
      "mean_token_accuracy": 0.16128226965665818,
      "num_tokens": 29527415.0,
      "step": 15095
    },
    {
      "entropy": 6.214818811416626,
      "epoch": 1.5094217024041585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000477866856462383,
      "loss": 5.8428,
      "mean_token_accuracy": 0.15656811147928237,
      "num_tokens": 29537296.0,
      "step": 15100
    },
    {
      "entropy": 6.161885070800781,
      "epoch": 1.5099215274653872,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004778514181165014,
      "loss": 5.9279,
      "mean_token_accuracy": 0.1557593360543251,
      "num_tokens": 29546709.0,
      "step": 15105
    },
    {
      "entropy": 6.133514070510865,
      "epoch": 1.5104213525266157,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047783597466693025,
      "loss": 5.9316,
      "mean_token_accuracy": 0.1554557427763939,
      "num_tokens": 29556627.0,
      "step": 15110
    },
    {
      "entropy": 6.180773496627808,
      "epoch": 1.5109211775878442,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004778205261140582,
      "loss": 5.9039,
      "mean_token_accuracy": 0.15329119861125945,
      "num_tokens": 29566381.0,
      "step": 15115
    },
    {
      "entropy": 6.174249696731567,
      "epoch": 1.5114210026490729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004778050724582739,
      "loss": 5.909,
      "mean_token_accuracy": 0.15893025174736977,
      "num_tokens": 29575634.0,
      "step": 15120
    },
    {
      "entropy": 6.109538984298706,
      "epoch": 1.5119208277103013,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047778961369996623,
      "loss": 5.8494,
      "mean_token_accuracy": 0.16182774901390076,
      "num_tokens": 29586103.0,
      "step": 15125
    },
    {
      "entropy": 6.101470851898194,
      "epoch": 1.5124206527715298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000477774149839524,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16595451682806014,
      "num_tokens": 29596764.0,
      "step": 15130
    },
    {
      "entropy": 6.171605062484741,
      "epoch": 1.5129204778327585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047775868087733634,
      "loss": 5.8534,
      "mean_token_accuracy": 0.15689775347709656,
      "num_tokens": 29606642.0,
      "step": 15135
    },
    {
      "entropy": 6.152146625518799,
      "epoch": 1.5134203028939872,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047774320681379237,
      "loss": 5.9288,
      "mean_token_accuracy": 0.15953888297080993,
      "num_tokens": 29616583.0,
      "step": 15140
    },
    {
      "entropy": 6.061578893661499,
      "epoch": 1.5139201279552157,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004777277276492816,
      "loss": 5.8653,
      "mean_token_accuracy": 0.1654105380177498,
      "num_tokens": 29625961.0,
      "step": 15145
    },
    {
      "entropy": 6.182234573364258,
      "epoch": 1.5144199530164442,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047771224338419326,
      "loss": 5.8383,
      "mean_token_accuracy": 0.15482267588377,
      "num_tokens": 29636089.0,
      "step": 15150
    },
    {
      "entropy": 6.1475245475769045,
      "epoch": 1.5149197780776729,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004776967540189171,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16506034433841704,
      "num_tokens": 29646068.0,
      "step": 15155
    },
    {
      "entropy": 6.063611650466919,
      "epoch": 1.5154196031389013,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004776812595538428,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16983757466077803,
      "num_tokens": 29655248.0,
      "step": 15160
    },
    {
      "entropy": 6.116879987716675,
      "epoch": 1.5159194282001298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047766575998936013,
      "loss": 5.9187,
      "mean_token_accuracy": 0.1526571989059448,
      "num_tokens": 29665073.0,
      "step": 15165
    },
    {
      "entropy": 6.193166923522949,
      "epoch": 1.5164192532613585,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047765025532585907,
      "loss": 5.9201,
      "mean_token_accuracy": 0.1598212242126465,
      "num_tokens": 29674714.0,
      "step": 15170
    },
    {
      "entropy": 6.183599853515625,
      "epoch": 1.5169190783225872,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004776347455637298,
      "loss": 5.9196,
      "mean_token_accuracy": 0.15961009114980698,
      "num_tokens": 29683658.0,
      "step": 15175
    },
    {
      "entropy": 6.093431234359741,
      "epoch": 1.5174189033838157,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004776192307033624,
      "loss": 5.7919,
      "mean_token_accuracy": 0.17428786307573318,
      "num_tokens": 29693089.0,
      "step": 15180
    },
    {
      "entropy": 6.1605597019195555,
      "epoch": 1.5179187284450442,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004776037107451473,
      "loss": 5.8574,
      "mean_token_accuracy": 0.1638511225581169,
      "num_tokens": 29702727.0,
      "step": 15185
    },
    {
      "entropy": 6.130493307113648,
      "epoch": 1.5184185535062729,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000477588185689475,
      "loss": 5.9149,
      "mean_token_accuracy": 0.16213694959878922,
      "num_tokens": 29713882.0,
      "step": 15190
    },
    {
      "entropy": 6.095305395126343,
      "epoch": 1.5189183785675013,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047757265553673606,
      "loss": 5.767,
      "mean_token_accuracy": 0.17057351768016815,
      "num_tokens": 29722792.0,
      "step": 15195
    },
    {
      "entropy": 6.111426639556885,
      "epoch": 1.5194182036287298,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047755712028732126,
      "loss": 5.9478,
      "mean_token_accuracy": 0.15854859203100205,
      "num_tokens": 29734246.0,
      "step": 15200
    },
    {
      "entropy": 6.201867294311524,
      "epoch": 1.5199180286899585,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004775415799416214,
      "loss": 5.9601,
      "mean_token_accuracy": 0.1598719373345375,
      "num_tokens": 29743747.0,
      "step": 15205
    },
    {
      "entropy": 6.141002750396728,
      "epoch": 1.5204178537511872,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004775260345000275,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16224075257778167,
      "num_tokens": 29754136.0,
      "step": 15210
    },
    {
      "entropy": 6.08375096321106,
      "epoch": 1.5209176788124157,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047751048396293077,
      "loss": 5.7786,
      "mean_token_accuracy": 0.16858799755573273,
      "num_tokens": 29763802.0,
      "step": 15215
    },
    {
      "entropy": 6.063307905197144,
      "epoch": 1.5214175038736442,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047749492833072226,
      "loss": 5.7937,
      "mean_token_accuracy": 0.15910851657390596,
      "num_tokens": 29773376.0,
      "step": 15220
    },
    {
      "entropy": 6.118761205673218,
      "epoch": 1.5219173289348729,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047747936760379354,
      "loss": 5.8903,
      "mean_token_accuracy": 0.16180018186569214,
      "num_tokens": 29784407.0,
      "step": 15225
    },
    {
      "entropy": 6.174037742614746,
      "epoch": 1.5224171539961013,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000477463801782536,
      "loss": 5.8636,
      "mean_token_accuracy": 0.15600069463253022,
      "num_tokens": 29794745.0,
      "step": 15230
    },
    {
      "entropy": 6.176032447814942,
      "epoch": 1.5229169790573298,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047744823086734136,
      "loss": 5.8285,
      "mean_token_accuracy": 0.16022895127534867,
      "num_tokens": 29804393.0,
      "step": 15235
    },
    {
      "entropy": 6.130827379226685,
      "epoch": 1.5234168041185585,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047743265485860123,
      "loss": 5.8452,
      "mean_token_accuracy": 0.16182911098003389,
      "num_tokens": 29814306.0,
      "step": 15240
    },
    {
      "entropy": 6.164071655273437,
      "epoch": 1.5239166291797872,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047741707375670766,
      "loss": 5.8701,
      "mean_token_accuracy": 0.15973699241876602,
      "num_tokens": 29823730.0,
      "step": 15245
    },
    {
      "entropy": 6.124432897567749,
      "epoch": 1.5244164542410157,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004774014875620526,
      "loss": 5.8653,
      "mean_token_accuracy": 0.15992121547460555,
      "num_tokens": 29833901.0,
      "step": 15250
    },
    {
      "entropy": 6.102708625793457,
      "epoch": 1.5249162793022442,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047738589627502814,
      "loss": 5.8019,
      "mean_token_accuracy": 0.16344711035490037,
      "num_tokens": 29843874.0,
      "step": 15255
    },
    {
      "entropy": 6.14438271522522,
      "epoch": 1.5254161043634729,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047737029989602665,
      "loss": 5.8361,
      "mean_token_accuracy": 0.15713583528995514,
      "num_tokens": 29853031.0,
      "step": 15260
    },
    {
      "entropy": 6.103888750076294,
      "epoch": 1.5259159294247013,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004773546984254405,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15814789831638337,
      "num_tokens": 29862437.0,
      "step": 15265
    },
    {
      "entropy": 6.0668822765350345,
      "epoch": 1.5264157544859298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004773390918636622,
      "loss": 5.8092,
      "mean_token_accuracy": 0.16669632941484452,
      "num_tokens": 29870914.0,
      "step": 15270
    },
    {
      "entropy": 6.108756971359253,
      "epoch": 1.5269155795471585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004773234802110844,
      "loss": 5.8167,
      "mean_token_accuracy": 0.1637582316994667,
      "num_tokens": 29880745.0,
      "step": 15275
    },
    {
      "entropy": 6.17589168548584,
      "epoch": 1.5274154046083872,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004773078634681,
      "loss": 6.0044,
      "mean_token_accuracy": 0.15438187643885612,
      "num_tokens": 29890049.0,
      "step": 15280
    },
    {
      "entropy": 6.2413076877594,
      "epoch": 1.5279152296696157,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047729224163510167,
      "loss": 6.0038,
      "mean_token_accuracy": 0.14840957075357436,
      "num_tokens": 29899664.0,
      "step": 15285
    },
    {
      "entropy": 6.097167873382569,
      "epoch": 1.5284150547308442,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047727661471248263,
      "loss": 5.8634,
      "mean_token_accuracy": 0.16577814593911172,
      "num_tokens": 29909261.0,
      "step": 15290
    },
    {
      "entropy": 6.076865386962891,
      "epoch": 1.5289148797920729,
      "grad_norm": 1.21875,
      "learning_rate": 0.000477260982700636,
      "loss": 5.7436,
      "mean_token_accuracy": 0.17050630152225493,
      "num_tokens": 29920137.0,
      "step": 15295
    },
    {
      "entropy": 6.115400409698486,
      "epoch": 1.5294147048533013,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004772453455999552,
      "loss": 5.8962,
      "mean_token_accuracy": 0.15941789001226425,
      "num_tokens": 29930708.0,
      "step": 15300
    },
    {
      "entropy": 6.12663779258728,
      "epoch": 1.5299145299145298,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004772297034108334,
      "loss": 5.9738,
      "mean_token_accuracy": 0.15663735270500184,
      "num_tokens": 29940804.0,
      "step": 15305
    },
    {
      "entropy": 6.074058103561401,
      "epoch": 1.5304143549757585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047721405613366434,
      "loss": 5.7989,
      "mean_token_accuracy": 0.1690444216132164,
      "num_tokens": 29951609.0,
      "step": 15310
    },
    {
      "entropy": 6.217831611633301,
      "epoch": 1.5309141800369872,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047719840376884173,
      "loss": 5.9683,
      "mean_token_accuracy": 0.1646478869020939,
      "num_tokens": 29961764.0,
      "step": 15315
    },
    {
      "entropy": 6.078429079055786,
      "epoch": 1.5314140050982155,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004771827463167593,
      "loss": 5.8895,
      "mean_token_accuracy": 0.1541639268398285,
      "num_tokens": 29972160.0,
      "step": 15320
    },
    {
      "entropy": 6.044488000869751,
      "epoch": 1.5319138301594442,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004771670837778109,
      "loss": 5.7854,
      "mean_token_accuracy": 0.16356478929519652,
      "num_tokens": 29982190.0,
      "step": 15325
    },
    {
      "entropy": 6.188935041427612,
      "epoch": 1.5324136552206729,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047715141615239075,
      "loss": 5.9082,
      "mean_token_accuracy": 0.15966019332408904,
      "num_tokens": 29992020.0,
      "step": 15330
    },
    {
      "entropy": 6.160448503494263,
      "epoch": 1.5329134802819013,
      "grad_norm": 1.65625,
      "learning_rate": 0.00047713574344089305,
      "loss": 5.9256,
      "mean_token_accuracy": 0.1570179656147957,
      "num_tokens": 30002585.0,
      "step": 15335
    },
    {
      "entropy": 6.1022566795349125,
      "epoch": 1.5334133053431298,
      "grad_norm": 1.234375,
      "learning_rate": 0.000477120065643712,
      "loss": 5.8806,
      "mean_token_accuracy": 0.16059028208255768,
      "num_tokens": 30012210.0,
      "step": 15340
    },
    {
      "entropy": 6.133050155639649,
      "epoch": 1.5339131304043585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004771043827612421,
      "loss": 5.9491,
      "mean_token_accuracy": 0.1614042341709137,
      "num_tokens": 30022339.0,
      "step": 15345
    },
    {
      "entropy": 6.055250644683838,
      "epoch": 1.5344129554655872,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047708869479387795,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16247523501515387,
      "num_tokens": 30032313.0,
      "step": 15350
    },
    {
      "entropy": 6.105166244506836,
      "epoch": 1.5349127805268155,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004770730017420143,
      "loss": 5.8283,
      "mean_token_accuracy": 0.1637309119105339,
      "num_tokens": 30041840.0,
      "step": 15355
    },
    {
      "entropy": 6.065309524536133,
      "epoch": 1.5354126055880442,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004770573036060458,
      "loss": 5.7965,
      "mean_token_accuracy": 0.1719755731523037,
      "num_tokens": 30052118.0,
      "step": 15360
    },
    {
      "entropy": 6.150920677185058,
      "epoch": 1.5359124306492729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004770416003863676,
      "loss": 5.9741,
      "mean_token_accuracy": 0.1529782921075821,
      "num_tokens": 30062352.0,
      "step": 15365
    },
    {
      "entropy": 6.174571084976196,
      "epoch": 1.5364122557105013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004770258920833748,
      "loss": 5.9444,
      "mean_token_accuracy": 0.1519168384373188,
      "num_tokens": 30072657.0,
      "step": 15370
    },
    {
      "entropy": 6.131659841537475,
      "epoch": 1.5369120807717298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047701017869746236,
      "loss": 5.8939,
      "mean_token_accuracy": 0.16448913663625717,
      "num_tokens": 30082759.0,
      "step": 15375
    },
    {
      "entropy": 6.158608675003052,
      "epoch": 1.5374119058329585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047699446022902596,
      "loss": 6.0134,
      "mean_token_accuracy": 0.14788779392838478,
      "num_tokens": 30092762.0,
      "step": 15380
    },
    {
      "entropy": 6.148561668395996,
      "epoch": 1.537911730894187,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047697873667846094,
      "loss": 5.8541,
      "mean_token_accuracy": 0.15895670801401138,
      "num_tokens": 30101492.0,
      "step": 15385
    },
    {
      "entropy": 6.219657850265503,
      "epoch": 1.5384115559554155,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047696300804616276,
      "loss": 5.892,
      "mean_token_accuracy": 0.1597977250814438,
      "num_tokens": 30111136.0,
      "step": 15390
    },
    {
      "entropy": 6.14922947883606,
      "epoch": 1.5389113810166442,
      "grad_norm": 1.5,
      "learning_rate": 0.0004769472743325273,
      "loss": 5.9175,
      "mean_token_accuracy": 0.15961354225873947,
      "num_tokens": 30120804.0,
      "step": 15395
    },
    {
      "entropy": 6.17525429725647,
      "epoch": 1.5394112060778729,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047693153553795046,
      "loss": 5.934,
      "mean_token_accuracy": 0.15859705656766893,
      "num_tokens": 30130976.0,
      "step": 15400
    },
    {
      "entropy": 6.109127187728882,
      "epoch": 1.5399110311391013,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004769157916628281,
      "loss": 5.8527,
      "mean_token_accuracy": 0.16378755867481232,
      "num_tokens": 30140060.0,
      "step": 15405
    },
    {
      "entropy": 6.141937112808227,
      "epoch": 1.5404108562003298,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047690004270755634,
      "loss": 5.9194,
      "mean_token_accuracy": 0.15956482589244841,
      "num_tokens": 30150474.0,
      "step": 15410
    },
    {
      "entropy": 6.181976842880249,
      "epoch": 1.5409106812615585,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004768842886725314,
      "loss": 5.9287,
      "mean_token_accuracy": 0.1592043846845627,
      "num_tokens": 30159529.0,
      "step": 15415
    },
    {
      "entropy": 6.109594345092773,
      "epoch": 1.541410506322787,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047686852955814986,
      "loss": 6.0417,
      "mean_token_accuracy": 0.15682915300130845,
      "num_tokens": 30169352.0,
      "step": 15420
    },
    {
      "entropy": 6.0949931144714355,
      "epoch": 1.5419103313840155,
      "grad_norm": 1.34375,
      "learning_rate": 0.000476852765364808,
      "loss": 5.8043,
      "mean_token_accuracy": 0.16553349494934083,
      "num_tokens": 30178190.0,
      "step": 15425
    },
    {
      "entropy": 6.110762310028076,
      "epoch": 1.5424101564452442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004768369960929025,
      "loss": 5.7757,
      "mean_token_accuracy": 0.16251659840345384,
      "num_tokens": 30187878.0,
      "step": 15430
    },
    {
      "entropy": 6.091281127929688,
      "epoch": 1.5429099815064728,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004768212217428301,
      "loss": 5.9038,
      "mean_token_accuracy": 0.16099462062120437,
      "num_tokens": 30197786.0,
      "step": 15435
    },
    {
      "entropy": 6.125912809371949,
      "epoch": 1.5434098065677013,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00047680544231498766,
      "loss": 5.8718,
      "mean_token_accuracy": 0.16214881390333175,
      "num_tokens": 30207837.0,
      "step": 15440
    },
    {
      "entropy": 6.086841106414795,
      "epoch": 1.5439096316289298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004767896578097723,
      "loss": 5.8499,
      "mean_token_accuracy": 0.16459144055843353,
      "num_tokens": 30217135.0,
      "step": 15445
    },
    {
      "entropy": 6.101604366302491,
      "epoch": 1.5444094566901585,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004767738682275811,
      "loss": 5.8401,
      "mean_token_accuracy": 0.15953147783875465,
      "num_tokens": 30227339.0,
      "step": 15450
    },
    {
      "entropy": 6.220514011383057,
      "epoch": 1.544909281751387,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004767580735688112,
      "loss": 6.0066,
      "mean_token_accuracy": 0.1526704952120781,
      "num_tokens": 30235735.0,
      "step": 15455
    },
    {
      "entropy": 6.105855512619018,
      "epoch": 1.5454091068126155,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004767422738338601,
      "loss": 5.8014,
      "mean_token_accuracy": 0.16696115732192993,
      "num_tokens": 30245921.0,
      "step": 15460
    },
    {
      "entropy": 6.09586124420166,
      "epoch": 1.5459089318738441,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004767264690231253,
      "loss": 5.8159,
      "mean_token_accuracy": 0.17011403888463975,
      "num_tokens": 30256047.0,
      "step": 15465
    },
    {
      "entropy": 6.112372922897339,
      "epoch": 1.5464087569350728,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004767106591370045,
      "loss": 5.8574,
      "mean_token_accuracy": 0.16208492666482927,
      "num_tokens": 30266460.0,
      "step": 15470
    },
    {
      "entropy": 6.1243916034698485,
      "epoch": 1.5469085819963013,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047669484417589537,
      "loss": 5.8523,
      "mean_token_accuracy": 0.16277459263801575,
      "num_tokens": 30275937.0,
      "step": 15475
    },
    {
      "entropy": 6.153229188919068,
      "epoch": 1.5474084070575298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047667902414019585,
      "loss": 5.79,
      "mean_token_accuracy": 0.17215044498443605,
      "num_tokens": 30285608.0,
      "step": 15480
    },
    {
      "entropy": 6.0987708568573,
      "epoch": 1.5479082321187585,
      "grad_norm": 1.4296875,
      "learning_rate": 0.000476663199030304,
      "loss": 5.8422,
      "mean_token_accuracy": 0.16495077311992645,
      "num_tokens": 30295071.0,
      "step": 15485
    },
    {
      "entropy": 5.964448118209839,
      "epoch": 1.548408057179987,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047664736884661804,
      "loss": 5.8344,
      "mean_token_accuracy": 0.16528006196022033,
      "num_tokens": 30305878.0,
      "step": 15490
    },
    {
      "entropy": 6.151038551330567,
      "epoch": 1.5489078822412155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047663153358953604,
      "loss": 5.9376,
      "mean_token_accuracy": 0.16205189228057862,
      "num_tokens": 30316544.0,
      "step": 15495
    },
    {
      "entropy": 6.203384304046631,
      "epoch": 1.5494077073024441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004766156932594566,
      "loss": 5.8364,
      "mean_token_accuracy": 0.1585783876478672,
      "num_tokens": 30325819.0,
      "step": 15500
    },
    {
      "entropy": 6.110129117965698,
      "epoch": 1.5499075323636728,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047659984785677824,
      "loss": 5.7982,
      "mean_token_accuracy": 0.16433823108673096,
      "num_tokens": 30335160.0,
      "step": 15505
    },
    {
      "entropy": 6.120265340805053,
      "epoch": 1.5504073574249013,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047658399738189955,
      "loss": 5.8934,
      "mean_token_accuracy": 0.16182106733322144,
      "num_tokens": 30343412.0,
      "step": 15510
    },
    {
      "entropy": 6.079827737808228,
      "epoch": 1.5509071824861298,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047656814183521937,
      "loss": 5.8061,
      "mean_token_accuracy": 0.16738064140081405,
      "num_tokens": 30352336.0,
      "step": 15515
    },
    {
      "entropy": 6.1105701446533205,
      "epoch": 1.5514070075473585,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004765522812171366,
      "loss": 5.8035,
      "mean_token_accuracy": 0.16175583153963088,
      "num_tokens": 30362476.0,
      "step": 15520
    },
    {
      "entropy": 6.081426477432251,
      "epoch": 1.551906832608587,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047653641552805023,
      "loss": 5.8841,
      "mean_token_accuracy": 0.15940205454826356,
      "num_tokens": 30372639.0,
      "step": 15525
    },
    {
      "entropy": 6.092545461654663,
      "epoch": 1.5524066576698154,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004765205447683595,
      "loss": 5.7507,
      "mean_token_accuracy": 0.16170932054519654,
      "num_tokens": 30382443.0,
      "step": 15530
    },
    {
      "entropy": 6.118351221084595,
      "epoch": 1.5529064827310441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004765046689384638,
      "loss": 5.8057,
      "mean_token_accuracy": 0.15895779505372049,
      "num_tokens": 30390969.0,
      "step": 15535
    },
    {
      "entropy": 6.0632524490356445,
      "epoch": 1.5534063077922728,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004764887880387625,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15700671672821045,
      "num_tokens": 30401773.0,
      "step": 15540
    },
    {
      "entropy": 6.169080829620361,
      "epoch": 1.5539061328535013,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047647290206965506,
      "loss": 5.9496,
      "mean_token_accuracy": 0.15057041719555855,
      "num_tokens": 30412393.0,
      "step": 15545
    },
    {
      "entropy": 6.144336843490601,
      "epoch": 1.5544059579147298,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047645701103154125,
      "loss": 5.86,
      "mean_token_accuracy": 0.15633816570043563,
      "num_tokens": 30422274.0,
      "step": 15550
    },
    {
      "entropy": 6.168834447860718,
      "epoch": 1.5549057829759585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047644111492482094,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15778811573982238,
      "num_tokens": 30432345.0,
      "step": 15555
    },
    {
      "entropy": 6.114602899551391,
      "epoch": 1.555405608037187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000476425213749894,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16749835908412933,
      "num_tokens": 30443016.0,
      "step": 15560
    },
    {
      "entropy": 6.040955924987793,
      "epoch": 1.5559054330984154,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004764093075071605,
      "loss": 5.7272,
      "mean_token_accuracy": 0.168129163980484,
      "num_tokens": 30452392.0,
      "step": 15565
    },
    {
      "entropy": 6.042561674118042,
      "epoch": 1.5564052581596441,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004763933961970206,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1632886216044426,
      "num_tokens": 30461178.0,
      "step": 15570
    },
    {
      "entropy": 6.104418659210205,
      "epoch": 1.5569050832208728,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004763774798198748,
      "loss": 5.837,
      "mean_token_accuracy": 0.1607632040977478,
      "num_tokens": 30470616.0,
      "step": 15575
    },
    {
      "entropy": 6.107015371322632,
      "epoch": 1.5574049082821013,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004763615583761234,
      "loss": 5.8199,
      "mean_token_accuracy": 0.1621793434023857,
      "num_tokens": 30479393.0,
      "step": 15580
    },
    {
      "entropy": 6.112373018264771,
      "epoch": 1.5579047333433298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004763456318661669,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16356650292873381,
      "num_tokens": 30488927.0,
      "step": 15585
    },
    {
      "entropy": 6.093655014038086,
      "epoch": 1.5584045584045585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004763297002904062,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15661060735583304,
      "num_tokens": 30498973.0,
      "step": 15590
    },
    {
      "entropy": 6.155541467666626,
      "epoch": 1.558904383465787,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047631376364924195,
      "loss": 5.8061,
      "mean_token_accuracy": 0.16573309600353242,
      "num_tokens": 30508643.0,
      "step": 15595
    },
    {
      "entropy": 6.021056699752807,
      "epoch": 1.5594042085270154,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004762978219430753,
      "loss": 5.764,
      "mean_token_accuracy": 0.1709181159734726,
      "num_tokens": 30518321.0,
      "step": 15600
    },
    {
      "entropy": 6.049333095550537,
      "epoch": 1.5599040335882441,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004762818751723072,
      "loss": 5.7779,
      "mean_token_accuracy": 0.16424501985311507,
      "num_tokens": 30528127.0,
      "step": 15605
    },
    {
      "entropy": 6.083439207077026,
      "epoch": 1.5604038586494728,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047626592333733895,
      "loss": 5.744,
      "mean_token_accuracy": 0.17178516536951066,
      "num_tokens": 30538411.0,
      "step": 15610
    },
    {
      "entropy": 6.114041566848755,
      "epoch": 1.560903683710701,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004762499664385719,
      "loss": 5.8757,
      "mean_token_accuracy": 0.15756051540374755,
      "num_tokens": 30548118.0,
      "step": 15615
    },
    {
      "entropy": 6.143196487426758,
      "epoch": 1.5614035087719298,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047623400447640733,
      "loss": 5.8876,
      "mean_token_accuracy": 0.1578742355108261,
      "num_tokens": 30557920.0,
      "step": 15620
    },
    {
      "entropy": 6.2058733940124515,
      "epoch": 1.5619033338331585,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047621803745124717,
      "loss": 5.922,
      "mean_token_accuracy": 0.15717290341854095,
      "num_tokens": 30567189.0,
      "step": 15625
    },
    {
      "entropy": 6.125095844268799,
      "epoch": 1.562403158894387,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047620206536349294,
      "loss": 5.9236,
      "mean_token_accuracy": 0.16054659932851792,
      "num_tokens": 30578208.0,
      "step": 15630
    },
    {
      "entropy": 6.127906417846679,
      "epoch": 1.5629029839556154,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047618608821354636,
      "loss": 5.8534,
      "mean_token_accuracy": 0.1648792579770088,
      "num_tokens": 30587845.0,
      "step": 15635
    },
    {
      "entropy": 6.145274066925049,
      "epoch": 1.5634028090168441,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004761701060018097,
      "loss": 5.8734,
      "mean_token_accuracy": 0.16200701296329498,
      "num_tokens": 30596601.0,
      "step": 15640
    },
    {
      "entropy": 6.155472660064698,
      "epoch": 1.5639026340780728,
      "grad_norm": 1.15625,
      "learning_rate": 0.000476154118728685,
      "loss": 5.9466,
      "mean_token_accuracy": 0.15516510009765624,
      "num_tokens": 30606860.0,
      "step": 15645
    },
    {
      "entropy": 6.097356653213501,
      "epoch": 1.564402459139301,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047613812639457425,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16390591114759445,
      "num_tokens": 30614980.0,
      "step": 15650
    },
    {
      "entropy": 6.124844264984131,
      "epoch": 1.5649022842005298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004761221289998801,
      "loss": 5.9152,
      "mean_token_accuracy": 0.16294804960489273,
      "num_tokens": 30625833.0,
      "step": 15655
    },
    {
      "entropy": 6.132277727127075,
      "epoch": 1.5654021092617585,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004761061265450049,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16455502957105636,
      "num_tokens": 30636052.0,
      "step": 15660
    },
    {
      "entropy": 6.186130666732788,
      "epoch": 1.565901934322987,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047609011903035136,
      "loss": 5.8968,
      "mean_token_accuracy": 0.15746278911828995,
      "num_tokens": 30646421.0,
      "step": 15665
    },
    {
      "entropy": 6.102401494979858,
      "epoch": 1.5664017593842154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047607410645632216,
      "loss": 5.9683,
      "mean_token_accuracy": 0.15722895860671998,
      "num_tokens": 30655903.0,
      "step": 15670
    },
    {
      "entropy": 6.0914726734161375,
      "epoch": 1.5669015844454441,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047605808882332013,
      "loss": 5.7386,
      "mean_token_accuracy": 0.16537982523441314,
      "num_tokens": 30664963.0,
      "step": 15675
    },
    {
      "entropy": 6.126423978805542,
      "epoch": 1.5674014095066726,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004760420661317484,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1620078608393669,
      "num_tokens": 30675182.0,
      "step": 15680
    },
    {
      "entropy": 6.161059427261352,
      "epoch": 1.567901234567901,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004760260383820099,
      "loss": 5.9779,
      "mean_token_accuracy": 0.1561131402850151,
      "num_tokens": 30685267.0,
      "step": 15685
    },
    {
      "entropy": 6.23167142868042,
      "epoch": 1.5684010596291298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047601000557450816,
      "loss": 5.9711,
      "mean_token_accuracy": 0.15159882605075836,
      "num_tokens": 30697462.0,
      "step": 15690
    },
    {
      "entropy": 6.159661293029785,
      "epoch": 1.5689008846903585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004759939677096463,
      "loss": 5.7889,
      "mean_token_accuracy": 0.16417139768600464,
      "num_tokens": 30706754.0,
      "step": 15695
    },
    {
      "entropy": 6.059874391555786,
      "epoch": 1.569400709751587,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004759779247878279,
      "loss": 5.9093,
      "mean_token_accuracy": 0.157210473716259,
      "num_tokens": 30717136.0,
      "step": 15700
    },
    {
      "entropy": 6.058730030059815,
      "epoch": 1.5699005348128154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047596187680945675,
      "loss": 5.7831,
      "mean_token_accuracy": 0.16183977723121643,
      "num_tokens": 30725830.0,
      "step": 15705
    },
    {
      "entropy": 6.192148685455322,
      "epoch": 1.5704003598740441,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004759458237749364,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16339411586523056,
      "num_tokens": 30736494.0,
      "step": 15710
    },
    {
      "entropy": 6.189281988143921,
      "epoch": 1.5709001849352726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004759297656846708,
      "loss": 5.8767,
      "mean_token_accuracy": 0.15415189638733864,
      "num_tokens": 30746006.0,
      "step": 15715
    },
    {
      "entropy": 6.107012128829956,
      "epoch": 1.571400009996501,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047591370253906415,
      "loss": 5.8434,
      "mean_token_accuracy": 0.1566074311733246,
      "num_tokens": 30754776.0,
      "step": 15720
    },
    {
      "entropy": 6.000264739990234,
      "epoch": 1.5718998350577298,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047589763433852027,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16884029060602188,
      "num_tokens": 30764342.0,
      "step": 15725
    },
    {
      "entropy": 6.076050853729248,
      "epoch": 1.5723996601189585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004758815610834437,
      "loss": 5.8424,
      "mean_token_accuracy": 0.15902737975120546,
      "num_tokens": 30773282.0,
      "step": 15730
    },
    {
      "entropy": 6.191977834701538,
      "epoch": 1.572899485180187,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047586548277423873,
      "loss": 5.8143,
      "mean_token_accuracy": 0.16420478224754334,
      "num_tokens": 30783407.0,
      "step": 15735
    },
    {
      "entropy": 6.101856470108032,
      "epoch": 1.5733993102414154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047584939941130994,
      "loss": 5.82,
      "mean_token_accuracy": 0.16188424527645112,
      "num_tokens": 30793922.0,
      "step": 15740
    },
    {
      "entropy": 6.071271991729736,
      "epoch": 1.5738991353026441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004758333109950619,
      "loss": 5.7892,
      "mean_token_accuracy": 0.162991201877594,
      "num_tokens": 30802721.0,
      "step": 15745
    },
    {
      "entropy": 6.0835436344146725,
      "epoch": 1.5743989603638726,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047581721752589943,
      "loss": 5.9495,
      "mean_token_accuracy": 0.15712231099605561,
      "num_tokens": 30813229.0,
      "step": 15750
    },
    {
      "entropy": 6.118486499786377,
      "epoch": 1.574898785425101,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004758011190042274,
      "loss": 5.9084,
      "mean_token_accuracy": 0.16131560653448104,
      "num_tokens": 30822989.0,
      "step": 15755
    },
    {
      "entropy": 6.130643081665039,
      "epoch": 1.5753986104863298,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004757850154304509,
      "loss": 5.9038,
      "mean_token_accuracy": 0.1543731227517128,
      "num_tokens": 30832134.0,
      "step": 15760
    },
    {
      "entropy": 6.181361532211303,
      "epoch": 1.5758984355475585,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004757689068049752,
      "loss": 5.9479,
      "mean_token_accuracy": 0.16006479859352113,
      "num_tokens": 30841528.0,
      "step": 15765
    },
    {
      "entropy": 6.070429420471191,
      "epoch": 1.576398260608787,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047575279312820533,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1588297575712204,
      "num_tokens": 30852166.0,
      "step": 15770
    },
    {
      "entropy": 6.128502655029297,
      "epoch": 1.5768980856700154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004757366744005469,
      "loss": 5.896,
      "mean_token_accuracy": 0.15851580649614333,
      "num_tokens": 30862188.0,
      "step": 15775
    },
    {
      "entropy": 6.166647577285767,
      "epoch": 1.5773979107312441,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047572055062240537,
      "loss": 5.905,
      "mean_token_accuracy": 0.15618523359298705,
      "num_tokens": 30871057.0,
      "step": 15780
    },
    {
      "entropy": 6.147783803939819,
      "epoch": 1.5778977357924726,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004757044217941865,
      "loss": 5.7553,
      "mean_token_accuracy": 0.16735859513282775,
      "num_tokens": 30880884.0,
      "step": 15785
    },
    {
      "entropy": 6.0723051071167,
      "epoch": 1.578397560853701,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047568828791629605,
      "loss": 5.8376,
      "mean_token_accuracy": 0.1607985094189644,
      "num_tokens": 30891017.0,
      "step": 15790
    },
    {
      "entropy": 6.08468337059021,
      "epoch": 1.5788973859149298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004756721489891398,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16075626760721207,
      "num_tokens": 30901008.0,
      "step": 15795
    },
    {
      "entropy": 6.164804124832154,
      "epoch": 1.5793972109761585,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004756560050131239,
      "loss": 5.9364,
      "mean_token_accuracy": 0.15656034648418427,
      "num_tokens": 30911266.0,
      "step": 15800
    },
    {
      "entropy": 6.111330223083496,
      "epoch": 1.579897036037387,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047563985598865466,
      "loss": 5.8258,
      "mean_token_accuracy": 0.1636205404996872,
      "num_tokens": 30921758.0,
      "step": 15805
    },
    {
      "entropy": 6.102659130096436,
      "epoch": 1.5803968610986154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047562370191613816,
      "loss": 5.8806,
      "mean_token_accuracy": 0.15896131247282028,
      "num_tokens": 30931446.0,
      "step": 15810
    },
    {
      "entropy": 6.137127780914307,
      "epoch": 1.5808966861598441,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004756075427959811,
      "loss": 5.9886,
      "mean_token_accuracy": 0.15946197956800462,
      "num_tokens": 30942651.0,
      "step": 15815
    },
    {
      "entropy": 6.09373722076416,
      "epoch": 1.5813965112210726,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004755913786285898,
      "loss": 5.7438,
      "mean_token_accuracy": 0.16538350731134416,
      "num_tokens": 30952039.0,
      "step": 15820
    },
    {
      "entropy": 6.0641357421875,
      "epoch": 1.581896336282301,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047557520941437093,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1651611313223839,
      "num_tokens": 30961834.0,
      "step": 15825
    },
    {
      "entropy": 6.120369386672974,
      "epoch": 1.5823961613435298,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047555903515373156,
      "loss": 5.8982,
      "mean_token_accuracy": 0.15750859379768373,
      "num_tokens": 30972015.0,
      "step": 15830
    },
    {
      "entropy": 6.0977850437164305,
      "epoch": 1.5828959864047585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047554285584707834,
      "loss": 5.7797,
      "mean_token_accuracy": 0.16424737870693207,
      "num_tokens": 30982082.0,
      "step": 15835
    },
    {
      "entropy": 6.071477937698364,
      "epoch": 1.583395811465987,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004755266714948186,
      "loss": 5.7705,
      "mean_token_accuracy": 0.16758534759283067,
      "num_tokens": 30991273.0,
      "step": 15840
    },
    {
      "entropy": 6.093302583694458,
      "epoch": 1.5838956365272154,
      "grad_norm": 1.125,
      "learning_rate": 0.0004755104820973593,
      "loss": 5.8056,
      "mean_token_accuracy": 0.16220231205224991,
      "num_tokens": 31001062.0,
      "step": 15845
    },
    {
      "entropy": 6.042902851104737,
      "epoch": 1.5843954615884441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047549428765510793,
      "loss": 5.7792,
      "mean_token_accuracy": 0.16632893830537795,
      "num_tokens": 31009905.0,
      "step": 15850
    },
    {
      "entropy": 6.067074632644653,
      "epoch": 1.5848952866496726,
      "grad_norm": 1.25,
      "learning_rate": 0.00047547808816847186,
      "loss": 5.8733,
      "mean_token_accuracy": 0.15995508283376694,
      "num_tokens": 31020517.0,
      "step": 15855
    },
    {
      "entropy": 6.175113677978516,
      "epoch": 1.585395111710901,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004754618836378587,
      "loss": 5.876,
      "mean_token_accuracy": 0.15295375883579254,
      "num_tokens": 31030789.0,
      "step": 15860
    },
    {
      "entropy": 6.170568799972534,
      "epoch": 1.5858949367721298,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004754456740636761,
      "loss": 5.8198,
      "mean_token_accuracy": 0.16940248608589173,
      "num_tokens": 31040459.0,
      "step": 15865
    },
    {
      "entropy": 6.097773170471191,
      "epoch": 1.5863947618333585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047542945944633187,
      "loss": 5.845,
      "mean_token_accuracy": 0.16541507244110107,
      "num_tokens": 31050049.0,
      "step": 15870
    },
    {
      "entropy": 6.032937288284302,
      "epoch": 1.5868945868945867,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047541323978623405,
      "loss": 5.8484,
      "mean_token_accuracy": 0.16259590089321135,
      "num_tokens": 31058678.0,
      "step": 15875
    },
    {
      "entropy": 6.10330057144165,
      "epoch": 1.5873944119558154,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047539701508379073,
      "loss": 5.895,
      "mean_token_accuracy": 0.15701993256807328,
      "num_tokens": 31067878.0,
      "step": 15880
    },
    {
      "entropy": 6.171307754516602,
      "epoch": 1.5878942370170441,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004753807853394101,
      "loss": 5.8801,
      "mean_token_accuracy": 0.15960941463708878,
      "num_tokens": 31077632.0,
      "step": 15885
    },
    {
      "entropy": 6.103956651687622,
      "epoch": 1.5883940620782726,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047536455055350034,
      "loss": 5.9337,
      "mean_token_accuracy": 0.1592994101345539,
      "num_tokens": 31087576.0,
      "step": 15890
    },
    {
      "entropy": 6.075461912155151,
      "epoch": 1.588893887139501,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047534831072647016,
      "loss": 5.8201,
      "mean_token_accuracy": 0.16470474302768706,
      "num_tokens": 31097053.0,
      "step": 15895
    },
    {
      "entropy": 6.09102087020874,
      "epoch": 1.5893937122007298,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047533206585872805,
      "loss": 5.8363,
      "mean_token_accuracy": 0.1650298446416855,
      "num_tokens": 31106426.0,
      "step": 15900
    },
    {
      "entropy": 6.115528535842896,
      "epoch": 1.5898935372619585,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004753158159506827,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16932081282138825,
      "num_tokens": 31115645.0,
      "step": 15905
    },
    {
      "entropy": 6.062646770477295,
      "epoch": 1.5903933623231867,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004752995610027429,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16525014489889145,
      "num_tokens": 31125275.0,
      "step": 15910
    },
    {
      "entropy": 6.068267869949341,
      "epoch": 1.5908931873844154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047528330101531777,
      "loss": 5.813,
      "mean_token_accuracy": 0.1624324843287468,
      "num_tokens": 31136444.0,
      "step": 15915
    },
    {
      "entropy": 6.0834729194641115,
      "epoch": 1.5913930124456441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047526703598881623,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16538735926151277,
      "num_tokens": 31145489.0,
      "step": 15920
    },
    {
      "entropy": 6.177283477783203,
      "epoch": 1.5918928375068726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004752507659236476,
      "loss": 5.9116,
      "mean_token_accuracy": 0.15779643058776854,
      "num_tokens": 31154239.0,
      "step": 15925
    },
    {
      "entropy": 6.161400175094604,
      "epoch": 1.592392662568101,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004752344908202213,
      "loss": 5.829,
      "mean_token_accuracy": 0.1661909833550453,
      "num_tokens": 31163401.0,
      "step": 15930
    },
    {
      "entropy": 6.041340970993042,
      "epoch": 1.5928924876293298,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004752182106789467,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16832517087459564,
      "num_tokens": 31173567.0,
      "step": 15935
    },
    {
      "entropy": 6.10235013961792,
      "epoch": 1.5933923126905583,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047520192550023344,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16383893936872482,
      "num_tokens": 31183654.0,
      "step": 15940
    },
    {
      "entropy": 6.09974308013916,
      "epoch": 1.5938921377517867,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004751856352844913,
      "loss": 5.887,
      "mean_token_accuracy": 0.16141289919614793,
      "num_tokens": 31193814.0,
      "step": 15945
    },
    {
      "entropy": 6.09717493057251,
      "epoch": 1.5943919628130154,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047516934003213,
      "loss": 5.8733,
      "mean_token_accuracy": 0.15989348739385606,
      "num_tokens": 31203834.0,
      "step": 15950
    },
    {
      "entropy": 6.123807334899903,
      "epoch": 1.5948917878742441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047515303974355965,
      "loss": 5.9691,
      "mean_token_accuracy": 0.15232257097959517,
      "num_tokens": 31214625.0,
      "step": 15955
    },
    {
      "entropy": 6.217723035812378,
      "epoch": 1.5953916129354726,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047513673441919036,
      "loss": 5.9145,
      "mean_token_accuracy": 0.15624975115060807,
      "num_tokens": 31224125.0,
      "step": 15960
    },
    {
      "entropy": 6.10897536277771,
      "epoch": 1.595891437996701,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004751204240594323,
      "loss": 5.8833,
      "mean_token_accuracy": 0.16300169378519058,
      "num_tokens": 31233869.0,
      "step": 15965
    },
    {
      "entropy": 6.150825548171997,
      "epoch": 1.5963912630579298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004751041086646959,
      "loss": 5.9999,
      "mean_token_accuracy": 0.15502840131521226,
      "num_tokens": 31244652.0,
      "step": 15970
    },
    {
      "entropy": 6.168025875091553,
      "epoch": 1.5968910881191583,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004750877882353915,
      "loss": 5.9185,
      "mean_token_accuracy": 0.16045887023210526,
      "num_tokens": 31254754.0,
      "step": 15975
    },
    {
      "entropy": 6.186917161941528,
      "epoch": 1.5973909131803867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047507146277192997,
      "loss": 5.9533,
      "mean_token_accuracy": 0.16073550283908844,
      "num_tokens": 31263768.0,
      "step": 15980
    },
    {
      "entropy": 6.060762739181518,
      "epoch": 1.5978907382416154,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004750551322747219,
      "loss": 5.8544,
      "mean_token_accuracy": 0.1638706371188164,
      "num_tokens": 31273295.0,
      "step": 15985
    },
    {
      "entropy": 6.1380720138549805,
      "epoch": 1.5983905633028441,
      "grad_norm": 1.125,
      "learning_rate": 0.0004750387967441781,
      "loss": 5.94,
      "mean_token_accuracy": 0.15558256804943085,
      "num_tokens": 31284996.0,
      "step": 15990
    },
    {
      "entropy": 6.1803545475006105,
      "epoch": 1.5988903883640726,
      "grad_norm": 1.6875,
      "learning_rate": 0.00047502245618070976,
      "loss": 5.8728,
      "mean_token_accuracy": 0.1548008918762207,
      "num_tokens": 31294496.0,
      "step": 15995
    },
    {
      "entropy": 6.093459224700927,
      "epoch": 1.599390213425301,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047500611058472783,
      "loss": 5.7765,
      "mean_token_accuracy": 0.1649110808968544,
      "num_tokens": 31303655.0,
      "step": 16000
    },
    {
      "entropy": 6.060240697860718,
      "epoch": 1.5998900384865298,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047498975995664356,
      "loss": 5.8258,
      "mean_token_accuracy": 0.16718256324529648,
      "num_tokens": 31313058.0,
      "step": 16005
    },
    {
      "entropy": 6.090587186813354,
      "epoch": 1.6003898635477583,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004749734042968685,
      "loss": 5.8166,
      "mean_token_accuracy": 0.163027885556221,
      "num_tokens": 31323214.0,
      "step": 16010
    },
    {
      "entropy": 6.159936094284058,
      "epoch": 1.6008896886089867,
      "grad_norm": 1.171875,
      "learning_rate": 0.000474957043605814,
      "loss": 5.9489,
      "mean_token_accuracy": 0.159230537712574,
      "num_tokens": 31333691.0,
      "step": 16015
    },
    {
      "entropy": 6.241786479949951,
      "epoch": 1.6013895136702154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047494067788389174,
      "loss": 5.8989,
      "mean_token_accuracy": 0.1530660316348076,
      "num_tokens": 31342240.0,
      "step": 16020
    },
    {
      "entropy": 6.125791120529175,
      "epoch": 1.6018893387314441,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047492430713151347,
      "loss": 5.7988,
      "mean_token_accuracy": 0.1713169813156128,
      "num_tokens": 31351955.0,
      "step": 16025
    },
    {
      "entropy": 6.009016275405884,
      "epoch": 1.6023891637926726,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004749079313490911,
      "loss": 5.7917,
      "mean_token_accuracy": 0.1676168918609619,
      "num_tokens": 31361583.0,
      "step": 16030
    },
    {
      "entropy": 6.04360671043396,
      "epoch": 1.602888988853901,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047489155053703655,
      "loss": 5.8588,
      "mean_token_accuracy": 0.161922425031662,
      "num_tokens": 31371219.0,
      "step": 16035
    },
    {
      "entropy": 6.1707839012146,
      "epoch": 1.6033888139151298,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004748751646957621,
      "loss": 5.8687,
      "mean_token_accuracy": 0.1556646019220352,
      "num_tokens": 31381561.0,
      "step": 16040
    },
    {
      "entropy": 6.215362405776977,
      "epoch": 1.6038886389763582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004748587738256799,
      "loss": 5.8291,
      "mean_token_accuracy": 0.16217069327831268,
      "num_tokens": 31391385.0,
      "step": 16045
    },
    {
      "entropy": 6.129397249221801,
      "epoch": 1.6043884640375867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004748423779272024,
      "loss": 5.8546,
      "mean_token_accuracy": 0.1555620014667511,
      "num_tokens": 31400817.0,
      "step": 16050
    },
    {
      "entropy": 6.07524847984314,
      "epoch": 1.6048882890988154,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047482597700074204,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15766070932149887,
      "num_tokens": 31411460.0,
      "step": 16055
    },
    {
      "entropy": 6.062694978713989,
      "epoch": 1.6053881141600441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047480957104671157,
      "loss": 5.7927,
      "mean_token_accuracy": 0.1720308169722557,
      "num_tokens": 31420407.0,
      "step": 16060
    },
    {
      "entropy": 6.164383363723755,
      "epoch": 1.6058879392212726,
      "grad_norm": 1.5,
      "learning_rate": 0.0004747931600655237,
      "loss": 5.9563,
      "mean_token_accuracy": 0.1534573644399643,
      "num_tokens": 31430964.0,
      "step": 16065
    },
    {
      "entropy": 6.03290958404541,
      "epoch": 1.606387764282501,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047477674405759136,
      "loss": 5.7669,
      "mean_token_accuracy": 0.16743373572826387,
      "num_tokens": 31440143.0,
      "step": 16070
    },
    {
      "entropy": 6.0419135093688965,
      "epoch": 1.6068875893437298,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004747603230233275,
      "loss": 5.85,
      "mean_token_accuracy": 0.1649542585015297,
      "num_tokens": 31449615.0,
      "step": 16075
    },
    {
      "entropy": 6.161441850662231,
      "epoch": 1.6073874144049582,
      "grad_norm": 1.25,
      "learning_rate": 0.0004747438969631453,
      "loss": 5.9351,
      "mean_token_accuracy": 0.14877657666802407,
      "num_tokens": 31459238.0,
      "step": 16080
    },
    {
      "entropy": 6.191607427597046,
      "epoch": 1.6078872394661867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004747274658774581,
      "loss": 5.8413,
      "mean_token_accuracy": 0.17061431854963302,
      "num_tokens": 31468540.0,
      "step": 16085
    },
    {
      "entropy": 6.113143539428711,
      "epoch": 1.6083870645274154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004747110297666793,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16448079347610473,
      "num_tokens": 31478999.0,
      "step": 16090
    },
    {
      "entropy": 6.081920337677002,
      "epoch": 1.6088868895886441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004746945886312223,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16099572479724883,
      "num_tokens": 31490237.0,
      "step": 16095
    },
    {
      "entropy": 6.111219072341919,
      "epoch": 1.6093867146498726,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004746781424715009,
      "loss": 5.8946,
      "mean_token_accuracy": 0.16183358132839204,
      "num_tokens": 31500250.0,
      "step": 16100
    },
    {
      "entropy": 6.2043990135192875,
      "epoch": 1.609886539711101,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047466169128792873,
      "loss": 5.9606,
      "mean_token_accuracy": 0.1561687022447586,
      "num_tokens": 31511559.0,
      "step": 16105
    },
    {
      "entropy": 6.195315217971801,
      "epoch": 1.6103863647723298,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004746452350809198,
      "loss": 5.9075,
      "mean_token_accuracy": 0.15455575436353683,
      "num_tokens": 31520189.0,
      "step": 16110
    },
    {
      "entropy": 6.021544742584228,
      "epoch": 1.6108861898335582,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047462877385088815,
      "loss": 5.7551,
      "mean_token_accuracy": 0.1642581880092621,
      "num_tokens": 31529361.0,
      "step": 16115
    },
    {
      "entropy": 6.0735876083374025,
      "epoch": 1.6113860148947867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047461230759824794,
      "loss": 5.9072,
      "mean_token_accuracy": 0.15475402921438217,
      "num_tokens": 31539530.0,
      "step": 16120
    },
    {
      "entropy": 6.2165834426879885,
      "epoch": 1.6118858399560154,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047459583632341343,
      "loss": 5.9555,
      "mean_token_accuracy": 0.15416406244039535,
      "num_tokens": 31549166.0,
      "step": 16125
    },
    {
      "entropy": 6.110639572143555,
      "epoch": 1.6123856650172441,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047457936002679905,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16346166282892227,
      "num_tokens": 31558643.0,
      "step": 16130
    },
    {
      "entropy": 6.117588806152344,
      "epoch": 1.6128854900784724,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047456287870881924,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16515784710645676,
      "num_tokens": 31568752.0,
      "step": 16135
    },
    {
      "entropy": 6.128545570373535,
      "epoch": 1.613385315139701,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004745463923698888,
      "loss": 5.9232,
      "mean_token_accuracy": 0.1560824006795883,
      "num_tokens": 31577894.0,
      "step": 16140
    },
    {
      "entropy": 6.061765956878662,
      "epoch": 1.6138851402009298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004745299010104226,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16323441565036773,
      "num_tokens": 31586348.0,
      "step": 16145
    },
    {
      "entropy": 6.112632131576538,
      "epoch": 1.6143849652621582,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047451340463083527,
      "loss": 5.8905,
      "mean_token_accuracy": 0.15955076217651368,
      "num_tokens": 31596472.0,
      "step": 16150
    },
    {
      "entropy": 6.178978013992309,
      "epoch": 1.6148847903233867,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047449690323154206,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16583316922187805,
      "num_tokens": 31606342.0,
      "step": 16155
    },
    {
      "entropy": 6.136794233322144,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047448039681295814,
      "loss": 5.9253,
      "mean_token_accuracy": 0.1568926379084587,
      "num_tokens": 31615548.0,
      "step": 16160
    },
    {
      "entropy": 6.068896913528443,
      "epoch": 1.6158844404458441,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047446388537549867,
      "loss": 5.8977,
      "mean_token_accuracy": 0.1589074045419693,
      "num_tokens": 31626290.0,
      "step": 16165
    },
    {
      "entropy": 6.137011861801147,
      "epoch": 1.6163842655070724,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004744473689195793,
      "loss": 5.8749,
      "mean_token_accuracy": 0.1628820925951004,
      "num_tokens": 31635620.0,
      "step": 16170
    },
    {
      "entropy": 6.09233512878418,
      "epoch": 1.616884090568301,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004744308474456154,
      "loss": 5.9009,
      "mean_token_accuracy": 0.16834541410207748,
      "num_tokens": 31647168.0,
      "step": 16175
    },
    {
      "entropy": 6.14128475189209,
      "epoch": 1.6173839156295298,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004744143209540227,
      "loss": 5.7856,
      "mean_token_accuracy": 0.16179870814085007,
      "num_tokens": 31656173.0,
      "step": 16180
    },
    {
      "entropy": 6.115216541290283,
      "epoch": 1.6178837406907582,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004743977894452169,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16130209863185882,
      "num_tokens": 31665224.0,
      "step": 16185
    },
    {
      "entropy": 6.0346660137176515,
      "epoch": 1.6183835657519867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004743812529196142,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16271012425422668,
      "num_tokens": 31674623.0,
      "step": 16190
    },
    {
      "entropy": 6.124181032180786,
      "epoch": 1.6188833908132154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004743647113776304,
      "loss": 5.8561,
      "mean_token_accuracy": 0.16743426769971848,
      "num_tokens": 31684565.0,
      "step": 16195
    },
    {
      "entropy": 6.140634250640869,
      "epoch": 1.619383215874444,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047434816481968175,
      "loss": 5.8834,
      "mean_token_accuracy": 0.1602107010781765,
      "num_tokens": 31694885.0,
      "step": 16200
    },
    {
      "entropy": 6.133418130874634,
      "epoch": 1.6198830409356724,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047433161324618456,
      "loss": 5.8822,
      "mean_token_accuracy": 0.15877734124660492,
      "num_tokens": 31704077.0,
      "step": 16205
    },
    {
      "entropy": 6.19438624382019,
      "epoch": 1.620382865996901,
      "grad_norm": 1.25,
      "learning_rate": 0.0004743150566575553,
      "loss": 5.9176,
      "mean_token_accuracy": 0.15162158161401748,
      "num_tokens": 31713773.0,
      "step": 16210
    },
    {
      "entropy": 6.149185276031494,
      "epoch": 1.6208826910581298,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004742984950542105,
      "loss": 5.9418,
      "mean_token_accuracy": 0.16136410534381868,
      "num_tokens": 31724020.0,
      "step": 16215
    },
    {
      "entropy": 6.093222522735596,
      "epoch": 1.6213825161193582,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047428192843656684,
      "loss": 5.8878,
      "mean_token_accuracy": 0.15722094476222992,
      "num_tokens": 31733992.0,
      "step": 16220
    },
    {
      "entropy": 6.180321788787841,
      "epoch": 1.6218823411805867,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047426535680504105,
      "loss": 5.9117,
      "mean_token_accuracy": 0.15999206602573396,
      "num_tokens": 31745153.0,
      "step": 16225
    },
    {
      "entropy": 6.148810529708863,
      "epoch": 1.6223821662418154,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047424878016005036,
      "loss": 5.8292,
      "mean_token_accuracy": 0.16316247582435608,
      "num_tokens": 31755110.0,
      "step": 16230
    },
    {
      "entropy": 6.132559299468994,
      "epoch": 1.622881991303044,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004742321985020115,
      "loss": 5.859,
      "mean_token_accuracy": 0.1618788167834282,
      "num_tokens": 31764219.0,
      "step": 16235
    },
    {
      "entropy": 6.082424545288086,
      "epoch": 1.6233818163642724,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004742156118313418,
      "loss": 5.759,
      "mean_token_accuracy": 0.15875313431024551,
      "num_tokens": 31773597.0,
      "step": 16240
    },
    {
      "entropy": 6.145002698898315,
      "epoch": 1.623881641425501,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047419902014845855,
      "loss": 5.8687,
      "mean_token_accuracy": 0.16014937311410904,
      "num_tokens": 31783302.0,
      "step": 16245
    },
    {
      "entropy": 6.0814916610717775,
      "epoch": 1.6243814664867298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004741824234537793,
      "loss": 5.8809,
      "mean_token_accuracy": 0.15806918144226073,
      "num_tokens": 31793127.0,
      "step": 16250
    },
    {
      "entropy": 6.087811088562011,
      "epoch": 1.6248812915479582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047416582174772143,
      "loss": 5.8905,
      "mean_token_accuracy": 0.16259425282478332,
      "num_tokens": 31802756.0,
      "step": 16255
    },
    {
      "entropy": 6.133480596542358,
      "epoch": 1.6253811166091867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047414921503070276,
      "loss": 5.9332,
      "mean_token_accuracy": 0.1575881764292717,
      "num_tokens": 31813390.0,
      "step": 16260
    },
    {
      "entropy": 6.156129312515259,
      "epoch": 1.6258809416704154,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047413260330314116,
      "loss": 5.9113,
      "mean_token_accuracy": 0.16229541525244712,
      "num_tokens": 31823304.0,
      "step": 16265
    },
    {
      "entropy": 6.148132419586181,
      "epoch": 1.626380766731644,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004741159865654544,
      "loss": 5.8581,
      "mean_token_accuracy": 0.1669325590133667,
      "num_tokens": 31832878.0,
      "step": 16270
    },
    {
      "entropy": 6.10863766670227,
      "epoch": 1.6268805917928724,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047409936481806076,
      "loss": 5.8828,
      "mean_token_accuracy": 0.1642310842871666,
      "num_tokens": 31841753.0,
      "step": 16275
    },
    {
      "entropy": 6.024941301345825,
      "epoch": 1.627380416854101,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004740827380613782,
      "loss": 5.7254,
      "mean_token_accuracy": 0.16844381392002106,
      "num_tokens": 31851451.0,
      "step": 16280
    },
    {
      "entropy": 6.1089280128479,
      "epoch": 1.6278802419153298,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047406610629582526,
      "loss": 5.8463,
      "mean_token_accuracy": 0.16606414914131165,
      "num_tokens": 31860885.0,
      "step": 16285
    },
    {
      "entropy": 6.032342576980591,
      "epoch": 1.6283800669765582,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047404946952182034,
      "loss": 5.7832,
      "mean_token_accuracy": 0.1683499440550804,
      "num_tokens": 31870056.0,
      "step": 16290
    },
    {
      "entropy": 6.101997709274292,
      "epoch": 1.6288798920377867,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004740328277397819,
      "loss": 5.8493,
      "mean_token_accuracy": 0.16689916849136352,
      "num_tokens": 31879513.0,
      "step": 16295
    },
    {
      "entropy": 6.14493498802185,
      "epoch": 1.6293797170990154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004740161809501288,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1655614197254181,
      "num_tokens": 31890288.0,
      "step": 16300
    },
    {
      "entropy": 6.141741609573364,
      "epoch": 1.6298795421602439,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047399952915327975,
      "loss": 5.8276,
      "mean_token_accuracy": 0.15968619138002396,
      "num_tokens": 31900168.0,
      "step": 16305
    },
    {
      "entropy": 6.112282180786133,
      "epoch": 1.6303793672214724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004739828723496538,
      "loss": 5.8011,
      "mean_token_accuracy": 0.15956270098686218,
      "num_tokens": 31908788.0,
      "step": 16310
    },
    {
      "entropy": 6.0826027393341064,
      "epoch": 1.630879192282701,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047396621053967,
      "loss": 5.8274,
      "mean_token_accuracy": 0.1686628744006157,
      "num_tokens": 31916968.0,
      "step": 16315
    },
    {
      "entropy": 6.120118284225464,
      "epoch": 1.6313790173439298,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047394954372374743,
      "loss": 5.8922,
      "mean_token_accuracy": 0.15677086561918258,
      "num_tokens": 31927205.0,
      "step": 16320
    },
    {
      "entropy": 6.161373233795166,
      "epoch": 1.6318788424051582,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047393287190230566,
      "loss": 5.831,
      "mean_token_accuracy": 0.16338104605674744,
      "num_tokens": 31937001.0,
      "step": 16325
    },
    {
      "entropy": 6.128987836837768,
      "epoch": 1.6323786674663867,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004739161950757639,
      "loss": 5.8521,
      "mean_token_accuracy": 0.16076452583074569,
      "num_tokens": 31946915.0,
      "step": 16330
    },
    {
      "entropy": 6.141094636917114,
      "epoch": 1.6328784925276154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000473899513244542,
      "loss": 5.8468,
      "mean_token_accuracy": 0.15981499701738358,
      "num_tokens": 31956615.0,
      "step": 16335
    },
    {
      "entropy": 6.117867422103882,
      "epoch": 1.6333783175888439,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004738828264090594,
      "loss": 5.9059,
      "mean_token_accuracy": 0.15665557831525803,
      "num_tokens": 31967203.0,
      "step": 16340
    },
    {
      "entropy": 6.097457361221314,
      "epoch": 1.6338781426500724,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047386613456973614,
      "loss": 5.7691,
      "mean_token_accuracy": 0.16266042739152908,
      "num_tokens": 31976736.0,
      "step": 16345
    },
    {
      "entropy": 5.999594831466675,
      "epoch": 1.634377967711301,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004738494377269921,
      "loss": 5.7999,
      "mean_token_accuracy": 0.17103495746850966,
      "num_tokens": 31987000.0,
      "step": 16350
    },
    {
      "entropy": 6.140643501281739,
      "epoch": 1.6348777927725298,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047383273588124736,
      "loss": 5.9814,
      "mean_token_accuracy": 0.16301832646131514,
      "num_tokens": 31996238.0,
      "step": 16355
    },
    {
      "entropy": 6.209257459640503,
      "epoch": 1.6353776178337582,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004738160290329222,
      "loss": 5.8986,
      "mean_token_accuracy": 0.15926598608493805,
      "num_tokens": 32005785.0,
      "step": 16360
    },
    {
      "entropy": 6.149421834945679,
      "epoch": 1.6358774428949867,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047379931718243684,
      "loss": 5.9468,
      "mean_token_accuracy": 0.15843429863452912,
      "num_tokens": 32017140.0,
      "step": 16365
    },
    {
      "entropy": 6.037563800811768,
      "epoch": 1.6363772679562154,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004737826003302118,
      "loss": 5.7389,
      "mean_token_accuracy": 0.17751652449369432,
      "num_tokens": 32026431.0,
      "step": 16370
    },
    {
      "entropy": 6.173066234588623,
      "epoch": 1.6368770930174439,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047376587847666767,
      "loss": 6.0066,
      "mean_token_accuracy": 0.15997146666049958,
      "num_tokens": 32036211.0,
      "step": 16375
    },
    {
      "entropy": 6.189323377609253,
      "epoch": 1.6373769180786724,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004737491516222253,
      "loss": 5.8857,
      "mean_token_accuracy": 0.15492396503686906,
      "num_tokens": 32047231.0,
      "step": 16380
    },
    {
      "entropy": 6.108688831329346,
      "epoch": 1.637876743139901,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047373241976730537,
      "loss": 5.87,
      "mean_token_accuracy": 0.16174225211143495,
      "num_tokens": 32057649.0,
      "step": 16385
    },
    {
      "entropy": 6.159617614746094,
      "epoch": 1.6383765682011298,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004737156829123289,
      "loss": 5.8311,
      "mean_token_accuracy": 0.16293939277529718,
      "num_tokens": 32066208.0,
      "step": 16390
    },
    {
      "entropy": 6.17410717010498,
      "epoch": 1.638876393262358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047369894105771693,
      "loss": 5.898,
      "mean_token_accuracy": 0.1624485582113266,
      "num_tokens": 32076451.0,
      "step": 16395
    },
    {
      "entropy": 6.17270131111145,
      "epoch": 1.6393762183235867,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047368219420389077,
      "loss": 5.9283,
      "mean_token_accuracy": 0.15363367199897765,
      "num_tokens": 32085642.0,
      "step": 16400
    },
    {
      "entropy": 6.135274600982666,
      "epoch": 1.6398760433848154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004736654423512717,
      "loss": 5.9008,
      "mean_token_accuracy": 0.1581696465611458,
      "num_tokens": 32094350.0,
      "step": 16405
    },
    {
      "entropy": 6.149425315856933,
      "epoch": 1.6403758684460439,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004736486855002813,
      "loss": 5.8746,
      "mean_token_accuracy": 0.1614062562584877,
      "num_tokens": 32104981.0,
      "step": 16410
    },
    {
      "entropy": 6.1327746391296385,
      "epoch": 1.6408756935072724,
      "grad_norm": 1.4765625,
      "learning_rate": 0.000473631923651341,
      "loss": 5.8423,
      "mean_token_accuracy": 0.16224468797445296,
      "num_tokens": 32114198.0,
      "step": 16415
    },
    {
      "entropy": 6.147230529785157,
      "epoch": 1.641375518568501,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004736151568048726,
      "loss": 5.8517,
      "mean_token_accuracy": 0.1621714487671852,
      "num_tokens": 32124622.0,
      "step": 16420
    },
    {
      "entropy": 6.141799402236939,
      "epoch": 1.6418753436297298,
      "grad_norm": 1.25,
      "learning_rate": 0.000473598384961298,
      "loss": 5.8976,
      "mean_token_accuracy": 0.152253158390522,
      "num_tokens": 32134466.0,
      "step": 16425
    },
    {
      "entropy": 6.195237159729004,
      "epoch": 1.642375168690958,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047358160812103925,
      "loss": 5.9206,
      "mean_token_accuracy": 0.15897852778434754,
      "num_tokens": 32144104.0,
      "step": 16430
    },
    {
      "entropy": 6.105606269836426,
      "epoch": 1.6428749937521867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004735648262845182,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16012743189930917,
      "num_tokens": 32156432.0,
      "step": 16435
    },
    {
      "entropy": 6.0935523986816404,
      "epoch": 1.6433748188134154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047354803945215727,
      "loss": 5.9146,
      "mean_token_accuracy": 0.1606764554977417,
      "num_tokens": 32166853.0,
      "step": 16440
    },
    {
      "entropy": 6.080415821075439,
      "epoch": 1.6438746438746439,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047353124762437865,
      "loss": 5.78,
      "mean_token_accuracy": 0.17210648208856583,
      "num_tokens": 32175604.0,
      "step": 16445
    },
    {
      "entropy": 6.091891956329346,
      "epoch": 1.6443744689358724,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004735144508016051,
      "loss": 5.8507,
      "mean_token_accuracy": 0.162253800034523,
      "num_tokens": 32185878.0,
      "step": 16450
    },
    {
      "entropy": 6.15599889755249,
      "epoch": 1.644874293997101,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004734976489842589,
      "loss": 5.9563,
      "mean_token_accuracy": 0.153241053968668,
      "num_tokens": 32196061.0,
      "step": 16455
    },
    {
      "entropy": 6.160298204421997,
      "epoch": 1.6453741190583295,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000473480842172763,
      "loss": 5.8634,
      "mean_token_accuracy": 0.1667049780488014,
      "num_tokens": 32207470.0,
      "step": 16460
    },
    {
      "entropy": 6.1717366695404055,
      "epoch": 1.645873944119558,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047346403036754023,
      "loss": 5.9602,
      "mean_token_accuracy": 0.15624635219573973,
      "num_tokens": 32216660.0,
      "step": 16465
    },
    {
      "entropy": 6.145235395431518,
      "epoch": 1.6463737691807867,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00047344721356901344,
      "loss": 5.8237,
      "mean_token_accuracy": 0.16484262943267822,
      "num_tokens": 32226472.0,
      "step": 16470
    },
    {
      "entropy": 6.159499263763427,
      "epoch": 1.6468735942420154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004734303917776059,
      "loss": 5.9434,
      "mean_token_accuracy": 0.15112022012472154,
      "num_tokens": 32235841.0,
      "step": 16475
    },
    {
      "entropy": 6.049981498718262,
      "epoch": 1.6473734193032439,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047341356499374065,
      "loss": 5.7124,
      "mean_token_accuracy": 0.167866487801075,
      "num_tokens": 32245388.0,
      "step": 16480
    },
    {
      "entropy": 6.1179088115692135,
      "epoch": 1.6478732443644724,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004733967332178412,
      "loss": 5.9051,
      "mean_token_accuracy": 0.16497117429971694,
      "num_tokens": 32255089.0,
      "step": 16485
    },
    {
      "entropy": 6.032177448272705,
      "epoch": 1.648373069425701,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004733798964503311,
      "loss": 5.7548,
      "mean_token_accuracy": 0.16922059655189514,
      "num_tokens": 32265196.0,
      "step": 16490
    },
    {
      "entropy": 6.035744762420654,
      "epoch": 1.6488728944869295,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004733630546916338,
      "loss": 5.7409,
      "mean_token_accuracy": 0.16643005460500718,
      "num_tokens": 32276734.0,
      "step": 16495
    },
    {
      "entropy": 6.12573823928833,
      "epoch": 1.649372719548158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047334620794217306,
      "loss": 5.883,
      "mean_token_accuracy": 0.16074480563402177,
      "num_tokens": 32286098.0,
      "step": 16500
    },
    {
      "entropy": 6.098812103271484,
      "epoch": 1.6498725446093867,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004733293562023727,
      "loss": 5.9141,
      "mean_token_accuracy": 0.15626098960638046,
      "num_tokens": 32295123.0,
      "step": 16505
    },
    {
      "entropy": 6.108363008499145,
      "epoch": 1.6503723696706154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047331249947265685,
      "loss": 5.8224,
      "mean_token_accuracy": 0.17050332576036453,
      "num_tokens": 32304016.0,
      "step": 16510
    },
    {
      "entropy": 6.116782474517822,
      "epoch": 1.6508721947318439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004732956377534496,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1608186572790146,
      "num_tokens": 32313437.0,
      "step": 16515
    },
    {
      "entropy": 6.175250625610351,
      "epoch": 1.6513720197930724,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000473278771045175,
      "loss": 5.8753,
      "mean_token_accuracy": 0.16361228227615357,
      "num_tokens": 32322966.0,
      "step": 16520
    },
    {
      "entropy": 6.083446407318116,
      "epoch": 1.651871844854301,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047326189934825764,
      "loss": 5.8835,
      "mean_token_accuracy": 0.157442407310009,
      "num_tokens": 32333555.0,
      "step": 16525
    },
    {
      "entropy": 6.1388014793396,
      "epoch": 1.6523716699155295,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004732450226631219,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16032904684543609,
      "num_tokens": 32343946.0,
      "step": 16530
    },
    {
      "entropy": 6.162550926208496,
      "epoch": 1.652871494976758,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047322814099019236,
      "loss": 5.8461,
      "mean_token_accuracy": 0.15958501100540162,
      "num_tokens": 32353932.0,
      "step": 16535
    },
    {
      "entropy": 6.011513662338257,
      "epoch": 1.6533713200379867,
      "grad_norm": 1.25,
      "learning_rate": 0.0004732112543298938,
      "loss": 5.747,
      "mean_token_accuracy": 0.17167413383722305,
      "num_tokens": 32363156.0,
      "step": 16540
    },
    {
      "entropy": 6.047575855255127,
      "epoch": 1.6538711450992154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004731943626826511,
      "loss": 5.8009,
      "mean_token_accuracy": 0.16844077855348588,
      "num_tokens": 32371987.0,
      "step": 16545
    },
    {
      "entropy": 6.168126201629638,
      "epoch": 1.6543709701604439,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047317746604888923,
      "loss": 5.8934,
      "mean_token_accuracy": 0.15469537675380707,
      "num_tokens": 32381391.0,
      "step": 16550
    },
    {
      "entropy": 6.151556968688965,
      "epoch": 1.6548707952216724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047316056442903333,
      "loss": 5.8943,
      "mean_token_accuracy": 0.16380312144756318,
      "num_tokens": 32390980.0,
      "step": 16555
    },
    {
      "entropy": 6.061680698394776,
      "epoch": 1.655370620282901,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004731436578235086,
      "loss": 5.8049,
      "mean_token_accuracy": 0.16317197382450105,
      "num_tokens": 32399240.0,
      "step": 16560
    },
    {
      "entropy": 6.132208967208863,
      "epoch": 1.6558704453441295,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047312674623274046,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16012369841337204,
      "num_tokens": 32409793.0,
      "step": 16565
    },
    {
      "entropy": 6.1559814453125,
      "epoch": 1.656370270405358,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004731098296571543,
      "loss": 5.8514,
      "mean_token_accuracy": 0.1619309403002262,
      "num_tokens": 32420658.0,
      "step": 16570
    },
    {
      "entropy": 6.258848571777344,
      "epoch": 1.6568700954665867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047309290809717584,
      "loss": 6.0333,
      "mean_token_accuracy": 0.15285867303609849,
      "num_tokens": 32430044.0,
      "step": 16575
    },
    {
      "entropy": 6.133506441116333,
      "epoch": 1.6573699205278154,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047307598155323083,
      "loss": 5.8828,
      "mean_token_accuracy": 0.16140941083431243,
      "num_tokens": 32439913.0,
      "step": 16580
    },
    {
      "entropy": 6.136096382141114,
      "epoch": 1.6578697455890439,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000473059050025745,
      "loss": 5.8217,
      "mean_token_accuracy": 0.15958474576473236,
      "num_tokens": 32449125.0,
      "step": 16585
    },
    {
      "entropy": 6.133864784240723,
      "epoch": 1.6583695706502724,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004730421135151445,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16147243082523347,
      "num_tokens": 32458549.0,
      "step": 16590
    },
    {
      "entropy": 6.088213729858398,
      "epoch": 1.658869395711501,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004730251720218554,
      "loss": 5.807,
      "mean_token_accuracy": 0.16491982489824294,
      "num_tokens": 32468860.0,
      "step": 16595
    },
    {
      "entropy": 6.15969181060791,
      "epoch": 1.6593692207727295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047300822554630393,
      "loss": 5.8215,
      "mean_token_accuracy": 0.16563009917736055,
      "num_tokens": 32479090.0,
      "step": 16600
    },
    {
      "entropy": 6.082017183303833,
      "epoch": 1.659869045833958,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004729912740889164,
      "loss": 5.7672,
      "mean_token_accuracy": 0.17205040752887726,
      "num_tokens": 32488611.0,
      "step": 16605
    },
    {
      "entropy": 6.058082437515258,
      "epoch": 1.6603688708951867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047297431765011945,
      "loss": 5.7833,
      "mean_token_accuracy": 0.16363301426172255,
      "num_tokens": 32497821.0,
      "step": 16610
    },
    {
      "entropy": 6.009025239944458,
      "epoch": 1.6608686959564154,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047295735623033953,
      "loss": 5.7743,
      "mean_token_accuracy": 0.17511193752288817,
      "num_tokens": 32507324.0,
      "step": 16615
    },
    {
      "entropy": 6.059767055511474,
      "epoch": 1.6613685210176439,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004729403898300035,
      "loss": 5.7729,
      "mean_token_accuracy": 0.17438238710165024,
      "num_tokens": 32517498.0,
      "step": 16620
    },
    {
      "entropy": 6.073728275299072,
      "epoch": 1.6618683460788723,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047292341844953813,
      "loss": 5.8331,
      "mean_token_accuracy": 0.16623251736164094,
      "num_tokens": 32526527.0,
      "step": 16625
    },
    {
      "entropy": 6.124826669692993,
      "epoch": 1.662368171140101,
      "grad_norm": 2.3125,
      "learning_rate": 0.0004729064420893706,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16758653223514558,
      "num_tokens": 32535017.0,
      "step": 16630
    },
    {
      "entropy": 5.94693284034729,
      "epoch": 1.6628679962013295,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004728894607499278,
      "loss": 5.6691,
      "mean_token_accuracy": 0.18654691874980928,
      "num_tokens": 32545687.0,
      "step": 16635
    },
    {
      "entropy": 6.085852336883545,
      "epoch": 1.663367821262558,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004728724744316371,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1632249027490616,
      "num_tokens": 32554876.0,
      "step": 16640
    },
    {
      "entropy": 6.182626056671142,
      "epoch": 1.6638676463237867,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047285548313492594,
      "loss": 5.9113,
      "mean_token_accuracy": 0.1599644050002098,
      "num_tokens": 32565453.0,
      "step": 16645
    },
    {
      "entropy": 6.15461277961731,
      "epoch": 1.6643674713850154,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004728384868602217,
      "loss": 5.8086,
      "mean_token_accuracy": 0.1610132187604904,
      "num_tokens": 32573882.0,
      "step": 16650
    },
    {
      "entropy": 6.149443531036377,
      "epoch": 1.6648672964462436,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000472821485607952,
      "loss": 5.9494,
      "mean_token_accuracy": 0.15137023776769637,
      "num_tokens": 32583182.0,
      "step": 16655
    },
    {
      "entropy": 6.130323266983032,
      "epoch": 1.6653671215074723,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004728044793785446,
      "loss": 5.9129,
      "mean_token_accuracy": 0.153822723031044,
      "num_tokens": 32593405.0,
      "step": 16660
    },
    {
      "entropy": 6.169586086273194,
      "epoch": 1.665866946568701,
      "grad_norm": 1.609375,
      "learning_rate": 0.00047278746817242754,
      "loss": 5.8752,
      "mean_token_accuracy": 0.15612266287207605,
      "num_tokens": 32603451.0,
      "step": 16665
    },
    {
      "entropy": 6.157718324661255,
      "epoch": 1.6663667716299295,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004727704519900286,
      "loss": 5.9781,
      "mean_token_accuracy": 0.16066828444600106,
      "num_tokens": 32614054.0,
      "step": 16670
    },
    {
      "entropy": 6.060660886764526,
      "epoch": 1.666866596691158,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000472753430831776,
      "loss": 5.8365,
      "mean_token_accuracy": 0.1647493675351143,
      "num_tokens": 32623154.0,
      "step": 16675
    },
    {
      "entropy": 6.146534490585327,
      "epoch": 1.6673664217523867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047273640469809786,
      "loss": 5.9999,
      "mean_token_accuracy": 0.1561283752322197,
      "num_tokens": 32633252.0,
      "step": 16680
    },
    {
      "entropy": 6.058470630645752,
      "epoch": 1.6678662468136154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004727193735894227,
      "loss": 5.7822,
      "mean_token_accuracy": 0.16794995814561844,
      "num_tokens": 32643681.0,
      "step": 16685
    },
    {
      "entropy": 6.050247716903686,
      "epoch": 1.6683660718748436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047270233750617907,
      "loss": 5.6962,
      "mean_token_accuracy": 0.1738314300775528,
      "num_tokens": 32653094.0,
      "step": 16690
    },
    {
      "entropy": 6.03511528968811,
      "epoch": 1.6688658969360723,
      "grad_norm": 1.671875,
      "learning_rate": 0.00047268529644879545,
      "loss": 5.7336,
      "mean_token_accuracy": 0.16250773817300795,
      "num_tokens": 32663795.0,
      "step": 16695
    },
    {
      "entropy": 6.101651763916015,
      "epoch": 1.669365721997301,
      "grad_norm": 1.75,
      "learning_rate": 0.00047266825041770066,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1655741348862648,
      "num_tokens": 32673763.0,
      "step": 16700
    },
    {
      "entropy": 6.014226627349854,
      "epoch": 1.6698655470585295,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004726511994133235,
      "loss": 5.795,
      "mean_token_accuracy": 0.16741044968366622,
      "num_tokens": 32683436.0,
      "step": 16705
    },
    {
      "entropy": 6.113631105422973,
      "epoch": 1.670365372119758,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047263414343609307,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1643044039607048,
      "num_tokens": 32692563.0,
      "step": 16710
    },
    {
      "entropy": 6.082835912704468,
      "epoch": 1.6708651971809867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047261708248643846,
      "loss": 5.8465,
      "mean_token_accuracy": 0.1624145582318306,
      "num_tokens": 32700973.0,
      "step": 16715
    },
    {
      "entropy": 6.047251462936401,
      "epoch": 1.6713650222422152,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004726000165647888,
      "loss": 5.7722,
      "mean_token_accuracy": 0.1700923338532448,
      "num_tokens": 32710299.0,
      "step": 16720
    },
    {
      "entropy": 6.1271240234375,
      "epoch": 1.6718648473034436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004725829456715737,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16144029945135116,
      "num_tokens": 32719421.0,
      "step": 16725
    },
    {
      "entropy": 6.130331420898438,
      "epoch": 1.6723646723646723,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004725658698072225,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16117009967565538,
      "num_tokens": 32729052.0,
      "step": 16730
    },
    {
      "entropy": 6.010404014587403,
      "epoch": 1.672864497425901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004725487889721648,
      "loss": 5.7428,
      "mean_token_accuracy": 0.17063481956720353,
      "num_tokens": 32738953.0,
      "step": 16735
    },
    {
      "entropy": 6.0865702629089355,
      "epoch": 1.6733643224871295,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004725317031668304,
      "loss": 5.8717,
      "mean_token_accuracy": 0.15795593708753586,
      "num_tokens": 32748728.0,
      "step": 16740
    },
    {
      "entropy": 6.113907527923584,
      "epoch": 1.673864147548358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047251461239164916,
      "loss": 5.8474,
      "mean_token_accuracy": 0.15611372143030167,
      "num_tokens": 32758305.0,
      "step": 16745
    },
    {
      "entropy": 6.1228752613067625,
      "epoch": 1.6743639726095867,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004724975166470512,
      "loss": 5.8749,
      "mean_token_accuracy": 0.16306701600551604,
      "num_tokens": 32768398.0,
      "step": 16750
    },
    {
      "entropy": 6.118631601333618,
      "epoch": 1.6748637976708152,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004724804159334665,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16221136823296547,
      "num_tokens": 32778970.0,
      "step": 16755
    },
    {
      "entropy": 6.08395938873291,
      "epoch": 1.6753636227320436,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004724633102513253,
      "loss": 5.7636,
      "mean_token_accuracy": 0.1637127861380577,
      "num_tokens": 32789815.0,
      "step": 16760
    },
    {
      "entropy": 6.069661331176758,
      "epoch": 1.6758634477932723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004724461996010581,
      "loss": 5.8638,
      "mean_token_accuracy": 0.16111362129449844,
      "num_tokens": 32799378.0,
      "step": 16765
    },
    {
      "entropy": 6.027296018600464,
      "epoch": 1.676363272854501,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004724290839830953,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1740318074822426,
      "num_tokens": 32809380.0,
      "step": 16770
    },
    {
      "entropy": 6.063532114028931,
      "epoch": 1.6768630979157295,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004724119633978675,
      "loss": 5.7591,
      "mean_token_accuracy": 0.17031576484441757,
      "num_tokens": 32818645.0,
      "step": 16775
    },
    {
      "entropy": 6.087962245941162,
      "epoch": 1.677362922976958,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047239483784580554,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16549896001815795,
      "num_tokens": 32827832.0,
      "step": 16780
    },
    {
      "entropy": 6.151183652877807,
      "epoch": 1.6778627480381867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004723777073273403,
      "loss": 5.9106,
      "mean_token_accuracy": 0.1544394999742508,
      "num_tokens": 32839312.0,
      "step": 16785
    },
    {
      "entropy": 6.091826581954956,
      "epoch": 1.6783625730994152,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004723605718429026,
      "loss": 5.7775,
      "mean_token_accuracy": 0.17149653136730195,
      "num_tokens": 32849065.0,
      "step": 16790
    },
    {
      "entropy": 6.034113073348999,
      "epoch": 1.6788623981606436,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004723434313929238,
      "loss": 5.7925,
      "mean_token_accuracy": 0.1567334435880184,
      "num_tokens": 32858469.0,
      "step": 16795
    },
    {
      "entropy": 6.128162479400634,
      "epoch": 1.6793622232218723,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047232628597783507,
      "loss": 5.9068,
      "mean_token_accuracy": 0.15622759312391282,
      "num_tokens": 32868700.0,
      "step": 16800
    },
    {
      "entropy": 6.125461769104004,
      "epoch": 1.679862048283101,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047230913559806773,
      "loss": 5.7975,
      "mean_token_accuracy": 0.16761286109685897,
      "num_tokens": 32878403.0,
      "step": 16805
    },
    {
      "entropy": 6.151577377319336,
      "epoch": 1.6803618733443295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004722919802540533,
      "loss": 5.8675,
      "mean_token_accuracy": 0.15826747119426726,
      "num_tokens": 32888322.0,
      "step": 16810
    },
    {
      "entropy": 6.090513849258423,
      "epoch": 1.680861698405558,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004722748199462234,
      "loss": 5.8364,
      "mean_token_accuracy": 0.16106705367565155,
      "num_tokens": 32897170.0,
      "step": 16815
    },
    {
      "entropy": 5.990663671493531,
      "epoch": 1.6813615234667867,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004722576546750098,
      "loss": 5.6826,
      "mean_token_accuracy": 0.16785183548927307,
      "num_tokens": 32906503.0,
      "step": 16820
    },
    {
      "entropy": 6.102352523803711,
      "epoch": 1.6818613485280152,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047224048444084434,
      "loss": 5.8023,
      "mean_token_accuracy": 0.16765003353357316,
      "num_tokens": 32915723.0,
      "step": 16825
    },
    {
      "entropy": 6.1057356834411625,
      "epoch": 1.6823611735892436,
      "grad_norm": 1.25,
      "learning_rate": 0.00047222330924415906,
      "loss": 5.8889,
      "mean_token_accuracy": 0.17007970809936523,
      "num_tokens": 32925589.0,
      "step": 16830
    },
    {
      "entropy": 6.129248762130738,
      "epoch": 1.6828609986504723,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004722061290853862,
      "loss": 5.8888,
      "mean_token_accuracy": 0.16438579708337783,
      "num_tokens": 32935727.0,
      "step": 16835
    },
    {
      "entropy": 6.018674612045288,
      "epoch": 1.683360823711701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047218894396495765,
      "loss": 5.7915,
      "mean_token_accuracy": 0.1671060025691986,
      "num_tokens": 32944257.0,
      "step": 16840
    },
    {
      "entropy": 6.0796959400177,
      "epoch": 1.6838606487729295,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004721717538833061,
      "loss": 5.8638,
      "mean_token_accuracy": 0.15776713639497758,
      "num_tokens": 32954341.0,
      "step": 16845
    },
    {
      "entropy": 6.095499753952026,
      "epoch": 1.684360473834158,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000472154558840864,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16357080340385438,
      "num_tokens": 32963997.0,
      "step": 16850
    },
    {
      "entropy": 6.007463645935059,
      "epoch": 1.6848602988953867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000472137358838064,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16467310339212418,
      "num_tokens": 32974301.0,
      "step": 16855
    },
    {
      "entropy": 6.185200214385986,
      "epoch": 1.6853601239566152,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004721201538753387,
      "loss": 5.8573,
      "mean_token_accuracy": 0.15662312209606172,
      "num_tokens": 32983535.0,
      "step": 16860
    },
    {
      "entropy": 6.125569152832031,
      "epoch": 1.6858599490178436,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047210294395312103,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16595944836735727,
      "num_tokens": 32993392.0,
      "step": 16865
    },
    {
      "entropy": 5.939804553985596,
      "epoch": 1.6863597740790723,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000472085729071844,
      "loss": 5.589,
      "mean_token_accuracy": 0.186769337952137,
      "num_tokens": 33002103.0,
      "step": 16870
    },
    {
      "entropy": 6.066113567352295,
      "epoch": 1.686859599140301,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047206850923194076,
      "loss": 5.8601,
      "mean_token_accuracy": 0.16124456971883774,
      "num_tokens": 33011557.0,
      "step": 16875
    },
    {
      "entropy": 6.162896013259887,
      "epoch": 1.6873594242015295,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047205128443384456,
      "loss": 5.857,
      "mean_token_accuracy": 0.1638089582324028,
      "num_tokens": 33021015.0,
      "step": 16880
    },
    {
      "entropy": 6.1377428531646725,
      "epoch": 1.687859249262758,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004720340546779887,
      "loss": 5.8147,
      "mean_token_accuracy": 0.16303296983242035,
      "num_tokens": 33029719.0,
      "step": 16885
    },
    {
      "entropy": 6.119587421417236,
      "epoch": 1.6883590743239867,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004720168199648068,
      "loss": 5.872,
      "mean_token_accuracy": 0.16215866059064865,
      "num_tokens": 33039424.0,
      "step": 16890
    },
    {
      "entropy": 6.07195725440979,
      "epoch": 1.6888588993852152,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047199958029473237,
      "loss": 5.8517,
      "mean_token_accuracy": 0.16355995535850526,
      "num_tokens": 33050175.0,
      "step": 16895
    },
    {
      "entropy": 6.116205215454102,
      "epoch": 1.6893587244464436,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047198233566819915,
      "loss": 5.8719,
      "mean_token_accuracy": 0.1620234914124012,
      "num_tokens": 33061352.0,
      "step": 16900
    },
    {
      "entropy": 6.1510108470916744,
      "epoch": 1.6898585495076723,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004719650860856411,
      "loss": 5.8339,
      "mean_token_accuracy": 0.16097532510757445,
      "num_tokens": 33069946.0,
      "step": 16905
    },
    {
      "entropy": 6.077900743484497,
      "epoch": 1.690358374568901,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047194783154749217,
      "loss": 5.8227,
      "mean_token_accuracy": 0.16575631499290466,
      "num_tokens": 33078478.0,
      "step": 16910
    },
    {
      "entropy": 6.072055196762085,
      "epoch": 1.6908581996301293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004719305720541865,
      "loss": 5.781,
      "mean_token_accuracy": 0.1756015732884407,
      "num_tokens": 33088073.0,
      "step": 16915
    },
    {
      "entropy": 6.041840505599976,
      "epoch": 1.691358024691358,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047191330760615823,
      "loss": 5.769,
      "mean_token_accuracy": 0.16911605447530748,
      "num_tokens": 33097127.0,
      "step": 16920
    },
    {
      "entropy": 6.028849935531616,
      "epoch": 1.6918578497525867,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047189603820384186,
      "loss": 5.786,
      "mean_token_accuracy": 0.16654183715581894,
      "num_tokens": 33107250.0,
      "step": 16925
    },
    {
      "entropy": 6.093971252441406,
      "epoch": 1.6923576748138152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047187876384767187,
      "loss": 5.7976,
      "mean_token_accuracy": 0.16852006912231446,
      "num_tokens": 33117585.0,
      "step": 16930
    },
    {
      "entropy": 6.107379674911499,
      "epoch": 1.6928574998750436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047186148453808274,
      "loss": 5.8015,
      "mean_token_accuracy": 0.16152559518814086,
      "num_tokens": 33127077.0,
      "step": 16935
    },
    {
      "entropy": 6.050421810150146,
      "epoch": 1.6933573249362723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004718442002755094,
      "loss": 5.8546,
      "mean_token_accuracy": 0.1570062518119812,
      "num_tokens": 33136327.0,
      "step": 16940
    },
    {
      "entropy": 6.032283735275269,
      "epoch": 1.693857149997501,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004718269110603866,
      "loss": 5.7308,
      "mean_token_accuracy": 0.1669022709131241,
      "num_tokens": 33145528.0,
      "step": 16945
    },
    {
      "entropy": 6.035370111465454,
      "epoch": 1.6943569750587293,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047180961689314937,
      "loss": 5.8042,
      "mean_token_accuracy": 0.17139956206083298,
      "num_tokens": 33154651.0,
      "step": 16950
    },
    {
      "entropy": 6.045061922073364,
      "epoch": 1.694856800119958,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004717923177742329,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16861154288053512,
      "num_tokens": 33164602.0,
      "step": 16955
    },
    {
      "entropy": 6.146135997772217,
      "epoch": 1.6953566251811867,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047177501370407223,
      "loss": 5.9329,
      "mean_token_accuracy": 0.15784537941217422,
      "num_tokens": 33173534.0,
      "step": 16960
    },
    {
      "entropy": 6.090135812759399,
      "epoch": 1.6958564502424152,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00047175770468310294,
      "loss": 5.8781,
      "mean_token_accuracy": 0.1633387476205826,
      "num_tokens": 33183528.0,
      "step": 16965
    },
    {
      "entropy": 6.171725702285767,
      "epoch": 1.6963562753036436,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047174039071176037,
      "loss": 5.9515,
      "mean_token_accuracy": 0.14600507616996766,
      "num_tokens": 33193762.0,
      "step": 16970
    },
    {
      "entropy": 6.125813150405884,
      "epoch": 1.6968561003648723,
      "grad_norm": 1.640625,
      "learning_rate": 0.00047172307179048025,
      "loss": 5.8022,
      "mean_token_accuracy": 0.16525878757238388,
      "num_tokens": 33203050.0,
      "step": 16975
    },
    {
      "entropy": 6.1911460876464846,
      "epoch": 1.697355925426101,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004717057479196982,
      "loss": 5.9363,
      "mean_token_accuracy": 0.16578419208526612,
      "num_tokens": 33212975.0,
      "step": 16980
    },
    {
      "entropy": 6.131929540634156,
      "epoch": 1.6978557504873293,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004716884190998502,
      "loss": 5.8535,
      "mean_token_accuracy": 0.1625105082988739,
      "num_tokens": 33222967.0,
      "step": 16985
    },
    {
      "entropy": 6.088766098022461,
      "epoch": 1.698355575548558,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004716710853313721,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15834417641162873,
      "num_tokens": 33233000.0,
      "step": 16990
    },
    {
      "entropy": 6.174155282974243,
      "epoch": 1.6988554006097867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047165374661470023,
      "loss": 5.9372,
      "mean_token_accuracy": 0.1594241589307785,
      "num_tokens": 33242429.0,
      "step": 16995
    },
    {
      "entropy": 6.10385684967041,
      "epoch": 1.6993552256710152,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047163640295027064,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16058932691812516,
      "num_tokens": 33252474.0,
      "step": 17000
    },
    {
      "entropy": 6.0725727558135985,
      "epoch": 1.6998550507322436,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047161905433851974,
      "loss": 5.8964,
      "mean_token_accuracy": 0.15775106698274613,
      "num_tokens": 33261296.0,
      "step": 17005
    },
    {
      "entropy": 6.032800388336182,
      "epoch": 1.7003548757934723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047160170077988403,
      "loss": 5.7421,
      "mean_token_accuracy": 0.16881646513938903,
      "num_tokens": 33271382.0,
      "step": 17010
    },
    {
      "entropy": 6.1386474609375,
      "epoch": 1.7008547008547008,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004715843422748001,
      "loss": 5.9005,
      "mean_token_accuracy": 0.15558534786105155,
      "num_tokens": 33281512.0,
      "step": 17015
    },
    {
      "entropy": 6.1379866123199465,
      "epoch": 1.7013545259159293,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047156697882370474,
      "loss": 5.8534,
      "mean_token_accuracy": 0.16162669211626052,
      "num_tokens": 33290974.0,
      "step": 17020
    },
    {
      "entropy": 6.132971668243409,
      "epoch": 1.701854350977158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004715496104270348,
      "loss": 5.8998,
      "mean_token_accuracy": 0.15249569863080978,
      "num_tokens": 33301576.0,
      "step": 17025
    },
    {
      "entropy": 6.0337732315063475,
      "epoch": 1.7023541760383867,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004715322370852272,
      "loss": 5.7112,
      "mean_token_accuracy": 0.17594804167747496,
      "num_tokens": 33311277.0,
      "step": 17030
    },
    {
      "entropy": 6.004238080978394,
      "epoch": 1.7028540010996152,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047151485879871907,
      "loss": 5.789,
      "mean_token_accuracy": 0.17186643332242965,
      "num_tokens": 33322018.0,
      "step": 17035
    },
    {
      "entropy": 6.117154979705811,
      "epoch": 1.7033538261608436,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047149747556794763,
      "loss": 5.8742,
      "mean_token_accuracy": 0.16148997247219085,
      "num_tokens": 33331804.0,
      "step": 17040
    },
    {
      "entropy": 6.170132684707641,
      "epoch": 1.7038536512220723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047148008739335035,
      "loss": 5.8661,
      "mean_token_accuracy": 0.15481021404266357,
      "num_tokens": 33342014.0,
      "step": 17045
    },
    {
      "entropy": 6.100893926620484,
      "epoch": 1.7043534762833008,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004714626942753646,
      "loss": 5.7921,
      "mean_token_accuracy": 0.16379691809415817,
      "num_tokens": 33352086.0,
      "step": 17050
    },
    {
      "entropy": 6.107739782333374,
      "epoch": 1.7048533013445293,
      "grad_norm": 1.4375,
      "learning_rate": 0.000471445296214428,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1691975176334381,
      "num_tokens": 33362067.0,
      "step": 17055
    },
    {
      "entropy": 6.054507446289063,
      "epoch": 1.705353126405758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047142789321097834,
      "loss": 5.8002,
      "mean_token_accuracy": 0.16584909707307816,
      "num_tokens": 33371803.0,
      "step": 17060
    },
    {
      "entropy": 6.10920557975769,
      "epoch": 1.7058529514669867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004714104852654534,
      "loss": 5.8415,
      "mean_token_accuracy": 0.16659219861030578,
      "num_tokens": 33381258.0,
      "step": 17065
    },
    {
      "entropy": 6.115944337844849,
      "epoch": 1.7063527765282152,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004713930723782912,
      "loss": 5.8364,
      "mean_token_accuracy": 0.16401960998773574,
      "num_tokens": 33390709.0,
      "step": 17070
    },
    {
      "entropy": 6.147069644927979,
      "epoch": 1.7068526015894436,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047137565454992983,
      "loss": 5.8593,
      "mean_token_accuracy": 0.16338293254375458,
      "num_tokens": 33400831.0,
      "step": 17075
    },
    {
      "entropy": 6.119107341766357,
      "epoch": 1.7073524266506723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047135823178080745,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16362357884645462,
      "num_tokens": 33410974.0,
      "step": 17080
    },
    {
      "entropy": 6.167402124404907,
      "epoch": 1.7078522517119008,
      "grad_norm": 1.25,
      "learning_rate": 0.0004713408040713626,
      "loss": 5.9481,
      "mean_token_accuracy": 0.1555358126759529,
      "num_tokens": 33420555.0,
      "step": 17085
    },
    {
      "entropy": 6.073935317993164,
      "epoch": 1.7083520767731293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047132337142203365,
      "loss": 5.94,
      "mean_token_accuracy": 0.15684898048639298,
      "num_tokens": 33431064.0,
      "step": 17090
    },
    {
      "entropy": 6.097357130050659,
      "epoch": 1.708851901834358,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004713059338332591,
      "loss": 5.8056,
      "mean_token_accuracy": 0.16830147206783294,
      "num_tokens": 33441223.0,
      "step": 17095
    },
    {
      "entropy": 6.161975002288818,
      "epoch": 1.7093517268955867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047128849130547774,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16569649428129196,
      "num_tokens": 33451520.0,
      "step": 17100
    },
    {
      "entropy": 6.1017725467681885,
      "epoch": 1.7098515519568152,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047127104383912856,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1687925636768341,
      "num_tokens": 33461310.0,
      "step": 17105
    },
    {
      "entropy": 6.141603660583496,
      "epoch": 1.7103513770180436,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004712535914346504,
      "loss": 5.8723,
      "mean_token_accuracy": 0.1637128010392189,
      "num_tokens": 33471250.0,
      "step": 17110
    },
    {
      "entropy": 6.060923910140991,
      "epoch": 1.7108512020792723,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047123613409248233,
      "loss": 5.7482,
      "mean_token_accuracy": 0.16454707235097885,
      "num_tokens": 33480950.0,
      "step": 17115
    },
    {
      "entropy": 6.071712398529053,
      "epoch": 1.7113510271405008,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047121867181306365,
      "loss": 5.7138,
      "mean_token_accuracy": 0.16773526072502137,
      "num_tokens": 33490394.0,
      "step": 17120
    },
    {
      "entropy": 6.06605281829834,
      "epoch": 1.7118508522017293,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004712012045968337,
      "loss": 5.832,
      "mean_token_accuracy": 0.15956658720970154,
      "num_tokens": 33500130.0,
      "step": 17125
    },
    {
      "entropy": 6.094152641296387,
      "epoch": 1.712350677262958,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00047118373244423197,
      "loss": 5.9053,
      "mean_token_accuracy": 0.1656326338648796,
      "num_tokens": 33510698.0,
      "step": 17130
    },
    {
      "entropy": 6.115771055221558,
      "epoch": 1.7128505023241867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047116625535569787,
      "loss": 5.8093,
      "mean_token_accuracy": 0.16489032059907913,
      "num_tokens": 33520161.0,
      "step": 17135
    },
    {
      "entropy": 6.085592031478882,
      "epoch": 1.7133503273854152,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047114877333167133,
      "loss": 5.8498,
      "mean_token_accuracy": 0.1692027822136879,
      "num_tokens": 33530190.0,
      "step": 17140
    },
    {
      "entropy": 6.096170663833618,
      "epoch": 1.7138501524466436,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047113128637259216,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16599542796611785,
      "num_tokens": 33539376.0,
      "step": 17145
    },
    {
      "entropy": 6.120475196838379,
      "epoch": 1.7143499775078723,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004711137944789003,
      "loss": 5.8214,
      "mean_token_accuracy": 0.16473720818758011,
      "num_tokens": 33549290.0,
      "step": 17150
    },
    {
      "entropy": 6.1167360782623295,
      "epoch": 1.7148498025691008,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047109629765103577,
      "loss": 5.9025,
      "mean_token_accuracy": 0.15792422890663146,
      "num_tokens": 33559483.0,
      "step": 17155
    },
    {
      "entropy": 6.106214475631714,
      "epoch": 1.7153496276303293,
      "grad_norm": 1.25,
      "learning_rate": 0.00047107879588943887,
      "loss": 5.8825,
      "mean_token_accuracy": 0.15954498648643495,
      "num_tokens": 33569464.0,
      "step": 17160
    },
    {
      "entropy": 6.194419002532959,
      "epoch": 1.715849452691558,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047106128919455,
      "loss": 5.9595,
      "mean_token_accuracy": 0.15557035207748413,
      "num_tokens": 33579759.0,
      "step": 17165
    },
    {
      "entropy": 6.154019451141357,
      "epoch": 1.7163492777527867,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004710437775668094,
      "loss": 5.8012,
      "mean_token_accuracy": 0.16216324269771576,
      "num_tokens": 33590054.0,
      "step": 17170
    },
    {
      "entropy": 6.06962251663208,
      "epoch": 1.7168491028140151,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004710262610066579,
      "loss": 5.8109,
      "mean_token_accuracy": 0.16861556321382523,
      "num_tokens": 33600014.0,
      "step": 17175
    },
    {
      "entropy": 6.1378460884094235,
      "epoch": 1.7173489278752436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047100873951453617,
      "loss": 5.8124,
      "mean_token_accuracy": 0.16211322098970413,
      "num_tokens": 33609579.0,
      "step": 17180
    },
    {
      "entropy": 6.11484546661377,
      "epoch": 1.7178487529364723,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047099121309088487,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16280268579721452,
      "num_tokens": 33618082.0,
      "step": 17185
    },
    {
      "entropy": 6.0571905136108395,
      "epoch": 1.7183485779977008,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047097368173614515,
      "loss": 5.857,
      "mean_token_accuracy": 0.15929074585437775,
      "num_tokens": 33627835.0,
      "step": 17190
    },
    {
      "entropy": 6.10786805152893,
      "epoch": 1.7188484030589293,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047095614545075807,
      "loss": 5.8592,
      "mean_token_accuracy": 0.15759921967983245,
      "num_tokens": 33638187.0,
      "step": 17195
    },
    {
      "entropy": 6.161505079269409,
      "epoch": 1.719348228120158,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004709386042351647,
      "loss": 5.8288,
      "mean_token_accuracy": 0.16300379037857055,
      "num_tokens": 33647805.0,
      "step": 17200
    },
    {
      "entropy": 6.134348011016845,
      "epoch": 1.7198480531813867,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004709210580898065,
      "loss": 5.835,
      "mean_token_accuracy": 0.16131857484579087,
      "num_tokens": 33656657.0,
      "step": 17205
    },
    {
      "entropy": 6.073101472854614,
      "epoch": 1.720347878242615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004709035070151249,
      "loss": 5.79,
      "mean_token_accuracy": 0.16222733110189438,
      "num_tokens": 33666025.0,
      "step": 17210
    },
    {
      "entropy": 6.070948934555053,
      "epoch": 1.7208477033038436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004708859510115614,
      "loss": 5.7649,
      "mean_token_accuracy": 0.16763748973608017,
      "num_tokens": 33676378.0,
      "step": 17215
    },
    {
      "entropy": 6.167617988586426,
      "epoch": 1.7213475283650723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004708683900795579,
      "loss": 5.9439,
      "mean_token_accuracy": 0.1524502530694008,
      "num_tokens": 33685899.0,
      "step": 17220
    },
    {
      "entropy": 6.0968364715576175,
      "epoch": 1.7218473534263008,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047085082421955603,
      "loss": 5.8314,
      "mean_token_accuracy": 0.16250892281532286,
      "num_tokens": 33696155.0,
      "step": 17225
    },
    {
      "entropy": 6.172334098815918,
      "epoch": 1.7223471784875293,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047083325343199786,
      "loss": 5.8317,
      "mean_token_accuracy": 0.16396375894546508,
      "num_tokens": 33704685.0,
      "step": 17230
    },
    {
      "entropy": 6.248908185958863,
      "epoch": 1.722847003548758,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004708156777173254,
      "loss": 5.9554,
      "mean_token_accuracy": 0.15723454356193542,
      "num_tokens": 33715189.0,
      "step": 17235
    },
    {
      "entropy": 6.105761384963989,
      "epoch": 1.7233468286099867,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004707980970759809,
      "loss": 5.8528,
      "mean_token_accuracy": 0.15928208827972412,
      "num_tokens": 33724619.0,
      "step": 17240
    },
    {
      "entropy": 6.070801544189453,
      "epoch": 1.723846653671215,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047078051150840666,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16701079308986663,
      "num_tokens": 33734371.0,
      "step": 17245
    },
    {
      "entropy": 6.0060371398925785,
      "epoch": 1.7243464787324436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004707629210150451,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1682990550994873,
      "num_tokens": 33743592.0,
      "step": 17250
    },
    {
      "entropy": 6.160249042510986,
      "epoch": 1.7248463037936723,
      "grad_norm": 1.7578125,
      "learning_rate": 0.0004707453255963388,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16355862021446227,
      "num_tokens": 33753686.0,
      "step": 17255
    },
    {
      "entropy": 6.167244005203247,
      "epoch": 1.7253461288549008,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047072772525273053,
      "loss": 5.8909,
      "mean_token_accuracy": 0.1625453144311905,
      "num_tokens": 33763076.0,
      "step": 17260
    },
    {
      "entropy": 6.024517822265625,
      "epoch": 1.7258459539161293,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047071011998466304,
      "loss": 5.6852,
      "mean_token_accuracy": 0.1721458300948143,
      "num_tokens": 33772283.0,
      "step": 17265
    },
    {
      "entropy": 6.072497129440308,
      "epoch": 1.726345778977358,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047069250979257934,
      "loss": 5.7746,
      "mean_token_accuracy": 0.1665676161646843,
      "num_tokens": 33782208.0,
      "step": 17270
    },
    {
      "entropy": 6.066922855377197,
      "epoch": 1.7268456040385864,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047067489467692236,
      "loss": 5.7689,
      "mean_token_accuracy": 0.17144491374492646,
      "num_tokens": 33792048.0,
      "step": 17275
    },
    {
      "entropy": 6.06114559173584,
      "epoch": 1.727345429099815,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004706572746381354,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16889194995164872,
      "num_tokens": 33800903.0,
      "step": 17280
    },
    {
      "entropy": 6.128403520584106,
      "epoch": 1.7278452541610436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047063964967666177,
      "loss": 5.8674,
      "mean_token_accuracy": 0.16363469362258912,
      "num_tokens": 33810575.0,
      "step": 17285
    },
    {
      "entropy": 6.0639153003692625,
      "epoch": 1.7283450792222723,
      "grad_norm": 1.25,
      "learning_rate": 0.0004706220197929449,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16402264237403869,
      "num_tokens": 33819365.0,
      "step": 17290
    },
    {
      "entropy": 6.136084508895874,
      "epoch": 1.7288449042835008,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004706043849874284,
      "loss": 5.9021,
      "mean_token_accuracy": 0.15926491022109984,
      "num_tokens": 33828941.0,
      "step": 17295
    },
    {
      "entropy": 6.087293291091919,
      "epoch": 1.7293447293447293,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047058674526055585,
      "loss": 5.8102,
      "mean_token_accuracy": 0.16338820457458497,
      "num_tokens": 33838700.0,
      "step": 17300
    },
    {
      "entropy": 6.1835216045379635,
      "epoch": 1.729844554405958,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00047056910061277115,
      "loss": 5.8457,
      "mean_token_accuracy": 0.15880443155765533,
      "num_tokens": 33848592.0,
      "step": 17305
    },
    {
      "entropy": 6.100729703903198,
      "epoch": 1.7303443794671864,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004705514510445182,
      "loss": 5.833,
      "mean_token_accuracy": 0.1671199917793274,
      "num_tokens": 33858037.0,
      "step": 17310
    },
    {
      "entropy": 6.0883807182312015,
      "epoch": 1.730844204528415,
      "grad_norm": 1.484375,
      "learning_rate": 0.000470533796556241,
      "loss": 5.8722,
      "mean_token_accuracy": 0.16382436752319335,
      "num_tokens": 33867284.0,
      "step": 17315
    },
    {
      "entropy": 6.118350410461426,
      "epoch": 1.7313440295896436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004705161371483839,
      "loss": 5.7709,
      "mean_token_accuracy": 0.1691048964858055,
      "num_tokens": 33876006.0,
      "step": 17320
    },
    {
      "entropy": 6.071259069442749,
      "epoch": 1.7318438546508723,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047049847282139095,
      "loss": 5.6669,
      "mean_token_accuracy": 0.17449919432401656,
      "num_tokens": 33885273.0,
      "step": 17325
    },
    {
      "entropy": 6.1583904266357425,
      "epoch": 1.7323436797121008,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047048080357570685,
      "loss": 5.8946,
      "mean_token_accuracy": 0.156680728495121,
      "num_tokens": 33894332.0,
      "step": 17330
    },
    {
      "entropy": 6.071640205383301,
      "epoch": 1.7328435047733293,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000470463129411776,
      "loss": 5.7907,
      "mean_token_accuracy": 0.16725963801145555,
      "num_tokens": 33903556.0,
      "step": 17335
    },
    {
      "entropy": 6.193722438812256,
      "epoch": 1.733343329834558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004704454503300431,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15192199051380156,
      "num_tokens": 33913129.0,
      "step": 17340
    },
    {
      "entropy": 6.126398086547852,
      "epoch": 1.7338431548957864,
      "grad_norm": 1.4375,
      "learning_rate": 0.000470427766330953,
      "loss": 5.8503,
      "mean_token_accuracy": 0.16198702156543732,
      "num_tokens": 33923016.0,
      "step": 17345
    },
    {
      "entropy": 6.13785228729248,
      "epoch": 1.734342979957015,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004704100774149506,
      "loss": 5.794,
      "mean_token_accuracy": 0.16730466037988662,
      "num_tokens": 33932336.0,
      "step": 17350
    },
    {
      "entropy": 6.1876918315887455,
      "epoch": 1.7348428050182436,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047039238358248086,
      "loss": 5.8278,
      "mean_token_accuracy": 0.16481807827949524,
      "num_tokens": 33941488.0,
      "step": 17355
    },
    {
      "entropy": 6.070731782913208,
      "epoch": 1.7353426300794723,
      "grad_norm": 1.375,
      "learning_rate": 0.000470374684833989,
      "loss": 5.6839,
      "mean_token_accuracy": 0.16993282437324525,
      "num_tokens": 33950919.0,
      "step": 17360
    },
    {
      "entropy": 5.991929912567139,
      "epoch": 1.7358424551407008,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004703569811699205,
      "loss": 5.7475,
      "mean_token_accuracy": 0.1737784266471863,
      "num_tokens": 33960609.0,
      "step": 17365
    },
    {
      "entropy": 6.1530698299407955,
      "epoch": 1.7363422802019293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047033927259072046,
      "loss": 5.8824,
      "mean_token_accuracy": 0.1581564076244831,
      "num_tokens": 33970454.0,
      "step": 17370
    },
    {
      "entropy": 6.051774644851685,
      "epoch": 1.736842105263158,
      "grad_norm": 1.703125,
      "learning_rate": 0.00047032155909683464,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17300886660814285,
      "num_tokens": 33979029.0,
      "step": 17375
    },
    {
      "entropy": 6.161842203140258,
      "epoch": 1.7373419303243864,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004703038406887087,
      "loss": 5.9845,
      "mean_token_accuracy": 0.15261003673076629,
      "num_tokens": 33990525.0,
      "step": 17380
    },
    {
      "entropy": 6.127135992050171,
      "epoch": 1.737841755385615,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047028611736678837,
      "loss": 5.7789,
      "mean_token_accuracy": 0.17074088007211685,
      "num_tokens": 34000022.0,
      "step": 17385
    },
    {
      "entropy": 6.106827116012573,
      "epoch": 1.7383415804468436,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004702683891315196,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16239847242832184,
      "num_tokens": 34010082.0,
      "step": 17390
    },
    {
      "entropy": 6.153608179092407,
      "epoch": 1.7388414055080723,
      "grad_norm": 1.375,
      "learning_rate": 0.00047025065598334834,
      "loss": 5.8322,
      "mean_token_accuracy": 0.15658228769898414,
      "num_tokens": 34020245.0,
      "step": 17395
    },
    {
      "entropy": 6.088201665878296,
      "epoch": 1.7393412305693008,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004702329179227208,
      "loss": 5.8191,
      "mean_token_accuracy": 0.15659847855567932,
      "num_tokens": 34029359.0,
      "step": 17400
    },
    {
      "entropy": 6.1941039085388185,
      "epoch": 1.7398410556305293,
      "grad_norm": 1.375,
      "learning_rate": 0.00047021517495008335,
      "loss": 5.8774,
      "mean_token_accuracy": 0.15642124116420747,
      "num_tokens": 34039778.0,
      "step": 17405
    },
    {
      "entropy": 6.192787313461304,
      "epoch": 1.740340880691758,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047019742706588233,
      "loss": 5.9057,
      "mean_token_accuracy": 0.16007311195135115,
      "num_tokens": 34050102.0,
      "step": 17410
    },
    {
      "entropy": 6.07509126663208,
      "epoch": 1.7408407057529864,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004701796742705642,
      "loss": 5.7607,
      "mean_token_accuracy": 0.17326110005378723,
      "num_tokens": 34059292.0,
      "step": 17415
    },
    {
      "entropy": 6.101332378387451,
      "epoch": 1.741340530814215,
      "grad_norm": 1.0625,
      "learning_rate": 0.00047016191656457573,
      "loss": 5.8828,
      "mean_token_accuracy": 0.158489128947258,
      "num_tokens": 34068547.0,
      "step": 17420
    },
    {
      "entropy": 6.0987097263336185,
      "epoch": 1.7418403558754436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047014415394836374,
      "loss": 5.7538,
      "mean_token_accuracy": 0.17125205099582672,
      "num_tokens": 34078142.0,
      "step": 17425
    },
    {
      "entropy": 6.103067255020141,
      "epoch": 1.7423401809366723,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004701263864223749,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1573589637875557,
      "num_tokens": 34087780.0,
      "step": 17430
    },
    {
      "entropy": 6.158544254302979,
      "epoch": 1.7428400059979008,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004701086139870565,
      "loss": 5.8815,
      "mean_token_accuracy": 0.15706681460142136,
      "num_tokens": 34097988.0,
      "step": 17435
    },
    {
      "entropy": 6.1324492454528805,
      "epoch": 1.7433398310591293,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004700908366428556,
      "loss": 5.8557,
      "mean_token_accuracy": 0.15688350647687913,
      "num_tokens": 34107310.0,
      "step": 17440
    },
    {
      "entropy": 6.1935327529907225,
      "epoch": 1.743839656120358,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004700730543902194,
      "loss": 5.95,
      "mean_token_accuracy": 0.1552974671125412,
      "num_tokens": 34117063.0,
      "step": 17445
    },
    {
      "entropy": 6.053671503067017,
      "epoch": 1.7443394811815864,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047005526722959537,
      "loss": 5.7854,
      "mean_token_accuracy": 0.16529690474271774,
      "num_tokens": 34125391.0,
      "step": 17450
    },
    {
      "entropy": 6.036420488357544,
      "epoch": 1.744839306242815,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047003747516143095,
      "loss": 5.883,
      "mean_token_accuracy": 0.16082841008901597,
      "num_tokens": 34135816.0,
      "step": 17455
    },
    {
      "entropy": 6.145236492156982,
      "epoch": 1.7453391313040436,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047001967818617385,
      "loss": 5.9235,
      "mean_token_accuracy": 0.1586654618382454,
      "num_tokens": 34146366.0,
      "step": 17460
    },
    {
      "entropy": 6.158540105819702,
      "epoch": 1.7458389563652723,
      "grad_norm": 1.9921875,
      "learning_rate": 0.0004700018763042719,
      "loss": 5.8446,
      "mean_token_accuracy": 0.16614151298999785,
      "num_tokens": 34155982.0,
      "step": 17465
    },
    {
      "entropy": 6.135180902481079,
      "epoch": 1.7463387814265006,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004699840695161728,
      "loss": 5.8727,
      "mean_token_accuracy": 0.15896083563566207,
      "num_tokens": 34164808.0,
      "step": 17470
    },
    {
      "entropy": 6.071656084060669,
      "epoch": 1.7468386064877293,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004699662578223248,
      "loss": 5.7939,
      "mean_token_accuracy": 0.16869038790464402,
      "num_tokens": 34174010.0,
      "step": 17475
    },
    {
      "entropy": 6.155232906341553,
      "epoch": 1.747338431548958,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046994844122317587,
      "loss": 5.7685,
      "mean_token_accuracy": 0.1609248623251915,
      "num_tokens": 34183116.0,
      "step": 17480
    },
    {
      "entropy": 6.082840061187744,
      "epoch": 1.7478382566101864,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046993061971917433,
      "loss": 5.8325,
      "mean_token_accuracy": 0.1709253028035164,
      "num_tokens": 34191976.0,
      "step": 17485
    },
    {
      "entropy": 6.111375713348389,
      "epoch": 1.748338081671415,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046991279331076865,
      "loss": 5.842,
      "mean_token_accuracy": 0.1612362489104271,
      "num_tokens": 34200211.0,
      "step": 17490
    },
    {
      "entropy": 6.115491724014282,
      "epoch": 1.7488379067326436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004698949619984071,
      "loss": 5.841,
      "mean_token_accuracy": 0.16141808331012725,
      "num_tokens": 34210392.0,
      "step": 17495
    },
    {
      "entropy": 6.096302795410156,
      "epoch": 1.7493377317938723,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004698771257825385,
      "loss": 5.8417,
      "mean_token_accuracy": 0.16122303605079652,
      "num_tokens": 34220139.0,
      "step": 17500
    },
    {
      "entropy": 6.126408672332763,
      "epoch": 1.7498375568551006,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004698592846636116,
      "loss": 5.8549,
      "mean_token_accuracy": 0.16112832054495813,
      "num_tokens": 34230413.0,
      "step": 17505
    },
    {
      "entropy": 6.031373453140259,
      "epoch": 1.7503373819163293,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004698414386420752,
      "loss": 5.8542,
      "mean_token_accuracy": 0.17045213282108307,
      "num_tokens": 34240312.0,
      "step": 17510
    },
    {
      "entropy": 6.086150312423706,
      "epoch": 1.750837206977558,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004698235877183784,
      "loss": 5.9006,
      "mean_token_accuracy": 0.15644796937704086,
      "num_tokens": 34250043.0,
      "step": 17515
    },
    {
      "entropy": 6.181535816192627,
      "epoch": 1.7513370320387864,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046980573189297017,
      "loss": 5.8397,
      "mean_token_accuracy": 0.16683902442455292,
      "num_tokens": 34259424.0,
      "step": 17520
    },
    {
      "entropy": 6.11281189918518,
      "epoch": 1.751836857100015,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004697878711662999,
      "loss": 5.7876,
      "mean_token_accuracy": 0.16655325293540954,
      "num_tokens": 34268133.0,
      "step": 17525
    },
    {
      "entropy": 6.06966757774353,
      "epoch": 1.7523366821612436,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046977000553881696,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16929287910461427,
      "num_tokens": 34277090.0,
      "step": 17530
    },
    {
      "entropy": 6.126381158828735,
      "epoch": 1.752836507222472,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046975213501097076,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16520670652389527,
      "num_tokens": 34285684.0,
      "step": 17535
    },
    {
      "entropy": 6.074184799194336,
      "epoch": 1.7533363322837006,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004697342595832109,
      "loss": 5.6921,
      "mean_token_accuracy": 0.17246692329645158,
      "num_tokens": 34294483.0,
      "step": 17540
    },
    {
      "entropy": 6.059299325942993,
      "epoch": 1.7538361573449293,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046971637925598725,
      "loss": 5.8137,
      "mean_token_accuracy": 0.16387852877378464,
      "num_tokens": 34303027.0,
      "step": 17545
    },
    {
      "entropy": 6.102261686325074,
      "epoch": 1.754335982406158,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004696984940297496,
      "loss": 5.8597,
      "mean_token_accuracy": 0.15393581688404084,
      "num_tokens": 34313034.0,
      "step": 17550
    },
    {
      "entropy": 6.084933423995972,
      "epoch": 1.7548358074673864,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000469680603904948,
      "loss": 5.8075,
      "mean_token_accuracy": 0.16775193214416503,
      "num_tokens": 34323345.0,
      "step": 17555
    },
    {
      "entropy": 6.094639825820923,
      "epoch": 1.755335632528615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046966270888203237,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16052644550800324,
      "num_tokens": 34332273.0,
      "step": 17560
    },
    {
      "entropy": 6.221106624603271,
      "epoch": 1.7558354575898436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004696448089614532,
      "loss": 5.9131,
      "mean_token_accuracy": 0.15925005376338958,
      "num_tokens": 34342071.0,
      "step": 17565
    },
    {
      "entropy": 6.135105037689209,
      "epoch": 1.756335282651072,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004696269041436606,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17191740423440932,
      "num_tokens": 34352204.0,
      "step": 17570
    },
    {
      "entropy": 6.030926322937011,
      "epoch": 1.7568351077123006,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004696089944291052,
      "loss": 5.749,
      "mean_token_accuracy": 0.17063892036676406,
      "num_tokens": 34362033.0,
      "step": 17575
    },
    {
      "entropy": 6.142614269256592,
      "epoch": 1.7573349327735293,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004695910798182377,
      "loss": 5.845,
      "mean_token_accuracy": 0.16558307856321336,
      "num_tokens": 34371770.0,
      "step": 17580
    },
    {
      "entropy": 6.0844078540802,
      "epoch": 1.757834757834758,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004695731603115086,
      "loss": 5.8576,
      "mean_token_accuracy": 0.16450049877166747,
      "num_tokens": 34381755.0,
      "step": 17585
    },
    {
      "entropy": 6.067360687255859,
      "epoch": 1.7583345828959864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046955523590936884,
      "loss": 5.747,
      "mean_token_accuracy": 0.16674330532550813,
      "num_tokens": 34390909.0,
      "step": 17590
    },
    {
      "entropy": 6.156034994125366,
      "epoch": 1.758834407957215,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004695373066122694,
      "loss": 5.8442,
      "mean_token_accuracy": 0.16521249264478682,
      "num_tokens": 34401051.0,
      "step": 17595
    },
    {
      "entropy": 6.181607055664062,
      "epoch": 1.7593342330184436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004695193724206615,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16492556035518646,
      "num_tokens": 34410099.0,
      "step": 17600
    },
    {
      "entropy": 6.103878021240234,
      "epoch": 1.759834058079672,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046950143333499613,
      "loss": 5.797,
      "mean_token_accuracy": 0.16166936457157136,
      "num_tokens": 34418895.0,
      "step": 17605
    },
    {
      "entropy": 5.9660454273223875,
      "epoch": 1.7603338831409006,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046948348935572476,
      "loss": 5.8229,
      "mean_token_accuracy": 0.16356450021266938,
      "num_tokens": 34429369.0,
      "step": 17610
    },
    {
      "entropy": 6.101383876800537,
      "epoch": 1.7608337082021293,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004694655404832988,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16689707785844804,
      "num_tokens": 34439773.0,
      "step": 17615
    },
    {
      "entropy": 6.26047682762146,
      "epoch": 1.761333533263358,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046944758671817,
      "loss": 5.9071,
      "mean_token_accuracy": 0.1601081147789955,
      "num_tokens": 34450613.0,
      "step": 17620
    },
    {
      "entropy": 6.1565845012664795,
      "epoch": 1.7618333583245864,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004694296280607899,
      "loss": 5.9083,
      "mean_token_accuracy": 0.163635815680027,
      "num_tokens": 34460785.0,
      "step": 17625
    },
    {
      "entropy": 6.228675794601441,
      "epoch": 1.762333183385815,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004694116645116104,
      "loss": 5.9414,
      "mean_token_accuracy": 0.15669906586408616,
      "num_tokens": 34471152.0,
      "step": 17630
    },
    {
      "entropy": 6.110484790802002,
      "epoch": 1.7628330084470436,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004693936960710834,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17268788516521455,
      "num_tokens": 34481511.0,
      "step": 17635
    },
    {
      "entropy": 6.0648352146148685,
      "epoch": 1.763332833508272,
      "grad_norm": 1.609375,
      "learning_rate": 0.00046937572273966104,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16912080645561217,
      "num_tokens": 34490556.0,
      "step": 17640
    },
    {
      "entropy": 6.058170032501221,
      "epoch": 1.7638326585695006,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046935774451779556,
      "loss": 5.8187,
      "mean_token_accuracy": 0.16395978182554244,
      "num_tokens": 34500363.0,
      "step": 17645
    },
    {
      "entropy": 6.052880716323853,
      "epoch": 1.7643324836307293,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004693397614059392,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16746517568826674,
      "num_tokens": 34511009.0,
      "step": 17650
    },
    {
      "entropy": 6.19283971786499,
      "epoch": 1.764832308691958,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004693217734045445,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15600780621170998,
      "num_tokens": 34520806.0,
      "step": 17655
    },
    {
      "entropy": 6.161008405685425,
      "epoch": 1.7653321337531864,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046930378051406393,
      "loss": 5.8063,
      "mean_token_accuracy": 0.16458062082529068,
      "num_tokens": 34531266.0,
      "step": 17660
    },
    {
      "entropy": 6.2179412841796875,
      "epoch": 1.765831958814415,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046928578273495026,
      "loss": 5.9796,
      "mean_token_accuracy": 0.15189775973558425,
      "num_tokens": 34542503.0,
      "step": 17665
    },
    {
      "entropy": 6.129137468338013,
      "epoch": 1.7663317838756436,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004692677800676563,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1636650398373604,
      "num_tokens": 34552639.0,
      "step": 17670
    },
    {
      "entropy": 6.0427570819854735,
      "epoch": 1.766831608936872,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046924977251263495,
      "loss": 5.7673,
      "mean_token_accuracy": 0.17232124954462053,
      "num_tokens": 34562411.0,
      "step": 17675
    },
    {
      "entropy": 6.059590911865234,
      "epoch": 1.7673314339981006,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046923176007033937,
      "loss": 5.828,
      "mean_token_accuracy": 0.16443783044815063,
      "num_tokens": 34571532.0,
      "step": 17680
    },
    {
      "entropy": 6.130098104476929,
      "epoch": 1.7678312590593293,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046921374274122265,
      "loss": 5.7548,
      "mean_token_accuracy": 0.16341262310743332,
      "num_tokens": 34580338.0,
      "step": 17685
    },
    {
      "entropy": 6.02744312286377,
      "epoch": 1.768331084120558,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046919572052573814,
      "loss": 5.774,
      "mean_token_accuracy": 0.16904015392065047,
      "num_tokens": 34589353.0,
      "step": 17690
    },
    {
      "entropy": 6.129482173919678,
      "epoch": 1.7688309091817864,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046917769342433925,
      "loss": 5.958,
      "mean_token_accuracy": 0.15644664019346238,
      "num_tokens": 34599981.0,
      "step": 17695
    },
    {
      "entropy": 6.006915426254272,
      "epoch": 1.769330734243015,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004691596614374796,
      "loss": 5.6776,
      "mean_token_accuracy": 0.17406300455331802,
      "num_tokens": 34608927.0,
      "step": 17700
    },
    {
      "entropy": 6.072988557815552,
      "epoch": 1.7698305593042436,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004691416245656128,
      "loss": 5.8568,
      "mean_token_accuracy": 0.16101256310939788,
      "num_tokens": 34619455.0,
      "step": 17705
    },
    {
      "entropy": 6.075609588623047,
      "epoch": 1.770330384365472,
      "grad_norm": 1.125,
      "learning_rate": 0.00046912358280919267,
      "loss": 5.7236,
      "mean_token_accuracy": 0.1709471747279167,
      "num_tokens": 34628761.0,
      "step": 17710
    },
    {
      "entropy": 6.110510206222534,
      "epoch": 1.7708302094267006,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046910553616867314,
      "loss": 5.8636,
      "mean_token_accuracy": 0.15698741376399994,
      "num_tokens": 34638859.0,
      "step": 17715
    },
    {
      "entropy": 6.0164775371551515,
      "epoch": 1.7713300344879293,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004690874846445083,
      "loss": 5.7304,
      "mean_token_accuracy": 0.169376240670681,
      "num_tokens": 34648466.0,
      "step": 17720
    },
    {
      "entropy": 6.00524582862854,
      "epoch": 1.771829859549158,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004690694282371523,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16982523053884507,
      "num_tokens": 34658567.0,
      "step": 17725
    },
    {
      "entropy": 6.016337490081787,
      "epoch": 1.7723296846103862,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004690513669470594,
      "loss": 5.6772,
      "mean_token_accuracy": 0.1756102368235588,
      "num_tokens": 34667555.0,
      "step": 17730
    },
    {
      "entropy": 6.139855813980103,
      "epoch": 1.772829509671615,
      "grad_norm": 1.34375,
      "learning_rate": 0.000469033300774684,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16426636129617692,
      "num_tokens": 34678002.0,
      "step": 17735
    },
    {
      "entropy": 6.070802068710327,
      "epoch": 1.7733293347328436,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004690152297204808,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1741837218403816,
      "num_tokens": 34688112.0,
      "step": 17740
    },
    {
      "entropy": 6.063505363464356,
      "epoch": 1.773829159794072,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046899715378490427,
      "loss": 5.8485,
      "mean_token_accuracy": 0.16169725656509398,
      "num_tokens": 34697960.0,
      "step": 17745
    },
    {
      "entropy": 6.081278276443482,
      "epoch": 1.7743289848553006,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046897907296840927,
      "loss": 5.6872,
      "mean_token_accuracy": 0.1768806129693985,
      "num_tokens": 34708072.0,
      "step": 17750
    },
    {
      "entropy": 6.079470682144165,
      "epoch": 1.7748288099165292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004689609872714507,
      "loss": 5.74,
      "mean_token_accuracy": 0.16905046999454498,
      "num_tokens": 34717723.0,
      "step": 17755
    },
    {
      "entropy": 6.10003809928894,
      "epoch": 1.775328634977758,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046894289669448366,
      "loss": 5.8669,
      "mean_token_accuracy": 0.1630461722612381,
      "num_tokens": 34727326.0,
      "step": 17760
    },
    {
      "entropy": 6.075009346008301,
      "epoch": 1.7758284600389862,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046892480123796314,
      "loss": 5.8208,
      "mean_token_accuracy": 0.16311060935258864,
      "num_tokens": 34737203.0,
      "step": 17765
    },
    {
      "entropy": 6.072621774673462,
      "epoch": 1.776328285100215,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004689067009023446,
      "loss": 5.8188,
      "mean_token_accuracy": 0.1718278184533119,
      "num_tokens": 34746287.0,
      "step": 17770
    },
    {
      "entropy": 6.073715496063232,
      "epoch": 1.7768281101614436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004688885956880834,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1668981909751892,
      "num_tokens": 34757478.0,
      "step": 17775
    },
    {
      "entropy": 6.067406511306762,
      "epoch": 1.777327935222672,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004688704855956349,
      "loss": 5.7749,
      "mean_token_accuracy": 0.16988726258277892,
      "num_tokens": 34766474.0,
      "step": 17780
    },
    {
      "entropy": 6.099605989456177,
      "epoch": 1.7778277602839005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046885237062545493,
      "loss": 5.7472,
      "mean_token_accuracy": 0.16580326557159425,
      "num_tokens": 34775500.0,
      "step": 17785
    },
    {
      "entropy": 6.030214405059814,
      "epoch": 1.7783275853451292,
      "grad_norm": 1.25,
      "learning_rate": 0.00046883425077799923,
      "loss": 5.8689,
      "mean_token_accuracy": 0.16291454881429673,
      "num_tokens": 34785290.0,
      "step": 17790
    },
    {
      "entropy": 6.177517032623291,
      "epoch": 1.7788274104063577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004688161260537236,
      "loss": 6.0439,
      "mean_token_accuracy": 0.15314865559339524,
      "num_tokens": 34794485.0,
      "step": 17795
    },
    {
      "entropy": 6.1718339920043945,
      "epoch": 1.7793272354675862,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004687979964530842,
      "loss": 5.7807,
      "mean_token_accuracy": 0.1661211684346199,
      "num_tokens": 34803825.0,
      "step": 17800
    },
    {
      "entropy": 6.194694137573242,
      "epoch": 1.779827060528815,
      "grad_norm": 1.234375,
      "learning_rate": 0.000468779861976537,
      "loss": 5.919,
      "mean_token_accuracy": 0.15517112612724304,
      "num_tokens": 34812553.0,
      "step": 17805
    },
    {
      "entropy": 6.110612487792968,
      "epoch": 1.7803268855900436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046876172262453843,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16224512606859207,
      "num_tokens": 34822427.0,
      "step": 17810
    },
    {
      "entropy": 6.170865678787232,
      "epoch": 1.780826710651272,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046874357839754463,
      "loss": 6.0286,
      "mean_token_accuracy": 0.15399813503026963,
      "num_tokens": 34831918.0,
      "step": 17815
    },
    {
      "entropy": 6.159872770309448,
      "epoch": 1.7813265357125005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046872542929601234,
      "loss": 5.8007,
      "mean_token_accuracy": 0.16087912321090697,
      "num_tokens": 34840429.0,
      "step": 17820
    },
    {
      "entropy": 6.098322916030884,
      "epoch": 1.7818263607737292,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046870727532039816,
      "loss": 5.7735,
      "mean_token_accuracy": 0.16529863178730012,
      "num_tokens": 34850455.0,
      "step": 17825
    },
    {
      "entropy": 6.0868353843688965,
      "epoch": 1.7823261858349577,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004686891164711586,
      "loss": 5.7788,
      "mean_token_accuracy": 0.16819332540035248,
      "num_tokens": 34860062.0,
      "step": 17830
    },
    {
      "entropy": 6.112264919281006,
      "epoch": 1.7828260108961862,
      "grad_norm": 1.25,
      "learning_rate": 0.00046867095274875085,
      "loss": 5.7892,
      "mean_token_accuracy": 0.17066195011138915,
      "num_tokens": 34869987.0,
      "step": 17835
    },
    {
      "entropy": 6.109940147399902,
      "epoch": 1.783325835957415,
      "grad_norm": 1.5,
      "learning_rate": 0.00046865278415363174,
      "loss": 5.9215,
      "mean_token_accuracy": 0.1586624003946781,
      "num_tokens": 34881593.0,
      "step": 17840
    },
    {
      "entropy": 6.1461750030517575,
      "epoch": 1.7838256610186436,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004686346106862584,
      "loss": 5.8498,
      "mean_token_accuracy": 0.16108776032924652,
      "num_tokens": 34891468.0,
      "step": 17845
    },
    {
      "entropy": 6.113051223754883,
      "epoch": 1.784325486079872,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004686164323470881,
      "loss": 5.739,
      "mean_token_accuracy": 0.1710408553481102,
      "num_tokens": 34900591.0,
      "step": 17850
    },
    {
      "entropy": 6.095070314407349,
      "epoch": 1.7848253111411005,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004685982491365781,
      "loss": 5.7497,
      "mean_token_accuracy": 0.17346751540899277,
      "num_tokens": 34909514.0,
      "step": 17855
    },
    {
      "entropy": 6.133065271377563,
      "epoch": 1.7853251362023292,
      "grad_norm": 1.25,
      "learning_rate": 0.00046858006105518615,
      "loss": 5.7548,
      "mean_token_accuracy": 0.16432792395353318,
      "num_tokens": 34918836.0,
      "step": 17860
    },
    {
      "entropy": 6.008483409881592,
      "epoch": 1.7858249612635577,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004685618681033695,
      "loss": 5.6928,
      "mean_token_accuracy": 0.17484699189662933,
      "num_tokens": 34928040.0,
      "step": 17865
    },
    {
      "entropy": 6.048050165176392,
      "epoch": 1.7863247863247862,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004685436702815861,
      "loss": 5.7523,
      "mean_token_accuracy": 0.16790154129266738,
      "num_tokens": 34936879.0,
      "step": 17870
    },
    {
      "entropy": 6.090900707244873,
      "epoch": 1.786824611386015,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004685254675902938,
      "loss": 5.7849,
      "mean_token_accuracy": 0.1602693259716034,
      "num_tokens": 34946986.0,
      "step": 17875
    },
    {
      "entropy": 6.104190301895142,
      "epoch": 1.7873244364472436,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004685072600299505,
      "loss": 5.7991,
      "mean_token_accuracy": 0.16674963980913163,
      "num_tokens": 34956848.0,
      "step": 17880
    },
    {
      "entropy": 6.161003112792969,
      "epoch": 1.787824261508472,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004684890476010144,
      "loss": 5.8338,
      "mean_token_accuracy": 0.16626414507627488,
      "num_tokens": 34966656.0,
      "step": 17885
    },
    {
      "entropy": 6.101506042480469,
      "epoch": 1.7883240865697005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004684708303039435,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17311881929636003,
      "num_tokens": 34975672.0,
      "step": 17890
    },
    {
      "entropy": 6.031322002410889,
      "epoch": 1.7888239116309292,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046845260813919645,
      "loss": 5.7702,
      "mean_token_accuracy": 0.1732256144285202,
      "num_tokens": 34985253.0,
      "step": 17895
    },
    {
      "entropy": 6.05954008102417,
      "epoch": 1.7893237366921577,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046843438110723143,
      "loss": 5.7759,
      "mean_token_accuracy": 0.16959327459335327,
      "num_tokens": 34995115.0,
      "step": 17900
    },
    {
      "entropy": 6.109324932098389,
      "epoch": 1.7898235617533862,
      "grad_norm": 1.703125,
      "learning_rate": 0.00046841614920850724,
      "loss": 5.8024,
      "mean_token_accuracy": 0.16545298099517822,
      "num_tokens": 35005317.0,
      "step": 17905
    },
    {
      "entropy": 6.144032716751099,
      "epoch": 1.790323386814615,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046839791244348243,
      "loss": 5.8357,
      "mean_token_accuracy": 0.1691131740808487,
      "num_tokens": 35015296.0,
      "step": 17910
    },
    {
      "entropy": 6.017535209655762,
      "epoch": 1.7908232118758436,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046837967081261606,
      "loss": 5.5966,
      "mean_token_accuracy": 0.18248314112424852,
      "num_tokens": 35023575.0,
      "step": 17915
    },
    {
      "entropy": 6.09282693862915,
      "epoch": 1.791323036937072,
      "grad_norm": 1.25,
      "learning_rate": 0.0004683614243163668,
      "loss": 5.9062,
      "mean_token_accuracy": 0.16196359544992447,
      "num_tokens": 35034350.0,
      "step": 17920
    },
    {
      "entropy": 6.025299310684204,
      "epoch": 1.7918228619983005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004683431729551939,
      "loss": 5.7893,
      "mean_token_accuracy": 0.16732770651578904,
      "num_tokens": 35045125.0,
      "step": 17925
    },
    {
      "entropy": 6.088627004623413,
      "epoch": 1.7923226870595292,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004683249167295565,
      "loss": 5.7431,
      "mean_token_accuracy": 0.171597720682621,
      "num_tokens": 35055160.0,
      "step": 17930
    },
    {
      "entropy": 6.153690433502197,
      "epoch": 1.7928225121207577,
      "grad_norm": 1.21875,
      "learning_rate": 0.000468306655639914,
      "loss": 5.8954,
      "mean_token_accuracy": 0.15682633966207504,
      "num_tokens": 35065626.0,
      "step": 17935
    },
    {
      "entropy": 6.176308250427246,
      "epoch": 1.7933223371819862,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004682883896867257,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15984192341566086,
      "num_tokens": 35077294.0,
      "step": 17940
    },
    {
      "entropy": 6.067380046844482,
      "epoch": 1.793822162243215,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004682701188704513,
      "loss": 5.7419,
      "mean_token_accuracy": 0.1734014555811882,
      "num_tokens": 35086022.0,
      "step": 17945
    },
    {
      "entropy": 6.1342826843261715,
      "epoch": 1.7943219873044436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004682518431915505,
      "loss": 5.9295,
      "mean_token_accuracy": 0.1611875869333744,
      "num_tokens": 35095966.0,
      "step": 17950
    },
    {
      "entropy": 6.104007434844971,
      "epoch": 1.794821812365672,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046823356265048293,
      "loss": 5.7519,
      "mean_token_accuracy": 0.1616478905081749,
      "num_tokens": 35105554.0,
      "step": 17955
    },
    {
      "entropy": 6.106305408477783,
      "epoch": 1.7953216374269005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004682152772477087,
      "loss": 5.7309,
      "mean_token_accuracy": 0.1767505794763565,
      "num_tokens": 35114485.0,
      "step": 17960
    },
    {
      "entropy": 6.111332082748413,
      "epoch": 1.7958214624881292,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046819698698368785,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16337924599647521,
      "num_tokens": 35124765.0,
      "step": 17965
    },
    {
      "entropy": 6.111536502838135,
      "epoch": 1.7963212875493577,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046817869185888044,
      "loss": 5.8522,
      "mean_token_accuracy": 0.15905121862888336,
      "num_tokens": 35134474.0,
      "step": 17970
    },
    {
      "entropy": 6.116213607788086,
      "epoch": 1.7968211126105862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004681603918737469,
      "loss": 5.7996,
      "mean_token_accuracy": 0.17053399682044984,
      "num_tokens": 35144717.0,
      "step": 17975
    },
    {
      "entropy": 6.1886969089508055,
      "epoch": 1.7973209376718149,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046814208702874766,
      "loss": 5.8148,
      "mean_token_accuracy": 0.1715927705168724,
      "num_tokens": 35153889.0,
      "step": 17980
    },
    {
      "entropy": 6.038231897354126,
      "epoch": 1.7978207627330436,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046812377732434316,
      "loss": 5.7493,
      "mean_token_accuracy": 0.16788053661584854,
      "num_tokens": 35163909.0,
      "step": 17985
    },
    {
      "entropy": 6.125533628463745,
      "epoch": 1.7983205877942718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046810546276099415,
      "loss": 5.8538,
      "mean_token_accuracy": 0.15573169738054277,
      "num_tokens": 35174614.0,
      "step": 17990
    },
    {
      "entropy": 6.1276843547821045,
      "epoch": 1.7988204128555005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004680871433391613,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16732051819562913,
      "num_tokens": 35184568.0,
      "step": 17995
    },
    {
      "entropy": 6.073163414001465,
      "epoch": 1.7993202379167292,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046806881905930565,
      "loss": 5.7972,
      "mean_token_accuracy": 0.17095420360565186,
      "num_tokens": 35194594.0,
      "step": 18000
    },
    {
      "epoch": 1.7993202379167292,
      "eval_entropy": 5.878379203031678,
      "eval_loss": 5.912787914276123,
      "eval_mean_token_accuracy": 0.1680029729163791,
      "eval_num_tokens": 35194594.0,
      "eval_runtime": 23.8276,
      "eval_samples_per_second": 1302.901,
      "eval_steps_per_second": 162.878,
      "step": 18000
    },
    {
      "entropy": 6.0592217445373535,
      "epoch": 1.7998200629779577,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046805048992188823,
      "loss": 5.7559,
      "mean_token_accuracy": 0.1684352993965149,
      "num_tokens": 35204710.0,
      "step": 18005
    },
    {
      "entropy": 6.205924940109253,
      "epoch": 1.8003198880391862,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046803215592737014,
      "loss": 5.9625,
      "mean_token_accuracy": 0.1491615131497383,
      "num_tokens": 35215531.0,
      "step": 18010
    },
    {
      "entropy": 6.15948166847229,
      "epoch": 1.8008197131004149,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046801381707621266,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16124841719865798,
      "num_tokens": 35224274.0,
      "step": 18015
    },
    {
      "entropy": 6.067934942245484,
      "epoch": 1.8013195381616436,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046799547336887727,
      "loss": 5.7664,
      "mean_token_accuracy": 0.1718728393316269,
      "num_tokens": 35234341.0,
      "step": 18020
    },
    {
      "entropy": 6.101414537429809,
      "epoch": 1.8018193632228718,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004679771248058254,
      "loss": 5.7899,
      "mean_token_accuracy": 0.15688160955905914,
      "num_tokens": 35244841.0,
      "step": 18025
    },
    {
      "entropy": 6.139003038406372,
      "epoch": 1.8023191882841005,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004679587713875187,
      "loss": 5.8824,
      "mean_token_accuracy": 0.15966786742210387,
      "num_tokens": 35254770.0,
      "step": 18030
    },
    {
      "entropy": 6.180947303771973,
      "epoch": 1.8028190133453292,
      "grad_norm": 1.375,
      "learning_rate": 0.000467940413114419,
      "loss": 5.7995,
      "mean_token_accuracy": 0.1655028611421585,
      "num_tokens": 35263676.0,
      "step": 18035
    },
    {
      "entropy": 6.095750570297241,
      "epoch": 1.8033188384065577,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004679220499869881,
      "loss": 5.7975,
      "mean_token_accuracy": 0.1576593965291977,
      "num_tokens": 35273845.0,
      "step": 18040
    },
    {
      "entropy": 6.15850601196289,
      "epoch": 1.8038186634677862,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004679036820056881,
      "loss": 5.9649,
      "mean_token_accuracy": 0.15750037506222725,
      "num_tokens": 35285121.0,
      "step": 18045
    },
    {
      "entropy": 6.190390968322754,
      "epoch": 1.8043184885290149,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046788530917098105,
      "loss": 5.8525,
      "mean_token_accuracy": 0.16261489391326905,
      "num_tokens": 35294601.0,
      "step": 18050
    },
    {
      "entropy": 6.126473236083984,
      "epoch": 1.8048183135902434,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004678669314833293,
      "loss": 5.8462,
      "mean_token_accuracy": 0.161384903639555,
      "num_tokens": 35304456.0,
      "step": 18055
    },
    {
      "entropy": 6.073666667938232,
      "epoch": 1.8053181386514718,
      "grad_norm": 1.125,
      "learning_rate": 0.00046784854894319525,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16789950281381608,
      "num_tokens": 35314333.0,
      "step": 18060
    },
    {
      "entropy": 6.143755006790161,
      "epoch": 1.8058179637127005,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046783016155104123,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16419412940740585,
      "num_tokens": 35325267.0,
      "step": 18065
    },
    {
      "entropy": 6.183286380767822,
      "epoch": 1.8063177887739292,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046781176930732997,
      "loss": 5.8124,
      "mean_token_accuracy": 0.1641531318426132,
      "num_tokens": 35334232.0,
      "step": 18070
    },
    {
      "entropy": 6.116700792312622,
      "epoch": 1.8068176138351577,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004677933722125243,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16146454811096192,
      "num_tokens": 35344961.0,
      "step": 18075
    },
    {
      "entropy": 6.129605960845947,
      "epoch": 1.8073174388963862,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004677749702670869,
      "loss": 5.8939,
      "mean_token_accuracy": 0.16636534705758094,
      "num_tokens": 35354719.0,
      "step": 18080
    },
    {
      "entropy": 6.0455484867095945,
      "epoch": 1.8078172639576149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004677565634714809,
      "loss": 5.7597,
      "mean_token_accuracy": 0.17157918810844422,
      "num_tokens": 35365301.0,
      "step": 18085
    },
    {
      "entropy": 6.045155000686646,
      "epoch": 1.8083170890188434,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004677381518261693,
      "loss": 5.8112,
      "mean_token_accuracy": 0.1658891260623932,
      "num_tokens": 35375586.0,
      "step": 18090
    },
    {
      "entropy": 6.112083101272583,
      "epoch": 1.8088169140800718,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046771973533161543,
      "loss": 5.8328,
      "mean_token_accuracy": 0.16432284712791442,
      "num_tokens": 35385226.0,
      "step": 18095
    },
    {
      "entropy": 6.081790256500244,
      "epoch": 1.8093167391413005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046770131398828264,
      "loss": 5.76,
      "mean_token_accuracy": 0.1685410887002945,
      "num_tokens": 35394872.0,
      "step": 18100
    },
    {
      "entropy": 6.009259462356567,
      "epoch": 1.8098165642025292,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046768288779663436,
      "loss": 5.7059,
      "mean_token_accuracy": 0.16963390707969667,
      "num_tokens": 35403790.0,
      "step": 18105
    },
    {
      "entropy": 6.124264287948608,
      "epoch": 1.8103163892637577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004676644567571342,
      "loss": 5.8577,
      "mean_token_accuracy": 0.1636961206793785,
      "num_tokens": 35414042.0,
      "step": 18110
    },
    {
      "entropy": 6.106362247467041,
      "epoch": 1.8108162143249862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046764602087024596,
      "loss": 5.6987,
      "mean_token_accuracy": 0.16814630180597306,
      "num_tokens": 35423345.0,
      "step": 18115
    },
    {
      "entropy": 6.06546835899353,
      "epoch": 1.8113160393862149,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004676275801364332,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16468493789434432,
      "num_tokens": 35433284.0,
      "step": 18120
    },
    {
      "entropy": 6.109310483932495,
      "epoch": 1.8118158644474434,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004676091345561603,
      "loss": 5.8327,
      "mean_token_accuracy": 0.158689384162426,
      "num_tokens": 35443258.0,
      "step": 18125
    },
    {
      "entropy": 6.09218282699585,
      "epoch": 1.8123156895086718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000467590684129891,
      "loss": 5.8664,
      "mean_token_accuracy": 0.16371462792158126,
      "num_tokens": 35453214.0,
      "step": 18130
    },
    {
      "entropy": 6.163098669052124,
      "epoch": 1.8128155145699005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046757222885808973,
      "loss": 5.7722,
      "mean_token_accuracy": 0.16917706280946732,
      "num_tokens": 35462802.0,
      "step": 18135
    },
    {
      "entropy": 6.037716674804687,
      "epoch": 1.8133153396311292,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046755376874122075,
      "loss": 5.6825,
      "mean_token_accuracy": 0.17584676593542098,
      "num_tokens": 35472423.0,
      "step": 18140
    },
    {
      "entropy": 6.110927248001099,
      "epoch": 1.8138151646923577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004675353037797485,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16236671954393386,
      "num_tokens": 35481835.0,
      "step": 18145
    },
    {
      "entropy": 6.09594235420227,
      "epoch": 1.8143149897535862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004675168339741375,
      "loss": 5.8327,
      "mean_token_accuracy": 0.1634435161948204,
      "num_tokens": 35491834.0,
      "step": 18150
    },
    {
      "entropy": 6.093013858795166,
      "epoch": 1.8148148148148149,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004674983593248525,
      "loss": 5.8214,
      "mean_token_accuracy": 0.16101322919130326,
      "num_tokens": 35501591.0,
      "step": 18155
    },
    {
      "entropy": 6.018006896972656,
      "epoch": 1.8153146398760434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046747987983235836,
      "loss": 5.7672,
      "mean_token_accuracy": 0.16541654765605926,
      "num_tokens": 35511720.0,
      "step": 18160
    },
    {
      "entropy": 6.155573558807373,
      "epoch": 1.8158144649372718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046746139549711993,
      "loss": 5.8566,
      "mean_token_accuracy": 0.16430281102657318,
      "num_tokens": 35521040.0,
      "step": 18165
    },
    {
      "entropy": 6.049437618255615,
      "epoch": 1.8163142899985005,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004674429063196023,
      "loss": 5.7343,
      "mean_token_accuracy": 0.17319744527339936,
      "num_tokens": 35530919.0,
      "step": 18170
    },
    {
      "entropy": 6.025699663162231,
      "epoch": 1.8168141150597292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046742441230027067,
      "loss": 5.707,
      "mean_token_accuracy": 0.1764920473098755,
      "num_tokens": 35539964.0,
      "step": 18175
    },
    {
      "entropy": 6.082205820083618,
      "epoch": 1.8173139401209577,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004674059134395904,
      "loss": 5.8238,
      "mean_token_accuracy": 0.16726171374320983,
      "num_tokens": 35549679.0,
      "step": 18180
    },
    {
      "entropy": 6.162162017822266,
      "epoch": 1.8178137651821862,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004673874097380268,
      "loss": 5.915,
      "mean_token_accuracy": 0.15371926575899125,
      "num_tokens": 35560159.0,
      "step": 18185
    },
    {
      "entropy": 6.188117408752442,
      "epoch": 1.8183135902434149,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004673689011960455,
      "loss": 5.9426,
      "mean_token_accuracy": 0.15697686225175858,
      "num_tokens": 35571591.0,
      "step": 18190
    },
    {
      "entropy": 6.11955509185791,
      "epoch": 1.8188134153046434,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046735038781411204,
      "loss": 5.7331,
      "mean_token_accuracy": 0.17004652470350265,
      "num_tokens": 35581641.0,
      "step": 18195
    },
    {
      "entropy": 6.11912088394165,
      "epoch": 1.8193132403658718,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004673318695926925,
      "loss": 5.9206,
      "mean_token_accuracy": 0.16183921992778777,
      "num_tokens": 35591738.0,
      "step": 18200
    },
    {
      "entropy": 6.18521637916565,
      "epoch": 1.8198130654271005,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004673133465322525,
      "loss": 5.9356,
      "mean_token_accuracy": 0.1587623432278633,
      "num_tokens": 35600818.0,
      "step": 18205
    },
    {
      "entropy": 6.167316532135009,
      "epoch": 1.8203128904883292,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004672948186332582,
      "loss": 5.7999,
      "mean_token_accuracy": 0.16857050955295563,
      "num_tokens": 35610180.0,
      "step": 18210
    },
    {
      "entropy": 6.055397033691406,
      "epoch": 1.8208127155495577,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046727628589617574,
      "loss": 5.8014,
      "mean_token_accuracy": 0.1679349347949028,
      "num_tokens": 35620894.0,
      "step": 18215
    },
    {
      "entropy": 6.1317016124725345,
      "epoch": 1.8213125406107862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046725774832147147,
      "loss": 5.9097,
      "mean_token_accuracy": 0.16099715679883958,
      "num_tokens": 35630976.0,
      "step": 18220
    },
    {
      "entropy": 6.1583582878112795,
      "epoch": 1.8218123656720149,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046723920590961167,
      "loss": 5.8819,
      "mean_token_accuracy": 0.16028386503458023,
      "num_tokens": 35640436.0,
      "step": 18225
    },
    {
      "entropy": 6.164221143722534,
      "epoch": 1.8223121907332434,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004672206586610629,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16340779066085814,
      "num_tokens": 35650533.0,
      "step": 18230
    },
    {
      "entropy": 6.043798303604126,
      "epoch": 1.8228120157944718,
      "grad_norm": 1.25,
      "learning_rate": 0.00046720210657629185,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16939725875854492,
      "num_tokens": 35660748.0,
      "step": 18235
    },
    {
      "entropy": 6.073694562911987,
      "epoch": 1.8233118408557005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046718354965576527,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16632049083709716,
      "num_tokens": 35671186.0,
      "step": 18240
    },
    {
      "entropy": 6.088987541198731,
      "epoch": 1.8238116659169292,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046716498789995,
      "loss": 5.7498,
      "mean_token_accuracy": 0.1727837383747101,
      "num_tokens": 35681784.0,
      "step": 18245
    },
    {
      "entropy": 6.119880533218383,
      "epoch": 1.8243114909781575,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046714642130931306,
      "loss": 5.8089,
      "mean_token_accuracy": 0.16464951783418655,
      "num_tokens": 35692110.0,
      "step": 18250
    },
    {
      "entropy": 6.050120162963867,
      "epoch": 1.8248113160393862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046712784988432173,
      "loss": 5.7699,
      "mean_token_accuracy": 0.16364702880382537,
      "num_tokens": 35702051.0,
      "step": 18255
    },
    {
      "entropy": 6.07155442237854,
      "epoch": 1.8253111411006149,
      "grad_norm": 1.359375,
      "learning_rate": 0.000467109273625443,
      "loss": 5.8008,
      "mean_token_accuracy": 0.1665101021528244,
      "num_tokens": 35711368.0,
      "step": 18260
    },
    {
      "entropy": 6.126394128799438,
      "epoch": 1.8258109661618434,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004670906925331444,
      "loss": 5.9048,
      "mean_token_accuracy": 0.1658518597483635,
      "num_tokens": 35720868.0,
      "step": 18265
    },
    {
      "entropy": 6.112776184082032,
      "epoch": 1.8263107912230718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046707210660789343,
      "loss": 5.808,
      "mean_token_accuracy": 0.16679180264472962,
      "num_tokens": 35730697.0,
      "step": 18270
    },
    {
      "entropy": 6.035299158096313,
      "epoch": 1.8268106162843005,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004670535158501577,
      "loss": 5.7722,
      "mean_token_accuracy": 0.1672576352953911,
      "num_tokens": 35740341.0,
      "step": 18275
    },
    {
      "entropy": 5.992048215866089,
      "epoch": 1.8273104413455292,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046703492026040495,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17962094992399216,
      "num_tokens": 35749324.0,
      "step": 18280
    },
    {
      "entropy": 6.1446703433990475,
      "epoch": 1.8278102664067575,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046701631983910296,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16052650958299636,
      "num_tokens": 35759650.0,
      "step": 18285
    },
    {
      "entropy": 6.065414237976074,
      "epoch": 1.8283100914679862,
      "grad_norm": 1.375,
      "learning_rate": 0.00046699771458671974,
      "loss": 5.7455,
      "mean_token_accuracy": 0.16704541593790054,
      "num_tokens": 35770586.0,
      "step": 18290
    },
    {
      "entropy": 6.121974849700928,
      "epoch": 1.8288099165292149,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046697910450372356,
      "loss": 5.8508,
      "mean_token_accuracy": 0.1641160249710083,
      "num_tokens": 35780204.0,
      "step": 18295
    },
    {
      "entropy": 6.092642402648925,
      "epoch": 1.8293097415904434,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046696048959058236,
      "loss": 5.79,
      "mean_token_accuracy": 0.1616687908768654,
      "num_tokens": 35790008.0,
      "step": 18300
    },
    {
      "entropy": 5.993779516220092,
      "epoch": 1.8298095666516718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004669418698477647,
      "loss": 5.7613,
      "mean_token_accuracy": 0.16684951931238173,
      "num_tokens": 35799752.0,
      "step": 18305
    },
    {
      "entropy": 6.143229913711548,
      "epoch": 1.8303093917129005,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046692324527573907,
      "loss": 5.8644,
      "mean_token_accuracy": 0.15887764543294908,
      "num_tokens": 35809779.0,
      "step": 18310
    },
    {
      "entropy": 6.137984418869019,
      "epoch": 1.830809216774129,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004669046158749738,
      "loss": 5.8056,
      "mean_token_accuracy": 0.1654566317796707,
      "num_tokens": 35819461.0,
      "step": 18315
    },
    {
      "entropy": 6.134438991546631,
      "epoch": 1.8313090418353575,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046688598164593793,
      "loss": 5.8603,
      "mean_token_accuracy": 0.16746909618377687,
      "num_tokens": 35828868.0,
      "step": 18320
    },
    {
      "entropy": 6.1127923965454105,
      "epoch": 1.8318088668965862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004668673425891001,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16024307161569595,
      "num_tokens": 35839102.0,
      "step": 18325
    },
    {
      "entropy": 6.066645908355713,
      "epoch": 1.8323086919578149,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046684869870492927,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17193718403577804,
      "num_tokens": 35848884.0,
      "step": 18330
    },
    {
      "entropy": 6.047945642471314,
      "epoch": 1.8328085170190433,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046683004999389455,
      "loss": 5.7792,
      "mean_token_accuracy": 0.1764673426747322,
      "num_tokens": 35858820.0,
      "step": 18335
    },
    {
      "entropy": 6.152068567276001,
      "epoch": 1.8333083420802718,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004668113964564651,
      "loss": 5.8257,
      "mean_token_accuracy": 0.15874118953943253,
      "num_tokens": 35868167.0,
      "step": 18340
    },
    {
      "entropy": 6.100328016281128,
      "epoch": 1.8338081671415005,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004667927380931104,
      "loss": 5.8605,
      "mean_token_accuracy": 0.15748917162418366,
      "num_tokens": 35877544.0,
      "step": 18345
    },
    {
      "entropy": 6.038480138778686,
      "epoch": 1.834307992202729,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004667740749042997,
      "loss": 5.7656,
      "mean_token_accuracy": 0.16848409920930862,
      "num_tokens": 35887263.0,
      "step": 18350
    },
    {
      "entropy": 6.1425727844238285,
      "epoch": 1.8348078172639575,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046675540689050253,
      "loss": 5.8456,
      "mean_token_accuracy": 0.1598464846611023,
      "num_tokens": 35896833.0,
      "step": 18355
    },
    {
      "entropy": 6.074813270568848,
      "epoch": 1.8353076423251862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046673673405218876,
      "loss": 5.7402,
      "mean_token_accuracy": 0.1707631155848503,
      "num_tokens": 35905477.0,
      "step": 18360
    },
    {
      "entropy": 6.11413803100586,
      "epoch": 1.8358074673864149,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004667180563898281,
      "loss": 5.8961,
      "mean_token_accuracy": 0.15892637073993682,
      "num_tokens": 35916261.0,
      "step": 18365
    },
    {
      "entropy": 6.071942663192749,
      "epoch": 1.8363072924476433,
      "grad_norm": 1.25,
      "learning_rate": 0.00046669937390389043,
      "loss": 5.7353,
      "mean_token_accuracy": 0.17189619243144988,
      "num_tokens": 35926188.0,
      "step": 18370
    },
    {
      "entropy": 6.012853574752808,
      "epoch": 1.8368071175088718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046668068659484584,
      "loss": 5.661,
      "mean_token_accuracy": 0.17279405891895294,
      "num_tokens": 35935970.0,
      "step": 18375
    },
    {
      "entropy": 6.082989454269409,
      "epoch": 1.8373069425701005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046666199446316454,
      "loss": 5.8466,
      "mean_token_accuracy": 0.1640591338276863,
      "num_tokens": 35946220.0,
      "step": 18380
    },
    {
      "entropy": 6.036368560791016,
      "epoch": 1.837806767631329,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046664329750931675,
      "loss": 5.6818,
      "mean_token_accuracy": 0.17640633136034012,
      "num_tokens": 35955918.0,
      "step": 18385
    },
    {
      "entropy": 6.199508905410767,
      "epoch": 1.8383065926925575,
      "grad_norm": 1.265625,
      "learning_rate": 0.000466624595733773,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15450822040438653,
      "num_tokens": 35966136.0,
      "step": 18390
    },
    {
      "entropy": 6.093500804901123,
      "epoch": 1.8388064177537862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004666058891370036,
      "loss": 5.7624,
      "mean_token_accuracy": 0.16331479400396348,
      "num_tokens": 35975775.0,
      "step": 18395
    },
    {
      "entropy": 6.103196716308593,
      "epoch": 1.8393062428150149,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004665871777194794,
      "loss": 5.7767,
      "mean_token_accuracy": 0.17077198773622512,
      "num_tokens": 35984637.0,
      "step": 18400
    },
    {
      "entropy": 6.127803230285645,
      "epoch": 1.8398060678762433,
      "grad_norm": 1.59375,
      "learning_rate": 0.00046656846148167107,
      "loss": 5.8043,
      "mean_token_accuracy": 0.16308410912752153,
      "num_tokens": 35993944.0,
      "step": 18405
    },
    {
      "entropy": 6.122818517684936,
      "epoch": 1.8403058929374718,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046654974042404957,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16634065806865692,
      "num_tokens": 36003442.0,
      "step": 18410
    },
    {
      "entropy": 6.074848604202271,
      "epoch": 1.8408057179987005,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046653101454708593,
      "loss": 5.7858,
      "mean_token_accuracy": 0.16688190400600433,
      "num_tokens": 36014354.0,
      "step": 18415
    },
    {
      "entropy": 6.2420876026153564,
      "epoch": 1.841305543059929,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004665122838512512,
      "loss": 5.9578,
      "mean_token_accuracy": 0.15925640016794204,
      "num_tokens": 36025499.0,
      "step": 18420
    },
    {
      "entropy": 6.220799207687378,
      "epoch": 1.8418053681211575,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004664935483370167,
      "loss": 5.876,
      "mean_token_accuracy": 0.16462518721818925,
      "num_tokens": 36036261.0,
      "step": 18425
    },
    {
      "entropy": 6.130304479598999,
      "epoch": 1.8423051931823862,
      "grad_norm": 1.25,
      "learning_rate": 0.0004664748080048537,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16379659175872802,
      "num_tokens": 36045663.0,
      "step": 18430
    },
    {
      "entropy": 6.001801586151123,
      "epoch": 1.8428050182436149,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046645606285523395,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1708531051874161,
      "num_tokens": 36055491.0,
      "step": 18435
    },
    {
      "entropy": 6.102457046508789,
      "epoch": 1.8433048433048433,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046643731288862884,
      "loss": 5.8849,
      "mean_token_accuracy": 0.15847297310829161,
      "num_tokens": 36066379.0,
      "step": 18440
    },
    {
      "entropy": 6.128042364120484,
      "epoch": 1.8438046683660718,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004664185581055103,
      "loss": 5.8788,
      "mean_token_accuracy": 0.16758137792348862,
      "num_tokens": 36075160.0,
      "step": 18445
    },
    {
      "entropy": 6.118989419937134,
      "epoch": 1.8443044934273005,
      "grad_norm": 1.4375,
      "learning_rate": 0.00046639979850634995,
      "loss": 5.8163,
      "mean_token_accuracy": 0.16133926883339883,
      "num_tokens": 36084620.0,
      "step": 18450
    },
    {
      "entropy": 6.127137231826782,
      "epoch": 1.844804318488529,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046638103409162,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15540297627449035,
      "num_tokens": 36095289.0,
      "step": 18455
    },
    {
      "entropy": 6.118347787857056,
      "epoch": 1.8453041435497575,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046636226486179244,
      "loss": 5.8575,
      "mean_token_accuracy": 0.16609905660152435,
      "num_tokens": 36105352.0,
      "step": 18460
    },
    {
      "entropy": 6.067371273040772,
      "epoch": 1.8458039686109862,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004663434908173396,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17448633909225464,
      "num_tokens": 36114722.0,
      "step": 18465
    },
    {
      "entropy": 6.060323476791382,
      "epoch": 1.8463037936722149,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004663247119587338,
      "loss": 5.8334,
      "mean_token_accuracy": 0.1652819573879242,
      "num_tokens": 36123745.0,
      "step": 18470
    },
    {
      "entropy": 6.067274332046509,
      "epoch": 1.8468036187334433,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004663059282864474,
      "loss": 5.8019,
      "mean_token_accuracy": 0.1602480486035347,
      "num_tokens": 36133315.0,
      "step": 18475
    },
    {
      "entropy": 6.149957370758057,
      "epoch": 1.8473034437946718,
      "grad_norm": 1.25,
      "learning_rate": 0.0004662871398009531,
      "loss": 5.7802,
      "mean_token_accuracy": 0.16328724026679992,
      "num_tokens": 36142990.0,
      "step": 18480
    },
    {
      "entropy": 6.13552827835083,
      "epoch": 1.8478032688559005,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004662683465027235,
      "loss": 5.8012,
      "mean_token_accuracy": 0.1631546601653099,
      "num_tokens": 36152804.0,
      "step": 18485
    },
    {
      "entropy": 6.111448669433594,
      "epoch": 1.848303093917129,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00046624954839223165,
      "loss": 5.8811,
      "mean_token_accuracy": 0.1598628729581833,
      "num_tokens": 36161772.0,
      "step": 18490
    },
    {
      "entropy": 6.026418399810791,
      "epoch": 1.8488029189783575,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046623074546995024,
      "loss": 5.7912,
      "mean_token_accuracy": 0.17030014395713805,
      "num_tokens": 36170726.0,
      "step": 18495
    },
    {
      "entropy": 6.0811797142028805,
      "epoch": 1.8493027440395862,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046621193773635256,
      "loss": 5.8398,
      "mean_token_accuracy": 0.15627398937940598,
      "num_tokens": 36181972.0,
      "step": 18500
    },
    {
      "entropy": 6.12644853591919,
      "epoch": 1.8498025691008149,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004661931251919118,
      "loss": 5.7903,
      "mean_token_accuracy": 0.16466337144374849,
      "num_tokens": 36191527.0,
      "step": 18505
    },
    {
      "entropy": 6.181639099121094,
      "epoch": 1.8503023941620433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046617430783710105,
      "loss": 5.9519,
      "mean_token_accuracy": 0.1571066662669182,
      "num_tokens": 36201591.0,
      "step": 18510
    },
    {
      "entropy": 6.09990177154541,
      "epoch": 1.8508022192232718,
      "grad_norm": 1.390625,
      "learning_rate": 0.000466155485672394,
      "loss": 5.8229,
      "mean_token_accuracy": 0.169594407081604,
      "num_tokens": 36211893.0,
      "step": 18515
    },
    {
      "entropy": 6.091359186172485,
      "epoch": 1.8513020442845005,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004661366586982641,
      "loss": 5.8171,
      "mean_token_accuracy": 0.16523477733135222,
      "num_tokens": 36221520.0,
      "step": 18520
    },
    {
      "entropy": 6.095824432373047,
      "epoch": 1.851801869345729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004661178269151851,
      "loss": 5.8122,
      "mean_token_accuracy": 0.17202051281929015,
      "num_tokens": 36231633.0,
      "step": 18525
    },
    {
      "entropy": 6.180442905426025,
      "epoch": 1.8523016944069575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046609899032363074,
      "loss": 5.8924,
      "mean_token_accuracy": 0.1621579185128212,
      "num_tokens": 36242514.0,
      "step": 18530
    },
    {
      "entropy": 6.171078729629516,
      "epoch": 1.8528015194681862,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004660801489240749,
      "loss": 5.8472,
      "mean_token_accuracy": 0.1600303366780281,
      "num_tokens": 36252631.0,
      "step": 18535
    },
    {
      "entropy": 6.095125150680542,
      "epoch": 1.8533013445294149,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046606130271699175,
      "loss": 5.7605,
      "mean_token_accuracy": 0.1647081807255745,
      "num_tokens": 36261894.0,
      "step": 18540
    },
    {
      "entropy": 6.070320177078247,
      "epoch": 1.8538011695906431,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004660424517028553,
      "loss": 5.9125,
      "mean_token_accuracy": 0.16081089228391648,
      "num_tokens": 36271483.0,
      "step": 18545
    },
    {
      "entropy": 6.152283143997193,
      "epoch": 1.8543009946518718,
      "grad_norm": 1.46875,
      "learning_rate": 0.00046602359588214,
      "loss": 5.849,
      "mean_token_accuracy": 0.1642509937286377,
      "num_tokens": 36281038.0,
      "step": 18550
    },
    {
      "entropy": 6.041108179092407,
      "epoch": 1.8548008197131005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004660047352553202,
      "loss": 5.6527,
      "mean_token_accuracy": 0.1831817612051964,
      "num_tokens": 36289718.0,
      "step": 18555
    },
    {
      "entropy": 6.0626462459564205,
      "epoch": 1.855300644774329,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046598586982287034,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16974541395902634,
      "num_tokens": 36299912.0,
      "step": 18560
    },
    {
      "entropy": 6.078205871582031,
      "epoch": 1.8558004698355575,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046596699958526515,
      "loss": 5.8208,
      "mean_token_accuracy": 0.15900982916355133,
      "num_tokens": 36309686.0,
      "step": 18565
    },
    {
      "entropy": 6.142557382583618,
      "epoch": 1.8563002948967862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046594812454297944,
      "loss": 5.825,
      "mean_token_accuracy": 0.16723460853099822,
      "num_tokens": 36319247.0,
      "step": 18570
    },
    {
      "entropy": 6.026694822311401,
      "epoch": 1.8568001199580149,
      "grad_norm": 1.5,
      "learning_rate": 0.000465929244696488,
      "loss": 5.7621,
      "mean_token_accuracy": 0.16881083995103835,
      "num_tokens": 36329149.0,
      "step": 18575
    },
    {
      "entropy": 6.064014863967896,
      "epoch": 1.8572999450192431,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004659103600462659,
      "loss": 5.812,
      "mean_token_accuracy": 0.1681393191218376,
      "num_tokens": 36338857.0,
      "step": 18580
    },
    {
      "entropy": 6.135358095169067,
      "epoch": 1.8577997700804718,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004658914705927883,
      "loss": 5.8526,
      "mean_token_accuracy": 0.1629018932580948,
      "num_tokens": 36349415.0,
      "step": 18585
    },
    {
      "entropy": 6.107405519485473,
      "epoch": 1.8582995951417005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004658725763365304,
      "loss": 5.816,
      "mean_token_accuracy": 0.16507086604833604,
      "num_tokens": 36359597.0,
      "step": 18590
    },
    {
      "entropy": 6.045396947860718,
      "epoch": 1.858799420202929,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004658536772779676,
      "loss": 5.7781,
      "mean_token_accuracy": 0.17121962010860442,
      "num_tokens": 36369816.0,
      "step": 18595
    },
    {
      "entropy": 6.110804700851441,
      "epoch": 1.8592992452641575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004658347734175754,
      "loss": 5.8577,
      "mean_token_accuracy": 0.15888345390558242,
      "num_tokens": 36380624.0,
      "step": 18600
    },
    {
      "entropy": 6.119035482406616,
      "epoch": 1.8597990703253862,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046581586475582946,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16826989203691484,
      "num_tokens": 36390597.0,
      "step": 18605
    },
    {
      "entropy": 6.134745693206787,
      "epoch": 1.8602988953866146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004657969512932054,
      "loss": 5.7889,
      "mean_token_accuracy": 0.1601613864302635,
      "num_tokens": 36400404.0,
      "step": 18610
    },
    {
      "entropy": 6.081156396865845,
      "epoch": 1.8607987204478431,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004657780330301791,
      "loss": 5.7816,
      "mean_token_accuracy": 0.16672659367322923,
      "num_tokens": 36409599.0,
      "step": 18615
    },
    {
      "entropy": 6.0154530048370365,
      "epoch": 1.8612985455090718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046575910996722657,
      "loss": 5.7317,
      "mean_token_accuracy": 0.1695237085223198,
      "num_tokens": 36419217.0,
      "step": 18620
    },
    {
      "entropy": 6.082343816757202,
      "epoch": 1.8617983705703005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000465740182104824,
      "loss": 5.8165,
      "mean_token_accuracy": 0.16734157428145408,
      "num_tokens": 36429097.0,
      "step": 18625
    },
    {
      "entropy": 6.086805963516236,
      "epoch": 1.862298195631529,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046572124944344745,
      "loss": 5.7867,
      "mean_token_accuracy": 0.17236936390399932,
      "num_tokens": 36439260.0,
      "step": 18630
    },
    {
      "entropy": 6.089321231842041,
      "epoch": 1.8627980206927575,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046570231198357344,
      "loss": 5.6982,
      "mean_token_accuracy": 0.16841880232095718,
      "num_tokens": 36449083.0,
      "step": 18635
    },
    {
      "entropy": 6.125030517578125,
      "epoch": 1.8632978457539862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004656833697256783,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16987740695476533,
      "num_tokens": 36459337.0,
      "step": 18640
    },
    {
      "entropy": 6.076115846633911,
      "epoch": 1.8637976708152146,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004656644226702386,
      "loss": 5.8408,
      "mean_token_accuracy": 0.16338631957769395,
      "num_tokens": 36469132.0,
      "step": 18645
    },
    {
      "entropy": 6.094110345840454,
      "epoch": 1.8642974958764431,
      "grad_norm": 1.125,
      "learning_rate": 0.00046564547081773115,
      "loss": 5.8391,
      "mean_token_accuracy": 0.16581190079450608,
      "num_tokens": 36479111.0,
      "step": 18650
    },
    {
      "entropy": 6.113719272613525,
      "epoch": 1.8647973209376718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004656265141686326,
      "loss": 5.7519,
      "mean_token_accuracy": 0.16952608078718184,
      "num_tokens": 36488553.0,
      "step": 18655
    },
    {
      "entropy": 6.112332487106324,
      "epoch": 1.8652971459989005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046560755272342014,
      "loss": 5.8472,
      "mean_token_accuracy": 0.165032497048378,
      "num_tokens": 36498719.0,
      "step": 18660
    },
    {
      "entropy": 6.14487361907959,
      "epoch": 1.865796971060129,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004655885864825706,
      "loss": 5.801,
      "mean_token_accuracy": 0.16636282056570054,
      "num_tokens": 36508332.0,
      "step": 18665
    },
    {
      "entropy": 6.063628435134888,
      "epoch": 1.8662967961213575,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004655696154465613,
      "loss": 5.795,
      "mean_token_accuracy": 0.1699405238032341,
      "num_tokens": 36517867.0,
      "step": 18670
    },
    {
      "entropy": 6.129523658752442,
      "epoch": 1.8667966211825862,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046555063961586956,
      "loss": 5.7826,
      "mean_token_accuracy": 0.16653148233890533,
      "num_tokens": 36526734.0,
      "step": 18675
    },
    {
      "entropy": 6.086728429794311,
      "epoch": 1.8672964462438146,
      "grad_norm": 1.609375,
      "learning_rate": 0.0004655316589909727,
      "loss": 5.7886,
      "mean_token_accuracy": 0.16238559037446976,
      "num_tokens": 36536381.0,
      "step": 18680
    },
    {
      "entropy": 6.1039000988006595,
      "epoch": 1.867796271305043,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004655126735723484,
      "loss": 5.803,
      "mean_token_accuracy": 0.16139552295207976,
      "num_tokens": 36545708.0,
      "step": 18685
    },
    {
      "entropy": 6.166023921966553,
      "epoch": 1.8682960963662718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046549368336047415,
      "loss": 5.8642,
      "mean_token_accuracy": 0.16313666105270386,
      "num_tokens": 36555616.0,
      "step": 18690
    },
    {
      "entropy": 6.164104318618774,
      "epoch": 1.8687959214275005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004654746883558279,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15622997879981995,
      "num_tokens": 36565394.0,
      "step": 18695
    },
    {
      "entropy": 6.112481260299683,
      "epoch": 1.869295746488729,
      "grad_norm": 1.25,
      "learning_rate": 0.0004654556885588875,
      "loss": 5.7386,
      "mean_token_accuracy": 0.17511805891990662,
      "num_tokens": 36576329.0,
      "step": 18700
    },
    {
      "entropy": 6.016382837295533,
      "epoch": 1.8697955715499575,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046543668397013103,
      "loss": 5.7495,
      "mean_token_accuracy": 0.16809244751930236,
      "num_tokens": 36585402.0,
      "step": 18705
    },
    {
      "entropy": 6.1395848274230955,
      "epoch": 1.8702953966111862,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004654176745900365,
      "loss": 5.9056,
      "mean_token_accuracy": 0.1601640149950981,
      "num_tokens": 36594318.0,
      "step": 18710
    },
    {
      "entropy": 6.0966023921966555,
      "epoch": 1.8707952216724146,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046539866041908233,
      "loss": 5.7878,
      "mean_token_accuracy": 0.16099971234798433,
      "num_tokens": 36605377.0,
      "step": 18715
    },
    {
      "entropy": 6.020800924301147,
      "epoch": 1.871295046733643,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00046537964145774687,
      "loss": 5.6576,
      "mean_token_accuracy": 0.17989901602268218,
      "num_tokens": 36614860.0,
      "step": 18720
    },
    {
      "entropy": 5.979335927963257,
      "epoch": 1.8717948717948718,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004653606177065086,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17474114447832106,
      "num_tokens": 36625968.0,
      "step": 18725
    },
    {
      "entropy": 6.091188430786133,
      "epoch": 1.8722946968561005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004653415891658462,
      "loss": 5.7971,
      "mean_token_accuracy": 0.16955651640892028,
      "num_tokens": 36633966.0,
      "step": 18730
    },
    {
      "entropy": 6.109600305557251,
      "epoch": 1.872794521917329,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046532255583623837,
      "loss": 5.8318,
      "mean_token_accuracy": 0.1661555290222168,
      "num_tokens": 36644409.0,
      "step": 18735
    },
    {
      "entropy": 6.132808971405029,
      "epoch": 1.8732943469785575,
      "grad_norm": 1.25,
      "learning_rate": 0.000465303517718164,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16500858664512635,
      "num_tokens": 36653660.0,
      "step": 18740
    },
    {
      "entropy": 6.1374023914337155,
      "epoch": 1.8737941720397862,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004652844748121021,
      "loss": 5.7576,
      "mean_token_accuracy": 0.16945238262414933,
      "num_tokens": 36663590.0,
      "step": 18745
    },
    {
      "entropy": 6.170210313796997,
      "epoch": 1.8742939971010146,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004652654271185318,
      "loss": 5.8434,
      "mean_token_accuracy": 0.16191455721855164,
      "num_tokens": 36672448.0,
      "step": 18750
    },
    {
      "entropy": 6.165058946609497,
      "epoch": 1.874793822162243,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046524637463793233,
      "loss": 5.93,
      "mean_token_accuracy": 0.15798409730196,
      "num_tokens": 36682901.0,
      "step": 18755
    },
    {
      "entropy": 6.082245397567749,
      "epoch": 1.8752936472234718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000465227317370783,
      "loss": 5.8167,
      "mean_token_accuracy": 0.1602613940834999,
      "num_tokens": 36693768.0,
      "step": 18760
    },
    {
      "entropy": 6.143332433700562,
      "epoch": 1.8757934722847005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046520825531756326,
      "loss": 5.8954,
      "mean_token_accuracy": 0.1596700966358185,
      "num_tokens": 36704005.0,
      "step": 18765
    },
    {
      "entropy": 6.141286325454712,
      "epoch": 1.876293297345929,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004651891884787528,
      "loss": 5.8058,
      "mean_token_accuracy": 0.16798323690891265,
      "num_tokens": 36712966.0,
      "step": 18770
    },
    {
      "entropy": 6.076968336105347,
      "epoch": 1.8767931224071575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046517011685483136,
      "loss": 5.7453,
      "mean_token_accuracy": 0.17516446560621263,
      "num_tokens": 36722160.0,
      "step": 18775
    },
    {
      "entropy": 6.075450372695923,
      "epoch": 1.8772929474683862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046515104044627867,
      "loss": 5.7643,
      "mean_token_accuracy": 0.16639539152383803,
      "num_tokens": 36731810.0,
      "step": 18780
    },
    {
      "entropy": 6.081890392303467,
      "epoch": 1.8777927725296146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004651319592535747,
      "loss": 5.8414,
      "mean_token_accuracy": 0.16477157026529313,
      "num_tokens": 36741772.0,
      "step": 18785
    },
    {
      "entropy": 6.145393896102905,
      "epoch": 1.878292597590843,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004651128732771996,
      "loss": 5.8646,
      "mean_token_accuracy": 0.1559132695198059,
      "num_tokens": 36752060.0,
      "step": 18790
    },
    {
      "entropy": 6.138278198242188,
      "epoch": 1.8787924226520718,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046509378251763346,
      "loss": 5.8082,
      "mean_token_accuracy": 0.16159359365701675,
      "num_tokens": 36760590.0,
      "step": 18795
    },
    {
      "entropy": 6.063232851028443,
      "epoch": 1.8792922477133005,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004650746869753567,
      "loss": 5.7245,
      "mean_token_accuracy": 0.16989997029304504,
      "num_tokens": 36769628.0,
      "step": 18800
    },
    {
      "entropy": 6.05721344947815,
      "epoch": 1.8797920727745288,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004650555866508498,
      "loss": 5.669,
      "mean_token_accuracy": 0.17731261849403382,
      "num_tokens": 36779498.0,
      "step": 18805
    },
    {
      "entropy": 6.07973690032959,
      "epoch": 1.8802918978357575,
      "grad_norm": 2.03125,
      "learning_rate": 0.00046503648154459315,
      "loss": 5.8128,
      "mean_token_accuracy": 0.162276990711689,
      "num_tokens": 36789789.0,
      "step": 18810
    },
    {
      "entropy": 6.083994960784912,
      "epoch": 1.8807917228969862,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004650173716570676,
      "loss": 5.714,
      "mean_token_accuracy": 0.16590510457754135,
      "num_tokens": 36800316.0,
      "step": 18815
    },
    {
      "entropy": 6.0476569652557375,
      "epoch": 1.8812915479582146,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00046499825698875384,
      "loss": 5.7946,
      "mean_token_accuracy": 0.16673074215650557,
      "num_tokens": 36810811.0,
      "step": 18820
    },
    {
      "entropy": 6.072043371200562,
      "epoch": 1.881791373019443,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046497913754013284,
      "loss": 5.7642,
      "mean_token_accuracy": 0.17145866602659227,
      "num_tokens": 36820439.0,
      "step": 18825
    },
    {
      "entropy": 6.1045128345489506,
      "epoch": 1.8822911980806718,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004649600133116856,
      "loss": 5.7428,
      "mean_token_accuracy": 0.16236790716648103,
      "num_tokens": 36830363.0,
      "step": 18830
    },
    {
      "entropy": 6.100769424438477,
      "epoch": 1.8827910231419005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046494088430389335,
      "loss": 5.7542,
      "mean_token_accuracy": 0.16570524722337723,
      "num_tokens": 36840854.0,
      "step": 18835
    },
    {
      "entropy": 6.066996002197266,
      "epoch": 1.8832908482031288,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004649217505172373,
      "loss": 5.7258,
      "mean_token_accuracy": 0.16406189352273942,
      "num_tokens": 36850719.0,
      "step": 18840
    },
    {
      "entropy": 6.018286323547363,
      "epoch": 1.8837906732643575,
      "grad_norm": 1.375,
      "learning_rate": 0.0004649026119521989,
      "loss": 5.8228,
      "mean_token_accuracy": 0.16590381264686585,
      "num_tokens": 36861074.0,
      "step": 18845
    },
    {
      "entropy": 6.085642242431641,
      "epoch": 1.8842904983255861,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046488346860925955,
      "loss": 5.8195,
      "mean_token_accuracy": 0.16124351620674132,
      "num_tokens": 36870432.0,
      "step": 18850
    },
    {
      "entropy": 6.129970216751099,
      "epoch": 1.8847903233868146,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046486432048890104,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16405318975448607,
      "num_tokens": 36880213.0,
      "step": 18855
    },
    {
      "entropy": 6.121780824661255,
      "epoch": 1.885290148448043,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004648451675916051,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1654884248971939,
      "num_tokens": 36889948.0,
      "step": 18860
    },
    {
      "entropy": 6.1047993183135985,
      "epoch": 1.8857899735092718,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046482600991785365,
      "loss": 5.8618,
      "mean_token_accuracy": 0.16027091443538666,
      "num_tokens": 36900211.0,
      "step": 18865
    },
    {
      "entropy": 6.083430671691895,
      "epoch": 1.8862897985705003,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046480684746812856,
      "loss": 5.8244,
      "mean_token_accuracy": 0.16933130472898483,
      "num_tokens": 36909515.0,
      "step": 18870
    },
    {
      "entropy": 6.055294561386108,
      "epoch": 1.8867896236317288,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046478768024291205,
      "loss": 5.7446,
      "mean_token_accuracy": 0.17418822646141052,
      "num_tokens": 36918057.0,
      "step": 18875
    },
    {
      "entropy": 6.138399410247803,
      "epoch": 1.8872894486929574,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046476850824268633,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16996460855007173,
      "num_tokens": 36926960.0,
      "step": 18880
    },
    {
      "entropy": 6.0994080066680905,
      "epoch": 1.8877892737541861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004647493314679338,
      "loss": 5.8377,
      "mean_token_accuracy": 0.16818828731775284,
      "num_tokens": 36936542.0,
      "step": 18885
    },
    {
      "entropy": 6.019711971282959,
      "epoch": 1.8882890988154146,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000464730149919137,
      "loss": 5.668,
      "mean_token_accuracy": 0.18008733689785003,
      "num_tokens": 36946621.0,
      "step": 18890
    },
    {
      "entropy": 6.105749654769897,
      "epoch": 1.888788923876643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004647109635967783,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16602389216423036,
      "num_tokens": 36955813.0,
      "step": 18895
    },
    {
      "entropy": 6.039545583724975,
      "epoch": 1.8892887489378718,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004646917725013406,
      "loss": 5.7798,
      "mean_token_accuracy": 0.17009483724832536,
      "num_tokens": 36965924.0,
      "step": 18900
    },
    {
      "entropy": 6.088921642303466,
      "epoch": 1.8897885739991003,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004646725766333068,
      "loss": 5.8306,
      "mean_token_accuracy": 0.1621186301112175,
      "num_tokens": 36975471.0,
      "step": 18905
    },
    {
      "entropy": 6.045242071151733,
      "epoch": 1.8902883990603288,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046465337599315966,
      "loss": 5.7024,
      "mean_token_accuracy": 0.16765774488449098,
      "num_tokens": 36985979.0,
      "step": 18910
    },
    {
      "entropy": 6.106415510177612,
      "epoch": 1.8907882241215574,
      "grad_norm": 1.25,
      "learning_rate": 0.00046463417058138247,
      "loss": 5.9148,
      "mean_token_accuracy": 0.16436421796679496,
      "num_tokens": 36996357.0,
      "step": 18915
    },
    {
      "entropy": 6.08010573387146,
      "epoch": 1.8912880491827861,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004646149603984583,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1689757898449898,
      "num_tokens": 37006193.0,
      "step": 18920
    },
    {
      "entropy": 6.165759181976318,
      "epoch": 1.8917878742440146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004645957454448706,
      "loss": 5.9043,
      "mean_token_accuracy": 0.16149502694606782,
      "num_tokens": 37015887.0,
      "step": 18925
    },
    {
      "entropy": 6.231616258621216,
      "epoch": 1.892287699305243,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004645765257211026,
      "loss": 5.8868,
      "mean_token_accuracy": 0.1559996023774147,
      "num_tokens": 37026869.0,
      "step": 18930
    },
    {
      "entropy": 6.148327875137329,
      "epoch": 1.8927875243664718,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046455730122763804,
      "loss": 5.8515,
      "mean_token_accuracy": 0.16367475241422652,
      "num_tokens": 37036170.0,
      "step": 18935
    },
    {
      "entropy": 6.087929582595825,
      "epoch": 1.8932873494277003,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004645380719649606,
      "loss": 5.8045,
      "mean_token_accuracy": 0.16373928785324096,
      "num_tokens": 37046843.0,
      "step": 18940
    },
    {
      "entropy": 6.100049543380737,
      "epoch": 1.8937871744889287,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046451883793355406,
      "loss": 5.8071,
      "mean_token_accuracy": 0.1635363757610321,
      "num_tokens": 37056948.0,
      "step": 18945
    },
    {
      "entropy": 6.0254580020904545,
      "epoch": 1.8942869995501574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046449959913390227,
      "loss": 5.7103,
      "mean_token_accuracy": 0.172881580889225,
      "num_tokens": 37066600.0,
      "step": 18950
    },
    {
      "entropy": 6.05014934539795,
      "epoch": 1.8947868246113861,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046448035556648936,
      "loss": 5.7814,
      "mean_token_accuracy": 0.1708945393562317,
      "num_tokens": 37076192.0,
      "step": 18955
    },
    {
      "entropy": 6.066793298721313,
      "epoch": 1.8952866496726146,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046446110723179945,
      "loss": 5.7983,
      "mean_token_accuracy": 0.17070139348506927,
      "num_tokens": 37087403.0,
      "step": 18960
    },
    {
      "entropy": 6.168452262878418,
      "epoch": 1.895786474733843,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004644418541303169,
      "loss": 5.9129,
      "mean_token_accuracy": 0.16137207448482513,
      "num_tokens": 37096643.0,
      "step": 18965
    },
    {
      "entropy": 6.130008888244629,
      "epoch": 1.8962862997950718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046442259626252593,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16828189343214034,
      "num_tokens": 37106088.0,
      "step": 18970
    },
    {
      "entropy": 6.089364528656006,
      "epoch": 1.8967861248563003,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046440333362891123,
      "loss": 5.7863,
      "mean_token_accuracy": 0.16744329482316972,
      "num_tokens": 37115830.0,
      "step": 18975
    },
    {
      "entropy": 6.057965230941773,
      "epoch": 1.8972859499175287,
      "grad_norm": 1.25,
      "learning_rate": 0.0004643840662299574,
      "loss": 5.7961,
      "mean_token_accuracy": 0.16563354283571244,
      "num_tokens": 37125311.0,
      "step": 18980
    },
    {
      "entropy": 6.003949594497681,
      "epoch": 1.8977857749787574,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004643647940661492,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17461594492197036,
      "num_tokens": 37135716.0,
      "step": 18985
    },
    {
      "entropy": 6.068088150024414,
      "epoch": 1.8982856000399861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046434551713797146,
      "loss": 5.6967,
      "mean_token_accuracy": 0.17249745875597,
      "num_tokens": 37145112.0,
      "step": 18990
    },
    {
      "entropy": 6.0893199920654295,
      "epoch": 1.8987854251012146,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004643262354459093,
      "loss": 5.8433,
      "mean_token_accuracy": 0.15569010972976685,
      "num_tokens": 37155081.0,
      "step": 18995
    },
    {
      "entropy": 5.998398733139038,
      "epoch": 1.899285250162443,
      "grad_norm": 1.375,
      "learning_rate": 0.00046430694899044774,
      "loss": 5.7775,
      "mean_token_accuracy": 0.17246189713478088,
      "num_tokens": 37165633.0,
      "step": 19000
    },
    {
      "entropy": 6.164998483657837,
      "epoch": 1.8997850752236718,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000464287657772072,
      "loss": 5.7436,
      "mean_token_accuracy": 0.16390874311327935,
      "num_tokens": 37175842.0,
      "step": 19005
    },
    {
      "entropy": 6.15986442565918,
      "epoch": 1.9002849002849003,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004642683617912676,
      "loss": 5.8823,
      "mean_token_accuracy": 0.15291849672794341,
      "num_tokens": 37185076.0,
      "step": 19010
    },
    {
      "entropy": 6.055150270462036,
      "epoch": 1.9007847253461287,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046424906104851994,
      "loss": 5.7695,
      "mean_token_accuracy": 0.16537159979343413,
      "num_tokens": 37195088.0,
      "step": 19015
    },
    {
      "entropy": 6.100834369659424,
      "epoch": 1.9012845504073574,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046422975554431453,
      "loss": 5.826,
      "mean_token_accuracy": 0.15980107188224793,
      "num_tokens": 37205082.0,
      "step": 19020
    },
    {
      "entropy": 6.115382242202759,
      "epoch": 1.9017843754685861,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046421044527913713,
      "loss": 5.7539,
      "mean_token_accuracy": 0.16508985161781312,
      "num_tokens": 37215097.0,
      "step": 19025
    },
    {
      "entropy": 6.126351165771484,
      "epoch": 1.9022842005298146,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004641911302534737,
      "loss": 5.8459,
      "mean_token_accuracy": 0.16221281588077546,
      "num_tokens": 37225461.0,
      "step": 19030
    },
    {
      "entropy": 6.121295595169068,
      "epoch": 1.902784025591043,
      "grad_norm": 1.5625,
      "learning_rate": 0.00046417181046781004,
      "loss": 5.9059,
      "mean_token_accuracy": 0.16110976934432983,
      "num_tokens": 37235033.0,
      "step": 19035
    },
    {
      "entropy": 6.127628803253174,
      "epoch": 1.9032838506522718,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004641524859226324,
      "loss": 5.8602,
      "mean_token_accuracy": 0.16038018018007277,
      "num_tokens": 37245407.0,
      "step": 19040
    },
    {
      "entropy": 6.215808725357055,
      "epoch": 1.9037836757135003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046413315661842676,
      "loss": 5.9476,
      "mean_token_accuracy": 0.15960542112588882,
      "num_tokens": 37255654.0,
      "step": 19045
    },
    {
      "entropy": 6.1914585590362545,
      "epoch": 1.9042835007747287,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004641138225556797,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16599395275115966,
      "num_tokens": 37264461.0,
      "step": 19050
    },
    {
      "entropy": 6.053776788711548,
      "epoch": 1.9047833258359574,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004640944837348774,
      "loss": 5.707,
      "mean_token_accuracy": 0.17361035645008088,
      "num_tokens": 37273893.0,
      "step": 19055
    },
    {
      "entropy": 6.143212366104126,
      "epoch": 1.9052831508971861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046407514015650667,
      "loss": 5.8783,
      "mean_token_accuracy": 0.15502255111932756,
      "num_tokens": 37282784.0,
      "step": 19060
    },
    {
      "entropy": 6.119091606140136,
      "epoch": 1.9057829759584144,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00046405579182105393,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1612707942724228,
      "num_tokens": 37291990.0,
      "step": 19065
    },
    {
      "entropy": 6.076426649093628,
      "epoch": 1.906282801019643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004640364387290062,
      "loss": 5.7371,
      "mean_token_accuracy": 0.17252199053764344,
      "num_tokens": 37302451.0,
      "step": 19070
    },
    {
      "entropy": 6.135573434829712,
      "epoch": 1.9067826260808718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046401708088085036,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16603320389986037,
      "num_tokens": 37313035.0,
      "step": 19075
    },
    {
      "entropy": 6.087472057342529,
      "epoch": 1.9072824511421003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046399771827707335,
      "loss": 5.7843,
      "mean_token_accuracy": 0.1663082480430603,
      "num_tokens": 37323107.0,
      "step": 19080
    },
    {
      "entropy": 6.04610595703125,
      "epoch": 1.9077822762033287,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046397835091816235,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16630543172359466,
      "num_tokens": 37332180.0,
      "step": 19085
    },
    {
      "entropy": 5.987257480621338,
      "epoch": 1.9082821012645574,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046395897880460474,
      "loss": 5.7313,
      "mean_token_accuracy": 0.18263283967971802,
      "num_tokens": 37341588.0,
      "step": 19090
    },
    {
      "entropy": 6.017257118225098,
      "epoch": 1.9087819263257861,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046393960193688783,
      "loss": 5.7042,
      "mean_token_accuracy": 0.17169519811868666,
      "num_tokens": 37350969.0,
      "step": 19095
    },
    {
      "entropy": 6.1935478210449215,
      "epoch": 1.9092817513870144,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046392022031549907,
      "loss": 5.8897,
      "mean_token_accuracy": 0.15908896923065186,
      "num_tokens": 37360387.0,
      "step": 19100
    },
    {
      "entropy": 6.107531642913818,
      "epoch": 1.909781576448243,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004639008339409263,
      "loss": 5.8211,
      "mean_token_accuracy": 0.17266913652420043,
      "num_tokens": 37369443.0,
      "step": 19105
    },
    {
      "entropy": 6.076848793029785,
      "epoch": 1.9102814015094718,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004638814428136571,
      "loss": 5.7458,
      "mean_token_accuracy": 0.17046762704849244,
      "num_tokens": 37379111.0,
      "step": 19110
    },
    {
      "entropy": 6.086488914489746,
      "epoch": 1.9107812265707003,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004638620469341794,
      "loss": 5.7992,
      "mean_token_accuracy": 0.17039063572883606,
      "num_tokens": 37389880.0,
      "step": 19115
    },
    {
      "entropy": 6.016513681411743,
      "epoch": 1.9112810516319287,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004638426463029812,
      "loss": 5.6868,
      "mean_token_accuracy": 0.17062409669160844,
      "num_tokens": 37400055.0,
      "step": 19120
    },
    {
      "entropy": 6.119580459594727,
      "epoch": 1.9117808766931574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004638232409205506,
      "loss": 5.8469,
      "mean_token_accuracy": 0.16507965475320815,
      "num_tokens": 37409223.0,
      "step": 19125
    },
    {
      "entropy": 6.115202617645264,
      "epoch": 1.912280701754386,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046380383078737584,
      "loss": 5.871,
      "mean_token_accuracy": 0.1657530203461647,
      "num_tokens": 37418480.0,
      "step": 19130
    },
    {
      "entropy": 6.0909405708312985,
      "epoch": 1.9127805268156144,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046378441590394525,
      "loss": 5.793,
      "mean_token_accuracy": 0.16599923074245454,
      "num_tokens": 37428578.0,
      "step": 19135
    },
    {
      "entropy": 6.1003023147583,
      "epoch": 1.913280351876843,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004637649962707474,
      "loss": 5.7613,
      "mean_token_accuracy": 0.16395787447690963,
      "num_tokens": 37438847.0,
      "step": 19140
    },
    {
      "entropy": 6.076535224914551,
      "epoch": 1.9137801769380718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046374557188827084,
      "loss": 5.7721,
      "mean_token_accuracy": 0.1665279433131218,
      "num_tokens": 37448369.0,
      "step": 19145
    },
    {
      "entropy": 6.118758058547973,
      "epoch": 1.9142800019993003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004637261427570042,
      "loss": 5.9069,
      "mean_token_accuracy": 0.16308291107416154,
      "num_tokens": 37457899.0,
      "step": 19150
    },
    {
      "entropy": 6.106628894805908,
      "epoch": 1.9147798270605287,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004637067088774364,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16470449268817902,
      "num_tokens": 37468429.0,
      "step": 19155
    },
    {
      "entropy": 6.165749883651733,
      "epoch": 1.9152796521217574,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046368727025005634,
      "loss": 5.9154,
      "mean_token_accuracy": 0.16133702397346497,
      "num_tokens": 37479009.0,
      "step": 19160
    },
    {
      "entropy": 6.079428243637085,
      "epoch": 1.915779477182986,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046366782687535315,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16454349011182784,
      "num_tokens": 37488114.0,
      "step": 19165
    },
    {
      "entropy": 6.10204005241394,
      "epoch": 1.9162793022442144,
      "grad_norm": 1.140625,
      "learning_rate": 0.000463648378753816,
      "loss": 5.7445,
      "mean_token_accuracy": 0.16577766835689545,
      "num_tokens": 37497477.0,
      "step": 19170
    },
    {
      "entropy": 6.140926933288574,
      "epoch": 1.916779127305443,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046362892588593406,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16205693483352662,
      "num_tokens": 37507126.0,
      "step": 19175
    },
    {
      "entropy": 6.152692699432373,
      "epoch": 1.9172789523666718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046360946827219703,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15926390439271926,
      "num_tokens": 37517625.0,
      "step": 19180
    },
    {
      "entropy": 6.162769365310669,
      "epoch": 1.9177787774279003,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004635900059130942,
      "loss": 5.9203,
      "mean_token_accuracy": 0.1567814901471138,
      "num_tokens": 37528761.0,
      "step": 19185
    },
    {
      "entropy": 6.138395690917969,
      "epoch": 1.9182786024891287,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004635705388091154,
      "loss": 5.822,
      "mean_token_accuracy": 0.1686330258846283,
      "num_tokens": 37538513.0,
      "step": 19190
    },
    {
      "entropy": 6.029062032699585,
      "epoch": 1.9187784275503574,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004635510669607504,
      "loss": 5.8142,
      "mean_token_accuracy": 0.16950635910034179,
      "num_tokens": 37548266.0,
      "step": 19195
    },
    {
      "entropy": 6.074082565307617,
      "epoch": 1.919278252611586,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000463531590368489,
      "loss": 5.747,
      "mean_token_accuracy": 0.17121636867523193,
      "num_tokens": 37558544.0,
      "step": 19200
    },
    {
      "entropy": 6.148809289932251,
      "epoch": 1.9197780776728144,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046351210903282136,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16527968347072602,
      "num_tokens": 37568430.0,
      "step": 19205
    },
    {
      "entropy": 6.123152542114258,
      "epoch": 1.920277902734043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046349262295423754,
      "loss": 5.851,
      "mean_token_accuracy": 0.15221864134073257,
      "num_tokens": 37579286.0,
      "step": 19210
    },
    {
      "entropy": 6.133230495452881,
      "epoch": 1.9207777277952718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046347313213322786,
      "loss": 5.8142,
      "mean_token_accuracy": 0.16417718231678008,
      "num_tokens": 37590142.0,
      "step": 19215
    },
    {
      "entropy": 6.14308443069458,
      "epoch": 1.9212775528565003,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004634536365702826,
      "loss": 5.973,
      "mean_token_accuracy": 0.15055600106716155,
      "num_tokens": 37600498.0,
      "step": 19220
    },
    {
      "entropy": 6.142151880264282,
      "epoch": 1.9217773779177287,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046343413626589243,
      "loss": 5.8692,
      "mean_token_accuracy": 0.15776242762804032,
      "num_tokens": 37610530.0,
      "step": 19225
    },
    {
      "entropy": 6.091331243515015,
      "epoch": 1.9222772029789574,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004634146312205478,
      "loss": 5.7643,
      "mean_token_accuracy": 0.17067279517650605,
      "num_tokens": 37621267.0,
      "step": 19230
    },
    {
      "entropy": 6.189578819274902,
      "epoch": 1.922777028040186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046339512143473956,
      "loss": 5.9128,
      "mean_token_accuracy": 0.15583074390888213,
      "num_tokens": 37632609.0,
      "step": 19235
    },
    {
      "entropy": 6.112420511245728,
      "epoch": 1.9232768531014144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046337560690895856,
      "loss": 5.71,
      "mean_token_accuracy": 0.17219986766576767,
      "num_tokens": 37641757.0,
      "step": 19240
    },
    {
      "entropy": 6.165593767166138,
      "epoch": 1.923776678162643,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046335608764369574,
      "loss": 5.8771,
      "mean_token_accuracy": 0.16389907151460648,
      "num_tokens": 37651021.0,
      "step": 19245
    },
    {
      "entropy": 6.113535594940186,
      "epoch": 1.9242765032238718,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046333656363944226,
      "loss": 5.8101,
      "mean_token_accuracy": 0.16303881406784057,
      "num_tokens": 37659965.0,
      "step": 19250
    },
    {
      "entropy": 6.055449438095093,
      "epoch": 1.9247763282851003,
      "grad_norm": 1.53125,
      "learning_rate": 0.00046331703489668925,
      "loss": 5.7909,
      "mean_token_accuracy": 0.1794048950076103,
      "num_tokens": 37669957.0,
      "step": 19255
    },
    {
      "entropy": 6.111278390884399,
      "epoch": 1.9252761533463287,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046329750141592814,
      "loss": 5.7166,
      "mean_token_accuracy": 0.1727800652384758,
      "num_tokens": 37678585.0,
      "step": 19260
    },
    {
      "entropy": 6.142039346694946,
      "epoch": 1.9257759784075574,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004632779631976503,
      "loss": 5.8168,
      "mean_token_accuracy": 0.16633886396884917,
      "num_tokens": 37688315.0,
      "step": 19265
    },
    {
      "entropy": 6.180815696716309,
      "epoch": 1.926275803468786,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046325842024234735,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16133116632699968,
      "num_tokens": 37699042.0,
      "step": 19270
    },
    {
      "entropy": 6.149508476257324,
      "epoch": 1.9267756285300144,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046323887255051105,
      "loss": 5.7871,
      "mean_token_accuracy": 0.16957712471485137,
      "num_tokens": 37709607.0,
      "step": 19275
    },
    {
      "entropy": 6.114201927185059,
      "epoch": 1.927275453591243,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046321932012263317,
      "loss": 5.7559,
      "mean_token_accuracy": 0.16329443827271461,
      "num_tokens": 37718893.0,
      "step": 19280
    },
    {
      "entropy": 6.114528703689575,
      "epoch": 1.9277752786524718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004631997629592055,
      "loss": 5.8105,
      "mean_token_accuracy": 0.16510997712612152,
      "num_tokens": 37730786.0,
      "step": 19285
    },
    {
      "entropy": 6.1808020114898685,
      "epoch": 1.9282751037137003,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004631802010607204,
      "loss": 5.8816,
      "mean_token_accuracy": 0.15401194021105766,
      "num_tokens": 37740801.0,
      "step": 19290
    },
    {
      "entropy": 6.094652509689331,
      "epoch": 1.9287749287749287,
      "grad_norm": 1.25,
      "learning_rate": 0.00046316063442766975,
      "loss": 5.7067,
      "mean_token_accuracy": 0.17165134102106094,
      "num_tokens": 37750580.0,
      "step": 19295
    },
    {
      "entropy": 5.985230159759522,
      "epoch": 1.9292747538361574,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046314106306054596,
      "loss": 5.6574,
      "mean_token_accuracy": 0.1811885342001915,
      "num_tokens": 37759915.0,
      "step": 19300
    },
    {
      "entropy": 6.222445917129517,
      "epoch": 1.929774578897386,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046312148695984147,
      "loss": 5.9419,
      "mean_token_accuracy": 0.15551216676831245,
      "num_tokens": 37770648.0,
      "step": 19305
    },
    {
      "entropy": 6.192369604110718,
      "epoch": 1.9302744039586144,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004631019061260488,
      "loss": 5.686,
      "mean_token_accuracy": 0.17453116327524185,
      "num_tokens": 37779612.0,
      "step": 19310
    },
    {
      "entropy": 6.111704397201538,
      "epoch": 1.930774229019843,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004630823205596605,
      "loss": 5.795,
      "mean_token_accuracy": 0.16676497161388398,
      "num_tokens": 37789813.0,
      "step": 19315
    },
    {
      "entropy": 6.060680866241455,
      "epoch": 1.9312740540810718,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046306273026116944,
      "loss": 5.8029,
      "mean_token_accuracy": 0.16438739001750946,
      "num_tokens": 37799706.0,
      "step": 19320
    },
    {
      "entropy": 6.126711797714234,
      "epoch": 1.9317738791423,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004630431352310685,
      "loss": 5.8651,
      "mean_token_accuracy": 0.15766213089227676,
      "num_tokens": 37809787.0,
      "step": 19325
    },
    {
      "entropy": 6.119903135299682,
      "epoch": 1.9322737042035287,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046302353546985066,
      "loss": 5.8309,
      "mean_token_accuracy": 0.16458355486392975,
      "num_tokens": 37821091.0,
      "step": 19330
    },
    {
      "entropy": 6.04904613494873,
      "epoch": 1.9327735292647574,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000463003930978009,
      "loss": 5.7336,
      "mean_token_accuracy": 0.16500821858644485,
      "num_tokens": 37830467.0,
      "step": 19335
    },
    {
      "entropy": 6.099285697937011,
      "epoch": 1.933273354325986,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004629843217560369,
      "loss": 5.7802,
      "mean_token_accuracy": 0.16258333772420883,
      "num_tokens": 37839666.0,
      "step": 19340
    },
    {
      "entropy": 6.164001703262329,
      "epoch": 1.9337731793872144,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004629647078044276,
      "loss": 5.884,
      "mean_token_accuracy": 0.15902386903762816,
      "num_tokens": 37848255.0,
      "step": 19345
    },
    {
      "entropy": 5.949913787841797,
      "epoch": 1.934273004448443,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004629450891236746,
      "loss": 5.6389,
      "mean_token_accuracy": 0.17550608962774278,
      "num_tokens": 37858077.0,
      "step": 19350
    },
    {
      "entropy": 6.060276889801026,
      "epoch": 1.9347728295096718,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004629254657142715,
      "loss": 5.8568,
      "mean_token_accuracy": 0.16377055048942565,
      "num_tokens": 37868027.0,
      "step": 19355
    },
    {
      "entropy": 6.131686115264893,
      "epoch": 1.9352726545709,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004629058375767121,
      "loss": 5.808,
      "mean_token_accuracy": 0.1689050540328026,
      "num_tokens": 37877952.0,
      "step": 19360
    },
    {
      "entropy": 6.087370347976685,
      "epoch": 1.9357724796321287,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046288620471149014,
      "loss": 5.8243,
      "mean_token_accuracy": 0.16281457394361495,
      "num_tokens": 37888873.0,
      "step": 19365
    },
    {
      "entropy": 6.087195682525635,
      "epoch": 1.9362723046933574,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046286656711909966,
      "loss": 5.6994,
      "mean_token_accuracy": 0.1678590290248394,
      "num_tokens": 37898360.0,
      "step": 19370
    },
    {
      "entropy": 5.969558191299439,
      "epoch": 1.936772129754586,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046284692480003467,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16496194005012513,
      "num_tokens": 37908620.0,
      "step": 19375
    },
    {
      "entropy": 6.0906483173370365,
      "epoch": 1.9372719548158144,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046282727775478935,
      "loss": 5.8798,
      "mean_token_accuracy": 0.1643528625369072,
      "num_tokens": 37917939.0,
      "step": 19380
    },
    {
      "entropy": 6.178954458236694,
      "epoch": 1.937771779877043,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004628076259838581,
      "loss": 5.8167,
      "mean_token_accuracy": 0.165118907392025,
      "num_tokens": 37928044.0,
      "step": 19385
    },
    {
      "entropy": 6.179335975646973,
      "epoch": 1.9382716049382716,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046278796948773525,
      "loss": 5.764,
      "mean_token_accuracy": 0.16966381967067717,
      "num_tokens": 37937853.0,
      "step": 19390
    },
    {
      "entropy": 6.047819995880127,
      "epoch": 1.9387714299995,
      "grad_norm": 1.546875,
      "learning_rate": 0.00046276830826691553,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16177310198545455,
      "num_tokens": 37946949.0,
      "step": 19395
    },
    {
      "entropy": 6.11584095954895,
      "epoch": 1.9392712550607287,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046274864232189334,
      "loss": 5.8639,
      "mean_token_accuracy": 0.15673984959721565,
      "num_tokens": 37958038.0,
      "step": 19400
    },
    {
      "entropy": 6.202079391479492,
      "epoch": 1.9397710801219574,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046272897165316367,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1682674318552017,
      "num_tokens": 37968751.0,
      "step": 19405
    },
    {
      "entropy": 6.042039155960083,
      "epoch": 1.940270905183186,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004627092962612215,
      "loss": 5.7706,
      "mean_token_accuracy": 0.16504312455654144,
      "num_tokens": 37979758.0,
      "step": 19410
    },
    {
      "entropy": 6.121612977981568,
      "epoch": 1.9407707302444144,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004626896161465616,
      "loss": 5.8559,
      "mean_token_accuracy": 0.15896472036838533,
      "num_tokens": 37989776.0,
      "step": 19415
    },
    {
      "entropy": 6.0955650329589846,
      "epoch": 1.941270555305643,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046266993130967925,
      "loss": 5.7352,
      "mean_token_accuracy": 0.17091960608959197,
      "num_tokens": 37999715.0,
      "step": 19420
    },
    {
      "entropy": 6.07746729850769,
      "epoch": 1.9417703803668716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004626502417510698,
      "loss": 5.81,
      "mean_token_accuracy": 0.1665532574057579,
      "num_tokens": 38010161.0,
      "step": 19425
    },
    {
      "entropy": 6.100387287139893,
      "epoch": 1.9422702054281,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046263054747122857,
      "loss": 5.7815,
      "mean_token_accuracy": 0.17227751165628433,
      "num_tokens": 38019569.0,
      "step": 19430
    },
    {
      "entropy": 6.147130250930786,
      "epoch": 1.9427700304893287,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004626108484706509,
      "loss": 5.8259,
      "mean_token_accuracy": 0.1630483403801918,
      "num_tokens": 38029366.0,
      "step": 19435
    },
    {
      "entropy": 6.057442998886108,
      "epoch": 1.9432698555505574,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00046259114474983266,
      "loss": 5.7879,
      "mean_token_accuracy": 0.1677574560046196,
      "num_tokens": 38038606.0,
      "step": 19440
    },
    {
      "entropy": 6.017407703399658,
      "epoch": 1.943769680611786,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046257143630926953,
      "loss": 5.6695,
      "mean_token_accuracy": 0.1749446615576744,
      "num_tokens": 38049045.0,
      "step": 19445
    },
    {
      "entropy": 6.161724948883057,
      "epoch": 1.9442695056730144,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004625517231494573,
      "loss": 5.8927,
      "mean_token_accuracy": 0.15795782655477525,
      "num_tokens": 38059052.0,
      "step": 19450
    },
    {
      "entropy": 6.102239942550659,
      "epoch": 1.944769330734243,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004625320052708919,
      "loss": 5.6901,
      "mean_token_accuracy": 0.17087386846542357,
      "num_tokens": 38069316.0,
      "step": 19455
    },
    {
      "entropy": 6.074679231643676,
      "epoch": 1.9452691557954715,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004625122826740696,
      "loss": 5.8284,
      "mean_token_accuracy": 0.1669756904244423,
      "num_tokens": 38079175.0,
      "step": 19460
    },
    {
      "entropy": 6.048158788681031,
      "epoch": 1.9457689808567,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004624925553594865,
      "loss": 5.7541,
      "mean_token_accuracy": 0.17112686783075332,
      "num_tokens": 38089701.0,
      "step": 19465
    },
    {
      "entropy": 6.197945785522461,
      "epoch": 1.9462688059179287,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004624728233276389,
      "loss": 5.8864,
      "mean_token_accuracy": 0.1583539441227913,
      "num_tokens": 38099948.0,
      "step": 19470
    },
    {
      "entropy": 6.111060285568238,
      "epoch": 1.9467686309791574,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046245308657902334,
      "loss": 5.8297,
      "mean_token_accuracy": 0.16897072941064833,
      "num_tokens": 38110195.0,
      "step": 19475
    },
    {
      "entropy": 6.048681974411011,
      "epoch": 1.947268456040386,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004624333451141364,
      "loss": 5.8468,
      "mean_token_accuracy": 0.161198927462101,
      "num_tokens": 38121027.0,
      "step": 19480
    },
    {
      "entropy": 6.078368520736694,
      "epoch": 1.9477682811016144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004624135989334746,
      "loss": 5.7488,
      "mean_token_accuracy": 0.1663047194480896,
      "num_tokens": 38130684.0,
      "step": 19485
    },
    {
      "entropy": 6.105405521392822,
      "epoch": 1.948268106162843,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046239384803753496,
      "loss": 5.6837,
      "mean_token_accuracy": 0.1738941803574562,
      "num_tokens": 38140376.0,
      "step": 19490
    },
    {
      "entropy": 6.157326412200928,
      "epoch": 1.9487679312240715,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046237409242681435,
      "loss": 5.9337,
      "mean_token_accuracy": 0.15576133504509926,
      "num_tokens": 38150687.0,
      "step": 19495
    },
    {
      "entropy": 6.09609432220459,
      "epoch": 1.9492677562853,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004623543321018097,
      "loss": 5.753,
      "mean_token_accuracy": 0.16447756439447403,
      "num_tokens": 38160853.0,
      "step": 19500
    },
    {
      "entropy": 6.033198118209839,
      "epoch": 1.9497675813465287,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004623345670630184,
      "loss": 5.7802,
      "mean_token_accuracy": 0.16720011234283447,
      "num_tokens": 38170702.0,
      "step": 19505
    },
    {
      "entropy": 6.078425359725952,
      "epoch": 1.9502674064077574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004623147973109375,
      "loss": 5.7958,
      "mean_token_accuracy": 0.16505419611930847,
      "num_tokens": 38179425.0,
      "step": 19510
    },
    {
      "entropy": 6.055085945129394,
      "epoch": 1.950767231468986,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046229502284606455,
      "loss": 5.8178,
      "mean_token_accuracy": 0.1672055020928383,
      "num_tokens": 38189700.0,
      "step": 19515
    },
    {
      "entropy": 6.118195486068726,
      "epoch": 1.9512670565302144,
      "grad_norm": 1.28125,
      "learning_rate": 0.000462275243668897,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16186968684196473,
      "num_tokens": 38198944.0,
      "step": 19520
    },
    {
      "entropy": 6.076929330825806,
      "epoch": 1.951766881591443,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046225545977993245,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1682885095477104,
      "num_tokens": 38207237.0,
      "step": 19525
    },
    {
      "entropy": 6.0861433982849125,
      "epoch": 1.9522667066526715,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004622356711796688,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16367159336805343,
      "num_tokens": 38217234.0,
      "step": 19530
    },
    {
      "entropy": 6.017566680908203,
      "epoch": 1.9527665317139,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004622158778686038,
      "loss": 5.7068,
      "mean_token_accuracy": 0.17454992681741716,
      "num_tokens": 38226463.0,
      "step": 19535
    },
    {
      "entropy": 5.973503637313843,
      "epoch": 1.9532663567751287,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046219607984723554,
      "loss": 5.732,
      "mean_token_accuracy": 0.17158081531524658,
      "num_tokens": 38235313.0,
      "step": 19540
    },
    {
      "entropy": 6.150286722183227,
      "epoch": 1.9537661818363574,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046217627711606205,
      "loss": 5.7391,
      "mean_token_accuracy": 0.16661752983927727,
      "num_tokens": 38243769.0,
      "step": 19545
    },
    {
      "entropy": 6.22131028175354,
      "epoch": 1.954266006897586,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046215646967558155,
      "loss": 5.8946,
      "mean_token_accuracy": 0.1631706953048706,
      "num_tokens": 38253522.0,
      "step": 19550
    },
    {
      "entropy": 6.070188379287719,
      "epoch": 1.9547658319588144,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046213665752629246,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16333727687597274,
      "num_tokens": 38263816.0,
      "step": 19555
    },
    {
      "entropy": 6.009294176101685,
      "epoch": 1.955265657020043,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004621168406686933,
      "loss": 5.7414,
      "mean_token_accuracy": 0.16696014553308486,
      "num_tokens": 38273139.0,
      "step": 19560
    },
    {
      "entropy": 6.050578355789185,
      "epoch": 1.9557654820812715,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046209701910328245,
      "loss": 5.7817,
      "mean_token_accuracy": 0.16455430835485457,
      "num_tokens": 38282355.0,
      "step": 19565
    },
    {
      "entropy": 6.08901424407959,
      "epoch": 1.9562653071425,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004620771928305588,
      "loss": 5.8046,
      "mean_token_accuracy": 0.1629269912838936,
      "num_tokens": 38291140.0,
      "step": 19570
    },
    {
      "entropy": 6.014429521560669,
      "epoch": 1.9567651322037287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004620573618510211,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17879654169082643,
      "num_tokens": 38300117.0,
      "step": 19575
    },
    {
      "entropy": 6.063425731658936,
      "epoch": 1.9572649572649574,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046203752616516834,
      "loss": 5.7656,
      "mean_token_accuracy": 0.17328575849533082,
      "num_tokens": 38309978.0,
      "step": 19580
    },
    {
      "entropy": 6.017241334915161,
      "epoch": 1.9577647823261857,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046201768577349957,
      "loss": 5.7967,
      "mean_token_accuracy": 0.1748356819152832,
      "num_tokens": 38318875.0,
      "step": 19585
    },
    {
      "entropy": 6.146378993988037,
      "epoch": 1.9582646073874144,
      "grad_norm": 1.375,
      "learning_rate": 0.0004619978406765139,
      "loss": 5.9255,
      "mean_token_accuracy": 0.15983956903219224,
      "num_tokens": 38328354.0,
      "step": 19590
    },
    {
      "entropy": 6.17619047164917,
      "epoch": 1.958764432448643,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004619779908747107,
      "loss": 5.8688,
      "mean_token_accuracy": 0.16250703036785125,
      "num_tokens": 38338196.0,
      "step": 19595
    },
    {
      "entropy": 6.09071569442749,
      "epoch": 1.9592642575098715,
      "grad_norm": 1.25,
      "learning_rate": 0.0004619581363685893,
      "loss": 5.8123,
      "mean_token_accuracy": 0.16221605390310287,
      "num_tokens": 38349077.0,
      "step": 19600
    },
    {
      "entropy": 6.10459303855896,
      "epoch": 1.9597640825711,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004619382771586494,
      "loss": 5.7966,
      "mean_token_accuracy": 0.1718171402812004,
      "num_tokens": 38359507.0,
      "step": 19605
    },
    {
      "entropy": 6.235325860977173,
      "epoch": 1.9602639076323287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004619184132453905,
      "loss": 5.9731,
      "mean_token_accuracy": 0.15831194892525674,
      "num_tokens": 38370452.0,
      "step": 19610
    },
    {
      "entropy": 6.140395164489746,
      "epoch": 1.9607637326935574,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004618985446293125,
      "loss": 5.7648,
      "mean_token_accuracy": 0.17209935188293457,
      "num_tokens": 38379734.0,
      "step": 19615
    },
    {
      "entropy": 5.969383716583252,
      "epoch": 1.9612635577547857,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046187867131091504,
      "loss": 5.6141,
      "mean_token_accuracy": 0.179729688167572,
      "num_tokens": 38388589.0,
      "step": 19620
    },
    {
      "entropy": 6.078347396850586,
      "epoch": 1.9617633828160144,
      "grad_norm": 1.25,
      "learning_rate": 0.0004618587932906984,
      "loss": 5.8452,
      "mean_token_accuracy": 0.15785517990589143,
      "num_tokens": 38399158.0,
      "step": 19625
    },
    {
      "entropy": 6.117124891281128,
      "epoch": 1.962263207877243,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046183891056916256,
      "loss": 5.768,
      "mean_token_accuracy": 0.1698819249868393,
      "num_tokens": 38409254.0,
      "step": 19630
    },
    {
      "entropy": 6.120018529891968,
      "epoch": 1.9627630329384715,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004618190231468079,
      "loss": 5.8331,
      "mean_token_accuracy": 0.1654137924313545,
      "num_tokens": 38419308.0,
      "step": 19635
    },
    {
      "entropy": 6.028248643875122,
      "epoch": 1.9632628579997,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046179913102413456,
      "loss": 5.773,
      "mean_token_accuracy": 0.1722830727696419,
      "num_tokens": 38428245.0,
      "step": 19640
    },
    {
      "entropy": 6.077307271957397,
      "epoch": 1.9637626830609287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004617792342016433,
      "loss": 5.8595,
      "mean_token_accuracy": 0.1617307558655739,
      "num_tokens": 38438675.0,
      "step": 19645
    },
    {
      "entropy": 6.157096433639526,
      "epoch": 1.9642625081221572,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004617593326798344,
      "loss": 5.7549,
      "mean_token_accuracy": 0.1709567502140999,
      "num_tokens": 38447592.0,
      "step": 19650
    },
    {
      "entropy": 6.081485366821289,
      "epoch": 1.9647623331833857,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004617394264592089,
      "loss": 5.7857,
      "mean_token_accuracy": 0.17066720724105836,
      "num_tokens": 38456290.0,
      "step": 19655
    },
    {
      "entropy": 6.038445472717285,
      "epoch": 1.9652621582446144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004617195155402674,
      "loss": 5.7282,
      "mean_token_accuracy": 0.16724506318569182,
      "num_tokens": 38465995.0,
      "step": 19660
    },
    {
      "entropy": 6.188421297073364,
      "epoch": 1.965761983305843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046169959992351103,
      "loss": 5.8787,
      "mean_token_accuracy": 0.15713273584842682,
      "num_tokens": 38475181.0,
      "step": 19665
    },
    {
      "entropy": 6.168809795379639,
      "epoch": 1.9662618083670715,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046167967960944066,
      "loss": 5.7866,
      "mean_token_accuracy": 0.16444614082574843,
      "num_tokens": 38484325.0,
      "step": 19670
    },
    {
      "entropy": 6.173360967636109,
      "epoch": 1.9667616334283,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004616597545985577,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16015665382146835,
      "num_tokens": 38492908.0,
      "step": 19675
    },
    {
      "entropy": 6.120298910140991,
      "epoch": 1.9672614584895287,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004616398248913632,
      "loss": 5.8333,
      "mean_token_accuracy": 0.1688578650355339,
      "num_tokens": 38501295.0,
      "step": 19680
    },
    {
      "entropy": 6.1254740238189695,
      "epoch": 1.9677612835507572,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046161989048835886,
      "loss": 5.792,
      "mean_token_accuracy": 0.16358178555965425,
      "num_tokens": 38511267.0,
      "step": 19685
    },
    {
      "entropy": 6.134437561035156,
      "epoch": 1.9682611086119857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046159995139004594,
      "loss": 5.7883,
      "mean_token_accuracy": 0.16685799807310103,
      "num_tokens": 38521476.0,
      "step": 19690
    },
    {
      "entropy": 6.141875314712524,
      "epoch": 1.9687609336732144,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004615800075969264,
      "loss": 5.6833,
      "mean_token_accuracy": 0.17721893936395644,
      "num_tokens": 38531731.0,
      "step": 19695
    },
    {
      "entropy": 6.04005389213562,
      "epoch": 1.969260758734443,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046156005910950187,
      "loss": 5.7913,
      "mean_token_accuracy": 0.1641255125403404,
      "num_tokens": 38541428.0,
      "step": 19700
    },
    {
      "entropy": 6.099810791015625,
      "epoch": 1.9697605837956715,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004615401059282742,
      "loss": 5.9068,
      "mean_token_accuracy": 0.1579722449183464,
      "num_tokens": 38550947.0,
      "step": 19705
    },
    {
      "entropy": 6.168578720092773,
      "epoch": 1.9702604088569,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046152014805374546,
      "loss": 5.8469,
      "mean_token_accuracy": 0.1636666998267174,
      "num_tokens": 38561090.0,
      "step": 19710
    },
    {
      "entropy": 6.1886249542236325,
      "epoch": 1.9707602339181287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004615001854864179,
      "loss": 5.853,
      "mean_token_accuracy": 0.1597514718770981,
      "num_tokens": 38570555.0,
      "step": 19715
    },
    {
      "entropy": 6.146050548553466,
      "epoch": 1.9712600589793572,
      "grad_norm": 1.25,
      "learning_rate": 0.00046148021822679347,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1696525737643242,
      "num_tokens": 38579392.0,
      "step": 19720
    },
    {
      "entropy": 6.048257493972779,
      "epoch": 1.9717598840405857,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004614602462753749,
      "loss": 5.7439,
      "mean_token_accuracy": 0.17161306142807006,
      "num_tokens": 38590352.0,
      "step": 19725
    },
    {
      "entropy": 6.10000958442688,
      "epoch": 1.9722597091018144,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004614402696326645,
      "loss": 5.802,
      "mean_token_accuracy": 0.1631348252296448,
      "num_tokens": 38600261.0,
      "step": 19730
    },
    {
      "entropy": 6.095134592056274,
      "epoch": 1.972759534163043,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046142028829916485,
      "loss": 5.7924,
      "mean_token_accuracy": 0.16561865210533142,
      "num_tokens": 38610409.0,
      "step": 19735
    },
    {
      "entropy": 6.141270351409912,
      "epoch": 1.9732593592242715,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004614003022753787,
      "loss": 5.8858,
      "mean_token_accuracy": 0.1619822420179844,
      "num_tokens": 38619737.0,
      "step": 19740
    },
    {
      "entropy": 6.132047653198242,
      "epoch": 1.9737591842855,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004613803115618089,
      "loss": 5.8074,
      "mean_token_accuracy": 0.17108708769083023,
      "num_tokens": 38628994.0,
      "step": 19745
    },
    {
      "entropy": 6.015794658660889,
      "epoch": 1.9742590093467287,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046136031615895846,
      "loss": 5.7004,
      "mean_token_accuracy": 0.1778621032834053,
      "num_tokens": 38638598.0,
      "step": 19750
    },
    {
      "entropy": 6.134431838989258,
      "epoch": 1.9747588344079572,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004613403160673305,
      "loss": 5.8213,
      "mean_token_accuracy": 0.16457957923412322,
      "num_tokens": 38649956.0,
      "step": 19755
    },
    {
      "entropy": 6.062316274642944,
      "epoch": 1.9752586594691857,
      "grad_norm": 1.296875,
      "learning_rate": 0.000461320311287428,
      "loss": 5.813,
      "mean_token_accuracy": 0.17163700014352798,
      "num_tokens": 38660006.0,
      "step": 19760
    },
    {
      "entropy": 6.108672189712524,
      "epoch": 1.9757584845304144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046130030181975446,
      "loss": 5.8449,
      "mean_token_accuracy": 0.16314135044813155,
      "num_tokens": 38669714.0,
      "step": 19765
    },
    {
      "entropy": 6.117444944381714,
      "epoch": 1.976258309591643,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004612802876648133,
      "loss": 5.8702,
      "mean_token_accuracy": 0.16691718101501465,
      "num_tokens": 38679766.0,
      "step": 19770
    },
    {
      "entropy": 6.07814211845398,
      "epoch": 1.9767581346528715,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000461260268823108,
      "loss": 5.773,
      "mean_token_accuracy": 0.16750557571649552,
      "num_tokens": 38689440.0,
      "step": 19775
    },
    {
      "entropy": 6.131039476394653,
      "epoch": 1.9772579597141,
      "grad_norm": 1.25,
      "learning_rate": 0.00046124024529514224,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16737780570983887,
      "num_tokens": 38700105.0,
      "step": 19780
    },
    {
      "entropy": 6.115006017684936,
      "epoch": 1.9777577847753287,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004612202170814198,
      "loss": 5.8084,
      "mean_token_accuracy": 0.1633569583296776,
      "num_tokens": 38709876.0,
      "step": 19785
    },
    {
      "entropy": 6.1560221195220945,
      "epoch": 1.9782576098365572,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004612001841824447,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16209092885255813,
      "num_tokens": 38718857.0,
      "step": 19790
    },
    {
      "entropy": 6.081400537490845,
      "epoch": 1.9787574348977857,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046118014659872085,
      "loss": 5.7688,
      "mean_token_accuracy": 0.1753775179386139,
      "num_tokens": 38729353.0,
      "step": 19795
    },
    {
      "entropy": 6.073758268356324,
      "epoch": 1.9792572599590144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004611601043307525,
      "loss": 5.7341,
      "mean_token_accuracy": 0.17395384907722472,
      "num_tokens": 38738641.0,
      "step": 19800
    },
    {
      "entropy": 6.063201427459717,
      "epoch": 1.979757085020243,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046114005737904376,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16959316283464432,
      "num_tokens": 38748423.0,
      "step": 19805
    },
    {
      "entropy": 6.081057739257813,
      "epoch": 1.9802569100814715,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000461120005744099,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16737391650676728,
      "num_tokens": 38758063.0,
      "step": 19810
    },
    {
      "entropy": 6.121038389205933,
      "epoch": 1.9807567351427,
      "grad_norm": 1.78125,
      "learning_rate": 0.00046109994942642287,
      "loss": 5.7913,
      "mean_token_accuracy": 0.16610306352376938,
      "num_tokens": 38766769.0,
      "step": 19815
    },
    {
      "entropy": 6.062558269500732,
      "epoch": 1.9812565602039287,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004610798884265199,
      "loss": 5.7036,
      "mean_token_accuracy": 0.17706155478954316,
      "num_tokens": 38775665.0,
      "step": 19820
    },
    {
      "entropy": 6.053650379180908,
      "epoch": 1.9817563852651572,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046105982274489483,
      "loss": 5.6601,
      "mean_token_accuracy": 0.17734412252902984,
      "num_tokens": 38784044.0,
      "step": 19825
    },
    {
      "entropy": 6.108375024795532,
      "epoch": 1.9822562103263857,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046103975238205244,
      "loss": 5.8016,
      "mean_token_accuracy": 0.17127639949321746,
      "num_tokens": 38793666.0,
      "step": 19830
    },
    {
      "entropy": 6.139939451217652,
      "epoch": 1.9827560353876144,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004610196773384978,
      "loss": 5.9073,
      "mean_token_accuracy": 0.15617815852165223,
      "num_tokens": 38804193.0,
      "step": 19835
    },
    {
      "entropy": 6.194441795349121,
      "epoch": 1.983255860448843,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046099959761473593,
      "loss": 5.8439,
      "mean_token_accuracy": 0.1614262968301773,
      "num_tokens": 38814200.0,
      "step": 19840
    },
    {
      "entropy": 6.205484199523926,
      "epoch": 1.9837556855100713,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046097951321127207,
      "loss": 5.8358,
      "mean_token_accuracy": 0.17049632519483565,
      "num_tokens": 38824390.0,
      "step": 19845
    },
    {
      "entropy": 5.9859381198883055,
      "epoch": 1.9842555105713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046095942412861145,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17730267345905304,
      "num_tokens": 38832733.0,
      "step": 19850
    },
    {
      "entropy": 6.036586332321167,
      "epoch": 1.9847553356325287,
      "grad_norm": 1.25,
      "learning_rate": 0.0004609393303672596,
      "loss": 5.8106,
      "mean_token_accuracy": 0.16726166307926177,
      "num_tokens": 38843368.0,
      "step": 19855
    },
    {
      "entropy": 6.036651945114135,
      "epoch": 1.9852551606937572,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046091923192772205,
      "loss": 5.6854,
      "mean_token_accuracy": 0.1794489085674286,
      "num_tokens": 38854066.0,
      "step": 19860
    },
    {
      "entropy": 6.1720459938049315,
      "epoch": 1.9857549857549857,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004608991288105045,
      "loss": 5.8399,
      "mean_token_accuracy": 0.16449529826641082,
      "num_tokens": 38864509.0,
      "step": 19865
    },
    {
      "entropy": 6.133470773696899,
      "epoch": 1.9862548108162144,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046087902101611265,
      "loss": 5.7646,
      "mean_token_accuracy": 0.17320798337459564,
      "num_tokens": 38874287.0,
      "step": 19870
    },
    {
      "entropy": 5.98052487373352,
      "epoch": 1.986754635877443,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004608589085450525,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17588917016983033,
      "num_tokens": 38884517.0,
      "step": 19875
    },
    {
      "entropy": 6.020698356628418,
      "epoch": 1.9872544609386713,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046083879139783,
      "loss": 5.6792,
      "mean_token_accuracy": 0.17020216584205627,
      "num_tokens": 38894440.0,
      "step": 19880
    },
    {
      "entropy": 6.030180549621582,
      "epoch": 1.9877542859999,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004608186695749514,
      "loss": 5.8451,
      "mean_token_accuracy": 0.165393728017807,
      "num_tokens": 38904500.0,
      "step": 19885
    },
    {
      "entropy": 6.1271350383758545,
      "epoch": 1.9882541110611287,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004607985430769228,
      "loss": 5.79,
      "mean_token_accuracy": 0.17110151201486587,
      "num_tokens": 38913695.0,
      "step": 19890
    },
    {
      "entropy": 6.05939679145813,
      "epoch": 1.9887539361223572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004607784119042507,
      "loss": 5.769,
      "mean_token_accuracy": 0.16264808773994446,
      "num_tokens": 38923071.0,
      "step": 19895
    },
    {
      "entropy": 6.010162019729615,
      "epoch": 1.9892537611835857,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004607582760574416,
      "loss": 5.7038,
      "mean_token_accuracy": 0.17400213032960893,
      "num_tokens": 38932663.0,
      "step": 19900
    },
    {
      "entropy": 6.067906475067138,
      "epoch": 1.9897535862448144,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000460738135537002,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16375896483659744,
      "num_tokens": 38941230.0,
      "step": 19905
    },
    {
      "entropy": 6.104377031326294,
      "epoch": 1.9902534113060428,
      "grad_norm": 1.421875,
      "learning_rate": 0.00046071799034343874,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17179404348134994,
      "num_tokens": 38951004.0,
      "step": 19910
    },
    {
      "entropy": 6.0490540027618405,
      "epoch": 1.9907532363672713,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004606978404772587,
      "loss": 5.7591,
      "mean_token_accuracy": 0.1676931858062744,
      "num_tokens": 38960440.0,
      "step": 19915
    },
    {
      "entropy": 6.031106281280517,
      "epoch": 1.9912530614285,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046067768593896865,
      "loss": 5.7326,
      "mean_token_accuracy": 0.17511798739433287,
      "num_tokens": 38970012.0,
      "step": 19920
    },
    {
      "entropy": 6.097012567520141,
      "epoch": 1.9917528864897287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004606575267290759,
      "loss": 5.8451,
      "mean_token_accuracy": 0.16589672416448592,
      "num_tokens": 38979842.0,
      "step": 19925
    },
    {
      "entropy": 5.889960289001465,
      "epoch": 1.9922527115509572,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004606373628480875,
      "loss": 5.5716,
      "mean_token_accuracy": 0.19813066720962524,
      "num_tokens": 38990123.0,
      "step": 19930
    },
    {
      "entropy": 6.019259691238403,
      "epoch": 1.9927525366121857,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004606171942965108,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16927750408649445,
      "num_tokens": 39000500.0,
      "step": 19935
    },
    {
      "entropy": 6.1255899429321286,
      "epoch": 1.9932523616734144,
      "grad_norm": 1.375,
      "learning_rate": 0.0004605970210748533,
      "loss": 5.8237,
      "mean_token_accuracy": 0.1658168539404869,
      "num_tokens": 39009795.0,
      "step": 19940
    },
    {
      "entropy": 6.142306137084961,
      "epoch": 1.9937521867346428,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004605768431836225,
      "loss": 5.7462,
      "mean_token_accuracy": 0.172055684030056,
      "num_tokens": 39018969.0,
      "step": 19945
    },
    {
      "entropy": 6.038141536712646,
      "epoch": 1.9942520117958713,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046055666062332605,
      "loss": 5.657,
      "mean_token_accuracy": 0.16788831874728202,
      "num_tokens": 39028167.0,
      "step": 19950
    },
    {
      "entropy": 5.992218112945556,
      "epoch": 1.9947518368571,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046053647339447184,
      "loss": 5.6916,
      "mean_token_accuracy": 0.17341987043619156,
      "num_tokens": 39037513.0,
      "step": 19955
    },
    {
      "entropy": 6.0559595108032225,
      "epoch": 1.9952516619183287,
      "grad_norm": 1.65625,
      "learning_rate": 0.00046051628149756767,
      "loss": 5.7655,
      "mean_token_accuracy": 0.1699365884065628,
      "num_tokens": 39046258.0,
      "step": 19960
    },
    {
      "entropy": 6.0115156173706055,
      "epoch": 1.9957514869795572,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004604960849331216,
      "loss": 5.6618,
      "mean_token_accuracy": 0.18070581704378127,
      "num_tokens": 39056778.0,
      "step": 19965
    },
    {
      "entropy": 6.125102710723877,
      "epoch": 1.9962513120407857,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004604758837016417,
      "loss": 5.6803,
      "mean_token_accuracy": 0.1712339535355568,
      "num_tokens": 39066091.0,
      "step": 19970
    },
    {
      "entropy": 6.077715730667114,
      "epoch": 1.9967511371020144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004604556778036363,
      "loss": 5.774,
      "mean_token_accuracy": 0.16924739480018616,
      "num_tokens": 39075769.0,
      "step": 19975
    },
    {
      "entropy": 6.107685470581055,
      "epoch": 1.9972509621632428,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046043546723961384,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16607929021120071,
      "num_tokens": 39085690.0,
      "step": 19980
    },
    {
      "entropy": 6.037949705123902,
      "epoch": 1.9977507872244713,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046041525201008266,
      "loss": 5.7057,
      "mean_token_accuracy": 0.1746404156088829,
      "num_tokens": 39094143.0,
      "step": 19985
    },
    {
      "entropy": 6.061005115509033,
      "epoch": 1.9982506122857,
      "grad_norm": 1.25,
      "learning_rate": 0.00046039503211555154,
      "loss": 5.8406,
      "mean_token_accuracy": 0.15932462811470033,
      "num_tokens": 39104980.0,
      "step": 19990
    },
    {
      "entropy": 6.14730486869812,
      "epoch": 1.9987504373469287,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046037480755652913,
      "loss": 5.8644,
      "mean_token_accuracy": 0.1639137953519821,
      "num_tokens": 39115093.0,
      "step": 19995
    },
    {
      "entropy": 6.193211650848388,
      "epoch": 1.9992502624081572,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046035457833352414,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16385732740163803,
      "num_tokens": 39124482.0,
      "step": 20000
    },
    {
      "entropy": 6.104484176635742,
      "epoch": 1.9997500874693857,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046033434444704577,
      "loss": 5.7599,
      "mean_token_accuracy": 0.1624110609292984,
      "num_tokens": 39135029.0,
      "step": 20005
    },
    {
      "entropy": 6.065491676330566,
      "epoch": 2.0001999300244915,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046031410589760294,
      "loss": 5.6926,
      "mean_token_accuracy": 0.1740923755698734,
      "num_tokens": 39142785.0,
      "step": 20010
    },
    {
      "entropy": 6.091444444656372,
      "epoch": 2.00069975508572,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046029386268570485,
      "loss": 5.7245,
      "mean_token_accuracy": 0.16244571506977082,
      "num_tokens": 39151917.0,
      "step": 20015
    },
    {
      "entropy": 6.0638532638549805,
      "epoch": 2.0011995801469484,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004602736148118609,
      "loss": 5.6032,
      "mean_token_accuracy": 0.18724473267793657,
      "num_tokens": 39161782.0,
      "step": 20020
    },
    {
      "entropy": 6.078358268737793,
      "epoch": 2.001699405208177,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046025336227658047,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17937309741973878,
      "num_tokens": 39170954.0,
      "step": 20025
    },
    {
      "entropy": 6.04412579536438,
      "epoch": 2.002199230269406,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046023310508037316,
      "loss": 5.6365,
      "mean_token_accuracy": 0.17893040627241136,
      "num_tokens": 39179405.0,
      "step": 20030
    },
    {
      "entropy": 6.04565315246582,
      "epoch": 2.002699055330634,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046021284322374847,
      "loss": 5.7142,
      "mean_token_accuracy": 0.16846811175346374,
      "num_tokens": 39188837.0,
      "step": 20035
    },
    {
      "entropy": 6.069019746780396,
      "epoch": 2.003198880391863,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004601925767072164,
      "loss": 5.71,
      "mean_token_accuracy": 0.1754305437207222,
      "num_tokens": 39198562.0,
      "step": 20040
    },
    {
      "entropy": 6.110895395278931,
      "epoch": 2.0036987054530915,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046017230553128675,
      "loss": 5.664,
      "mean_token_accuracy": 0.16881361305713655,
      "num_tokens": 39207503.0,
      "step": 20045
    },
    {
      "entropy": 6.134699153900146,
      "epoch": 2.00419853051432,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046015202969646953,
      "loss": 5.7078,
      "mean_token_accuracy": 0.17290228456258774,
      "num_tokens": 39216992.0,
      "step": 20050
    },
    {
      "entropy": 6.084650373458862,
      "epoch": 2.0046983555755484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004601317492032749,
      "loss": 5.7209,
      "mean_token_accuracy": 0.16952417641878129,
      "num_tokens": 39227202.0,
      "step": 20055
    },
    {
      "entropy": 6.014699220657349,
      "epoch": 2.005198180636777,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000460111464052213,
      "loss": 5.7353,
      "mean_token_accuracy": 0.16773567944765091,
      "num_tokens": 39236843.0,
      "step": 20060
    },
    {
      "entropy": 6.071777820587158,
      "epoch": 2.005698005698006,
      "grad_norm": 1.25,
      "learning_rate": 0.00046009117424379445,
      "loss": 5.6851,
      "mean_token_accuracy": 0.17643558233976364,
      "num_tokens": 39246162.0,
      "step": 20065
    },
    {
      "entropy": 6.183686351776123,
      "epoch": 2.006197830759234,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046007087977852947,
      "loss": 5.84,
      "mean_token_accuracy": 0.1641357958316803,
      "num_tokens": 39256295.0,
      "step": 20070
    },
    {
      "entropy": 6.190367364883423,
      "epoch": 2.0066976558204628,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046005058065692883,
      "loss": 5.75,
      "mean_token_accuracy": 0.17239010632038115,
      "num_tokens": 39266704.0,
      "step": 20075
    },
    {
      "entropy": 6.1328624248504635,
      "epoch": 2.0071974808816915,
      "grad_norm": 1.5,
      "learning_rate": 0.00046003027687950316,
      "loss": 5.7094,
      "mean_token_accuracy": 0.16430145800113677,
      "num_tokens": 39276308.0,
      "step": 20080
    },
    {
      "entropy": 6.0264410972595215,
      "epoch": 2.00769730594292,
      "grad_norm": 1.375,
      "learning_rate": 0.0004600099684467634,
      "loss": 5.691,
      "mean_token_accuracy": 0.1615176320075989,
      "num_tokens": 39287357.0,
      "step": 20085
    },
    {
      "entropy": 6.014863061904907,
      "epoch": 2.0081971310041484,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004599896553592204,
      "loss": 5.658,
      "mean_token_accuracy": 0.17799095958471298,
      "num_tokens": 39297209.0,
      "step": 20090
    },
    {
      "entropy": 6.096435117721557,
      "epoch": 2.008696956065377,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004599693376173853,
      "loss": 5.8036,
      "mean_token_accuracy": 0.15944433808326722,
      "num_tokens": 39305884.0,
      "step": 20095
    },
    {
      "entropy": 6.075083494186401,
      "epoch": 2.009196781126606,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004599490152217692,
      "loss": 5.7195,
      "mean_token_accuracy": 0.17086339145898818,
      "num_tokens": 39314761.0,
      "step": 20100
    },
    {
      "entropy": 6.151798677444458,
      "epoch": 2.009696606187834,
      "grad_norm": 1.5625,
      "learning_rate": 0.00045992868817288355,
      "loss": 5.7458,
      "mean_token_accuracy": 0.16921046525239944,
      "num_tokens": 39323752.0,
      "step": 20105
    },
    {
      "entropy": 6.1474110126495365,
      "epoch": 2.0101964312490628,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004599083564712396,
      "loss": 5.7581,
      "mean_token_accuracy": 0.171538308262825,
      "num_tokens": 39333296.0,
      "step": 20110
    },
    {
      "entropy": 6.125001430511475,
      "epoch": 2.0106962563102915,
      "grad_norm": 1.515625,
      "learning_rate": 0.00045988802011734896,
      "loss": 5.7961,
      "mean_token_accuracy": 0.16353830248117446,
      "num_tokens": 39343693.0,
      "step": 20115
    },
    {
      "entropy": 6.027267599105835,
      "epoch": 2.01119608137152,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045986767911172343,
      "loss": 5.6925,
      "mean_token_accuracy": 0.17170249670743942,
      "num_tokens": 39353443.0,
      "step": 20120
    },
    {
      "entropy": 6.09709620475769,
      "epoch": 2.0116959064327484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045984733345487465,
      "loss": 5.6598,
      "mean_token_accuracy": 0.17263263165950776,
      "num_tokens": 39363236.0,
      "step": 20125
    },
    {
      "entropy": 6.110104131698608,
      "epoch": 2.012195731493977,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045982698314731447,
      "loss": 5.6782,
      "mean_token_accuracy": 0.16893595308065415,
      "num_tokens": 39372764.0,
      "step": 20130
    },
    {
      "entropy": 6.155585098266601,
      "epoch": 2.012695556555206,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045980662818955495,
      "loss": 5.7386,
      "mean_token_accuracy": 0.1687980055809021,
      "num_tokens": 39383087.0,
      "step": 20135
    },
    {
      "entropy": 6.082065200805664,
      "epoch": 2.013195381616434,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045978626858210833,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17314461022615432,
      "num_tokens": 39392356.0,
      "step": 20140
    },
    {
      "entropy": 6.065162324905396,
      "epoch": 2.0136952066776628,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004597659043254866,
      "loss": 5.7619,
      "mean_token_accuracy": 0.16792158782482147,
      "num_tokens": 39401782.0,
      "step": 20145
    },
    {
      "entropy": 6.068939590454102,
      "epoch": 2.0141950317388915,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004597455354202023,
      "loss": 5.6757,
      "mean_token_accuracy": 0.1735281154513359,
      "num_tokens": 39410819.0,
      "step": 20150
    },
    {
      "entropy": 5.990709686279297,
      "epoch": 2.01469485680012,
      "grad_norm": 1.25,
      "learning_rate": 0.0004597251618667679,
      "loss": 5.6382,
      "mean_token_accuracy": 0.17810062021017076,
      "num_tokens": 39419902.0,
      "step": 20155
    },
    {
      "entropy": 6.1240647315979,
      "epoch": 2.0151946818613484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045970478366569594,
      "loss": 5.7599,
      "mean_token_accuracy": 0.16983793377876283,
      "num_tokens": 39429265.0,
      "step": 20160
    },
    {
      "entropy": 6.054523181915283,
      "epoch": 2.015694506922577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045968440081749917,
      "loss": 5.687,
      "mean_token_accuracy": 0.16839855164289474,
      "num_tokens": 39439515.0,
      "step": 20165
    },
    {
      "entropy": 6.040804672241211,
      "epoch": 2.016194331983806,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004596640133226904,
      "loss": 5.6802,
      "mean_token_accuracy": 0.17307647615671157,
      "num_tokens": 39448970.0,
      "step": 20170
    },
    {
      "entropy": 6.066859006881714,
      "epoch": 2.016694157045034,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045964362118178253,
      "loss": 5.6583,
      "mean_token_accuracy": 0.1729396551847458,
      "num_tokens": 39460359.0,
      "step": 20175
    },
    {
      "entropy": 6.1631019592285154,
      "epoch": 2.0171939821062628,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004596232243952887,
      "loss": 5.8381,
      "mean_token_accuracy": 0.1643498495221138,
      "num_tokens": 39470175.0,
      "step": 20180
    },
    {
      "entropy": 5.951327562332153,
      "epoch": 2.0176938071674915,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000459602822963722,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17904564440250398,
      "num_tokens": 39479834.0,
      "step": 20185
    },
    {
      "entropy": 6.038627195358276,
      "epoch": 2.01819363222872,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045958241688759583,
      "loss": 5.6765,
      "mean_token_accuracy": 0.16634062826633453,
      "num_tokens": 39489108.0,
      "step": 20190
    },
    {
      "entropy": 6.079608821868897,
      "epoch": 2.0186934572899484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045956200616742356,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17182571291923524,
      "num_tokens": 39499098.0,
      "step": 20195
    },
    {
      "entropy": 6.117280769348144,
      "epoch": 2.019193282351177,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004595415908037187,
      "loss": 5.7361,
      "mean_token_accuracy": 0.17202571928501129,
      "num_tokens": 39508470.0,
      "step": 20200
    },
    {
      "entropy": 6.165522766113281,
      "epoch": 2.019693107412406,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004595211707969948,
      "loss": 5.7735,
      "mean_token_accuracy": 0.16382265239953994,
      "num_tokens": 39518941.0,
      "step": 20205
    },
    {
      "entropy": 6.089139366149903,
      "epoch": 2.020192932473634,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004595007461477658,
      "loss": 5.6665,
      "mean_token_accuracy": 0.16793506890535354,
      "num_tokens": 39528613.0,
      "step": 20210
    },
    {
      "entropy": 6.066159868240357,
      "epoch": 2.0206927575348628,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045948031685654557,
      "loss": 5.6865,
      "mean_token_accuracy": 0.16720812320709227,
      "num_tokens": 39537831.0,
      "step": 20215
    },
    {
      "entropy": 6.01226840019226,
      "epoch": 2.0211925825960915,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004594598829238479,
      "loss": 5.6397,
      "mean_token_accuracy": 0.177620629966259,
      "num_tokens": 39547439.0,
      "step": 20220
    },
    {
      "entropy": 6.030704307556152,
      "epoch": 2.0216924076573197,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004594394443501872,
      "loss": 5.65,
      "mean_token_accuracy": 0.17353388667106628,
      "num_tokens": 39556898.0,
      "step": 20225
    },
    {
      "entropy": 6.1166980266571045,
      "epoch": 2.0221922327185484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004594190011360775,
      "loss": 5.6895,
      "mean_token_accuracy": 0.16907729655504228,
      "num_tokens": 39565882.0,
      "step": 20230
    },
    {
      "entropy": 6.071561193466186,
      "epoch": 2.022692057779777,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004593985532820331,
      "loss": 5.7774,
      "mean_token_accuracy": 0.1645402044057846,
      "num_tokens": 39575903.0,
      "step": 20235
    },
    {
      "entropy": 6.026433229446411,
      "epoch": 2.023191882841006,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004593781007885686,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17251231223344804,
      "num_tokens": 39586058.0,
      "step": 20240
    },
    {
      "entropy": 6.165940046310425,
      "epoch": 2.023691707902234,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004593576436561985,
      "loss": 5.7566,
      "mean_token_accuracy": 0.1674616739153862,
      "num_tokens": 39596241.0,
      "step": 20245
    },
    {
      "entropy": 6.1626345157623295,
      "epoch": 2.0241915329634628,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00045933718188543763,
      "loss": 5.814,
      "mean_token_accuracy": 0.16136470288038254,
      "num_tokens": 39605359.0,
      "step": 20250
    },
    {
      "entropy": 6.10009732246399,
      "epoch": 2.0246913580246915,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004593167154768006,
      "loss": 5.7161,
      "mean_token_accuracy": 0.17562412321567536,
      "num_tokens": 39615168.0,
      "step": 20255
    },
    {
      "entropy": 6.04750623703003,
      "epoch": 2.0251911830859197,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004592962444308024,
      "loss": 5.6454,
      "mean_token_accuracy": 0.1723933219909668,
      "num_tokens": 39624308.0,
      "step": 20260
    },
    {
      "entropy": 6.0584512710571286,
      "epoch": 2.0256910081471484,
      "grad_norm": 1.5,
      "learning_rate": 0.0004592757687479582,
      "loss": 5.7345,
      "mean_token_accuracy": 0.17411022633314133,
      "num_tokens": 39633422.0,
      "step": 20265
    },
    {
      "entropy": 6.056822013854981,
      "epoch": 2.026190833208377,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000459255288428783,
      "loss": 5.7049,
      "mean_token_accuracy": 0.17158133536577225,
      "num_tokens": 39642725.0,
      "step": 20270
    },
    {
      "entropy": 6.097681331634521,
      "epoch": 2.026690658269606,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004592348034737922,
      "loss": 5.7354,
      "mean_token_accuracy": 0.16793563216924667,
      "num_tokens": 39652974.0,
      "step": 20275
    },
    {
      "entropy": 6.138616752624512,
      "epoch": 2.027190483330834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004592143138835012,
      "loss": 5.8296,
      "mean_token_accuracy": 0.15870533138513565,
      "num_tokens": 39661947.0,
      "step": 20280
    },
    {
      "entropy": 6.03874659538269,
      "epoch": 2.0276903083920628,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004591938196584254,
      "loss": 5.5586,
      "mean_token_accuracy": 0.18923692256212235,
      "num_tokens": 39671179.0,
      "step": 20285
    },
    {
      "entropy": 6.079241037368774,
      "epoch": 2.0281901334532915,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045917332079908055,
      "loss": 5.7712,
      "mean_token_accuracy": 0.16783061623573303,
      "num_tokens": 39681874.0,
      "step": 20290
    },
    {
      "entropy": 6.103134441375732,
      "epoch": 2.0286899585145197,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004591528173059823,
      "loss": 5.7001,
      "mean_token_accuracy": 0.16710326224565505,
      "num_tokens": 39691351.0,
      "step": 20295
    },
    {
      "entropy": 6.121220111846924,
      "epoch": 2.0291897835757484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004591323091796465,
      "loss": 5.7179,
      "mean_token_accuracy": 0.17291091233491898,
      "num_tokens": 39702315.0,
      "step": 20300
    },
    {
      "entropy": 6.081080055236816,
      "epoch": 2.029689608636977,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004591117964205893,
      "loss": 5.7499,
      "mean_token_accuracy": 0.16749942004680635,
      "num_tokens": 39712048.0,
      "step": 20305
    },
    {
      "entropy": 6.037313079833984,
      "epoch": 2.030189433698206,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045909127902932657,
      "loss": 5.7474,
      "mean_token_accuracy": 0.1674064114689827,
      "num_tokens": 39722013.0,
      "step": 20310
    },
    {
      "entropy": 6.126332521438599,
      "epoch": 2.030689258759434,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004590707570063747,
      "loss": 5.707,
      "mean_token_accuracy": 0.16622736752033235,
      "num_tokens": 39730867.0,
      "step": 20315
    },
    {
      "entropy": 6.047085952758789,
      "epoch": 2.0311890838206628,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045905023035224993,
      "loss": 5.664,
      "mean_token_accuracy": 0.17227870225906372,
      "num_tokens": 39740451.0,
      "step": 20320
    },
    {
      "entropy": 6.003446340560913,
      "epoch": 2.0316889088818915,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045902969906746875,
      "loss": 5.6944,
      "mean_token_accuracy": 0.17919987440109253,
      "num_tokens": 39750580.0,
      "step": 20325
    },
    {
      "entropy": 6.02822527885437,
      "epoch": 2.0321887339431197,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045900916315254765,
      "loss": 5.5831,
      "mean_token_accuracy": 0.1791377618908882,
      "num_tokens": 39760943.0,
      "step": 20330
    },
    {
      "entropy": 6.077625894546509,
      "epoch": 2.0326885590043484,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045898862260800337,
      "loss": 5.6895,
      "mean_token_accuracy": 0.17175418585538865,
      "num_tokens": 39770686.0,
      "step": 20335
    },
    {
      "entropy": 6.094436407089233,
      "epoch": 2.033188384065577,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045896807743435263,
      "loss": 5.765,
      "mean_token_accuracy": 0.1644475981593132,
      "num_tokens": 39781121.0,
      "step": 20340
    },
    {
      "entropy": 6.1101771831512455,
      "epoch": 2.033688209126806,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004589475276321125,
      "loss": 5.6918,
      "mean_token_accuracy": 0.17267676293849946,
      "num_tokens": 39791284.0,
      "step": 20345
    },
    {
      "entropy": 6.046210193634034,
      "epoch": 2.034188034188034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045892697320179987,
      "loss": 5.6302,
      "mean_token_accuracy": 0.17318816035985946,
      "num_tokens": 39800475.0,
      "step": 20350
    },
    {
      "entropy": 6.054823541641236,
      "epoch": 2.0346878592492628,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004589064141439319,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16768928170204161,
      "num_tokens": 39809077.0,
      "step": 20355
    },
    {
      "entropy": 6.0096721172332765,
      "epoch": 2.0351876843104915,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004588858504590259,
      "loss": 5.6027,
      "mean_token_accuracy": 0.18204623013734816,
      "num_tokens": 39819201.0,
      "step": 20360
    },
    {
      "entropy": 6.047993135452271,
      "epoch": 2.0356875093717197,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004588652821475992,
      "loss": 5.6547,
      "mean_token_accuracy": 0.16438134014606476,
      "num_tokens": 39829067.0,
      "step": 20365
    },
    {
      "entropy": 6.060696458816528,
      "epoch": 2.0361873344329484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045884470921016935,
      "loss": 5.686,
      "mean_token_accuracy": 0.16664479672908783,
      "num_tokens": 39838766.0,
      "step": 20370
    },
    {
      "entropy": 6.045194673538208,
      "epoch": 2.036687159494177,
      "grad_norm": 1.5,
      "learning_rate": 0.00045882413164725395,
      "loss": 5.6351,
      "mean_token_accuracy": 0.17102732956409455,
      "num_tokens": 39849108.0,
      "step": 20375
    },
    {
      "entropy": 6.076483201980591,
      "epoch": 2.037186984555406,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045880354945937056,
      "loss": 5.699,
      "mean_token_accuracy": 0.17262165248394012,
      "num_tokens": 39857556.0,
      "step": 20380
    },
    {
      "entropy": 5.9948389530181885,
      "epoch": 2.037686809616634,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004587829626470373,
      "loss": 5.6222,
      "mean_token_accuracy": 0.1760129749774933,
      "num_tokens": 39868729.0,
      "step": 20385
    },
    {
      "entropy": 6.010599565505982,
      "epoch": 2.0381866346778628,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045876237121077196,
      "loss": 5.637,
      "mean_token_accuracy": 0.17419475466012954,
      "num_tokens": 39878463.0,
      "step": 20390
    },
    {
      "entropy": 6.032925081253052,
      "epoch": 2.0386864597390915,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004587417751510926,
      "loss": 5.6245,
      "mean_token_accuracy": 0.171415376663208,
      "num_tokens": 39888836.0,
      "step": 20395
    },
    {
      "entropy": 6.042307806015015,
      "epoch": 2.0391862848003197,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045872117446851753,
      "loss": 5.6632,
      "mean_token_accuracy": 0.17482268810272217,
      "num_tokens": 39899567.0,
      "step": 20400
    },
    {
      "entropy": 6.070378541946411,
      "epoch": 2.0396861098615484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004587005691635649,
      "loss": 5.6678,
      "mean_token_accuracy": 0.16796865612268447,
      "num_tokens": 39908324.0,
      "step": 20405
    },
    {
      "entropy": 6.109102869033814,
      "epoch": 2.040185934922777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045867995923675326,
      "loss": 5.7516,
      "mean_token_accuracy": 0.16485819071531296,
      "num_tokens": 39918725.0,
      "step": 20410
    },
    {
      "entropy": 6.103215074539184,
      "epoch": 2.040685759984006,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004586593446886011,
      "loss": 5.665,
      "mean_token_accuracy": 0.17185193449258804,
      "num_tokens": 39928317.0,
      "step": 20415
    },
    {
      "entropy": 6.032854080200195,
      "epoch": 2.041185585045234,
      "grad_norm": 1.25,
      "learning_rate": 0.0004586387255196271,
      "loss": 5.694,
      "mean_token_accuracy": 0.1778609871864319,
      "num_tokens": 39937314.0,
      "step": 20420
    },
    {
      "entropy": 6.045533466339111,
      "epoch": 2.0416854101064628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045861810173035,
      "loss": 5.7267,
      "mean_token_accuracy": 0.16540264934301377,
      "num_tokens": 39947323.0,
      "step": 20425
    },
    {
      "entropy": 6.108846378326416,
      "epoch": 2.0421852351676915,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045859747332128883,
      "loss": 5.7506,
      "mean_token_accuracy": 0.16234126091003417,
      "num_tokens": 39956982.0,
      "step": 20430
    },
    {
      "entropy": 6.108264350891114,
      "epoch": 2.0426850602289197,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004585768402929623,
      "loss": 5.7052,
      "mean_token_accuracy": 0.16726094037294387,
      "num_tokens": 39967619.0,
      "step": 20435
    },
    {
      "entropy": 6.032459878921509,
      "epoch": 2.0431848852901484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004585562026458898,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1703927993774414,
      "num_tokens": 39977916.0,
      "step": 20440
    },
    {
      "entropy": 5.989623403549194,
      "epoch": 2.043684710351377,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045853556038059044,
      "loss": 5.6764,
      "mean_token_accuracy": 0.17366916090250015,
      "num_tokens": 39986918.0,
      "step": 20445
    },
    {
      "entropy": 6.060161352157593,
      "epoch": 2.0441845354126054,
      "grad_norm": 1.375,
      "learning_rate": 0.00045851491349758366,
      "loss": 5.6568,
      "mean_token_accuracy": 0.16945516616106032,
      "num_tokens": 39995641.0,
      "step": 20450
    },
    {
      "entropy": 6.085879325866699,
      "epoch": 2.044684360473834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045849426199738884,
      "loss": 5.7161,
      "mean_token_accuracy": 0.16685705184936522,
      "num_tokens": 40005525.0,
      "step": 20455
    },
    {
      "entropy": 6.096784067153931,
      "epoch": 2.0451841855350628,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004584736058805256,
      "loss": 5.7834,
      "mean_token_accuracy": 0.1642579898238182,
      "num_tokens": 40014474.0,
      "step": 20460
    },
    {
      "entropy": 6.067488956451416,
      "epoch": 2.0456840105962915,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004584529451475137,
      "loss": 5.7543,
      "mean_token_accuracy": 0.16844117194414138,
      "num_tokens": 40024231.0,
      "step": 20465
    },
    {
      "entropy": 6.000402593612671,
      "epoch": 2.0461838356575197,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045843227979887286,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18282969146966935,
      "num_tokens": 40033947.0,
      "step": 20470
    },
    {
      "entropy": 6.10225477218628,
      "epoch": 2.0466836607187484,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045841160983512303,
      "loss": 5.7244,
      "mean_token_accuracy": 0.16556269228458403,
      "num_tokens": 40044800.0,
      "step": 20475
    },
    {
      "entropy": 6.041636037826538,
      "epoch": 2.047183485779977,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004583909352567843,
      "loss": 5.6108,
      "mean_token_accuracy": 0.1680411383509636,
      "num_tokens": 40053629.0,
      "step": 20480
    },
    {
      "entropy": 6.0296934127807615,
      "epoch": 2.0476833108412054,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004583702560643769,
      "loss": 5.7614,
      "mean_token_accuracy": 0.16502762585878372,
      "num_tokens": 40063014.0,
      "step": 20485
    },
    {
      "entropy": 6.074089050292969,
      "epoch": 2.048183135902434,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000458349572258421,
      "loss": 5.7291,
      "mean_token_accuracy": 0.16842673122882842,
      "num_tokens": 40072332.0,
      "step": 20490
    },
    {
      "entropy": 6.122220039367676,
      "epoch": 2.0486829609636628,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045832888383943706,
      "loss": 5.8245,
      "mean_token_accuracy": 0.16729456931352615,
      "num_tokens": 40083144.0,
      "step": 20495
    },
    {
      "entropy": 6.0720278263092045,
      "epoch": 2.0491827860248915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004583081908079455,
      "loss": 5.5854,
      "mean_token_accuracy": 0.18319520056247712,
      "num_tokens": 40092035.0,
      "step": 20500
    },
    {
      "entropy": 6.118153762817383,
      "epoch": 2.0496826110861197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004582874931644671,
      "loss": 5.7157,
      "mean_token_accuracy": 0.17185365110635759,
      "num_tokens": 40101518.0,
      "step": 20505
    },
    {
      "entropy": 6.109589290618897,
      "epoch": 2.0501824361473484,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045826679090952253,
      "loss": 5.7111,
      "mean_token_accuracy": 0.16958885192871093,
      "num_tokens": 40110856.0,
      "step": 20510
    },
    {
      "entropy": 6.023039817810059,
      "epoch": 2.050682261208577,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045824608404363267,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16256580948829652,
      "num_tokens": 40120977.0,
      "step": 20515
    },
    {
      "entropy": 6.1628539085388185,
      "epoch": 2.0511820862698054,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004582253725673184,
      "loss": 5.7692,
      "mean_token_accuracy": 0.1612524002790451,
      "num_tokens": 40130358.0,
      "step": 20520
    },
    {
      "entropy": 6.074954080581665,
      "epoch": 2.051681911331034,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000458204656481101,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17618220746517183,
      "num_tokens": 40139889.0,
      "step": 20525
    },
    {
      "entropy": 6.081759309768676,
      "epoch": 2.0521817363922628,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004581839357855015,
      "loss": 5.7601,
      "mean_token_accuracy": 0.16753579676151276,
      "num_tokens": 40150262.0,
      "step": 20530
    },
    {
      "entropy": 6.077924728393555,
      "epoch": 2.0526815614534915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045816321048104143,
      "loss": 5.689,
      "mean_token_accuracy": 0.17176225334405898,
      "num_tokens": 40159256.0,
      "step": 20535
    },
    {
      "entropy": 6.1021020889282225,
      "epoch": 2.0531813865147197,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004581424805682421,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16563212275505065,
      "num_tokens": 40169283.0,
      "step": 20540
    },
    {
      "entropy": 6.068378925323486,
      "epoch": 2.0536812115759484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000458121746047625,
      "loss": 5.6181,
      "mean_token_accuracy": 0.17696426212787628,
      "num_tokens": 40180621.0,
      "step": 20545
    },
    {
      "entropy": 6.186681604385376,
      "epoch": 2.054181036637177,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004581010069197119,
      "loss": 5.7909,
      "mean_token_accuracy": 0.16109200417995453,
      "num_tokens": 40189601.0,
      "step": 20550
    },
    {
      "entropy": 6.108200931549073,
      "epoch": 2.0546808616984054,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045808026318502464,
      "loss": 5.7669,
      "mean_token_accuracy": 0.16586344093084335,
      "num_tokens": 40200762.0,
      "step": 20555
    },
    {
      "entropy": 6.131374979019165,
      "epoch": 2.055180686759634,
      "grad_norm": 1.25,
      "learning_rate": 0.0004580595148440851,
      "loss": 5.7535,
      "mean_token_accuracy": 0.16859683394432068,
      "num_tokens": 40210667.0,
      "step": 20560
    },
    {
      "entropy": 6.027220726013184,
      "epoch": 2.0556805118208628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004580387618974152,
      "loss": 5.5531,
      "mean_token_accuracy": 0.1843808799982071,
      "num_tokens": 40220239.0,
      "step": 20565
    },
    {
      "entropy": 6.087039852142334,
      "epoch": 2.0561803368820915,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045801800434553715,
      "loss": 5.7434,
      "mean_token_accuracy": 0.16519773304462432,
      "num_tokens": 40231099.0,
      "step": 20570
    },
    {
      "entropy": 6.148837375640869,
      "epoch": 2.0566801619433197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004579972421889732,
      "loss": 5.7415,
      "mean_token_accuracy": 0.1746173806488514,
      "num_tokens": 40240372.0,
      "step": 20575
    },
    {
      "entropy": 6.055112075805664,
      "epoch": 2.0571799870045484,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004579764754282458,
      "loss": 5.7427,
      "mean_token_accuracy": 0.16501038819551467,
      "num_tokens": 40250819.0,
      "step": 20580
    },
    {
      "entropy": 6.041369247436523,
      "epoch": 2.057679812065777,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045795570406387726,
      "loss": 5.6759,
      "mean_token_accuracy": 0.17759001404047012,
      "num_tokens": 40260306.0,
      "step": 20585
    },
    {
      "entropy": 5.855421829223633,
      "epoch": 2.0581796371270054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004579349280963904,
      "loss": 5.6306,
      "mean_token_accuracy": 0.18251317441463472,
      "num_tokens": 40269828.0,
      "step": 20590
    },
    {
      "entropy": 6.104312753677368,
      "epoch": 2.058679462188234,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045791414752630765,
      "loss": 5.8019,
      "mean_token_accuracy": 0.16974111944437026,
      "num_tokens": 40279947.0,
      "step": 20595
    },
    {
      "entropy": 6.118199348449707,
      "epoch": 2.0591792872494628,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004578933623541522,
      "loss": 5.7534,
      "mean_token_accuracy": 0.1670648068189621,
      "num_tokens": 40289170.0,
      "step": 20600
    },
    {
      "entropy": 6.128918266296386,
      "epoch": 2.0596791123106915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045787257258044666,
      "loss": 5.7272,
      "mean_token_accuracy": 0.16797063425183295,
      "num_tokens": 40299561.0,
      "step": 20605
    },
    {
      "entropy": 6.13249454498291,
      "epoch": 2.0601789373719197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045785177820571434,
      "loss": 5.7139,
      "mean_token_accuracy": 0.1634964644908905,
      "num_tokens": 40309222.0,
      "step": 20610
    },
    {
      "entropy": 6.159003210067749,
      "epoch": 2.0606787624331484,
      "grad_norm": 1.25,
      "learning_rate": 0.00045783097923047833,
      "loss": 5.7519,
      "mean_token_accuracy": 0.16955784410238267,
      "num_tokens": 40318868.0,
      "step": 20615
    },
    {
      "entropy": 6.050571393966675,
      "epoch": 2.061178587494377,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045781017565526194,
      "loss": 5.7175,
      "mean_token_accuracy": 0.16865400820970536,
      "num_tokens": 40328184.0,
      "step": 20620
    },
    {
      "entropy": 6.047176742553711,
      "epoch": 2.0616784125556054,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045778936748058855,
      "loss": 5.7426,
      "mean_token_accuracy": 0.17912014573812485,
      "num_tokens": 40338927.0,
      "step": 20625
    },
    {
      "entropy": 6.132220125198364,
      "epoch": 2.062178237616834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045776855470698163,
      "loss": 5.7824,
      "mean_token_accuracy": 0.16382409185171126,
      "num_tokens": 40349174.0,
      "step": 20630
    },
    {
      "entropy": 6.002584218978882,
      "epoch": 2.0626780626780628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045774773733496505,
      "loss": 5.6334,
      "mean_token_accuracy": 0.17389543503522872,
      "num_tokens": 40359460.0,
      "step": 20635
    },
    {
      "entropy": 6.051400661468506,
      "epoch": 2.0631778877392914,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004577269153650624,
      "loss": 5.6336,
      "mean_token_accuracy": 0.17750032544136046,
      "num_tokens": 40368943.0,
      "step": 20640
    },
    {
      "entropy": 6.089885139465332,
      "epoch": 2.0636777128005197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004577060887977975,
      "loss": 5.6026,
      "mean_token_accuracy": 0.1737050950527191,
      "num_tokens": 40378285.0,
      "step": 20645
    },
    {
      "entropy": 6.055282354354858,
      "epoch": 2.0641775378617484,
      "grad_norm": 1.375,
      "learning_rate": 0.00045768525763369443,
      "loss": 5.689,
      "mean_token_accuracy": 0.17665852308273317,
      "num_tokens": 40386638.0,
      "step": 20650
    },
    {
      "entropy": 6.048986196517944,
      "epoch": 2.064677362922977,
      "grad_norm": 1.25,
      "learning_rate": 0.00045766442187327735,
      "loss": 5.7637,
      "mean_token_accuracy": 0.16850300431251525,
      "num_tokens": 40396149.0,
      "step": 20655
    },
    {
      "entropy": 6.059125375747681,
      "epoch": 2.0651771879842054,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045764358151707043,
      "loss": 5.6501,
      "mean_token_accuracy": 0.17396908402442932,
      "num_tokens": 40405915.0,
      "step": 20660
    },
    {
      "entropy": 6.12072262763977,
      "epoch": 2.065677013045434,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004576227365655979,
      "loss": 5.6914,
      "mean_token_accuracy": 0.17625551968812941,
      "num_tokens": 40415292.0,
      "step": 20665
    },
    {
      "entropy": 6.086262607574463,
      "epoch": 2.0661768381066627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004576018870193844,
      "loss": 5.7805,
      "mean_token_accuracy": 0.16395557224750518,
      "num_tokens": 40426729.0,
      "step": 20670
    },
    {
      "entropy": 6.040805339813232,
      "epoch": 2.0666766631678914,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004575810328789543,
      "loss": 5.6696,
      "mean_token_accuracy": 0.17328388392925262,
      "num_tokens": 40436016.0,
      "step": 20675
    },
    {
      "entropy": 6.1476507663726805,
      "epoch": 2.0671764882291197,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004575601741448325,
      "loss": 5.7678,
      "mean_token_accuracy": 0.16506674140691757,
      "num_tokens": 40444894.0,
      "step": 20680
    },
    {
      "entropy": 6.126026964187622,
      "epoch": 2.0676763132903484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045753931081754366,
      "loss": 5.7283,
      "mean_token_accuracy": 0.17252593338489533,
      "num_tokens": 40454650.0,
      "step": 20685
    },
    {
      "entropy": 6.1054089069366455,
      "epoch": 2.068176138351577,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004575184428976127,
      "loss": 5.7538,
      "mean_token_accuracy": 0.1626607745885849,
      "num_tokens": 40463809.0,
      "step": 20690
    },
    {
      "entropy": 6.032472896575928,
      "epoch": 2.0686759634128054,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004574975703855647,
      "loss": 5.7164,
      "mean_token_accuracy": 0.1690675362944603,
      "num_tokens": 40474497.0,
      "step": 20695
    },
    {
      "entropy": 6.059039878845215,
      "epoch": 2.069175788474034,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045747669328192477,
      "loss": 5.6897,
      "mean_token_accuracy": 0.17160843163728715,
      "num_tokens": 40484617.0,
      "step": 20700
    },
    {
      "entropy": 6.066926383972168,
      "epoch": 2.0696756135352627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004574558115872183,
      "loss": 5.6497,
      "mean_token_accuracy": 0.16986705958843232,
      "num_tokens": 40494620.0,
      "step": 20705
    },
    {
      "entropy": 6.023681497573852,
      "epoch": 2.0701754385964914,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045743492530197046,
      "loss": 5.621,
      "mean_token_accuracy": 0.18300495594739913,
      "num_tokens": 40503546.0,
      "step": 20710
    },
    {
      "entropy": 6.083880615234375,
      "epoch": 2.0706752636577197,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004574140344267069,
      "loss": 5.7184,
      "mean_token_accuracy": 0.1722347468137741,
      "num_tokens": 40513373.0,
      "step": 20715
    },
    {
      "entropy": 6.0756556510925295,
      "epoch": 2.0711750887189484,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004573931389619531,
      "loss": 5.7225,
      "mean_token_accuracy": 0.16912156492471694,
      "num_tokens": 40523468.0,
      "step": 20720
    },
    {
      "entropy": 6.088141012191772,
      "epoch": 2.071674913780177,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045737223890823494,
      "loss": 5.7372,
      "mean_token_accuracy": 0.16296301782131195,
      "num_tokens": 40533570.0,
      "step": 20725
    },
    {
      "entropy": 6.064460515975952,
      "epoch": 2.0721747388414054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045735133426607813,
      "loss": 5.7601,
      "mean_token_accuracy": 0.1662217766046524,
      "num_tokens": 40544223.0,
      "step": 20730
    },
    {
      "entropy": 6.078164434432983,
      "epoch": 2.072674563902634,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045733042503600865,
      "loss": 5.6684,
      "mean_token_accuracy": 0.17542317509651184,
      "num_tokens": 40554201.0,
      "step": 20735
    },
    {
      "entropy": 5.994706106185913,
      "epoch": 2.0731743889638627,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045730951121855255,
      "loss": 5.6246,
      "mean_token_accuracy": 0.18095745146274567,
      "num_tokens": 40564134.0,
      "step": 20740
    },
    {
      "entropy": 6.13175139427185,
      "epoch": 2.0736742140250914,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004572885928142362,
      "loss": 5.8264,
      "mean_token_accuracy": 0.16297447681427002,
      "num_tokens": 40575466.0,
      "step": 20745
    },
    {
      "entropy": 6.183193111419678,
      "epoch": 2.0741740390863197,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004572676698235857,
      "loss": 5.7957,
      "mean_token_accuracy": 0.16958122700452805,
      "num_tokens": 40586296.0,
      "step": 20750
    },
    {
      "entropy": 6.072747898101807,
      "epoch": 2.0746738641475484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004572467422471275,
      "loss": 5.7438,
      "mean_token_accuracy": 0.17147841006517411,
      "num_tokens": 40597721.0,
      "step": 20755
    },
    {
      "entropy": 6.040781736373901,
      "epoch": 2.075173689208777,
      "grad_norm": 1.3125,
      "learning_rate": 0.00045722581008538816,
      "loss": 5.7134,
      "mean_token_accuracy": 0.16845156252384186,
      "num_tokens": 40607804.0,
      "step": 20760
    },
    {
      "entropy": 6.053397130966187,
      "epoch": 2.0756735142700053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004572048733388944,
      "loss": 5.7025,
      "mean_token_accuracy": 0.17177778482437134,
      "num_tokens": 40617652.0,
      "step": 20765
    },
    {
      "entropy": 6.115008783340454,
      "epoch": 2.076173339331234,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045718393200817284,
      "loss": 5.7236,
      "mean_token_accuracy": 0.16818695217370988,
      "num_tokens": 40626949.0,
      "step": 20770
    },
    {
      "entropy": 6.029764080047608,
      "epoch": 2.0766731643924627,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045716298609375044,
      "loss": 5.7686,
      "mean_token_accuracy": 0.1721368670463562,
      "num_tokens": 40636176.0,
      "step": 20775
    },
    {
      "entropy": 6.114700794219971,
      "epoch": 2.077172989453691,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045714203559615423,
      "loss": 5.7865,
      "mean_token_accuracy": 0.17197150886058807,
      "num_tokens": 40646268.0,
      "step": 20780
    },
    {
      "entropy": 6.15015926361084,
      "epoch": 2.0776728145149197,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004571210805159112,
      "loss": 5.8044,
      "mean_token_accuracy": 0.16312361806631087,
      "num_tokens": 40655352.0,
      "step": 20785
    },
    {
      "entropy": 6.085427904129029,
      "epoch": 2.0781726395761484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004571001208535487,
      "loss": 5.717,
      "mean_token_accuracy": 0.16633364409208298,
      "num_tokens": 40665166.0,
      "step": 20790
    },
    {
      "entropy": 6.019620943069458,
      "epoch": 2.078672464637377,
      "grad_norm": 1.21875,
      "learning_rate": 0.000457079156609594,
      "loss": 5.5883,
      "mean_token_accuracy": 0.1793285831809044,
      "num_tokens": 40675149.0,
      "step": 20795
    },
    {
      "entropy": 6.103204441070557,
      "epoch": 2.0791722896986053,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004570581877845746,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17178708165884018,
      "num_tokens": 40685116.0,
      "step": 20800
    },
    {
      "entropy": 6.027391767501831,
      "epoch": 2.079672114759834,
      "grad_norm": 1.3125,
      "learning_rate": 0.000457037214379018,
      "loss": 5.5879,
      "mean_token_accuracy": 0.18337980210781096,
      "num_tokens": 40694834.0,
      "step": 20805
    },
    {
      "entropy": 6.1471747875213625,
      "epoch": 2.0801719398210627,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045701623639345204,
      "loss": 5.8493,
      "mean_token_accuracy": 0.16641114950180053,
      "num_tokens": 40704294.0,
      "step": 20810
    },
    {
      "entropy": 6.04757399559021,
      "epoch": 2.080671764882291,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045699525382840423,
      "loss": 5.7141,
      "mean_token_accuracy": 0.16801653504371644,
      "num_tokens": 40714269.0,
      "step": 20815
    },
    {
      "entropy": 6.174980688095093,
      "epoch": 2.0811715899435197,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045697426668440277,
      "loss": 5.7699,
      "mean_token_accuracy": 0.15996603965759276,
      "num_tokens": 40723795.0,
      "step": 20820
    },
    {
      "entropy": 6.103749942779541,
      "epoch": 2.0816714150047484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045695327496197557,
      "loss": 5.7939,
      "mean_token_accuracy": 0.16598533242940902,
      "num_tokens": 40734702.0,
      "step": 20825
    },
    {
      "entropy": 6.051512289047241,
      "epoch": 2.082171240065977,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004569322786616508,
      "loss": 5.6278,
      "mean_token_accuracy": 0.17919976264238358,
      "num_tokens": 40744174.0,
      "step": 20830
    },
    {
      "entropy": 6.111645650863648,
      "epoch": 2.0826710651272053,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004569112777839567,
      "loss": 5.7679,
      "mean_token_accuracy": 0.16601127088069917,
      "num_tokens": 40754474.0,
      "step": 20835
    },
    {
      "entropy": 6.088803005218506,
      "epoch": 2.083170890188434,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045689027232942165,
      "loss": 5.6968,
      "mean_token_accuracy": 0.17266090214252472,
      "num_tokens": 40764051.0,
      "step": 20840
    },
    {
      "entropy": 6.139381837844849,
      "epoch": 2.0836707152496627,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004568692622985742,
      "loss": 5.7431,
      "mean_token_accuracy": 0.16517102271318435,
      "num_tokens": 40774518.0,
      "step": 20845
    },
    {
      "entropy": 6.032560682296753,
      "epoch": 2.084170540310891,
      "grad_norm": 1.25,
      "learning_rate": 0.00045684824769194284,
      "loss": 5.7492,
      "mean_token_accuracy": 0.1695293143391609,
      "num_tokens": 40783887.0,
      "step": 20850
    },
    {
      "entropy": 6.102872085571289,
      "epoch": 2.0846703653721197,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004568272285100564,
      "loss": 5.773,
      "mean_token_accuracy": 0.16484374403953553,
      "num_tokens": 40793607.0,
      "step": 20855
    },
    {
      "entropy": 6.095940065383911,
      "epoch": 2.0851701904333484,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004568062047534437,
      "loss": 5.7122,
      "mean_token_accuracy": 0.1681744635105133,
      "num_tokens": 40802948.0,
      "step": 20860
    },
    {
      "entropy": 6.054484796524048,
      "epoch": 2.085670015494577,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004567851764226337,
      "loss": 5.7451,
      "mean_token_accuracy": 0.162997043132782,
      "num_tokens": 40812733.0,
      "step": 20865
    },
    {
      "entropy": 6.131355142593383,
      "epoch": 2.0861698405558053,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045676414351815533,
      "loss": 5.686,
      "mean_token_accuracy": 0.16982703953981398,
      "num_tokens": 40821074.0,
      "step": 20870
    },
    {
      "entropy": 6.156388759613037,
      "epoch": 2.086669665617034,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004567431060405379,
      "loss": 5.743,
      "mean_token_accuracy": 0.17190967351198197,
      "num_tokens": 40831427.0,
      "step": 20875
    },
    {
      "entropy": 6.135774326324463,
      "epoch": 2.0871694906782627,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004567220639903108,
      "loss": 5.7153,
      "mean_token_accuracy": 0.16997757852077483,
      "num_tokens": 40841433.0,
      "step": 20880
    },
    {
      "entropy": 6.100941324234009,
      "epoch": 2.087669315739491,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045670101736800325,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1667348176240921,
      "num_tokens": 40851176.0,
      "step": 20885
    },
    {
      "entropy": 6.157898855209351,
      "epoch": 2.0881691408007197,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045667996617414485,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16626515090465546,
      "num_tokens": 40860436.0,
      "step": 20890
    },
    {
      "entropy": 6.203657627105713,
      "epoch": 2.0886689658619484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004566589104092653,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1653279185295105,
      "num_tokens": 40870650.0,
      "step": 20895
    },
    {
      "entropy": 6.109425687789917,
      "epoch": 2.089168790923177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004566378500738943,
      "loss": 5.7275,
      "mean_token_accuracy": 0.1756486788392067,
      "num_tokens": 40879967.0,
      "step": 20900
    },
    {
      "entropy": 6.012493371963501,
      "epoch": 2.0896686159844053,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045661678516856164,
      "loss": 5.6418,
      "mean_token_accuracy": 0.1745280832052231,
      "num_tokens": 40889299.0,
      "step": 20905
    },
    {
      "entropy": 5.99773325920105,
      "epoch": 2.090168441045634,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004565957156937975,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18397777676582336,
      "num_tokens": 40899165.0,
      "step": 20910
    },
    {
      "entropy": 6.05017056465149,
      "epoch": 2.0906682661068627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004565746416501318,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1676620677113533,
      "num_tokens": 40908834.0,
      "step": 20915
    },
    {
      "entropy": 6.123997497558594,
      "epoch": 2.091168091168091,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045655356303809494,
      "loss": 5.7433,
      "mean_token_accuracy": 0.16568946987390518,
      "num_tokens": 40918132.0,
      "step": 20920
    },
    {
      "entropy": 6.083917284011841,
      "epoch": 2.0916679162293197,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045653247985821707,
      "loss": 5.6953,
      "mean_token_accuracy": 0.16744003295898438,
      "num_tokens": 40929050.0,
      "step": 20925
    },
    {
      "entropy": 6.111245059967041,
      "epoch": 2.0921677412905484,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045651139211102875,
      "loss": 5.8016,
      "mean_token_accuracy": 0.1605992317199707,
      "num_tokens": 40938876.0,
      "step": 20930
    },
    {
      "entropy": 6.125909948348999,
      "epoch": 2.092667566351777,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00045649029979706054,
      "loss": 5.7512,
      "mean_token_accuracy": 0.16838214099407195,
      "num_tokens": 40948639.0,
      "step": 20935
    },
    {
      "entropy": 6.106826591491699,
      "epoch": 2.0931673914130053,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00045646920291684303,
      "loss": 5.7341,
      "mean_token_accuracy": 0.1634480372071266,
      "num_tokens": 40958718.0,
      "step": 20940
    },
    {
      "entropy": 6.094656181335449,
      "epoch": 2.093667216474234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004564481014709072,
      "loss": 5.7191,
      "mean_token_accuracy": 0.1653069645166397,
      "num_tokens": 40967962.0,
      "step": 20945
    },
    {
      "entropy": 6.09478178024292,
      "epoch": 2.0941670415354627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004564269954597836,
      "loss": 5.6789,
      "mean_token_accuracy": 0.17250234335660936,
      "num_tokens": 40978566.0,
      "step": 20950
    },
    {
      "entropy": 6.037272596359253,
      "epoch": 2.094666866596691,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045640588488400364,
      "loss": 5.669,
      "mean_token_accuracy": 0.1781452164053917,
      "num_tokens": 40989439.0,
      "step": 20955
    },
    {
      "entropy": 6.009652376174927,
      "epoch": 2.0951666916579197,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004563847697440983,
      "loss": 5.6884,
      "mean_token_accuracy": 0.1758020743727684,
      "num_tokens": 40998934.0,
      "step": 20960
    },
    {
      "entropy": 6.09810585975647,
      "epoch": 2.0956665167191484,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045636365004059876,
      "loss": 5.6439,
      "mean_token_accuracy": 0.18098915368318558,
      "num_tokens": 41007965.0,
      "step": 20965
    },
    {
      "entropy": 6.134417486190796,
      "epoch": 2.0961663417803766,
      "grad_norm": 1.25,
      "learning_rate": 0.00045634252577403644,
      "loss": 5.7669,
      "mean_token_accuracy": 0.15776253044605254,
      "num_tokens": 41018572.0,
      "step": 20970
    },
    {
      "entropy": 6.048743391036988,
      "epoch": 2.0966661668416053,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045632139694494294,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17729633748531343,
      "num_tokens": 41028272.0,
      "step": 20975
    },
    {
      "entropy": 5.995200824737549,
      "epoch": 2.097165991902834,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004563002635538496,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18557381480932236,
      "num_tokens": 41037282.0,
      "step": 20980
    },
    {
      "entropy": 6.094186544418335,
      "epoch": 2.0976658169640627,
      "grad_norm": 1.5,
      "learning_rate": 0.0004562791256012884,
      "loss": 5.6863,
      "mean_token_accuracy": 0.17826733142137527,
      "num_tokens": 41046590.0,
      "step": 20985
    },
    {
      "entropy": 5.938064861297607,
      "epoch": 2.098165642025291,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004562579830877911,
      "loss": 5.5253,
      "mean_token_accuracy": 0.1897111117839813,
      "num_tokens": 41057014.0,
      "step": 20990
    },
    {
      "entropy": 6.13736629486084,
      "epoch": 2.0986654670865197,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004562368360138894,
      "loss": 5.7063,
      "mean_token_accuracy": 0.17003546804189681,
      "num_tokens": 41065882.0,
      "step": 20995
    },
    {
      "entropy": 6.065377235412598,
      "epoch": 2.0991652921477484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004562156843801157,
      "loss": 5.6948,
      "mean_token_accuracy": 0.16880695670843124,
      "num_tokens": 41075971.0,
      "step": 21000
    },
    {
      "epoch": 2.0991652921477484,
      "eval_entropy": 5.883823129388052,
      "eval_loss": 5.853286266326904,
      "eval_mean_token_accuracy": 0.17254896714315837,
      "eval_num_tokens": 41075971.0,
      "eval_runtime": 23.9152,
      "eval_samples_per_second": 1298.128,
      "eval_steps_per_second": 162.282,
      "step": 21000
    },
    {
      "entropy": 6.049785327911377,
      "epoch": 2.0996651172089766,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045619452818700194,
      "loss": 5.5983,
      "mean_token_accuracy": 0.17930205911397934,
      "num_tokens": 41085305.0,
      "step": 21005
    },
    {
      "entropy": 6.151028394699097,
      "epoch": 2.1001649422702053,
      "grad_norm": 1.4375,
      "learning_rate": 0.00045617336743508056,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16336980313062668,
      "num_tokens": 41094995.0,
      "step": 21010
    },
    {
      "entropy": 6.110802793502808,
      "epoch": 2.100664767331434,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004561522021248837,
      "loss": 5.7091,
      "mean_token_accuracy": 0.1704626426100731,
      "num_tokens": 41104809.0,
      "step": 21015
    },
    {
      "entropy": 6.133065223693848,
      "epoch": 2.1011645923926627,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004561310322569442,
      "loss": 5.8164,
      "mean_token_accuracy": 0.15958776473999023,
      "num_tokens": 41116002.0,
      "step": 21020
    },
    {
      "entropy": 6.049251890182495,
      "epoch": 2.101664417453891,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045610985783179454,
      "loss": 5.6905,
      "mean_token_accuracy": 0.16500808596611022,
      "num_tokens": 41127657.0,
      "step": 21025
    },
    {
      "entropy": 6.071070003509521,
      "epoch": 2.1021642425151197,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004560886788499674,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17546890079975128,
      "num_tokens": 41137151.0,
      "step": 21030
    },
    {
      "entropy": 6.0857562065124515,
      "epoch": 2.1026640675763484,
      "grad_norm": 1.328125,
      "learning_rate": 0.00045606749531199573,
      "loss": 5.6944,
      "mean_token_accuracy": 0.17500328421592712,
      "num_tokens": 41146365.0,
      "step": 21035
    },
    {
      "entropy": 6.068025541305542,
      "epoch": 2.1031638926375766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045604630721841243,
      "loss": 5.7164,
      "mean_token_accuracy": 0.16332240477204324,
      "num_tokens": 41155506.0,
      "step": 21040
    },
    {
      "entropy": 6.062033748626709,
      "epoch": 2.1036637176988053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045602511456975064,
      "loss": 5.673,
      "mean_token_accuracy": 0.16884481757879258,
      "num_tokens": 41166382.0,
      "step": 21045
    },
    {
      "entropy": 6.048169994354248,
      "epoch": 2.104163542760034,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004560039173665436,
      "loss": 5.6813,
      "mean_token_accuracy": 0.17500511109828948,
      "num_tokens": 41176658.0,
      "step": 21050
    },
    {
      "entropy": 5.952599668502808,
      "epoch": 2.1046633678212627,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004559827156093245,
      "loss": 5.5021,
      "mean_token_accuracy": 0.19114165008068085,
      "num_tokens": 41187528.0,
      "step": 21055
    },
    {
      "entropy": 6.116586971282959,
      "epoch": 2.105163192882491,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004559615092986268,
      "loss": 5.7109,
      "mean_token_accuracy": 0.1739984706044197,
      "num_tokens": 41197923.0,
      "step": 21060
    },
    {
      "entropy": 6.035449457168579,
      "epoch": 2.1056630179437197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045594029843498425,
      "loss": 5.6544,
      "mean_token_accuracy": 0.17582201361656188,
      "num_tokens": 41207358.0,
      "step": 21065
    },
    {
      "entropy": 6.06853928565979,
      "epoch": 2.1061628430049484,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045591908301893027,
      "loss": 5.7609,
      "mean_token_accuracy": 0.16642290651798247,
      "num_tokens": 41217733.0,
      "step": 21070
    },
    {
      "entropy": 6.150815773010254,
      "epoch": 2.1066626680661766,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045589786305099875,
      "loss": 5.828,
      "mean_token_accuracy": 0.1677322804927826,
      "num_tokens": 41228065.0,
      "step": 21075
    },
    {
      "entropy": 6.170335912704468,
      "epoch": 2.1071624931274053,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004558766385317235,
      "loss": 5.7715,
      "mean_token_accuracy": 0.16590268760919571,
      "num_tokens": 41237192.0,
      "step": 21080
    },
    {
      "entropy": 5.975124454498291,
      "epoch": 2.107662318188634,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045585540946163855,
      "loss": 5.5088,
      "mean_token_accuracy": 0.18679441660642623,
      "num_tokens": 41247583.0,
      "step": 21085
    },
    {
      "entropy": 6.04443325996399,
      "epoch": 2.1081621432498627,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004558341758412781,
      "loss": 5.7036,
      "mean_token_accuracy": 0.16916703581809997,
      "num_tokens": 41257179.0,
      "step": 21090
    },
    {
      "entropy": 6.101089191436768,
      "epoch": 2.108661968311091,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004558129376711763,
      "loss": 5.7517,
      "mean_token_accuracy": 0.16538990288972855,
      "num_tokens": 41266737.0,
      "step": 21095
    },
    {
      "entropy": 6.058122301101685,
      "epoch": 2.1091617933723197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004557916949518675,
      "loss": 5.6293,
      "mean_token_accuracy": 0.17445387095212936,
      "num_tokens": 41275449.0,
      "step": 21100
    },
    {
      "entropy": 6.093290948867798,
      "epoch": 2.1096616184335484,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004557704476838862,
      "loss": 5.7556,
      "mean_token_accuracy": 0.1699414387345314,
      "num_tokens": 41285555.0,
      "step": 21105
    },
    {
      "entropy": 6.004319143295288,
      "epoch": 2.1101614434947766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045574919586776686,
      "loss": 5.5807,
      "mean_token_accuracy": 0.1822939470410347,
      "num_tokens": 41295649.0,
      "step": 21110
    },
    {
      "entropy": 6.06004753112793,
      "epoch": 2.1106612685560053,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045572793950404434,
      "loss": 5.7316,
      "mean_token_accuracy": 0.16908913254737853,
      "num_tokens": 41304762.0,
      "step": 21115
    },
    {
      "entropy": 6.0354126453399655,
      "epoch": 2.111161093617234,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045570667859325337,
      "loss": 5.7049,
      "mean_token_accuracy": 0.16935069411993026,
      "num_tokens": 41314840.0,
      "step": 21120
    },
    {
      "entropy": 6.15801510810852,
      "epoch": 2.1116609186784627,
      "grad_norm": 2.65625,
      "learning_rate": 0.00045568541313592884,
      "loss": 5.7125,
      "mean_token_accuracy": 0.17145930081605912,
      "num_tokens": 41326056.0,
      "step": 21125
    },
    {
      "entropy": 6.092984914779663,
      "epoch": 2.112160743739691,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045566414313260586,
      "loss": 5.6596,
      "mean_token_accuracy": 0.17258236408233643,
      "num_tokens": 41336348.0,
      "step": 21130
    },
    {
      "entropy": 6.054533767700195,
      "epoch": 2.1126605688009197,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004556428685838194,
      "loss": 5.7294,
      "mean_token_accuracy": 0.16754430383443833,
      "num_tokens": 41347112.0,
      "step": 21135
    },
    {
      "entropy": 6.056876182556152,
      "epoch": 2.1131603938621484,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000455621589490105,
      "loss": 5.7062,
      "mean_token_accuracy": 0.16552826762199402,
      "num_tokens": 41356872.0,
      "step": 21140
    },
    {
      "entropy": 6.087315797805786,
      "epoch": 2.1136602189233766,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004556003058519979,
      "loss": 5.6681,
      "mean_token_accuracy": 0.1752202406525612,
      "num_tokens": 41365710.0,
      "step": 21145
    },
    {
      "entropy": 6.032313823699951,
      "epoch": 2.1141600439846053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045557901767003347,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17100085765123368,
      "num_tokens": 41375676.0,
      "step": 21150
    },
    {
      "entropy": 6.081670141220092,
      "epoch": 2.114659869045834,
      "grad_norm": 2.1875,
      "learning_rate": 0.00045555772494474745,
      "loss": 5.732,
      "mean_token_accuracy": 0.16997245848178863,
      "num_tokens": 41386134.0,
      "step": 21155
    },
    {
      "entropy": 6.094786977767944,
      "epoch": 2.1151596941070627,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004555364276766756,
      "loss": 5.6753,
      "mean_token_accuracy": 0.17503724694252015,
      "num_tokens": 41396318.0,
      "step": 21160
    },
    {
      "entropy": 6.000144672393799,
      "epoch": 2.115659519168291,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004555151258663535,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1759974792599678,
      "num_tokens": 41405921.0,
      "step": 21165
    },
    {
      "entropy": 6.048991966247558,
      "epoch": 2.1161593442295197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004554938195143175,
      "loss": 5.6438,
      "mean_token_accuracy": 0.1724298670887947,
      "num_tokens": 41415868.0,
      "step": 21170
    },
    {
      "entropy": 6.021948337554932,
      "epoch": 2.1166591692907484,
      "grad_norm": 1.25,
      "learning_rate": 0.0004554725086211033,
      "loss": 5.6987,
      "mean_token_accuracy": 0.1749032750725746,
      "num_tokens": 41425585.0,
      "step": 21175
    },
    {
      "entropy": 6.032670640945435,
      "epoch": 2.1171589943519766,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045545119318724734,
      "loss": 5.6926,
      "mean_token_accuracy": 0.17649065107107162,
      "num_tokens": 41434675.0,
      "step": 21180
    },
    {
      "entropy": 6.232807445526123,
      "epoch": 2.1176588194132053,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004554298732132857,
      "loss": 5.8382,
      "mean_token_accuracy": 0.15925035029649734,
      "num_tokens": 41445644.0,
      "step": 21185
    },
    {
      "entropy": 6.062338972091675,
      "epoch": 2.118158644474434,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004554085486997549,
      "loss": 5.6343,
      "mean_token_accuracy": 0.1768007442355156,
      "num_tokens": 41455416.0,
      "step": 21190
    },
    {
      "entropy": 6.061080265045166,
      "epoch": 2.1186584695356627,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004553872196471914,
      "loss": 5.7461,
      "mean_token_accuracy": 0.17173900306224824,
      "num_tokens": 41465988.0,
      "step": 21195
    },
    {
      "entropy": 6.008725261688232,
      "epoch": 2.119158294596891,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045536588605613195,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17656796425580978,
      "num_tokens": 41475704.0,
      "step": 21200
    },
    {
      "entropy": 6.08421630859375,
      "epoch": 2.1196581196581197,
      "grad_norm": 1.578125,
      "learning_rate": 0.00045534454792711324,
      "loss": 5.8014,
      "mean_token_accuracy": 0.16511528044939042,
      "num_tokens": 41485723.0,
      "step": 21205
    },
    {
      "entropy": 6.146075868606568,
      "epoch": 2.1201579447193484,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045532320526067206,
      "loss": 5.7523,
      "mean_token_accuracy": 0.1675642728805542,
      "num_tokens": 41495610.0,
      "step": 21210
    },
    {
      "entropy": 6.099324512481689,
      "epoch": 2.1206577697805766,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004553018580573454,
      "loss": 5.7151,
      "mean_token_accuracy": 0.16849422603845596,
      "num_tokens": 41504799.0,
      "step": 21215
    },
    {
      "entropy": 6.097462511062622,
      "epoch": 2.1211575948418053,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00045528050631767037,
      "loss": 5.6953,
      "mean_token_accuracy": 0.16615478396415712,
      "num_tokens": 41514335.0,
      "step": 21220
    },
    {
      "entropy": 6.06825008392334,
      "epoch": 2.121657419903034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045525915004218426,
      "loss": 5.705,
      "mean_token_accuracy": 0.16547246277332306,
      "num_tokens": 41523955.0,
      "step": 21225
    },
    {
      "entropy": 6.076111507415772,
      "epoch": 2.1221572449642627,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045523778923142435,
      "loss": 5.6894,
      "mean_token_accuracy": 0.1738893836736679,
      "num_tokens": 41532647.0,
      "step": 21230
    },
    {
      "entropy": 6.089785814285278,
      "epoch": 2.122657070025491,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004552164238859279,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17104080468416213,
      "num_tokens": 41542885.0,
      "step": 21235
    },
    {
      "entropy": 6.077095174789429,
      "epoch": 2.1231568950867197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045519505400623274,
      "loss": 5.8147,
      "mean_token_accuracy": 0.17540149688720702,
      "num_tokens": 41553381.0,
      "step": 21240
    },
    {
      "entropy": 6.155397987365722,
      "epoch": 2.1236567201479484,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045517367959287635,
      "loss": 5.7416,
      "mean_token_accuracy": 0.1674499049782753,
      "num_tokens": 41563740.0,
      "step": 21245
    },
    {
      "entropy": 6.115851259231567,
      "epoch": 2.1241565452091766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045515230064639657,
      "loss": 5.6357,
      "mean_token_accuracy": 0.17527905404567717,
      "num_tokens": 41574404.0,
      "step": 21250
    },
    {
      "entropy": 6.0682367324829105,
      "epoch": 2.1246563702704053,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045513091716733117,
      "loss": 5.659,
      "mean_token_accuracy": 0.17287605553865432,
      "num_tokens": 41583532.0,
      "step": 21255
    },
    {
      "entropy": 6.037268447875976,
      "epoch": 2.125156195331634,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00045510952915621834,
      "loss": 5.6521,
      "mean_token_accuracy": 0.17227199524641038,
      "num_tokens": 41594003.0,
      "step": 21260
    },
    {
      "entropy": 6.094523525238037,
      "epoch": 2.1256560203928627,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004550881366135961,
      "loss": 5.6624,
      "mean_token_accuracy": 0.1684977665543556,
      "num_tokens": 41603779.0,
      "step": 21265
    },
    {
      "entropy": 5.962189245223999,
      "epoch": 2.126155845454091,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045506673954000265,
      "loss": 5.4893,
      "mean_token_accuracy": 0.19601529687643052,
      "num_tokens": 41613666.0,
      "step": 21270
    },
    {
      "entropy": 6.097946882247925,
      "epoch": 2.1266556705153197,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004550453379359763,
      "loss": 5.7144,
      "mean_token_accuracy": 0.16949256360530854,
      "num_tokens": 41622312.0,
      "step": 21275
    },
    {
      "entropy": 6.135853004455567,
      "epoch": 2.1271554955765484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045502393180205564,
      "loss": 5.7936,
      "mean_token_accuracy": 0.16290003657341004,
      "num_tokens": 41632798.0,
      "step": 21280
    },
    {
      "entropy": 6.074649286270142,
      "epoch": 2.1276553206377766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004550025211387792,
      "loss": 5.7201,
      "mean_token_accuracy": 0.17108866423368455,
      "num_tokens": 41641396.0,
      "step": 21285
    },
    {
      "entropy": 6.1466837406158445,
      "epoch": 2.1281551456990053,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004549811059466857,
      "loss": 5.7596,
      "mean_token_accuracy": 0.1683203399181366,
      "num_tokens": 41650136.0,
      "step": 21290
    },
    {
      "entropy": 6.114579534530639,
      "epoch": 2.128654970760234,
      "grad_norm": 1.25,
      "learning_rate": 0.0004549596862263138,
      "loss": 5.7399,
      "mean_token_accuracy": 0.16693121939897537,
      "num_tokens": 41660001.0,
      "step": 21295
    },
    {
      "entropy": 6.005676317214966,
      "epoch": 2.1291547958214627,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004549382619782025,
      "loss": 5.7175,
      "mean_token_accuracy": 0.17273443043231965,
      "num_tokens": 41669846.0,
      "step": 21300
    },
    {
      "entropy": 6.082149934768677,
      "epoch": 2.129654620882691,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004549168332028908,
      "loss": 5.6416,
      "mean_token_accuracy": 0.17776622772216796,
      "num_tokens": 41678684.0,
      "step": 21305
    },
    {
      "entropy": 6.017084741592408,
      "epoch": 2.1301544459439197,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045489539990091796,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1746670663356781,
      "num_tokens": 41688475.0,
      "step": 21310
    },
    {
      "entropy": 6.039726877212525,
      "epoch": 2.1306542710051484,
      "grad_norm": 1.453125,
      "learning_rate": 0.00045487396207282316,
      "loss": 5.6446,
      "mean_token_accuracy": 0.18180587887763977,
      "num_tokens": 41698541.0,
      "step": 21315
    },
    {
      "entropy": 6.017726707458496,
      "epoch": 2.1311540960663766,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045485251971914565,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1727582186460495,
      "num_tokens": 41707138.0,
      "step": 21320
    },
    {
      "entropy": 6.086601781845093,
      "epoch": 2.1316539211276053,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004548310728404251,
      "loss": 5.7435,
      "mean_token_accuracy": 0.16415712386369705,
      "num_tokens": 41718103.0,
      "step": 21325
    },
    {
      "entropy": 6.105875492095947,
      "epoch": 2.132153746188834,
      "grad_norm": 1.21875,
      "learning_rate": 0.000454809621437201,
      "loss": 5.6797,
      "mean_token_accuracy": 0.17060742825269698,
      "num_tokens": 41727889.0,
      "step": 21330
    },
    {
      "entropy": 6.091103172302246,
      "epoch": 2.1326535712500623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004547881655100131,
      "loss": 5.7745,
      "mean_token_accuracy": 0.17081725150346755,
      "num_tokens": 41738421.0,
      "step": 21335
    },
    {
      "entropy": 5.990882205963135,
      "epoch": 2.133153396311291,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004547667050594013,
      "loss": 5.6132,
      "mean_token_accuracy": 0.1814442828297615,
      "num_tokens": 41748192.0,
      "step": 21340
    },
    {
      "entropy": 6.0295015335083,
      "epoch": 2.1336532213725197,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045474524008590536,
      "loss": 5.701,
      "mean_token_accuracy": 0.17400831133127212,
      "num_tokens": 41757000.0,
      "step": 21345
    },
    {
      "entropy": 6.071630620956421,
      "epoch": 2.1341530464337484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004547237705900655,
      "loss": 5.6385,
      "mean_token_accuracy": 0.17987994849681854,
      "num_tokens": 41766163.0,
      "step": 21350
    },
    {
      "entropy": 6.050393152236938,
      "epoch": 2.1346528714949766,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045470229657242176,
      "loss": 5.6276,
      "mean_token_accuracy": 0.1771720826625824,
      "num_tokens": 41775600.0,
      "step": 21355
    },
    {
      "entropy": 6.066697883605957,
      "epoch": 2.1351526965562053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045468081803351447,
      "loss": 5.8289,
      "mean_token_accuracy": 0.1603616625070572,
      "num_tokens": 41784535.0,
      "step": 21360
    },
    {
      "entropy": 6.090231037139892,
      "epoch": 2.135652521617434,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004546593349738841,
      "loss": 5.7154,
      "mean_token_accuracy": 0.16477185785770415,
      "num_tokens": 41794534.0,
      "step": 21365
    },
    {
      "entropy": 6.014848423004151,
      "epoch": 2.1361523466786623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000454637847394071,
      "loss": 5.6419,
      "mean_token_accuracy": 0.17683679312467576,
      "num_tokens": 41804155.0,
      "step": 21370
    },
    {
      "entropy": 5.984817028045654,
      "epoch": 2.136652171739891,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045461635529461597,
      "loss": 5.6283,
      "mean_token_accuracy": 0.176930770277977,
      "num_tokens": 41814813.0,
      "step": 21375
    },
    {
      "entropy": 6.034980344772339,
      "epoch": 2.1371519968011197,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004545948586760596,
      "loss": 5.6434,
      "mean_token_accuracy": 0.17669469863176346,
      "num_tokens": 41824660.0,
      "step": 21380
    },
    {
      "entropy": 6.073409748077393,
      "epoch": 2.1376518218623484,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004545733575389428,
      "loss": 5.6584,
      "mean_token_accuracy": 0.18451297879219056,
      "num_tokens": 41834139.0,
      "step": 21385
    },
    {
      "entropy": 6.010978031158447,
      "epoch": 2.1381516469235766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004545518518838066,
      "loss": 5.7327,
      "mean_token_accuracy": 0.17084549367427826,
      "num_tokens": 41845183.0,
      "step": 21390
    },
    {
      "entropy": 5.9997998714447025,
      "epoch": 2.1386514719848053,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000454530341711192,
      "loss": 5.6525,
      "mean_token_accuracy": 0.17596149891614915,
      "num_tokens": 41854569.0,
      "step": 21395
    },
    {
      "entropy": 6.09321551322937,
      "epoch": 2.139151297046034,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004545088270216401,
      "loss": 5.618,
      "mean_token_accuracy": 0.18022694438695908,
      "num_tokens": 41863469.0,
      "step": 21400
    },
    {
      "entropy": 6.092192983627319,
      "epoch": 2.1396511221072623,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004544873078156924,
      "loss": 5.7248,
      "mean_token_accuracy": 0.1730280488729477,
      "num_tokens": 41873172.0,
      "step": 21405
    },
    {
      "entropy": 6.008124160766601,
      "epoch": 2.140150947168491,
      "grad_norm": 1.25,
      "learning_rate": 0.00045446578409389016,
      "loss": 5.6542,
      "mean_token_accuracy": 0.16920198500156403,
      "num_tokens": 41882276.0,
      "step": 21410
    },
    {
      "entropy": 6.07354507446289,
      "epoch": 2.1406507722297197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045444425585677503,
      "loss": 5.7941,
      "mean_token_accuracy": 0.16323873847723008,
      "num_tokens": 41893240.0,
      "step": 21415
    },
    {
      "entropy": 6.16746973991394,
      "epoch": 2.1411505972909484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004544227231048886,
      "loss": 5.69,
      "mean_token_accuracy": 0.17111559361219406,
      "num_tokens": 41902889.0,
      "step": 21420
    },
    {
      "entropy": 6.09088168144226,
      "epoch": 2.1416504223521766,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004544011858387726,
      "loss": 5.7358,
      "mean_token_accuracy": 0.1704891175031662,
      "num_tokens": 41911887.0,
      "step": 21425
    },
    {
      "entropy": 5.950848484039307,
      "epoch": 2.1421502474134053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045437964405896895,
      "loss": 5.6197,
      "mean_token_accuracy": 0.18271611630916595,
      "num_tokens": 41921563.0,
      "step": 21430
    },
    {
      "entropy": 6.019585990905762,
      "epoch": 2.142650072474634,
      "grad_norm": 1.40625,
      "learning_rate": 0.00045435809776601954,
      "loss": 5.6338,
      "mean_token_accuracy": 0.17090944647789003,
      "num_tokens": 41930454.0,
      "step": 21435
    },
    {
      "entropy": 6.096118307113647,
      "epoch": 2.1431498975358623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004543365469604666,
      "loss": 5.7562,
      "mean_token_accuracy": 0.16460837572813034,
      "num_tokens": 41939289.0,
      "step": 21440
    },
    {
      "entropy": 6.007991933822632,
      "epoch": 2.143649722597091,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004543149916428523,
      "loss": 5.6488,
      "mean_token_accuracy": 0.1725596472620964,
      "num_tokens": 41949569.0,
      "step": 21445
    },
    {
      "entropy": 6.030355024337768,
      "epoch": 2.1441495476583197,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004542934318137189,
      "loss": 5.7088,
      "mean_token_accuracy": 0.1726304903626442,
      "num_tokens": 41960767.0,
      "step": 21450
    },
    {
      "entropy": 6.0758223056793215,
      "epoch": 2.1446493727195484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004542718674736089,
      "loss": 5.6811,
      "mean_token_accuracy": 0.16844454258680344,
      "num_tokens": 41969985.0,
      "step": 21455
    },
    {
      "entropy": 6.102953481674194,
      "epoch": 2.1451491977807766,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004542502986230649,
      "loss": 5.6791,
      "mean_token_accuracy": 0.1699275329709053,
      "num_tokens": 41980156.0,
      "step": 21460
    },
    {
      "entropy": 6.093511724472046,
      "epoch": 2.1456490228420053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004542287252626295,
      "loss": 5.7794,
      "mean_token_accuracy": 0.17191445529460908,
      "num_tokens": 41990416.0,
      "step": 21465
    },
    {
      "entropy": 6.086202716827392,
      "epoch": 2.146148847903234,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004542071473928455,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16503936499357225,
      "num_tokens": 42000065.0,
      "step": 21470
    },
    {
      "entropy": 6.12524380683899,
      "epoch": 2.1466486729644623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004541855650142557,
      "loss": 5.6544,
      "mean_token_accuracy": 0.17495515644550325,
      "num_tokens": 42010683.0,
      "step": 21475
    },
    {
      "entropy": 6.058239698410034,
      "epoch": 2.147148498025691,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045416397812740323,
      "loss": 5.6924,
      "mean_token_accuracy": 0.17533515244722367,
      "num_tokens": 42020191.0,
      "step": 21480
    },
    {
      "entropy": 6.075342512130737,
      "epoch": 2.1476483230869197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004541423867328312,
      "loss": 5.7161,
      "mean_token_accuracy": 0.16611202508211137,
      "num_tokens": 42029875.0,
      "step": 21485
    },
    {
      "entropy": 6.053252172470093,
      "epoch": 2.148148148148148,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004541207908310829,
      "loss": 5.6663,
      "mean_token_accuracy": 0.17682785540819168,
      "num_tokens": 42040265.0,
      "step": 21490
    },
    {
      "entropy": 6.0104998588562015,
      "epoch": 2.1486479732093766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004540991904227015,
      "loss": 5.5721,
      "mean_token_accuracy": 0.1850965440273285,
      "num_tokens": 42050370.0,
      "step": 21495
    },
    {
      "entropy": 6.0474776268005375,
      "epoch": 2.1491477982706053,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045407758550823055,
      "loss": 5.7013,
      "mean_token_accuracy": 0.1755475714802742,
      "num_tokens": 42059865.0,
      "step": 21500
    },
    {
      "entropy": 6.018110752105713,
      "epoch": 2.149647623331834,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00045405597608821365,
      "loss": 5.7051,
      "mean_token_accuracy": 0.17302850931882857,
      "num_tokens": 42071973.0,
      "step": 21505
    },
    {
      "entropy": 6.107784175872803,
      "epoch": 2.1501474483930623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004540343621631944,
      "loss": 5.6875,
      "mean_token_accuracy": 0.17684863209724427,
      "num_tokens": 42082810.0,
      "step": 21510
    },
    {
      "entropy": 6.116256666183472,
      "epoch": 2.150647273454291,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004540127437337167,
      "loss": 5.6821,
      "mean_token_accuracy": 0.17214419841766357,
      "num_tokens": 42092644.0,
      "step": 21515
    },
    {
      "entropy": 5.973757982254028,
      "epoch": 2.1511470985155197,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004539911208003245,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1764889433979988,
      "num_tokens": 42103393.0,
      "step": 21520
    },
    {
      "entropy": 5.9138304710388185,
      "epoch": 2.151646923576748,
      "grad_norm": 1.5,
      "learning_rate": 0.0004539694933635617,
      "loss": 5.6315,
      "mean_token_accuracy": 0.17131459563970566,
      "num_tokens": 42113759.0,
      "step": 21525
    },
    {
      "entropy": 6.046679210662842,
      "epoch": 2.1521467486379766,
      "grad_norm": 1.5,
      "learning_rate": 0.00045394786142397253,
      "loss": 5.7031,
      "mean_token_accuracy": 0.16836813390254973,
      "num_tokens": 42124092.0,
      "step": 21530
    },
    {
      "entropy": 5.9762450695037845,
      "epoch": 2.1526465736992053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004539262249821012,
      "loss": 5.5421,
      "mean_token_accuracy": 0.18571716994047166,
      "num_tokens": 42132750.0,
      "step": 21535
    },
    {
      "entropy": 6.056936264038086,
      "epoch": 2.153146398760434,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045390458403849203,
      "loss": 5.6932,
      "mean_token_accuracy": 0.1749413266777992,
      "num_tokens": 42141766.0,
      "step": 21540
    },
    {
      "entropy": 6.121698999404908,
      "epoch": 2.1536462238216623,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004538829385936896,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16371053606271743,
      "num_tokens": 42152057.0,
      "step": 21545
    },
    {
      "entropy": 6.027368497848511,
      "epoch": 2.154146048882891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004538612886482385,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1740359142422676,
      "num_tokens": 42162414.0,
      "step": 21550
    },
    {
      "entropy": 6.094902038574219,
      "epoch": 2.1546458739441197,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004538396342026834,
      "loss": 5.6773,
      "mean_token_accuracy": 0.16994105726480485,
      "num_tokens": 42172217.0,
      "step": 21555
    },
    {
      "entropy": 6.072580575942993,
      "epoch": 2.155145699005348,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004538179752575691,
      "loss": 5.7114,
      "mean_token_accuracy": 0.16411114931106568,
      "num_tokens": 42182267.0,
      "step": 21560
    },
    {
      "entropy": 6.047468900680542,
      "epoch": 2.1556455240665766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004537963118134406,
      "loss": 5.7656,
      "mean_token_accuracy": 0.1765565648674965,
      "num_tokens": 42192279.0,
      "step": 21565
    },
    {
      "entropy": 6.16641058921814,
      "epoch": 2.1561453491278053,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045377464387084287,
      "loss": 5.8276,
      "mean_token_accuracy": 0.16476137340068817,
      "num_tokens": 42202209.0,
      "step": 21570
    },
    {
      "entropy": 6.123558616638183,
      "epoch": 2.156645174189034,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004537529714303211,
      "loss": 5.7841,
      "mean_token_accuracy": 0.16599818021059037,
      "num_tokens": 42212378.0,
      "step": 21575
    },
    {
      "entropy": 6.045112991333008,
      "epoch": 2.1571449992502623,
      "grad_norm": 1.25,
      "learning_rate": 0.00045373129449242054,
      "loss": 5.7052,
      "mean_token_accuracy": 0.17154521644115447,
      "num_tokens": 42222725.0,
      "step": 21580
    },
    {
      "entropy": 5.994068384170532,
      "epoch": 2.157644824311491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004537096130576866,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17128431051969528,
      "num_tokens": 42233510.0,
      "step": 21585
    },
    {
      "entropy": 6.0925334930419925,
      "epoch": 2.1581446493727197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045368792712666486,
      "loss": 5.7704,
      "mean_token_accuracy": 0.16953141540288924,
      "num_tokens": 42242589.0,
      "step": 21590
    },
    {
      "entropy": 6.097486448287964,
      "epoch": 2.158644474433948,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004536662366999008,
      "loss": 5.6325,
      "mean_token_accuracy": 0.16861331164836885,
      "num_tokens": 42251663.0,
      "step": 21595
    },
    {
      "entropy": 6.078781700134277,
      "epoch": 2.1591442994951766,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004536445417779402,
      "loss": 5.6262,
      "mean_token_accuracy": 0.1790884330868721,
      "num_tokens": 42260507.0,
      "step": 21600
    },
    {
      "entropy": 6.0428753852844235,
      "epoch": 2.1596441245564053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045362284236132895,
      "loss": 5.7229,
      "mean_token_accuracy": 0.17202132046222687,
      "num_tokens": 42270967.0,
      "step": 21605
    },
    {
      "entropy": 6.025993394851684,
      "epoch": 2.160143949617634,
      "grad_norm": 1.25,
      "learning_rate": 0.00045360113845061283,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17342241555452348,
      "num_tokens": 42280418.0,
      "step": 21610
    },
    {
      "entropy": 6.0390305519104,
      "epoch": 2.1606437746788623,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045357943004633813,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1706431746482849,
      "num_tokens": 42290371.0,
      "step": 21615
    },
    {
      "entropy": 6.007556438446045,
      "epoch": 2.161143599740091,
      "grad_norm": 1.8046875,
      "learning_rate": 0.00045355771714905086,
      "loss": 5.6515,
      "mean_token_accuracy": 0.1736885502934456,
      "num_tokens": 42300259.0,
      "step": 21620
    },
    {
      "entropy": 6.036654329299926,
      "epoch": 2.1616434248013197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045353599975929746,
      "loss": 5.669,
      "mean_token_accuracy": 0.1727564215660095,
      "num_tokens": 42309817.0,
      "step": 21625
    },
    {
      "entropy": 6.128452110290527,
      "epoch": 2.162143249862548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004535142778776241,
      "loss": 5.7658,
      "mean_token_accuracy": 0.16774977892637252,
      "num_tokens": 42319231.0,
      "step": 21630
    },
    {
      "entropy": 6.014666652679443,
      "epoch": 2.1626430749237766,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004534925515045776,
      "loss": 5.6401,
      "mean_token_accuracy": 0.17607091069221498,
      "num_tokens": 42328943.0,
      "step": 21635
    },
    {
      "entropy": 6.0321050643920895,
      "epoch": 2.1631428999850053,
      "grad_norm": 1.25,
      "learning_rate": 0.00045347082064070435,
      "loss": 5.6484,
      "mean_token_accuracy": 0.17371224313974382,
      "num_tokens": 42337938.0,
      "step": 21640
    },
    {
      "entropy": 6.080334949493408,
      "epoch": 2.163642725046234,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004534490852865512,
      "loss": 5.76,
      "mean_token_accuracy": 0.17043631374835969,
      "num_tokens": 42348906.0,
      "step": 21645
    },
    {
      "entropy": 6.11230878829956,
      "epoch": 2.1641425501074623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045342734544266493,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17258514016866683,
      "num_tokens": 42359026.0,
      "step": 21650
    },
    {
      "entropy": 6.140669870376587,
      "epoch": 2.164642375168691,
      "grad_norm": 1.25,
      "learning_rate": 0.0004534056011095926,
      "loss": 5.773,
      "mean_token_accuracy": 0.1680891051888466,
      "num_tokens": 42368697.0,
      "step": 21655
    },
    {
      "entropy": 6.06164779663086,
      "epoch": 2.1651422002299197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045338385228788124,
      "loss": 5.6267,
      "mean_token_accuracy": 0.17672697752714156,
      "num_tokens": 42377961.0,
      "step": 21660
    },
    {
      "entropy": 6.06563367843628,
      "epoch": 2.165642025291148,
      "grad_norm": 1.828125,
      "learning_rate": 0.00045336209897807813,
      "loss": 5.682,
      "mean_token_accuracy": 0.1685069441795349,
      "num_tokens": 42388219.0,
      "step": 21665
    },
    {
      "entropy": 6.044868516921997,
      "epoch": 2.1661418503523766,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045334034118073047,
      "loss": 5.7167,
      "mean_token_accuracy": 0.1743968427181244,
      "num_tokens": 42397626.0,
      "step": 21670
    },
    {
      "entropy": 6.151108884811402,
      "epoch": 2.1666416754136053,
      "grad_norm": 1.53125,
      "learning_rate": 0.00045331857889638566,
      "loss": 5.7265,
      "mean_token_accuracy": 0.16495366171002387,
      "num_tokens": 42406806.0,
      "step": 21675
    },
    {
      "entropy": 5.965486288070679,
      "epoch": 2.167141500474834,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004532968121255914,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1877909004688263,
      "num_tokens": 42416544.0,
      "step": 21680
    },
    {
      "entropy": 6.010894775390625,
      "epoch": 2.1676413255360623,
      "grad_norm": 1.375,
      "learning_rate": 0.00045327504086889513,
      "loss": 5.6702,
      "mean_token_accuracy": 0.16993131190538407,
      "num_tokens": 42426447.0,
      "step": 21685
    },
    {
      "entropy": 6.077881240844727,
      "epoch": 2.168141150597291,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045325326512684466,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16515269577503205,
      "num_tokens": 42435411.0,
      "step": 21690
    },
    {
      "entropy": 6.103131055831909,
      "epoch": 2.1686409756585197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045323148489998807,
      "loss": 5.7201,
      "mean_token_accuracy": 0.1616802141070366,
      "num_tokens": 42445280.0,
      "step": 21695
    },
    {
      "entropy": 6.074206304550171,
      "epoch": 2.169140800719748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000453209700188873,
      "loss": 5.6515,
      "mean_token_accuracy": 0.17444013804197311,
      "num_tokens": 42455182.0,
      "step": 21700
    },
    {
      "entropy": 6.0931610584259035,
      "epoch": 2.1696406257809766,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045318791099404785,
      "loss": 5.7713,
      "mean_token_accuracy": 0.16296152472496034,
      "num_tokens": 42464492.0,
      "step": 21705
    },
    {
      "entropy": 6.097999811172485,
      "epoch": 2.1701404508422053,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045316611731606075,
      "loss": 5.7774,
      "mean_token_accuracy": 0.1632525622844696,
      "num_tokens": 42475461.0,
      "step": 21710
    },
    {
      "entropy": 6.0696868896484375,
      "epoch": 2.170640275903434,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004531443191554598,
      "loss": 5.7201,
      "mean_token_accuracy": 0.1755180314183235,
      "num_tokens": 42484383.0,
      "step": 21715
    },
    {
      "entropy": 6.07525315284729,
      "epoch": 2.1711401009646623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004531225165127938,
      "loss": 5.7262,
      "mean_token_accuracy": 0.17382334768772126,
      "num_tokens": 42494268.0,
      "step": 21720
    },
    {
      "entropy": 6.1246593475341795,
      "epoch": 2.171639926025891,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00045310070938861107,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16650617569684983,
      "num_tokens": 42503035.0,
      "step": 21725
    },
    {
      "entropy": 5.992408227920532,
      "epoch": 2.1721397510871197,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045307889778346035,
      "loss": 5.5864,
      "mean_token_accuracy": 0.18432072550058365,
      "num_tokens": 42512205.0,
      "step": 21730
    },
    {
      "entropy": 6.099648523330688,
      "epoch": 2.172639576148348,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004530570816978903,
      "loss": 5.7638,
      "mean_token_accuracy": 0.16651995182037355,
      "num_tokens": 42523153.0,
      "step": 21735
    },
    {
      "entropy": 6.171558094024658,
      "epoch": 2.1731394012095766,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045303526113245,
      "loss": 5.7386,
      "mean_token_accuracy": 0.1703706905245781,
      "num_tokens": 42532457.0,
      "step": 21740
    },
    {
      "entropy": 6.069806289672852,
      "epoch": 2.1736392262708053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004530134360876882,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17562314867973328,
      "num_tokens": 42542179.0,
      "step": 21745
    },
    {
      "entropy": 6.0285979270935055,
      "epoch": 2.174139051332034,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004529916065641543,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1715813860297203,
      "num_tokens": 42552441.0,
      "step": 21750
    },
    {
      "entropy": 6.014201068878174,
      "epoch": 2.1746388763932623,
      "grad_norm": 1.25,
      "learning_rate": 0.0004529697725623973,
      "loss": 5.571,
      "mean_token_accuracy": 0.1822548270225525,
      "num_tokens": 42562600.0,
      "step": 21755
    },
    {
      "entropy": 6.118940925598144,
      "epoch": 2.175138701454491,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045294793408296656,
      "loss": 5.6647,
      "mean_token_accuracy": 0.1747439309954643,
      "num_tokens": 42571790.0,
      "step": 21760
    },
    {
      "entropy": 6.061412715911866,
      "epoch": 2.1756385265157197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004529260911264117,
      "loss": 5.6706,
      "mean_token_accuracy": 0.18529340177774428,
      "num_tokens": 42581816.0,
      "step": 21765
    },
    {
      "entropy": 6.026689100265503,
      "epoch": 2.176138351576948,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004529042436932822,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1767548516392708,
      "num_tokens": 42591501.0,
      "step": 21770
    },
    {
      "entropy": 6.091596460342407,
      "epoch": 2.1766381766381766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045288239178412754,
      "loss": 5.6203,
      "mean_token_accuracy": 0.1797621801495552,
      "num_tokens": 42601237.0,
      "step": 21775
    },
    {
      "entropy": 5.989274740219116,
      "epoch": 2.1771380016994053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004528605353994978,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17625731378793716,
      "num_tokens": 42611239.0,
      "step": 21780
    },
    {
      "entropy": 6.074346542358398,
      "epoch": 2.177637826760634,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004528386745399428,
      "loss": 5.6922,
      "mean_token_accuracy": 0.17732557207345961,
      "num_tokens": 42621153.0,
      "step": 21785
    },
    {
      "entropy": 6.096200370788575,
      "epoch": 2.1781376518218623,
      "grad_norm": 1.078125,
      "learning_rate": 0.00045281680920601235,
      "loss": 5.6936,
      "mean_token_accuracy": 0.17341695427894593,
      "num_tokens": 42631508.0,
      "step": 21790
    },
    {
      "entropy": 6.105245923995971,
      "epoch": 2.178637476883091,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004527949393982569,
      "loss": 5.712,
      "mean_token_accuracy": 0.1670197382569313,
      "num_tokens": 42641401.0,
      "step": 21795
    },
    {
      "entropy": 6.082239723205566,
      "epoch": 2.1791373019443196,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045277306511722645,
      "loss": 5.7209,
      "mean_token_accuracy": 0.16531577184796334,
      "num_tokens": 42651631.0,
      "step": 21800
    },
    {
      "entropy": 6.041162204742432,
      "epoch": 2.179637127005548,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004527511863634714,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17460914850234985,
      "num_tokens": 42661506.0,
      "step": 21805
    },
    {
      "entropy": 6.033502769470215,
      "epoch": 2.1801369520667766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004527293031375422,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17212567031383513,
      "num_tokens": 42671403.0,
      "step": 21810
    },
    {
      "entropy": 6.10186996459961,
      "epoch": 2.1806367771280053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004527074154399896,
      "loss": 5.8009,
      "mean_token_accuracy": 0.16897159367799758,
      "num_tokens": 42681360.0,
      "step": 21815
    },
    {
      "entropy": 6.055552339553833,
      "epoch": 2.181136602189234,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045268552327136415,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17467959225177765,
      "num_tokens": 42691381.0,
      "step": 21820
    },
    {
      "entropy": 6.07011342048645,
      "epoch": 2.1816364272504623,
      "grad_norm": 1.375,
      "learning_rate": 0.00045266362663221656,
      "loss": 5.7385,
      "mean_token_accuracy": 0.17270308881998062,
      "num_tokens": 42701857.0,
      "step": 21825
    },
    {
      "entropy": 6.144731616973877,
      "epoch": 2.182136252311691,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045264172552309786,
      "loss": 5.7315,
      "mean_token_accuracy": 0.16550954207777976,
      "num_tokens": 42711526.0,
      "step": 21830
    },
    {
      "entropy": 6.11213264465332,
      "epoch": 2.1826360773729196,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004526198199445591,
      "loss": 5.6869,
      "mean_token_accuracy": 0.16968348920345305,
      "num_tokens": 42721552.0,
      "step": 21835
    },
    {
      "entropy": 6.053549194335938,
      "epoch": 2.183135902434148,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045259790989715133,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17522703111171722,
      "num_tokens": 42732161.0,
      "step": 21840
    },
    {
      "entropy": 6.042395544052124,
      "epoch": 2.1836357274953766,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045257599538142587,
      "loss": 5.7299,
      "mean_token_accuracy": 0.1716204524040222,
      "num_tokens": 42741140.0,
      "step": 21845
    },
    {
      "entropy": 6.061429405212403,
      "epoch": 2.1841355525566053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045255407639793407,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17254729568958282,
      "num_tokens": 42750000.0,
      "step": 21850
    },
    {
      "entropy": 6.061018323898315,
      "epoch": 2.1846353776178336,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004525321529472274,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17104449421167373,
      "num_tokens": 42760430.0,
      "step": 21855
    },
    {
      "entropy": 6.055662727355957,
      "epoch": 2.1851352026790622,
      "grad_norm": 1.25,
      "learning_rate": 0.00045251022502985737,
      "loss": 5.6987,
      "mean_token_accuracy": 0.17426712661981583,
      "num_tokens": 42770904.0,
      "step": 21860
    },
    {
      "entropy": 6.068159484863282,
      "epoch": 2.185635027740291,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004524882926463758,
      "loss": 5.7353,
      "mean_token_accuracy": 0.16607723832130433,
      "num_tokens": 42781297.0,
      "step": 21865
    },
    {
      "entropy": 6.047343158721924,
      "epoch": 2.1861348528015196,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004524663557973345,
      "loss": 5.7278,
      "mean_token_accuracy": 0.17435943484306335,
      "num_tokens": 42790781.0,
      "step": 21870
    },
    {
      "entropy": 6.0649292945861815,
      "epoch": 2.186634677862748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045244441448328535,
      "loss": 5.7151,
      "mean_token_accuracy": 0.16528489142656327,
      "num_tokens": 42800387.0,
      "step": 21875
    },
    {
      "entropy": 6.121155834197998,
      "epoch": 2.1871345029239766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045242246870478035,
      "loss": 5.7752,
      "mean_token_accuracy": 0.16461876183748245,
      "num_tokens": 42809204.0,
      "step": 21880
    },
    {
      "entropy": 6.100079774856567,
      "epoch": 2.1876343279852053,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004524005184623717,
      "loss": 5.7565,
      "mean_token_accuracy": 0.17068672478199004,
      "num_tokens": 42819431.0,
      "step": 21885
    },
    {
      "entropy": 5.942271184921265,
      "epoch": 2.1881341530464335,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004523785637566117,
      "loss": 5.5693,
      "mean_token_accuracy": 0.18336160928010942,
      "num_tokens": 42830072.0,
      "step": 21890
    },
    {
      "entropy": 6.018280839920044,
      "epoch": 2.1886339781076622,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004523566045880527,
      "loss": 5.6894,
      "mean_token_accuracy": 0.16862375438213348,
      "num_tokens": 42840201.0,
      "step": 21895
    },
    {
      "entropy": 5.991559743881226,
      "epoch": 2.189133803168891,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045233464095724706,
      "loss": 5.6464,
      "mean_token_accuracy": 0.17447273433208466,
      "num_tokens": 42849695.0,
      "step": 21900
    },
    {
      "entropy": 6.125292253494263,
      "epoch": 2.1896336282301196,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045231267286474755,
      "loss": 5.7768,
      "mean_token_accuracy": 0.1702263042330742,
      "num_tokens": 42858674.0,
      "step": 21905
    },
    {
      "entropy": 6.102081680297852,
      "epoch": 2.190133453291348,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004522907003111068,
      "loss": 5.7291,
      "mean_token_accuracy": 0.17314068153500556,
      "num_tokens": 42869879.0,
      "step": 21910
    },
    {
      "entropy": 6.036731719970703,
      "epoch": 2.1906332783525766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004522687232968776,
      "loss": 5.7032,
      "mean_token_accuracy": 0.168434077501297,
      "num_tokens": 42879652.0,
      "step": 21915
    },
    {
      "entropy": 6.023220682144165,
      "epoch": 2.1911331034138053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045224674182261306,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17858026921749115,
      "num_tokens": 42889081.0,
      "step": 21920
    },
    {
      "entropy": 6.080866575241089,
      "epoch": 2.1916329284750335,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045222475588886597,
      "loss": 5.6747,
      "mean_token_accuracy": 0.17574896812438964,
      "num_tokens": 42899621.0,
      "step": 21925
    },
    {
      "entropy": 6.047423505783081,
      "epoch": 2.1921327535362622,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045220276549618973,
      "loss": 5.775,
      "mean_token_accuracy": 0.1685224875807762,
      "num_tokens": 42910722.0,
      "step": 21930
    },
    {
      "entropy": 6.091746187210083,
      "epoch": 2.192632578597491,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045218077064513744,
      "loss": 5.7686,
      "mean_token_accuracy": 0.16880830526351928,
      "num_tokens": 42920788.0,
      "step": 21935
    },
    {
      "entropy": 6.036105012893676,
      "epoch": 2.1931324036587196,
      "grad_norm": 1.25,
      "learning_rate": 0.0004521587713362626,
      "loss": 5.6989,
      "mean_token_accuracy": 0.1733218476176262,
      "num_tokens": 42930183.0,
      "step": 21940
    },
    {
      "entropy": 6.041737031936646,
      "epoch": 2.193632228719948,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004521367675701187,
      "loss": 5.6798,
      "mean_token_accuracy": 0.17645209282636642,
      "num_tokens": 42939886.0,
      "step": 21945
    },
    {
      "entropy": 6.096748399734497,
      "epoch": 2.1941320537811766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004521147593472592,
      "loss": 5.7094,
      "mean_token_accuracy": 0.1677560806274414,
      "num_tokens": 42948756.0,
      "step": 21950
    },
    {
      "entropy": 6.0517213344573975,
      "epoch": 2.1946318788424053,
      "grad_norm": 1.25,
      "learning_rate": 0.00045209274666823795,
      "loss": 5.7511,
      "mean_token_accuracy": 0.17416853010654448,
      "num_tokens": 42957540.0,
      "step": 21955
    },
    {
      "entropy": 6.117839097976685,
      "epoch": 2.1951317039036335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004520707295336088,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17202810943126678,
      "num_tokens": 42967628.0,
      "step": 21960
    },
    {
      "entropy": 6.128359985351563,
      "epoch": 2.1956315289648622,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045204870794392574,
      "loss": 5.7433,
      "mean_token_accuracy": 0.16954494416713714,
      "num_tokens": 42975850.0,
      "step": 21965
    },
    {
      "entropy": 6.039148998260498,
      "epoch": 2.196131354026091,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045202668189974263,
      "loss": 5.7905,
      "mean_token_accuracy": 0.16320885270833968,
      "num_tokens": 42986640.0,
      "step": 21970
    },
    {
      "entropy": 6.025380706787109,
      "epoch": 2.1966311790873196,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004520046514016138,
      "loss": 5.5804,
      "mean_token_accuracy": 0.18114374727010726,
      "num_tokens": 42995876.0,
      "step": 21975
    },
    {
      "entropy": 6.019872617721558,
      "epoch": 2.197131004148548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045198261645009347,
      "loss": 5.6589,
      "mean_token_accuracy": 0.17336872667074205,
      "num_tokens": 43005843.0,
      "step": 21980
    },
    {
      "entropy": 6.080774164199829,
      "epoch": 2.1976308292097766,
      "grad_norm": 1.296875,
      "learning_rate": 0.000451960577045736,
      "loss": 5.7356,
      "mean_token_accuracy": 0.16359950602054596,
      "num_tokens": 43015866.0,
      "step": 21985
    },
    {
      "entropy": 6.043430662155151,
      "epoch": 2.1981306542710053,
      "grad_norm": 1.21875,
      "learning_rate": 0.000451938533189096,
      "loss": 5.6926,
      "mean_token_accuracy": 0.17309972196817397,
      "num_tokens": 43025669.0,
      "step": 21990
    },
    {
      "entropy": 6.086376714706421,
      "epoch": 2.1986304793322335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004519164848807281,
      "loss": 5.7517,
      "mean_token_accuracy": 0.16561348140239715,
      "num_tokens": 43036120.0,
      "step": 21995
    },
    {
      "entropy": 6.138607311248779,
      "epoch": 2.1991303043934622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045189443212118686,
      "loss": 5.7785,
      "mean_token_accuracy": 0.1683248072862625,
      "num_tokens": 43045799.0,
      "step": 22000
    },
    {
      "entropy": 6.003041172027588,
      "epoch": 2.199630129454691,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045187237491102727,
      "loss": 5.6308,
      "mean_token_accuracy": 0.17727708518505098,
      "num_tokens": 43055341.0,
      "step": 22005
    },
    {
      "entropy": 6.02529616355896,
      "epoch": 2.200129954515919,
      "grad_norm": 1.25,
      "learning_rate": 0.00045185031325080426,
      "loss": 5.7555,
      "mean_token_accuracy": 0.16980350017547607,
      "num_tokens": 43064899.0,
      "step": 22010
    },
    {
      "entropy": 6.04108510017395,
      "epoch": 2.200629779577148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045182824714107284,
      "loss": 5.6179,
      "mean_token_accuracy": 0.17996545732021332,
      "num_tokens": 43074789.0,
      "step": 22015
    },
    {
      "entropy": 6.1798858642578125,
      "epoch": 2.2011296046383766,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045180617658238827,
      "loss": 5.7384,
      "mean_token_accuracy": 0.17071097642183303,
      "num_tokens": 43084107.0,
      "step": 22020
    },
    {
      "entropy": 6.094903135299683,
      "epoch": 2.2016294296996053,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004517841015753057,
      "loss": 5.67,
      "mean_token_accuracy": 0.17519693076610565,
      "num_tokens": 43093120.0,
      "step": 22025
    },
    {
      "entropy": 5.9775059700012205,
      "epoch": 2.2021292547608335,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004517620221203807,
      "loss": 5.7377,
      "mean_token_accuracy": 0.17057846933603288,
      "num_tokens": 43103251.0,
      "step": 22030
    },
    {
      "entropy": 6.072139549255371,
      "epoch": 2.2026290798220622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045173993821816873,
      "loss": 5.7027,
      "mean_token_accuracy": 0.1694188728928566,
      "num_tokens": 43113162.0,
      "step": 22035
    },
    {
      "entropy": 6.079460525512696,
      "epoch": 2.203128904883291,
      "grad_norm": 1.25,
      "learning_rate": 0.0004517178498692253,
      "loss": 5.6903,
      "mean_token_accuracy": 0.17559992223978044,
      "num_tokens": 43122170.0,
      "step": 22040
    },
    {
      "entropy": 6.129349040985107,
      "epoch": 2.203628729944519,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045169575707410637,
      "loss": 5.7141,
      "mean_token_accuracy": 0.16435018181800842,
      "num_tokens": 43132271.0,
      "step": 22045
    },
    {
      "entropy": 6.065129613876342,
      "epoch": 2.204128555005748,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045167365983336756,
      "loss": 5.6459,
      "mean_token_accuracy": 0.17564557045698165,
      "num_tokens": 43141593.0,
      "step": 22050
    },
    {
      "entropy": 6.002485513687134,
      "epoch": 2.2046283800669766,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045165155814756497,
      "loss": 5.6891,
      "mean_token_accuracy": 0.16883871257305144,
      "num_tokens": 43150659.0,
      "step": 22055
    },
    {
      "entropy": 6.075923871994019,
      "epoch": 2.2051282051282053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045162945201725457,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17701226770877837,
      "num_tokens": 43160575.0,
      "step": 22060
    },
    {
      "entropy": 6.148479270935058,
      "epoch": 2.2056280301894335,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004516073414429927,
      "loss": 5.7068,
      "mean_token_accuracy": 0.1684955731034279,
      "num_tokens": 43169703.0,
      "step": 22065
    },
    {
      "entropy": 6.1927244663238525,
      "epoch": 2.2061278552506622,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004515852264253356,
      "loss": 5.8133,
      "mean_token_accuracy": 0.1619138926267624,
      "num_tokens": 43180685.0,
      "step": 22070
    },
    {
      "entropy": 6.015853500366211,
      "epoch": 2.206627680311891,
      "grad_norm": 1.25,
      "learning_rate": 0.00045156310696483953,
      "loss": 5.6234,
      "mean_token_accuracy": 0.181362783908844,
      "num_tokens": 43190398.0,
      "step": 22075
    },
    {
      "entropy": 6.076595687866211,
      "epoch": 2.207127505373119,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004515409830620612,
      "loss": 5.8126,
      "mean_token_accuracy": 0.16640556305646897,
      "num_tokens": 43200867.0,
      "step": 22080
    },
    {
      "entropy": 6.132361507415771,
      "epoch": 2.207627330434348,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004515188547175571,
      "loss": 5.7886,
      "mean_token_accuracy": 0.16479425132274628,
      "num_tokens": 43211236.0,
      "step": 22085
    },
    {
      "entropy": 6.169962978363037,
      "epoch": 2.2081271554955766,
      "grad_norm": 1.234375,
      "learning_rate": 0.000451496721931884,
      "loss": 5.8498,
      "mean_token_accuracy": 0.16564706861972808,
      "num_tokens": 43220713.0,
      "step": 22090
    },
    {
      "entropy": 6.061186265945435,
      "epoch": 2.2086269805568053,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004514745847055989,
      "loss": 5.6953,
      "mean_token_accuracy": 0.1718796655535698,
      "num_tokens": 43231258.0,
      "step": 22095
    },
    {
      "entropy": 6.10233359336853,
      "epoch": 2.2091268056180335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004514524430392586,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17494950890541078,
      "num_tokens": 43240523.0,
      "step": 22100
    },
    {
      "entropy": 6.0847245216369625,
      "epoch": 2.2096266306792622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004514302969334202,
      "loss": 5.8107,
      "mean_token_accuracy": 0.1699256643652916,
      "num_tokens": 43250044.0,
      "step": 22105
    },
    {
      "entropy": 6.089792108535766,
      "epoch": 2.210126455740491,
      "grad_norm": 1.375,
      "learning_rate": 0.00045140814638864105,
      "loss": 5.7273,
      "mean_token_accuracy": 0.1691812351346016,
      "num_tokens": 43260393.0,
      "step": 22110
    },
    {
      "entropy": 6.0787981986999515,
      "epoch": 2.210626280801719,
      "grad_norm": 1.25,
      "learning_rate": 0.0004513859914054782,
      "loss": 5.6442,
      "mean_token_accuracy": 0.18110208660364152,
      "num_tokens": 43270554.0,
      "step": 22115
    },
    {
      "entropy": 6.010758638381958,
      "epoch": 2.211126105862948,
      "grad_norm": 1.25,
      "learning_rate": 0.00045136383198448926,
      "loss": 5.5966,
      "mean_token_accuracy": 0.17571573257446288,
      "num_tokens": 43280256.0,
      "step": 22120
    },
    {
      "entropy": 6.066720342636108,
      "epoch": 2.2116259309241766,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004513416681262317,
      "loss": 5.7258,
      "mean_token_accuracy": 0.17462086975574492,
      "num_tokens": 43289399.0,
      "step": 22125
    },
    {
      "entropy": 6.111040735244751,
      "epoch": 2.2121257559854053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004513194998312631,
      "loss": 5.7634,
      "mean_token_accuracy": 0.1696481555700302,
      "num_tokens": 43299349.0,
      "step": 22130
    },
    {
      "entropy": 6.02924075126648,
      "epoch": 2.2126255810466335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004512973271001413,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17870509326457978,
      "num_tokens": 43309199.0,
      "step": 22135
    },
    {
      "entropy": 6.01779613494873,
      "epoch": 2.2131254061078622,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045127514993342413,
      "loss": 5.6829,
      "mean_token_accuracy": 0.18165641874074936,
      "num_tokens": 43319637.0,
      "step": 22140
    },
    {
      "entropy": 6.034041833877564,
      "epoch": 2.213625231169091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004512529683316695,
      "loss": 5.6681,
      "mean_token_accuracy": 0.16943552792072297,
      "num_tokens": 43329005.0,
      "step": 22145
    },
    {
      "entropy": 6.005779790878296,
      "epoch": 2.214125056230319,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045123078229543557,
      "loss": 5.6016,
      "mean_token_accuracy": 0.18037374168634415,
      "num_tokens": 43339299.0,
      "step": 22150
    },
    {
      "entropy": 6.02998194694519,
      "epoch": 2.214624881291548,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004512085918252805,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1728320986032486,
      "num_tokens": 43350317.0,
      "step": 22155
    },
    {
      "entropy": 6.071431970596313,
      "epoch": 2.2151247063527766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004511863969217625,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17422719895839692,
      "num_tokens": 43360607.0,
      "step": 22160
    },
    {
      "entropy": 6.10122127532959,
      "epoch": 2.2156245314140053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045116419758544024,
      "loss": 5.7285,
      "mean_token_accuracy": 0.16648525148630142,
      "num_tokens": 43370622.0,
      "step": 22165
    },
    {
      "entropy": 6.049825382232666,
      "epoch": 2.2161243564752335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045114199381687207,
      "loss": 5.6952,
      "mean_token_accuracy": 0.17655676752328872,
      "num_tokens": 43379841.0,
      "step": 22170
    },
    {
      "entropy": 5.9938160419464115,
      "epoch": 2.2166241815364622,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004511197856166166,
      "loss": 5.614,
      "mean_token_accuracy": 0.17537855952978135,
      "num_tokens": 43389575.0,
      "step": 22175
    },
    {
      "entropy": 6.070791816711425,
      "epoch": 2.217124006597691,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004510975729852327,
      "loss": 5.6669,
      "mean_token_accuracy": 0.17679303586483003,
      "num_tokens": 43399686.0,
      "step": 22180
    },
    {
      "entropy": 6.078019332885742,
      "epoch": 2.217623831658919,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004510753559232792,
      "loss": 5.7673,
      "mean_token_accuracy": 0.16574058383703233,
      "num_tokens": 43408918.0,
      "step": 22185
    },
    {
      "entropy": 6.108441400527954,
      "epoch": 2.218123656720148,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045105313443131496,
      "loss": 5.7882,
      "mean_token_accuracy": 0.16797700226306916,
      "num_tokens": 43419285.0,
      "step": 22190
    },
    {
      "entropy": 6.048844289779663,
      "epoch": 2.2186234817813766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004510309085098992,
      "loss": 5.6184,
      "mean_token_accuracy": 0.1789767175912857,
      "num_tokens": 43428909.0,
      "step": 22195
    },
    {
      "entropy": 6.050942087173462,
      "epoch": 2.2191233068426053,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045100867815959107,
      "loss": 5.6724,
      "mean_token_accuracy": 0.16924350559711457,
      "num_tokens": 43438092.0,
      "step": 22200
    },
    {
      "entropy": 5.913561248779297,
      "epoch": 2.2196231319038335,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004509864433809499,
      "loss": 5.5251,
      "mean_token_accuracy": 0.18366561383008956,
      "num_tokens": 43447566.0,
      "step": 22205
    },
    {
      "entropy": 6.065630483627319,
      "epoch": 2.2201229569650622,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004509642041745351,
      "loss": 5.6852,
      "mean_token_accuracy": 0.17648178935050965,
      "num_tokens": 43457765.0,
      "step": 22210
    },
    {
      "entropy": 6.067888402938843,
      "epoch": 2.220622782026291,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004509419605409062,
      "loss": 5.5658,
      "mean_token_accuracy": 0.18134808987379075,
      "num_tokens": 43467664.0,
      "step": 22215
    },
    {
      "entropy": 6.080079174041748,
      "epoch": 2.221122607087519,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004509197124806229,
      "loss": 5.7651,
      "mean_token_accuracy": 0.16917336583137513,
      "num_tokens": 43477732.0,
      "step": 22220
    },
    {
      "entropy": 5.95259370803833,
      "epoch": 2.221622432148748,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045089745999424486,
      "loss": 5.6885,
      "mean_token_accuracy": 0.17906248271465303,
      "num_tokens": 43487005.0,
      "step": 22225
    },
    {
      "entropy": 6.010063552856446,
      "epoch": 2.2221222572099766,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000450875203082332,
      "loss": 5.679,
      "mean_token_accuracy": 0.17322757691144944,
      "num_tokens": 43496167.0,
      "step": 22230
    },
    {
      "entropy": 6.116608619689941,
      "epoch": 2.2226220822712053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004508529417454442,
      "loss": 5.6724,
      "mean_token_accuracy": 0.17614705562591554,
      "num_tokens": 43506464.0,
      "step": 22235
    },
    {
      "entropy": 6.0567319869995115,
      "epoch": 2.2231219073324335,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045083067598414176,
      "loss": 5.6535,
      "mean_token_accuracy": 0.16819214522838594,
      "num_tokens": 43516554.0,
      "step": 22240
    },
    {
      "entropy": 5.979336214065552,
      "epoch": 2.2236217323936622,
      "grad_norm": 1.390625,
      "learning_rate": 0.00045080840579898466,
      "loss": 5.6127,
      "mean_token_accuracy": 0.1810126632452011,
      "num_tokens": 43525233.0,
      "step": 22245
    },
    {
      "entropy": 6.118981742858887,
      "epoch": 2.224121557454891,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004507861311905334,
      "loss": 5.8506,
      "mean_token_accuracy": 0.15224290490150452,
      "num_tokens": 43536075.0,
      "step": 22250
    },
    {
      "entropy": 6.027043247222901,
      "epoch": 2.224621382516119,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004507638521593482,
      "loss": 5.5904,
      "mean_token_accuracy": 0.1804900050163269,
      "num_tokens": 43545478.0,
      "step": 22255
    },
    {
      "entropy": 6.105366945266724,
      "epoch": 2.225121207577348,
      "grad_norm": 1.25,
      "learning_rate": 0.0004507415687059898,
      "loss": 5.7052,
      "mean_token_accuracy": 0.15931729078292847,
      "num_tokens": 43554261.0,
      "step": 22260
    },
    {
      "entropy": 5.9661452293396,
      "epoch": 2.2256210326385766,
      "grad_norm": 1.25,
      "learning_rate": 0.00045071928083101874,
      "loss": 5.6107,
      "mean_token_accuracy": 0.17459188252687455,
      "num_tokens": 43563235.0,
      "step": 22265
    },
    {
      "entropy": 6.068112993240357,
      "epoch": 2.2261208576998053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004506969885349957,
      "loss": 5.7252,
      "mean_token_accuracy": 0.16759674698114396,
      "num_tokens": 43572385.0,
      "step": 22270
    },
    {
      "entropy": 6.050657415390015,
      "epoch": 2.2266206827610335,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004506746918184818,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17530978173017503,
      "num_tokens": 43580843.0,
      "step": 22275
    },
    {
      "entropy": 6.02613639831543,
      "epoch": 2.2271205078222622,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004506523906820378,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17955580055713655,
      "num_tokens": 43590500.0,
      "step": 22280
    },
    {
      "entropy": 5.980172967910766,
      "epoch": 2.227620332883491,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004506300851262248,
      "loss": 5.6228,
      "mean_token_accuracy": 0.17643571645021439,
      "num_tokens": 43601306.0,
      "step": 22285
    },
    {
      "entropy": 6.0255202293396,
      "epoch": 2.228120157944719,
      "grad_norm": 1.125,
      "learning_rate": 0.00045060777515160404,
      "loss": 5.5747,
      "mean_token_accuracy": 0.17988001108169555,
      "num_tokens": 43612728.0,
      "step": 22290
    },
    {
      "entropy": 6.089349794387817,
      "epoch": 2.228619983005948,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004505854607587369,
      "loss": 5.6934,
      "mean_token_accuracy": 0.17434689700603484,
      "num_tokens": 43622178.0,
      "step": 22295
    },
    {
      "entropy": 6.143679904937744,
      "epoch": 2.2291198080671766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004505631419481847,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1587417483329773,
      "num_tokens": 43632622.0,
      "step": 22300
    },
    {
      "entropy": 6.141744232177734,
      "epoch": 2.2296196331284053,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004505408187205091,
      "loss": 5.8289,
      "mean_token_accuracy": 0.15835371986031532,
      "num_tokens": 43642703.0,
      "step": 22305
    },
    {
      "entropy": 6.055547571182251,
      "epoch": 2.2301194581896335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004505184910762716,
      "loss": 5.6914,
      "mean_token_accuracy": 0.1730354383587837,
      "num_tokens": 43652075.0,
      "step": 22310
    },
    {
      "entropy": 6.058986854553223,
      "epoch": 2.2306192832508622,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00045049615901603406,
      "loss": 5.7043,
      "mean_token_accuracy": 0.17136113941669465,
      "num_tokens": 43660459.0,
      "step": 22315
    },
    {
      "entropy": 6.005695962905884,
      "epoch": 2.231119108312091,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004504738225403583,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18519778400659562,
      "num_tokens": 43670528.0,
      "step": 22320
    },
    {
      "entropy": 5.975799703598023,
      "epoch": 2.231618933373319,
      "grad_norm": 1.125,
      "learning_rate": 0.0004504514816498064,
      "loss": 5.6257,
      "mean_token_accuracy": 0.17635545879602432,
      "num_tokens": 43680383.0,
      "step": 22325
    },
    {
      "entropy": 6.054727268218994,
      "epoch": 2.232118758434548,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045042913634494024,
      "loss": 5.6495,
      "mean_token_accuracy": 0.18034762740135193,
      "num_tokens": 43690922.0,
      "step": 22330
    },
    {
      "entropy": 6.071556234359742,
      "epoch": 2.2326185834957766,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004504067866263222,
      "loss": 5.7636,
      "mean_token_accuracy": 0.1683909311890602,
      "num_tokens": 43700444.0,
      "step": 22335
    },
    {
      "entropy": 6.1318611145019535,
      "epoch": 2.2331184085570053,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004503844324945146,
      "loss": 5.6744,
      "mean_token_accuracy": 0.1713843509554863,
      "num_tokens": 43711946.0,
      "step": 22340
    },
    {
      "entropy": 6.066158199310303,
      "epoch": 2.2336182336182335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004503620739500797,
      "loss": 5.6906,
      "mean_token_accuracy": 0.16843247413635254,
      "num_tokens": 43722368.0,
      "step": 22345
    },
    {
      "entropy": 5.932968521118164,
      "epoch": 2.2341180586794622,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004503397109935802,
      "loss": 5.5237,
      "mean_token_accuracy": 0.1851904273033142,
      "num_tokens": 43731035.0,
      "step": 22350
    },
    {
      "entropy": 6.041280364990234,
      "epoch": 2.234617883740691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004503173436255787,
      "loss": 5.6788,
      "mean_token_accuracy": 0.1676893100142479,
      "num_tokens": 43740737.0,
      "step": 22355
    },
    {
      "entropy": 6.019621276855469,
      "epoch": 2.235117708801919,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000450294971846638,
      "loss": 5.7342,
      "mean_token_accuracy": 0.1750851258635521,
      "num_tokens": 43750184.0,
      "step": 22360
    },
    {
      "entropy": 6.019226789474487,
      "epoch": 2.235617533863148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004502725956573208,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17361545115709304,
      "num_tokens": 43759573.0,
      "step": 22365
    },
    {
      "entropy": 6.0200074195861815,
      "epoch": 2.2361173589243766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004502502150581903,
      "loss": 5.6335,
      "mean_token_accuracy": 0.17137114107608795,
      "num_tokens": 43769065.0,
      "step": 22370
    },
    {
      "entropy": 6.086877298355103,
      "epoch": 2.2366171839856053,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004502278300498094,
      "loss": 5.7287,
      "mean_token_accuracy": 0.17025910019874574,
      "num_tokens": 43779279.0,
      "step": 22375
    },
    {
      "entropy": 6.0687113285064695,
      "epoch": 2.2371170090468335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045020544063274147,
      "loss": 5.7343,
      "mean_token_accuracy": 0.16966008841991426,
      "num_tokens": 43789606.0,
      "step": 22380
    },
    {
      "entropy": 6.061851882934571,
      "epoch": 2.2376168341080622,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045018304680754975,
      "loss": 5.7068,
      "mean_token_accuracy": 0.17301966100931168,
      "num_tokens": 43798309.0,
      "step": 22385
    },
    {
      "entropy": 6.0265124320983885,
      "epoch": 2.238116659169291,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004501606485747975,
      "loss": 5.6983,
      "mean_token_accuracy": 0.17446670383214952,
      "num_tokens": 43809378.0,
      "step": 22390
    },
    {
      "entropy": 6.0399909019470215,
      "epoch": 2.238616484230519,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004501382459350485,
      "loss": 5.6435,
      "mean_token_accuracy": 0.1717437133193016,
      "num_tokens": 43819258.0,
      "step": 22395
    },
    {
      "entropy": 6.031619882583618,
      "epoch": 2.239116309291748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004501158388888663,
      "loss": 5.6636,
      "mean_token_accuracy": 0.17531964778900147,
      "num_tokens": 43829015.0,
      "step": 22400
    },
    {
      "entropy": 6.129696178436279,
      "epoch": 2.2396161343529766,
      "grad_norm": 1.375,
      "learning_rate": 0.0004500934274368147,
      "loss": 5.7584,
      "mean_token_accuracy": 0.16970013380050658,
      "num_tokens": 43838347.0,
      "step": 22405
    },
    {
      "entropy": 6.186740922927856,
      "epoch": 2.240115959414205,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00045007101157945747,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1635530635714531,
      "num_tokens": 43847302.0,
      "step": 22410
    },
    {
      "entropy": 6.033930492401123,
      "epoch": 2.2406157844754335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045004859131735867,
      "loss": 5.6701,
      "mean_token_accuracy": 0.17288992404937745,
      "num_tokens": 43857803.0,
      "step": 22415
    },
    {
      "entropy": 6.043728017807007,
      "epoch": 2.2411156095366622,
      "grad_norm": 1.375,
      "learning_rate": 0.00045002616665108233,
      "loss": 5.7311,
      "mean_token_accuracy": 0.16480331867933273,
      "num_tokens": 43867682.0,
      "step": 22420
    },
    {
      "entropy": 6.03595495223999,
      "epoch": 2.241615434597891,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004500037375811926,
      "loss": 5.762,
      "mean_token_accuracy": 0.16159674227237703,
      "num_tokens": 43878699.0,
      "step": 22425
    },
    {
      "entropy": 6.1810768127441404,
      "epoch": 2.242115259659119,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044998130410825395,
      "loss": 5.7692,
      "mean_token_accuracy": 0.16305905282497407,
      "num_tokens": 43887752.0,
      "step": 22430
    },
    {
      "entropy": 6.157530975341797,
      "epoch": 2.242615084720348,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004499588662328307,
      "loss": 5.7393,
      "mean_token_accuracy": 0.16450757533311844,
      "num_tokens": 43897518.0,
      "step": 22435
    },
    {
      "entropy": 6.1025168895721436,
      "epoch": 2.2431149097815766,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004499364239554874,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17424908876419068,
      "num_tokens": 43907054.0,
      "step": 22440
    },
    {
      "entropy": 6.125002193450928,
      "epoch": 2.243614734842805,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004499139772767887,
      "loss": 5.7591,
      "mean_token_accuracy": 0.16122409254312514,
      "num_tokens": 43916926.0,
      "step": 22445
    },
    {
      "entropy": 6.068198108673096,
      "epoch": 2.2441145599040335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004498915261972993,
      "loss": 5.7234,
      "mean_token_accuracy": 0.1728411942720413,
      "num_tokens": 43926615.0,
      "step": 22450
    },
    {
      "entropy": 6.0543872833251955,
      "epoch": 2.244614384965262,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044986907071758417,
      "loss": 5.5842,
      "mean_token_accuracy": 0.17856413573026658,
      "num_tokens": 43936768.0,
      "step": 22455
    },
    {
      "entropy": 6.005413579940796,
      "epoch": 2.245114210026491,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004498466108382082,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18311822712421416,
      "num_tokens": 43945812.0,
      "step": 22460
    },
    {
      "entropy": 6.136846971511841,
      "epoch": 2.245614035087719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004498241465597365,
      "loss": 5.846,
      "mean_token_accuracy": 0.15964664071798323,
      "num_tokens": 43955848.0,
      "step": 22465
    },
    {
      "entropy": 6.116476202011109,
      "epoch": 2.246113860148948,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004498016778827342,
      "loss": 5.6995,
      "mean_token_accuracy": 0.16555339395999907,
      "num_tokens": 43964375.0,
      "step": 22470
    },
    {
      "entropy": 6.08310832977295,
      "epoch": 2.2466136852101766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004497792048077667,
      "loss": 5.7581,
      "mean_token_accuracy": 0.16371650844812394,
      "num_tokens": 43973992.0,
      "step": 22475
    },
    {
      "entropy": 6.133019208908081,
      "epoch": 2.247113510271405,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004497567273353994,
      "loss": 5.7685,
      "mean_token_accuracy": 0.1659737229347229,
      "num_tokens": 43983283.0,
      "step": 22480
    },
    {
      "entropy": 6.028037118911743,
      "epoch": 2.2476133353326335,
      "grad_norm": 1.40625,
      "learning_rate": 0.00044973424546619774,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18218233436346054,
      "num_tokens": 43992175.0,
      "step": 22485
    },
    {
      "entropy": 6.090828943252563,
      "epoch": 2.248113160393862,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004497117592007275,
      "loss": 5.6615,
      "mean_token_accuracy": 0.1757904052734375,
      "num_tokens": 44001684.0,
      "step": 22490
    },
    {
      "entropy": 6.059793424606323,
      "epoch": 2.248612985455091,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004496892685395543,
      "loss": 5.723,
      "mean_token_accuracy": 0.169662407040596,
      "num_tokens": 44011381.0,
      "step": 22495
    },
    {
      "entropy": 6.0827079772949215,
      "epoch": 2.249112810516319,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004496667734832441,
      "loss": 5.6766,
      "mean_token_accuracy": 0.17446137368679046,
      "num_tokens": 44020702.0,
      "step": 22500
    },
    {
      "entropy": 6.041311740875244,
      "epoch": 2.249612635577548,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044964427403236274,
      "loss": 5.6388,
      "mean_token_accuracy": 0.17422616332769394,
      "num_tokens": 44030123.0,
      "step": 22505
    },
    {
      "entropy": 5.9842767238616945,
      "epoch": 2.2501124606387766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004496217701874765,
      "loss": 5.631,
      "mean_token_accuracy": 0.17304603457450868,
      "num_tokens": 44039673.0,
      "step": 22510
    },
    {
      "entropy": 5.977402973175049,
      "epoch": 2.250612285700005,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004495992619491514,
      "loss": 5.5533,
      "mean_token_accuracy": 0.1791591927409172,
      "num_tokens": 44050446.0,
      "step": 22515
    },
    {
      "entropy": 6.012661647796631,
      "epoch": 2.2511121107612335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004495767493179538,
      "loss": 5.642,
      "mean_token_accuracy": 0.1738092765212059,
      "num_tokens": 44059711.0,
      "step": 22520
    },
    {
      "entropy": 6.114560222625732,
      "epoch": 2.251611935822462,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004495542322944501,
      "loss": 5.7358,
      "mean_token_accuracy": 0.16339579969644547,
      "num_tokens": 44069573.0,
      "step": 22525
    },
    {
      "entropy": 6.092714548110962,
      "epoch": 2.2521117608836905,
      "grad_norm": 1.25,
      "learning_rate": 0.00044953171087920693,
      "loss": 5.7202,
      "mean_token_accuracy": 0.17207184582948684,
      "num_tokens": 44078834.0,
      "step": 22530
    },
    {
      "entropy": 6.030003213882447,
      "epoch": 2.252611585944919,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004495091850727907,
      "loss": 5.668,
      "mean_token_accuracy": 0.17535384744405746,
      "num_tokens": 44089007.0,
      "step": 22535
    },
    {
      "entropy": 6.0552496910095215,
      "epoch": 2.253111411006148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044948665487576833,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17471062242984772,
      "num_tokens": 44099482.0,
      "step": 22540
    },
    {
      "entropy": 6.082548475265503,
      "epoch": 2.2536112360673766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044946412028870657,
      "loss": 5.7449,
      "mean_token_accuracy": 0.16935905516147615,
      "num_tokens": 44108767.0,
      "step": 22545
    },
    {
      "entropy": 6.069309043884277,
      "epoch": 2.254111061128605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004494415813121725,
      "loss": 5.5647,
      "mean_token_accuracy": 0.1831383928656578,
      "num_tokens": 44119086.0,
      "step": 22550
    },
    {
      "entropy": 6.072295618057251,
      "epoch": 2.2546108861898335,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004494190379467331,
      "loss": 5.7289,
      "mean_token_accuracy": 0.17360569089651107,
      "num_tokens": 44129392.0,
      "step": 22555
    },
    {
      "entropy": 6.146345520019532,
      "epoch": 2.255110711251062,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044939649019295564,
      "loss": 5.8481,
      "mean_token_accuracy": 0.16550399661064147,
      "num_tokens": 44141226.0,
      "step": 22560
    },
    {
      "entropy": 6.0753926753997805,
      "epoch": 2.2556105363122905,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004493739380514073,
      "loss": 5.6364,
      "mean_token_accuracy": 0.18202758878469466,
      "num_tokens": 44151142.0,
      "step": 22565
    },
    {
      "entropy": 6.0422461986541744,
      "epoch": 2.256110361373519,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004493513815226556,
      "loss": 5.6238,
      "mean_token_accuracy": 0.17931187599897386,
      "num_tokens": 44161431.0,
      "step": 22570
    },
    {
      "entropy": 5.98138656616211,
      "epoch": 2.256610186434748,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000449328820607268,
      "loss": 5.6514,
      "mean_token_accuracy": 0.18012242168188095,
      "num_tokens": 44171220.0,
      "step": 22575
    },
    {
      "entropy": 6.0695531368255615,
      "epoch": 2.2571100114959766,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004493062553058121,
      "loss": 5.7611,
      "mean_token_accuracy": 0.1662132427096367,
      "num_tokens": 44181314.0,
      "step": 22580
    },
    {
      "entropy": 6.07424578666687,
      "epoch": 2.257609836557205,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044928368561885566,
      "loss": 5.6572,
      "mean_token_accuracy": 0.17335680574178697,
      "num_tokens": 44190575.0,
      "step": 22585
    },
    {
      "entropy": 6.061916494369507,
      "epoch": 2.2581096616184335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044926111154696655,
      "loss": 5.7244,
      "mean_token_accuracy": 0.16768932789564134,
      "num_tokens": 44199769.0,
      "step": 22590
    },
    {
      "entropy": 6.0376749515533445,
      "epoch": 2.258609486679662,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004492385330907127,
      "loss": 5.6169,
      "mean_token_accuracy": 0.18004419803619384,
      "num_tokens": 44209255.0,
      "step": 22595
    },
    {
      "entropy": 6.08119683265686,
      "epoch": 2.2591093117408905,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004492159502506622,
      "loss": 5.7604,
      "mean_token_accuracy": 0.17187785357236862,
      "num_tokens": 44217742.0,
      "step": 22600
    },
    {
      "entropy": 5.9936192512512205,
      "epoch": 2.259609136802119,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044919336302738324,
      "loss": 5.6736,
      "mean_token_accuracy": 0.17271197140216826,
      "num_tokens": 44227277.0,
      "step": 22605
    },
    {
      "entropy": 6.074738025665283,
      "epoch": 2.260108961863348,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044917077142144407,
      "loss": 5.7021,
      "mean_token_accuracy": 0.17731619775295257,
      "num_tokens": 44236460.0,
      "step": 22610
    },
    {
      "entropy": 5.953318166732788,
      "epoch": 2.2606087869245766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044914817543341315,
      "loss": 5.6414,
      "mean_token_accuracy": 0.17848921269178392,
      "num_tokens": 44246840.0,
      "step": 22615
    },
    {
      "entropy": 6.142713260650635,
      "epoch": 2.261108611985805,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004491255750638589,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16318626701831818,
      "num_tokens": 44257258.0,
      "step": 22620
    },
    {
      "entropy": 6.160098600387573,
      "epoch": 2.2616084370470335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044910297031335006,
      "loss": 5.7709,
      "mean_token_accuracy": 0.17275480628013612,
      "num_tokens": 44267155.0,
      "step": 22625
    },
    {
      "entropy": 5.930028486251831,
      "epoch": 2.262108262108262,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044908036118245523,
      "loss": 5.5076,
      "mean_token_accuracy": 0.18199223279953003,
      "num_tokens": 44276432.0,
      "step": 22630
    },
    {
      "entropy": 6.022866868972779,
      "epoch": 2.2626080871694905,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004490577476717433,
      "loss": 5.627,
      "mean_token_accuracy": 0.1819588229060173,
      "num_tokens": 44287340.0,
      "step": 22635
    },
    {
      "entropy": 6.0334547996521,
      "epoch": 2.263107912230719,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004490351297817832,
      "loss": 5.6849,
      "mean_token_accuracy": 0.1740166053175926,
      "num_tokens": 44296252.0,
      "step": 22640
    },
    {
      "entropy": 6.064418983459473,
      "epoch": 2.263607737291948,
      "grad_norm": 1.140625,
      "learning_rate": 0.000449012507513144,
      "loss": 5.6895,
      "mean_token_accuracy": 0.17840614765882493,
      "num_tokens": 44305779.0,
      "step": 22645
    },
    {
      "entropy": 6.036673927307129,
      "epoch": 2.2641075623531766,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044898988086639497,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17423006892204285,
      "num_tokens": 44315729.0,
      "step": 22650
    },
    {
      "entropy": 6.005787515640259,
      "epoch": 2.264607387414405,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004489672498421053,
      "loss": 5.6595,
      "mean_token_accuracy": 0.17248037010431289,
      "num_tokens": 44325525.0,
      "step": 22655
    },
    {
      "entropy": 5.990830612182617,
      "epoch": 2.2651072124756335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004489446144408444,
      "loss": 5.5762,
      "mean_token_accuracy": 0.1865074262022972,
      "num_tokens": 44335860.0,
      "step": 22660
    },
    {
      "entropy": 5.9571503639221195,
      "epoch": 2.265607037536862,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004489219746631817,
      "loss": 5.5682,
      "mean_token_accuracy": 0.18396010994911194,
      "num_tokens": 44344944.0,
      "step": 22665
    },
    {
      "entropy": 6.046880006790161,
      "epoch": 2.2661068625980905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044889933050968695,
      "loss": 5.6699,
      "mean_token_accuracy": 0.17131990790367127,
      "num_tokens": 44355293.0,
      "step": 22670
    },
    {
      "entropy": 6.034197759628296,
      "epoch": 2.266606687659319,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044887668198092975,
      "loss": 5.5927,
      "mean_token_accuracy": 0.18326831758022308,
      "num_tokens": 44364252.0,
      "step": 22675
    },
    {
      "entropy": 6.095677423477173,
      "epoch": 2.267106512720548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044885402907748,
      "loss": 5.7071,
      "mean_token_accuracy": 0.17406044006347657,
      "num_tokens": 44374483.0,
      "step": 22680
    },
    {
      "entropy": 6.082588720321655,
      "epoch": 2.2676063377817766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044883137179990755,
      "loss": 5.7011,
      "mean_token_accuracy": 0.1705227628350258,
      "num_tokens": 44385163.0,
      "step": 22685
    },
    {
      "entropy": 6.047369956970215,
      "epoch": 2.268106162843005,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044880871014878253,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1776985317468643,
      "num_tokens": 44394993.0,
      "step": 22690
    },
    {
      "entropy": 6.0860381603240965,
      "epoch": 2.2686059879042335,
      "grad_norm": 1.25,
      "learning_rate": 0.00044878604412467517,
      "loss": 5.7347,
      "mean_token_accuracy": 0.17210544347763063,
      "num_tokens": 44404954.0,
      "step": 22695
    },
    {
      "entropy": 6.080325555801392,
      "epoch": 2.269105812965462,
      "grad_norm": 1.25,
      "learning_rate": 0.00044876337372815555,
      "loss": 5.7035,
      "mean_token_accuracy": 0.16961118578910828,
      "num_tokens": 44413911.0,
      "step": 22700
    },
    {
      "entropy": 6.136998891830444,
      "epoch": 2.2696056380266905,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00044874069895979417,
      "loss": 5.8076,
      "mean_token_accuracy": 0.16257812231779098,
      "num_tokens": 44423156.0,
      "step": 22705
    },
    {
      "entropy": 6.022463178634643,
      "epoch": 2.270105463087919,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004487180198201616,
      "loss": 5.7154,
      "mean_token_accuracy": 0.171135550737381,
      "num_tokens": 44432303.0,
      "step": 22710
    },
    {
      "entropy": 5.972885513305664,
      "epoch": 2.270605288149148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044869533630982833,
      "loss": 5.5517,
      "mean_token_accuracy": 0.17934047877788545,
      "num_tokens": 44442097.0,
      "step": 22715
    },
    {
      "entropy": 5.958808755874633,
      "epoch": 2.2711051132103766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000448672648429365,
      "loss": 5.6247,
      "mean_token_accuracy": 0.1791617527604103,
      "num_tokens": 44452168.0,
      "step": 22720
    },
    {
      "entropy": 6.010843658447266,
      "epoch": 2.271604938271605,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044864995617934265,
      "loss": 5.6518,
      "mean_token_accuracy": 0.16865999400615692,
      "num_tokens": 44461612.0,
      "step": 22725
    },
    {
      "entropy": 6.063406658172608,
      "epoch": 2.2721047633328335,
      "grad_norm": 1.234375,
      "learning_rate": 0.000448627259560332,
      "loss": 5.7258,
      "mean_token_accuracy": 0.16694357693195344,
      "num_tokens": 44473397.0,
      "step": 22730
    },
    {
      "entropy": 6.038657760620117,
      "epoch": 2.272604588394062,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044860455857290415,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17758417129516602,
      "num_tokens": 44482677.0,
      "step": 22735
    },
    {
      "entropy": 6.097441530227661,
      "epoch": 2.2731044134552905,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004485818532176304,
      "loss": 5.7326,
      "mean_token_accuracy": 0.16886211037635804,
      "num_tokens": 44492918.0,
      "step": 22740
    },
    {
      "entropy": 6.090908193588257,
      "epoch": 2.273604238516519,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004485591434950818,
      "loss": 5.7142,
      "mean_token_accuracy": 0.16557948887348176,
      "num_tokens": 44502514.0,
      "step": 22745
    },
    {
      "entropy": 6.01677622795105,
      "epoch": 2.274104063577748,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004485364294058298,
      "loss": 5.5847,
      "mean_token_accuracy": 0.18155863285064697,
      "num_tokens": 44513078.0,
      "step": 22750
    },
    {
      "entropy": 5.987873363494873,
      "epoch": 2.2746038886389766,
      "grad_norm": 1.25,
      "learning_rate": 0.00044851371095044597,
      "loss": 5.6687,
      "mean_token_accuracy": 0.17145102024078368,
      "num_tokens": 44523318.0,
      "step": 22755
    },
    {
      "entropy": 6.055798530578613,
      "epoch": 2.275103713700205,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044849098812950174,
      "loss": 5.7418,
      "mean_token_accuracy": 0.17206021696329116,
      "num_tokens": 44533910.0,
      "step": 22760
    },
    {
      "entropy": 6.0717298030853275,
      "epoch": 2.2756035387614335,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000448468260943569,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17326900362968445,
      "num_tokens": 44543733.0,
      "step": 22765
    },
    {
      "entropy": 6.03480486869812,
      "epoch": 2.276103363822662,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044844552939321937,
      "loss": 5.7135,
      "mean_token_accuracy": 0.17321734726428986,
      "num_tokens": 44554041.0,
      "step": 22770
    },
    {
      "entropy": 6.011075162887574,
      "epoch": 2.2766031888838905,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004484227934790248,
      "loss": 5.677,
      "mean_token_accuracy": 0.17687660455703735,
      "num_tokens": 44564136.0,
      "step": 22775
    },
    {
      "entropy": 5.995497846603394,
      "epoch": 2.277103013945119,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004484000532015575,
      "loss": 5.5977,
      "mean_token_accuracy": 0.1805384486913681,
      "num_tokens": 44572706.0,
      "step": 22780
    },
    {
      "entropy": 6.06974196434021,
      "epoch": 2.277602839006348,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044837730856138935,
      "loss": 5.7019,
      "mean_token_accuracy": 0.17030956596136093,
      "num_tokens": 44581748.0,
      "step": 22785
    },
    {
      "entropy": 6.033576440811157,
      "epoch": 2.2781026640675766,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004483545595590928,
      "loss": 5.6329,
      "mean_token_accuracy": 0.17813839614391327,
      "num_tokens": 44591107.0,
      "step": 22790
    },
    {
      "entropy": 6.136862564086914,
      "epoch": 2.278602489128805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004483318061952401,
      "loss": 5.7573,
      "mean_token_accuracy": 0.16432317197322846,
      "num_tokens": 44600718.0,
      "step": 22795
    },
    {
      "entropy": 5.972891807556152,
      "epoch": 2.2791023141900335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004483090484704038,
      "loss": 5.6039,
      "mean_token_accuracy": 0.17773894667625428,
      "num_tokens": 44610702.0,
      "step": 22800
    },
    {
      "entropy": 6.069174432754517,
      "epoch": 2.279602139251262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044828628638515647,
      "loss": 5.6839,
      "mean_token_accuracy": 0.17562095671892167,
      "num_tokens": 44621060.0,
      "step": 22805
    },
    {
      "entropy": 6.037214279174805,
      "epoch": 2.2801019643124905,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004482635199400707,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1826841875910759,
      "num_tokens": 44631060.0,
      "step": 22810
    },
    {
      "entropy": 6.06053352355957,
      "epoch": 2.280601789373719,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044824074913571936,
      "loss": 5.6392,
      "mean_token_accuracy": 0.18135792165994644,
      "num_tokens": 44640025.0,
      "step": 22815
    },
    {
      "entropy": 6.013695812225341,
      "epoch": 2.281101614434948,
      "grad_norm": 1.25,
      "learning_rate": 0.00044821797397267535,
      "loss": 5.7412,
      "mean_token_accuracy": 0.17610736191272736,
      "num_tokens": 44650168.0,
      "step": 22820
    },
    {
      "entropy": 6.089326047897339,
      "epoch": 2.2816014394961766,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004481951944515117,
      "loss": 5.7283,
      "mean_token_accuracy": 0.16308040022850037,
      "num_tokens": 44660370.0,
      "step": 22825
    },
    {
      "entropy": 6.040594959259034,
      "epoch": 2.282101264557405,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044817241057280155,
      "loss": 5.6919,
      "mean_token_accuracy": 0.1686481863260269,
      "num_tokens": 44669776.0,
      "step": 22830
    },
    {
      "entropy": 6.018581199645996,
      "epoch": 2.2826010896186335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004481496223371181,
      "loss": 5.6498,
      "mean_token_accuracy": 0.1762848123908043,
      "num_tokens": 44678891.0,
      "step": 22835
    },
    {
      "entropy": 6.009015893936157,
      "epoch": 2.283100914679862,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00044812682974503464,
      "loss": 5.6182,
      "mean_token_accuracy": 0.18193535059690474,
      "num_tokens": 44688750.0,
      "step": 22840
    },
    {
      "entropy": 6.044629335403442,
      "epoch": 2.2836007397410905,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004481040327971248,
      "loss": 5.651,
      "mean_token_accuracy": 0.17397948652505874,
      "num_tokens": 44698570.0,
      "step": 22845
    },
    {
      "entropy": 6.091488409042358,
      "epoch": 2.284100564802319,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044808123149396193,
      "loss": 5.6798,
      "mean_token_accuracy": 0.17537674456834793,
      "num_tokens": 44707423.0,
      "step": 22850
    },
    {
      "entropy": 6.0692291259765625,
      "epoch": 2.284600389863548,
      "grad_norm": 1.25,
      "learning_rate": 0.00044805842583611993,
      "loss": 5.721,
      "mean_token_accuracy": 0.16528848856687545,
      "num_tokens": 44717526.0,
      "step": 22855
    },
    {
      "entropy": 5.996245718002319,
      "epoch": 2.2851002149247766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044803561582417237,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18408962935209275,
      "num_tokens": 44726713.0,
      "step": 22860
    },
    {
      "entropy": 6.085090017318725,
      "epoch": 2.285600039986005,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044801280145869327,
      "loss": 5.7784,
      "mean_token_accuracy": 0.1661432057619095,
      "num_tokens": 44736399.0,
      "step": 22865
    },
    {
      "entropy": 6.093620777130127,
      "epoch": 2.2860998650472335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004479899827402566,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16218296885490419,
      "num_tokens": 44746291.0,
      "step": 22870
    },
    {
      "entropy": 6.098348712921142,
      "epoch": 2.286599690108462,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044796715966943654,
      "loss": 5.7259,
      "mean_token_accuracy": 0.16971249282360076,
      "num_tokens": 44755937.0,
      "step": 22875
    },
    {
      "entropy": 6.040825033187867,
      "epoch": 2.2870995151696905,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004479443322468072,
      "loss": 5.6899,
      "mean_token_accuracy": 0.17231782674789428,
      "num_tokens": 44765764.0,
      "step": 22880
    },
    {
      "entropy": 6.067892217636109,
      "epoch": 2.287599340230919,
      "grad_norm": 1.125,
      "learning_rate": 0.0004479215004729429,
      "loss": 5.7614,
      "mean_token_accuracy": 0.16674137860536575,
      "num_tokens": 44776134.0,
      "step": 22885
    },
    {
      "entropy": 6.0653977394104,
      "epoch": 2.288099165292148,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004478986643484183,
      "loss": 5.689,
      "mean_token_accuracy": 0.17864413261413575,
      "num_tokens": 44786356.0,
      "step": 22890
    },
    {
      "entropy": 6.083305025100708,
      "epoch": 2.2885989903533765,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044787582387380764,
      "loss": 5.5877,
      "mean_token_accuracy": 0.17913303524255753,
      "num_tokens": 44795838.0,
      "step": 22895
    },
    {
      "entropy": 6.030621671676636,
      "epoch": 2.289098815414605,
      "grad_norm": 1.25,
      "learning_rate": 0.0004478529790496858,
      "loss": 5.6861,
      "mean_token_accuracy": 0.17426675260066987,
      "num_tokens": 44804571.0,
      "step": 22900
    },
    {
      "entropy": 6.096728038787842,
      "epoch": 2.2895986404758335,
      "grad_norm": 1.25,
      "learning_rate": 0.00044783012987662755,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17425755709409713,
      "num_tokens": 44814546.0,
      "step": 22905
    },
    {
      "entropy": 6.031918573379516,
      "epoch": 2.290098465537062,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004478072763552076,
      "loss": 5.5824,
      "mean_token_accuracy": 0.17931883335113524,
      "num_tokens": 44825196.0,
      "step": 22910
    },
    {
      "entropy": 6.02703628540039,
      "epoch": 2.2905982905982905,
      "grad_norm": 1.25,
      "learning_rate": 0.0004477844184860011,
      "loss": 5.6327,
      "mean_token_accuracy": 0.1738823503255844,
      "num_tokens": 44835502.0,
      "step": 22915
    },
    {
      "entropy": 5.994491243362427,
      "epoch": 2.291098115659519,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044776155626958305,
      "loss": 5.6887,
      "mean_token_accuracy": 0.17429987639188765,
      "num_tokens": 44845574.0,
      "step": 22920
    },
    {
      "entropy": 5.905504035949707,
      "epoch": 2.291597940720748,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044773868970652875,
      "loss": 5.5456,
      "mean_token_accuracy": 0.1858523964881897,
      "num_tokens": 44855946.0,
      "step": 22925
    },
    {
      "entropy": 6.057617712020874,
      "epoch": 2.2920977657819765,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044771581879741334,
      "loss": 5.7442,
      "mean_token_accuracy": 0.16830600947141647,
      "num_tokens": 44865753.0,
      "step": 22930
    },
    {
      "entropy": 6.058864545822144,
      "epoch": 2.292597590843205,
      "grad_norm": 1.71875,
      "learning_rate": 0.00044769294354281253,
      "loss": 5.6614,
      "mean_token_accuracy": 0.17320311069488525,
      "num_tokens": 44876296.0,
      "step": 22935
    },
    {
      "entropy": 6.180350065231323,
      "epoch": 2.2930974159044335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044767006394330166,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16393024772405623,
      "num_tokens": 44886207.0,
      "step": 22940
    },
    {
      "entropy": 6.048877143859864,
      "epoch": 2.293597240965662,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044764717999945644,
      "loss": 5.6505,
      "mean_token_accuracy": 0.18178277909755708,
      "num_tokens": 44894911.0,
      "step": 22945
    },
    {
      "entropy": 5.993880033493042,
      "epoch": 2.2940970660268905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044762429171185246,
      "loss": 5.6677,
      "mean_token_accuracy": 0.17632439434528352,
      "num_tokens": 44904513.0,
      "step": 22950
    },
    {
      "entropy": 5.944888544082642,
      "epoch": 2.294596891088119,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044760139908106583,
      "loss": 5.6015,
      "mean_token_accuracy": 0.17590239495038987,
      "num_tokens": 44914039.0,
      "step": 22955
    },
    {
      "entropy": 5.996934270858764,
      "epoch": 2.295096716149348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004475785021076724,
      "loss": 5.6843,
      "mean_token_accuracy": 0.17826638370752335,
      "num_tokens": 44924149.0,
      "step": 22960
    },
    {
      "entropy": 5.973503112792969,
      "epoch": 2.2955965412105765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044755560079224834,
      "loss": 5.5862,
      "mean_token_accuracy": 0.18703240752220154,
      "num_tokens": 44933929.0,
      "step": 22965
    },
    {
      "entropy": 6.065384244918823,
      "epoch": 2.296096366271805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004475326951353697,
      "loss": 5.6231,
      "mean_token_accuracy": 0.1779414877295494,
      "num_tokens": 44943199.0,
      "step": 22970
    },
    {
      "entropy": 6.086389350891113,
      "epoch": 2.2965961913330335,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044750978513761276,
      "loss": 5.7255,
      "mean_token_accuracy": 0.16743334680795668,
      "num_tokens": 44952660.0,
      "step": 22975
    },
    {
      "entropy": 5.991690540313721,
      "epoch": 2.297096016394262,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004474868707995541,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1795956015586853,
      "num_tokens": 44962907.0,
      "step": 22980
    },
    {
      "entropy": 6.044561433792114,
      "epoch": 2.2975958414554905,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044746395212177015,
      "loss": 5.6922,
      "mean_token_accuracy": 0.1763395309448242,
      "num_tokens": 44973319.0,
      "step": 22985
    },
    {
      "entropy": 6.051795721054077,
      "epoch": 2.298095666516719,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044744102910483755,
      "loss": 5.6284,
      "mean_token_accuracy": 0.1791013792157173,
      "num_tokens": 44982818.0,
      "step": 22990
    },
    {
      "entropy": 5.982341527938843,
      "epoch": 2.298595491577948,
      "grad_norm": 1.140625,
      "learning_rate": 0.000447418101749333,
      "loss": 5.5167,
      "mean_token_accuracy": 0.186491659283638,
      "num_tokens": 44992189.0,
      "step": 22995
    },
    {
      "entropy": 6.009306335449219,
      "epoch": 2.299095316639176,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004473951700558334,
      "loss": 5.6601,
      "mean_token_accuracy": 0.18032930195331573,
      "num_tokens": 45002296.0,
      "step": 23000
    },
    {
      "entropy": 6.105239486694336,
      "epoch": 2.299595141700405,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044737223402491574,
      "loss": 5.7091,
      "mean_token_accuracy": 0.16951675862073898,
      "num_tokens": 45011500.0,
      "step": 23005
    },
    {
      "entropy": 6.119904041290283,
      "epoch": 2.3000949667616335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044734929365715685,
      "loss": 5.7841,
      "mean_token_accuracy": 0.16232468336820602,
      "num_tokens": 45021816.0,
      "step": 23010
    },
    {
      "entropy": 6.120476770401001,
      "epoch": 2.3005947918228618,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044732634895313424,
      "loss": 5.768,
      "mean_token_accuracy": 0.16844913512468337,
      "num_tokens": 45031227.0,
      "step": 23015
    },
    {
      "entropy": 6.00724425315857,
      "epoch": 2.3010946168840904,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000447303399913425,
      "loss": 5.6152,
      "mean_token_accuracy": 0.18231048434972763,
      "num_tokens": 45040076.0,
      "step": 23020
    },
    {
      "entropy": 6.100012636184692,
      "epoch": 2.301594441945319,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044728044653860653,
      "loss": 5.7893,
      "mean_token_accuracy": 0.17103687077760696,
      "num_tokens": 45049905.0,
      "step": 23025
    },
    {
      "entropy": 6.177232980728149,
      "epoch": 2.302094267006548,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004472574888292564,
      "loss": 5.8297,
      "mean_token_accuracy": 0.1632405236363411,
      "num_tokens": 45059722.0,
      "step": 23030
    },
    {
      "entropy": 6.055925893783569,
      "epoch": 2.302594092067776,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004472345267859521,
      "loss": 5.6605,
      "mean_token_accuracy": 0.1753182590007782,
      "num_tokens": 45068734.0,
      "step": 23035
    },
    {
      "entropy": 6.044592380523682,
      "epoch": 2.303093917129005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004472115604092715,
      "loss": 5.6548,
      "mean_token_accuracy": 0.17338049858808519,
      "num_tokens": 45078774.0,
      "step": 23040
    },
    {
      "entropy": 6.002565383911133,
      "epoch": 2.3035937421902335,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044718858969979236,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17240967303514482,
      "num_tokens": 45088351.0,
      "step": 23045
    },
    {
      "entropy": 6.195791101455688,
      "epoch": 2.3040935672514617,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044716561465809256,
      "loss": 5.8661,
      "mean_token_accuracy": 0.15939389020204545,
      "num_tokens": 45098530.0,
      "step": 23050
    },
    {
      "entropy": 6.044642543792724,
      "epoch": 2.3045933923126904,
      "grad_norm": 1.25,
      "learning_rate": 0.00044714263528475024,
      "loss": 5.6939,
      "mean_token_accuracy": 0.17320514023303984,
      "num_tokens": 45108950.0,
      "step": 23055
    },
    {
      "entropy": 5.977533292770386,
      "epoch": 2.305093217373919,
      "grad_norm": 1.25,
      "learning_rate": 0.00044711965158034345,
      "loss": 5.7274,
      "mean_token_accuracy": 0.17250028997659683,
      "num_tokens": 45117548.0,
      "step": 23060
    },
    {
      "entropy": 6.073072624206543,
      "epoch": 2.305593042435148,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004470966635454506,
      "loss": 5.6787,
      "mean_token_accuracy": 0.168802110850811,
      "num_tokens": 45126882.0,
      "step": 23065
    },
    {
      "entropy": 6.150829887390136,
      "epoch": 2.306092867496376,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004470736711806499,
      "loss": 5.7112,
      "mean_token_accuracy": 0.16492611914873123,
      "num_tokens": 45137751.0,
      "step": 23070
    },
    {
      "entropy": 6.011454010009766,
      "epoch": 2.306592692557605,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044705067448651996,
      "loss": 5.5135,
      "mean_token_accuracy": 0.19064177721738815,
      "num_tokens": 45147051.0,
      "step": 23075
    },
    {
      "entropy": 6.016005754470825,
      "epoch": 2.3070925176188335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004470276734636393,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17156544774770738,
      "num_tokens": 45156859.0,
      "step": 23080
    },
    {
      "entropy": 6.034773349761963,
      "epoch": 2.3075923426800617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044700466811258666,
      "loss": 5.6543,
      "mean_token_accuracy": 0.1768588125705719,
      "num_tokens": 45167555.0,
      "step": 23085
    },
    {
      "entropy": 6.08986873626709,
      "epoch": 2.3080921677412904,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004469816584339408,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17727627158164977,
      "num_tokens": 45177777.0,
      "step": 23090
    },
    {
      "entropy": 6.037686443328857,
      "epoch": 2.308591992802519,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044695864442828066,
      "loss": 5.6278,
      "mean_token_accuracy": 0.17769563049077988,
      "num_tokens": 45187902.0,
      "step": 23095
    },
    {
      "entropy": 6.109902954101562,
      "epoch": 2.309091817863748,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044693562609618534,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17385202944278716,
      "num_tokens": 45198012.0,
      "step": 23100
    },
    {
      "entropy": 6.069012069702149,
      "epoch": 2.309591642924976,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004469126034382339,
      "loss": 5.6328,
      "mean_token_accuracy": 0.183845716714859,
      "num_tokens": 45208327.0,
      "step": 23105
    },
    {
      "entropy": 6.049908065795899,
      "epoch": 2.310091467986205,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004468895764550055,
      "loss": 5.7828,
      "mean_token_accuracy": 0.16745977997779846,
      "num_tokens": 45218748.0,
      "step": 23110
    },
    {
      "entropy": 6.0672197341918945,
      "epoch": 2.3105912930474335,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004468665451470797,
      "loss": 5.7077,
      "mean_token_accuracy": 0.17513021379709243,
      "num_tokens": 45228803.0,
      "step": 23115
    },
    {
      "entropy": 6.088761281967163,
      "epoch": 2.3110911181086617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004468435095150358,
      "loss": 5.6774,
      "mean_token_accuracy": 0.17834333926439286,
      "num_tokens": 45238855.0,
      "step": 23120
    },
    {
      "entropy": 6.056411600112915,
      "epoch": 2.3115909431698904,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044682046955945345,
      "loss": 5.6456,
      "mean_token_accuracy": 0.17723999321460723,
      "num_tokens": 45248848.0,
      "step": 23125
    },
    {
      "entropy": 6.067784786224365,
      "epoch": 2.312090768231119,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004467974252809123,
      "loss": 5.7008,
      "mean_token_accuracy": 0.16763491481542586,
      "num_tokens": 45258170.0,
      "step": 23130
    },
    {
      "entropy": 6.0079484462738035,
      "epoch": 2.312590593292348,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044677437667999204,
      "loss": 5.6473,
      "mean_token_accuracy": 0.17765099108219146,
      "num_tokens": 45268033.0,
      "step": 23135
    },
    {
      "entropy": 6.036505794525146,
      "epoch": 2.313090418353576,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044675132375727273,
      "loss": 5.6482,
      "mean_token_accuracy": 0.1798658013343811,
      "num_tokens": 45277318.0,
      "step": 23140
    },
    {
      "entropy": 6.091793870925903,
      "epoch": 2.313590243414805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004467282665133343,
      "loss": 5.6958,
      "mean_token_accuracy": 0.1692764401435852,
      "num_tokens": 45286810.0,
      "step": 23145
    },
    {
      "entropy": 6.008791542053222,
      "epoch": 2.3140900684760335,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004467052049487569,
      "loss": 5.5814,
      "mean_token_accuracy": 0.17930554896593093,
      "num_tokens": 45296292.0,
      "step": 23150
    },
    {
      "entropy": 6.092882299423218,
      "epoch": 2.3145898935372617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044668213906412074,
      "loss": 5.7019,
      "mean_token_accuracy": 0.17092470973730087,
      "num_tokens": 45305580.0,
      "step": 23155
    },
    {
      "entropy": 5.921294212341309,
      "epoch": 2.3150897185984904,
      "grad_norm": 2.375,
      "learning_rate": 0.00044665906886000604,
      "loss": 5.5177,
      "mean_token_accuracy": 0.1913107618689537,
      "num_tokens": 45314831.0,
      "step": 23160
    },
    {
      "entropy": 5.97441611289978,
      "epoch": 2.315589543659719,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044663599433699346,
      "loss": 5.629,
      "mean_token_accuracy": 0.17939716428518296,
      "num_tokens": 45324503.0,
      "step": 23165
    },
    {
      "entropy": 6.143839740753174,
      "epoch": 2.316089368720948,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044661291549566337,
      "loss": 5.7625,
      "mean_token_accuracy": 0.1685203120112419,
      "num_tokens": 45335459.0,
      "step": 23170
    },
    {
      "entropy": 6.121432113647461,
      "epoch": 2.316589193782176,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004465898323365965,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16739365607500076,
      "num_tokens": 45344951.0,
      "step": 23175
    },
    {
      "entropy": 6.118986797332764,
      "epoch": 2.317089018843405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004465667448603736,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16021621078252793,
      "num_tokens": 45354953.0,
      "step": 23180
    },
    {
      "entropy": 6.091918516159057,
      "epoch": 2.3175888439046335,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044654365306757553,
      "loss": 5.7047,
      "mean_token_accuracy": 0.1706259921193123,
      "num_tokens": 45364567.0,
      "step": 23185
    },
    {
      "entropy": 6.156133317947388,
      "epoch": 2.3180886689658617,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044652055695878335,
      "loss": 5.7146,
      "mean_token_accuracy": 0.16759245097637177,
      "num_tokens": 45374147.0,
      "step": 23190
    },
    {
      "entropy": 6.076023769378662,
      "epoch": 2.3185884940270904,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044649745653457806,
      "loss": 5.6341,
      "mean_token_accuracy": 0.1777077227830887,
      "num_tokens": 45383304.0,
      "step": 23195
    },
    {
      "entropy": 5.725328207015991,
      "epoch": 2.319088319088319,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044647435179554087,
      "loss": 5.4009,
      "mean_token_accuracy": 0.20492978990077973,
      "num_tokens": 45392508.0,
      "step": 23200
    },
    {
      "entropy": 6.0586405277252195,
      "epoch": 2.319588144149548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004464512427422531,
      "loss": 5.7352,
      "mean_token_accuracy": 0.17134341597557068,
      "num_tokens": 45402584.0,
      "step": 23205
    },
    {
      "entropy": 6.128275060653687,
      "epoch": 2.320087969210776,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044642812937529624,
      "loss": 5.7735,
      "mean_token_accuracy": 0.1616213694214821,
      "num_tokens": 45413270.0,
      "step": 23210
    },
    {
      "entropy": 6.092425012588501,
      "epoch": 2.320587794272005,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004464050116952517,
      "loss": 5.7261,
      "mean_token_accuracy": 0.17754533886909485,
      "num_tokens": 45423444.0,
      "step": 23215
    },
    {
      "entropy": 6.108678770065308,
      "epoch": 2.3210876193332335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044638188970270114,
      "loss": 5.7667,
      "mean_token_accuracy": 0.16391527205705642,
      "num_tokens": 45433737.0,
      "step": 23220
    },
    {
      "entropy": 6.071215200424194,
      "epoch": 2.3215874443944617,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00044635876339822647,
      "loss": 5.6974,
      "mean_token_accuracy": 0.1702743276953697,
      "num_tokens": 45442948.0,
      "step": 23225
    },
    {
      "entropy": 6.012935590744019,
      "epoch": 2.3220872694556904,
      "grad_norm": 1.546875,
      "learning_rate": 0.00044633563278240933,
      "loss": 5.6252,
      "mean_token_accuracy": 0.17357975691556932,
      "num_tokens": 45453181.0,
      "step": 23230
    },
    {
      "entropy": 6.039011764526367,
      "epoch": 2.322587094516919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044631249785583173,
      "loss": 5.7178,
      "mean_token_accuracy": 0.16836444437503814,
      "num_tokens": 45462888.0,
      "step": 23235
    },
    {
      "entropy": 6.103275537490845,
      "epoch": 2.323086919578148,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004462893586190758,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1719035267829895,
      "num_tokens": 45471269.0,
      "step": 23240
    },
    {
      "entropy": 6.0467942237854,
      "epoch": 2.323586744639376,
      "grad_norm": 1.25,
      "learning_rate": 0.00044626621507272364,
      "loss": 5.6833,
      "mean_token_accuracy": 0.1716432511806488,
      "num_tokens": 45480799.0,
      "step": 23245
    },
    {
      "entropy": 5.984700918197632,
      "epoch": 2.324086569700605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044624306721735763,
      "loss": 5.6725,
      "mean_token_accuracy": 0.17296350598335267,
      "num_tokens": 45490729.0,
      "step": 23250
    },
    {
      "entropy": 6.123805952072144,
      "epoch": 2.3245863947618335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044621991505356014,
      "loss": 5.7548,
      "mean_token_accuracy": 0.1650667741894722,
      "num_tokens": 45500020.0,
      "step": 23255
    },
    {
      "entropy": 6.124481153488159,
      "epoch": 2.3250862198230617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044619675858191353,
      "loss": 5.71,
      "mean_token_accuracy": 0.17179297804832458,
      "num_tokens": 45509175.0,
      "step": 23260
    },
    {
      "entropy": 6.0842968940734865,
      "epoch": 2.3255860448842904,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004461735978030006,
      "loss": 5.7188,
      "mean_token_accuracy": 0.16932351589202882,
      "num_tokens": 45519089.0,
      "step": 23265
    },
    {
      "entropy": 5.919111728668213,
      "epoch": 2.326085869945519,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044615043271740404,
      "loss": 5.5185,
      "mean_token_accuracy": 0.19256361573934555,
      "num_tokens": 45528623.0,
      "step": 23270
    },
    {
      "entropy": 6.030800199508667,
      "epoch": 2.326585695006748,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004461272633257065,
      "loss": 5.6649,
      "mean_token_accuracy": 0.1722028747200966,
      "num_tokens": 45538260.0,
      "step": 23275
    },
    {
      "entropy": 6.042866563796997,
      "epoch": 2.327085520067976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004461040896284912,
      "loss": 5.6808,
      "mean_token_accuracy": 0.17097893580794335,
      "num_tokens": 45547874.0,
      "step": 23280
    },
    {
      "entropy": 6.144810724258423,
      "epoch": 2.327585345129205,
      "grad_norm": 1.15625,
      "learning_rate": 0.000446080911626341,
      "loss": 5.6566,
      "mean_token_accuracy": 0.18004069328308106,
      "num_tokens": 45557229.0,
      "step": 23285
    },
    {
      "entropy": 6.102698850631714,
      "epoch": 2.3280851701904335,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004460577293198391,
      "loss": 5.7996,
      "mean_token_accuracy": 0.16215088218450546,
      "num_tokens": 45567803.0,
      "step": 23290
    },
    {
      "entropy": 5.977037525177002,
      "epoch": 2.3285849952516617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044603454270956863,
      "loss": 5.5351,
      "mean_token_accuracy": 0.18374047726392745,
      "num_tokens": 45576412.0,
      "step": 23295
    },
    {
      "entropy": 6.003872394561768,
      "epoch": 2.3290848203128904,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004460113517961132,
      "loss": 5.653,
      "mean_token_accuracy": 0.17269170582294463,
      "num_tokens": 45584951.0,
      "step": 23300
    },
    {
      "entropy": 6.027482748031616,
      "epoch": 2.329584645374119,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004459881565800561,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17566693872213363,
      "num_tokens": 45594520.0,
      "step": 23305
    },
    {
      "entropy": 6.001627969741821,
      "epoch": 2.330084470435348,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004459649570619809,
      "loss": 5.5643,
      "mean_token_accuracy": 0.18265742510557176,
      "num_tokens": 45604438.0,
      "step": 23310
    },
    {
      "entropy": 6.051852893829346,
      "epoch": 2.330584295496576,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004459417532424715,
      "loss": 5.7219,
      "mean_token_accuracy": 0.17558204531669616,
      "num_tokens": 45614898.0,
      "step": 23315
    },
    {
      "entropy": 6.125941133499145,
      "epoch": 2.331084120557805,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004459185451221116,
      "loss": 5.7742,
      "mean_token_accuracy": 0.16784680634737015,
      "num_tokens": 45624812.0,
      "step": 23320
    },
    {
      "entropy": 6.002766180038452,
      "epoch": 2.3315839456190335,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000445895332701485,
      "loss": 5.594,
      "mean_token_accuracy": 0.17593426406383514,
      "num_tokens": 45635238.0,
      "step": 23325
    },
    {
      "entropy": 6.108658123016357,
      "epoch": 2.3320837706802617,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044587211598117577,
      "loss": 5.6875,
      "mean_token_accuracy": 0.17447464168071747,
      "num_tokens": 45644928.0,
      "step": 23330
    },
    {
      "entropy": 6.160111713409424,
      "epoch": 2.3325835957414904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044584889496176815,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16796404272317886,
      "num_tokens": 45654708.0,
      "step": 23335
    },
    {
      "entropy": 6.063060760498047,
      "epoch": 2.333083420802719,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004458256696438463,
      "loss": 5.7025,
      "mean_token_accuracy": 0.17224891036748885,
      "num_tokens": 45664000.0,
      "step": 23340
    },
    {
      "entropy": 6.072907876968384,
      "epoch": 2.333583245863948,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044580244002799454,
      "loss": 5.7054,
      "mean_token_accuracy": 0.16995314210653306,
      "num_tokens": 45674563.0,
      "step": 23345
    },
    {
      "entropy": 6.016777658462525,
      "epoch": 2.334083070925176,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004457792061147973,
      "loss": 5.6479,
      "mean_token_accuracy": 0.1798236310482025,
      "num_tokens": 45683525.0,
      "step": 23350
    },
    {
      "entropy": 6.0997673034667965,
      "epoch": 2.334582895986405,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004457559679048392,
      "loss": 5.7791,
      "mean_token_accuracy": 0.16786349266767503,
      "num_tokens": 45693767.0,
      "step": 23355
    },
    {
      "entropy": 6.140542840957641,
      "epoch": 2.3350827210476335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000445732725398705,
      "loss": 5.7495,
      "mean_token_accuracy": 0.17368314489722253,
      "num_tokens": 45702999.0,
      "step": 23360
    },
    {
      "entropy": 6.037727975845337,
      "epoch": 2.3355825461088617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004457094785969792,
      "loss": 5.6225,
      "mean_token_accuracy": 0.17639402076601982,
      "num_tokens": 45712370.0,
      "step": 23365
    },
    {
      "entropy": 6.008673143386841,
      "epoch": 2.3360823711700904,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044568622750024693,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1764042615890503,
      "num_tokens": 45721321.0,
      "step": 23370
    },
    {
      "entropy": 6.014508247375488,
      "epoch": 2.336582196231319,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004456629721090931,
      "loss": 5.681,
      "mean_token_accuracy": 0.1714154452085495,
      "num_tokens": 45731043.0,
      "step": 23375
    },
    {
      "entropy": 6.0245105743408205,
      "epoch": 2.337082021292548,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004456397124241028,
      "loss": 5.6509,
      "mean_token_accuracy": 0.1768755540251732,
      "num_tokens": 45739916.0,
      "step": 23380
    },
    {
      "entropy": 6.04182071685791,
      "epoch": 2.337581846353776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004456164484458612,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1789645105600357,
      "num_tokens": 45749680.0,
      "step": 23385
    },
    {
      "entropy": 6.058217716217041,
      "epoch": 2.338081671415005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004455931801749537,
      "loss": 5.7289,
      "mean_token_accuracy": 0.1626124069094658,
      "num_tokens": 45758457.0,
      "step": 23390
    },
    {
      "entropy": 6.177946615219116,
      "epoch": 2.3385814964762335,
      "grad_norm": 2.03125,
      "learning_rate": 0.00044556990761196566,
      "loss": 5.7902,
      "mean_token_accuracy": 0.15996935963630676,
      "num_tokens": 45768793.0,
      "step": 23395
    },
    {
      "entropy": 6.178095197677612,
      "epoch": 2.3390813215374617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044554663075748266,
      "loss": 5.7793,
      "mean_token_accuracy": 0.17039839029312134,
      "num_tokens": 45778724.0,
      "step": 23400
    },
    {
      "entropy": 6.0857696533203125,
      "epoch": 2.3395811465986904,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004455233496120903,
      "loss": 5.7447,
      "mean_token_accuracy": 0.1728442996740341,
      "num_tokens": 45788186.0,
      "step": 23405
    },
    {
      "entropy": 6.049584245681762,
      "epoch": 2.340080971659919,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004455000641763743,
      "loss": 5.6613,
      "mean_token_accuracy": 0.170838662981987,
      "num_tokens": 45797496.0,
      "step": 23410
    },
    {
      "entropy": 6.077252054214478,
      "epoch": 2.340580796721148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044547677445092064,
      "loss": 5.6631,
      "mean_token_accuracy": 0.17840794026851653,
      "num_tokens": 45807450.0,
      "step": 23415
    },
    {
      "entropy": 6.013986825942993,
      "epoch": 2.341080621782376,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044545348043631525,
      "loss": 5.6367,
      "mean_token_accuracy": 0.18305356055498123,
      "num_tokens": 45817149.0,
      "step": 23420
    },
    {
      "entropy": 6.0737419605255125,
      "epoch": 2.341580446843605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044543018213314404,
      "loss": 5.7463,
      "mean_token_accuracy": 0.17303720563650132,
      "num_tokens": 45828102.0,
      "step": 23425
    },
    {
      "entropy": 6.061511611938476,
      "epoch": 2.3420802719048335,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044540687954199335,
      "loss": 5.6993,
      "mean_token_accuracy": 0.17023872286081315,
      "num_tokens": 45837961.0,
      "step": 23430
    },
    {
      "entropy": 5.973452568054199,
      "epoch": 2.3425800969660617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044538357266344936,
      "loss": 5.4693,
      "mean_token_accuracy": 0.1926322489976883,
      "num_tokens": 45848190.0,
      "step": 23435
    },
    {
      "entropy": 6.045256090164185,
      "epoch": 2.3430799220272904,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004453602614980986,
      "loss": 5.6771,
      "mean_token_accuracy": 0.17478940039873123,
      "num_tokens": 45857019.0,
      "step": 23440
    },
    {
      "entropy": 6.0640054702758786,
      "epoch": 2.343579747088519,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004453369460465274,
      "loss": 5.6745,
      "mean_token_accuracy": 0.1752214714884758,
      "num_tokens": 45865835.0,
      "step": 23445
    },
    {
      "entropy": 6.0751244068145756,
      "epoch": 2.344079572149748,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004453136263093225,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16525964438915253,
      "num_tokens": 45876027.0,
      "step": 23450
    },
    {
      "entropy": 6.090083122253418,
      "epoch": 2.344579397210976,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004452903022870707,
      "loss": 5.726,
      "mean_token_accuracy": 0.1668257534503937,
      "num_tokens": 45886270.0,
      "step": 23455
    },
    {
      "entropy": 6.0442986488342285,
      "epoch": 2.3450792222722048,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004452669739803586,
      "loss": 5.6178,
      "mean_token_accuracy": 0.18402289301157,
      "num_tokens": 45895741.0,
      "step": 23460
    },
    {
      "entropy": 5.984249114990234,
      "epoch": 2.3455790473334335,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004452436413897733,
      "loss": 5.58,
      "mean_token_accuracy": 0.18382303267717362,
      "num_tokens": 45905119.0,
      "step": 23465
    },
    {
      "entropy": 6.026241111755371,
      "epoch": 2.3460788723946617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004452203045159018,
      "loss": 5.6942,
      "mean_token_accuracy": 0.16921270489692689,
      "num_tokens": 45915677.0,
      "step": 23470
    },
    {
      "entropy": 6.072244310379029,
      "epoch": 2.3465786974558904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004451969633593312,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17296656966209412,
      "num_tokens": 45925256.0,
      "step": 23475
    },
    {
      "entropy": 6.050121688842774,
      "epoch": 2.347078522517119,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044517361792064883,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17447609603404998,
      "num_tokens": 45933623.0,
      "step": 23480
    },
    {
      "entropy": 6.076349496841431,
      "epoch": 2.347578347578348,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044515026820044195,
      "loss": 5.7661,
      "mean_token_accuracy": 0.16509698182344437,
      "num_tokens": 45945226.0,
      "step": 23485
    },
    {
      "entropy": 6.108572149276734,
      "epoch": 2.348078172639576,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004451269141992982,
      "loss": 5.7032,
      "mean_token_accuracy": 0.16948267668485642,
      "num_tokens": 45954744.0,
      "step": 23490
    },
    {
      "entropy": 6.056755304336548,
      "epoch": 2.3485779977008048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044510355591780503,
      "loss": 5.6595,
      "mean_token_accuracy": 0.1756557673215866,
      "num_tokens": 45964216.0,
      "step": 23495
    },
    {
      "entropy": 6.046732950210571,
      "epoch": 2.3490778227620335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004450801933565502,
      "loss": 5.7164,
      "mean_token_accuracy": 0.17080431431531906,
      "num_tokens": 45974342.0,
      "step": 23500
    },
    {
      "entropy": 6.039375257492066,
      "epoch": 2.3495776478232617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004450568265161214,
      "loss": 5.6231,
      "mean_token_accuracy": 0.1766813427209854,
      "num_tokens": 45983392.0,
      "step": 23505
    },
    {
      "entropy": 6.094000625610351,
      "epoch": 2.3500774728844904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044503345539710664,
      "loss": 5.7035,
      "mean_token_accuracy": 0.17260759323835373,
      "num_tokens": 45993717.0,
      "step": 23510
    },
    {
      "entropy": 6.038467168807983,
      "epoch": 2.350577297945719,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004450100800000939,
      "loss": 5.6687,
      "mean_token_accuracy": 0.1789119854569435,
      "num_tokens": 46003780.0,
      "step": 23515
    },
    {
      "entropy": 6.0903528213500975,
      "epoch": 2.3510771230069474,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004449867003256713,
      "loss": 5.7579,
      "mean_token_accuracy": 0.1640891656279564,
      "num_tokens": 46015003.0,
      "step": 23520
    },
    {
      "entropy": 6.035603141784668,
      "epoch": 2.351576948068176,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044496331637442697,
      "loss": 5.6811,
      "mean_token_accuracy": 0.1649954304099083,
      "num_tokens": 46025425.0,
      "step": 23525
    },
    {
      "entropy": 6.105323791503906,
      "epoch": 2.3520767731294048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004449399281469495,
      "loss": 5.721,
      "mean_token_accuracy": 0.1745951369404793,
      "num_tokens": 46035978.0,
      "step": 23530
    },
    {
      "entropy": 6.069410085678101,
      "epoch": 2.3525765981906335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044491653564382704,
      "loss": 5.685,
      "mean_token_accuracy": 0.17277275174856185,
      "num_tokens": 46046184.0,
      "step": 23535
    },
    {
      "entropy": 6.074217176437378,
      "epoch": 2.3530764232518617,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004448931388656483,
      "loss": 5.6123,
      "mean_token_accuracy": 0.1717921167612076,
      "num_tokens": 46057274.0,
      "step": 23540
    },
    {
      "entropy": 6.152638101577759,
      "epoch": 2.3535762483130904,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044486973781300196,
      "loss": 5.8044,
      "mean_token_accuracy": 0.1620525598526001,
      "num_tokens": 46067302.0,
      "step": 23545
    },
    {
      "entropy": 6.037632894515991,
      "epoch": 2.354076073374319,
      "grad_norm": 1.125,
      "learning_rate": 0.00044484633248647665,
      "loss": 5.6984,
      "mean_token_accuracy": 0.1702074110507965,
      "num_tokens": 46078076.0,
      "step": 23550
    },
    {
      "entropy": 6.0329851627349855,
      "epoch": 2.3545758984355474,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044482292288666134,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17584035843610762,
      "num_tokens": 46088008.0,
      "step": 23555
    },
    {
      "entropy": 6.0814415454864506,
      "epoch": 2.355075723496776,
      "grad_norm": 1.15625,
      "learning_rate": 0.000444799509014145,
      "loss": 5.6909,
      "mean_token_accuracy": 0.1701152428984642,
      "num_tokens": 46097418.0,
      "step": 23560
    },
    {
      "entropy": 6.005620050430298,
      "epoch": 2.3555755485580048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004447760908695167,
      "loss": 5.6414,
      "mean_token_accuracy": 0.1771276667714119,
      "num_tokens": 46106906.0,
      "step": 23565
    },
    {
      "entropy": 6.015163707733154,
      "epoch": 2.356075373619233,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044475266845336566,
      "loss": 5.6582,
      "mean_token_accuracy": 0.17504483312368393,
      "num_tokens": 46116859.0,
      "step": 23570
    },
    {
      "entropy": 6.02967209815979,
      "epoch": 2.3565751986804617,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004447292417662811,
      "loss": 5.7134,
      "mean_token_accuracy": 0.1715968668460846,
      "num_tokens": 46127633.0,
      "step": 23575
    },
    {
      "entropy": 6.073098134994507,
      "epoch": 2.3570750237416904,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004447058108088525,
      "loss": 5.6583,
      "mean_token_accuracy": 0.17284958958625793,
      "num_tokens": 46136539.0,
      "step": 23580
    },
    {
      "entropy": 6.057160711288452,
      "epoch": 2.357574848802919,
      "grad_norm": 1.125,
      "learning_rate": 0.00044468237558166945,
      "loss": 5.6869,
      "mean_token_accuracy": 0.16955903470516204,
      "num_tokens": 46147204.0,
      "step": 23585
    },
    {
      "entropy": 6.014602994918823,
      "epoch": 2.3580746738641474,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044465893608532146,
      "loss": 5.6241,
      "mean_token_accuracy": 0.18116725534200667,
      "num_tokens": 46157638.0,
      "step": 23590
    },
    {
      "entropy": 6.097106313705444,
      "epoch": 2.358574498925376,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044463549232039824,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17373716980218887,
      "num_tokens": 46167820.0,
      "step": 23595
    },
    {
      "entropy": 6.097931957244873,
      "epoch": 2.3590743239866048,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004446120442874898,
      "loss": 5.6829,
      "mean_token_accuracy": 0.17294856756925583,
      "num_tokens": 46177964.0,
      "step": 23600
    },
    {
      "entropy": 6.061763000488281,
      "epoch": 2.359574149047833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004445885919871859,
      "loss": 5.6895,
      "mean_token_accuracy": 0.17141249477863313,
      "num_tokens": 46187265.0,
      "step": 23605
    },
    {
      "entropy": 6.032730436325073,
      "epoch": 2.3600739741090617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004445651354200766,
      "loss": 5.6866,
      "mean_token_accuracy": 0.1744834452867508,
      "num_tokens": 46196683.0,
      "step": 23610
    },
    {
      "entropy": 5.983437967300415,
      "epoch": 2.3605737991702904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044454167458675225,
      "loss": 5.5657,
      "mean_token_accuracy": 0.1839605912566185,
      "num_tokens": 46205619.0,
      "step": 23615
    },
    {
      "entropy": 6.0360715866088865,
      "epoch": 2.361073624231519,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044451820948780295,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17571811825037004,
      "num_tokens": 46215121.0,
      "step": 23620
    },
    {
      "entropy": 6.012126350402832,
      "epoch": 2.3615734492927474,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044449474012381903,
      "loss": 5.6036,
      "mean_token_accuracy": 0.1813683643937111,
      "num_tokens": 46224773.0,
      "step": 23625
    },
    {
      "entropy": 6.026264905929565,
      "epoch": 2.362073274353976,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044447126649539123,
      "loss": 5.7001,
      "mean_token_accuracy": 0.17422660291194916,
      "num_tokens": 46234297.0,
      "step": 23630
    },
    {
      "entropy": 6.020096635818481,
      "epoch": 2.3625730994152048,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004444477886031098,
      "loss": 5.6335,
      "mean_token_accuracy": 0.17862877994775772,
      "num_tokens": 46243947.0,
      "step": 23635
    },
    {
      "entropy": 6.0570197105407715,
      "epoch": 2.363072924476433,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004444243064475657,
      "loss": 5.6048,
      "mean_token_accuracy": 0.1733655497431755,
      "num_tokens": 46253604.0,
      "step": 23640
    },
    {
      "entropy": 6.055138158798218,
      "epoch": 2.3635727495376617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004444008200293496,
      "loss": 5.6643,
      "mean_token_accuracy": 0.18190052211284638,
      "num_tokens": 46263460.0,
      "step": 23645
    },
    {
      "entropy": 6.1363365650177,
      "epoch": 2.3640725745988904,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044437732934905256,
      "loss": 5.7415,
      "mean_token_accuracy": 0.17146467864513398,
      "num_tokens": 46272780.0,
      "step": 23650
    },
    {
      "entropy": 6.039152050018311,
      "epoch": 2.364572399660119,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004443538344072654,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18557597398757936,
      "num_tokens": 46283278.0,
      "step": 23655
    },
    {
      "entropy": 5.980262994766235,
      "epoch": 2.3650722247213474,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044433033520457937,
      "loss": 5.6379,
      "mean_token_accuracy": 0.18031202256679535,
      "num_tokens": 46291900.0,
      "step": 23660
    },
    {
      "entropy": 6.005715370178223,
      "epoch": 2.365572049782576,
      "grad_norm": 1.375,
      "learning_rate": 0.0004443068317415857,
      "loss": 5.6539,
      "mean_token_accuracy": 0.16810369342565537,
      "num_tokens": 46301517.0,
      "step": 23665
    },
    {
      "entropy": 6.085965919494629,
      "epoch": 2.3660718748438048,
      "grad_norm": 1.46875,
      "learning_rate": 0.00044428332401887566,
      "loss": 5.701,
      "mean_token_accuracy": 0.1735311672091484,
      "num_tokens": 46311267.0,
      "step": 23670
    },
    {
      "entropy": 6.0854298114776615,
      "epoch": 2.366571699905033,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004442598120370408,
      "loss": 5.7378,
      "mean_token_accuracy": 0.1687912181019783,
      "num_tokens": 46320280.0,
      "step": 23675
    },
    {
      "entropy": 6.055421161651611,
      "epoch": 2.3670715249662617,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004442362957966726,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17941712737083435,
      "num_tokens": 46328566.0,
      "step": 23680
    },
    {
      "entropy": 6.1079882144927975,
      "epoch": 2.3675713500274904,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044421277529836276,
      "loss": 5.7572,
      "mean_token_accuracy": 0.17267882376909255,
      "num_tokens": 46337586.0,
      "step": 23685
    },
    {
      "entropy": 6.023986291885376,
      "epoch": 2.368071175088719,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004441892505427031,
      "loss": 5.6105,
      "mean_token_accuracy": 0.17789245545864105,
      "num_tokens": 46346505.0,
      "step": 23690
    },
    {
      "entropy": 5.975319147109985,
      "epoch": 2.3685710001499474,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004441657215302853,
      "loss": 5.5194,
      "mean_token_accuracy": 0.18700728863477706,
      "num_tokens": 46355809.0,
      "step": 23695
    },
    {
      "entropy": 6.029742574691772,
      "epoch": 2.369070825211176,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044414218826170155,
      "loss": 5.6446,
      "mean_token_accuracy": 0.1746444061398506,
      "num_tokens": 46366844.0,
      "step": 23700
    },
    {
      "entropy": 6.071159601211548,
      "epoch": 2.3695706502724048,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004441186507375439,
      "loss": 5.7934,
      "mean_token_accuracy": 0.1637785941362381,
      "num_tokens": 46376762.0,
      "step": 23705
    },
    {
      "entropy": 6.078175115585327,
      "epoch": 2.370070475333633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044409510895840446,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17225260734558107,
      "num_tokens": 46385716.0,
      "step": 23710
    },
    {
      "entropy": 6.008757448196411,
      "epoch": 2.3705703003948617,
      "grad_norm": 1.25,
      "learning_rate": 0.00044407156292487553,
      "loss": 5.6063,
      "mean_token_accuracy": 0.18035060465335845,
      "num_tokens": 46394442.0,
      "step": 23715
    },
    {
      "entropy": 5.9554516792297365,
      "epoch": 2.3710701254560904,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004440480126375497,
      "loss": 5.6656,
      "mean_token_accuracy": 0.172440804541111,
      "num_tokens": 46405380.0,
      "step": 23720
    },
    {
      "entropy": 6.0448887825012205,
      "epoch": 2.371569950517319,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004440244580970193,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17637459486722945,
      "num_tokens": 46416134.0,
      "step": 23725
    },
    {
      "entropy": 6.143343210220337,
      "epoch": 2.3720697755785474,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044400089930387703,
      "loss": 5.7138,
      "mean_token_accuracy": 0.1717278093099594,
      "num_tokens": 46425793.0,
      "step": 23730
    },
    {
      "entropy": 6.051113510131836,
      "epoch": 2.372569600639776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044397733625871566,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17600757926702498,
      "num_tokens": 46435548.0,
      "step": 23735
    },
    {
      "entropy": 6.0395244598388675,
      "epoch": 2.3730694257010048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044395376896212805,
      "loss": 5.6846,
      "mean_token_accuracy": 0.17226164191961288,
      "num_tokens": 46445674.0,
      "step": 23740
    },
    {
      "entropy": 6.090441083908081,
      "epoch": 2.373569250762233,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044393019741470696,
      "loss": 5.6235,
      "mean_token_accuracy": 0.17673216462135316,
      "num_tokens": 46455711.0,
      "step": 23745
    },
    {
      "entropy": 6.057266807556152,
      "epoch": 2.3740690758234617,
      "grad_norm": 1.8984375,
      "learning_rate": 0.0004439066216170456,
      "loss": 5.6275,
      "mean_token_accuracy": 0.1782774582505226,
      "num_tokens": 46466998.0,
      "step": 23750
    },
    {
      "entropy": 6.07445592880249,
      "epoch": 2.3745689008846904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044388304156973713,
      "loss": 5.5763,
      "mean_token_accuracy": 0.18300435096025466,
      "num_tokens": 46475802.0,
      "step": 23755
    },
    {
      "entropy": 6.021193599700927,
      "epoch": 2.375068725945919,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044385945727337477,
      "loss": 5.712,
      "mean_token_accuracy": 0.18041964024305343,
      "num_tokens": 46485678.0,
      "step": 23760
    },
    {
      "entropy": 5.9908106327056885,
      "epoch": 2.3755685510071474,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000443835868728552,
      "loss": 5.5636,
      "mean_token_accuracy": 0.17858246564865113,
      "num_tokens": 46495062.0,
      "step": 23765
    },
    {
      "entropy": 6.0981697082519535,
      "epoch": 2.376068376068376,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004438122759358621,
      "loss": 5.6869,
      "mean_token_accuracy": 0.17095334976911544,
      "num_tokens": 46503588.0,
      "step": 23770
    },
    {
      "entropy": 6.118279647827149,
      "epoch": 2.3765682011296048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044378867889589884,
      "loss": 5.7205,
      "mean_token_accuracy": 0.16866388618946077,
      "num_tokens": 46512855.0,
      "step": 23775
    },
    {
      "entropy": 6.141739988327027,
      "epoch": 2.377068026190833,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004437650776092558,
      "loss": 5.8518,
      "mean_token_accuracy": 0.1590802937746048,
      "num_tokens": 46525243.0,
      "step": 23780
    },
    {
      "entropy": 6.092657423019409,
      "epoch": 2.3775678512520617,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044374147207652695,
      "loss": 5.6694,
      "mean_token_accuracy": 0.17486888021230698,
      "num_tokens": 46535308.0,
      "step": 23785
    },
    {
      "entropy": 6.133058357238769,
      "epoch": 2.3780676763132904,
      "grad_norm": 1.171875,
      "learning_rate": 0.000443717862298306,
      "loss": 5.7304,
      "mean_token_accuracy": 0.17031027674674987,
      "num_tokens": 46544434.0,
      "step": 23790
    },
    {
      "entropy": 6.1418312072753904,
      "epoch": 2.378567501374519,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044369424827518704,
      "loss": 5.7232,
      "mean_token_accuracy": 0.16965923756361007,
      "num_tokens": 46553588.0,
      "step": 23795
    },
    {
      "entropy": 6.109742259979248,
      "epoch": 2.3790673264357474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044367063000776423,
      "loss": 5.724,
      "mean_token_accuracy": 0.16983386129140854,
      "num_tokens": 46563383.0,
      "step": 23800
    },
    {
      "entropy": 6.0310625553131105,
      "epoch": 2.379567151496976,
      "grad_norm": 1.125,
      "learning_rate": 0.0004436470074966317,
      "loss": 5.7084,
      "mean_token_accuracy": 0.18013746216893195,
      "num_tokens": 46572837.0,
      "step": 23805
    },
    {
      "entropy": 6.085662508010865,
      "epoch": 2.3800669765582048,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000443623380742384,
      "loss": 5.6607,
      "mean_token_accuracy": 0.17736538052558898,
      "num_tokens": 46582040.0,
      "step": 23810
    },
    {
      "entropy": 6.121455287933349,
      "epoch": 2.380566801619433,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004435997497456153,
      "loss": 5.6851,
      "mean_token_accuracy": 0.1818220004439354,
      "num_tokens": 46592785.0,
      "step": 23815
    },
    {
      "entropy": 6.111083507537842,
      "epoch": 2.3810666266806617,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004435761145069203,
      "loss": 5.652,
      "mean_token_accuracy": 0.17141897082328797,
      "num_tokens": 46603625.0,
      "step": 23820
    },
    {
      "entropy": 6.13180775642395,
      "epoch": 2.3815664517418904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044355247502689374,
      "loss": 5.6939,
      "mean_token_accuracy": 0.1742783322930336,
      "num_tokens": 46612705.0,
      "step": 23825
    },
    {
      "entropy": 6.029627084732056,
      "epoch": 2.382066276803119,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004435288313061302,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17568543404340745,
      "num_tokens": 46623178.0,
      "step": 23830
    },
    {
      "entropy": 5.972030353546143,
      "epoch": 2.3825661018643474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004435051833452246,
      "loss": 5.592,
      "mean_token_accuracy": 0.173100808262825,
      "num_tokens": 46633063.0,
      "step": 23835
    },
    {
      "entropy": 6.092933464050293,
      "epoch": 2.383065926925576,
      "grad_norm": 1.203125,
      "learning_rate": 0.000443481531144772,
      "loss": 5.6598,
      "mean_token_accuracy": 0.1784608244895935,
      "num_tokens": 46641830.0,
      "step": 23840
    },
    {
      "entropy": 6.114895963668824,
      "epoch": 2.3835657519868048,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004434578747053674,
      "loss": 5.6562,
      "mean_token_accuracy": 0.16600145548582076,
      "num_tokens": 46651421.0,
      "step": 23845
    },
    {
      "entropy": 5.9578510284423825,
      "epoch": 2.384065577048033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044343421402760606,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18203773051500322,
      "num_tokens": 46661453.0,
      "step": 23850
    },
    {
      "entropy": 6.0850615978240965,
      "epoch": 2.3845654021092617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004434105491120832,
      "loss": 5.7002,
      "mean_token_accuracy": 0.17171524688601494,
      "num_tokens": 46671355.0,
      "step": 23855
    },
    {
      "entropy": 6.110534620285034,
      "epoch": 2.3850652271704904,
      "grad_norm": 1.3125,
      "learning_rate": 0.00044338687995939424,
      "loss": 5.7304,
      "mean_token_accuracy": 0.17297100126743317,
      "num_tokens": 46680966.0,
      "step": 23860
    },
    {
      "entropy": 6.081911087036133,
      "epoch": 2.385565052231719,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004433632065701348,
      "loss": 5.714,
      "mean_token_accuracy": 0.16984258890151976,
      "num_tokens": 46691350.0,
      "step": 23865
    },
    {
      "entropy": 6.060596132278443,
      "epoch": 2.3860648772929474,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004433395289449003,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1693006470799446,
      "num_tokens": 46701324.0,
      "step": 23870
    },
    {
      "entropy": 6.04092493057251,
      "epoch": 2.386564702354176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004433158470842867,
      "loss": 5.6613,
      "mean_token_accuracy": 0.17517112493515014,
      "num_tokens": 46710713.0,
      "step": 23875
    },
    {
      "entropy": 5.914649724960327,
      "epoch": 2.3870645274154048,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004432921609888896,
      "loss": 5.5032,
      "mean_token_accuracy": 0.18964454382658005,
      "num_tokens": 46720798.0,
      "step": 23880
    },
    {
      "entropy": 6.058990526199341,
      "epoch": 2.387564352476633,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044326847065930507,
      "loss": 5.7438,
      "mean_token_accuracy": 0.1786477118730545,
      "num_tokens": 46730337.0,
      "step": 23885
    },
    {
      "entropy": 6.086106538772583,
      "epoch": 2.3880641775378617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044324477609612907,
      "loss": 5.7571,
      "mean_token_accuracy": 0.16690945029258727,
      "num_tokens": 46740457.0,
      "step": 23890
    },
    {
      "entropy": 6.058048248291016,
      "epoch": 2.3885640025990904,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004432210772999578,
      "loss": 5.6652,
      "mean_token_accuracy": 0.1711254045367241,
      "num_tokens": 46749764.0,
      "step": 23895
    },
    {
      "entropy": 6.093569993972778,
      "epoch": 2.389063827660319,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004431973742713876,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17579243928194047,
      "num_tokens": 46760343.0,
      "step": 23900
    },
    {
      "entropy": 6.118886661529541,
      "epoch": 2.3895636527215474,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004431736670110146,
      "loss": 5.723,
      "mean_token_accuracy": 0.17064702063798903,
      "num_tokens": 46770187.0,
      "step": 23905
    },
    {
      "entropy": 6.0594617366790775,
      "epoch": 2.390063477782776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044314995551943554,
      "loss": 5.6731,
      "mean_token_accuracy": 0.17807217836380004,
      "num_tokens": 46779984.0,
      "step": 23910
    },
    {
      "entropy": 5.996427249908447,
      "epoch": 2.3905633028440048,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004431262397972468,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18239436149597169,
      "num_tokens": 46788940.0,
      "step": 23915
    },
    {
      "entropy": 6.096741151809693,
      "epoch": 2.391063127905233,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044310251984504517,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17439302206039428,
      "num_tokens": 46798606.0,
      "step": 23920
    },
    {
      "entropy": 6.110261154174805,
      "epoch": 2.3915629529664617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044307879566342736,
      "loss": 5.7339,
      "mean_token_accuracy": 0.16655119508504868,
      "num_tokens": 46809366.0,
      "step": 23925
    },
    {
      "entropy": 6.11518702507019,
      "epoch": 2.3920627780276904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004430550672529903,
      "loss": 5.693,
      "mean_token_accuracy": 0.16754647344350815,
      "num_tokens": 46819174.0,
      "step": 23930
    },
    {
      "entropy": 6.043776130676269,
      "epoch": 2.392562603088919,
      "grad_norm": 1.21875,
      "learning_rate": 0.000443031334614331,
      "loss": 5.5645,
      "mean_token_accuracy": 0.18398601263761521,
      "num_tokens": 46828194.0,
      "step": 23935
    },
    {
      "entropy": 6.006314659118653,
      "epoch": 2.3930624281501474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004430075977480466,
      "loss": 5.6224,
      "mean_token_accuracy": 0.18121256679296494,
      "num_tokens": 46838038.0,
      "step": 23940
    },
    {
      "entropy": 5.952013158798218,
      "epoch": 2.393562253211376,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004429838566547343,
      "loss": 5.6251,
      "mean_token_accuracy": 0.17376225739717482,
      "num_tokens": 46847495.0,
      "step": 23945
    },
    {
      "entropy": 6.025554132461548,
      "epoch": 2.3940620782726048,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004429601113349913,
      "loss": 5.7046,
      "mean_token_accuracy": 0.17728944271802902,
      "num_tokens": 46857759.0,
      "step": 23950
    },
    {
      "entropy": 6.130538749694824,
      "epoch": 2.394561903333833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044293636178941516,
      "loss": 5.7158,
      "mean_token_accuracy": 0.16898464113473893,
      "num_tokens": 46868147.0,
      "step": 23955
    },
    {
      "entropy": 6.076845645904541,
      "epoch": 2.3950617283950617,
      "grad_norm": 1.125,
      "learning_rate": 0.0004429126080186034,
      "loss": 5.6921,
      "mean_token_accuracy": 0.17706722021102905,
      "num_tokens": 46877863.0,
      "step": 23960
    },
    {
      "entropy": 6.1012091636657715,
      "epoch": 2.3955615534562904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044288885002315354,
      "loss": 5.7131,
      "mean_token_accuracy": 0.1661419540643692,
      "num_tokens": 46887542.0,
      "step": 23965
    },
    {
      "entropy": 6.1888673305511475,
      "epoch": 2.396061378517519,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044286508780366345,
      "loss": 5.6787,
      "mean_token_accuracy": 0.1716465935111046,
      "num_tokens": 46897194.0,
      "step": 23970
    },
    {
      "entropy": 6.126384830474853,
      "epoch": 2.3965612035787474,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044284132136073097,
      "loss": 5.5925,
      "mean_token_accuracy": 0.1772977814078331,
      "num_tokens": 46905800.0,
      "step": 23975
    },
    {
      "entropy": 6.004502439498902,
      "epoch": 2.397061028639976,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004428175506949541,
      "loss": 5.657,
      "mean_token_accuracy": 0.17375702708959578,
      "num_tokens": 46915587.0,
      "step": 23980
    },
    {
      "entropy": 6.104705667495727,
      "epoch": 2.3975608537012048,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044279377580693077,
      "loss": 5.7372,
      "mean_token_accuracy": 0.1654622882604599,
      "num_tokens": 46925886.0,
      "step": 23985
    },
    {
      "entropy": 6.120051717758178,
      "epoch": 2.398060678762433,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004427699966972592,
      "loss": 5.6655,
      "mean_token_accuracy": 0.18130151629447938,
      "num_tokens": 46936640.0,
      "step": 23990
    },
    {
      "entropy": 6.1146468162536625,
      "epoch": 2.3985605038236617,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044274621336653765,
      "loss": 5.7122,
      "mean_token_accuracy": 0.17225565165281295,
      "num_tokens": 46947699.0,
      "step": 23995
    },
    {
      "entropy": 6.046917581558228,
      "epoch": 2.3990603288848904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044272242581536457,
      "loss": 5.6526,
      "mean_token_accuracy": 0.17453950941562651,
      "num_tokens": 46956723.0,
      "step": 24000
    },
    {
      "epoch": 2.3990603288848904,
      "eval_entropy": 5.8283089205519,
      "eval_loss": 5.795994758605957,
      "eval_mean_token_accuracy": 0.17726871646082576,
      "eval_num_tokens": 46956723.0,
      "eval_runtime": 23.899,
      "eval_samples_per_second": 1299.008,
      "eval_steps_per_second": 162.392,
      "step": 24000
    },
    {
      "entropy": 6.042965745925903,
      "epoch": 2.399560153946119,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044269863404433846,
      "loss": 5.6653,
      "mean_token_accuracy": 0.17608170956373215,
      "num_tokens": 46966528.0,
      "step": 24005
    },
    {
      "entropy": 6.197557735443115,
      "epoch": 2.4000599790073474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004426748380540578,
      "loss": 5.7398,
      "mean_token_accuracy": 0.1658778265118599,
      "num_tokens": 46975249.0,
      "step": 24010
    },
    {
      "entropy": 6.043919610977173,
      "epoch": 2.400559804068576,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004426510378451215,
      "loss": 5.7024,
      "mean_token_accuracy": 0.17237787395715715,
      "num_tokens": 46985351.0,
      "step": 24015
    },
    {
      "entropy": 6.060683870315552,
      "epoch": 2.4010596291298048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044262723341812805,
      "loss": 5.7057,
      "mean_token_accuracy": 0.17141251713037492,
      "num_tokens": 46995529.0,
      "step": 24020
    },
    {
      "entropy": 5.992416572570801,
      "epoch": 2.401559454191033,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004426034247736766,
      "loss": 5.5874,
      "mean_token_accuracy": 0.18093190491199493,
      "num_tokens": 47005058.0,
      "step": 24025
    },
    {
      "entropy": 6.011659049987793,
      "epoch": 2.4020592792522617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004425796119123662,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17181796282529832,
      "num_tokens": 47015389.0,
      "step": 24030
    },
    {
      "entropy": 6.147126150131226,
      "epoch": 2.4025591043134904,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044255579483479577,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17140749990940093,
      "num_tokens": 47024918.0,
      "step": 24035
    },
    {
      "entropy": 6.111945056915284,
      "epoch": 2.4030589293747187,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004425319735415647,
      "loss": 5.7706,
      "mean_token_accuracy": 0.1638035833835602,
      "num_tokens": 47035078.0,
      "step": 24040
    },
    {
      "entropy": 6.085130882263184,
      "epoch": 2.4035587544359474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004425081480332723,
      "loss": 5.6989,
      "mean_token_accuracy": 0.1764901414513588,
      "num_tokens": 47045004.0,
      "step": 24045
    },
    {
      "entropy": 6.025552701950073,
      "epoch": 2.404058579497176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000442484318310518,
      "loss": 5.5884,
      "mean_token_accuracy": 0.18073896169662476,
      "num_tokens": 47053996.0,
      "step": 24050
    },
    {
      "entropy": 6.04428882598877,
      "epoch": 2.4045584045584047,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044246048437390133,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17061605900526047,
      "num_tokens": 47063668.0,
      "step": 24055
    },
    {
      "entropy": 6.08431921005249,
      "epoch": 2.405058229619633,
      "grad_norm": 1.109375,
      "learning_rate": 0.000442436646224022,
      "loss": 5.6883,
      "mean_token_accuracy": 0.17241081446409226,
      "num_tokens": 47073125.0,
      "step": 24060
    },
    {
      "entropy": 6.122834205627441,
      "epoch": 2.4055580546808617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044241280386147965,
      "loss": 5.7607,
      "mean_token_accuracy": 0.16893422454595566,
      "num_tokens": 47083502.0,
      "step": 24065
    },
    {
      "entropy": 6.123512887954712,
      "epoch": 2.4060578797420904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044238895728687433,
      "loss": 5.6756,
      "mean_token_accuracy": 0.1751937136054039,
      "num_tokens": 47093474.0,
      "step": 24070
    },
    {
      "entropy": 5.986929416656494,
      "epoch": 2.4065577048033187,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004423651065008059,
      "loss": 5.5531,
      "mean_token_accuracy": 0.18417953401803971,
      "num_tokens": 47103349.0,
      "step": 24075
    },
    {
      "entropy": 5.971023368835449,
      "epoch": 2.4070575298645474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004423412515038744,
      "loss": 5.6305,
      "mean_token_accuracy": 0.17522869557142257,
      "num_tokens": 47112746.0,
      "step": 24080
    },
    {
      "entropy": 6.0923398494720455,
      "epoch": 2.407557354925776,
      "grad_norm": 1.125,
      "learning_rate": 0.0004423173922966802,
      "loss": 5.7316,
      "mean_token_accuracy": 0.1720352828502655,
      "num_tokens": 47122622.0,
      "step": 24085
    },
    {
      "entropy": 6.089981174468994,
      "epoch": 2.4080571799870043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044229352887982325,
      "loss": 5.6537,
      "mean_token_accuracy": 0.17346566468477248,
      "num_tokens": 47131529.0,
      "step": 24090
    },
    {
      "entropy": 6.090803766250611,
      "epoch": 2.408557005048233,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044226966125390435,
      "loss": 5.753,
      "mean_token_accuracy": 0.173018117249012,
      "num_tokens": 47141173.0,
      "step": 24095
    },
    {
      "entropy": 6.140329360961914,
      "epoch": 2.4090568301094617,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004422457894195237,
      "loss": 5.8043,
      "mean_token_accuracy": 0.17296157106757165,
      "num_tokens": 47150276.0,
      "step": 24100
    },
    {
      "entropy": 6.065801191329956,
      "epoch": 2.4095566551706904,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004422219133772821,
      "loss": 5.6784,
      "mean_token_accuracy": 0.17610106468200684,
      "num_tokens": 47159614.0,
      "step": 24105
    },
    {
      "entropy": 6.131489658355713,
      "epoch": 2.4100564802319187,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004421980331277802,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17023035883903503,
      "num_tokens": 47169082.0,
      "step": 24110
    },
    {
      "entropy": 5.969507217407227,
      "epoch": 2.4105563052931473,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044217414867161873,
      "loss": 5.5758,
      "mean_token_accuracy": 0.1921337828040123,
      "num_tokens": 47178457.0,
      "step": 24115
    },
    {
      "entropy": 6.064727687835694,
      "epoch": 2.411056130354376,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004421502600093987,
      "loss": 5.6473,
      "mean_token_accuracy": 0.1766381397843361,
      "num_tokens": 47187580.0,
      "step": 24120
    },
    {
      "entropy": 6.0598602294921875,
      "epoch": 2.4115559554156043,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044212636714172115,
      "loss": 5.6295,
      "mean_token_accuracy": 0.17606940269470214,
      "num_tokens": 47197709.0,
      "step": 24125
    },
    {
      "entropy": 6.119775581359863,
      "epoch": 2.412055780476833,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004421024700691873,
      "loss": 5.742,
      "mean_token_accuracy": 0.17031013071537018,
      "num_tokens": 47207816.0,
      "step": 24130
    },
    {
      "entropy": 6.071758699417114,
      "epoch": 2.4125556055380617,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004420785687923982,
      "loss": 5.6934,
      "mean_token_accuracy": 0.16863930821418763,
      "num_tokens": 47217434.0,
      "step": 24135
    },
    {
      "entropy": 6.018271732330322,
      "epoch": 2.4130554305992904,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004420546633119554,
      "loss": 5.7212,
      "mean_token_accuracy": 0.16969618946313858,
      "num_tokens": 47227102.0,
      "step": 24140
    },
    {
      "entropy": 5.976340293884277,
      "epoch": 2.4135552556605187,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004420307536284602,
      "loss": 5.577,
      "mean_token_accuracy": 0.1846646323800087,
      "num_tokens": 47235826.0,
      "step": 24145
    },
    {
      "entropy": 6.056925487518311,
      "epoch": 2.4140550807217473,
      "grad_norm": 1.25,
      "learning_rate": 0.00044200683974251424,
      "loss": 5.6847,
      "mean_token_accuracy": 0.1801686853170395,
      "num_tokens": 47244168.0,
      "step": 24150
    },
    {
      "entropy": 6.038318395614624,
      "epoch": 2.414554905782976,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004419829216547191,
      "loss": 5.6927,
      "mean_token_accuracy": 0.17230963259935378,
      "num_tokens": 47253437.0,
      "step": 24155
    },
    {
      "entropy": 6.0749646663665775,
      "epoch": 2.4150547308442043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004419589993656767,
      "loss": 5.6886,
      "mean_token_accuracy": 0.17216431945562363,
      "num_tokens": 47263137.0,
      "step": 24160
    },
    {
      "entropy": 6.209331083297729,
      "epoch": 2.415554555905433,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004419350728759888,
      "loss": 5.8281,
      "mean_token_accuracy": 0.1648449942469597,
      "num_tokens": 47273081.0,
      "step": 24165
    },
    {
      "entropy": 6.080365467071533,
      "epoch": 2.4160543809666617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044191114218625736,
      "loss": 5.6402,
      "mean_token_accuracy": 0.1776473730802536,
      "num_tokens": 47282500.0,
      "step": 24170
    },
    {
      "entropy": 6.038380289077759,
      "epoch": 2.4165542060278904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004418872072970846,
      "loss": 5.6021,
      "mean_token_accuracy": 0.18433862030506135,
      "num_tokens": 47291467.0,
      "step": 24175
    },
    {
      "entropy": 5.988425016403198,
      "epoch": 2.4170540310891186,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004418632682090727,
      "loss": 5.6532,
      "mean_token_accuracy": 0.1738206997513771,
      "num_tokens": 47302286.0,
      "step": 24180
    },
    {
      "entropy": 6.013473987579346,
      "epoch": 2.4175538561503473,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004418393249228238,
      "loss": 5.5859,
      "mean_token_accuracy": 0.1825210005044937,
      "num_tokens": 47311271.0,
      "step": 24185
    },
    {
      "entropy": 5.9916126251220705,
      "epoch": 2.418053681211576,
      "grad_norm": 1.25,
      "learning_rate": 0.0004418153774389405,
      "loss": 5.606,
      "mean_token_accuracy": 0.18613332659006118,
      "num_tokens": 47319909.0,
      "step": 24190
    },
    {
      "entropy": 6.02083740234375,
      "epoch": 2.4185535062728043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004417914257580252,
      "loss": 5.599,
      "mean_token_accuracy": 0.17795877754688263,
      "num_tokens": 47329422.0,
      "step": 24195
    },
    {
      "entropy": 6.046617650985718,
      "epoch": 2.419053331334033,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004417674698806805,
      "loss": 5.6848,
      "mean_token_accuracy": 0.16805751770734786,
      "num_tokens": 47341287.0,
      "step": 24200
    },
    {
      "entropy": 6.062111854553223,
      "epoch": 2.4195531563952617,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004417435098075093,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17700352221727372,
      "num_tokens": 47350550.0,
      "step": 24205
    },
    {
      "entropy": 6.096392250061035,
      "epoch": 2.4200529814564904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004417195455391142,
      "loss": 5.6841,
      "mean_token_accuracy": 0.1736200138926506,
      "num_tokens": 47359867.0,
      "step": 24210
    },
    {
      "entropy": 6.021597051620484,
      "epoch": 2.4205528065177186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044169557707609826,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1724132254719734,
      "num_tokens": 47368713.0,
      "step": 24215
    },
    {
      "entropy": 5.99743275642395,
      "epoch": 2.4210526315789473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004416716044190644,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1774180144071579,
      "num_tokens": 47378404.0,
      "step": 24220
    },
    {
      "entropy": 6.124552774429321,
      "epoch": 2.421552456640176,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000441647627568616,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17293068766593933,
      "num_tokens": 47387403.0,
      "step": 24225
    },
    {
      "entropy": 6.057734632492066,
      "epoch": 2.4220522817014043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044162364652535613,
      "loss": 5.6464,
      "mean_token_accuracy": 0.1752588555216789,
      "num_tokens": 47397322.0,
      "step": 24230
    },
    {
      "entropy": 6.04959888458252,
      "epoch": 2.422552106762633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004415996612898882,
      "loss": 5.6291,
      "mean_token_accuracy": 0.17939954549074172,
      "num_tokens": 47407827.0,
      "step": 24235
    },
    {
      "entropy": 6.0642658233642575,
      "epoch": 2.4230519318238617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044157567186281553,
      "loss": 5.7329,
      "mean_token_accuracy": 0.1720676451921463,
      "num_tokens": 47418664.0,
      "step": 24240
    },
    {
      "entropy": 5.988983583450318,
      "epoch": 2.4235517568850904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000441551678244742,
      "loss": 5.6435,
      "mean_token_accuracy": 0.17674168646335603,
      "num_tokens": 47428399.0,
      "step": 24245
    },
    {
      "entropy": 6.03638105392456,
      "epoch": 2.4240515819463186,
      "grad_norm": 1.21875,
      "learning_rate": 0.000441527680436271,
      "loss": 5.6522,
      "mean_token_accuracy": 0.17913534939289094,
      "num_tokens": 47438608.0,
      "step": 24250
    },
    {
      "entropy": 6.043403768539429,
      "epoch": 2.4245514070075473,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044150367843800645,
      "loss": 5.7155,
      "mean_token_accuracy": 0.17610953152179717,
      "num_tokens": 47447551.0,
      "step": 24255
    },
    {
      "entropy": 5.953941297531128,
      "epoch": 2.425051232068776,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004414796722505522,
      "loss": 5.5399,
      "mean_token_accuracy": 0.17878128588199615,
      "num_tokens": 47456886.0,
      "step": 24260
    },
    {
      "entropy": 6.036540222167969,
      "epoch": 2.4255510571300043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004414556618745122,
      "loss": 5.5884,
      "mean_token_accuracy": 0.1760821148753166,
      "num_tokens": 47466452.0,
      "step": 24265
    },
    {
      "entropy": 5.975663042068481,
      "epoch": 2.426050882191233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044143164731049063,
      "loss": 5.6348,
      "mean_token_accuracy": 0.18218640834093094,
      "num_tokens": 47477390.0,
      "step": 24270
    },
    {
      "entropy": 6.018576002120971,
      "epoch": 2.4265507072524617,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004414076285590916,
      "loss": 5.5941,
      "mean_token_accuracy": 0.1825854793190956,
      "num_tokens": 47487198.0,
      "step": 24275
    },
    {
      "entropy": 6.027209901809693,
      "epoch": 2.4270505323136904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004413836056209195,
      "loss": 5.5774,
      "mean_token_accuracy": 0.18375833779573442,
      "num_tokens": 47496597.0,
      "step": 24280
    },
    {
      "entropy": 5.923119831085205,
      "epoch": 2.4275503573749186,
      "grad_norm": 1.125,
      "learning_rate": 0.00044135957849657866,
      "loss": 5.5522,
      "mean_token_accuracy": 0.1854797899723053,
      "num_tokens": 47506658.0,
      "step": 24285
    },
    {
      "entropy": 6.0781087398529055,
      "epoch": 2.4280501824361473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044133554718667363,
      "loss": 5.6818,
      "mean_token_accuracy": 0.16987173408269882,
      "num_tokens": 47516830.0,
      "step": 24290
    },
    {
      "entropy": 6.004931163787842,
      "epoch": 2.428550007497376,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044131151169180905,
      "loss": 5.5687,
      "mean_token_accuracy": 0.17832327336072923,
      "num_tokens": 47526051.0,
      "step": 24295
    },
    {
      "entropy": 6.018750429153442,
      "epoch": 2.4290498325586043,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044128747201258956,
      "loss": 5.7022,
      "mean_token_accuracy": 0.17027734220027924,
      "num_tokens": 47535958.0,
      "step": 24300
    },
    {
      "entropy": 6.119160747528076,
      "epoch": 2.429549657619833,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044126342814962015,
      "loss": 5.6757,
      "mean_token_accuracy": 0.17158644199371337,
      "num_tokens": 47545596.0,
      "step": 24305
    },
    {
      "entropy": 6.12937822341919,
      "epoch": 2.4300494826810617,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044123938010350575,
      "loss": 5.7434,
      "mean_token_accuracy": 0.1694709315896034,
      "num_tokens": 47555143.0,
      "step": 24310
    },
    {
      "entropy": 6.066638326644897,
      "epoch": 2.4305493077422904,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044121532787485114,
      "loss": 5.6271,
      "mean_token_accuracy": 0.1813995882868767,
      "num_tokens": 47564676.0,
      "step": 24315
    },
    {
      "entropy": 6.007778072357178,
      "epoch": 2.4310491328035186,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044119127146426167,
      "loss": 5.6441,
      "mean_token_accuracy": 0.1842271700501442,
      "num_tokens": 47574082.0,
      "step": 24320
    },
    {
      "entropy": 6.01358437538147,
      "epoch": 2.4315489578647473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044116721087234266,
      "loss": 5.637,
      "mean_token_accuracy": 0.18669049590826034,
      "num_tokens": 47583886.0,
      "step": 24325
    },
    {
      "entropy": 6.110100841522216,
      "epoch": 2.432048782925976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004411431460996994,
      "loss": 5.7724,
      "mean_token_accuracy": 0.16440752297639846,
      "num_tokens": 47593428.0,
      "step": 24330
    },
    {
      "entropy": 6.036356496810913,
      "epoch": 2.4325486079872043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004411190771469373,
      "loss": 5.6351,
      "mean_token_accuracy": 0.1805667519569397,
      "num_tokens": 47602321.0,
      "step": 24335
    },
    {
      "entropy": 6.089473152160645,
      "epoch": 2.433048433048433,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044109500401466194,
      "loss": 5.6286,
      "mean_token_accuracy": 0.17483365833759307,
      "num_tokens": 47611231.0,
      "step": 24340
    },
    {
      "entropy": 6.204289817810059,
      "epoch": 2.4335482581096617,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004410709267034791,
      "loss": 5.7729,
      "mean_token_accuracy": 0.16689778566360475,
      "num_tokens": 47622606.0,
      "step": 24345
    },
    {
      "entropy": 6.074892282485962,
      "epoch": 2.4340480831708904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044104684521399434,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17702654004096985,
      "num_tokens": 47632054.0,
      "step": 24350
    },
    {
      "entropy": 6.152192258834839,
      "epoch": 2.4345479082321186,
      "grad_norm": 1.625,
      "learning_rate": 0.0004410227595468138,
      "loss": 5.8524,
      "mean_token_accuracy": 0.1603536143898964,
      "num_tokens": 47642318.0,
      "step": 24355
    },
    {
      "entropy": 6.111044216156006,
      "epoch": 2.4350477332933473,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044099866970254327,
      "loss": 5.689,
      "mean_token_accuracy": 0.17418668866157533,
      "num_tokens": 47652467.0,
      "step": 24360
    },
    {
      "entropy": 6.104608058929443,
      "epoch": 2.435547558354576,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044097457568178903,
      "loss": 5.7092,
      "mean_token_accuracy": 0.16408001780509948,
      "num_tokens": 47661729.0,
      "step": 24365
    },
    {
      "entropy": 5.86772518157959,
      "epoch": 2.4360473834158043,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044095047748515706,
      "loss": 5.4078,
      "mean_token_accuracy": 0.20024411529302596,
      "num_tokens": 47670937.0,
      "step": 24370
    },
    {
      "entropy": 5.9670439720153805,
      "epoch": 2.436547208477033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044092637511325384,
      "loss": 5.6377,
      "mean_token_accuracy": 0.17574608623981475,
      "num_tokens": 47680414.0,
      "step": 24375
    },
    {
      "entropy": 6.006484127044677,
      "epoch": 2.4370470335382617,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004409022685666857,
      "loss": 5.6382,
      "mean_token_accuracy": 0.18162106424570085,
      "num_tokens": 47689381.0,
      "step": 24380
    },
    {
      "entropy": 6.111452913284301,
      "epoch": 2.4375468585994904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004408781578460592,
      "loss": 5.6638,
      "mean_token_accuracy": 0.18204910904169083,
      "num_tokens": 47698810.0,
      "step": 24385
    },
    {
      "entropy": 6.0291859149932865,
      "epoch": 2.4380466836607186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044085404295198097,
      "loss": 5.6203,
      "mean_token_accuracy": 0.17739759683609008,
      "num_tokens": 47708396.0,
      "step": 24390
    },
    {
      "entropy": 6.03398289680481,
      "epoch": 2.4385465087219473,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004408299238850576,
      "loss": 5.6085,
      "mean_token_accuracy": 0.1745418578386307,
      "num_tokens": 47717487.0,
      "step": 24395
    },
    {
      "entropy": 6.007273626327515,
      "epoch": 2.439046333783176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004408058006458961,
      "loss": 5.7183,
      "mean_token_accuracy": 0.17155901193618775,
      "num_tokens": 47727717.0,
      "step": 24400
    },
    {
      "entropy": 5.873701286315918,
      "epoch": 2.4395461588444043,
      "grad_norm": 2.46875,
      "learning_rate": 0.0004407816732351032,
      "loss": 5.4847,
      "mean_token_accuracy": 0.1867561161518097,
      "num_tokens": 47738129.0,
      "step": 24405
    },
    {
      "entropy": 5.984415340423584,
      "epoch": 2.440045983905633,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044075754165328614,
      "loss": 5.6642,
      "mean_token_accuracy": 0.1789252981543541,
      "num_tokens": 47748130.0,
      "step": 24410
    },
    {
      "entropy": 6.057972860336304,
      "epoch": 2.4405458089668617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000440733405901052,
      "loss": 5.6822,
      "mean_token_accuracy": 0.17312609702348708,
      "num_tokens": 47758066.0,
      "step": 24415
    },
    {
      "entropy": 6.043606328964233,
      "epoch": 2.4410456340280904,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044070926597900804,
      "loss": 5.577,
      "mean_token_accuracy": 0.18207037299871445,
      "num_tokens": 47767771.0,
      "step": 24420
    },
    {
      "entropy": 6.079096555709839,
      "epoch": 2.4415454590893186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004406851218877615,
      "loss": 5.6343,
      "mean_token_accuracy": 0.17277196943759918,
      "num_tokens": 47777893.0,
      "step": 24425
    },
    {
      "entropy": 6.037047243118286,
      "epoch": 2.4420452841505473,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044066097362792,
      "loss": 5.5636,
      "mean_token_accuracy": 0.18002692461013795,
      "num_tokens": 47787804.0,
      "step": 24430
    },
    {
      "entropy": 5.992109584808349,
      "epoch": 2.442545109211776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000440636821200091,
      "loss": 5.6145,
      "mean_token_accuracy": 0.18433353304862976,
      "num_tokens": 47796644.0,
      "step": 24435
    },
    {
      "entropy": 5.9571301460266115,
      "epoch": 2.4430449342730043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004406126646048822,
      "loss": 5.5692,
      "mean_token_accuracy": 0.18582808524370192,
      "num_tokens": 47806918.0,
      "step": 24440
    },
    {
      "entropy": 6.050378656387329,
      "epoch": 2.443544759334233,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004405885038429014,
      "loss": 5.5802,
      "mean_token_accuracy": 0.19132273644208908,
      "num_tokens": 47817167.0,
      "step": 24445
    },
    {
      "entropy": 6.06876130104065,
      "epoch": 2.4440445843954617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044056433891475645,
      "loss": 5.6917,
      "mean_token_accuracy": 0.1757412701845169,
      "num_tokens": 47826363.0,
      "step": 24450
    },
    {
      "entropy": 5.984944772720337,
      "epoch": 2.4445444094566904,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004405401698210553,
      "loss": 5.6114,
      "mean_token_accuracy": 0.16856224611401557,
      "num_tokens": 47837427.0,
      "step": 24455
    },
    {
      "entropy": 6.030162143707275,
      "epoch": 2.4450442345179186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004405159965624061,
      "loss": 5.7358,
      "mean_token_accuracy": 0.17041965574026108,
      "num_tokens": 47846619.0,
      "step": 24460
    },
    {
      "entropy": 6.0926696300506595,
      "epoch": 2.4455440595791473,
      "grad_norm": 1.34375,
      "learning_rate": 0.000440491819139417,
      "loss": 5.6997,
      "mean_token_accuracy": 0.1669275239109993,
      "num_tokens": 47856577.0,
      "step": 24465
    },
    {
      "entropy": 6.025983905792236,
      "epoch": 2.446043884640376,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004404676375526963,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18151356130838395,
      "num_tokens": 47865855.0,
      "step": 24470
    },
    {
      "entropy": 6.038949346542358,
      "epoch": 2.4465437097016043,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044044345180285245,
      "loss": 5.6349,
      "mean_token_accuracy": 0.17816444337368012,
      "num_tokens": 47875074.0,
      "step": 24475
    },
    {
      "entropy": 6.03907618522644,
      "epoch": 2.447043534762833,
      "grad_norm": 1.125,
      "learning_rate": 0.000440419261890494,
      "loss": 5.7237,
      "mean_token_accuracy": 0.1641565218567848,
      "num_tokens": 47885159.0,
      "step": 24480
    },
    {
      "entropy": 5.968872976303101,
      "epoch": 2.4475433598240617,
      "grad_norm": 1.390625,
      "learning_rate": 0.00044039506781622936,
      "loss": 5.6707,
      "mean_token_accuracy": 0.18110481351614,
      "num_tokens": 47895082.0,
      "step": 24485
    },
    {
      "entropy": 6.030393075942993,
      "epoch": 2.4480431848852904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004403708695806674,
      "loss": 5.6072,
      "mean_token_accuracy": 0.18806161135435104,
      "num_tokens": 47905428.0,
      "step": 24490
    },
    {
      "entropy": 6.032662582397461,
      "epoch": 2.4485430099465186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004403466671844169,
      "loss": 5.4484,
      "mean_token_accuracy": 0.20029826164245607,
      "num_tokens": 47914803.0,
      "step": 24495
    },
    {
      "entropy": 6.024805164337158,
      "epoch": 2.4490428350077473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004403224606280869,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17402581125497818,
      "num_tokens": 47924276.0,
      "step": 24500
    },
    {
      "entropy": 6.057995986938477,
      "epoch": 2.449542660068976,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004402982499122862,
      "loss": 5.7105,
      "mean_token_accuracy": 0.17116962671279906,
      "num_tokens": 47934980.0,
      "step": 24505
    },
    {
      "entropy": 6.06968674659729,
      "epoch": 2.4500424851302043,
      "grad_norm": 0.9921875,
      "learning_rate": 0.00044027403503762416,
      "loss": 5.6866,
      "mean_token_accuracy": 0.16838560923933982,
      "num_tokens": 47944925.0,
      "step": 24510
    },
    {
      "entropy": 6.0958404541015625,
      "epoch": 2.450542310191433,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004402498160047099,
      "loss": 5.6879,
      "mean_token_accuracy": 0.17333202958106994,
      "num_tokens": 47954324.0,
      "step": 24515
    },
    {
      "entropy": 6.061291694641113,
      "epoch": 2.4510421352526617,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004402255928141528,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1782301962375641,
      "num_tokens": 47964333.0,
      "step": 24520
    },
    {
      "entropy": 6.090138626098633,
      "epoch": 2.4515419603138904,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004402013654665623,
      "loss": 5.7356,
      "mean_token_accuracy": 0.16270026862621306,
      "num_tokens": 47973790.0,
      "step": 24525
    },
    {
      "entropy": 6.039667844772339,
      "epoch": 2.4520417853751186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044017713396254796,
      "loss": 5.7031,
      "mean_token_accuracy": 0.17556628435850144,
      "num_tokens": 47983756.0,
      "step": 24530
    },
    {
      "entropy": 6.0816093444824215,
      "epoch": 2.4525416104363473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044015289830271945,
      "loss": 5.6859,
      "mean_token_accuracy": 0.16868012100458146,
      "num_tokens": 47993605.0,
      "step": 24535
    },
    {
      "entropy": 6.046799850463867,
      "epoch": 2.453041435497576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044012865848768644,
      "loss": 5.6541,
      "mean_token_accuracy": 0.17736852914094925,
      "num_tokens": 48003553.0,
      "step": 24540
    },
    {
      "entropy": 6.089321517944336,
      "epoch": 2.4535412605588043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044010441451805896,
      "loss": 5.7569,
      "mean_token_accuracy": 0.1679273799061775,
      "num_tokens": 48015074.0,
      "step": 24545
    },
    {
      "entropy": 6.102528381347656,
      "epoch": 2.454041085620033,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004400801663944468,
      "loss": 5.6851,
      "mean_token_accuracy": 0.17276339530944823,
      "num_tokens": 48024392.0,
      "step": 24550
    },
    {
      "entropy": 6.048371028900147,
      "epoch": 2.4545409106812617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044005591411746023,
      "loss": 5.6137,
      "mean_token_accuracy": 0.17515185922384263,
      "num_tokens": 48034626.0,
      "step": 24555
    },
    {
      "entropy": 6.0221342086792,
      "epoch": 2.4550407357424904,
      "grad_norm": 1.359375,
      "learning_rate": 0.00044003165768770927,
      "loss": 5.5415,
      "mean_token_accuracy": 0.18535973280668258,
      "num_tokens": 48044542.0,
      "step": 24560
    },
    {
      "entropy": 6.01475157737732,
      "epoch": 2.4555405608037186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004400073971058043,
      "loss": 5.6398,
      "mean_token_accuracy": 0.1754768267273903,
      "num_tokens": 48053951.0,
      "step": 24565
    },
    {
      "entropy": 6.100653791427613,
      "epoch": 2.4560403858649473,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043998313237235566,
      "loss": 5.7434,
      "mean_token_accuracy": 0.17010299414396285,
      "num_tokens": 48064587.0,
      "step": 24570
    },
    {
      "entropy": 6.05978946685791,
      "epoch": 2.456540210926176,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004399588634879739,
      "loss": 5.6638,
      "mean_token_accuracy": 0.17779201716184617,
      "num_tokens": 48074204.0,
      "step": 24575
    },
    {
      "entropy": 6.076124668121338,
      "epoch": 2.4570400359874043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004399345904532695,
      "loss": 5.6552,
      "mean_token_accuracy": 0.1774495631456375,
      "num_tokens": 48084617.0,
      "step": 24580
    },
    {
      "entropy": 6.120323705673218,
      "epoch": 2.457539861048633,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004399103132688533,
      "loss": 5.6722,
      "mean_token_accuracy": 0.17949090003967286,
      "num_tokens": 48094333.0,
      "step": 24585
    },
    {
      "entropy": 6.009313774108887,
      "epoch": 2.4580396861098617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043988603193533606,
      "loss": 5.6937,
      "mean_token_accuracy": 0.1757107064127922,
      "num_tokens": 48104639.0,
      "step": 24590
    },
    {
      "entropy": 5.987154769897461,
      "epoch": 2.45853951117109,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004398617464533287,
      "loss": 5.629,
      "mean_token_accuracy": 0.18079759329557418,
      "num_tokens": 48115291.0,
      "step": 24595
    },
    {
      "entropy": 6.090948438644409,
      "epoch": 2.4590393362323186,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004398374568234422,
      "loss": 5.734,
      "mean_token_accuracy": 0.17239010632038115,
      "num_tokens": 48124351.0,
      "step": 24600
    },
    {
      "entropy": 6.0368133068084715,
      "epoch": 2.4595391612935473,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004398131630462877,
      "loss": 5.6317,
      "mean_token_accuracy": 0.17778980880975723,
      "num_tokens": 48134236.0,
      "step": 24605
    },
    {
      "entropy": 6.0452805519104,
      "epoch": 2.4600389863547756,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043978886512247646,
      "loss": 5.652,
      "mean_token_accuracy": 0.17362954169511796,
      "num_tokens": 48143569.0,
      "step": 24610
    },
    {
      "entropy": 6.063014316558838,
      "epoch": 2.4605388114160043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004397645630526197,
      "loss": 5.667,
      "mean_token_accuracy": 0.1771695390343666,
      "num_tokens": 48153856.0,
      "step": 24615
    },
    {
      "entropy": 6.01398983001709,
      "epoch": 2.461038636477233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000439740256837329,
      "loss": 5.6435,
      "mean_token_accuracy": 0.1792007014155388,
      "num_tokens": 48163131.0,
      "step": 24620
    },
    {
      "entropy": 6.0610511779785154,
      "epoch": 2.4615384615384617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004397159464772158,
      "loss": 5.7263,
      "mean_token_accuracy": 0.1612721264362335,
      "num_tokens": 48174717.0,
      "step": 24625
    },
    {
      "entropy": 6.0289490699768065,
      "epoch": 2.46203828659969,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043969163197289174,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1802788570523262,
      "num_tokens": 48186207.0,
      "step": 24630
    },
    {
      "entropy": 6.064333915710449,
      "epoch": 2.4625381116609186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004396673133249687,
      "loss": 5.6604,
      "mean_token_accuracy": 0.1712229147553444,
      "num_tokens": 48196939.0,
      "step": 24635
    },
    {
      "entropy": 6.026987314224243,
      "epoch": 2.4630379367221473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004396429905340584,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17693585604429246,
      "num_tokens": 48208388.0,
      "step": 24640
    },
    {
      "entropy": 6.085429239273071,
      "epoch": 2.4635377617833756,
      "grad_norm": 1.25,
      "learning_rate": 0.00043961866360077266,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17899845391511918,
      "num_tokens": 48218272.0,
      "step": 24645
    },
    {
      "entropy": 6.047089576721191,
      "epoch": 2.4640375868446043,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004395943325257239,
      "loss": 5.6619,
      "mean_token_accuracy": 0.18169104009866716,
      "num_tokens": 48227526.0,
      "step": 24650
    },
    {
      "entropy": 6.0078144550323485,
      "epoch": 2.464537411905833,
      "grad_norm": 1.53125,
      "learning_rate": 0.00043956999730952403,
      "loss": 5.6259,
      "mean_token_accuracy": 0.17888711541891097,
      "num_tokens": 48237181.0,
      "step": 24655
    },
    {
      "entropy": 6.0129180431365965,
      "epoch": 2.4650372369670617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043954565795278536,
      "loss": 5.557,
      "mean_token_accuracy": 0.19218465238809584,
      "num_tokens": 48245338.0,
      "step": 24660
    },
    {
      "entropy": 6.03862509727478,
      "epoch": 2.46553706202829,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004395213144561203,
      "loss": 5.6913,
      "mean_token_accuracy": 0.1737093985080719,
      "num_tokens": 48254927.0,
      "step": 24665
    },
    {
      "entropy": 6.102736377716065,
      "epoch": 2.4660368870895186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004394969668201413,
      "loss": 5.719,
      "mean_token_accuracy": 0.18043704926967621,
      "num_tokens": 48266170.0,
      "step": 24670
    },
    {
      "entropy": 6.061919069290161,
      "epoch": 2.4665367121507473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004394726150454609,
      "loss": 5.6084,
      "mean_token_accuracy": 0.17653534859418868,
      "num_tokens": 48276216.0,
      "step": 24675
    },
    {
      "entropy": 6.029545688629151,
      "epoch": 2.4670365372119756,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004394482591326918,
      "loss": 5.7233,
      "mean_token_accuracy": 0.17139132618904113,
      "num_tokens": 48285677.0,
      "step": 24680
    },
    {
      "entropy": 6.074597501754761,
      "epoch": 2.4675363622732043,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043942389908244693,
      "loss": 5.6696,
      "mean_token_accuracy": 0.1774395987391472,
      "num_tokens": 48297252.0,
      "step": 24685
    },
    {
      "entropy": 6.0129622459411625,
      "epoch": 2.468036187334433,
      "grad_norm": 1.140625,
      "learning_rate": 0.000439399534895339,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18436138033866883,
      "num_tokens": 48306507.0,
      "step": 24690
    },
    {
      "entropy": 5.984614658355713,
      "epoch": 2.4685360123956617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004393751665719811,
      "loss": 5.6077,
      "mean_token_accuracy": 0.18099451661109925,
      "num_tokens": 48316323.0,
      "step": 24695
    },
    {
      "entropy": 6.044980812072754,
      "epoch": 2.46903583745689,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004393507941129863,
      "loss": 5.6468,
      "mean_token_accuracy": 0.17129335850477218,
      "num_tokens": 48326007.0,
      "step": 24700
    },
    {
      "entropy": 6.0125902652740475,
      "epoch": 2.4695356625181186,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004393264175189678,
      "loss": 5.6471,
      "mean_token_accuracy": 0.177483469247818,
      "num_tokens": 48335291.0,
      "step": 24705
    },
    {
      "entropy": 6.131001806259155,
      "epoch": 2.4700354875793473,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043930203679053885,
      "loss": 5.6973,
      "mean_token_accuracy": 0.1676548585295677,
      "num_tokens": 48345078.0,
      "step": 24710
    },
    {
      "entropy": 6.069846773147583,
      "epoch": 2.4705353126405756,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043927765192831305,
      "loss": 5.7171,
      "mean_token_accuracy": 0.1744041070342064,
      "num_tokens": 48354329.0,
      "step": 24715
    },
    {
      "entropy": 6.065669918060303,
      "epoch": 2.4710351377018043,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004392532629329038,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17765288352966307,
      "num_tokens": 48363261.0,
      "step": 24720
    },
    {
      "entropy": 6.098497295379639,
      "epoch": 2.471534962763033,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043922886980492465,
      "loss": 5.671,
      "mean_token_accuracy": 0.17505188137292862,
      "num_tokens": 48372389.0,
      "step": 24725
    },
    {
      "entropy": 6.218533229827881,
      "epoch": 2.4720347878242617,
      "grad_norm": 1.25,
      "learning_rate": 0.00043920447254498947,
      "loss": 5.7788,
      "mean_token_accuracy": 0.16726745963096618,
      "num_tokens": 48381553.0,
      "step": 24730
    },
    {
      "entropy": 6.135932588577271,
      "epoch": 2.47253461288549,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004391800711537119,
      "loss": 5.6286,
      "mean_token_accuracy": 0.1793373256921768,
      "num_tokens": 48391356.0,
      "step": 24735
    },
    {
      "entropy": 6.07943696975708,
      "epoch": 2.4730344379467186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043915566563170605,
      "loss": 5.7284,
      "mean_token_accuracy": 0.1723628595471382,
      "num_tokens": 48401092.0,
      "step": 24740
    },
    {
      "entropy": 6.063373184204101,
      "epoch": 2.4735342630079473,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043913125597958594,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16918949857354165,
      "num_tokens": 48411891.0,
      "step": 24745
    },
    {
      "entropy": 6.107413291931152,
      "epoch": 2.4740340880691756,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043910684219796564,
      "loss": 5.7052,
      "mean_token_accuracy": 0.17323433458805085,
      "num_tokens": 48421393.0,
      "step": 24750
    },
    {
      "entropy": 6.048322677612305,
      "epoch": 2.4745339131304043,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043908242428745937,
      "loss": 5.651,
      "mean_token_accuracy": 0.17287407517433168,
      "num_tokens": 48431485.0,
      "step": 24755
    },
    {
      "entropy": 6.083398628234863,
      "epoch": 2.475033738191633,
      "grad_norm": 1.25,
      "learning_rate": 0.00043905800224868153,
      "loss": 5.6755,
      "mean_token_accuracy": 0.17667129188776015,
      "num_tokens": 48440997.0,
      "step": 24760
    },
    {
      "entropy": 6.058137512207031,
      "epoch": 2.4755335632528617,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004390335760822466,
      "loss": 5.6834,
      "mean_token_accuracy": 0.1712002784013748,
      "num_tokens": 48451043.0,
      "step": 24765
    },
    {
      "entropy": 5.983648157119751,
      "epoch": 2.47603338831409,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004390091457887691,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17412691414356232,
      "num_tokens": 48460776.0,
      "step": 24770
    },
    {
      "entropy": 5.999238157272339,
      "epoch": 2.4765332133753186,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004389847113688637,
      "loss": 5.6465,
      "mean_token_accuracy": 0.1707058683037758,
      "num_tokens": 48470095.0,
      "step": 24775
    },
    {
      "entropy": 6.032614517211914,
      "epoch": 2.4770330384365473,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043896027282314516,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17214203029870986,
      "num_tokens": 48480977.0,
      "step": 24780
    },
    {
      "entropy": 5.973993301391602,
      "epoch": 2.4775328634977756,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004389358301522283,
      "loss": 5.6433,
      "mean_token_accuracy": 0.176000614464283,
      "num_tokens": 48490547.0,
      "step": 24785
    },
    {
      "entropy": 6.099792098999023,
      "epoch": 2.4780326885590043,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043891138335672825,
      "loss": 5.6709,
      "mean_token_accuracy": 0.17133862227201463,
      "num_tokens": 48500358.0,
      "step": 24790
    },
    {
      "entropy": 6.048324680328369,
      "epoch": 2.478532513620233,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004388869324372599,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17295517176389694,
      "num_tokens": 48509175.0,
      "step": 24795
    },
    {
      "entropy": 6.043154287338257,
      "epoch": 2.4790323386814617,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004388624773944385,
      "loss": 5.6991,
      "mean_token_accuracy": 0.17507774829864503,
      "num_tokens": 48518620.0,
      "step": 24800
    },
    {
      "entropy": 6.049569082260132,
      "epoch": 2.47953216374269,
      "grad_norm": 1.25,
      "learning_rate": 0.00043883801822887933,
      "loss": 5.6563,
      "mean_token_accuracy": 0.18195595741271972,
      "num_tokens": 48527609.0,
      "step": 24805
    },
    {
      "entropy": 6.097655868530273,
      "epoch": 2.4800319888039186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043881355494119773,
      "loss": 5.641,
      "mean_token_accuracy": 0.17164506167173385,
      "num_tokens": 48538002.0,
      "step": 24810
    },
    {
      "entropy": 6.06972861289978,
      "epoch": 2.4805318138651473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004387890875320093,
      "loss": 5.6205,
      "mean_token_accuracy": 0.1751384124159813,
      "num_tokens": 48546939.0,
      "step": 24815
    },
    {
      "entropy": 6.018962144851685,
      "epoch": 2.4810316389263756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043876461600192955,
      "loss": 5.6122,
      "mean_token_accuracy": 0.17949387580156326,
      "num_tokens": 48556431.0,
      "step": 24820
    },
    {
      "entropy": 5.978823280334472,
      "epoch": 2.4815314639876043,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004387401403515743,
      "loss": 5.644,
      "mean_token_accuracy": 0.17840481251478196,
      "num_tokens": 48565993.0,
      "step": 24825
    },
    {
      "entropy": 6.016033077239991,
      "epoch": 2.482031289048833,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043871566058155913,
      "loss": 5.6313,
      "mean_token_accuracy": 0.18486625254154204,
      "num_tokens": 48575324.0,
      "step": 24830
    },
    {
      "entropy": 6.0697669506073,
      "epoch": 2.4825311141100617,
      "grad_norm": 1.25,
      "learning_rate": 0.00043869117669250015,
      "loss": 5.6868,
      "mean_token_accuracy": 0.17492999136447906,
      "num_tokens": 48584432.0,
      "step": 24835
    },
    {
      "entropy": 6.096307182312012,
      "epoch": 2.48303093917129,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004386666886850132,
      "loss": 5.6908,
      "mean_token_accuracy": 0.1694542571902275,
      "num_tokens": 48595459.0,
      "step": 24840
    },
    {
      "entropy": 6.083243274688721,
      "epoch": 2.4835307642325186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043864219655971457,
      "loss": 5.6652,
      "mean_token_accuracy": 0.17644311040639876,
      "num_tokens": 48605505.0,
      "step": 24845
    },
    {
      "entropy": 6.042428731918335,
      "epoch": 2.4840305892937473,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004386177003172203,
      "loss": 5.5854,
      "mean_token_accuracy": 0.18321512192487716,
      "num_tokens": 48615237.0,
      "step": 24850
    },
    {
      "entropy": 6.096088838577271,
      "epoch": 2.4845304143549756,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043859319995814686,
      "loss": 5.6971,
      "mean_token_accuracy": 0.1738990679383278,
      "num_tokens": 48625711.0,
      "step": 24855
    },
    {
      "entropy": 6.092739725112915,
      "epoch": 2.4850302394162043,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004385686954831105,
      "loss": 5.6409,
      "mean_token_accuracy": 0.18015241026878356,
      "num_tokens": 48636091.0,
      "step": 24860
    },
    {
      "entropy": 5.943760538101197,
      "epoch": 2.485530064477433,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043854418689272793,
      "loss": 5.5242,
      "mean_token_accuracy": 0.18355729877948762,
      "num_tokens": 48645891.0,
      "step": 24865
    },
    {
      "entropy": 5.989504718780518,
      "epoch": 2.4860298895386617,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043851967418761574,
      "loss": 5.6811,
      "mean_token_accuracy": 0.17721570879220963,
      "num_tokens": 48657197.0,
      "step": 24870
    },
    {
      "entropy": 6.036539363861084,
      "epoch": 2.48652971459989,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043849515736839057,
      "loss": 5.6416,
      "mean_token_accuracy": 0.18034935891628265,
      "num_tokens": 48666325.0,
      "step": 24875
    },
    {
      "entropy": 6.079435729980469,
      "epoch": 2.4870295396611186,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043847063643566935,
      "loss": 5.6545,
      "mean_token_accuracy": 0.17652601152658462,
      "num_tokens": 48676208.0,
      "step": 24880
    },
    {
      "entropy": 6.041273593902588,
      "epoch": 2.4875293647223473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043844611139006893,
      "loss": 5.7181,
      "mean_token_accuracy": 0.17897819429636003,
      "num_tokens": 48685809.0,
      "step": 24885
    },
    {
      "entropy": 6.103718328475952,
      "epoch": 2.4880291897835756,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043842158223220644,
      "loss": 5.7463,
      "mean_token_accuracy": 0.17112945020198822,
      "num_tokens": 48696029.0,
      "step": 24890
    },
    {
      "entropy": 6.043841981887818,
      "epoch": 2.4885290148448043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000438397048962699,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18305823355913162,
      "num_tokens": 48704858.0,
      "step": 24895
    },
    {
      "entropy": 6.089008092880249,
      "epoch": 2.489028839906033,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00043837251158216386,
      "loss": 5.7333,
      "mean_token_accuracy": 0.17796538770198822,
      "num_tokens": 48715120.0,
      "step": 24900
    },
    {
      "entropy": 6.085360908508301,
      "epoch": 2.4895286649672617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043834797009121836,
      "loss": 5.6761,
      "mean_token_accuracy": 0.17476103007793425,
      "num_tokens": 48725156.0,
      "step": 24905
    },
    {
      "entropy": 6.040737724304199,
      "epoch": 2.49002849002849,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043832342449047994,
      "loss": 5.6341,
      "mean_token_accuracy": 0.18081084936857222,
      "num_tokens": 48734799.0,
      "step": 24910
    },
    {
      "entropy": 6.043331861495972,
      "epoch": 2.4905283150897186,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004382988747805661,
      "loss": 5.6569,
      "mean_token_accuracy": 0.17906107902526855,
      "num_tokens": 48744414.0,
      "step": 24915
    },
    {
      "entropy": 6.057382678985595,
      "epoch": 2.4910281401509473,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004382743209620948,
      "loss": 5.7532,
      "mean_token_accuracy": 0.17202349603176117,
      "num_tokens": 48754390.0,
      "step": 24920
    },
    {
      "entropy": 6.094738960266113,
      "epoch": 2.4915279652121756,
      "grad_norm": 1.125,
      "learning_rate": 0.0004382497630356835,
      "loss": 5.6419,
      "mean_token_accuracy": 0.174369740486145,
      "num_tokens": 48764517.0,
      "step": 24925
    },
    {
      "entropy": 6.138540458679199,
      "epoch": 2.4920277902734043,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043822520100195015,
      "loss": 5.6849,
      "mean_token_accuracy": 0.17088401317596436,
      "num_tokens": 48774767.0,
      "step": 24930
    },
    {
      "entropy": 6.107260513305664,
      "epoch": 2.492527615334633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043820063486151267,
      "loss": 5.7043,
      "mean_token_accuracy": 0.17480051442980765,
      "num_tokens": 48784279.0,
      "step": 24935
    },
    {
      "entropy": 6.089031982421875,
      "epoch": 2.4930274403958617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043817606461498933,
      "loss": 5.59,
      "mean_token_accuracy": 0.18504401594400405,
      "num_tokens": 48794179.0,
      "step": 24940
    },
    {
      "entropy": 6.017665529251099,
      "epoch": 2.49352726545709,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043815149026299823,
      "loss": 5.607,
      "mean_token_accuracy": 0.18101875633001327,
      "num_tokens": 48804593.0,
      "step": 24945
    },
    {
      "entropy": 6.009657621383667,
      "epoch": 2.4940270905183186,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004381269118061575,
      "loss": 5.5856,
      "mean_token_accuracy": 0.1840279921889305,
      "num_tokens": 48813280.0,
      "step": 24950
    },
    {
      "entropy": 6.068167018890381,
      "epoch": 2.4945269155795473,
      "grad_norm": 1.125,
      "learning_rate": 0.0004381023292450857,
      "loss": 5.6038,
      "mean_token_accuracy": 0.18123367577791213,
      "num_tokens": 48823830.0,
      "step": 24955
    },
    {
      "entropy": 5.946252059936524,
      "epoch": 2.4950267406407756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004380777425804012,
      "loss": 5.559,
      "mean_token_accuracy": 0.1831057071685791,
      "num_tokens": 48833277.0,
      "step": 24960
    },
    {
      "entropy": 6.014184427261353,
      "epoch": 2.4955265657020043,
      "grad_norm": 1.5859375,
      "learning_rate": 0.00043805315181272275,
      "loss": 5.6323,
      "mean_token_accuracy": 0.1677643060684204,
      "num_tokens": 48842718.0,
      "step": 24965
    },
    {
      "entropy": 6.08176999092102,
      "epoch": 2.496026390763233,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004380285569426689,
      "loss": 5.7513,
      "mean_token_accuracy": 0.16864261478185655,
      "num_tokens": 48852831.0,
      "step": 24970
    },
    {
      "entropy": 5.999747037887573,
      "epoch": 2.4965262158244617,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004380039579708586,
      "loss": 5.5992,
      "mean_token_accuracy": 0.18112493604421614,
      "num_tokens": 48862188.0,
      "step": 24975
    },
    {
      "entropy": 6.093376159667969,
      "epoch": 2.49702604088569,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004379793548979105,
      "loss": 5.5953,
      "mean_token_accuracy": 0.18181442469358444,
      "num_tokens": 48872817.0,
      "step": 24980
    },
    {
      "entropy": 5.994704675674439,
      "epoch": 2.4975258659469186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004379547477244439,
      "loss": 5.5987,
      "mean_token_accuracy": 0.1848782181739807,
      "num_tokens": 48882233.0,
      "step": 24985
    },
    {
      "entropy": 6.01886248588562,
      "epoch": 2.4980256910081473,
      "grad_norm": 1.125,
      "learning_rate": 0.0004379301364510777,
      "loss": 5.6722,
      "mean_token_accuracy": 0.17680893689393998,
      "num_tokens": 48893534.0,
      "step": 24990
    },
    {
      "entropy": 6.02821912765503,
      "epoch": 2.4985255160693756,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004379055210784313,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18057587444782258,
      "num_tokens": 48903067.0,
      "step": 24995
    },
    {
      "entropy": 6.007367897033691,
      "epoch": 2.4990253411306043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004378809016071238,
      "loss": 5.5664,
      "mean_token_accuracy": 0.18147855252027512,
      "num_tokens": 48911966.0,
      "step": 25000
    },
    {
      "entropy": 5.9378691673278805,
      "epoch": 2.499525166191833,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004378562780377748,
      "loss": 5.4709,
      "mean_token_accuracy": 0.19920501559972764,
      "num_tokens": 48921600.0,
      "step": 25005
    },
    {
      "entropy": 6.049672794342041,
      "epoch": 2.5000249912530617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004378316503710037,
      "loss": 5.6358,
      "mean_token_accuracy": 0.1807971179485321,
      "num_tokens": 48931071.0,
      "step": 25010
    },
    {
      "entropy": 6.007986116409302,
      "epoch": 2.50052481631429,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004378070186074302,
      "loss": 5.6412,
      "mean_token_accuracy": 0.18435328155755998,
      "num_tokens": 48941957.0,
      "step": 25015
    },
    {
      "entropy": 6.055571174621582,
      "epoch": 2.5010246413755186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043778238274767395,
      "loss": 5.6042,
      "mean_token_accuracy": 0.18137228041887282,
      "num_tokens": 48951466.0,
      "step": 25020
    },
    {
      "entropy": 6.1255968570709225,
      "epoch": 2.501524466436747,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004377577427923549,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17655265480279922,
      "num_tokens": 48961967.0,
      "step": 25025
    },
    {
      "entropy": 6.066107320785522,
      "epoch": 2.5020242914979756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043773309874209294,
      "loss": 5.6954,
      "mean_token_accuracy": 0.18041328638792037,
      "num_tokens": 48973391.0,
      "step": 25030
    },
    {
      "entropy": 5.985300445556641,
      "epoch": 2.5025241165592043,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043770845059750807,
      "loss": 5.6263,
      "mean_token_accuracy": 0.1783823251724243,
      "num_tokens": 48983047.0,
      "step": 25035
    },
    {
      "entropy": 6.04102053642273,
      "epoch": 2.503023941620433,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043768379835922044,
      "loss": 5.6986,
      "mean_token_accuracy": 0.16725168228149415,
      "num_tokens": 48992140.0,
      "step": 25040
    },
    {
      "entropy": 5.999973583221435,
      "epoch": 2.5035237666816617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004376591420278503,
      "loss": 5.5099,
      "mean_token_accuracy": 0.19277865886688234,
      "num_tokens": 49002323.0,
      "step": 25045
    },
    {
      "entropy": 6.0490964412689205,
      "epoch": 2.50402359174289,
      "grad_norm": 1.234375,
      "learning_rate": 0.000437634481604018,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17914209216833116,
      "num_tokens": 49011727.0,
      "step": 25050
    },
    {
      "entropy": 6.025519514083863,
      "epoch": 2.5045234168041186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000437609817088344,
      "loss": 5.5673,
      "mean_token_accuracy": 0.18025189638137817,
      "num_tokens": 49020723.0,
      "step": 25055
    },
    {
      "entropy": 6.015647554397583,
      "epoch": 2.505023241865347,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043758514848144886,
      "loss": 5.7047,
      "mean_token_accuracy": 0.17517778873443604,
      "num_tokens": 49030270.0,
      "step": 25060
    },
    {
      "entropy": 6.004714441299439,
      "epoch": 2.5055230669265756,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004375604757839532,
      "loss": 5.5559,
      "mean_token_accuracy": 0.18690720200538635,
      "num_tokens": 49039726.0,
      "step": 25065
    },
    {
      "entropy": 6.082171440124512,
      "epoch": 2.5060228919878043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043753579899647783,
      "loss": 5.6384,
      "mean_token_accuracy": 0.1713825136423111,
      "num_tokens": 49050191.0,
      "step": 25070
    },
    {
      "entropy": 6.123177003860474,
      "epoch": 2.506522717049033,
      "grad_norm": 1.25,
      "learning_rate": 0.00043751111811964353,
      "loss": 5.7329,
      "mean_token_accuracy": 0.16965943872928618,
      "num_tokens": 49060748.0,
      "step": 25075
    },
    {
      "entropy": 5.997367858886719,
      "epoch": 2.5070225421102617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004374864331540713,
      "loss": 5.6684,
      "mean_token_accuracy": 0.18173669874668122,
      "num_tokens": 49070582.0,
      "step": 25080
    },
    {
      "entropy": 5.975010967254638,
      "epoch": 2.50752236717149,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004374617441003823,
      "loss": 5.6062,
      "mean_token_accuracy": 0.1851858988404274,
      "num_tokens": 49080729.0,
      "step": 25085
    },
    {
      "entropy": 6.098676681518555,
      "epoch": 2.5080221922327186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004374370509591976,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17519107460975647,
      "num_tokens": 49090814.0,
      "step": 25090
    },
    {
      "entropy": 6.063648986816406,
      "epoch": 2.508522017293947,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043741235373113854,
      "loss": 5.6454,
      "mean_token_accuracy": 0.17712452858686448,
      "num_tokens": 49100236.0,
      "step": 25095
    },
    {
      "entropy": 6.047211217880249,
      "epoch": 2.5090218423551756,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043738765241682637,
      "loss": 5.6173,
      "mean_token_accuracy": 0.17933498769998552,
      "num_tokens": 49111403.0,
      "step": 25100
    },
    {
      "entropy": 6.056380558013916,
      "epoch": 2.5095216674164043,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004373629470168826,
      "loss": 5.6748,
      "mean_token_accuracy": 0.1759803220629692,
      "num_tokens": 49121678.0,
      "step": 25105
    },
    {
      "entropy": 5.993218517303466,
      "epoch": 2.510021492477633,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000437338237531929,
      "loss": 5.5272,
      "mean_token_accuracy": 0.1916448399424553,
      "num_tokens": 49131171.0,
      "step": 25110
    },
    {
      "entropy": 5.983142280578614,
      "epoch": 2.5105213175388617,
      "grad_norm": 1.25,
      "learning_rate": 0.0004373135239625871,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18537678569555283,
      "num_tokens": 49140300.0,
      "step": 25115
    },
    {
      "entropy": 5.93507719039917,
      "epoch": 2.51102114260009,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043728880630947864,
      "loss": 5.5935,
      "mean_token_accuracy": 0.1801465168595314,
      "num_tokens": 49150028.0,
      "step": 25120
    },
    {
      "entropy": 6.085080623626709,
      "epoch": 2.5115209676613186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043726408457322557,
      "loss": 5.6848,
      "mean_token_accuracy": 0.17511211782693864,
      "num_tokens": 49160971.0,
      "step": 25125
    },
    {
      "entropy": 6.098710346221924,
      "epoch": 2.512020792722547,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043723935875445,
      "loss": 5.6957,
      "mean_token_accuracy": 0.1706320747733116,
      "num_tokens": 49170834.0,
      "step": 25130
    },
    {
      "entropy": 5.985915994644165,
      "epoch": 2.5125206177837756,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004372146288537738,
      "loss": 5.5399,
      "mean_token_accuracy": 0.18494061827659608,
      "num_tokens": 49180459.0,
      "step": 25135
    },
    {
      "entropy": 6.018298625946045,
      "epoch": 2.5130204428450043,
      "grad_norm": 1.765625,
      "learning_rate": 0.00043718989487181927,
      "loss": 5.581,
      "mean_token_accuracy": 0.17955516576766967,
      "num_tokens": 49190767.0,
      "step": 25140
    },
    {
      "entropy": 6.097707366943359,
      "epoch": 2.513520267906233,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004371651568092088,
      "loss": 5.6474,
      "mean_token_accuracy": 0.1713810682296753,
      "num_tokens": 49200368.0,
      "step": 25145
    },
    {
      "entropy": 6.059710741043091,
      "epoch": 2.5140200929674617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004371404146665647,
      "loss": 5.6414,
      "mean_token_accuracy": 0.17824871689081193,
      "num_tokens": 49210731.0,
      "step": 25150
    },
    {
      "entropy": 6.002552127838134,
      "epoch": 2.51451991802869,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043711566844450934,
      "loss": 5.6105,
      "mean_token_accuracy": 0.17849625647068024,
      "num_tokens": 49222783.0,
      "step": 25155
    },
    {
      "entropy": 5.987192630767822,
      "epoch": 2.5150197430899186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043709091814366565,
      "loss": 5.6082,
      "mean_token_accuracy": 0.18067038357257842,
      "num_tokens": 49232738.0,
      "step": 25160
    },
    {
      "entropy": 6.101254510879516,
      "epoch": 2.515519568151147,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043706616376465614,
      "loss": 5.7113,
      "mean_token_accuracy": 0.16903769075870514,
      "num_tokens": 49242666.0,
      "step": 25165
    },
    {
      "entropy": 5.992093944549561,
      "epoch": 2.5160193932123756,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004370414053081036,
      "loss": 5.6046,
      "mean_token_accuracy": 0.17978104948997498,
      "num_tokens": 49253227.0,
      "step": 25170
    },
    {
      "entropy": 5.9924116134643555,
      "epoch": 2.5165192182736043,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000437016642774631,
      "loss": 5.548,
      "mean_token_accuracy": 0.18907111883163452,
      "num_tokens": 49262635.0,
      "step": 25175
    },
    {
      "entropy": 6.01654748916626,
      "epoch": 2.517019043334833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043699187616486144,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17894240617752075,
      "num_tokens": 49271952.0,
      "step": 25180
    },
    {
      "entropy": 6.070410108566284,
      "epoch": 2.5175188683960616,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004369671054794178,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16995348185300826,
      "num_tokens": 49281282.0,
      "step": 25185
    },
    {
      "entropy": 5.980491590499878,
      "epoch": 2.51801869345729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043694233071892366,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1800130158662796,
      "num_tokens": 49290892.0,
      "step": 25190
    },
    {
      "entropy": 6.124154281616211,
      "epoch": 2.5185185185185186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043691755188400203,
      "loss": 5.6835,
      "mean_token_accuracy": 0.176180337369442,
      "num_tokens": 49300816.0,
      "step": 25195
    },
    {
      "entropy": 6.09184079170227,
      "epoch": 2.519018343579747,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004368927689752765,
      "loss": 5.7181,
      "mean_token_accuracy": 0.17144746482372283,
      "num_tokens": 49309849.0,
      "step": 25200
    },
    {
      "entropy": 6.043341875076294,
      "epoch": 2.5195181686409756,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043686798199337053,
      "loss": 5.5608,
      "mean_token_accuracy": 0.17642135471105574,
      "num_tokens": 49319785.0,
      "step": 25205
    },
    {
      "entropy": 6.040855503082275,
      "epoch": 2.5200179937022043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043684319093890777,
      "loss": 5.6181,
      "mean_token_accuracy": 0.17705944925546646,
      "num_tokens": 49329117.0,
      "step": 25210
    },
    {
      "entropy": 6.088783216476441,
      "epoch": 2.520517818763433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043681839581251206,
      "loss": 5.6879,
      "mean_token_accuracy": 0.17530667632818223,
      "num_tokens": 49338527.0,
      "step": 25215
    },
    {
      "entropy": 6.113578605651855,
      "epoch": 2.521017643824661,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043679359661480703,
      "loss": 5.8313,
      "mean_token_accuracy": 0.16349812746047973,
      "num_tokens": 49348976.0,
      "step": 25220
    },
    {
      "entropy": 6.008475112915039,
      "epoch": 2.52151746888589,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043676879334641686,
      "loss": 5.6707,
      "mean_token_accuracy": 0.17991546392440796,
      "num_tokens": 49358817.0,
      "step": 25225
    },
    {
      "entropy": 6.002351999282837,
      "epoch": 2.5220172939471186,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004367439860079654,
      "loss": 5.6231,
      "mean_token_accuracy": 0.18047865182161332,
      "num_tokens": 49369463.0,
      "step": 25230
    },
    {
      "entropy": 6.114549446105957,
      "epoch": 2.522517119008347,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043671917460007693,
      "loss": 5.6896,
      "mean_token_accuracy": 0.1778136670589447,
      "num_tokens": 49379198.0,
      "step": 25235
    },
    {
      "entropy": 5.990658950805664,
      "epoch": 2.5230169440695756,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043669435912337557,
      "loss": 5.5757,
      "mean_token_accuracy": 0.18568713515996932,
      "num_tokens": 49388018.0,
      "step": 25240
    },
    {
      "entropy": 6.0181293964385985,
      "epoch": 2.5235167691308042,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004366695395784858,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17623900324106218,
      "num_tokens": 49397082.0,
      "step": 25245
    },
    {
      "entropy": 5.983112382888794,
      "epoch": 2.524016594192033,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000436644715966032,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1773070976138115,
      "num_tokens": 49405763.0,
      "step": 25250
    },
    {
      "entropy": 5.944419717788696,
      "epoch": 2.524516419253261,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004366198882866387,
      "loss": 5.5263,
      "mean_token_accuracy": 0.1846076637506485,
      "num_tokens": 49415296.0,
      "step": 25255
    },
    {
      "entropy": 6.10158257484436,
      "epoch": 2.52501624431449,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043659505654093065,
      "loss": 5.7243,
      "mean_token_accuracy": 0.1735975608229637,
      "num_tokens": 49425066.0,
      "step": 25260
    },
    {
      "entropy": 6.018766403198242,
      "epoch": 2.5255160693757186,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043657022072953253,
      "loss": 5.598,
      "mean_token_accuracy": 0.1824384182691574,
      "num_tokens": 49434056.0,
      "step": 25265
    },
    {
      "entropy": 6.16401891708374,
      "epoch": 2.526015894436947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004365453808530693,
      "loss": 5.7285,
      "mean_token_accuracy": 0.1768331304192543,
      "num_tokens": 49443822.0,
      "step": 25270
    },
    {
      "entropy": 5.992084789276123,
      "epoch": 2.5265157194981755,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043652053691216574,
      "loss": 5.5622,
      "mean_token_accuracy": 0.18281512409448625,
      "num_tokens": 49453455.0,
      "step": 25275
    },
    {
      "entropy": 6.013744831085205,
      "epoch": 2.5270155445594042,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004364956889074471,
      "loss": 5.6029,
      "mean_token_accuracy": 0.18042178303003312,
      "num_tokens": 49462069.0,
      "step": 25280
    },
    {
      "entropy": 6.061666059494018,
      "epoch": 2.527515369620633,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004364708368395385,
      "loss": 5.6627,
      "mean_token_accuracy": 0.17637724429368973,
      "num_tokens": 49471802.0,
      "step": 25285
    },
    {
      "entropy": 5.97733588218689,
      "epoch": 2.528015194681861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043644598070906516,
      "loss": 5.6124,
      "mean_token_accuracy": 0.1835997372865677,
      "num_tokens": 49481181.0,
      "step": 25290
    },
    {
      "entropy": 6.054309129714966,
      "epoch": 2.52851501974309,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004364211205166525,
      "loss": 5.6602,
      "mean_token_accuracy": 0.1737711489200592,
      "num_tokens": 49491769.0,
      "step": 25295
    },
    {
      "entropy": 6.102015352249145,
      "epoch": 2.5290148448043186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043639625626292595,
      "loss": 5.7092,
      "mean_token_accuracy": 0.1742127373814583,
      "num_tokens": 49501711.0,
      "step": 25300
    },
    {
      "entropy": 6.027209377288818,
      "epoch": 2.529514669865547,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004363713879485112,
      "loss": 5.6786,
      "mean_token_accuracy": 0.17557390928268432,
      "num_tokens": 49511248.0,
      "step": 25305
    },
    {
      "entropy": 5.968700504302978,
      "epoch": 2.5300144949267755,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004363465155740338,
      "loss": 5.513,
      "mean_token_accuracy": 0.19184069335460663,
      "num_tokens": 49520349.0,
      "step": 25310
    },
    {
      "entropy": 6.097583055496216,
      "epoch": 2.5305143199880042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004363216391401196,
      "loss": 5.7253,
      "mean_token_accuracy": 0.18251978904008864,
      "num_tokens": 49530233.0,
      "step": 25315
    },
    {
      "entropy": 5.999865484237671,
      "epoch": 2.531014145049233,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362967586473945,
      "loss": 5.5484,
      "mean_token_accuracy": 0.1863280341029167,
      "num_tokens": 49540059.0,
      "step": 25320
    },
    {
      "entropy": 6.079344749450684,
      "epoch": 2.531513970110461,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004362718740964844,
      "loss": 5.6112,
      "mean_token_accuracy": 0.17963734567165374,
      "num_tokens": 49549113.0,
      "step": 25325
    },
    {
      "entropy": 6.1234822273254395,
      "epoch": 2.53201379517169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362469854880155,
      "loss": 5.7137,
      "mean_token_accuracy": 0.17103082686662674,
      "num_tokens": 49558954.0,
      "step": 25330
    },
    {
      "entropy": 5.925192975997925,
      "epoch": 2.5325136202329186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004362220928226139,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18811638206243514,
      "num_tokens": 49568905.0,
      "step": 25335
    },
    {
      "entropy": 6.017057037353515,
      "epoch": 2.533013445294147,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043619719610090595,
      "loss": 5.5836,
      "mean_token_accuracy": 0.18272799104452134,
      "num_tokens": 49579581.0,
      "step": 25340
    },
    {
      "entropy": 6.047063302993775,
      "epoch": 2.5335132703553755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000436172295323518,
      "loss": 5.6404,
      "mean_token_accuracy": 0.17867728620767592,
      "num_tokens": 49589169.0,
      "step": 25345
    },
    {
      "entropy": 6.087251329421997,
      "epoch": 2.5340130954166042,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043614739049107666,
      "loss": 5.7517,
      "mean_token_accuracy": 0.17447396069765092,
      "num_tokens": 49598910.0,
      "step": 25350
    },
    {
      "entropy": 6.072708511352539,
      "epoch": 2.534512920477833,
      "grad_norm": 1.421875,
      "learning_rate": 0.00043612248160420833,
      "loss": 5.671,
      "mean_token_accuracy": 0.17427663654088973,
      "num_tokens": 49608570.0,
      "step": 25355
    },
    {
      "entropy": 6.087412929534912,
      "epoch": 2.535012745539061,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043609756866353985,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1717851936817169,
      "num_tokens": 49618034.0,
      "step": 25360
    },
    {
      "entropy": 6.087347459793091,
      "epoch": 2.53551257060029,
      "grad_norm": 1.3125,
      "learning_rate": 0.000436072651669698,
      "loss": 5.7083,
      "mean_token_accuracy": 0.17378555089235306,
      "num_tokens": 49628297.0,
      "step": 25365
    },
    {
      "entropy": 6.046040058135986,
      "epoch": 2.5360123956615186,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004360477306233097,
      "loss": 5.7189,
      "mean_token_accuracy": 0.17441911101341248,
      "num_tokens": 49638504.0,
      "step": 25370
    },
    {
      "entropy": 6.034769678115845,
      "epoch": 2.536512220722747,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043602280552500195,
      "loss": 5.6712,
      "mean_token_accuracy": 0.17964299619197846,
      "num_tokens": 49648597.0,
      "step": 25375
    },
    {
      "entropy": 6.01648736000061,
      "epoch": 2.5370120457839755,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043599787637540186,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17834883481264113,
      "num_tokens": 49657767.0,
      "step": 25380
    },
    {
      "entropy": 6.0184088230133055,
      "epoch": 2.5375118708452042,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043597294317513654,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18792062848806382,
      "num_tokens": 49667739.0,
      "step": 25385
    },
    {
      "entropy": 6.004859066009521,
      "epoch": 2.538011695906433,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043594800592483345,
      "loss": 5.5349,
      "mean_token_accuracy": 0.18249509036540984,
      "num_tokens": 49676350.0,
      "step": 25390
    },
    {
      "entropy": 6.029582595825195,
      "epoch": 2.538511520967661,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043592306462511993,
      "loss": 5.6686,
      "mean_token_accuracy": 0.1783662259578705,
      "num_tokens": 49684654.0,
      "step": 25395
    },
    {
      "entropy": 6.081376552581787,
      "epoch": 2.53901134602889,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043589811927662356,
      "loss": 5.6604,
      "mean_token_accuracy": 0.1789412096142769,
      "num_tokens": 49695138.0,
      "step": 25400
    },
    {
      "entropy": 6.07553243637085,
      "epoch": 2.5395111710901186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043587316987997183,
      "loss": 5.6126,
      "mean_token_accuracy": 0.17862605154514313,
      "num_tokens": 49704671.0,
      "step": 25405
    },
    {
      "entropy": 5.999970579147339,
      "epoch": 2.540010996151347,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043584821643579265,
      "loss": 5.6774,
      "mean_token_accuracy": 0.1771268367767334,
      "num_tokens": 49714998.0,
      "step": 25410
    },
    {
      "entropy": 6.0109460830688475,
      "epoch": 2.5405108212125755,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043582325894471366,
      "loss": 5.4949,
      "mean_token_accuracy": 0.18528762459754944,
      "num_tokens": 49725333.0,
      "step": 25415
    },
    {
      "entropy": 6.067316007614136,
      "epoch": 2.5410106462738042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004357982974073629,
      "loss": 5.6942,
      "mean_token_accuracy": 0.1738043799996376,
      "num_tokens": 49735919.0,
      "step": 25420
    },
    {
      "entropy": 6.055648469924927,
      "epoch": 2.541510471335033,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004357733318243683,
      "loss": 5.6724,
      "mean_token_accuracy": 0.17633128464221953,
      "num_tokens": 49746631.0,
      "step": 25425
    },
    {
      "entropy": 6.052191686630249,
      "epoch": 2.542010296396261,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043574836219635805,
      "loss": 5.6564,
      "mean_token_accuracy": 0.17704743295907974,
      "num_tokens": 49756241.0,
      "step": 25430
    },
    {
      "entropy": 6.0505446434021,
      "epoch": 2.54251012145749,
      "grad_norm": 1.125,
      "learning_rate": 0.00043572338852396035,
      "loss": 5.6424,
      "mean_token_accuracy": 0.18008130937814712,
      "num_tokens": 49766761.0,
      "step": 25435
    },
    {
      "entropy": 6.007811212539673,
      "epoch": 2.5430099465187186,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043569841080780367,
      "loss": 5.6139,
      "mean_token_accuracy": 0.1808549866080284,
      "num_tokens": 49775442.0,
      "step": 25440
    },
    {
      "entropy": 6.052313899993896,
      "epoch": 2.543509771579947,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00043567342904851625,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17091287076473236,
      "num_tokens": 49784071.0,
      "step": 25445
    },
    {
      "entropy": 6.040236520767212,
      "epoch": 2.5440095966411755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004356484432467267,
      "loss": 5.6804,
      "mean_token_accuracy": 0.17447432726621628,
      "num_tokens": 49793649.0,
      "step": 25450
    },
    {
      "entropy": 6.065722179412842,
      "epoch": 2.5445094217024042,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043562345340306363,
      "loss": 5.7002,
      "mean_token_accuracy": 0.17370177954435348,
      "num_tokens": 49802940.0,
      "step": 25455
    },
    {
      "entropy": 5.974815511703492,
      "epoch": 2.545009246763633,
      "grad_norm": 2.21875,
      "learning_rate": 0.0004355984595181559,
      "loss": 5.5791,
      "mean_token_accuracy": 0.1822923704981804,
      "num_tokens": 49813035.0,
      "step": 25460
    },
    {
      "entropy": 6.086843824386596,
      "epoch": 2.545509071824861,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004355734615926321,
      "loss": 5.7409,
      "mean_token_accuracy": 0.17039729356765748,
      "num_tokens": 49823154.0,
      "step": 25465
    },
    {
      "entropy": 6.005413913726807,
      "epoch": 2.54600889688609,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004355484596271214,
      "loss": 5.5772,
      "mean_token_accuracy": 0.17517422139644623,
      "num_tokens": 49833662.0,
      "step": 25470
    },
    {
      "entropy": 5.931272888183594,
      "epoch": 2.5465087219473186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004355234536222528,
      "loss": 5.5293,
      "mean_token_accuracy": 0.18179947584867479,
      "num_tokens": 49843322.0,
      "step": 25475
    },
    {
      "entropy": 5.990976762771607,
      "epoch": 2.547008547008547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004354984435786554,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18439106941223143,
      "num_tokens": 49853360.0,
      "step": 25480
    },
    {
      "entropy": 5.981731176376343,
      "epoch": 2.5475083720697755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004354734294969585,
      "loss": 5.6454,
      "mean_token_accuracy": 0.17640133053064347,
      "num_tokens": 49864205.0,
      "step": 25485
    },
    {
      "entropy": 6.026479578018188,
      "epoch": 2.5480081971310042,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043544841137779127,
      "loss": 5.6118,
      "mean_token_accuracy": 0.17646278142929078,
      "num_tokens": 49873629.0,
      "step": 25490
    },
    {
      "entropy": 6.012130784988403,
      "epoch": 2.548508022192233,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043542338922178343,
      "loss": 5.5712,
      "mean_token_accuracy": 0.18047667890787125,
      "num_tokens": 49883877.0,
      "step": 25495
    },
    {
      "entropy": 6.076614427566528,
      "epoch": 2.549007847253461,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043539836302956434,
      "loss": 5.7117,
      "mean_token_accuracy": 0.17337778359651565,
      "num_tokens": 49893664.0,
      "step": 25500
    },
    {
      "entropy": 5.98065767288208,
      "epoch": 2.54950767231469,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004353733328017637,
      "loss": 5.5969,
      "mean_token_accuracy": 0.18388498574495316,
      "num_tokens": 49903356.0,
      "step": 25505
    },
    {
      "entropy": 6.000235176086425,
      "epoch": 2.5500074973759186,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004353482985390113,
      "loss": 5.5648,
      "mean_token_accuracy": 0.18062531054019929,
      "num_tokens": 49912579.0,
      "step": 25510
    },
    {
      "entropy": 5.992606353759766,
      "epoch": 2.550507322437147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000435323260241937,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18678076863288878,
      "num_tokens": 49922123.0,
      "step": 25515
    },
    {
      "entropy": 6.1148090839385985,
      "epoch": 2.5510071474983755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043529821791117064,
      "loss": 5.6979,
      "mean_token_accuracy": 0.17485936135053634,
      "num_tokens": 49931340.0,
      "step": 25520
    },
    {
      "entropy": 6.119412040710449,
      "epoch": 2.5515069725596042,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004352731715473425,
      "loss": 5.694,
      "mean_token_accuracy": 0.1771440714597702,
      "num_tokens": 49940449.0,
      "step": 25525
    },
    {
      "entropy": 6.078823518753052,
      "epoch": 2.552006797620833,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043524812115108244,
      "loss": 5.6857,
      "mean_token_accuracy": 0.17441317737102507,
      "num_tokens": 49949914.0,
      "step": 25530
    },
    {
      "entropy": 6.039649486541748,
      "epoch": 2.552506622682061,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043522306672302103,
      "loss": 5.6433,
      "mean_token_accuracy": 0.18024411648511887,
      "num_tokens": 49960877.0,
      "step": 25535
    },
    {
      "entropy": 6.064227676391601,
      "epoch": 2.55300644774329,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004351980082637884,
      "loss": 5.5984,
      "mean_token_accuracy": 0.18599121421575546,
      "num_tokens": 49969272.0,
      "step": 25540
    },
    {
      "entropy": 6.068405055999756,
      "epoch": 2.5535062728045186,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043517294577401504,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17760029137134553,
      "num_tokens": 49978600.0,
      "step": 25545
    },
    {
      "entropy": 5.878106546401978,
      "epoch": 2.554006097865747,
      "grad_norm": 1.375,
      "learning_rate": 0.00043514787925433176,
      "loss": 5.5143,
      "mean_token_accuracy": 0.19106391370296477,
      "num_tokens": 49987772.0,
      "step": 25550
    },
    {
      "entropy": 6.005835771560669,
      "epoch": 2.5545059229269755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004351228087053689,
      "loss": 5.704,
      "mean_token_accuracy": 0.16908048540353776,
      "num_tokens": 49996826.0,
      "step": 25555
    },
    {
      "entropy": 6.096341276168824,
      "epoch": 2.5550057479882042,
      "grad_norm": 1.796875,
      "learning_rate": 0.00043509773412775746,
      "loss": 5.6359,
      "mean_token_accuracy": 0.17590566873550414,
      "num_tokens": 50006619.0,
      "step": 25560
    },
    {
      "entropy": 5.998745584487915,
      "epoch": 2.555505573049433,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004350726555221282,
      "loss": 5.5751,
      "mean_token_accuracy": 0.18535693287849425,
      "num_tokens": 50017050.0,
      "step": 25565
    },
    {
      "entropy": 6.016600847244263,
      "epoch": 2.556005398110661,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004350475728891121,
      "loss": 5.6204,
      "mean_token_accuracy": 0.18247933834791183,
      "num_tokens": 50026558.0,
      "step": 25570
    },
    {
      "entropy": 6.01793942451477,
      "epoch": 2.55650522317189,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004350224862293403,
      "loss": 5.6089,
      "mean_token_accuracy": 0.17761911749839782,
      "num_tokens": 50035540.0,
      "step": 25575
    },
    {
      "entropy": 6.021709823608399,
      "epoch": 2.557005048233118,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043499739554344393,
      "loss": 5.6403,
      "mean_token_accuracy": 0.17829322963953018,
      "num_tokens": 50045780.0,
      "step": 25580
    },
    {
      "entropy": 6.026425552368164,
      "epoch": 2.557504873294347,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00043497230083205425,
      "loss": 5.5242,
      "mean_token_accuracy": 0.183003930747509,
      "num_tokens": 50055977.0,
      "step": 25585
    },
    {
      "entropy": 6.033564472198487,
      "epoch": 2.5580046983555755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004349472020958026,
      "loss": 5.652,
      "mean_token_accuracy": 0.1751616269350052,
      "num_tokens": 50065974.0,
      "step": 25590
    },
    {
      "entropy": 5.9302510738372805,
      "epoch": 2.5585045234168042,
      "grad_norm": 1.125,
      "learning_rate": 0.00043492209933532054,
      "loss": 5.5514,
      "mean_token_accuracy": 0.18659948855638503,
      "num_tokens": 50075969.0,
      "step": 25595
    },
    {
      "entropy": 6.045860290527344,
      "epoch": 2.559004348478033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043489699255123954,
      "loss": 5.6162,
      "mean_token_accuracy": 0.1855676382780075,
      "num_tokens": 50085304.0,
      "step": 25600
    },
    {
      "entropy": 6.079273700714111,
      "epoch": 2.559504173539261,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004348718817441914,
      "loss": 5.6969,
      "mean_token_accuracy": 0.1788138911128044,
      "num_tokens": 50094903.0,
      "step": 25605
    },
    {
      "entropy": 6.077898263931274,
      "epoch": 2.56000399860049,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004348467669148078,
      "loss": 5.6035,
      "mean_token_accuracy": 0.17880319207906722,
      "num_tokens": 50106990.0,
      "step": 25610
    },
    {
      "entropy": 6.164908981323242,
      "epoch": 2.560503823661718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043482164806372075,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1692583054304123,
      "num_tokens": 50117655.0,
      "step": 25615
    },
    {
      "entropy": 6.12608232498169,
      "epoch": 2.561003648722947,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004347965251915622,
      "loss": 5.6843,
      "mean_token_accuracy": 0.1754228562116623,
      "num_tokens": 50127024.0,
      "step": 25620
    },
    {
      "entropy": 5.991326093673706,
      "epoch": 2.5615034737841755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043477139829896415,
      "loss": 5.6316,
      "mean_token_accuracy": 0.18353494256734848,
      "num_tokens": 50136524.0,
      "step": 25625
    },
    {
      "entropy": 6.0724719047546385,
      "epoch": 2.5620032988454042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043474626738655876,
      "loss": 5.6931,
      "mean_token_accuracy": 0.17504622638225556,
      "num_tokens": 50146242.0,
      "step": 25630
    },
    {
      "entropy": 5.991260719299317,
      "epoch": 2.562503123906633,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004347211324549785,
      "loss": 5.5974,
      "mean_token_accuracy": 0.1820445254445076,
      "num_tokens": 50156486.0,
      "step": 25635
    },
    {
      "entropy": 6.008723258972168,
      "epoch": 2.563002948967861,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043469599350485556,
      "loss": 5.6484,
      "mean_token_accuracy": 0.17606824934482573,
      "num_tokens": 50166479.0,
      "step": 25640
    },
    {
      "entropy": 6.042048501968384,
      "epoch": 2.56350277402909,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004346708505368225,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17520945966243745,
      "num_tokens": 50176207.0,
      "step": 25645
    },
    {
      "entropy": 6.0437140464782715,
      "epoch": 2.564002599090318,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004346457035515121,
      "loss": 5.6374,
      "mean_token_accuracy": 0.1785886377096176,
      "num_tokens": 50185577.0,
      "step": 25650
    },
    {
      "entropy": 6.121019983291626,
      "epoch": 2.564502424151547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043462055254955666,
      "loss": 5.6415,
      "mean_token_accuracy": 0.17691698670387268,
      "num_tokens": 50195473.0,
      "step": 25655
    },
    {
      "entropy": 6.0379187107086185,
      "epoch": 2.5650022492127755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004345953975315893,
      "loss": 5.5739,
      "mean_token_accuracy": 0.1824019655585289,
      "num_tokens": 50205684.0,
      "step": 25660
    },
    {
      "entropy": 6.075171995162964,
      "epoch": 2.5655020742740042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004345702384982429,
      "loss": 5.7325,
      "mean_token_accuracy": 0.1727476716041565,
      "num_tokens": 50215716.0,
      "step": 25665
    },
    {
      "entropy": 6.162225770950317,
      "epoch": 2.566001899335233,
      "grad_norm": 1.25,
      "learning_rate": 0.0004345450754501502,
      "loss": 5.7657,
      "mean_token_accuracy": 0.16791683584451675,
      "num_tokens": 50226240.0,
      "step": 25670
    },
    {
      "entropy": 6.090461349487304,
      "epoch": 2.566501724396461,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043451990838794453,
      "loss": 5.7082,
      "mean_token_accuracy": 0.1705689549446106,
      "num_tokens": 50234886.0,
      "step": 25675
    },
    {
      "entropy": 6.088668155670166,
      "epoch": 2.56700154945769,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000434494737312259,
      "loss": 5.6898,
      "mean_token_accuracy": 0.1762990117073059,
      "num_tokens": 50244594.0,
      "step": 25680
    },
    {
      "entropy": 6.026648950576782,
      "epoch": 2.567501374518918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043446956222372693,
      "loss": 5.6351,
      "mean_token_accuracy": 0.18366547375917436,
      "num_tokens": 50254423.0,
      "step": 25685
    },
    {
      "entropy": 6.024155807495117,
      "epoch": 2.568001199580147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004344443831229816,
      "loss": 5.6599,
      "mean_token_accuracy": 0.17633561789989471,
      "num_tokens": 50263280.0,
      "step": 25690
    },
    {
      "entropy": 6.049336576461792,
      "epoch": 2.5685010246413755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043441920001065676,
      "loss": 5.6344,
      "mean_token_accuracy": 0.176080159842968,
      "num_tokens": 50272802.0,
      "step": 25695
    },
    {
      "entropy": 6.050707530975342,
      "epoch": 2.5690008497026042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004343940128873858,
      "loss": 5.6875,
      "mean_token_accuracy": 0.17789363414049147,
      "num_tokens": 50282066.0,
      "step": 25700
    },
    {
      "entropy": 6.054574584960937,
      "epoch": 2.569500674763833,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043436882175380256,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18388429135084153,
      "num_tokens": 50292333.0,
      "step": 25705
    },
    {
      "entropy": 5.903323936462402,
      "epoch": 2.570000499825061,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004343436266105407,
      "loss": 5.4415,
      "mean_token_accuracy": 0.19323317110538482,
      "num_tokens": 50302001.0,
      "step": 25710
    },
    {
      "entropy": 6.118019866943359,
      "epoch": 2.57050032488629,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004343184274582343,
      "loss": 5.7582,
      "mean_token_accuracy": 0.17008230090141296,
      "num_tokens": 50312447.0,
      "step": 25715
    },
    {
      "entropy": 6.025377893447876,
      "epoch": 2.571000149947518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004342932242975172,
      "loss": 5.5949,
      "mean_token_accuracy": 0.17891275733709336,
      "num_tokens": 50321534.0,
      "step": 25720
    },
    {
      "entropy": 6.0414940357208256,
      "epoch": 2.571499975008747,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004342680171290235,
      "loss": 5.6135,
      "mean_token_accuracy": 0.17746197581291198,
      "num_tokens": 50331354.0,
      "step": 25725
    },
    {
      "entropy": 6.189075374603272,
      "epoch": 2.5719998000699755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004342428059533875,
      "loss": 5.7927,
      "mean_token_accuracy": 0.16364883929491042,
      "num_tokens": 50341463.0,
      "step": 25730
    },
    {
      "entropy": 6.103353214263916,
      "epoch": 2.5724996251312042,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004342175907712435,
      "loss": 5.5964,
      "mean_token_accuracy": 0.1764276385307312,
      "num_tokens": 50350374.0,
      "step": 25735
    },
    {
      "entropy": 6.002917146682739,
      "epoch": 2.572999450192433,
      "grad_norm": 1.125,
      "learning_rate": 0.00043419237158322596,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17603530436754228,
      "num_tokens": 50359749.0,
      "step": 25740
    },
    {
      "entropy": 6.063030672073364,
      "epoch": 2.573499275253661,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043416714838996915,
      "loss": 5.6935,
      "mean_token_accuracy": 0.1795203521847725,
      "num_tokens": 50369910.0,
      "step": 25745
    },
    {
      "entropy": 6.142698144912719,
      "epoch": 2.57399910031489,
      "grad_norm": 1.203125,
      "learning_rate": 0.000434141921192108,
      "loss": 5.7138,
      "mean_token_accuracy": 0.16885101050138474,
      "num_tokens": 50380840.0,
      "step": 25750
    },
    {
      "entropy": 6.046850347518921,
      "epoch": 2.574498925376118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000434116689990277,
      "loss": 5.6498,
      "mean_token_accuracy": 0.18276134729385377,
      "num_tokens": 50389310.0,
      "step": 25755
    },
    {
      "entropy": 6.0242094039917,
      "epoch": 2.574998750437347,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043409145478511105,
      "loss": 5.7226,
      "mean_token_accuracy": 0.17912080138921738,
      "num_tokens": 50398412.0,
      "step": 25760
    },
    {
      "entropy": 6.0102311134338375,
      "epoch": 2.5754985754985755,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043406621557724495,
      "loss": 5.5949,
      "mean_token_accuracy": 0.17683623284101485,
      "num_tokens": 50408525.0,
      "step": 25765
    },
    {
      "entropy": 6.1336463451385494,
      "epoch": 2.5759984005598042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043404097236731396,
      "loss": 5.7925,
      "mean_token_accuracy": 0.16813481152057647,
      "num_tokens": 50419578.0,
      "step": 25770
    },
    {
      "entropy": 6.079278087615966,
      "epoch": 2.5764982256210325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004340157251559529,
      "loss": 5.6564,
      "mean_token_accuracy": 0.17974288761615753,
      "num_tokens": 50428925.0,
      "step": 25775
    },
    {
      "entropy": 5.989203548431396,
      "epoch": 2.576998050682261,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043399047394379717,
      "loss": 5.5877,
      "mean_token_accuracy": 0.1844230592250824,
      "num_tokens": 50438769.0,
      "step": 25780
    },
    {
      "entropy": 6.0562351703643795,
      "epoch": 2.57749787574349,
      "grad_norm": 1.125,
      "learning_rate": 0.00043396521873148195,
      "loss": 5.7324,
      "mean_token_accuracy": 0.17407798916101455,
      "num_tokens": 50449210.0,
      "step": 25785
    },
    {
      "entropy": 6.050251579284668,
      "epoch": 2.577997700804718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043393995951964277,
      "loss": 5.6493,
      "mean_token_accuracy": 0.17404088824987413,
      "num_tokens": 50458703.0,
      "step": 25790
    },
    {
      "entropy": 5.984742450714111,
      "epoch": 2.578497525865947,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043391469630891513,
      "loss": 5.6279,
      "mean_token_accuracy": 0.18197864592075347,
      "num_tokens": 50468210.0,
      "step": 25795
    },
    {
      "entropy": 6.03759651184082,
      "epoch": 2.5789973509271755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004338894290999346,
      "loss": 5.6176,
      "mean_token_accuracy": 0.1853585809469223,
      "num_tokens": 50477726.0,
      "step": 25800
    },
    {
      "entropy": 6.089315938949585,
      "epoch": 2.5794971759884042,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043386415789333687,
      "loss": 5.6139,
      "mean_token_accuracy": 0.18043663054704667,
      "num_tokens": 50488637.0,
      "step": 25805
    },
    {
      "entropy": 6.04966721534729,
      "epoch": 2.5799970010496325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004338388826897579,
      "loss": 5.57,
      "mean_token_accuracy": 0.18727486878633498,
      "num_tokens": 50498544.0,
      "step": 25810
    },
    {
      "entropy": 6.025316190719605,
      "epoch": 2.580496826110861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004338136034898335,
      "loss": 5.7525,
      "mean_token_accuracy": 0.16407786607742308,
      "num_tokens": 50508482.0,
      "step": 25815
    },
    {
      "entropy": 6.017668104171753,
      "epoch": 2.58099665117209,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004337883202941997,
      "loss": 5.6191,
      "mean_token_accuracy": 0.17860807925462724,
      "num_tokens": 50518294.0,
      "step": 25820
    },
    {
      "entropy": 6.028322649002075,
      "epoch": 2.581496476233318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004337630331034926,
      "loss": 5.6339,
      "mean_token_accuracy": 0.17694128155708314,
      "num_tokens": 50528403.0,
      "step": 25825
    },
    {
      "entropy": 6.075345277786255,
      "epoch": 2.581996301294547,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043373774191834844,
      "loss": 5.6202,
      "mean_token_accuracy": 0.17656493335962295,
      "num_tokens": 50538103.0,
      "step": 25830
    },
    {
      "entropy": 6.003156900405884,
      "epoch": 2.5824961263557755,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043371244673940355,
      "loss": 5.5723,
      "mean_token_accuracy": 0.1827936589717865,
      "num_tokens": 50548020.0,
      "step": 25835
    },
    {
      "entropy": 6.063708162307739,
      "epoch": 2.582995951417004,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004336871475672943,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17520661652088165,
      "num_tokens": 50557843.0,
      "step": 25840
    },
    {
      "entropy": 6.106504583358765,
      "epoch": 2.5834957764782325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004336618444026573,
      "loss": 5.7026,
      "mean_token_accuracy": 0.16996080577373504,
      "num_tokens": 50567377.0,
      "step": 25845
    },
    {
      "entropy": 6.038467645645142,
      "epoch": 2.583995601539461,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043363653724612903,
      "loss": 5.5833,
      "mean_token_accuracy": 0.18137314468622207,
      "num_tokens": 50577184.0,
      "step": 25850
    },
    {
      "entropy": 6.02907772064209,
      "epoch": 2.58449542660069,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043361122609834646,
      "loss": 5.6358,
      "mean_token_accuracy": 0.177972449362278,
      "num_tokens": 50585859.0,
      "step": 25855
    },
    {
      "entropy": 6.047159099578858,
      "epoch": 2.584995251661918,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004335859109599462,
      "loss": 5.6273,
      "mean_token_accuracy": 0.18237798064947128,
      "num_tokens": 50595627.0,
      "step": 25860
    },
    {
      "entropy": 6.000776672363282,
      "epoch": 2.585495076723147,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043356059183156514,
      "loss": 5.5974,
      "mean_token_accuracy": 0.18218372762203217,
      "num_tokens": 50605733.0,
      "step": 25865
    },
    {
      "entropy": 6.069572496414184,
      "epoch": 2.5859949017843755,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00043353526871384054,
      "loss": 5.703,
      "mean_token_accuracy": 0.17337165176868438,
      "num_tokens": 50615341.0,
      "step": 25870
    },
    {
      "entropy": 6.057975959777832,
      "epoch": 2.586494726845604,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004335099416074092,
      "loss": 5.6295,
      "mean_token_accuracy": 0.17950332164764404,
      "num_tokens": 50625545.0,
      "step": 25875
    },
    {
      "entropy": 6.0462573051452635,
      "epoch": 2.5869945519068325,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004334846105129086,
      "loss": 5.7039,
      "mean_token_accuracy": 0.17092407494783401,
      "num_tokens": 50635518.0,
      "step": 25880
    },
    {
      "entropy": 5.990850305557251,
      "epoch": 2.587494376968061,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000433459275430976,
      "loss": 5.526,
      "mean_token_accuracy": 0.18877878785133362,
      "num_tokens": 50644353.0,
      "step": 25885
    },
    {
      "entropy": 6.094750452041626,
      "epoch": 2.58799420202929,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004334339363622488,
      "loss": 5.7315,
      "mean_token_accuracy": 0.17348005324602128,
      "num_tokens": 50653911.0,
      "step": 25890
    },
    {
      "entropy": 6.078180885314941,
      "epoch": 2.588494027090518,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043340859330736454,
      "loss": 5.653,
      "mean_token_accuracy": 0.17294460088014602,
      "num_tokens": 50664476.0,
      "step": 25895
    },
    {
      "entropy": 6.079866504669189,
      "epoch": 2.588993852151747,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00043338324626696084,
      "loss": 5.6023,
      "mean_token_accuracy": 0.17800079584121703,
      "num_tokens": 50673779.0,
      "step": 25900
    },
    {
      "entropy": 6.057315111160278,
      "epoch": 2.5894936772129755,
      "grad_norm": 3.046875,
      "learning_rate": 0.0004333578952416753,
      "loss": 5.6101,
      "mean_token_accuracy": 0.1768708661198616,
      "num_tokens": 50682572.0,
      "step": 25905
    },
    {
      "entropy": 5.959677839279175,
      "epoch": 2.589993502274204,
      "grad_norm": 1.203125,
      "learning_rate": 0.000433332540232146,
      "loss": 5.5402,
      "mean_token_accuracy": 0.18218840807676315,
      "num_tokens": 50691553.0,
      "step": 25910
    },
    {
      "entropy": 6.071141529083252,
      "epoch": 2.5904933273354325,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043330718123901074,
      "loss": 5.5736,
      "mean_token_accuracy": 0.1890376552939415,
      "num_tokens": 50701348.0,
      "step": 25915
    },
    {
      "entropy": 5.983849573135376,
      "epoch": 2.590993152396661,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043328181826290745,
      "loss": 5.5795,
      "mean_token_accuracy": 0.1800802543759346,
      "num_tokens": 50710586.0,
      "step": 25920
    },
    {
      "entropy": 5.96975564956665,
      "epoch": 2.59149297745789,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004332564513044744,
      "loss": 5.6265,
      "mean_token_accuracy": 0.18213405460119247,
      "num_tokens": 50720224.0,
      "step": 25925
    },
    {
      "entropy": 6.041332769393921,
      "epoch": 2.591992802519118,
      "grad_norm": 1.4375,
      "learning_rate": 0.00043323108036434974,
      "loss": 5.5855,
      "mean_token_accuracy": 0.1850375860929489,
      "num_tokens": 50730295.0,
      "step": 25930
    },
    {
      "entropy": 6.102500057220459,
      "epoch": 2.592492627580347,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004332057054431718,
      "loss": 5.6578,
      "mean_token_accuracy": 0.17403520047664642,
      "num_tokens": 50739516.0,
      "step": 25935
    },
    {
      "entropy": 6.028181362152099,
      "epoch": 2.5929924526415755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000433180326541579,
      "loss": 5.5723,
      "mean_token_accuracy": 0.18627001494169235,
      "num_tokens": 50748506.0,
      "step": 25940
    },
    {
      "entropy": 6.045462322235108,
      "epoch": 2.593492277702804,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004331549436602099,
      "loss": 5.672,
      "mean_token_accuracy": 0.1731103017926216,
      "num_tokens": 50758804.0,
      "step": 25945
    },
    {
      "entropy": 5.926534938812256,
      "epoch": 2.5939921027640325,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043312955679970317,
      "loss": 5.5228,
      "mean_token_accuracy": 0.19063213765621184,
      "num_tokens": 50768898.0,
      "step": 25950
    },
    {
      "entropy": 6.04374213218689,
      "epoch": 2.594491927825261,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043310416596069746,
      "loss": 5.6735,
      "mean_token_accuracy": 0.1808082014322281,
      "num_tokens": 50778625.0,
      "step": 25955
    },
    {
      "entropy": 5.9920543193817135,
      "epoch": 2.59499175288649,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043307877114383167,
      "loss": 5.5488,
      "mean_token_accuracy": 0.17906266152858735,
      "num_tokens": 50787162.0,
      "step": 25960
    },
    {
      "entropy": 5.945101165771485,
      "epoch": 2.595491577947718,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004330533723497445,
      "loss": 5.5748,
      "mean_token_accuracy": 0.18555621504783631,
      "num_tokens": 50796615.0,
      "step": 25965
    },
    {
      "entropy": 6.0943420886993405,
      "epoch": 2.595991403008947,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043302796957907523,
      "loss": 5.7079,
      "mean_token_accuracy": 0.17230619192123414,
      "num_tokens": 50807381.0,
      "step": 25970
    },
    {
      "entropy": 6.140337896347046,
      "epoch": 2.5964912280701755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043300256283246297,
      "loss": 5.659,
      "mean_token_accuracy": 0.17262806743383408,
      "num_tokens": 50818348.0,
      "step": 25975
    },
    {
      "entropy": 6.056605529785156,
      "epoch": 2.596991053131404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043297715211054684,
      "loss": 5.6699,
      "mean_token_accuracy": 0.17609446495771408,
      "num_tokens": 50828688.0,
      "step": 25980
    },
    {
      "entropy": 6.035661792755127,
      "epoch": 2.5974908781926325,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00043295173741396626,
      "loss": 5.6481,
      "mean_token_accuracy": 0.17754530906677246,
      "num_tokens": 50837831.0,
      "step": 25985
    },
    {
      "entropy": 6.066333532333374,
      "epoch": 2.597990703253861,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004329263187433605,
      "loss": 5.6126,
      "mean_token_accuracy": 0.17702539414167404,
      "num_tokens": 50846998.0,
      "step": 25990
    },
    {
      "entropy": 5.983028936386108,
      "epoch": 2.59849052831509,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004329008960993694,
      "loss": 5.5186,
      "mean_token_accuracy": 0.18867831975221633,
      "num_tokens": 50856310.0,
      "step": 25995
    },
    {
      "entropy": 5.885134840011597,
      "epoch": 2.598990353376318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004328754694826322,
      "loss": 5.5685,
      "mean_token_accuracy": 0.19061746895313264,
      "num_tokens": 50867376.0,
      "step": 26000
    },
    {
      "entropy": 6.0081392288208,
      "epoch": 2.599490178437547,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004328500388937889,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1862940028309822,
      "num_tokens": 50877007.0,
      "step": 26005
    },
    {
      "entropy": 6.0840977191925045,
      "epoch": 2.5999900034987755,
      "grad_norm": 2.21875,
      "learning_rate": 0.00043282460433347926,
      "loss": 5.5991,
      "mean_token_accuracy": 0.18165654838085174,
      "num_tokens": 50886221.0,
      "step": 26010
    },
    {
      "entropy": 6.003738880157471,
      "epoch": 2.600489828560004,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004327991658023431,
      "loss": 5.66,
      "mean_token_accuracy": 0.17306568920612336,
      "num_tokens": 50895534.0,
      "step": 26015
    },
    {
      "entropy": 5.983510160446167,
      "epoch": 2.6009896536212325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004327737233010205,
      "loss": 5.5893,
      "mean_token_accuracy": 0.18208255022764205,
      "num_tokens": 50905987.0,
      "step": 26020
    },
    {
      "entropy": 6.060332822799682,
      "epoch": 2.601489478682461,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043274827683015174,
      "loss": 5.7021,
      "mean_token_accuracy": 0.16769730895757676,
      "num_tokens": 50914810.0,
      "step": 26025
    },
    {
      "entropy": 6.11578779220581,
      "epoch": 2.60198930374369,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043272282639037683,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17308832406997682,
      "num_tokens": 50925115.0,
      "step": 26030
    },
    {
      "entropy": 6.158581829071045,
      "epoch": 2.602489128804918,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043269737198233627,
      "loss": 5.7183,
      "mean_token_accuracy": 0.1753817081451416,
      "num_tokens": 50934511.0,
      "step": 26035
    },
    {
      "entropy": 6.043734312057495,
      "epoch": 2.602988953866147,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004326719136066704,
      "loss": 5.6029,
      "mean_token_accuracy": 0.18095627874135972,
      "num_tokens": 50943937.0,
      "step": 26040
    },
    {
      "entropy": 6.10493540763855,
      "epoch": 2.6034887789273755,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043264645126401977,
      "loss": 5.7234,
      "mean_token_accuracy": 0.16442741751670836,
      "num_tokens": 50952952.0,
      "step": 26045
    },
    {
      "entropy": 6.036356496810913,
      "epoch": 2.603988603988604,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043262098495502497,
      "loss": 5.5794,
      "mean_token_accuracy": 0.17932440489530563,
      "num_tokens": 50962569.0,
      "step": 26050
    },
    {
      "entropy": 6.064872884750367,
      "epoch": 2.6044884290498325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004325955146803267,
      "loss": 5.6772,
      "mean_token_accuracy": 0.17790494859218597,
      "num_tokens": 50971520.0,
      "step": 26055
    },
    {
      "entropy": 6.003273344039917,
      "epoch": 2.604988254111061,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004325700404405659,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17576491385698317,
      "num_tokens": 50981209.0,
      "step": 26060
    },
    {
      "entropy": 6.060185861587525,
      "epoch": 2.60548807917229,
      "grad_norm": 1.125,
      "learning_rate": 0.0004325445622363834,
      "loss": 5.6391,
      "mean_token_accuracy": 0.17435826659202575,
      "num_tokens": 50991245.0,
      "step": 26065
    },
    {
      "entropy": 6.083366775512696,
      "epoch": 2.605987904233518,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004325190800684203,
      "loss": 5.6673,
      "mean_token_accuracy": 0.1749819278717041,
      "num_tokens": 51000341.0,
      "step": 26070
    },
    {
      "entropy": 6.175801563262939,
      "epoch": 2.606487729294747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004324935939373176,
      "loss": 5.7791,
      "mean_token_accuracy": 0.16671054661273957,
      "num_tokens": 51010136.0,
      "step": 26075
    },
    {
      "entropy": 6.108215188980102,
      "epoch": 2.6069875543559755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004324681038437166,
      "loss": 5.7148,
      "mean_token_accuracy": 0.17272776067256929,
      "num_tokens": 51020751.0,
      "step": 26080
    },
    {
      "entropy": 6.110399532318115,
      "epoch": 2.607487379417204,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004324426097882587,
      "loss": 5.7645,
      "mean_token_accuracy": 0.1672731399536133,
      "num_tokens": 51031546.0,
      "step": 26085
    },
    {
      "entropy": 6.07423095703125,
      "epoch": 2.6079872044784325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004324171117715852,
      "loss": 5.6109,
      "mean_token_accuracy": 0.18422822654247284,
      "num_tokens": 51041351.0,
      "step": 26090
    },
    {
      "entropy": 6.011583709716797,
      "epoch": 2.608487029539661,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043239160979433765,
      "loss": 5.6026,
      "mean_token_accuracy": 0.17618126422166824,
      "num_tokens": 51050450.0,
      "step": 26095
    },
    {
      "entropy": 5.971647787094116,
      "epoch": 2.6089868546008894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004323661038571578,
      "loss": 5.553,
      "mean_token_accuracy": 0.1863924264907837,
      "num_tokens": 51059895.0,
      "step": 26100
    },
    {
      "entropy": 6.0414632797241214,
      "epoch": 2.609486679662118,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00043234059396068725,
      "loss": 5.6516,
      "mean_token_accuracy": 0.1809893250465393,
      "num_tokens": 51068984.0,
      "step": 26105
    },
    {
      "entropy": 5.987191200256348,
      "epoch": 2.609986504723347,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004323150801055677,
      "loss": 5.6127,
      "mean_token_accuracy": 0.1800945743918419,
      "num_tokens": 51079087.0,
      "step": 26110
    },
    {
      "entropy": 6.06707649230957,
      "epoch": 2.6104863297845755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004322895622924414,
      "loss": 5.6336,
      "mean_token_accuracy": 0.1768326371908188,
      "num_tokens": 51089306.0,
      "step": 26115
    },
    {
      "entropy": 6.047014474868774,
      "epoch": 2.610986154845804,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004322640405219501,
      "loss": 5.6525,
      "mean_token_accuracy": 0.17667663395404815,
      "num_tokens": 51098998.0,
      "step": 26120
    },
    {
      "entropy": 6.013906002044678,
      "epoch": 2.6114859799070325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043223851479473596,
      "loss": 5.5536,
      "mean_token_accuracy": 0.19230637550354004,
      "num_tokens": 51108995.0,
      "step": 26125
    },
    {
      "entropy": 6.004920530319214,
      "epoch": 2.611985804968261,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043221298511144133,
      "loss": 5.497,
      "mean_token_accuracy": 0.18638110309839248,
      "num_tokens": 51118574.0,
      "step": 26130
    },
    {
      "entropy": 5.917248392105103,
      "epoch": 2.6124856300294894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043218745147270845,
      "loss": 5.5371,
      "mean_token_accuracy": 0.1898409530520439,
      "num_tokens": 51128438.0,
      "step": 26135
    },
    {
      "entropy": 6.11306357383728,
      "epoch": 2.612985455090718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043216191387917966,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17574283331632615,
      "num_tokens": 51137223.0,
      "step": 26140
    },
    {
      "entropy": 6.096914958953858,
      "epoch": 2.613485280151947,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004321363723314976,
      "loss": 5.6994,
      "mean_token_accuracy": 0.17306362837553024,
      "num_tokens": 51147299.0,
      "step": 26145
    },
    {
      "entropy": 6.057611036300659,
      "epoch": 2.6139851052131755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043211082683030483,
      "loss": 5.738,
      "mean_token_accuracy": 0.166473788022995,
      "num_tokens": 51157782.0,
      "step": 26150
    },
    {
      "entropy": 6.026036119461059,
      "epoch": 2.614484930274404,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043208527737624406,
      "loss": 5.6139,
      "mean_token_accuracy": 0.17704542577266694,
      "num_tokens": 51168328.0,
      "step": 26155
    },
    {
      "entropy": 6.01550817489624,
      "epoch": 2.6149847553356325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043205972396995825,
      "loss": 5.6577,
      "mean_token_accuracy": 0.17329904288053513,
      "num_tokens": 51177619.0,
      "step": 26160
    },
    {
      "entropy": 6.0162616729736325,
      "epoch": 2.615484580396861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004320341666120901,
      "loss": 5.5899,
      "mean_token_accuracy": 0.1876514345407486,
      "num_tokens": 51186039.0,
      "step": 26165
    },
    {
      "entropy": 6.000626564025879,
      "epoch": 2.6159844054580894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043200860530328287,
      "loss": 5.5163,
      "mean_token_accuracy": 0.1926588699221611,
      "num_tokens": 51195798.0,
      "step": 26170
    },
    {
      "entropy": 6.055613040924072,
      "epoch": 2.616484230519318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043198304004417933,
      "loss": 5.69,
      "mean_token_accuracy": 0.17520691752433776,
      "num_tokens": 51206020.0,
      "step": 26175
    },
    {
      "entropy": 6.066198301315308,
      "epoch": 2.616984055580547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043195747083542296,
      "loss": 5.6235,
      "mean_token_accuracy": 0.17814874053001403,
      "num_tokens": 51216376.0,
      "step": 26180
    },
    {
      "entropy": 5.924957656860352,
      "epoch": 2.6174838806417755,
      "grad_norm": 1.125,
      "learning_rate": 0.000431931897677657,
      "loss": 5.5047,
      "mean_token_accuracy": 0.1789536491036415,
      "num_tokens": 51225542.0,
      "step": 26185
    },
    {
      "entropy": 6.038915491104126,
      "epoch": 2.617983705703004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043190632057152497,
      "loss": 5.703,
      "mean_token_accuracy": 0.16902630776166916,
      "num_tokens": 51235649.0,
      "step": 26190
    },
    {
      "entropy": 6.0624676704406735,
      "epoch": 2.6184835307642325,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004318807395176702,
      "loss": 5.667,
      "mean_token_accuracy": 0.18006136864423752,
      "num_tokens": 51245665.0,
      "step": 26195
    },
    {
      "entropy": 6.046308946609497,
      "epoch": 2.618983355825461,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004318551545167364,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1721554070711136,
      "num_tokens": 51255213.0,
      "step": 26200
    },
    {
      "entropy": 5.979729747772216,
      "epoch": 2.6194831808866894,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004318295655693674,
      "loss": 5.5329,
      "mean_token_accuracy": 0.1902702793478966,
      "num_tokens": 51264742.0,
      "step": 26205
    },
    {
      "entropy": 5.990330410003662,
      "epoch": 2.619983005947918,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043180397267620675,
      "loss": 5.6778,
      "mean_token_accuracy": 0.1805764451622963,
      "num_tokens": 51275135.0,
      "step": 26210
    },
    {
      "entropy": 6.029267740249634,
      "epoch": 2.620482831009147,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004317783758378985,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18675299882888793,
      "num_tokens": 51285285.0,
      "step": 26215
    },
    {
      "entropy": 6.086340093612671,
      "epoch": 2.6209826560703755,
      "grad_norm": 1.25,
      "learning_rate": 0.00043175277505508677,
      "loss": 5.6834,
      "mean_token_accuracy": 0.1784561887383461,
      "num_tokens": 51294880.0,
      "step": 26220
    },
    {
      "entropy": 6.099629592895508,
      "epoch": 2.621482481131604,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004317271703284154,
      "loss": 5.7254,
      "mean_token_accuracy": 0.1724514529109001,
      "num_tokens": 51305585.0,
      "step": 26225
    },
    {
      "entropy": 6.091507244110107,
      "epoch": 2.6219823061928325,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043170156165852887,
      "loss": 5.6783,
      "mean_token_accuracy": 0.17896926105022432,
      "num_tokens": 51315377.0,
      "step": 26230
    },
    {
      "entropy": 6.013772964477539,
      "epoch": 2.622482131254061,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004316759490460713,
      "loss": 5.6243,
      "mean_token_accuracy": 0.18363578766584396,
      "num_tokens": 51325800.0,
      "step": 26235
    },
    {
      "entropy": 5.93272385597229,
      "epoch": 2.6229819563152894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043165033249168723,
      "loss": 5.5096,
      "mean_token_accuracy": 0.18874809741973878,
      "num_tokens": 51335138.0,
      "step": 26240
    },
    {
      "entropy": 6.059574317932129,
      "epoch": 2.623481781376518,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043162471199602103,
      "loss": 5.623,
      "mean_token_accuracy": 0.17749946415424347,
      "num_tokens": 51345871.0,
      "step": 26245
    },
    {
      "entropy": 6.058729887008667,
      "epoch": 2.623981606437747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043159908755971746,
      "loss": 5.6453,
      "mean_token_accuracy": 0.1735410362482071,
      "num_tokens": 51354774.0,
      "step": 26250
    },
    {
      "entropy": 5.942973899841308,
      "epoch": 2.6244814314989755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043157345918342103,
      "loss": 5.6025,
      "mean_token_accuracy": 0.17975897938013077,
      "num_tokens": 51364641.0,
      "step": 26255
    },
    {
      "entropy": 6.042588090896606,
      "epoch": 2.624981256560204,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004315478268677768,
      "loss": 5.6528,
      "mean_token_accuracy": 0.17660840302705766,
      "num_tokens": 51373821.0,
      "step": 26260
    },
    {
      "entropy": 6.129432535171508,
      "epoch": 2.6254810816214325,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004315221906134294,
      "loss": 5.7179,
      "mean_token_accuracy": 0.1735947087407112,
      "num_tokens": 51383926.0,
      "step": 26265
    },
    {
      "entropy": 5.954310178756714,
      "epoch": 2.625980906682661,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043149655042102404,
      "loss": 5.4819,
      "mean_token_accuracy": 0.1899528607726097,
      "num_tokens": 51393254.0,
      "step": 26270
    },
    {
      "entropy": 5.949499320983887,
      "epoch": 2.6264807317438894,
      "grad_norm": 0.9609375,
      "learning_rate": 0.00043147090629120575,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17105428725481034,
      "num_tokens": 51403890.0,
      "step": 26275
    },
    {
      "entropy": 6.0268126964569095,
      "epoch": 2.626980556805118,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004314452582246197,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18495482206344604,
      "num_tokens": 51413309.0,
      "step": 26280
    },
    {
      "entropy": 6.047746276855468,
      "epoch": 2.627480381866347,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043141960622191116,
      "loss": 5.639,
      "mean_token_accuracy": 0.1801455095410347,
      "num_tokens": 51423364.0,
      "step": 26285
    },
    {
      "entropy": 5.9819104194641115,
      "epoch": 2.6279802069275755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043139395028372565,
      "loss": 5.5499,
      "mean_token_accuracy": 0.1873622342944145,
      "num_tokens": 51434105.0,
      "step": 26290
    },
    {
      "entropy": 5.985121488571167,
      "epoch": 2.6284800319888038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043136829041070856,
      "loss": 5.634,
      "mean_token_accuracy": 0.1805615469813347,
      "num_tokens": 51442980.0,
      "step": 26295
    },
    {
      "entropy": 6.004526424407959,
      "epoch": 2.6289798570500325,
      "grad_norm": 1.125,
      "learning_rate": 0.00043134262660350555,
      "loss": 5.6373,
      "mean_token_accuracy": 0.1789153814315796,
      "num_tokens": 51452659.0,
      "step": 26300
    },
    {
      "entropy": 5.995804262161255,
      "epoch": 2.629479682111261,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004313169588627622,
      "loss": 5.5675,
      "mean_token_accuracy": 0.18237171471118926,
      "num_tokens": 51462212.0,
      "step": 26305
    },
    {
      "entropy": 6.021825218200684,
      "epoch": 2.6299795071724894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004312912871891245,
      "loss": 5.6794,
      "mean_token_accuracy": 0.17184750288724898,
      "num_tokens": 51472066.0,
      "step": 26310
    },
    {
      "entropy": 5.985482931137085,
      "epoch": 2.630479332233718,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004312656115832381,
      "loss": 5.6187,
      "mean_token_accuracy": 0.18049290925264358,
      "num_tokens": 51482345.0,
      "step": 26315
    },
    {
      "entropy": 6.133411169052124,
      "epoch": 2.630979157294947,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043123993204574926,
      "loss": 5.6651,
      "mean_token_accuracy": 0.1763326570391655,
      "num_tokens": 51493259.0,
      "step": 26320
    },
    {
      "entropy": 6.0639838695526125,
      "epoch": 2.6314789823561755,
      "grad_norm": 1.125,
      "learning_rate": 0.00043121424857730384,
      "loss": 5.6493,
      "mean_token_accuracy": 0.1740374505519867,
      "num_tokens": 51503105.0,
      "step": 26325
    },
    {
      "entropy": 5.9842857837677,
      "epoch": 2.6319788074174038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004311885611785482,
      "loss": 5.5754,
      "mean_token_accuracy": 0.18283674567937852,
      "num_tokens": 51512650.0,
      "step": 26330
    },
    {
      "entropy": 5.999758386611939,
      "epoch": 2.6324786324786325,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043116286985012847,
      "loss": 5.5616,
      "mean_token_accuracy": 0.1810513198375702,
      "num_tokens": 51521457.0,
      "step": 26335
    },
    {
      "entropy": 6.056594800949097,
      "epoch": 2.632978457539861,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043113717459269116,
      "loss": 5.6566,
      "mean_token_accuracy": 0.1822540670633316,
      "num_tokens": 51530715.0,
      "step": 26340
    },
    {
      "entropy": 6.069809484481811,
      "epoch": 2.6334782826010894,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004311114754068827,
      "loss": 5.6497,
      "mean_token_accuracy": 0.18359818011522294,
      "num_tokens": 51540243.0,
      "step": 26345
    },
    {
      "entropy": 6.040119695663452,
      "epoch": 2.633978107662318,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004310857722933497,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18781823217868804,
      "num_tokens": 51550191.0,
      "step": 26350
    },
    {
      "entropy": 6.0913573741912845,
      "epoch": 2.634477932723547,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043106006525273896,
      "loss": 5.695,
      "mean_token_accuracy": 0.17783313244581223,
      "num_tokens": 51559707.0,
      "step": 26355
    },
    {
      "entropy": 5.9849461078643795,
      "epoch": 2.6349777577847755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004310343542856969,
      "loss": 5.6108,
      "mean_token_accuracy": 0.18427322208881378,
      "num_tokens": 51569146.0,
      "step": 26360
    },
    {
      "entropy": 6.063048028945923,
      "epoch": 2.6354775828460038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043100863939287077,
      "loss": 5.697,
      "mean_token_accuracy": 0.17032020837068557,
      "num_tokens": 51578301.0,
      "step": 26365
    },
    {
      "entropy": 6.121313905715942,
      "epoch": 2.6359774079072325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004309829205749074,
      "loss": 5.7126,
      "mean_token_accuracy": 0.1718689411878586,
      "num_tokens": 51587474.0,
      "step": 26370
    },
    {
      "entropy": 6.043306350708008,
      "epoch": 2.636477232968461,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00043095719783245386,
      "loss": 5.6689,
      "mean_token_accuracy": 0.17691429406404496,
      "num_tokens": 51597166.0,
      "step": 26375
    },
    {
      "entropy": 5.99685845375061,
      "epoch": 2.6369770580296894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043093147116615737,
      "loss": 5.5925,
      "mean_token_accuracy": 0.17952672243118287,
      "num_tokens": 51606448.0,
      "step": 26380
    },
    {
      "entropy": 6.014078521728516,
      "epoch": 2.637476883090918,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004309057405766652,
      "loss": 5.6067,
      "mean_token_accuracy": 0.18896129131317138,
      "num_tokens": 51615901.0,
      "step": 26385
    },
    {
      "entropy": 6.073169994354248,
      "epoch": 2.637976708152147,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043088000606462474,
      "loss": 5.6174,
      "mean_token_accuracy": 0.1820953071117401,
      "num_tokens": 51624798.0,
      "step": 26390
    },
    {
      "entropy": 6.019411659240722,
      "epoch": 2.6384765332133755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043085426763068334,
      "loss": 5.6167,
      "mean_token_accuracy": 0.17732793390750884,
      "num_tokens": 51634359.0,
      "step": 26395
    },
    {
      "entropy": 6.079281044006348,
      "epoch": 2.6389763582746038,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043082852527548876,
      "loss": 5.6685,
      "mean_token_accuracy": 0.1769236668944359,
      "num_tokens": 51644251.0,
      "step": 26400
    },
    {
      "entropy": 6.136126184463501,
      "epoch": 2.6394761833358324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004308027789996886,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17395219206809998,
      "num_tokens": 51653741.0,
      "step": 26405
    },
    {
      "entropy": 6.0278924942016605,
      "epoch": 2.639976008397061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004307770288039306,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17497114092111588,
      "num_tokens": 51663174.0,
      "step": 26410
    },
    {
      "entropy": 6.119552373886108,
      "epoch": 2.6404758334582894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043075127468886264,
      "loss": 5.7592,
      "mean_token_accuracy": 0.16511047929525374,
      "num_tokens": 51673194.0,
      "step": 26415
    },
    {
      "entropy": 6.089165830612183,
      "epoch": 2.640975658519518,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043072551665513265,
      "loss": 5.604,
      "mean_token_accuracy": 0.18054888993501664,
      "num_tokens": 51682211.0,
      "step": 26420
    },
    {
      "entropy": 6.038824415206909,
      "epoch": 2.641475483580747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043069975470338875,
      "loss": 5.6257,
      "mean_token_accuracy": 0.18060154020786284,
      "num_tokens": 51692005.0,
      "step": 26425
    },
    {
      "entropy": 6.109280967712403,
      "epoch": 2.6419753086419755,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043067398883427907,
      "loss": 5.6615,
      "mean_token_accuracy": 0.18540504574775696,
      "num_tokens": 51701451.0,
      "step": 26430
    },
    {
      "entropy": 5.998253440856933,
      "epoch": 2.6424751337032037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043064821904845185,
      "loss": 5.5478,
      "mean_token_accuracy": 0.18302650153636932,
      "num_tokens": 51710894.0,
      "step": 26435
    },
    {
      "entropy": 6.076818037033081,
      "epoch": 2.6429749587644324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004306224453465556,
      "loss": 5.6947,
      "mean_token_accuracy": 0.17128417491912842,
      "num_tokens": 51721984.0,
      "step": 26440
    },
    {
      "entropy": 5.990346050262451,
      "epoch": 2.643474783825661,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043059666772923854,
      "loss": 5.5551,
      "mean_token_accuracy": 0.18785680681467057,
      "num_tokens": 51731816.0,
      "step": 26445
    },
    {
      "entropy": 6.0463011264801025,
      "epoch": 2.6439746088868894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004305708861971495,
      "loss": 5.7121,
      "mean_token_accuracy": 0.18274737745523453,
      "num_tokens": 51743661.0,
      "step": 26450
    },
    {
      "entropy": 6.047233819961548,
      "epoch": 2.644474433948118,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004305451007509369,
      "loss": 5.6814,
      "mean_token_accuracy": 0.171556556224823,
      "num_tokens": 51753468.0,
      "step": 26455
    },
    {
      "entropy": 6.066139221191406,
      "epoch": 2.644974259009347,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004305193113912496,
      "loss": 5.6319,
      "mean_token_accuracy": 0.1791599452495575,
      "num_tokens": 51763125.0,
      "step": 26460
    },
    {
      "entropy": 6.019484424591065,
      "epoch": 2.6454740840705755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043049351811873645,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1773580715060234,
      "num_tokens": 51773229.0,
      "step": 26465
    },
    {
      "entropy": 6.013549327850342,
      "epoch": 2.6459739091318037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004304677209340464,
      "loss": 5.5987,
      "mean_token_accuracy": 0.17930756360292435,
      "num_tokens": 51783296.0,
      "step": 26470
    },
    {
      "entropy": 6.0443346977233885,
      "epoch": 2.6464737341930324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043044191983782846,
      "loss": 5.58,
      "mean_token_accuracy": 0.1774120643734932,
      "num_tokens": 51792083.0,
      "step": 26475
    },
    {
      "entropy": 6.0843785285949705,
      "epoch": 2.646973559254261,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043041611483073187,
      "loss": 5.7388,
      "mean_token_accuracy": 0.17146477848291397,
      "num_tokens": 51802269.0,
      "step": 26480
    },
    {
      "entropy": 6.0587810516357425,
      "epoch": 2.6474733843154894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303903059134059,
      "loss": 5.6204,
      "mean_token_accuracy": 0.1857772648334503,
      "num_tokens": 51812192.0,
      "step": 26485
    },
    {
      "entropy": 6.030695104598999,
      "epoch": 2.647973209376718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043036449308649967,
      "loss": 5.6452,
      "mean_token_accuracy": 0.181847882270813,
      "num_tokens": 51822065.0,
      "step": 26490
    },
    {
      "entropy": 6.091955947875976,
      "epoch": 2.648473034437947,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043033867635066287,
      "loss": 5.6656,
      "mean_token_accuracy": 0.1755553215742111,
      "num_tokens": 51831470.0,
      "step": 26495
    },
    {
      "entropy": 6.096385812759399,
      "epoch": 2.6489728594991755,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004303128557065448,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17293713837862015,
      "num_tokens": 51841969.0,
      "step": 26500
    },
    {
      "entropy": 6.090498399734497,
      "epoch": 2.6494726845604037,
      "grad_norm": 1.125,
      "learning_rate": 0.0004302870311547954,
      "loss": 5.6619,
      "mean_token_accuracy": 0.1766855776309967,
      "num_tokens": 51851513.0,
      "step": 26505
    },
    {
      "entropy": 6.103252172470093,
      "epoch": 2.6499725096216324,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043026120269606423,
      "loss": 5.6177,
      "mean_token_accuracy": 0.17914734184741973,
      "num_tokens": 51860791.0,
      "step": 26510
    },
    {
      "entropy": 6.086277437210083,
      "epoch": 2.650472334682861,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043023537033100106,
      "loss": 5.6531,
      "mean_token_accuracy": 0.18038948327302934,
      "num_tokens": 51871124.0,
      "step": 26515
    },
    {
      "entropy": 6.040558195114135,
      "epoch": 2.6509721597440894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043020953406025606,
      "loss": 5.5561,
      "mean_token_accuracy": 0.18737245202064515,
      "num_tokens": 51879183.0,
      "step": 26520
    },
    {
      "entropy": 5.966899394989014,
      "epoch": 2.651471984805318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000430183693884479,
      "loss": 5.593,
      "mean_token_accuracy": 0.1845401793718338,
      "num_tokens": 51889824.0,
      "step": 26525
    },
    {
      "entropy": 5.953361845016479,
      "epoch": 2.651971809866547,
      "grad_norm": 1.125,
      "learning_rate": 0.0004301578498043202,
      "loss": 5.5409,
      "mean_token_accuracy": 0.1910584181547165,
      "num_tokens": 51899561.0,
      "step": 26530
    },
    {
      "entropy": 5.998831701278687,
      "epoch": 2.6524716349277755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043013200182042983,
      "loss": 5.6698,
      "mean_token_accuracy": 0.17054789811372756,
      "num_tokens": 51908359.0,
      "step": 26535
    },
    {
      "entropy": 6.026673126220703,
      "epoch": 2.6529714599890037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043010614993345815,
      "loss": 5.5309,
      "mean_token_accuracy": 0.18022550493478776,
      "num_tokens": 51917782.0,
      "step": 26540
    },
    {
      "entropy": 6.036654138565064,
      "epoch": 2.6534712850502324,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004300802941440557,
      "loss": 5.5998,
      "mean_token_accuracy": 0.1769339546561241,
      "num_tokens": 51926218.0,
      "step": 26545
    },
    {
      "entropy": 5.964511203765869,
      "epoch": 2.653971110111461,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004300544344528729,
      "loss": 5.4972,
      "mean_token_accuracy": 0.18609803020954133,
      "num_tokens": 51936541.0,
      "step": 26550
    },
    {
      "entropy": 6.1200672626495365,
      "epoch": 2.6544709351726894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043002857086056053,
      "loss": 5.7436,
      "mean_token_accuracy": 0.16934717148542405,
      "num_tokens": 51945158.0,
      "step": 26555
    },
    {
      "entropy": 6.064972162246704,
      "epoch": 2.654970760233918,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004300027033677691,
      "loss": 5.6848,
      "mean_token_accuracy": 0.1759493350982666,
      "num_tokens": 51955402.0,
      "step": 26560
    },
    {
      "entropy": 6.048012971878052,
      "epoch": 2.655470585295147,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004299768319751496,
      "loss": 5.6086,
      "mean_token_accuracy": 0.18332893252372742,
      "num_tokens": 51965038.0,
      "step": 26565
    },
    {
      "entropy": 6.093061923980713,
      "epoch": 2.6559704103563755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004299509566833529,
      "loss": 5.6653,
      "mean_token_accuracy": 0.17589034885168076,
      "num_tokens": 51974967.0,
      "step": 26570
    },
    {
      "entropy": 6.037825679779052,
      "epoch": 2.6564702354176037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042992507749302995,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17820480614900588,
      "num_tokens": 51985010.0,
      "step": 26575
    },
    {
      "entropy": 6.055655527114868,
      "epoch": 2.6569700604788324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042989919440483194,
      "loss": 5.5931,
      "mean_token_accuracy": 0.18227517157793044,
      "num_tokens": 51994620.0,
      "step": 26580
    },
    {
      "entropy": 6.071619367599487,
      "epoch": 2.657469885540061,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004298733074194101,
      "loss": 5.5638,
      "mean_token_accuracy": 0.18180508464574813,
      "num_tokens": 52003475.0,
      "step": 26585
    },
    {
      "entropy": 6.028344058990479,
      "epoch": 2.6579697106012894,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004298474165374156,
      "loss": 5.5937,
      "mean_token_accuracy": 0.17966216653585435,
      "num_tokens": 52013303.0,
      "step": 26590
    },
    {
      "entropy": 5.980200529098511,
      "epoch": 2.658469535662518,
      "grad_norm": 1.25,
      "learning_rate": 0.0004298215217595,
      "loss": 5.5562,
      "mean_token_accuracy": 0.1847623407840729,
      "num_tokens": 52021944.0,
      "step": 26595
    },
    {
      "entropy": 6.084877395629883,
      "epoch": 2.658969360723747,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042979562308631476,
      "loss": 5.6552,
      "mean_token_accuracy": 0.174858894944191,
      "num_tokens": 52030590.0,
      "step": 26600
    },
    {
      "entropy": 6.047455453872681,
      "epoch": 2.6594691857849755,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004297697205185115,
      "loss": 5.6322,
      "mean_token_accuracy": 0.17801419496536255,
      "num_tokens": 52039861.0,
      "step": 26605
    },
    {
      "entropy": 6.082759380340576,
      "epoch": 2.6599690108462037,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004297438140567419,
      "loss": 5.7151,
      "mean_token_accuracy": 0.1695578932762146,
      "num_tokens": 52051598.0,
      "step": 26610
    },
    {
      "entropy": 6.10412015914917,
      "epoch": 2.6604688359074324,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004297179037016577,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1780354470014572,
      "num_tokens": 52060662.0,
      "step": 26615
    },
    {
      "entropy": 6.1047831058502195,
      "epoch": 2.6609686609686607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042969198945391084,
      "loss": 5.6638,
      "mean_token_accuracy": 0.17644205093383789,
      "num_tokens": 52070166.0,
      "step": 26620
    },
    {
      "entropy": 5.98270263671875,
      "epoch": 2.6614684860298894,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004296660713141534,
      "loss": 5.5857,
      "mean_token_accuracy": 0.18556248992681504,
      "num_tokens": 52079409.0,
      "step": 26625
    },
    {
      "entropy": 6.0187280654907225,
      "epoch": 2.661968311091118,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042964014928303733,
      "loss": 5.6732,
      "mean_token_accuracy": 0.1807694524526596,
      "num_tokens": 52090419.0,
      "step": 26630
    },
    {
      "entropy": 6.10511531829834,
      "epoch": 2.662468136152347,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000429614223361215,
      "loss": 5.6765,
      "mean_token_accuracy": 0.177564400434494,
      "num_tokens": 52100681.0,
      "step": 26635
    },
    {
      "entropy": 6.047874021530151,
      "epoch": 2.6629679612135755,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004295882935493385,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17537523657083512,
      "num_tokens": 52109852.0,
      "step": 26640
    },
    {
      "entropy": 6.0269850730896,
      "epoch": 2.6634677862748037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042956235984806034,
      "loss": 5.6064,
      "mean_token_accuracy": 0.18157270699739456,
      "num_tokens": 52118967.0,
      "step": 26645
    },
    {
      "entropy": 6.106083345413208,
      "epoch": 2.6639676113360324,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042953642225803296,
      "loss": 5.6818,
      "mean_token_accuracy": 0.17949129343032838,
      "num_tokens": 52128570.0,
      "step": 26650
    },
    {
      "entropy": 5.983897686004639,
      "epoch": 2.6644674363972607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000429510480779909,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18703127652406693,
      "num_tokens": 52139683.0,
      "step": 26655
    },
    {
      "entropy": 6.053976917266846,
      "epoch": 2.6649672614584894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004294845354143411,
      "loss": 5.6018,
      "mean_token_accuracy": 0.18000504821538926,
      "num_tokens": 52150139.0,
      "step": 26660
    },
    {
      "entropy": 6.037559556961059,
      "epoch": 2.665467086519718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004294585861619821,
      "loss": 5.6734,
      "mean_token_accuracy": 0.17824286073446274,
      "num_tokens": 52159211.0,
      "step": 26665
    },
    {
      "entropy": 6.1177583694458,
      "epoch": 2.665966911580947,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042943263302348475,
      "loss": 5.7107,
      "mean_token_accuracy": 0.1706299901008606,
      "num_tokens": 52169100.0,
      "step": 26670
    },
    {
      "entropy": 6.0162872791290285,
      "epoch": 2.6664667366421755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042940667599950205,
      "loss": 5.5408,
      "mean_token_accuracy": 0.18543625622987747,
      "num_tokens": 52178336.0,
      "step": 26675
    },
    {
      "entropy": 6.02078914642334,
      "epoch": 2.6669665617034037,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00042938071509068717,
      "loss": 5.6255,
      "mean_token_accuracy": 0.181387035548687,
      "num_tokens": 52188579.0,
      "step": 26680
    },
    {
      "entropy": 6.081357526779175,
      "epoch": 2.6674663867646324,
      "grad_norm": 1.5,
      "learning_rate": 0.00042935475029769323,
      "loss": 5.6537,
      "mean_token_accuracy": 0.17428108751773835,
      "num_tokens": 52197706.0,
      "step": 26685
    },
    {
      "entropy": 6.096717405319214,
      "epoch": 2.6679662118258607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042932878162117347,
      "loss": 5.7132,
      "mean_token_accuracy": 0.17555491477251053,
      "num_tokens": 52207881.0,
      "step": 26690
    },
    {
      "entropy": 6.031832885742188,
      "epoch": 2.6684660368870894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293028090617813,
      "loss": 5.628,
      "mean_token_accuracy": 0.18694858103990555,
      "num_tokens": 52217842.0,
      "step": 26695
    },
    {
      "entropy": 6.092027378082276,
      "epoch": 2.668965861948318,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042927683262017016,
      "loss": 5.6925,
      "mean_token_accuracy": 0.17402174025774003,
      "num_tokens": 52228198.0,
      "step": 26700
    },
    {
      "entropy": 6.129705333709717,
      "epoch": 2.669465687009547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004292508522969935,
      "loss": 5.6361,
      "mean_token_accuracy": 0.17694351822137833,
      "num_tokens": 52238658.0,
      "step": 26705
    },
    {
      "entropy": 6.06276569366455,
      "epoch": 2.6699655120707755,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004292248680929052,
      "loss": 5.6011,
      "mean_token_accuracy": 0.18343767672777175,
      "num_tokens": 52247767.0,
      "step": 26710
    },
    {
      "entropy": 6.114051151275635,
      "epoch": 2.6704653371320037,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042919888000855885,
      "loss": 5.6923,
      "mean_token_accuracy": 0.1765110284090042,
      "num_tokens": 52257536.0,
      "step": 26715
    },
    {
      "entropy": 6.029721450805664,
      "epoch": 2.6709651621932324,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042917288804460835,
      "loss": 5.6476,
      "mean_token_accuracy": 0.18707684725522994,
      "num_tokens": 52269002.0,
      "step": 26720
    },
    {
      "entropy": 6.0494670391082765,
      "epoch": 2.6714649872544607,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004291468922017078,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17258246541023253,
      "num_tokens": 52279065.0,
      "step": 26725
    },
    {
      "entropy": 5.994875383377075,
      "epoch": 2.6719648123156894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000429120892480511,
      "loss": 5.57,
      "mean_token_accuracy": 0.18389893174171448,
      "num_tokens": 52289077.0,
      "step": 26730
    },
    {
      "entropy": 6.047633075714112,
      "epoch": 2.672464637376918,
      "grad_norm": 1.25,
      "learning_rate": 0.0004290948888816722,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1775241643190384,
      "num_tokens": 52298147.0,
      "step": 26735
    },
    {
      "entropy": 6.04411678314209,
      "epoch": 2.672964462438147,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004290688814058456,
      "loss": 5.6388,
      "mean_token_accuracy": 0.18214184790849686,
      "num_tokens": 52308131.0,
      "step": 26740
    },
    {
      "entropy": 5.982518815994263,
      "epoch": 2.6734642874993755,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004290428700536857,
      "loss": 5.5501,
      "mean_token_accuracy": 0.1868968889117241,
      "num_tokens": 52317211.0,
      "step": 26745
    },
    {
      "entropy": 6.119689226150513,
      "epoch": 2.6739641125606037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004290168548258467,
      "loss": 5.7451,
      "mean_token_accuracy": 0.17138187438249589,
      "num_tokens": 52326640.0,
      "step": 26750
    },
    {
      "entropy": 6.0265793800354,
      "epoch": 2.6744639376218324,
      "grad_norm": 1.25,
      "learning_rate": 0.00042899083572298335,
      "loss": 5.6024,
      "mean_token_accuracy": 0.18405210226774216,
      "num_tokens": 52335084.0,
      "step": 26755
    },
    {
      "entropy": 5.992807626724243,
      "epoch": 2.6749637626830607,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042896481274575014,
      "loss": 5.5608,
      "mean_token_accuracy": 0.19096204042434692,
      "num_tokens": 52344982.0,
      "step": 26760
    },
    {
      "entropy": 6.0317973613739015,
      "epoch": 2.6754635877442894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042893878589480195,
      "loss": 5.6171,
      "mean_token_accuracy": 0.1824658617377281,
      "num_tokens": 52353976.0,
      "step": 26765
    },
    {
      "entropy": 6.078367471694946,
      "epoch": 2.675963412805518,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042891275517079345,
      "loss": 5.7292,
      "mean_token_accuracy": 0.16895684450864792,
      "num_tokens": 52363917.0,
      "step": 26770
    },
    {
      "entropy": 6.005734300613403,
      "epoch": 2.676463237866747,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042888672057437964,
      "loss": 5.5544,
      "mean_token_accuracy": 0.18395720720291137,
      "num_tokens": 52373008.0,
      "step": 26775
    },
    {
      "entropy": 6.010800075531006,
      "epoch": 2.6769630629279755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042886068210621557,
      "loss": 5.5719,
      "mean_token_accuracy": 0.17743490636348724,
      "num_tokens": 52383872.0,
      "step": 26780
    },
    {
      "entropy": 6.089404201507568,
      "epoch": 2.6774628879892037,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004288346397669563,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17673787474632263,
      "num_tokens": 52393484.0,
      "step": 26785
    },
    {
      "entropy": 6.058883380889893,
      "epoch": 2.6779627130504324,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004288085935572571,
      "loss": 5.668,
      "mean_token_accuracy": 0.1749977946281433,
      "num_tokens": 52403350.0,
      "step": 26790
    },
    {
      "entropy": 5.9999714374542235,
      "epoch": 2.6784625381116607,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004287825434777733,
      "loss": 5.5835,
      "mean_token_accuracy": 0.18024312257766723,
      "num_tokens": 52412270.0,
      "step": 26795
    },
    {
      "entropy": 6.109327220916748,
      "epoch": 2.6789623631728894,
      "grad_norm": 1.25,
      "learning_rate": 0.00042875648952916024,
      "loss": 5.6253,
      "mean_token_accuracy": 0.18454151302576066,
      "num_tokens": 52422855.0,
      "step": 26800
    },
    {
      "entropy": 6.031548070907593,
      "epoch": 2.679462188234118,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004287304317120734,
      "loss": 5.5943,
      "mean_token_accuracy": 0.18212641924619674,
      "num_tokens": 52432254.0,
      "step": 26805
    },
    {
      "entropy": 5.97008638381958,
      "epoch": 2.679962013295347,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042870437002716854,
      "loss": 5.5905,
      "mean_token_accuracy": 0.1830124318599701,
      "num_tokens": 52441830.0,
      "step": 26810
    },
    {
      "entropy": 6.063038110733032,
      "epoch": 2.680461838356575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004286783044751013,
      "loss": 5.684,
      "mean_token_accuracy": 0.17394119799137114,
      "num_tokens": 52451913.0,
      "step": 26815
    },
    {
      "entropy": 6.125390720367432,
      "epoch": 2.6809616634178037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004286522350565274,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1881584495306015,
      "num_tokens": 52460737.0,
      "step": 26820
    },
    {
      "entropy": 5.977898454666137,
      "epoch": 2.6814614884790324,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042862616177210286,
      "loss": 5.5606,
      "mean_token_accuracy": 0.1826951950788498,
      "num_tokens": 52470398.0,
      "step": 26825
    },
    {
      "entropy": 5.9910815238952635,
      "epoch": 2.6819613135402607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004286000846224836,
      "loss": 5.6289,
      "mean_token_accuracy": 0.18206773102283477,
      "num_tokens": 52481362.0,
      "step": 26830
    },
    {
      "entropy": 6.087036275863648,
      "epoch": 2.6824611386014894,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042857400360832575,
      "loss": 5.5843,
      "mean_token_accuracy": 0.1861228570342064,
      "num_tokens": 52490724.0,
      "step": 26835
    },
    {
      "entropy": 6.062619018554687,
      "epoch": 2.682960963662718,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004285479187302854,
      "loss": 5.64,
      "mean_token_accuracy": 0.17833374440670013,
      "num_tokens": 52500038.0,
      "step": 26840
    },
    {
      "entropy": 6.060660552978516,
      "epoch": 2.6834607887239468,
      "grad_norm": 1.109375,
      "learning_rate": 0.000428521829989019,
      "loss": 5.5884,
      "mean_token_accuracy": 0.18244792073965072,
      "num_tokens": 52509796.0,
      "step": 26845
    },
    {
      "entropy": 6.0190112590789795,
      "epoch": 2.683960613785175,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004284957373851829,
      "loss": 5.6116,
      "mean_token_accuracy": 0.18119607865810394,
      "num_tokens": 52519373.0,
      "step": 26850
    },
    {
      "entropy": 6.041742420196533,
      "epoch": 2.6844604388464037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042846964091943347,
      "loss": 5.6182,
      "mean_token_accuracy": 0.18393932431936263,
      "num_tokens": 52530073.0,
      "step": 26855
    },
    {
      "entropy": 6.06521201133728,
      "epoch": 2.6849602639076324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042844354059242735,
      "loss": 5.7202,
      "mean_token_accuracy": 0.17468787282705306,
      "num_tokens": 52538995.0,
      "step": 26860
    },
    {
      "entropy": 6.053931951522827,
      "epoch": 2.6854600889688607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004284174364048213,
      "loss": 5.6026,
      "mean_token_accuracy": 0.1872763603925705,
      "num_tokens": 52547824.0,
      "step": 26865
    },
    {
      "entropy": 6.042893552780152,
      "epoch": 2.6859599140300894,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000428391328357272,
      "loss": 5.6076,
      "mean_token_accuracy": 0.18020243048667908,
      "num_tokens": 52557215.0,
      "step": 26870
    },
    {
      "entropy": 6.036085987091065,
      "epoch": 2.686459739091318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004283652164504364,
      "loss": 5.6732,
      "mean_token_accuracy": 0.17615024000406265,
      "num_tokens": 52567557.0,
      "step": 26875
    },
    {
      "entropy": 6.12913408279419,
      "epoch": 2.6869595641525468,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042833910068497134,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1795722022652626,
      "num_tokens": 52577597.0,
      "step": 26880
    },
    {
      "entropy": 6.101193857192993,
      "epoch": 2.687459389213775,
      "grad_norm": 1.21875,
      "learning_rate": 0.000428312981061534,
      "loss": 5.6522,
      "mean_token_accuracy": 0.1770983874797821,
      "num_tokens": 52589183.0,
      "step": 26885
    },
    {
      "entropy": 6.00059757232666,
      "epoch": 2.6879592142750037,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004282868575807815,
      "loss": 5.5893,
      "mean_token_accuracy": 0.18097124099731446,
      "num_tokens": 52598846.0,
      "step": 26890
    },
    {
      "entropy": 6.045957469940186,
      "epoch": 2.6884590393362324,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042826073024337117,
      "loss": 5.654,
      "mean_token_accuracy": 0.16930387169122696,
      "num_tokens": 52609072.0,
      "step": 26895
    },
    {
      "entropy": 6.102137517929077,
      "epoch": 2.6889588643974607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004282345990499602,
      "loss": 5.7566,
      "mean_token_accuracy": 0.1696426786482334,
      "num_tokens": 52619648.0,
      "step": 26900
    },
    {
      "entropy": 6.094373893737793,
      "epoch": 2.6894586894586894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004282084640012062,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17668427377939225,
      "num_tokens": 52629992.0,
      "step": 26905
    },
    {
      "entropy": 6.05767593383789,
      "epoch": 2.689958514519918,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004281823250977667,
      "loss": 5.5877,
      "mean_token_accuracy": 0.18002481907606124,
      "num_tokens": 52639174.0,
      "step": 26910
    },
    {
      "entropy": 5.911628532409668,
      "epoch": 2.6904583395811468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042815618234029923,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19879125952720642,
      "num_tokens": 52648254.0,
      "step": 26915
    },
    {
      "entropy": 5.955444812774658,
      "epoch": 2.690958164642375,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042813003572946163,
      "loss": 5.597,
      "mean_token_accuracy": 0.18432983309030532,
      "num_tokens": 52658072.0,
      "step": 26920
    },
    {
      "entropy": 6.064718866348267,
      "epoch": 2.6914579897036037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042810388526591184,
      "loss": 5.6804,
      "mean_token_accuracy": 0.1821260154247284,
      "num_tokens": 52668868.0,
      "step": 26925
    },
    {
      "entropy": 6.082945013046265,
      "epoch": 2.6919578147648324,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042807773095030757,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1781650111079216,
      "num_tokens": 52678848.0,
      "step": 26930
    },
    {
      "entropy": 5.900856924057007,
      "epoch": 2.6924576398260607,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004280515727833071,
      "loss": 5.4477,
      "mean_token_accuracy": 0.19660942107439042,
      "num_tokens": 52689690.0,
      "step": 26935
    },
    {
      "entropy": 5.9871828079223635,
      "epoch": 2.6929574648872894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004280254107655684,
      "loss": 5.584,
      "mean_token_accuracy": 0.18158626854419707,
      "num_tokens": 52699284.0,
      "step": 26940
    },
    {
      "entropy": 6.110007858276367,
      "epoch": 2.693457289948518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004279992448977497,
      "loss": 5.7205,
      "mean_token_accuracy": 0.17692452073097228,
      "num_tokens": 52709961.0,
      "step": 26945
    },
    {
      "entropy": 5.9969336032867435,
      "epoch": 2.6939571150097468,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004279730751805094,
      "loss": 5.553,
      "mean_token_accuracy": 0.18442232459783553,
      "num_tokens": 52720655.0,
      "step": 26950
    },
    {
      "entropy": 6.072439193725586,
      "epoch": 2.694456940070975,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042794690161450583,
      "loss": 5.6137,
      "mean_token_accuracy": 0.1878785640001297,
      "num_tokens": 52729628.0,
      "step": 26955
    },
    {
      "entropy": 5.942128229141235,
      "epoch": 2.6949567651322037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004279207242003976,
      "loss": 5.4684,
      "mean_token_accuracy": 0.19135302156209946,
      "num_tokens": 52738685.0,
      "step": 26960
    },
    {
      "entropy": 5.9455650806427,
      "epoch": 2.6954565901934324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042789454293884336,
      "loss": 5.5935,
      "mean_token_accuracy": 0.18152989000082015,
      "num_tokens": 52748249.0,
      "step": 26965
    },
    {
      "entropy": 6.060465621948242,
      "epoch": 2.6959564152546607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004278683578305017,
      "loss": 5.7362,
      "mean_token_accuracy": 0.17758633494377135,
      "num_tokens": 52758447.0,
      "step": 26970
    },
    {
      "entropy": 6.148628425598145,
      "epoch": 2.6964562403158894,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042784216887603154,
      "loss": 5.7025,
      "mean_token_accuracy": 0.17833295315504075,
      "num_tokens": 52768448.0,
      "step": 26975
    },
    {
      "entropy": 6.124154853820801,
      "epoch": 2.696956065377118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004278159760760917,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17527024298906327,
      "num_tokens": 52778831.0,
      "step": 26980
    },
    {
      "entropy": 6.016594123840332,
      "epoch": 2.6974558904383468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004277897794313412,
      "loss": 5.5859,
      "mean_token_accuracy": 0.17764904350042343,
      "num_tokens": 52788915.0,
      "step": 26985
    },
    {
      "entropy": 6.0113465785980225,
      "epoch": 2.697955715499575,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004277635789424393,
      "loss": 5.6335,
      "mean_token_accuracy": 0.1785183548927307,
      "num_tokens": 52798670.0,
      "step": 26990
    },
    {
      "entropy": 6.106602764129638,
      "epoch": 2.6984555405608037,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004277373746100449,
      "loss": 5.643,
      "mean_token_accuracy": 0.17983695417642592,
      "num_tokens": 52807783.0,
      "step": 26995
    },
    {
      "entropy": 6.108546447753906,
      "epoch": 2.6989553656220324,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042771116643481753,
      "loss": 5.6786,
      "mean_token_accuracy": 0.16714686006307602,
      "num_tokens": 52816664.0,
      "step": 27000
    },
    {
      "epoch": 2.6989553656220324,
      "eval_entropy": 5.843713487493903,
      "eval_loss": 5.73512601852417,
      "eval_mean_token_accuracy": 0.1820072148823609,
      "eval_num_tokens": 52816664.0,
      "eval_runtime": 23.9019,
      "eval_samples_per_second": 1298.852,
      "eval_steps_per_second": 162.372,
      "step": 27000
    },
    {
      "entropy": 6.03926420211792,
      "epoch": 2.6994551906832607,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004276849544174166,
      "loss": 5.6581,
      "mean_token_accuracy": 0.17595646381378174,
      "num_tokens": 52826241.0,
      "step": 27005
    },
    {
      "entropy": 6.092903804779053,
      "epoch": 2.6999550157444894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042765873855850133,
      "loss": 5.6099,
      "mean_token_accuracy": 0.18059709668159485,
      "num_tokens": 52836138.0,
      "step": 27010
    },
    {
      "entropy": 6.063237094879151,
      "epoch": 2.700454840805718,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004276325188587316,
      "loss": 5.5721,
      "mean_token_accuracy": 0.18232734352350236,
      "num_tokens": 52845679.0,
      "step": 27015
    },
    {
      "entropy": 6.007856845855713,
      "epoch": 2.7009546658669468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004276062953187669,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18395509868860244,
      "num_tokens": 52855455.0,
      "step": 27020
    },
    {
      "entropy": 6.083292675018311,
      "epoch": 2.701454490928175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004275800679392671,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18079634457826615,
      "num_tokens": 52865589.0,
      "step": 27025
    },
    {
      "entropy": 6.045676612854004,
      "epoch": 2.7019543159894037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004275538367208921,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17637289315462112,
      "num_tokens": 52877087.0,
      "step": 27030
    },
    {
      "entropy": 6.198502397537231,
      "epoch": 2.7024541410506324,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042752760166430185,
      "loss": 5.7534,
      "mean_token_accuracy": 0.16752005442976953,
      "num_tokens": 52887370.0,
      "step": 27035
    },
    {
      "entropy": 6.131304693222046,
      "epoch": 2.7029539661118607,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042750136277015625,
      "loss": 5.6493,
      "mean_token_accuracy": 0.1733310639858246,
      "num_tokens": 52897744.0,
      "step": 27040
    },
    {
      "entropy": 6.087254619598388,
      "epoch": 2.7034537911730894,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042747512003911577,
      "loss": 5.6061,
      "mean_token_accuracy": 0.18026783764362336,
      "num_tokens": 52907008.0,
      "step": 27045
    },
    {
      "entropy": 6.032212018966675,
      "epoch": 2.703953616234318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042744887347184035,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17535370439291,
      "num_tokens": 52916941.0,
      "step": 27050
    },
    {
      "entropy": 6.02375864982605,
      "epoch": 2.7044534412955468,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004274226230689906,
      "loss": 5.6356,
      "mean_token_accuracy": 0.17818123251199722,
      "num_tokens": 52927381.0,
      "step": 27055
    },
    {
      "entropy": 6.036500692367554,
      "epoch": 2.704953266356775,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042739636883122687,
      "loss": 5.605,
      "mean_token_accuracy": 0.18654220849275588,
      "num_tokens": 52937462.0,
      "step": 27060
    },
    {
      "entropy": 6.049922895431519,
      "epoch": 2.7054530914180037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042737011075920974,
      "loss": 5.6206,
      "mean_token_accuracy": 0.1778799071907997,
      "num_tokens": 52946612.0,
      "step": 27065
    },
    {
      "entropy": 6.017063045501709,
      "epoch": 2.7059529164792324,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004273438488535997,
      "loss": 5.5957,
      "mean_token_accuracy": 0.18237807154655455,
      "num_tokens": 52955814.0,
      "step": 27070
    },
    {
      "entropy": 5.945138311386108,
      "epoch": 2.7064527415404607,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042731758311505773,
      "loss": 5.57,
      "mean_token_accuracy": 0.185306616127491,
      "num_tokens": 52965586.0,
      "step": 27075
    },
    {
      "entropy": 6.004530954360962,
      "epoch": 2.7069525666016894,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00042729131354424455,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18798395693302156,
      "num_tokens": 52975805.0,
      "step": 27080
    },
    {
      "entropy": 6.120990562438965,
      "epoch": 2.707452391662918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004272650401418211,
      "loss": 5.6852,
      "mean_token_accuracy": 0.1712728187441826,
      "num_tokens": 52986040.0,
      "step": 27085
    },
    {
      "entropy": 6.018300771713257,
      "epoch": 2.7079522167241468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004272387629084484,
      "loss": 5.597,
      "mean_token_accuracy": 0.1818310499191284,
      "num_tokens": 52995660.0,
      "step": 27090
    },
    {
      "entropy": 6.064320135116577,
      "epoch": 2.708452041785375,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042721248184478765,
      "loss": 5.7427,
      "mean_token_accuracy": 0.17306471168994902,
      "num_tokens": 53005193.0,
      "step": 27095
    },
    {
      "entropy": 6.123877477645874,
      "epoch": 2.7089518668466037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004271861969515,
      "loss": 5.682,
      "mean_token_accuracy": 0.18094757348299026,
      "num_tokens": 53015624.0,
      "step": 27100
    },
    {
      "entropy": 6.032697582244873,
      "epoch": 2.7094516919078324,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042715990822924673,
      "loss": 5.5946,
      "mean_token_accuracy": 0.17955675423145295,
      "num_tokens": 53026673.0,
      "step": 27105
    },
    {
      "entropy": 6.112025165557862,
      "epoch": 2.7099515169690607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042713361567868946,
      "loss": 5.6294,
      "mean_token_accuracy": 0.17946127206087112,
      "num_tokens": 53035421.0,
      "step": 27110
    },
    {
      "entropy": 6.075194597244263,
      "epoch": 2.7104513420302894,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042710731930048945,
      "loss": 5.6582,
      "mean_token_accuracy": 0.1795593097805977,
      "num_tokens": 53045132.0,
      "step": 27115
    },
    {
      "entropy": 6.0122510433197025,
      "epoch": 2.710951167091518,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004270810190953085,
      "loss": 5.5838,
      "mean_token_accuracy": 0.18204703330993652,
      "num_tokens": 53055176.0,
      "step": 27120
    },
    {
      "entropy": 5.959026575088501,
      "epoch": 2.7114509921527468,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004270547150638082,
      "loss": 5.5966,
      "mean_token_accuracy": 0.18448735475540162,
      "num_tokens": 53064846.0,
      "step": 27125
    },
    {
      "entropy": 5.992881155014038,
      "epoch": 2.711950817213975,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042702840720665043,
      "loss": 5.6025,
      "mean_token_accuracy": 0.18808460384607315,
      "num_tokens": 53074669.0,
      "step": 27130
    },
    {
      "entropy": 6.141991758346558,
      "epoch": 2.7124506422752037,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042700209552449706,
      "loss": 5.636,
      "mean_token_accuracy": 0.17640046775341034,
      "num_tokens": 53084196.0,
      "step": 27135
    },
    {
      "entropy": 6.02795181274414,
      "epoch": 2.7129504673364324,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042697578001801,
      "loss": 5.6098,
      "mean_token_accuracy": 0.179369755089283,
      "num_tokens": 53094435.0,
      "step": 27140
    },
    {
      "entropy": 5.985828304290772,
      "epoch": 2.7134502923976607,
      "grad_norm": 1.125,
      "learning_rate": 0.0004269494606878515,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1859418734908104,
      "num_tokens": 53104221.0,
      "step": 27145
    },
    {
      "entropy": 6.046533870697021,
      "epoch": 2.7139501174588894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004269231375346836,
      "loss": 5.6479,
      "mean_token_accuracy": 0.17590133398771285,
      "num_tokens": 53113685.0,
      "step": 27150
    },
    {
      "entropy": 6.127332496643066,
      "epoch": 2.714449942520118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004268968105591687,
      "loss": 5.7149,
      "mean_token_accuracy": 0.16670687049627303,
      "num_tokens": 53125353.0,
      "step": 27155
    },
    {
      "entropy": 6.037249946594239,
      "epoch": 2.7149497675813468,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004268704797619692,
      "loss": 5.6259,
      "mean_token_accuracy": 0.18056243062019348,
      "num_tokens": 53133860.0,
      "step": 27160
    },
    {
      "entropy": 5.989230394363403,
      "epoch": 2.715449592642575,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042684414514374734,
      "loss": 5.5332,
      "mean_token_accuracy": 0.18514848500490189,
      "num_tokens": 53142700.0,
      "step": 27165
    },
    {
      "entropy": 6.067013692855835,
      "epoch": 2.7159494177038037,
      "grad_norm": 1.625,
      "learning_rate": 0.00042681780670516595,
      "loss": 5.5845,
      "mean_token_accuracy": 0.17540451884269714,
      "num_tokens": 53152517.0,
      "step": 27170
    },
    {
      "entropy": 5.984105062484741,
      "epoch": 2.716449242765032,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004267914644468877,
      "loss": 5.4994,
      "mean_token_accuracy": 0.18723472952842712,
      "num_tokens": 53162270.0,
      "step": 27175
    },
    {
      "entropy": 6.034225606918335,
      "epoch": 2.7169490678262607,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004267651183695752,
      "loss": 5.7461,
      "mean_token_accuracy": 0.17318395227193834,
      "num_tokens": 53173664.0,
      "step": 27180
    },
    {
      "entropy": 6.0678380012512205,
      "epoch": 2.7174488928874894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004267387684738913,
      "loss": 5.6385,
      "mean_token_accuracy": 0.1769708588719368,
      "num_tokens": 53184668.0,
      "step": 27185
    },
    {
      "entropy": 6.089393186569214,
      "epoch": 2.717948717948718,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004267124147604991,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18318649679422377,
      "num_tokens": 53194107.0,
      "step": 27190
    },
    {
      "entropy": 6.024494695663452,
      "epoch": 2.7184485430099468,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004266860572300615,
      "loss": 5.599,
      "mean_token_accuracy": 0.1838911309838295,
      "num_tokens": 53204248.0,
      "step": 27195
    },
    {
      "entropy": 6.085969734191894,
      "epoch": 2.718948368071175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004266596958832419,
      "loss": 5.681,
      "mean_token_accuracy": 0.16975592076778412,
      "num_tokens": 53213415.0,
      "step": 27200
    },
    {
      "entropy": 6.194616079330444,
      "epoch": 2.7194481931324037,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004266333307207032,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1718432068824768,
      "num_tokens": 53224403.0,
      "step": 27205
    },
    {
      "entropy": 6.113743257522583,
      "epoch": 2.719948018193632,
      "grad_norm": 1.34375,
      "learning_rate": 0.00042660696174310904,
      "loss": 5.6626,
      "mean_token_accuracy": 0.17695599645376206,
      "num_tokens": 53233890.0,
      "step": 27210
    },
    {
      "entropy": 6.039796686172485,
      "epoch": 2.7204478432548607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265805889511227,
      "loss": 5.6606,
      "mean_token_accuracy": 0.17375041544437408,
      "num_tokens": 53243783.0,
      "step": 27215
    },
    {
      "entropy": 6.066599941253662,
      "epoch": 2.7209476683160894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004265542123454077,
      "loss": 5.648,
      "mean_token_accuracy": 0.1789502054452896,
      "num_tokens": 53252941.0,
      "step": 27220
    },
    {
      "entropy": 6.084618711471558,
      "epoch": 2.721447493377318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042652783192662774,
      "loss": 5.653,
      "mean_token_accuracy": 0.17953821122646332,
      "num_tokens": 53262846.0,
      "step": 27225
    },
    {
      "entropy": 6.089793729782104,
      "epoch": 2.7219473184385468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004265014476954465,
      "loss": 5.7161,
      "mean_token_accuracy": 0.17524535059928895,
      "num_tokens": 53273416.0,
      "step": 27230
    },
    {
      "entropy": 6.095695161819458,
      "epoch": 2.722447143499775,
      "grad_norm": 1.25,
      "learning_rate": 0.0004264750596525279,
      "loss": 5.6658,
      "mean_token_accuracy": 0.18075388073921203,
      "num_tokens": 53282815.0,
      "step": 27235
    },
    {
      "entropy": 6.043470144271851,
      "epoch": 2.7229469685610037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042644866779853577,
      "loss": 5.6057,
      "mean_token_accuracy": 0.17653962224721909,
      "num_tokens": 53292355.0,
      "step": 27240
    },
    {
      "entropy": 6.010481977462769,
      "epoch": 2.723446793622232,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042642227213413405,
      "loss": 5.6527,
      "mean_token_accuracy": 0.18240527361631392,
      "num_tokens": 53301947.0,
      "step": 27245
    },
    {
      "entropy": 6.049094295501709,
      "epoch": 2.7239466186834607,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042639587265998696,
      "loss": 5.659,
      "mean_token_accuracy": 0.17736599892377852,
      "num_tokens": 53311753.0,
      "step": 27250
    },
    {
      "entropy": 6.1108269691467285,
      "epoch": 2.7244464437446894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004263694693767587,
      "loss": 5.6554,
      "mean_token_accuracy": 0.17721427083015442,
      "num_tokens": 53321447.0,
      "step": 27255
    },
    {
      "entropy": 6.04491491317749,
      "epoch": 2.724946268805918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004263430622851135,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18309365063905716,
      "num_tokens": 53331809.0,
      "step": 27260
    },
    {
      "entropy": 6.078882265090942,
      "epoch": 2.7254460938671468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004263166513857158,
      "loss": 5.6824,
      "mean_token_accuracy": 0.1805805280804634,
      "num_tokens": 53341674.0,
      "step": 27265
    },
    {
      "entropy": 5.949611520767212,
      "epoch": 2.725945918928375,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042629023667923016,
      "loss": 5.5008,
      "mean_token_accuracy": 0.19125837385654448,
      "num_tokens": 53350689.0,
      "step": 27270
    },
    {
      "entropy": 6.028753280639648,
      "epoch": 2.7264457439896037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042626381816632096,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17340307235717772,
      "num_tokens": 53360596.0,
      "step": 27275
    },
    {
      "entropy": 6.097342491149902,
      "epoch": 2.726945569050832,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042623739584765315,
      "loss": 5.6227,
      "mean_token_accuracy": 0.17678437232971192,
      "num_tokens": 53370277.0,
      "step": 27280
    },
    {
      "entropy": 6.017086315155029,
      "epoch": 2.7274453941120607,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042621096972389136,
      "loss": 5.5822,
      "mean_token_accuracy": 0.18103986084461213,
      "num_tokens": 53379479.0,
      "step": 27285
    },
    {
      "entropy": 6.069314289093017,
      "epoch": 2.7279452191732894,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042618453979570035,
      "loss": 5.6233,
      "mean_token_accuracy": 0.18032572269439698,
      "num_tokens": 53389175.0,
      "step": 27290
    },
    {
      "entropy": 5.961041736602783,
      "epoch": 2.728445044234518,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042615810606374535,
      "loss": 5.4949,
      "mean_token_accuracy": 0.19160650074481964,
      "num_tokens": 53398941.0,
      "step": 27295
    },
    {
      "entropy": 6.0216480731964115,
      "epoch": 2.7289448692957468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004261316685286913,
      "loss": 5.5486,
      "mean_token_accuracy": 0.1825973093509674,
      "num_tokens": 53408501.0,
      "step": 27300
    },
    {
      "entropy": 6.058285808563232,
      "epoch": 2.729444694356975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004261052271912033,
      "loss": 5.6063,
      "mean_token_accuracy": 0.1834433376789093,
      "num_tokens": 53418272.0,
      "step": 27305
    },
    {
      "entropy": 6.0416093349456785,
      "epoch": 2.7299445194182037,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004260787820519467,
      "loss": 5.6209,
      "mean_token_accuracy": 0.18418028205633163,
      "num_tokens": 53428433.0,
      "step": 27310
    },
    {
      "entropy": 6.1094530582427975,
      "epoch": 2.730444344479432,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004260523331115868,
      "loss": 5.7276,
      "mean_token_accuracy": 0.1703392446041107,
      "num_tokens": 53438023.0,
      "step": 27315
    },
    {
      "entropy": 6.083354187011719,
      "epoch": 2.7309441695406607,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00042602588037078903,
      "loss": 5.5881,
      "mean_token_accuracy": 0.18495915234088897,
      "num_tokens": 53447492.0,
      "step": 27320
    },
    {
      "entropy": 6.042680788040161,
      "epoch": 2.7314439946018894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042599942383021906,
      "loss": 5.675,
      "mean_token_accuracy": 0.17032116353511811,
      "num_tokens": 53457563.0,
      "step": 27325
    },
    {
      "entropy": 5.993863868713379,
      "epoch": 2.731943819663118,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004259729634905424,
      "loss": 5.6247,
      "mean_token_accuracy": 0.17819541841745376,
      "num_tokens": 53466616.0,
      "step": 27330
    },
    {
      "entropy": 6.083539772033691,
      "epoch": 2.7324436447243468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042594649935242493,
      "loss": 5.6393,
      "mean_token_accuracy": 0.17788549065589904,
      "num_tokens": 53476365.0,
      "step": 27335
    },
    {
      "entropy": 5.988248825073242,
      "epoch": 2.732943469785575,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004259200314165322,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1881335496902466,
      "num_tokens": 53485581.0,
      "step": 27340
    },
    {
      "entropy": 5.942099475860596,
      "epoch": 2.7334432948468037,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004258935596835303,
      "loss": 5.5728,
      "mean_token_accuracy": 0.17852835059165956,
      "num_tokens": 53495533.0,
      "step": 27345
    },
    {
      "entropy": 6.034225559234619,
      "epoch": 2.733943119908032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042586708415408547,
      "loss": 5.6101,
      "mean_token_accuracy": 0.17463312894105912,
      "num_tokens": 53505530.0,
      "step": 27350
    },
    {
      "entropy": 6.090355062484742,
      "epoch": 2.7344429449692607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004258406048288635,
      "loss": 5.6977,
      "mean_token_accuracy": 0.1751638576388359,
      "num_tokens": 53515276.0,
      "step": 27355
    },
    {
      "entropy": 6.089758634567261,
      "epoch": 2.7349427700304894,
      "grad_norm": 1.40625,
      "learning_rate": 0.00042581412170853073,
      "loss": 5.6599,
      "mean_token_accuracy": 0.18723128288984298,
      "num_tokens": 53525213.0,
      "step": 27360
    },
    {
      "entropy": 6.025127983093261,
      "epoch": 2.735442595091718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004257876347937535,
      "loss": 5.6508,
      "mean_token_accuracy": 0.18152974247932435,
      "num_tokens": 53534691.0,
      "step": 27365
    },
    {
      "entropy": 5.950275659561157,
      "epoch": 2.7359424201529463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004257611440851981,
      "loss": 5.4679,
      "mean_token_accuracy": 0.19857280552387238,
      "num_tokens": 53545878.0,
      "step": 27370
    },
    {
      "entropy": 6.045867824554444,
      "epoch": 2.736442245214175,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042573464958353116,
      "loss": 5.586,
      "mean_token_accuracy": 0.18304882049560547,
      "num_tokens": 53555757.0,
      "step": 27375
    },
    {
      "entropy": 6.060261631011963,
      "epoch": 2.7369420702754037,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004257081512894192,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17251193523406982,
      "num_tokens": 53565494.0,
      "step": 27380
    },
    {
      "entropy": 6.057809162139892,
      "epoch": 2.737441895336632,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042568164920352895,
      "loss": 5.6684,
      "mean_token_accuracy": 0.18045217245817186,
      "num_tokens": 53574302.0,
      "step": 27385
    },
    {
      "entropy": 6.014428806304932,
      "epoch": 2.7379417203978607,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004256551433265273,
      "loss": 5.6464,
      "mean_token_accuracy": 0.17961916327476501,
      "num_tokens": 53585447.0,
      "step": 27390
    },
    {
      "entropy": 6.121490526199341,
      "epoch": 2.7384415454590894,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004256286336590808,
      "loss": 5.634,
      "mean_token_accuracy": 0.17177849858999253,
      "num_tokens": 53594728.0,
      "step": 27395
    },
    {
      "entropy": 5.9265360832214355,
      "epoch": 2.738941370520318,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004256021202018568,
      "loss": 5.4699,
      "mean_token_accuracy": 0.18576357066631316,
      "num_tokens": 53604961.0,
      "step": 27400
    },
    {
      "entropy": 5.941939830780029,
      "epoch": 2.7394411955815463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004255756029555222,
      "loss": 5.5627,
      "mean_token_accuracy": 0.1932597666978836,
      "num_tokens": 53614736.0,
      "step": 27405
    },
    {
      "entropy": 5.922925233840942,
      "epoch": 2.739941020642775,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042554908192074406,
      "loss": 5.4384,
      "mean_token_accuracy": 0.20116442143917085,
      "num_tokens": 53623740.0,
      "step": 27410
    },
    {
      "entropy": 5.963981103897095,
      "epoch": 2.7404408457040037,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042552255709818975,
      "loss": 5.5818,
      "mean_token_accuracy": 0.18492401093244554,
      "num_tokens": 53633901.0,
      "step": 27415
    },
    {
      "entropy": 6.177663898468017,
      "epoch": 2.740940670765232,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042549602848852673,
      "loss": 5.8001,
      "mean_token_accuracy": 0.17128297835588455,
      "num_tokens": 53644242.0,
      "step": 27420
    },
    {
      "entropy": 6.027837610244751,
      "epoch": 2.7414404958264607,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004254694960924224,
      "loss": 5.5649,
      "mean_token_accuracy": 0.183864089846611,
      "num_tokens": 53654764.0,
      "step": 27425
    },
    {
      "entropy": 5.910456418991089,
      "epoch": 2.7419403208876894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004254429599105442,
      "loss": 5.4325,
      "mean_token_accuracy": 0.18616232573986052,
      "num_tokens": 53664746.0,
      "step": 27430
    },
    {
      "entropy": 6.055664587020874,
      "epoch": 2.742440145948918,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004254164199435598,
      "loss": 5.667,
      "mean_token_accuracy": 0.17773000448942183,
      "num_tokens": 53673586.0,
      "step": 27435
    },
    {
      "entropy": 6.017638301849365,
      "epoch": 2.7429399710101463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004253898761921371,
      "loss": 5.5714,
      "mean_token_accuracy": 0.1836363971233368,
      "num_tokens": 53683907.0,
      "step": 27440
    },
    {
      "entropy": 6.025735187530517,
      "epoch": 2.743439796071375,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004253633286569437,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17979005873203277,
      "num_tokens": 53694018.0,
      "step": 27445
    },
    {
      "entropy": 6.172735071182251,
      "epoch": 2.7439396211326037,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004253367773386476,
      "loss": 5.8233,
      "mean_token_accuracy": 0.16961902529001235,
      "num_tokens": 53705057.0,
      "step": 27450
    },
    {
      "entropy": 6.075808572769165,
      "epoch": 2.744439446193832,
      "grad_norm": 1.265625,
      "learning_rate": 0.000425310222237917,
      "loss": 5.557,
      "mean_token_accuracy": 0.18130016922950745,
      "num_tokens": 53713848.0,
      "step": 27455
    },
    {
      "entropy": 6.046567153930664,
      "epoch": 2.7449392712550607,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042528366335541984,
      "loss": 5.6317,
      "mean_token_accuracy": 0.17437636107206345,
      "num_tokens": 53722716.0,
      "step": 27460
    },
    {
      "entropy": 5.982735729217529,
      "epoch": 2.7454390963162894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004252571006918243,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19001764804124832,
      "num_tokens": 53732450.0,
      "step": 27465
    },
    {
      "entropy": 6.051136636734009,
      "epoch": 2.745938921377518,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004252305342477989,
      "loss": 5.6138,
      "mean_token_accuracy": 0.18190599381923675,
      "num_tokens": 53742050.0,
      "step": 27470
    },
    {
      "entropy": 6.070226049423217,
      "epoch": 2.7464387464387463,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042520396402401185,
      "loss": 5.64,
      "mean_token_accuracy": 0.17663216590881348,
      "num_tokens": 53751392.0,
      "step": 27475
    },
    {
      "entropy": 6.101192426681519,
      "epoch": 2.746938571499975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004251773900211317,
      "loss": 5.7091,
      "mean_token_accuracy": 0.1703242301940918,
      "num_tokens": 53761477.0,
      "step": 27480
    },
    {
      "entropy": 6.0996582984924315,
      "epoch": 2.7474383965612037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042515081223982716,
      "loss": 5.6178,
      "mean_token_accuracy": 0.17865118980407715,
      "num_tokens": 53771046.0,
      "step": 27485
    },
    {
      "entropy": 6.026313543319702,
      "epoch": 2.747938221622432,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042512423068076673,
      "loss": 5.6384,
      "mean_token_accuracy": 0.17947592139244078,
      "num_tokens": 53781259.0,
      "step": 27490
    },
    {
      "entropy": 6.061674451828003,
      "epoch": 2.7484380466836607,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004250976453446193,
      "loss": 5.6199,
      "mean_token_accuracy": 0.18232771009206772,
      "num_tokens": 53790895.0,
      "step": 27495
    },
    {
      "entropy": 6.05341682434082,
      "epoch": 2.7489378717448893,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004250710562320538,
      "loss": 5.6118,
      "mean_token_accuracy": 0.18045377880334854,
      "num_tokens": 53800120.0,
      "step": 27500
    },
    {
      "entropy": 5.992454719543457,
      "epoch": 2.749437696806118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042504446334373905,
      "loss": 5.6043,
      "mean_token_accuracy": 0.17790991365909575,
      "num_tokens": 53809960.0,
      "step": 27505
    },
    {
      "entropy": 5.99509825706482,
      "epoch": 2.7499375218673463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004250178666803442,
      "loss": 5.5827,
      "mean_token_accuracy": 0.1806960090994835,
      "num_tokens": 53819356.0,
      "step": 27510
    },
    {
      "entropy": 6.007902050018311,
      "epoch": 2.750437346928575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004249912662425385,
      "loss": 5.4952,
      "mean_token_accuracy": 0.19082632064819335,
      "num_tokens": 53830026.0,
      "step": 27515
    },
    {
      "entropy": 6.049623346328735,
      "epoch": 2.7509371719898037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042496466203099107,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17594922780990602,
      "num_tokens": 53840040.0,
      "step": 27520
    },
    {
      "entropy": 6.0627258777618405,
      "epoch": 2.751436997051032,
      "grad_norm": 1.390625,
      "learning_rate": 0.00042493805404637136,
      "loss": 5.654,
      "mean_token_accuracy": 0.175443634390831,
      "num_tokens": 53850106.0,
      "step": 27525
    },
    {
      "entropy": 6.055921125411987,
      "epoch": 2.7519368221122606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042491144228934885,
      "loss": 5.6173,
      "mean_token_accuracy": 0.18542161583900452,
      "num_tokens": 53861312.0,
      "step": 27530
    },
    {
      "entropy": 6.10058479309082,
      "epoch": 2.7524366471734893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000424884826760593,
      "loss": 5.7267,
      "mean_token_accuracy": 0.17635586261749267,
      "num_tokens": 53871914.0,
      "step": 27535
    },
    {
      "entropy": 6.048169088363648,
      "epoch": 2.752936472234718,
      "grad_norm": 1.25,
      "learning_rate": 0.00042485820746077346,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17616316825151443,
      "num_tokens": 53881241.0,
      "step": 27540
    },
    {
      "entropy": 6.030424880981445,
      "epoch": 2.7534362972959463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042483158439055995,
      "loss": 5.6676,
      "mean_token_accuracy": 0.17065680921077728,
      "num_tokens": 53891002.0,
      "step": 27545
    },
    {
      "entropy": 6.098529386520386,
      "epoch": 2.753936122357175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004248049575506224,
      "loss": 5.6615,
      "mean_token_accuracy": 0.17537735849618913,
      "num_tokens": 53900998.0,
      "step": 27550
    },
    {
      "entropy": 6.058357524871826,
      "epoch": 2.7544359474184037,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004247783269416306,
      "loss": 5.6211,
      "mean_token_accuracy": 0.17545729875564575,
      "num_tokens": 53910577.0,
      "step": 27555
    },
    {
      "entropy": 6.094953632354736,
      "epoch": 2.754935772479632,
      "grad_norm": 1.125,
      "learning_rate": 0.00042475169256425466,
      "loss": 5.7369,
      "mean_token_accuracy": 0.17246570587158203,
      "num_tokens": 53919757.0,
      "step": 27560
    },
    {
      "entropy": 6.09102349281311,
      "epoch": 2.7554355975408606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004247250544191647,
      "loss": 5.606,
      "mean_token_accuracy": 0.17900809794664382,
      "num_tokens": 53928521.0,
      "step": 27565
    },
    {
      "entropy": 6.119507789611816,
      "epoch": 2.7559354226020893,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042469841250703086,
      "loss": 5.7217,
      "mean_token_accuracy": 0.17197618633508682,
      "num_tokens": 53939127.0,
      "step": 27570
    },
    {
      "entropy": 6.029885005950928,
      "epoch": 2.756435247663318,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004246717668285235,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1936142325401306,
      "num_tokens": 53948572.0,
      "step": 27575
    },
    {
      "entropy": 6.103550958633423,
      "epoch": 2.7569350727245463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042464511738431296,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18222021609544753,
      "num_tokens": 53958332.0,
      "step": 27580
    },
    {
      "entropy": 6.068691873550415,
      "epoch": 2.757434897785775,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042461846417506983,
      "loss": 5.6041,
      "mean_token_accuracy": 0.18175142854452134,
      "num_tokens": 53968771.0,
      "step": 27585
    },
    {
      "entropy": 6.042122793197632,
      "epoch": 2.7579347228470037,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004245918072014646,
      "loss": 5.6309,
      "mean_token_accuracy": 0.17849535644054412,
      "num_tokens": 53979134.0,
      "step": 27590
    },
    {
      "entropy": 6.05056095123291,
      "epoch": 2.758434547908232,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000424565146464168,
      "loss": 5.6607,
      "mean_token_accuracy": 0.17738613486289978,
      "num_tokens": 53990668.0,
      "step": 27595
    },
    {
      "entropy": 6.096655178070068,
      "epoch": 2.7589343729694606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042453848196385085,
      "loss": 5.7208,
      "mean_token_accuracy": 0.17079819291830062,
      "num_tokens": 54001559.0,
      "step": 27600
    },
    {
      "entropy": 6.029014348983765,
      "epoch": 2.7594341980306893,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042451181370118394,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18224153071641921,
      "num_tokens": 54010893.0,
      "step": 27605
    },
    {
      "entropy": 6.067518949508667,
      "epoch": 2.759934023091918,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004244851416768383,
      "loss": 5.614,
      "mean_token_accuracy": 0.17474195510149002,
      "num_tokens": 54019592.0,
      "step": 27610
    },
    {
      "entropy": 6.078827571868897,
      "epoch": 2.7604338481531463,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004244584658914849,
      "loss": 5.6848,
      "mean_token_accuracy": 0.17317474782466888,
      "num_tokens": 54029915.0,
      "step": 27615
    },
    {
      "entropy": 6.068371391296386,
      "epoch": 2.760933673214375,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042443178634579497,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17362615317106248,
      "num_tokens": 54039554.0,
      "step": 27620
    },
    {
      "entropy": 5.975620555877685,
      "epoch": 2.7614334982756037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004244051030404398,
      "loss": 5.4078,
      "mean_token_accuracy": 0.1990509510040283,
      "num_tokens": 54050184.0,
      "step": 27625
    },
    {
      "entropy": 6.007684850692749,
      "epoch": 2.761933323336832,
      "grad_norm": 1.4375,
      "learning_rate": 0.00042437841597609054,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18410078138113023,
      "num_tokens": 54058895.0,
      "step": 27630
    },
    {
      "entropy": 6.092415046691895,
      "epoch": 2.7624331483980606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004243517251534189,
      "loss": 5.6322,
      "mean_token_accuracy": 0.17747240662574768,
      "num_tokens": 54068032.0,
      "step": 27635
    },
    {
      "entropy": 6.0497386932373045,
      "epoch": 2.7629329734592893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004243250305730963,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17594524323940278,
      "num_tokens": 54076435.0,
      "step": 27640
    },
    {
      "entropy": 6.0071821212768555,
      "epoch": 2.763432798520518,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00042429833223579436,
      "loss": 5.6291,
      "mean_token_accuracy": 0.17838697433471679,
      "num_tokens": 54086274.0,
      "step": 27645
    },
    {
      "entropy": 6.057536649703979,
      "epoch": 2.7639326235817463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042427163014218473,
      "loss": 5.6337,
      "mean_token_accuracy": 0.17806814834475518,
      "num_tokens": 54095699.0,
      "step": 27650
    },
    {
      "entropy": 6.047024583816528,
      "epoch": 2.764432448642975,
      "grad_norm": 1.3125,
      "learning_rate": 0.00042424492429293933,
      "loss": 5.5627,
      "mean_token_accuracy": 0.19254744052886963,
      "num_tokens": 54106424.0,
      "step": 27655
    },
    {
      "entropy": 6.0183733940124515,
      "epoch": 2.7649322737042037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042421821468873003,
      "loss": 5.5985,
      "mean_token_accuracy": 0.17971686869859696,
      "num_tokens": 54115226.0,
      "step": 27660
    },
    {
      "entropy": 6.02657117843628,
      "epoch": 2.765432098765432,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042419150133022883,
      "loss": 5.5675,
      "mean_token_accuracy": 0.1882973998785019,
      "num_tokens": 54124018.0,
      "step": 27665
    },
    {
      "entropy": 6.0933003425598145,
      "epoch": 2.7659319238266606,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004241647842181078,
      "loss": 5.6764,
      "mean_token_accuracy": 0.17228395640850067,
      "num_tokens": 54134096.0,
      "step": 27670
    },
    {
      "entropy": 6.076525115966797,
      "epoch": 2.7664317488878893,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042413806335303935,
      "loss": 5.6908,
      "mean_token_accuracy": 0.17090590447187423,
      "num_tokens": 54143207.0,
      "step": 27675
    },
    {
      "entropy": 5.916588830947876,
      "epoch": 2.766931573949118,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042411133873569544,
      "loss": 5.458,
      "mean_token_accuracy": 0.20261725932359695,
      "num_tokens": 54153647.0,
      "step": 27680
    },
    {
      "entropy": 6.138403034210205,
      "epoch": 2.7674313990103463,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004240846103667487,
      "loss": 5.637,
      "mean_token_accuracy": 0.17530468553304673,
      "num_tokens": 54163695.0,
      "step": 27685
    },
    {
      "entropy": 6.032921934127808,
      "epoch": 2.767931224071575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042405787824687153,
      "loss": 5.5769,
      "mean_token_accuracy": 0.1817946583032608,
      "num_tokens": 54174450.0,
      "step": 27690
    },
    {
      "entropy": 5.972270059585571,
      "epoch": 2.7684310491328032,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004240311423767364,
      "loss": 5.56,
      "mean_token_accuracy": 0.18667499274015426,
      "num_tokens": 54184328.0,
      "step": 27695
    },
    {
      "entropy": 5.983116960525512,
      "epoch": 2.768930874194032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004240044027570162,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18423062711954116,
      "num_tokens": 54193385.0,
      "step": 27700
    },
    {
      "entropy": 6.012155437469483,
      "epoch": 2.7694306992552606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004239776593883835,
      "loss": 5.563,
      "mean_token_accuracy": 0.17800952792167662,
      "num_tokens": 54203320.0,
      "step": 27705
    },
    {
      "entropy": 6.064953279495239,
      "epoch": 2.7699305243164893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042395091227151124,
      "loss": 5.664,
      "mean_token_accuracy": 0.17871954143047333,
      "num_tokens": 54213395.0,
      "step": 27710
    },
    {
      "entropy": 6.091742897033692,
      "epoch": 2.770430349377718,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042392416140707233,
      "loss": 5.6492,
      "mean_token_accuracy": 0.17566876411437987,
      "num_tokens": 54222093.0,
      "step": 27715
    },
    {
      "entropy": 6.067710447311401,
      "epoch": 2.7709301744389463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004238974067957399,
      "loss": 5.6354,
      "mean_token_accuracy": 0.17781140208244323,
      "num_tokens": 54231950.0,
      "step": 27720
    },
    {
      "entropy": 5.990950393676758,
      "epoch": 2.771429999500175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000423870648438187,
      "loss": 5.5378,
      "mean_token_accuracy": 0.18509217500686645,
      "num_tokens": 54241585.0,
      "step": 27725
    },
    {
      "entropy": 6.079980182647705,
      "epoch": 2.7719298245614032,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004238438863350869,
      "loss": 5.7091,
      "mean_token_accuracy": 0.17894463539123534,
      "num_tokens": 54251525.0,
      "step": 27730
    },
    {
      "entropy": 6.020295810699463,
      "epoch": 2.772429649622632,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004238171204871129,
      "loss": 5.5454,
      "mean_token_accuracy": 0.18907224833965303,
      "num_tokens": 54260777.0,
      "step": 27735
    },
    {
      "entropy": 6.086642265319824,
      "epoch": 2.7729294746838606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042379035089493844,
      "loss": 5.6306,
      "mean_token_accuracy": 0.1771319568157196,
      "num_tokens": 54270651.0,
      "step": 27740
    },
    {
      "entropy": 5.940955114364624,
      "epoch": 2.7734292997450893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004237635775592371,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19272326678037643,
      "num_tokens": 54280241.0,
      "step": 27745
    },
    {
      "entropy": 6.042846584320069,
      "epoch": 2.773929124806318,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004237368004806823,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17742201387882234,
      "num_tokens": 54290075.0,
      "step": 27750
    },
    {
      "entropy": 6.045971059799195,
      "epoch": 2.7744289498675463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000423710019659948,
      "loss": 5.5616,
      "mean_token_accuracy": 0.18728675991296767,
      "num_tokens": 54298452.0,
      "step": 27755
    },
    {
      "entropy": 6.008562183380127,
      "epoch": 2.774928774928775,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042368323509770785,
      "loss": 5.5292,
      "mean_token_accuracy": 0.1915299877524376,
      "num_tokens": 54308516.0,
      "step": 27760
    },
    {
      "entropy": 5.981932544708252,
      "epoch": 2.7754285999900032,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004236564467946357,
      "loss": 5.5295,
      "mean_token_accuracy": 0.1822269603610039,
      "num_tokens": 54317921.0,
      "step": 27765
    },
    {
      "entropy": 5.942964553833008,
      "epoch": 2.775928425051232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004236296547514056,
      "loss": 5.5092,
      "mean_token_accuracy": 0.18748657554388046,
      "num_tokens": 54327558.0,
      "step": 27770
    },
    {
      "entropy": 6.034267473220825,
      "epoch": 2.7764282501124606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004236028589686917,
      "loss": 5.6581,
      "mean_token_accuracy": 0.18286047875881195,
      "num_tokens": 54337638.0,
      "step": 27775
    },
    {
      "entropy": 6.019650506973266,
      "epoch": 2.7769280751736893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042357605944716816,
      "loss": 5.5461,
      "mean_token_accuracy": 0.1872737392783165,
      "num_tokens": 54345607.0,
      "step": 27780
    },
    {
      "entropy": 5.940119504928589,
      "epoch": 2.777427900234918,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042354925618750904,
      "loss": 5.4896,
      "mean_token_accuracy": 0.1905573919415474,
      "num_tokens": 54355230.0,
      "step": 27785
    },
    {
      "entropy": 6.049480056762695,
      "epoch": 2.7779277252961463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042352244919038895,
      "loss": 5.6767,
      "mean_token_accuracy": 0.17129889279603958,
      "num_tokens": 54365449.0,
      "step": 27790
    },
    {
      "entropy": 6.05459508895874,
      "epoch": 2.778427550357375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042349563845648226,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1819275364279747,
      "num_tokens": 54375822.0,
      "step": 27795
    },
    {
      "entropy": 6.092546844482422,
      "epoch": 2.7789273754186032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004234688239864634,
      "loss": 5.6942,
      "mean_token_accuracy": 0.17197567969560623,
      "num_tokens": 54385365.0,
      "step": 27800
    },
    {
      "entropy": 6.029508495330811,
      "epoch": 2.779427200479832,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004234420057810073,
      "loss": 5.5508,
      "mean_token_accuracy": 0.18579584658145903,
      "num_tokens": 54394466.0,
      "step": 27805
    },
    {
      "entropy": 6.01148533821106,
      "epoch": 2.7799270255410606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042341518384078847,
      "loss": 5.5593,
      "mean_token_accuracy": 0.1874656245112419,
      "num_tokens": 54404011.0,
      "step": 27810
    },
    {
      "entropy": 6.044650650024414,
      "epoch": 2.7804268506022893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004233883581664818,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18371761441230774,
      "num_tokens": 54413527.0,
      "step": 27815
    },
    {
      "entropy": 6.049122476577759,
      "epoch": 2.780926675663518,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004233615287587622,
      "loss": 5.6465,
      "mean_token_accuracy": 0.18432531803846358,
      "num_tokens": 54423301.0,
      "step": 27820
    },
    {
      "entropy": 6.072556018829346,
      "epoch": 2.7814265007247463,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042333469561830475,
      "loss": 5.6408,
      "mean_token_accuracy": 0.17596339136362077,
      "num_tokens": 54433132.0,
      "step": 27825
    },
    {
      "entropy": 6.061946201324463,
      "epoch": 2.781926325785975,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004233078587457845,
      "loss": 5.6556,
      "mean_token_accuracy": 0.1778102844953537,
      "num_tokens": 54442926.0,
      "step": 27830
    },
    {
      "entropy": 6.167185544967651,
      "epoch": 2.7824261508472032,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004232810181418767,
      "loss": 5.6639,
      "mean_token_accuracy": 0.16979291290044785,
      "num_tokens": 54452325.0,
      "step": 27835
    },
    {
      "entropy": 6.011190176010132,
      "epoch": 2.782925975908432,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004232541738072566,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18137818127870559,
      "num_tokens": 54462340.0,
      "step": 27840
    },
    {
      "entropy": 5.974647760391235,
      "epoch": 2.7834258009696606,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042322732574259976,
      "loss": 5.4786,
      "mean_token_accuracy": 0.1969167098402977,
      "num_tokens": 54471752.0,
      "step": 27845
    },
    {
      "entropy": 6.077604532241821,
      "epoch": 2.7839256260308893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042320047394858143,
      "loss": 5.6756,
      "mean_token_accuracy": 0.17790543735027314,
      "num_tokens": 54482712.0,
      "step": 27850
    },
    {
      "entropy": 6.089851331710816,
      "epoch": 2.784425451092118,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004231736184258773,
      "loss": 5.6014,
      "mean_token_accuracy": 0.18475119769573212,
      "num_tokens": 54492804.0,
      "step": 27855
    },
    {
      "entropy": 6.079278087615966,
      "epoch": 2.7849252761533463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004231467591751631,
      "loss": 5.6741,
      "mean_token_accuracy": 0.17565094083547592,
      "num_tokens": 54502869.0,
      "step": 27860
    },
    {
      "entropy": 6.06396803855896,
      "epoch": 2.785425101214575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004231198961971145,
      "loss": 5.6041,
      "mean_token_accuracy": 0.18312956839799882,
      "num_tokens": 54513089.0,
      "step": 27865
    },
    {
      "entropy": 6.064765310287475,
      "epoch": 2.7859249262758032,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004230930294924075,
      "loss": 5.6737,
      "mean_token_accuracy": 0.17587975561618804,
      "num_tokens": 54524312.0,
      "step": 27870
    },
    {
      "entropy": 6.017914342880249,
      "epoch": 2.786424751337032,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042306615906171794,
      "loss": 5.6395,
      "mean_token_accuracy": 0.17872171700000763,
      "num_tokens": 54534258.0,
      "step": 27875
    },
    {
      "entropy": 6.113019895553589,
      "epoch": 2.7869245763982606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000423039284905722,
      "loss": 5.6078,
      "mean_token_accuracy": 0.17804242819547653,
      "num_tokens": 54544411.0,
      "step": 27880
    },
    {
      "entropy": 6.096106719970703,
      "epoch": 2.7874244014594893,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042301240702509576,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18549129962921143,
      "num_tokens": 54553253.0,
      "step": 27885
    },
    {
      "entropy": 6.003637790679932,
      "epoch": 2.7879242265207176,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004229855254205154,
      "loss": 5.6001,
      "mean_token_accuracy": 0.17542673647403717,
      "num_tokens": 54563095.0,
      "step": 27890
    },
    {
      "entropy": 6.001569414138794,
      "epoch": 2.7884240515819463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042295864009265723,
      "loss": 5.5555,
      "mean_token_accuracy": 0.18266140222549437,
      "num_tokens": 54573969.0,
      "step": 27895
    },
    {
      "entropy": 6.112708997726441,
      "epoch": 2.788923876643175,
      "grad_norm": 1.125,
      "learning_rate": 0.00042293175104219776,
      "loss": 5.6665,
      "mean_token_accuracy": 0.16909576952457428,
      "num_tokens": 54583864.0,
      "step": 27900
    },
    {
      "entropy": 6.134289073944092,
      "epoch": 2.7894237017044032,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042290485826981355,
      "loss": 5.6844,
      "mean_token_accuracy": 0.17379383742809296,
      "num_tokens": 54594463.0,
      "step": 27905
    },
    {
      "entropy": 6.0349977016448975,
      "epoch": 2.789923526765632,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042287796177618115,
      "loss": 5.6152,
      "mean_token_accuracy": 0.1761782869696617,
      "num_tokens": 54603940.0,
      "step": 27910
    },
    {
      "entropy": 5.9830461025238035,
      "epoch": 2.7904233518268606,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042285106156197727,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18384443670511247,
      "num_tokens": 54613936.0,
      "step": 27915
    },
    {
      "entropy": 6.019849824905395,
      "epoch": 2.7909231768880893,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004228241576278787,
      "loss": 5.5777,
      "mean_token_accuracy": 0.18337658196687698,
      "num_tokens": 54622584.0,
      "step": 27920
    },
    {
      "entropy": 6.0578478336334225,
      "epoch": 2.7914230019493176,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042279724997456246,
      "loss": 5.6738,
      "mean_token_accuracy": 0.17813718914985657,
      "num_tokens": 54632899.0,
      "step": 27925
    },
    {
      "entropy": 6.178372955322265,
      "epoch": 2.7919228270105463,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042277033860270544,
      "loss": 5.7925,
      "mean_token_accuracy": 0.17119476050138474,
      "num_tokens": 54644025.0,
      "step": 27930
    },
    {
      "entropy": 6.102255535125733,
      "epoch": 2.792422652071775,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042274342351298467,
      "loss": 5.5883,
      "mean_token_accuracy": 0.1846216633915901,
      "num_tokens": 54652916.0,
      "step": 27935
    },
    {
      "entropy": 6.132054567337036,
      "epoch": 2.7929224771330032,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004227165047060773,
      "loss": 5.7568,
      "mean_token_accuracy": 0.16864680945873262,
      "num_tokens": 54662888.0,
      "step": 27940
    },
    {
      "entropy": 6.07523775100708,
      "epoch": 2.793422302194232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004226895821826608,
      "loss": 5.6183,
      "mean_token_accuracy": 0.18059612363576888,
      "num_tokens": 54672597.0,
      "step": 27945
    },
    {
      "entropy": 6.037691259384156,
      "epoch": 2.7939221272554606,
      "grad_norm": 1.25,
      "learning_rate": 0.0004226626559434124,
      "loss": 5.6138,
      "mean_token_accuracy": 0.18663717806339264,
      "num_tokens": 54683284.0,
      "step": 27950
    },
    {
      "entropy": 6.057548332214355,
      "epoch": 2.7944219523166893,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042263572598900956,
      "loss": 5.6633,
      "mean_token_accuracy": 0.17831076979637145,
      "num_tokens": 54693207.0,
      "step": 27955
    },
    {
      "entropy": 6.014265012741089,
      "epoch": 2.7949217773779176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042260879232012985,
      "loss": 5.5375,
      "mean_token_accuracy": 0.19002314358949662,
      "num_tokens": 54703122.0,
      "step": 27960
    },
    {
      "entropy": 5.975918531417847,
      "epoch": 2.7954216024391463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042258185493745086,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18414102345705033,
      "num_tokens": 54713034.0,
      "step": 27965
    },
    {
      "entropy": 6.092884826660156,
      "epoch": 2.795921427500375,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004225549138416504,
      "loss": 5.7882,
      "mean_token_accuracy": 0.16601140797138214,
      "num_tokens": 54722342.0,
      "step": 27970
    },
    {
      "entropy": 6.109921407699585,
      "epoch": 2.7964212525616032,
      "grad_norm": 1.125,
      "learning_rate": 0.0004225279690334063,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17193118780851363,
      "num_tokens": 54732346.0,
      "step": 27975
    },
    {
      "entropy": 6.087307262420654,
      "epoch": 2.796921077622832,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00042250102051339647,
      "loss": 5.6441,
      "mean_token_accuracy": 0.17260152250528335,
      "num_tokens": 54742859.0,
      "step": 27980
    },
    {
      "entropy": 6.037876129150391,
      "epoch": 2.7974209026840606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042247406828229897,
      "loss": 5.519,
      "mean_token_accuracy": 0.18853932172060012,
      "num_tokens": 54752782.0,
      "step": 27985
    },
    {
      "entropy": 6.040089130401611,
      "epoch": 2.7979207277452893,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004224471123407917,
      "loss": 5.6148,
      "mean_token_accuracy": 0.17979101091623306,
      "num_tokens": 54762415.0,
      "step": 27990
    },
    {
      "entropy": 6.042787885665893,
      "epoch": 2.7984205528065176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004224201526895531,
      "loss": 5.5669,
      "mean_token_accuracy": 0.18606301099061967,
      "num_tokens": 54773709.0,
      "step": 27995
    },
    {
      "entropy": 5.995025682449341,
      "epoch": 2.7989203778677463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042239318932926144,
      "loss": 5.5418,
      "mean_token_accuracy": 0.19408591240644454,
      "num_tokens": 54782895.0,
      "step": 28000
    },
    {
      "entropy": 6.011727237701416,
      "epoch": 2.799420202928975,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000422366222260595,
      "loss": 5.6019,
      "mean_token_accuracy": 0.18287300169467927,
      "num_tokens": 54793032.0,
      "step": 28005
    },
    {
      "entropy": 6.074238729476929,
      "epoch": 2.7999200279902032,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00042233925148423235,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17239965200424195,
      "num_tokens": 54803325.0,
      "step": 28010
    },
    {
      "entropy": 6.086663484573364,
      "epoch": 2.800419853051432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004223122770008521,
      "loss": 5.6684,
      "mean_token_accuracy": 0.1767553448677063,
      "num_tokens": 54813630.0,
      "step": 28015
    },
    {
      "entropy": 6.100894546508789,
      "epoch": 2.8009196781126606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042228529881113285,
      "loss": 5.6195,
      "mean_token_accuracy": 0.17632038295269012,
      "num_tokens": 54822916.0,
      "step": 28020
    },
    {
      "entropy": 5.982534551620484,
      "epoch": 2.8014195031738893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042225831691575336,
      "loss": 5.536,
      "mean_token_accuracy": 0.18883873969316484,
      "num_tokens": 54834650.0,
      "step": 28025
    },
    {
      "entropy": 6.020419788360596,
      "epoch": 2.8019193282351176,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004222313313153925,
      "loss": 5.6031,
      "mean_token_accuracy": 0.18424385786056519,
      "num_tokens": 54843347.0,
      "step": 28030
    },
    {
      "entropy": 5.870360279083252,
      "epoch": 2.8024191532963463,
      "grad_norm": 1.765625,
      "learning_rate": 0.00042220434201072917,
      "loss": 5.4848,
      "mean_token_accuracy": 0.1898971512913704,
      "num_tokens": 54853625.0,
      "step": 28035
    },
    {
      "entropy": 6.153947496414185,
      "epoch": 2.802918978357575,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004221773490024426,
      "loss": 5.7494,
      "mean_token_accuracy": 0.1776506006717682,
      "num_tokens": 54863796.0,
      "step": 28040
    },
    {
      "entropy": 6.16171498298645,
      "epoch": 2.8034188034188032,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004221503522912117,
      "loss": 5.6167,
      "mean_token_accuracy": 0.17762255817651748,
      "num_tokens": 54872766.0,
      "step": 28045
    },
    {
      "entropy": 5.971778297424317,
      "epoch": 2.803918628480032,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042212335187771587,
      "loss": 5.5228,
      "mean_token_accuracy": 0.19001631438732147,
      "num_tokens": 54881234.0,
      "step": 28050
    },
    {
      "entropy": 6.014069080352783,
      "epoch": 2.8044184535412606,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042209634776263434,
      "loss": 5.6113,
      "mean_token_accuracy": 0.18057523369789125,
      "num_tokens": 54890830.0,
      "step": 28055
    },
    {
      "entropy": 6.108667802810669,
      "epoch": 2.8049182786024893,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004220693399466466,
      "loss": 5.7266,
      "mean_token_accuracy": 0.1705661654472351,
      "num_tokens": 54900496.0,
      "step": 28060
    },
    {
      "entropy": 6.116151809692383,
      "epoch": 2.8054181036637176,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042204232843043207,
      "loss": 5.6706,
      "mean_token_accuracy": 0.17208039313554763,
      "num_tokens": 54911190.0,
      "step": 28065
    },
    {
      "entropy": 6.024494743347168,
      "epoch": 2.8059179287249463,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042201531321467027,
      "loss": 5.5462,
      "mean_token_accuracy": 0.18462601006031037,
      "num_tokens": 54919817.0,
      "step": 28070
    },
    {
      "entropy": 6.001372385025024,
      "epoch": 2.806417753786175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042198829430004114,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1764098137617111,
      "num_tokens": 54930549.0,
      "step": 28075
    },
    {
      "entropy": 5.990999412536621,
      "epoch": 2.8069175788474032,
      "grad_norm": 1.125,
      "learning_rate": 0.00042196127168722436,
      "loss": 5.6321,
      "mean_token_accuracy": 0.18827691227197646,
      "num_tokens": 54940016.0,
      "step": 28080
    },
    {
      "entropy": 5.984377193450928,
      "epoch": 2.807417403908632,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004219342453768998,
      "loss": 5.6612,
      "mean_token_accuracy": 0.17994602471590043,
      "num_tokens": 54949664.0,
      "step": 28085
    },
    {
      "entropy": 6.014549493789673,
      "epoch": 2.8079172289698606,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004219072153697474,
      "loss": 5.6729,
      "mean_token_accuracy": 0.1753360465168953,
      "num_tokens": 54959931.0,
      "step": 28090
    },
    {
      "entropy": 6.004751110076905,
      "epoch": 2.8084170540310893,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004218801816664473,
      "loss": 5.4806,
      "mean_token_accuracy": 0.1917170464992523,
      "num_tokens": 54969902.0,
      "step": 28095
    },
    {
      "entropy": 6.058023595809937,
      "epoch": 2.8089168790923176,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004218531442676796,
      "loss": 5.6073,
      "mean_token_accuracy": 0.1770539402961731,
      "num_tokens": 54978865.0,
      "step": 28100
    },
    {
      "entropy": 6.032883214950561,
      "epoch": 2.8094167041535463,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004218261031741246,
      "loss": 5.5488,
      "mean_token_accuracy": 0.18356322795152663,
      "num_tokens": 54987895.0,
      "step": 28105
    },
    {
      "entropy": 6.035139083862305,
      "epoch": 2.809916529214775,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042179905838646254,
      "loss": 5.5852,
      "mean_token_accuracy": 0.1810919776558876,
      "num_tokens": 54997703.0,
      "step": 28110
    },
    {
      "entropy": 5.96881103515625,
      "epoch": 2.8104163542760032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000421772009905374,
      "loss": 5.5197,
      "mean_token_accuracy": 0.18808430135250093,
      "num_tokens": 55007591.0,
      "step": 28115
    },
    {
      "entropy": 5.970992136001587,
      "epoch": 2.810916179337232,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004217449577315395,
      "loss": 5.6065,
      "mean_token_accuracy": 0.18182790279388428,
      "num_tokens": 55017784.0,
      "step": 28120
    },
    {
      "entropy": 5.975998497009277,
      "epoch": 2.8114160043984606,
      "grad_norm": 1.25,
      "learning_rate": 0.0004217179018656396,
      "loss": 5.5367,
      "mean_token_accuracy": 0.18907858580350875,
      "num_tokens": 55027336.0,
      "step": 28125
    },
    {
      "entropy": 6.042586469650269,
      "epoch": 2.8119158294596893,
      "grad_norm": 1.078125,
      "learning_rate": 0.000421690842308355,
      "loss": 5.6075,
      "mean_token_accuracy": 0.17773516476154327,
      "num_tokens": 55037102.0,
      "step": 28130
    },
    {
      "entropy": 6.011901044845581,
      "epoch": 2.8124156545209176,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004216637790603667,
      "loss": 5.688,
      "mean_token_accuracy": 0.17322746962308883,
      "num_tokens": 55046528.0,
      "step": 28135
    },
    {
      "entropy": 6.026383113861084,
      "epoch": 2.8129154795821463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004216367121223553,
      "loss": 5.6372,
      "mean_token_accuracy": 0.1791699171066284,
      "num_tokens": 55057175.0,
      "step": 28140
    },
    {
      "entropy": 6.074655246734619,
      "epoch": 2.813415304643375,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000421609641495002,
      "loss": 5.6928,
      "mean_token_accuracy": 0.1699300616979599,
      "num_tokens": 55067345.0,
      "step": 28145
    },
    {
      "entropy": 6.053578233718872,
      "epoch": 2.8139151297046032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042158256717898785,
      "loss": 5.5384,
      "mean_token_accuracy": 0.18642511665821077,
      "num_tokens": 55076728.0,
      "step": 28150
    },
    {
      "entropy": 6.018805646896363,
      "epoch": 2.814414954765832,
      "grad_norm": 1.078125,
      "learning_rate": 0.000421555489174994,
      "loss": 5.6398,
      "mean_token_accuracy": 0.1840595319867134,
      "num_tokens": 55086262.0,
      "step": 28155
    },
    {
      "entropy": 5.9397886276245115,
      "epoch": 2.8149147798270606,
      "grad_norm": 1.25,
      "learning_rate": 0.00042152840748370184,
      "loss": 5.467,
      "mean_token_accuracy": 0.19571727514266968,
      "num_tokens": 55095096.0,
      "step": 28160
    },
    {
      "entropy": 5.9701636791229244,
      "epoch": 2.8154146048882893,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004215013221057926,
      "loss": 5.5524,
      "mean_token_accuracy": 0.18813118636608123,
      "num_tokens": 55104961.0,
      "step": 28165
    },
    {
      "entropy": 6.045828008651734,
      "epoch": 2.8159144299495176,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004214742330419478,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18357456028461455,
      "num_tokens": 55114069.0,
      "step": 28170
    },
    {
      "entropy": 5.976323938369751,
      "epoch": 2.8164142550107463,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042144714029284893,
      "loss": 5.553,
      "mean_token_accuracy": 0.1865283191204071,
      "num_tokens": 55124775.0,
      "step": 28175
    },
    {
      "entropy": 6.127162075042724,
      "epoch": 2.816914080071975,
      "grad_norm": 1.125,
      "learning_rate": 0.00042142004385917783,
      "loss": 5.7342,
      "mean_token_accuracy": 0.17096866369247438,
      "num_tokens": 55134245.0,
      "step": 28180
    },
    {
      "entropy": 6.075980520248413,
      "epoch": 2.8174139051332032,
      "grad_norm": 1.359375,
      "learning_rate": 0.000421392943741616,
      "loss": 5.6743,
      "mean_token_accuracy": 0.17672564834356308,
      "num_tokens": 55145770.0,
      "step": 28185
    },
    {
      "entropy": 6.034541654586792,
      "epoch": 2.817913730194432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004213658399408454,
      "loss": 5.5865,
      "mean_token_accuracy": 0.18042981922626494,
      "num_tokens": 55156618.0,
      "step": 28190
    },
    {
      "entropy": 6.012775802612305,
      "epoch": 2.8184135552556606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004213387324575479,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18218191862106323,
      "num_tokens": 55166323.0,
      "step": 28195
    },
    {
      "entropy": 6.1087809085845945,
      "epoch": 2.8189133803168893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004213116212924056,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16922108381986617,
      "num_tokens": 55176037.0,
      "step": 28200
    },
    {
      "entropy": 6.123135566711426,
      "epoch": 2.8194132053781176,
      "grad_norm": 1.125,
      "learning_rate": 0.00042128450644610046,
      "loss": 5.6329,
      "mean_token_accuracy": 0.1786355569958687,
      "num_tokens": 55185861.0,
      "step": 28205
    },
    {
      "entropy": 6.04125075340271,
      "epoch": 2.8199130304393463,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042125738791931486,
      "loss": 5.6324,
      "mean_token_accuracy": 0.18046713769435882,
      "num_tokens": 55196126.0,
      "step": 28210
    },
    {
      "entropy": 6.015937566757202,
      "epoch": 2.8204128555005745,
      "grad_norm": 1.203125,
      "learning_rate": 0.000421230265712731,
      "loss": 5.5463,
      "mean_token_accuracy": 0.18805399537086487,
      "num_tokens": 55205108.0,
      "step": 28215
    },
    {
      "entropy": 6.027273321151734,
      "epoch": 2.820912680561803,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004212031398270313,
      "loss": 5.6547,
      "mean_token_accuracy": 0.17612036615610122,
      "num_tokens": 55215351.0,
      "step": 28220
    },
    {
      "entropy": 6.034762191772461,
      "epoch": 2.821412505623032,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042117601026289827,
      "loss": 5.5479,
      "mean_token_accuracy": 0.18741717785596848,
      "num_tokens": 55224594.0,
      "step": 28225
    },
    {
      "entropy": 6.073874521255493,
      "epoch": 2.8219123306842606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042114887702101434,
      "loss": 5.6146,
      "mean_token_accuracy": 0.17503816336393357,
      "num_tokens": 55234600.0,
      "step": 28230
    },
    {
      "entropy": 6.0536786079406735,
      "epoch": 2.8224121557454893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042112174010206226,
      "loss": 5.5359,
      "mean_token_accuracy": 0.18511317819356918,
      "num_tokens": 55243377.0,
      "step": 28235
    },
    {
      "entropy": 6.0219141960144045,
      "epoch": 2.8229119808067176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042109459950672487,
      "loss": 5.5693,
      "mean_token_accuracy": 0.17906364649534226,
      "num_tokens": 55252245.0,
      "step": 28240
    },
    {
      "entropy": 6.057013845443725,
      "epoch": 2.8234118058679463,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042106745523568494,
      "loss": 5.6419,
      "mean_token_accuracy": 0.17151365876197816,
      "num_tokens": 55261784.0,
      "step": 28245
    },
    {
      "entropy": 6.041095304489136,
      "epoch": 2.8239116309291745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042104030728962535,
      "loss": 5.6044,
      "mean_token_accuracy": 0.18500850349664688,
      "num_tokens": 55272151.0,
      "step": 28250
    },
    {
      "entropy": 6.091882276535034,
      "epoch": 2.824411455990403,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042101315566922927,
      "loss": 5.6056,
      "mean_token_accuracy": 0.1756083145737648,
      "num_tokens": 55281800.0,
      "step": 28255
    },
    {
      "entropy": 6.060226678848267,
      "epoch": 2.824911281051632,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042098600037517985,
      "loss": 5.6177,
      "mean_token_accuracy": 0.17673148065805436,
      "num_tokens": 55291297.0,
      "step": 28260
    },
    {
      "entropy": 5.960746002197266,
      "epoch": 2.8254111061128606,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004209588414081601,
      "loss": 5.5583,
      "mean_token_accuracy": 0.19282777458429337,
      "num_tokens": 55300265.0,
      "step": 28265
    },
    {
      "entropy": 6.013680791854858,
      "epoch": 2.8259109311740893,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042093167876885345,
      "loss": 5.5542,
      "mean_token_accuracy": 0.18394701927900314,
      "num_tokens": 55310802.0,
      "step": 28270
    },
    {
      "entropy": 6.082170104980468,
      "epoch": 2.8264107562353176,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004209045124579434,
      "loss": 5.6408,
      "mean_token_accuracy": 0.18487460911273956,
      "num_tokens": 55320871.0,
      "step": 28275
    },
    {
      "entropy": 6.084898138046265,
      "epoch": 2.8269105812965463,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042087734247611333,
      "loss": 5.6804,
      "mean_token_accuracy": 0.16653202772140502,
      "num_tokens": 55330419.0,
      "step": 28280
    },
    {
      "entropy": 6.087577533721924,
      "epoch": 2.8274104063577745,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004208501688240468,
      "loss": 5.6712,
      "mean_token_accuracy": 0.17979590296745301,
      "num_tokens": 55339930.0,
      "step": 28285
    },
    {
      "entropy": 6.047424268722534,
      "epoch": 2.827910231419003,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00042082299150242764,
      "loss": 5.6261,
      "mean_token_accuracy": 0.1795596659183502,
      "num_tokens": 55349871.0,
      "step": 28290
    },
    {
      "entropy": 6.086045503616333,
      "epoch": 2.828410056480232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004207958105119395,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17341867238283157,
      "num_tokens": 55360046.0,
      "step": 28295
    },
    {
      "entropy": 6.0423413753509525,
      "epoch": 2.8289098815414606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042076862585326626,
      "loss": 5.5445,
      "mean_token_accuracy": 0.18603315502405166,
      "num_tokens": 55370173.0,
      "step": 28300
    },
    {
      "entropy": 6.017988014221191,
      "epoch": 2.8294097066026893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042074143752709195,
      "loss": 5.6149,
      "mean_token_accuracy": 0.18256779313087462,
      "num_tokens": 55380947.0,
      "step": 28305
    },
    {
      "entropy": 6.117248487472534,
      "epoch": 2.8299095316639176,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004207142455341005,
      "loss": 5.6835,
      "mean_token_accuracy": 0.17596840411424636,
      "num_tokens": 55390317.0,
      "step": 28310
    },
    {
      "entropy": 6.034973049163819,
      "epoch": 2.8304093567251463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042068704987497627,
      "loss": 5.5124,
      "mean_token_accuracy": 0.18597981631755828,
      "num_tokens": 55400729.0,
      "step": 28315
    },
    {
      "entropy": 6.00244402885437,
      "epoch": 2.8309091817863745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042065985055040326,
      "loss": 5.5822,
      "mean_token_accuracy": 0.18212665766477584,
      "num_tokens": 55409770.0,
      "step": 28320
    },
    {
      "entropy": 6.017697525024414,
      "epoch": 2.831409006847603,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004206326475610658,
      "loss": 5.6506,
      "mean_token_accuracy": 0.1786036565899849,
      "num_tokens": 55419521.0,
      "step": 28325
    },
    {
      "entropy": 6.105055904388427,
      "epoch": 2.831908831908832,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042060544090764856,
      "loss": 5.7082,
      "mean_token_accuracy": 0.1795203745365143,
      "num_tokens": 55429496.0,
      "step": 28330
    },
    {
      "entropy": 6.025254106521606,
      "epoch": 2.8324086569700606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004205782305908358,
      "loss": 5.5997,
      "mean_token_accuracy": 0.17708902806043625,
      "num_tokens": 55438465.0,
      "step": 28335
    },
    {
      "entropy": 6.0270589828491214,
      "epoch": 2.8329084820312893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004205510166113122,
      "loss": 5.6075,
      "mean_token_accuracy": 0.18580835908651352,
      "num_tokens": 55448061.0,
      "step": 28340
    },
    {
      "entropy": 5.9125580310821535,
      "epoch": 2.8334083070925176,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042052379896976254,
      "loss": 5.5106,
      "mean_token_accuracy": 0.19075769782066346,
      "num_tokens": 55456363.0,
      "step": 28345
    },
    {
      "entropy": 5.972819757461548,
      "epoch": 2.8339081321537463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004204965776668716,
      "loss": 5.5552,
      "mean_token_accuracy": 0.18397632539272307,
      "num_tokens": 55466507.0,
      "step": 28350
    },
    {
      "entropy": 6.134210014343262,
      "epoch": 2.8344079572149745,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004204693527033241,
      "loss": 5.6434,
      "mean_token_accuracy": 0.180515293776989,
      "num_tokens": 55476723.0,
      "step": 28355
    },
    {
      "entropy": 6.06043438911438,
      "epoch": 2.834907782276203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042044212407980517,
      "loss": 5.6348,
      "mean_token_accuracy": 0.17950020134449005,
      "num_tokens": 55486227.0,
      "step": 28360
    },
    {
      "entropy": 6.060025596618653,
      "epoch": 2.835407607337432,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042041489179699983,
      "loss": 5.6278,
      "mean_token_accuracy": 0.18115690648555755,
      "num_tokens": 55495899.0,
      "step": 28365
    },
    {
      "entropy": 5.947348594665527,
      "epoch": 2.8359074323986606,
      "grad_norm": 1.96875,
      "learning_rate": 0.0004203876558555932,
      "loss": 5.6152,
      "mean_token_accuracy": 0.18923051059246063,
      "num_tokens": 55506164.0,
      "step": 28370
    },
    {
      "entropy": 6.05122971534729,
      "epoch": 2.8364072574598893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042036041625627055,
      "loss": 5.6459,
      "mean_token_accuracy": 0.17917312532663346,
      "num_tokens": 55515159.0,
      "step": 28375
    },
    {
      "entropy": 6.052556848526001,
      "epoch": 2.8369070825211176,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042033317299971726,
      "loss": 5.6574,
      "mean_token_accuracy": 0.17982162535190582,
      "num_tokens": 55524728.0,
      "step": 28380
    },
    {
      "entropy": 6.039006328582763,
      "epoch": 2.8374069075823463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004203059260866187,
      "loss": 5.6383,
      "mean_token_accuracy": 0.178266179561615,
      "num_tokens": 55534555.0,
      "step": 28385
    },
    {
      "entropy": 6.086304330825806,
      "epoch": 2.8379067326435745,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004202786755176604,
      "loss": 5.6716,
      "mean_token_accuracy": 0.18216136395931243,
      "num_tokens": 55544156.0,
      "step": 28390
    },
    {
      "entropy": 6.061879444122314,
      "epoch": 2.838406557704803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004202514212935281,
      "loss": 5.6167,
      "mean_token_accuracy": 0.18740918934345246,
      "num_tokens": 55553912.0,
      "step": 28395
    },
    {
      "entropy": 6.041742992401123,
      "epoch": 2.838906382766032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004202241634149074,
      "loss": 5.6004,
      "mean_token_accuracy": 0.18133955746889113,
      "num_tokens": 55564052.0,
      "step": 28400
    },
    {
      "entropy": 5.956309700012207,
      "epoch": 2.8394062078272606,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00042019690188248407,
      "loss": 5.391,
      "mean_token_accuracy": 0.21219144761562347,
      "num_tokens": 55574098.0,
      "step": 28405
    },
    {
      "entropy": 5.998479270935059,
      "epoch": 2.839906032888489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004201696366969441,
      "loss": 5.5158,
      "mean_token_accuracy": 0.18834036588668823,
      "num_tokens": 55584429.0,
      "step": 28410
    },
    {
      "entropy": 6.1165729522705075,
      "epoch": 2.8404058579497176,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004201423678589734,
      "loss": 5.6534,
      "mean_token_accuracy": 0.18058861941099166,
      "num_tokens": 55595990.0,
      "step": 28415
    },
    {
      "entropy": 6.078712606430054,
      "epoch": 2.8409056830109463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000420115095369258,
      "loss": 5.6716,
      "mean_token_accuracy": 0.18238510936498642,
      "num_tokens": 55606208.0,
      "step": 28420
    },
    {
      "entropy": 5.9988658905029295,
      "epoch": 2.8414055080721745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004200878192284843,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18814100921154023,
      "num_tokens": 55615790.0,
      "step": 28425
    },
    {
      "entropy": 6.076408958435058,
      "epoch": 2.841905333133403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004200605394373383,
      "loss": 5.6254,
      "mean_token_accuracy": 0.18294065743684768,
      "num_tokens": 55625914.0,
      "step": 28430
    },
    {
      "entropy": 6.01550784111023,
      "epoch": 2.842405158194632,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004200332559965065,
      "loss": 5.5666,
      "mean_token_accuracy": 0.18991390317678453,
      "num_tokens": 55636019.0,
      "step": 28435
    },
    {
      "entropy": 6.0366497993469235,
      "epoch": 2.8429049832558606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004200059689066752,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17612652629613876,
      "num_tokens": 55645713.0,
      "step": 28440
    },
    {
      "entropy": 6.105549716949463,
      "epoch": 2.843404808317089,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041997867816853115,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17322741597890853,
      "num_tokens": 55655773.0,
      "step": 28445
    },
    {
      "entropy": 6.047887516021729,
      "epoch": 2.8439046333783176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041995138378276084,
      "loss": 5.5905,
      "mean_token_accuracy": 0.18562873154878617,
      "num_tokens": 55664647.0,
      "step": 28450
    },
    {
      "entropy": 6.123758125305176,
      "epoch": 2.8444044584395463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041992408575005095,
      "loss": 5.7197,
      "mean_token_accuracy": 0.17188856154680252,
      "num_tokens": 55674610.0,
      "step": 28455
    },
    {
      "entropy": 6.0310587882995605,
      "epoch": 2.8449042835007745,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004198967840710884,
      "loss": 5.5985,
      "mean_token_accuracy": 0.1789831280708313,
      "num_tokens": 55684260.0,
      "step": 28460
    },
    {
      "entropy": 5.992735147476196,
      "epoch": 2.845404108562003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041986947874656,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18362425416707992,
      "num_tokens": 55693546.0,
      "step": 28465
    },
    {
      "entropy": 6.030821847915649,
      "epoch": 2.845903933623232,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004198421697771529,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18688427954912185,
      "num_tokens": 55702824.0,
      "step": 28470
    },
    {
      "entropy": 6.073639726638794,
      "epoch": 2.8464037586844606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000419814857163554,
      "loss": 5.6592,
      "mean_token_accuracy": 0.17300720661878585,
      "num_tokens": 55712522.0,
      "step": 28475
    },
    {
      "entropy": 6.010354804992676,
      "epoch": 2.846903583745689,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004197875409064506,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18150143772363664,
      "num_tokens": 55723093.0,
      "step": 28480
    },
    {
      "entropy": 5.841633987426758,
      "epoch": 2.8474034088069176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004197602210065299,
      "loss": 5.3896,
      "mean_token_accuracy": 0.20670489817857743,
      "num_tokens": 55732583.0,
      "step": 28485
    },
    {
      "entropy": 6.067646408081055,
      "epoch": 2.8479032338681463,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004197328974644793,
      "loss": 5.6693,
      "mean_token_accuracy": 0.1769425466656685,
      "num_tokens": 55742777.0,
      "step": 28490
    },
    {
      "entropy": 6.110142564773559,
      "epoch": 2.8484030589293745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004197055702809862,
      "loss": 5.6313,
      "mean_token_accuracy": 0.1759537249803543,
      "num_tokens": 55751587.0,
      "step": 28495
    },
    {
      "entropy": 6.051973485946656,
      "epoch": 2.848902883990603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004196782394567382,
      "loss": 5.6967,
      "mean_token_accuracy": 0.17215802669525146,
      "num_tokens": 55761959.0,
      "step": 28500
    },
    {
      "entropy": 6.034333562850952,
      "epoch": 2.849402709051832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000419650904992423,
      "loss": 5.6236,
      "mean_token_accuracy": 0.18205676972866058,
      "num_tokens": 55771055.0,
      "step": 28505
    },
    {
      "entropy": 6.119200325012207,
      "epoch": 2.8499025341130606,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041962356688872817,
      "loss": 5.6293,
      "mean_token_accuracy": 0.18178342431783676,
      "num_tokens": 55780625.0,
      "step": 28510
    },
    {
      "entropy": 6.070038366317749,
      "epoch": 2.850402359174289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004195962251463417,
      "loss": 5.6196,
      "mean_token_accuracy": 0.18585037738084792,
      "num_tokens": 55790290.0,
      "step": 28515
    },
    {
      "entropy": 6.069152164459228,
      "epoch": 2.8509021842355176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004195688797659513,
      "loss": 5.6604,
      "mean_token_accuracy": 0.1793117865920067,
      "num_tokens": 55800407.0,
      "step": 28520
    },
    {
      "entropy": 6.121233654022217,
      "epoch": 2.8514020092967463,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041954153074824516,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1737453430891037,
      "num_tokens": 55810205.0,
      "step": 28525
    },
    {
      "entropy": 6.003313255310059,
      "epoch": 2.8519018343579745,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004195141780939113,
      "loss": 5.5612,
      "mean_token_accuracy": 0.1885147660970688,
      "num_tokens": 55820115.0,
      "step": 28530
    },
    {
      "entropy": 6.036087560653686,
      "epoch": 2.852401659419203,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041948682180363784,
      "loss": 5.5805,
      "mean_token_accuracy": 0.18579138666391373,
      "num_tokens": 55829620.0,
      "step": 28535
    },
    {
      "entropy": 6.0271673679351805,
      "epoch": 2.852901484480432,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004194594618781132,
      "loss": 5.5585,
      "mean_token_accuracy": 0.1798023208975792,
      "num_tokens": 55839945.0,
      "step": 28540
    },
    {
      "entropy": 5.9198554992675785,
      "epoch": 2.8534013095416606,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041943209831802567,
      "loss": 5.4557,
      "mean_token_accuracy": 0.19327008426189424,
      "num_tokens": 55850276.0,
      "step": 28545
    },
    {
      "entropy": 6.115046691894531,
      "epoch": 2.853901134602889,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041940473112406365,
      "loss": 5.5534,
      "mean_token_accuracy": 0.18481171876192093,
      "num_tokens": 55859898.0,
      "step": 28550
    },
    {
      "entropy": 6.049998950958252,
      "epoch": 2.8544009596641176,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041937736029691575,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18838018774986268,
      "num_tokens": 55869339.0,
      "step": 28555
    },
    {
      "entropy": 6.035439825057983,
      "epoch": 2.8549007847253463,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004193499858372706,
      "loss": 5.6332,
      "mean_token_accuracy": 0.18104694783687592,
      "num_tokens": 55880161.0,
      "step": 28560
    },
    {
      "entropy": 6.06524691581726,
      "epoch": 2.8554006097865745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000419322607745817,
      "loss": 5.6312,
      "mean_token_accuracy": 0.18173981457948685,
      "num_tokens": 55889522.0,
      "step": 28565
    },
    {
      "entropy": 6.095998287200928,
      "epoch": 2.855900434847803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041929522602324377,
      "loss": 5.6943,
      "mean_token_accuracy": 0.17718494683504105,
      "num_tokens": 55899253.0,
      "step": 28570
    },
    {
      "entropy": 6.048129367828369,
      "epoch": 2.856400259909032,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004192678406702398,
      "loss": 5.5205,
      "mean_token_accuracy": 0.1876794084906578,
      "num_tokens": 55908385.0,
      "step": 28575
    },
    {
      "entropy": 6.015753507614136,
      "epoch": 2.8569000849702606,
      "grad_norm": 1.3125,
      "learning_rate": 0.000419240451687494,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1896716684103012,
      "num_tokens": 55917171.0,
      "step": 28580
    },
    {
      "entropy": 6.047233581542969,
      "epoch": 2.857399910031489,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041921305907569555,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1801106169819832,
      "num_tokens": 55926839.0,
      "step": 28585
    },
    {
      "entropy": 6.077057981491089,
      "epoch": 2.8578997350927176,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004191856628355337,
      "loss": 5.6692,
      "mean_token_accuracy": 0.18036141395568847,
      "num_tokens": 55936850.0,
      "step": 28590
    },
    {
      "entropy": 6.098279237747192,
      "epoch": 2.8583995601539463,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004191582629676977,
      "loss": 5.7121,
      "mean_token_accuracy": 0.1687959536910057,
      "num_tokens": 55947242.0,
      "step": 28595
    },
    {
      "entropy": 6.117022895812989,
      "epoch": 2.8588993852151745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041913085947287675,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17528360933065415,
      "num_tokens": 55957150.0,
      "step": 28600
    },
    {
      "entropy": 6.0765515804290775,
      "epoch": 2.859399210276403,
      "grad_norm": 1.25,
      "learning_rate": 0.0004191034523517606,
      "loss": 5.6426,
      "mean_token_accuracy": 0.1823470711708069,
      "num_tokens": 55966769.0,
      "step": 28605
    },
    {
      "entropy": 6.015485715866089,
      "epoch": 2.859899035337632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041907604160503864,
      "loss": 5.5194,
      "mean_token_accuracy": 0.19312027841806412,
      "num_tokens": 55976536.0,
      "step": 28610
    },
    {
      "entropy": 5.958929252624512,
      "epoch": 2.8603988603988606,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041904862723340046,
      "loss": 5.4795,
      "mean_token_accuracy": 0.18714341819286345,
      "num_tokens": 55986607.0,
      "step": 28615
    },
    {
      "entropy": 6.107750368118286,
      "epoch": 2.860898685460089,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041902120923753593,
      "loss": 5.6381,
      "mean_token_accuracy": 0.18216864466667176,
      "num_tokens": 55995767.0,
      "step": 28620
    },
    {
      "entropy": 6.1288923740386965,
      "epoch": 2.8613985105213176,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004189937876181349,
      "loss": 5.7187,
      "mean_token_accuracy": 0.17486169338226318,
      "num_tokens": 56005602.0,
      "step": 28625
    },
    {
      "entropy": 6.041937446594238,
      "epoch": 2.8618983355825462,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004189663623758872,
      "loss": 5.598,
      "mean_token_accuracy": 0.1834500789642334,
      "num_tokens": 56014991.0,
      "step": 28630
    },
    {
      "entropy": 6.0840552806854244,
      "epoch": 2.8623981606437745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041893893351148284,
      "loss": 5.6587,
      "mean_token_accuracy": 0.17908178865909577,
      "num_tokens": 56023486.0,
      "step": 28635
    },
    {
      "entropy": 6.093478775024414,
      "epoch": 2.862897985705003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000418911501025612,
      "loss": 5.7052,
      "mean_token_accuracy": 0.17747148722410203,
      "num_tokens": 56034017.0,
      "step": 28640
    },
    {
      "entropy": 6.123781347274781,
      "epoch": 2.863397810766232,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004188840649189648,
      "loss": 5.6784,
      "mean_token_accuracy": 0.1737393245100975,
      "num_tokens": 56043433.0,
      "step": 28645
    },
    {
      "entropy": 6.009289789199829,
      "epoch": 2.8638976358274606,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041885662519223154,
      "loss": 5.5765,
      "mean_token_accuracy": 0.1823050782084465,
      "num_tokens": 56053116.0,
      "step": 28650
    },
    {
      "entropy": 5.936497974395752,
      "epoch": 2.864397460888689,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004188291818461026,
      "loss": 5.5478,
      "mean_token_accuracy": 0.18304950445890428,
      "num_tokens": 56063728.0,
      "step": 28655
    },
    {
      "entropy": 6.006323528289795,
      "epoch": 2.8648972859499175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041880173488126847,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18665555864572525,
      "num_tokens": 56072987.0,
      "step": 28660
    },
    {
      "entropy": 6.06512336730957,
      "epoch": 2.8653971110111462,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041877428429841967,
      "loss": 5.5566,
      "mean_token_accuracy": 0.1835345894098282,
      "num_tokens": 56082819.0,
      "step": 28665
    },
    {
      "entropy": 6.091949319839477,
      "epoch": 2.8658969360723745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004187468300982469,
      "loss": 5.741,
      "mean_token_accuracy": 0.1724376678466797,
      "num_tokens": 56094104.0,
      "step": 28670
    },
    {
      "entropy": 6.024542999267578,
      "epoch": 2.866396761133603,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004187193722814409,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18917314261198043,
      "num_tokens": 56103595.0,
      "step": 28675
    },
    {
      "entropy": 6.084650325775146,
      "epoch": 2.866896586194832,
      "grad_norm": 1.25,
      "learning_rate": 0.0004186919108486925,
      "loss": 5.6557,
      "mean_token_accuracy": 0.17580038905143738,
      "num_tokens": 56113458.0,
      "step": 28680
    },
    {
      "entropy": 6.070828247070312,
      "epoch": 2.8673964112560606,
      "grad_norm": 1.125,
      "learning_rate": 0.0004186644458006925,
      "loss": 5.6421,
      "mean_token_accuracy": 0.18265070617198945,
      "num_tokens": 56123694.0,
      "step": 28685
    },
    {
      "entropy": 5.9982898235321045,
      "epoch": 2.867896236317289,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000418636977138132,
      "loss": 5.543,
      "mean_token_accuracy": 0.18814538717269896,
      "num_tokens": 56132392.0,
      "step": 28690
    },
    {
      "entropy": 6.048486423492432,
      "epoch": 2.8683960613785175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041860950486170227,
      "loss": 5.6296,
      "mean_token_accuracy": 0.18042662739753723,
      "num_tokens": 56141175.0,
      "step": 28695
    },
    {
      "entropy": 6.044412469863891,
      "epoch": 2.8688958864397462,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004185820289720942,
      "loss": 5.6585,
      "mean_token_accuracy": 0.17062430903315545,
      "num_tokens": 56151089.0,
      "step": 28700
    },
    {
      "entropy": 6.015093564987183,
      "epoch": 2.8693957115009745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041855454946999925,
      "loss": 5.5227,
      "mean_token_accuracy": 0.18558406233787536,
      "num_tokens": 56160800.0,
      "step": 28705
    },
    {
      "entropy": 6.050983715057373,
      "epoch": 2.869895536562203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041852706635610884,
      "loss": 5.5882,
      "mean_token_accuracy": 0.18641610443592072,
      "num_tokens": 56171385.0,
      "step": 28710
    },
    {
      "entropy": 5.998235988616943,
      "epoch": 2.870395361623432,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041849957963111434,
      "loss": 5.5308,
      "mean_token_accuracy": 0.19267251193523408,
      "num_tokens": 56181187.0,
      "step": 28715
    },
    {
      "entropy": 6.0031280517578125,
      "epoch": 2.8708951866846606,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004184720892957073,
      "loss": 5.521,
      "mean_token_accuracy": 0.18864137977361678,
      "num_tokens": 56189478.0,
      "step": 28720
    },
    {
      "entropy": 6.0652587890625,
      "epoch": 2.871395011745889,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004184445953505795,
      "loss": 5.6452,
      "mean_token_accuracy": 0.18082950562238692,
      "num_tokens": 56198809.0,
      "step": 28725
    },
    {
      "entropy": 6.015886831283569,
      "epoch": 2.8718948368071175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004184170977964225,
      "loss": 5.5587,
      "mean_token_accuracy": 0.18553516417741775,
      "num_tokens": 56208826.0,
      "step": 28730
    },
    {
      "entropy": 6.084173011779785,
      "epoch": 2.8723946618683462,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041838959663392827,
      "loss": 5.5948,
      "mean_token_accuracy": 0.17879559546709062,
      "num_tokens": 56217811.0,
      "step": 28735
    },
    {
      "entropy": 6.052037143707276,
      "epoch": 2.8728944869295745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004183620918637887,
      "loss": 5.5762,
      "mean_token_accuracy": 0.1833435982465744,
      "num_tokens": 56226624.0,
      "step": 28740
    },
    {
      "entropy": 5.952481269836426,
      "epoch": 2.873394311990803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004183345834866957,
      "loss": 5.5209,
      "mean_token_accuracy": 0.18233538419008255,
      "num_tokens": 56235829.0,
      "step": 28745
    },
    {
      "entropy": 6.020563316345215,
      "epoch": 2.873894137052032,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041830707150334164,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18443678319454193,
      "num_tokens": 56245387.0,
      "step": 28750
    },
    {
      "entropy": 6.045729255676269,
      "epoch": 2.8743939621132606,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004182795559144184,
      "loss": 5.6424,
      "mean_token_accuracy": 0.17834228724241258,
      "num_tokens": 56254936.0,
      "step": 28755
    },
    {
      "entropy": 6.149669218063354,
      "epoch": 2.874893787174489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004182520367206184,
      "loss": 5.7398,
      "mean_token_accuracy": 0.17517222762107848,
      "num_tokens": 56265070.0,
      "step": 28760
    },
    {
      "entropy": 5.997907543182373,
      "epoch": 2.8753936122357175,
      "grad_norm": 1.125,
      "learning_rate": 0.00041822451392263413,
      "loss": 5.5083,
      "mean_token_accuracy": 0.19171238094568252,
      "num_tokens": 56274602.0,
      "step": 28765
    },
    {
      "entropy": 5.994124507904052,
      "epoch": 2.875893437296946,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004181969875211578,
      "loss": 5.5966,
      "mean_token_accuracy": 0.18065776973962783,
      "num_tokens": 56284078.0,
      "step": 28770
    },
    {
      "entropy": 5.999985647201538,
      "epoch": 2.8763932623581745,
      "grad_norm": 1.125,
      "learning_rate": 0.00041816945751688215,
      "loss": 5.563,
      "mean_token_accuracy": 0.18270056694746017,
      "num_tokens": 56295132.0,
      "step": 28775
    },
    {
      "entropy": 5.95076265335083,
      "epoch": 2.876893087419403,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041814192391049983,
      "loss": 5.547,
      "mean_token_accuracy": 0.20034223049879074,
      "num_tokens": 56305156.0,
      "step": 28780
    },
    {
      "entropy": 6.092486476898193,
      "epoch": 2.877392912480632,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004181143867027036,
      "loss": 5.6875,
      "mean_token_accuracy": 0.18087583184242248,
      "num_tokens": 56314272.0,
      "step": 28785
    },
    {
      "entropy": 6.002312517166137,
      "epoch": 2.8778927375418606,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041808684589418615,
      "loss": 5.6264,
      "mean_token_accuracy": 0.18201750814914702,
      "num_tokens": 56325184.0,
      "step": 28790
    },
    {
      "entropy": 5.971588277816773,
      "epoch": 2.878392562603089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004180593014856404,
      "loss": 5.5266,
      "mean_token_accuracy": 0.18109861463308335,
      "num_tokens": 56335288.0,
      "step": 28795
    },
    {
      "entropy": 5.993987941741944,
      "epoch": 2.8788923876643175,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004180317534777596,
      "loss": 5.54,
      "mean_token_accuracy": 0.1891075000166893,
      "num_tokens": 56344038.0,
      "step": 28800
    },
    {
      "entropy": 6.089092493057251,
      "epoch": 2.879392212725546,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041800420187123656,
      "loss": 5.7323,
      "mean_token_accuracy": 0.17628139853477479,
      "num_tokens": 56355120.0,
      "step": 28805
    },
    {
      "entropy": 6.050000715255737,
      "epoch": 2.8798920377867745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004179766466667646,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18295959383249283,
      "num_tokens": 56364398.0,
      "step": 28810
    },
    {
      "entropy": 5.9735277652740475,
      "epoch": 2.880391862848003,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004179490878650371,
      "loss": 5.5291,
      "mean_token_accuracy": 0.19143731594085694,
      "num_tokens": 56373988.0,
      "step": 28815
    },
    {
      "entropy": 6.053165292739868,
      "epoch": 2.880891687909232,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004179215254667473,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17384148091077806,
      "num_tokens": 56383398.0,
      "step": 28820
    },
    {
      "entropy": 6.128585910797119,
      "epoch": 2.8813915129704606,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004178939594725887,
      "loss": 5.6837,
      "mean_token_accuracy": 0.1721334710717201,
      "num_tokens": 56392679.0,
      "step": 28825
    },
    {
      "entropy": 6.062990331649781,
      "epoch": 2.881891338031689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004178663898832548,
      "loss": 5.6429,
      "mean_token_accuracy": 0.1749894767999649,
      "num_tokens": 56402812.0,
      "step": 28830
    },
    {
      "entropy": 6.028980875015259,
      "epoch": 2.8823911630929175,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004178388166994392,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17710015624761583,
      "num_tokens": 56412799.0,
      "step": 28835
    },
    {
      "entropy": 6.001198816299438,
      "epoch": 2.882890988154146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004178112399218359,
      "loss": 5.5538,
      "mean_token_accuracy": 0.18379381000995637,
      "num_tokens": 56422200.0,
      "step": 28840
    },
    {
      "entropy": 5.998325395584106,
      "epoch": 2.8833908132153745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041778365955113844,
      "loss": 5.4912,
      "mean_token_accuracy": 0.19207023829221725,
      "num_tokens": 56432466.0,
      "step": 28845
    },
    {
      "entropy": 6.000420427322387,
      "epoch": 2.883890638276603,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004177560755880409,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18372411131858826,
      "num_tokens": 56442139.0,
      "step": 28850
    },
    {
      "entropy": 5.913214826583863,
      "epoch": 2.884390463337832,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004177284880332373,
      "loss": 5.4922,
      "mean_token_accuracy": 0.18736203759908676,
      "num_tokens": 56452509.0,
      "step": 28855
    },
    {
      "entropy": 6.004851627349853,
      "epoch": 2.8848902883990606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041770089688742155,
      "loss": 5.5226,
      "mean_token_accuracy": 0.18562132716178895,
      "num_tokens": 56462135.0,
      "step": 28860
    },
    {
      "entropy": 5.98517951965332,
      "epoch": 2.885390113460289,
      "grad_norm": 1.5,
      "learning_rate": 0.000417673302151288,
      "loss": 5.487,
      "mean_token_accuracy": 0.19038112461566925,
      "num_tokens": 56471544.0,
      "step": 28865
    },
    {
      "entropy": 6.095265245437622,
      "epoch": 2.8858899385215175,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004176457038255309,
      "loss": 5.6587,
      "mean_token_accuracy": 0.18310460299253464,
      "num_tokens": 56481293.0,
      "step": 28870
    },
    {
      "entropy": 5.975299882888794,
      "epoch": 2.886389763582746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004176181019108446,
      "loss": 5.5936,
      "mean_token_accuracy": 0.18104808926582336,
      "num_tokens": 56490265.0,
      "step": 28875
    },
    {
      "entropy": 5.95723967552185,
      "epoch": 2.8868895886439745,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004175904964079236,
      "loss": 5.4953,
      "mean_token_accuracy": 0.19385033398866652,
      "num_tokens": 56499617.0,
      "step": 28880
    },
    {
      "entropy": 5.988598918914795,
      "epoch": 2.887389413705203,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004175628873174623,
      "loss": 5.551,
      "mean_token_accuracy": 0.19019993990659714,
      "num_tokens": 56509962.0,
      "step": 28885
    },
    {
      "entropy": 6.091445970535278,
      "epoch": 2.887889238766432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041753527464015547,
      "loss": 5.6296,
      "mean_token_accuracy": 0.17801593840122223,
      "num_tokens": 56519786.0,
      "step": 28890
    },
    {
      "entropy": 6.022245645523071,
      "epoch": 2.8883890638276606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004175076583766978,
      "loss": 5.6042,
      "mean_token_accuracy": 0.17378997057676315,
      "num_tokens": 56528675.0,
      "step": 28895
    },
    {
      "entropy": 6.0434770584106445,
      "epoch": 2.888888888888889,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004174800385277842,
      "loss": 5.656,
      "mean_token_accuracy": 0.18026389181613922,
      "num_tokens": 56538783.0,
      "step": 28900
    },
    {
      "entropy": 6.100630807876587,
      "epoch": 2.8893887139501175,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004174524150941094,
      "loss": 5.7102,
      "mean_token_accuracy": 0.17204311341047288,
      "num_tokens": 56549063.0,
      "step": 28905
    },
    {
      "entropy": 6.083174753189087,
      "epoch": 2.889888539011346,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004174247880763685,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18402988463640213,
      "num_tokens": 56558414.0,
      "step": 28910
    },
    {
      "entropy": 6.029193782806397,
      "epoch": 2.8903883640725745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004173971574752566,
      "loss": 5.642,
      "mean_token_accuracy": 0.17932081073522568,
      "num_tokens": 56568493.0,
      "step": 28915
    },
    {
      "entropy": 6.054084539413452,
      "epoch": 2.890888189133803,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004173695232914688,
      "loss": 5.6481,
      "mean_token_accuracy": 0.16955292969942093,
      "num_tokens": 56578832.0,
      "step": 28920
    },
    {
      "entropy": 6.057915592193604,
      "epoch": 2.891388014195032,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004173418855257005,
      "loss": 5.6157,
      "mean_token_accuracy": 0.1834283858537674,
      "num_tokens": 56587707.0,
      "step": 28925
    },
    {
      "entropy": 5.980989646911621,
      "epoch": 2.8918878392562606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000417314244178647,
      "loss": 5.5691,
      "mean_token_accuracy": 0.18807289153337478,
      "num_tokens": 56597145.0,
      "step": 28930
    },
    {
      "entropy": 5.980626106262207,
      "epoch": 2.892387664317489,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004172865992510037,
      "loss": 5.5902,
      "mean_token_accuracy": 0.18407552242279052,
      "num_tokens": 56607148.0,
      "step": 28935
    },
    {
      "entropy": 5.984575414657593,
      "epoch": 2.8928874893787175,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041725895074346615,
      "loss": 5.5519,
      "mean_token_accuracy": 0.1799624592065811,
      "num_tokens": 56616574.0,
      "step": 28940
    },
    {
      "entropy": 6.043421792984009,
      "epoch": 2.893387314439946,
      "grad_norm": 1.125,
      "learning_rate": 0.0004172312986567301,
      "loss": 5.5804,
      "mean_token_accuracy": 0.1886957973241806,
      "num_tokens": 56625986.0,
      "step": 28945
    },
    {
      "entropy": 5.984831237792969,
      "epoch": 2.8938871395011745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004172036429914911,
      "loss": 5.5839,
      "mean_token_accuracy": 0.183633553981781,
      "num_tokens": 56634524.0,
      "step": 28950
    },
    {
      "entropy": 6.0394065380096436,
      "epoch": 2.894386964562403,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041717598374844505,
      "loss": 5.5891,
      "mean_token_accuracy": 0.17836853265762329,
      "num_tokens": 56643770.0,
      "step": 28955
    },
    {
      "entropy": 5.972534084320069,
      "epoch": 2.894886789623632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041714832092828785,
      "loss": 5.5478,
      "mean_token_accuracy": 0.19309113770723343,
      "num_tokens": 56654339.0,
      "step": 28960
    },
    {
      "entropy": 5.983754730224609,
      "epoch": 2.89538661468486,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041712065453171544,
      "loss": 5.5108,
      "mean_token_accuracy": 0.18054074048995972,
      "num_tokens": 56664814.0,
      "step": 28965
    },
    {
      "entropy": 6.026061868667602,
      "epoch": 2.895886439746089,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041709298455942394,
      "loss": 5.6429,
      "mean_token_accuracy": 0.1802267163991928,
      "num_tokens": 56674602.0,
      "step": 28970
    },
    {
      "entropy": 5.943478775024414,
      "epoch": 2.8963862648073175,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041706531101210956,
      "loss": 5.5918,
      "mean_token_accuracy": 0.19094932973384857,
      "num_tokens": 56684294.0,
      "step": 28975
    },
    {
      "entropy": 6.039227771759033,
      "epoch": 2.896886089868546,
      "grad_norm": 1.125,
      "learning_rate": 0.00041703763389046853,
      "loss": 5.6562,
      "mean_token_accuracy": 0.18034527748823165,
      "num_tokens": 56693626.0,
      "step": 28980
    },
    {
      "entropy": 6.087743043899536,
      "epoch": 2.8973859149297745,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041700995319519707,
      "loss": 5.6285,
      "mean_token_accuracy": 0.1783273324370384,
      "num_tokens": 56704574.0,
      "step": 28985
    },
    {
      "entropy": 5.989123868942261,
      "epoch": 2.897885739991003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004169822689269918,
      "loss": 5.5339,
      "mean_token_accuracy": 0.19056861102581024,
      "num_tokens": 56714990.0,
      "step": 28990
    },
    {
      "entropy": 6.029825830459595,
      "epoch": 2.898385565052232,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041695458108654914,
      "loss": 5.6229,
      "mean_token_accuracy": 0.17852021008729935,
      "num_tokens": 56724651.0,
      "step": 28995
    },
    {
      "entropy": 5.982212781906128,
      "epoch": 2.89888539011346,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041692688967456576,
      "loss": 5.545,
      "mean_token_accuracy": 0.18659220784902572,
      "num_tokens": 56733500.0,
      "step": 29000
    },
    {
      "entropy": 6.053435659408569,
      "epoch": 2.899385215174689,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041689919469173835,
      "loss": 5.7163,
      "mean_token_accuracy": 0.1790541708469391,
      "num_tokens": 56743010.0,
      "step": 29005
    },
    {
      "entropy": 6.081256151199341,
      "epoch": 2.8998850402359175,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004168714961387637,
      "loss": 5.6488,
      "mean_token_accuracy": 0.17956655025482177,
      "num_tokens": 56753441.0,
      "step": 29010
    },
    {
      "entropy": 6.102076244354248,
      "epoch": 2.900384865297146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004168437940163388,
      "loss": 5.6764,
      "mean_token_accuracy": 0.1782798394560814,
      "num_tokens": 56762767.0,
      "step": 29015
    },
    {
      "entropy": 6.07594313621521,
      "epoch": 2.9008846903583745,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004168160883251605,
      "loss": 5.6603,
      "mean_token_accuracy": 0.1837557449936867,
      "num_tokens": 56772354.0,
      "step": 29020
    },
    {
      "entropy": 6.080379247665405,
      "epoch": 2.901384515419603,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041678837906592586,
      "loss": 5.5629,
      "mean_token_accuracy": 0.1810840353369713,
      "num_tokens": 56782453.0,
      "step": 29025
    },
    {
      "entropy": 6.015006494522095,
      "epoch": 2.901884340480832,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041676066623933216,
      "loss": 5.6714,
      "mean_token_accuracy": 0.1798033371567726,
      "num_tokens": 56792762.0,
      "step": 29030
    },
    {
      "entropy": 5.989713525772094,
      "epoch": 2.90238416554206,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004167329498460766,
      "loss": 5.5707,
      "mean_token_accuracy": 0.1869126409292221,
      "num_tokens": 56802160.0,
      "step": 29035
    },
    {
      "entropy": 6.028722906112671,
      "epoch": 2.902883990603289,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00041670522988685637,
      "loss": 5.5509,
      "mean_token_accuracy": 0.1801064595580101,
      "num_tokens": 56811691.0,
      "step": 29040
    },
    {
      "entropy": 6.073664903640747,
      "epoch": 2.9033838156645175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041667750636236914,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18633545637130738,
      "num_tokens": 56821721.0,
      "step": 29045
    },
    {
      "entropy": 5.841474008560181,
      "epoch": 2.903883640725746,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004166497792733122,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20582425892353057,
      "num_tokens": 56832463.0,
      "step": 29050
    },
    {
      "entropy": 6.006496715545654,
      "epoch": 2.9043834657869745,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004166220486203834,
      "loss": 5.688,
      "mean_token_accuracy": 0.1694597378373146,
      "num_tokens": 56842814.0,
      "step": 29055
    },
    {
      "entropy": 6.135624504089355,
      "epoch": 2.904883290848203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004165943144042802,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17082666754722595,
      "num_tokens": 56853410.0,
      "step": 29060
    },
    {
      "entropy": 6.081226348876953,
      "epoch": 2.905383115909432,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041656657662570047,
      "loss": 5.6097,
      "mean_token_accuracy": 0.1854262799024582,
      "num_tokens": 56864264.0,
      "step": 29065
    },
    {
      "entropy": 6.029941892623901,
      "epoch": 2.90588294097066,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004165388352853421,
      "loss": 5.6315,
      "mean_token_accuracy": 0.179891200363636,
      "num_tokens": 56873275.0,
      "step": 29070
    },
    {
      "entropy": 5.984784030914307,
      "epoch": 2.906382766031889,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041651109038390305,
      "loss": 5.5288,
      "mean_token_accuracy": 0.19816692769527436,
      "num_tokens": 56882700.0,
      "step": 29075
    },
    {
      "entropy": 6.131310939788818,
      "epoch": 2.9068825910931175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041648334192208146,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17205194532871246,
      "num_tokens": 56892836.0,
      "step": 29080
    },
    {
      "entropy": 6.008128118515015,
      "epoch": 2.907382416154346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041645558990057536,
      "loss": 5.5812,
      "mean_token_accuracy": 0.1850943922996521,
      "num_tokens": 56904337.0,
      "step": 29085
    },
    {
      "entropy": 5.944661474227905,
      "epoch": 2.9078822412155745,
      "grad_norm": 1.578125,
      "learning_rate": 0.00041642783432008294,
      "loss": 5.546,
      "mean_token_accuracy": 0.18736034780740737,
      "num_tokens": 56914296.0,
      "step": 29090
    },
    {
      "entropy": 6.020099449157715,
      "epoch": 2.908382066276803,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041640007518130274,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1780528485774994,
      "num_tokens": 56924528.0,
      "step": 29095
    },
    {
      "entropy": 6.076849460601807,
      "epoch": 2.908881891338032,
      "grad_norm": 1.0625,
      "learning_rate": 0.000416372312484933,
      "loss": 5.5947,
      "mean_token_accuracy": 0.18825190365314484,
      "num_tokens": 56934585.0,
      "step": 29100
    },
    {
      "entropy": 6.056212997436523,
      "epoch": 2.90938171639926,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041634454623167217,
      "loss": 5.6394,
      "mean_token_accuracy": 0.1867221936583519,
      "num_tokens": 56944551.0,
      "step": 29105
    },
    {
      "entropy": 6.038660621643066,
      "epoch": 2.909881541460489,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041631677642221896,
      "loss": 5.6224,
      "mean_token_accuracy": 0.1788402184844017,
      "num_tokens": 56953120.0,
      "step": 29110
    },
    {
      "entropy": 6.0315605163574215,
      "epoch": 2.9103813665217175,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000416289003057272,
      "loss": 5.5911,
      "mean_token_accuracy": 0.1820249915122986,
      "num_tokens": 56963747.0,
      "step": 29115
    },
    {
      "entropy": 6.004735040664673,
      "epoch": 2.910881191582946,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004162612261375302,
      "loss": 5.5953,
      "mean_token_accuracy": 0.1826077476143837,
      "num_tokens": 56974192.0,
      "step": 29120
    },
    {
      "entropy": 6.076845407485962,
      "epoch": 2.9113810166441745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041623344566369216,
      "loss": 5.6846,
      "mean_token_accuracy": 0.171724534034729,
      "num_tokens": 56984672.0,
      "step": 29125
    },
    {
      "entropy": 6.0288245677948,
      "epoch": 2.911880841705403,
      "grad_norm": 1.25,
      "learning_rate": 0.000416205661636457,
      "loss": 5.578,
      "mean_token_accuracy": 0.17814985066652297,
      "num_tokens": 56993737.0,
      "step": 29130
    },
    {
      "entropy": 6.104853343963623,
      "epoch": 2.912380666766632,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004161778740565238,
      "loss": 5.6538,
      "mean_token_accuracy": 0.1775185912847519,
      "num_tokens": 57004253.0,
      "step": 29135
    },
    {
      "entropy": 6.039421796798706,
      "epoch": 2.91288049182786,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041615008292459157,
      "loss": 5.5627,
      "mean_token_accuracy": 0.18056328892707824,
      "num_tokens": 57013690.0,
      "step": 29140
    },
    {
      "entropy": 5.954478740692139,
      "epoch": 2.913380316889089,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041612228824135954,
      "loss": 5.4646,
      "mean_token_accuracy": 0.1982791766524315,
      "num_tokens": 57022702.0,
      "step": 29145
    },
    {
      "entropy": 6.034492683410645,
      "epoch": 2.9138801419503175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041609449000752706,
      "loss": 5.6476,
      "mean_token_accuracy": 0.18325491547584533,
      "num_tokens": 57032442.0,
      "step": 29150
    },
    {
      "entropy": 6.080494403839111,
      "epoch": 2.914379967011546,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004160666882237936,
      "loss": 5.5662,
      "mean_token_accuracy": 0.18340432047843933,
      "num_tokens": 57042284.0,
      "step": 29155
    },
    {
      "entropy": 6.003726387023926,
      "epoch": 2.9148797920727745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004160388828908585,
      "loss": 5.5214,
      "mean_token_accuracy": 0.18596233874559404,
      "num_tokens": 57051504.0,
      "step": 29160
    },
    {
      "entropy": 6.032172298431396,
      "epoch": 2.915379617134003,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041601107400942146,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18615123331546785,
      "num_tokens": 57060717.0,
      "step": 29165
    },
    {
      "entropy": 5.97490348815918,
      "epoch": 2.915879442195232,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041598326158018197,
      "loss": 5.5943,
      "mean_token_accuracy": 0.1789835900068283,
      "num_tokens": 57070767.0,
      "step": 29170
    },
    {
      "entropy": 6.021779918670655,
      "epoch": 2.91637926725646,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041595544560384004,
      "loss": 5.5419,
      "mean_token_accuracy": 0.18840219527482988,
      "num_tokens": 57080922.0,
      "step": 29175
    },
    {
      "entropy": 5.964443731307983,
      "epoch": 2.916879092317689,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004159276260810952,
      "loss": 5.5057,
      "mean_token_accuracy": 0.1916899099946022,
      "num_tokens": 57090703.0,
      "step": 29180
    },
    {
      "entropy": 5.997401571273803,
      "epoch": 2.9173789173789175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041589980301264765,
      "loss": 5.5934,
      "mean_token_accuracy": 0.1866826444864273,
      "num_tokens": 57100468.0,
      "step": 29185
    },
    {
      "entropy": 6.064566659927368,
      "epoch": 2.917878742440146,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004158719763991973,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18029579222202302,
      "num_tokens": 57110281.0,
      "step": 29190
    },
    {
      "entropy": 5.98733081817627,
      "epoch": 2.9183785675013745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041584414624144426,
      "loss": 5.6006,
      "mean_token_accuracy": 0.1827958956360817,
      "num_tokens": 57120280.0,
      "step": 29195
    },
    {
      "entropy": 5.962887620925903,
      "epoch": 2.918878392562603,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004158163125400887,
      "loss": 5.6119,
      "mean_token_accuracy": 0.18364714086055756,
      "num_tokens": 57129577.0,
      "step": 29200
    },
    {
      "entropy": 6.031758451461792,
      "epoch": 2.919378217623832,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041578847529583106,
      "loss": 5.645,
      "mean_token_accuracy": 0.18029373586177827,
      "num_tokens": 57139327.0,
      "step": 29205
    },
    {
      "entropy": 6.039500665664673,
      "epoch": 2.91987804268506,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004157606345093715,
      "loss": 5.5898,
      "mean_token_accuracy": 0.18205122351646424,
      "num_tokens": 57149288.0,
      "step": 29210
    },
    {
      "entropy": 6.069787406921387,
      "epoch": 2.920377867746289,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004157327901814107,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17866008579730988,
      "num_tokens": 57158308.0,
      "step": 29215
    },
    {
      "entropy": 6.006001281738281,
      "epoch": 2.9208776928075175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041570494231264906,
      "loss": 5.5422,
      "mean_token_accuracy": 0.18560656607151033,
      "num_tokens": 57167893.0,
      "step": 29220
    },
    {
      "entropy": 5.963773584365844,
      "epoch": 2.9213775178687458,
      "grad_norm": 1.125,
      "learning_rate": 0.00041567709090378725,
      "loss": 5.5356,
      "mean_token_accuracy": 0.19403080344200135,
      "num_tokens": 57178489.0,
      "step": 29225
    },
    {
      "entropy": 6.075389099121094,
      "epoch": 2.9218773429299745,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041564923595552604,
      "loss": 5.5825,
      "mean_token_accuracy": 0.17715671062469482,
      "num_tokens": 57187713.0,
      "step": 29230
    },
    {
      "entropy": 6.037934017181397,
      "epoch": 2.922377167991203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041562137746856625,
      "loss": 5.5405,
      "mean_token_accuracy": 0.1829712450504303,
      "num_tokens": 57197914.0,
      "step": 29235
    },
    {
      "entropy": 5.976506185531616,
      "epoch": 2.922876993052432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041559351544360893,
      "loss": 5.5663,
      "mean_token_accuracy": 0.18941060453653336,
      "num_tokens": 57207902.0,
      "step": 29240
    },
    {
      "entropy": 6.110430145263672,
      "epoch": 2.92337681811366,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041556564988135474,
      "loss": 5.6447,
      "mean_token_accuracy": 0.1759818360209465,
      "num_tokens": 57219000.0,
      "step": 29245
    },
    {
      "entropy": 6.072840785980224,
      "epoch": 2.923876643174889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004155377807825051,
      "loss": 5.6244,
      "mean_token_accuracy": 0.18413797467947007,
      "num_tokens": 57228333.0,
      "step": 29250
    },
    {
      "entropy": 6.01442551612854,
      "epoch": 2.9243764682361175,
      "grad_norm": 1.15625,
      "learning_rate": 0.000415509908147761,
      "loss": 5.628,
      "mean_token_accuracy": 0.18216869980096817,
      "num_tokens": 57237595.0,
      "step": 29255
    },
    {
      "entropy": 6.030862665176391,
      "epoch": 2.9248762932973458,
      "grad_norm": 1.125,
      "learning_rate": 0.00041548203197782377,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18807681798934936,
      "num_tokens": 57246797.0,
      "step": 29260
    },
    {
      "entropy": 6.077279949188233,
      "epoch": 2.9253761183585745,
      "grad_norm": 1.5625,
      "learning_rate": 0.00041545415227339474,
      "loss": 5.6263,
      "mean_token_accuracy": 0.18635233491659164,
      "num_tokens": 57258685.0,
      "step": 29265
    },
    {
      "entropy": 6.061473274230957,
      "epoch": 2.925875943419803,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004154262690351754,
      "loss": 5.6332,
      "mean_token_accuracy": 0.17685779929161072,
      "num_tokens": 57269103.0,
      "step": 29270
    },
    {
      "entropy": 6.051159334182739,
      "epoch": 2.926375768481032,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004153983822638673,
      "loss": 5.6826,
      "mean_token_accuracy": 0.1771888852119446,
      "num_tokens": 57279346.0,
      "step": 29275
    },
    {
      "entropy": 6.113370084762574,
      "epoch": 2.92687559354226,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041537049196017196,
      "loss": 5.6261,
      "mean_token_accuracy": 0.1774197056889534,
      "num_tokens": 57289438.0,
      "step": 29280
    },
    {
      "entropy": 5.962847089767456,
      "epoch": 2.927375418603489,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041534259812479125,
      "loss": 5.6507,
      "mean_token_accuracy": 0.18135122954845428,
      "num_tokens": 57299621.0,
      "step": 29285
    },
    {
      "entropy": 6.04791955947876,
      "epoch": 2.927875243664717,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004153147007584269,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18395321369171141,
      "num_tokens": 57309043.0,
      "step": 29290
    },
    {
      "entropy": 6.059103918075562,
      "epoch": 2.9283750687259458,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041528679986178066,
      "loss": 5.5653,
      "mean_token_accuracy": 0.18990875035524368,
      "num_tokens": 57318554.0,
      "step": 29295
    },
    {
      "entropy": 6.021565628051758,
      "epoch": 2.9288748937871745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041525889543555475,
      "loss": 5.5817,
      "mean_token_accuracy": 0.18486708998680115,
      "num_tokens": 57327798.0,
      "step": 29300
    },
    {
      "entropy": 6.004161643981933,
      "epoch": 2.929374718848403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000415230987480451,
      "loss": 5.631,
      "mean_token_accuracy": 0.1872919902205467,
      "num_tokens": 57337395.0,
      "step": 29305
    },
    {
      "entropy": 6.015544176101685,
      "epoch": 2.929874543909632,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004152030759971718,
      "loss": 5.5883,
      "mean_token_accuracy": 0.18769689053297042,
      "num_tokens": 57347025.0,
      "step": 29310
    },
    {
      "entropy": 5.933170318603516,
      "epoch": 2.93037436897086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004151751609864193,
      "loss": 5.4161,
      "mean_token_accuracy": 0.19684004187583923,
      "num_tokens": 57356093.0,
      "step": 29315
    },
    {
      "entropy": 6.008208990097046,
      "epoch": 2.930874194032089,
      "grad_norm": 1.25,
      "learning_rate": 0.0004151472424488957,
      "loss": 5.6098,
      "mean_token_accuracy": 0.18528386056423188,
      "num_tokens": 57365852.0,
      "step": 29320
    },
    {
      "entropy": 6.05967149734497,
      "epoch": 2.931374019093317,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004151193203853036,
      "loss": 5.6438,
      "mean_token_accuracy": 0.1743568003177643,
      "num_tokens": 57375598.0,
      "step": 29325
    },
    {
      "entropy": 5.98837718963623,
      "epoch": 2.9318738441545458,
      "grad_norm": 1.125,
      "learning_rate": 0.00041509139479634544,
      "loss": 5.5168,
      "mean_token_accuracy": 0.18752224147319793,
      "num_tokens": 57385533.0,
      "step": 29330
    },
    {
      "entropy": 6.030917596817017,
      "epoch": 2.9323736692157745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041506346568272385,
      "loss": 5.577,
      "mean_token_accuracy": 0.18863266110420226,
      "num_tokens": 57396109.0,
      "step": 29335
    },
    {
      "entropy": 6.141340589523315,
      "epoch": 2.932873494277003,
      "grad_norm": 1.296875,
      "learning_rate": 0.00041503553304514157,
      "loss": 5.6765,
      "mean_token_accuracy": 0.1785540223121643,
      "num_tokens": 57407186.0,
      "step": 29340
    },
    {
      "entropy": 6.040895557403564,
      "epoch": 2.933373319338232,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004150075968843012,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1865852266550064,
      "num_tokens": 57415934.0,
      "step": 29345
    },
    {
      "entropy": 6.010465621948242,
      "epoch": 2.93387314439946,
      "grad_norm": 1.125,
      "learning_rate": 0.00041497965720090573,
      "loss": 5.5863,
      "mean_token_accuracy": 0.190141761302948,
      "num_tokens": 57425650.0,
      "step": 29350
    },
    {
      "entropy": 5.95405535697937,
      "epoch": 2.934372969460689,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041495171399565816,
      "loss": 5.5281,
      "mean_token_accuracy": 0.18773210048675537,
      "num_tokens": 57436070.0,
      "step": 29355
    },
    {
      "entropy": 6.038443470001221,
      "epoch": 2.934872794521917,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004149237672692615,
      "loss": 5.5958,
      "mean_token_accuracy": 0.1842592626810074,
      "num_tokens": 57446050.0,
      "step": 29360
    },
    {
      "entropy": 6.000102949142456,
      "epoch": 2.9353726195831458,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004148958170224188,
      "loss": 5.5408,
      "mean_token_accuracy": 0.18711530417203903,
      "num_tokens": 57455898.0,
      "step": 29365
    },
    {
      "entropy": 5.962058115005493,
      "epoch": 2.9358724446443745,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041486786325583343,
      "loss": 5.5478,
      "mean_token_accuracy": 0.18816670924425125,
      "num_tokens": 57464754.0,
      "step": 29370
    },
    {
      "entropy": 6.016819429397583,
      "epoch": 2.936372269705603,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041483990597020854,
      "loss": 5.6185,
      "mean_token_accuracy": 0.18022072017192842,
      "num_tokens": 57474267.0,
      "step": 29375
    },
    {
      "entropy": 5.952918291091919,
      "epoch": 2.936872094766832,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041481194516624763,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1838544100522995,
      "num_tokens": 57484193.0,
      "step": 29380
    },
    {
      "entropy": 6.024671983718872,
      "epoch": 2.93737191982806,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004147839808446541,
      "loss": 5.4851,
      "mean_token_accuracy": 0.188570736348629,
      "num_tokens": 57493004.0,
      "step": 29385
    },
    {
      "entropy": 6.017355298995971,
      "epoch": 2.937871744889289,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041475601300613166,
      "loss": 5.5219,
      "mean_token_accuracy": 0.19598665684461594,
      "num_tokens": 57502509.0,
      "step": 29390
    },
    {
      "entropy": 6.044591426849365,
      "epoch": 2.938371569950517,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041472804165138384,
      "loss": 5.5649,
      "mean_token_accuracy": 0.18064139038324356,
      "num_tokens": 57511770.0,
      "step": 29395
    },
    {
      "entropy": 5.977962970733643,
      "epoch": 2.9388713950117458,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004147000667811145,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18188648223876952,
      "num_tokens": 57521724.0,
      "step": 29400
    },
    {
      "entropy": 5.99794111251831,
      "epoch": 2.9393712200729745,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004146720883960273,
      "loss": 5.575,
      "mean_token_accuracy": 0.18668529987335206,
      "num_tokens": 57531393.0,
      "step": 29405
    },
    {
      "entropy": 6.011261701583862,
      "epoch": 2.939871045134203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004146441064968264,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18025825619697572,
      "num_tokens": 57539793.0,
      "step": 29410
    },
    {
      "entropy": 6.132712030410767,
      "epoch": 2.940370870195432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041461612108421576,
      "loss": 5.7406,
      "mean_token_accuracy": 0.17193396389484406,
      "num_tokens": 57549099.0,
      "step": 29415
    },
    {
      "entropy": 6.075200414657592,
      "epoch": 2.94087069525666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041458813215889935,
      "loss": 5.5884,
      "mean_token_accuracy": 0.18158857822418212,
      "num_tokens": 57558146.0,
      "step": 29420
    },
    {
      "entropy": 6.0259661197662355,
      "epoch": 2.941370520317889,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004145601397215815,
      "loss": 5.5598,
      "mean_token_accuracy": 0.1833136796951294,
      "num_tokens": 57567448.0,
      "step": 29425
    },
    {
      "entropy": 6.029967880249023,
      "epoch": 2.941870345379117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041453214377296636,
      "loss": 5.608,
      "mean_token_accuracy": 0.1781615749001503,
      "num_tokens": 57577286.0,
      "step": 29430
    },
    {
      "entropy": 5.974725961685181,
      "epoch": 2.9423701704403458,
      "grad_norm": 1.25,
      "learning_rate": 0.0004145041443137585,
      "loss": 5.5512,
      "mean_token_accuracy": 0.17970898002386093,
      "num_tokens": 57587926.0,
      "step": 29435
    },
    {
      "entropy": 5.9086870670318605,
      "epoch": 2.9428699955015745,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004144761413446622,
      "loss": 5.4693,
      "mean_token_accuracy": 0.19614626616239547,
      "num_tokens": 57598533.0,
      "step": 29440
    },
    {
      "entropy": 5.999774694442749,
      "epoch": 2.943369820562803,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041444813486638215,
      "loss": 5.5427,
      "mean_token_accuracy": 0.19911328107118606,
      "num_tokens": 57607410.0,
      "step": 29445
    },
    {
      "entropy": 6.0105468273162845,
      "epoch": 2.943869645624032,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041442012487962284,
      "loss": 5.6673,
      "mean_token_accuracy": 0.1742256224155426,
      "num_tokens": 57617787.0,
      "step": 29450
    },
    {
      "entropy": 6.030466222763062,
      "epoch": 2.94436947068526,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004143921113850891,
      "loss": 5.606,
      "mean_token_accuracy": 0.18333171010017396,
      "num_tokens": 57627537.0,
      "step": 29455
    },
    {
      "entropy": 6.044958829879761,
      "epoch": 2.944869295746489,
      "grad_norm": 1.421875,
      "learning_rate": 0.00041436409438348564,
      "loss": 5.578,
      "mean_token_accuracy": 0.18482746183872223,
      "num_tokens": 57637281.0,
      "step": 29460
    },
    {
      "entropy": 5.964457988739014,
      "epoch": 2.945369120807717,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004143360738755175,
      "loss": 5.5369,
      "mean_token_accuracy": 0.18672540038824081,
      "num_tokens": 57646889.0,
      "step": 29465
    },
    {
      "entropy": 6.002085113525391,
      "epoch": 2.9458689458689458,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004143080498618895,
      "loss": 5.5266,
      "mean_token_accuracy": 0.1915872722864151,
      "num_tokens": 57655583.0,
      "step": 29470
    },
    {
      "entropy": 6.016175699234009,
      "epoch": 2.9463687709301745,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041428002234330694,
      "loss": 5.5628,
      "mean_token_accuracy": 0.18201389908790588,
      "num_tokens": 57664778.0,
      "step": 29475
    },
    {
      "entropy": 5.930409860610962,
      "epoch": 2.946868595991403,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004142519913204748,
      "loss": 5.5522,
      "mean_token_accuracy": 0.1823820188641548,
      "num_tokens": 57674663.0,
      "step": 29480
    },
    {
      "entropy": 6.019059085845948,
      "epoch": 2.9473684210526314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041422395679409834,
      "loss": 5.6099,
      "mean_token_accuracy": 0.18915409743785858,
      "num_tokens": 57684267.0,
      "step": 29485
    },
    {
      "entropy": 6.0954467296600345,
      "epoch": 2.94786824611386,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041419591876488304,
      "loss": 5.5728,
      "mean_token_accuracy": 0.183589468896389,
      "num_tokens": 57694513.0,
      "step": 29490
    },
    {
      "entropy": 6.043826627731323,
      "epoch": 2.948368071175089,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004141678772335342,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1919431358575821,
      "num_tokens": 57703474.0,
      "step": 29495
    },
    {
      "entropy": 5.9695837020874025,
      "epoch": 2.948867896236317,
      "grad_norm": 1.125,
      "learning_rate": 0.00041413983220075736,
      "loss": 5.5419,
      "mean_token_accuracy": 0.1866513416171074,
      "num_tokens": 57714371.0,
      "step": 29500
    },
    {
      "entropy": 6.110765647888184,
      "epoch": 2.9493677212975458,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041411178366725814,
      "loss": 5.6376,
      "mean_token_accuracy": 0.1829434961080551,
      "num_tokens": 57724403.0,
      "step": 29505
    },
    {
      "entropy": 6.0138226509094235,
      "epoch": 2.9498675463587745,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004140837316337423,
      "loss": 5.601,
      "mean_token_accuracy": 0.184197336435318,
      "num_tokens": 57733705.0,
      "step": 29510
    },
    {
      "entropy": 6.011688375473023,
      "epoch": 2.950367371420003,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004140556761009155,
      "loss": 5.5126,
      "mean_token_accuracy": 0.18886819630861282,
      "num_tokens": 57743597.0,
      "step": 29515
    },
    {
      "entropy": 5.995961856842041,
      "epoch": 2.9508671964812314,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004140276170694837,
      "loss": 5.6063,
      "mean_token_accuracy": 0.187365585565567,
      "num_tokens": 57754085.0,
      "step": 29520
    },
    {
      "entropy": 6.0861516952514645,
      "epoch": 2.95136702154246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041399955454015286,
      "loss": 5.6465,
      "mean_token_accuracy": 0.18268795311450958,
      "num_tokens": 57764140.0,
      "step": 29525
    },
    {
      "entropy": 6.099708795547485,
      "epoch": 2.951866846603689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041397148851362903,
      "loss": 5.581,
      "mean_token_accuracy": 0.19307701736688615,
      "num_tokens": 57772537.0,
      "step": 29530
    },
    {
      "entropy": 6.106938648223877,
      "epoch": 2.952366671664917,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00041394341899061814,
      "loss": 5.7328,
      "mean_token_accuracy": 0.16919091939926148,
      "num_tokens": 57782758.0,
      "step": 29535
    },
    {
      "entropy": 6.105792713165283,
      "epoch": 2.9528664967261458,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004139153459718267,
      "loss": 5.6767,
      "mean_token_accuracy": 0.1824124276638031,
      "num_tokens": 57791315.0,
      "step": 29540
    },
    {
      "entropy": 6.102380228042603,
      "epoch": 2.9533663217873745,
      "grad_norm": 1.234375,
      "learning_rate": 0.000413887269457961,
      "loss": 5.6632,
      "mean_token_accuracy": 0.17987623661756516,
      "num_tokens": 57800225.0,
      "step": 29545
    },
    {
      "entropy": 5.999696063995361,
      "epoch": 2.953866146848603,
      "grad_norm": 1.25,
      "learning_rate": 0.00041385918944972716,
      "loss": 5.5163,
      "mean_token_accuracy": 0.19184273332357407,
      "num_tokens": 57810319.0,
      "step": 29550
    },
    {
      "entropy": 5.929095697402954,
      "epoch": 2.9543659719098314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041383110594783197,
      "loss": 5.4579,
      "mean_token_accuracy": 0.19727450758218765,
      "num_tokens": 57819506.0,
      "step": 29555
    },
    {
      "entropy": 6.019326782226562,
      "epoch": 2.95486579697106,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041380301895298173,
      "loss": 5.6651,
      "mean_token_accuracy": 0.18406865149736404,
      "num_tokens": 57829047.0,
      "step": 29560
    },
    {
      "entropy": 6.049376678466797,
      "epoch": 2.955365622032289,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041377492846588333,
      "loss": 5.6615,
      "mean_token_accuracy": 0.1781827002763748,
      "num_tokens": 57839155.0,
      "step": 29565
    },
    {
      "entropy": 5.991638946533203,
      "epoch": 2.955865447093517,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041374683448724346,
      "loss": 5.5082,
      "mean_token_accuracy": 0.19444656372070312,
      "num_tokens": 57848705.0,
      "step": 29570
    },
    {
      "entropy": 5.9549880027771,
      "epoch": 2.9563652721547458,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004137187370177689,
      "loss": 5.5245,
      "mean_token_accuracy": 0.18695602267980577,
      "num_tokens": 57858231.0,
      "step": 29575
    },
    {
      "entropy": 6.037567090988159,
      "epoch": 2.9568650972159745,
      "grad_norm": 1.359375,
      "learning_rate": 0.00041369063605816667,
      "loss": 5.5952,
      "mean_token_accuracy": 0.18708688616752625,
      "num_tokens": 57868931.0,
      "step": 29580
    },
    {
      "entropy": 5.914687013626098,
      "epoch": 2.957364922277203,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004136625316091437,
      "loss": 5.4897,
      "mean_token_accuracy": 0.19121533036231994,
      "num_tokens": 57879075.0,
      "step": 29585
    },
    {
      "entropy": 6.021975946426392,
      "epoch": 2.9578647473384314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004136344236714071,
      "loss": 5.5571,
      "mean_token_accuracy": 0.18669759035110473,
      "num_tokens": 57889730.0,
      "step": 29590
    },
    {
      "entropy": 6.101756429672241,
      "epoch": 2.95836457239966,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041360631224566407,
      "loss": 5.6513,
      "mean_token_accuracy": 0.18102300763130189,
      "num_tokens": 57900269.0,
      "step": 29595
    },
    {
      "entropy": 6.040577697753906,
      "epoch": 2.958864397460889,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004135781973326219,
      "loss": 5.5532,
      "mean_token_accuracy": 0.1842327043414116,
      "num_tokens": 57909269.0,
      "step": 29600
    },
    {
      "entropy": 5.917732810974121,
      "epoch": 2.959364222522117,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041355007893298796,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1877902179956436,
      "num_tokens": 57918408.0,
      "step": 29605
    },
    {
      "entropy": 5.97173376083374,
      "epoch": 2.9598640475833458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004135219570474697,
      "loss": 5.5633,
      "mean_token_accuracy": 0.1856458455324173,
      "num_tokens": 57928864.0,
      "step": 29610
    },
    {
      "entropy": 6.0679301738739015,
      "epoch": 2.9603638726445745,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041349383167677456,
      "loss": 5.6374,
      "mean_token_accuracy": 0.17675033956766129,
      "num_tokens": 57937998.0,
      "step": 29615
    },
    {
      "entropy": 6.015809631347656,
      "epoch": 2.960863697705803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041346570282161033,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18628222942352296,
      "num_tokens": 57947630.0,
      "step": 29620
    },
    {
      "entropy": 6.067818117141724,
      "epoch": 2.9613635227670314,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041343757048268464,
      "loss": 5.6004,
      "mean_token_accuracy": 0.18684641271829605,
      "num_tokens": 57957013.0,
      "step": 29625
    },
    {
      "entropy": 6.0980079650878904,
      "epoch": 2.96186334782826,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004134094346607053,
      "loss": 5.6372,
      "mean_token_accuracy": 0.17684347480535506,
      "num_tokens": 57967445.0,
      "step": 29630
    },
    {
      "entropy": 6.063826322555542,
      "epoch": 2.962363172889489,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004133812953563802,
      "loss": 5.6911,
      "mean_token_accuracy": 0.1710430145263672,
      "num_tokens": 57976761.0,
      "step": 29635
    },
    {
      "entropy": 6.043433904647827,
      "epoch": 2.962862997950717,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041335315257041733,
      "loss": 5.5419,
      "mean_token_accuracy": 0.18580065965652465,
      "num_tokens": 57986743.0,
      "step": 29640
    },
    {
      "entropy": 6.043831300735474,
      "epoch": 2.9633628230119458,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041332500630352463,
      "loss": 5.6138,
      "mean_token_accuracy": 0.17722252160310745,
      "num_tokens": 57996606.0,
      "step": 29645
    },
    {
      "entropy": 6.020371770858764,
      "epoch": 2.9638626480731745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004132968565564105,
      "loss": 5.5744,
      "mean_token_accuracy": 0.18752964735031127,
      "num_tokens": 58005348.0,
      "step": 29650
    },
    {
      "entropy": 5.989268684387207,
      "epoch": 2.964362473134403,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000413268703329783,
      "loss": 5.5785,
      "mean_token_accuracy": 0.1813996285200119,
      "num_tokens": 58015034.0,
      "step": 29655
    },
    {
      "entropy": 5.996172666549683,
      "epoch": 2.9648622981956314,
      "grad_norm": 1.125,
      "learning_rate": 0.0004132405466243505,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19260846823453903,
      "num_tokens": 58023813.0,
      "step": 29660
    },
    {
      "entropy": 5.994862604141235,
      "epoch": 2.96536212325686,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041321238644082144,
      "loss": 5.5435,
      "mean_token_accuracy": 0.18804031163454055,
      "num_tokens": 58033567.0,
      "step": 29665
    },
    {
      "entropy": 6.039542865753174,
      "epoch": 2.965861948318089,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041318422277990426,
      "loss": 5.5363,
      "mean_token_accuracy": 0.18651091307401657,
      "num_tokens": 58043633.0,
      "step": 29670
    },
    {
      "entropy": 5.961838006973267,
      "epoch": 2.966361773379317,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004131560556423076,
      "loss": 5.4935,
      "mean_token_accuracy": 0.19071186929941178,
      "num_tokens": 58053932.0,
      "step": 29675
    },
    {
      "entropy": 6.033643388748169,
      "epoch": 2.9668615984405458,
      "grad_norm": 1.25,
      "learning_rate": 0.0004131278850287401,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18802286237478255,
      "num_tokens": 58064168.0,
      "step": 29680
    },
    {
      "entropy": 5.957163143157959,
      "epoch": 2.9673614235017745,
      "grad_norm": 1.125,
      "learning_rate": 0.00041309971093991067,
      "loss": 5.5694,
      "mean_token_accuracy": 0.18635204881429673,
      "num_tokens": 58074389.0,
      "step": 29685
    },
    {
      "entropy": 5.935129690170288,
      "epoch": 2.967861248563003,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041307153337652805,
      "loss": 5.4492,
      "mean_token_accuracy": 0.1909039631485939,
      "num_tokens": 58084028.0,
      "step": 29690
    },
    {
      "entropy": 6.017789745330811,
      "epoch": 2.9683610736242314,
      "grad_norm": 1.078125,
      "learning_rate": 0.000413043352339301,
      "loss": 5.6453,
      "mean_token_accuracy": 0.1777293026447296,
      "num_tokens": 58093842.0,
      "step": 29695
    },
    {
      "entropy": 6.087294435501098,
      "epoch": 2.96886089868546,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004130151678289388,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18804396241903304,
      "num_tokens": 58103775.0,
      "step": 29700
    },
    {
      "entropy": 6.022687005996704,
      "epoch": 2.969360723746689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004129869798461506,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18738717436790467,
      "num_tokens": 58113710.0,
      "step": 29705
    },
    {
      "entropy": 6.001804065704346,
      "epoch": 2.969860548807917,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004129587883916454,
      "loss": 5.6186,
      "mean_token_accuracy": 0.1751911610364914,
      "num_tokens": 58123408.0,
      "step": 29710
    },
    {
      "entropy": 6.149905967712402,
      "epoch": 2.9703603738691458,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004129305934661326,
      "loss": 5.6621,
      "mean_token_accuracy": 0.17876077443361282,
      "num_tokens": 58132305.0,
      "step": 29715
    },
    {
      "entropy": 6.024210214614868,
      "epoch": 2.9708601989303745,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041290239507032155,
      "loss": 5.5512,
      "mean_token_accuracy": 0.19074736833572387,
      "num_tokens": 58141655.0,
      "step": 29720
    },
    {
      "entropy": 5.932215213775635,
      "epoch": 2.971360023991603,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004128741932049218,
      "loss": 5.4815,
      "mean_token_accuracy": 0.19577588587999345,
      "num_tokens": 58151495.0,
      "step": 29725
    },
    {
      "entropy": 5.925086784362793,
      "epoch": 2.9718598490528314,
      "grad_norm": 1.25,
      "learning_rate": 0.0004128459878706428,
      "loss": 5.5771,
      "mean_token_accuracy": 0.1758035436272621,
      "num_tokens": 58161388.0,
      "step": 29730
    },
    {
      "entropy": 5.917692470550537,
      "epoch": 2.97235967411406,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004128177790681942,
      "loss": 5.4183,
      "mean_token_accuracy": 0.20507770478725434,
      "num_tokens": 58171563.0,
      "step": 29735
    },
    {
      "entropy": 6.03322467803955,
      "epoch": 2.972859499175289,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041278956679828566,
      "loss": 5.5823,
      "mean_token_accuracy": 0.18827921897172928,
      "num_tokens": 58181943.0,
      "step": 29740
    },
    {
      "entropy": 6.0860052585601805,
      "epoch": 2.973359324236517,
      "grad_norm": 1.125,
      "learning_rate": 0.00041276135106162713,
      "loss": 5.6837,
      "mean_token_accuracy": 0.18151523917913437,
      "num_tokens": 58192199.0,
      "step": 29745
    },
    {
      "entropy": 6.028452682495117,
      "epoch": 2.9738591492977458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004127331318589285,
      "loss": 5.6588,
      "mean_token_accuracy": 0.18039130866527558,
      "num_tokens": 58201612.0,
      "step": 29750
    },
    {
      "entropy": 6.038299179077148,
      "epoch": 2.9743589743589745,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041270490919089966,
      "loss": 5.6154,
      "mean_token_accuracy": 0.18370076566934584,
      "num_tokens": 58211567.0,
      "step": 29755
    },
    {
      "entropy": 6.106947755813598,
      "epoch": 2.974858799420203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004126766830582508,
      "loss": 5.6168,
      "mean_token_accuracy": 0.17853999882936478,
      "num_tokens": 58222739.0,
      "step": 29760
    },
    {
      "entropy": 6.057509517669677,
      "epoch": 2.9753586244814314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000412648453461692,
      "loss": 5.5604,
      "mean_token_accuracy": 0.18401384651660918,
      "num_tokens": 58232269.0,
      "step": 29765
    },
    {
      "entropy": 6.054175281524659,
      "epoch": 2.97585844954266,
      "grad_norm": 1.375,
      "learning_rate": 0.00041262022040193346,
      "loss": 5.6926,
      "mean_token_accuracy": 0.17619024813175202,
      "num_tokens": 58242451.0,
      "step": 29770
    },
    {
      "entropy": 6.0701127529144285,
      "epoch": 2.976358274603889,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041259198387968564,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16901984959840774,
      "num_tokens": 58252202.0,
      "step": 29775
    },
    {
      "entropy": 6.012253046035767,
      "epoch": 2.976858099665117,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041256374389565895,
      "loss": 5.5341,
      "mean_token_accuracy": 0.18820920139551162,
      "num_tokens": 58260864.0,
      "step": 29780
    },
    {
      "entropy": 6.04476318359375,
      "epoch": 2.9773579247263458,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041253550045056375,
      "loss": 5.5497,
      "mean_token_accuracy": 0.188883176445961,
      "num_tokens": 58271636.0,
      "step": 29785
    },
    {
      "entropy": 6.079018068313599,
      "epoch": 2.9778577497875744,
      "grad_norm": 1.375,
      "learning_rate": 0.00041250725354511085,
      "loss": 5.5212,
      "mean_token_accuracy": 0.19210850149393083,
      "num_tokens": 58281792.0,
      "step": 29790
    },
    {
      "entropy": 6.06784143447876,
      "epoch": 2.978357574848803,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004124790031800109,
      "loss": 5.6498,
      "mean_token_accuracy": 0.18399347513914108,
      "num_tokens": 58291528.0,
      "step": 29795
    },
    {
      "entropy": 5.998754215240479,
      "epoch": 2.9788573999100314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004124507493559745,
      "loss": 5.5946,
      "mean_token_accuracy": 0.18766550719738007,
      "num_tokens": 58302159.0,
      "step": 29800
    },
    {
      "entropy": 5.939973783493042,
      "epoch": 2.97935722497126,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004124224920737126,
      "loss": 5.5984,
      "mean_token_accuracy": 0.18666949570178987,
      "num_tokens": 58311621.0,
      "step": 29805
    },
    {
      "entropy": 6.0049824714660645,
      "epoch": 2.9798570500324884,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004123942313339361,
      "loss": 5.5967,
      "mean_token_accuracy": 0.18718411922454833,
      "num_tokens": 58320968.0,
      "step": 29810
    },
    {
      "entropy": 6.0454737663269045,
      "epoch": 2.980356875093717,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041236596713735624,
      "loss": 5.6004,
      "mean_token_accuracy": 0.17859688699245452,
      "num_tokens": 58330092.0,
      "step": 29815
    },
    {
      "entropy": 6.005214786529541,
      "epoch": 2.9808567001549457,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004123376994846839,
      "loss": 5.6761,
      "mean_token_accuracy": 0.17893062382936478,
      "num_tokens": 58339976.0,
      "step": 29820
    },
    {
      "entropy": 5.9774370193481445,
      "epoch": 2.9813565252161744,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004123094283766304,
      "loss": 5.4917,
      "mean_token_accuracy": 0.18945151418447495,
      "num_tokens": 58349343.0,
      "step": 29825
    },
    {
      "entropy": 6.033159065246582,
      "epoch": 2.981856350277403,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041228115381390705,
      "loss": 5.5803,
      "mean_token_accuracy": 0.18418329805135727,
      "num_tokens": 58358858.0,
      "step": 29830
    },
    {
      "entropy": 6.10026593208313,
      "epoch": 2.9823561753386314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041225287579722526,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17555319517850876,
      "num_tokens": 58368532.0,
      "step": 29835
    },
    {
      "entropy": 5.986203145980835,
      "epoch": 2.98285600039986,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004122245943272963,
      "loss": 5.4509,
      "mean_token_accuracy": 0.19303636848926545,
      "num_tokens": 58377997.0,
      "step": 29840
    },
    {
      "entropy": 5.957690715789795,
      "epoch": 2.9833558254610884,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041219630940483187,
      "loss": 5.5417,
      "mean_token_accuracy": 0.19067085832357406,
      "num_tokens": 58387136.0,
      "step": 29845
    },
    {
      "entropy": 6.059955501556397,
      "epoch": 2.983855650522317,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004121680210305436,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18541160374879836,
      "num_tokens": 58396110.0,
      "step": 29850
    },
    {
      "entropy": 6.162730360031128,
      "epoch": 2.9843554755835457,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004121397292051433,
      "loss": 5.7114,
      "mean_token_accuracy": 0.1743092879652977,
      "num_tokens": 58406411.0,
      "step": 29855
    },
    {
      "entropy": 6.049276447296142,
      "epoch": 2.9848553006447744,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004121114339293427,
      "loss": 5.5742,
      "mean_token_accuracy": 0.18447191566228865,
      "num_tokens": 58416172.0,
      "step": 29860
    },
    {
      "entropy": 5.966488075256348,
      "epoch": 2.985355125706003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004120831352038537,
      "loss": 5.4911,
      "mean_token_accuracy": 0.18953044712543488,
      "num_tokens": 58426245.0,
      "step": 29865
    },
    {
      "entropy": 6.003190803527832,
      "epoch": 2.9858549507672314,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004120548330293882,
      "loss": 5.5712,
      "mean_token_accuracy": 0.1826629802584648,
      "num_tokens": 58435796.0,
      "step": 29870
    },
    {
      "entropy": 6.065922975540161,
      "epoch": 2.98635477582846,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00041202652740665845,
      "loss": 5.6463,
      "mean_token_accuracy": 0.17436708360910416,
      "num_tokens": 58445360.0,
      "step": 29875
    },
    {
      "entropy": 6.075990486145019,
      "epoch": 2.9868546008896883,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004119982183363765,
      "loss": 5.6198,
      "mean_token_accuracy": 0.1826986104249954,
      "num_tokens": 58453665.0,
      "step": 29880
    },
    {
      "entropy": 6.029713869094849,
      "epoch": 2.987354425950917,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041196990581925466,
      "loss": 5.5946,
      "mean_token_accuracy": 0.17953983694314957,
      "num_tokens": 58462121.0,
      "step": 29885
    },
    {
      "entropy": 5.998000955581665,
      "epoch": 2.9878542510121457,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004119415898560052,
      "loss": 5.5773,
      "mean_token_accuracy": 0.1763468712568283,
      "num_tokens": 58471678.0,
      "step": 29890
    },
    {
      "entropy": 6.127283334732056,
      "epoch": 2.9883540760733744,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004119132704473405,
      "loss": 5.6488,
      "mean_token_accuracy": 0.17526998072862626,
      "num_tokens": 58481140.0,
      "step": 29895
    },
    {
      "entropy": 6.1273415088653564,
      "epoch": 2.988853901134603,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004118849475939733,
      "loss": 5.6368,
      "mean_token_accuracy": 0.17847371250391006,
      "num_tokens": 58491163.0,
      "step": 29900
    },
    {
      "entropy": 6.002063512802124,
      "epoch": 2.9893537261958314,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041185662129661593,
      "loss": 5.598,
      "mean_token_accuracy": 0.1793772354722023,
      "num_tokens": 58500433.0,
      "step": 29905
    },
    {
      "entropy": 6.054016494750977,
      "epoch": 2.98985355125706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004118282915559813,
      "loss": 5.6081,
      "mean_token_accuracy": 0.1764826312661171,
      "num_tokens": 58510618.0,
      "step": 29910
    },
    {
      "entropy": 6.0296814918518065,
      "epoch": 2.9903533763182883,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004117999583727819,
      "loss": 5.5238,
      "mean_token_accuracy": 0.18426795303821564,
      "num_tokens": 58519658.0,
      "step": 29915
    },
    {
      "entropy": 6.0040544986724855,
      "epoch": 2.990853201379517,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041177162174773084,
      "loss": 5.4904,
      "mean_token_accuracy": 0.19149068593978882,
      "num_tokens": 58530102.0,
      "step": 29920
    },
    {
      "entropy": 6.057943820953369,
      "epoch": 2.9913530264407457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041174328168154094,
      "loss": 5.6009,
      "mean_token_accuracy": 0.18007841110229492,
      "num_tokens": 58539300.0,
      "step": 29925
    },
    {
      "entropy": 5.966517066955566,
      "epoch": 2.9918528515019744,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004117149381749253,
      "loss": 5.5964,
      "mean_token_accuracy": 0.17788388282060624,
      "num_tokens": 58548592.0,
      "step": 29930
    },
    {
      "entropy": 6.020139169692993,
      "epoch": 2.992352676563203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004116865912285971,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18424477130174638,
      "num_tokens": 58559335.0,
      "step": 29935
    },
    {
      "entropy": 6.063624095916748,
      "epoch": 2.9928525016244314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041165824084326923,
      "loss": 5.5176,
      "mean_token_accuracy": 0.18739703744649888,
      "num_tokens": 58568470.0,
      "step": 29940
    },
    {
      "entropy": 5.95731258392334,
      "epoch": 2.99335232668566,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004116298870196553,
      "loss": 5.4624,
      "mean_token_accuracy": 0.19610629230737686,
      "num_tokens": 58577514.0,
      "step": 29945
    },
    {
      "entropy": 6.013263368606568,
      "epoch": 2.9938521517468883,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00041160152975846853,
      "loss": 5.5273,
      "mean_token_accuracy": 0.1894572675228119,
      "num_tokens": 58587241.0,
      "step": 29950
    },
    {
      "entropy": 6.0325035572052,
      "epoch": 2.994351976808117,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004115731690604225,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1827186718583107,
      "num_tokens": 58596923.0,
      "step": 29955
    },
    {
      "entropy": 6.029146909713745,
      "epoch": 2.9948518018693457,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004115448049262306,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1883434534072876,
      "num_tokens": 58607316.0,
      "step": 29960
    },
    {
      "entropy": 5.95598030090332,
      "epoch": 2.9953516269305744,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004115164373566065,
      "loss": 5.4955,
      "mean_token_accuracy": 0.19682665467262267,
      "num_tokens": 58616601.0,
      "step": 29965
    },
    {
      "entropy": 6.017064809799194,
      "epoch": 2.995851451991803,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041148806635226407,
      "loss": 5.6581,
      "mean_token_accuracy": 0.18383354246616362,
      "num_tokens": 58628094.0,
      "step": 29970
    },
    {
      "entropy": 6.051974439620972,
      "epoch": 2.9963512770530314,
      "grad_norm": 1.09375,
      "learning_rate": 0.000411459691913917,
      "loss": 5.6036,
      "mean_token_accuracy": 0.18051334619522094,
      "num_tokens": 58638005.0,
      "step": 29975
    },
    {
      "entropy": 6.02429895401001,
      "epoch": 2.99685110211426,
      "grad_norm": 1.125,
      "learning_rate": 0.00041143131404227913,
      "loss": 5.6037,
      "mean_token_accuracy": 0.18025618493556977,
      "num_tokens": 58648552.0,
      "step": 29980
    },
    {
      "entropy": 5.983723068237305,
      "epoch": 2.9973509271754883,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004114029327380645,
      "loss": 5.4985,
      "mean_token_accuracy": 0.19326425045728685,
      "num_tokens": 58657661.0,
      "step": 29985
    },
    {
      "entropy": 6.05328049659729,
      "epoch": 2.997850752236717,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004113745480019872,
      "loss": 5.6652,
      "mean_token_accuracy": 0.18217990696430206,
      "num_tokens": 58667553.0,
      "step": 29990
    },
    {
      "entropy": 6.134248971939087,
      "epoch": 2.9983505772979457,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004113461598347614,
      "loss": 5.7105,
      "mean_token_accuracy": 0.1779572546482086,
      "num_tokens": 58677641.0,
      "step": 29995
    },
    {
      "entropy": 6.041600561141967,
      "epoch": 2.9988504023591744,
      "grad_norm": 1.6015625,
      "learning_rate": 0.00041131776823710134,
      "loss": 5.5667,
      "mean_token_accuracy": 0.18589375168085098,
      "num_tokens": 58687729.0,
      "step": 30000
    },
    {
      "epoch": 2.9988504023591744,
      "eval_entropy": 5.85013244822295,
      "eval_loss": 5.682925224304199,
      "eval_mean_token_accuracy": 0.1872464606768306,
      "eval_num_tokens": 58687729.0,
      "eval_runtime": 24.1176,
      "eval_samples_per_second": 1287.233,
      "eval_steps_per_second": 160.92,
      "step": 30000
    },
    {
      "entropy": 6.015658807754517,
      "epoch": 2.9993502274204027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004112893732097213,
      "loss": 5.5313,
      "mean_token_accuracy": 0.19306307286024094,
      "num_tokens": 58697164.0,
      "step": 30005
    },
    {
      "entropy": 5.928306627273559,
      "epoch": 2.9998500524816314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004112609747533356,
      "loss": 5.479,
      "mean_token_accuracy": 0.19978833347558975,
      "num_tokens": 58705702.0,
      "step": 30010
    },
    {
      "entropy": 6.011835098266602,
      "epoch": 3.000299895036737,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004112325728686589,
      "loss": 5.5754,
      "mean_token_accuracy": 0.18248557878865135,
      "num_tokens": 58714408.0,
      "step": 30015
    },
    {
      "entropy": 6.010691356658936,
      "epoch": 3.000799720097966,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041120416755640574,
      "loss": 5.4989,
      "mean_token_accuracy": 0.19321251064538955,
      "num_tokens": 58724185.0,
      "step": 30020
    },
    {
      "entropy": 5.997696828842163,
      "epoch": 3.001299545159194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041117575881729075,
      "loss": 5.4052,
      "mean_token_accuracy": 0.1977203831076622,
      "num_tokens": 58733998.0,
      "step": 30025
    },
    {
      "entropy": 6.00664324760437,
      "epoch": 3.001799370220423,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041114734665202875,
      "loss": 5.5178,
      "mean_token_accuracy": 0.18863451182842256,
      "num_tokens": 58743881.0,
      "step": 30030
    },
    {
      "entropy": 6.021720457077026,
      "epoch": 3.0022991952816516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004111189310613344,
      "loss": 5.56,
      "mean_token_accuracy": 0.18617595732212067,
      "num_tokens": 58754106.0,
      "step": 30035
    },
    {
      "entropy": 6.09415488243103,
      "epoch": 3.00279902034288,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041109051204592294,
      "loss": 5.5176,
      "mean_token_accuracy": 0.1876286029815674,
      "num_tokens": 58763830.0,
      "step": 30040
    },
    {
      "entropy": 6.001323175430298,
      "epoch": 3.0032988454041085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041106208960650913,
      "loss": 5.4775,
      "mean_token_accuracy": 0.1971851885318756,
      "num_tokens": 58773478.0,
      "step": 30045
    },
    {
      "entropy": 5.958635139465332,
      "epoch": 3.003798670465337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041103366374380817,
      "loss": 5.5247,
      "mean_token_accuracy": 0.19465427845716476,
      "num_tokens": 58784529.0,
      "step": 30050
    },
    {
      "entropy": 6.048456621170044,
      "epoch": 3.004298495526566,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041100523445853525,
      "loss": 5.6245,
      "mean_token_accuracy": 0.1840030297636986,
      "num_tokens": 58794087.0,
      "step": 30055
    },
    {
      "entropy": 5.880138206481933,
      "epoch": 3.004798320587794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004109768017514056,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2096722364425659,
      "num_tokens": 58804350.0,
      "step": 30060
    },
    {
      "entropy": 6.007662725448609,
      "epoch": 3.005298145649023,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004109483656231346,
      "loss": 5.5836,
      "mean_token_accuracy": 0.18172685652971268,
      "num_tokens": 58814560.0,
      "step": 30065
    },
    {
      "entropy": 6.028471565246582,
      "epoch": 3.0057979707102516,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041091992607443777,
      "loss": 5.5759,
      "mean_token_accuracy": 0.17819382548332213,
      "num_tokens": 58825016.0,
      "step": 30070
    },
    {
      "entropy": 5.919505023956299,
      "epoch": 3.00629779577148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041089148310603053,
      "loss": 5.3779,
      "mean_token_accuracy": 0.19917938709259034,
      "num_tokens": 58834673.0,
      "step": 30075
    },
    {
      "entropy": 5.98980565071106,
      "epoch": 3.0067976208327085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004108630367186285,
      "loss": 5.392,
      "mean_token_accuracy": 0.19573962539434434,
      "num_tokens": 58843461.0,
      "step": 30080
    },
    {
      "entropy": 6.011548280715942,
      "epoch": 3.007297445893937,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041083458691294755,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18904543220996856,
      "num_tokens": 58853048.0,
      "step": 30085
    },
    {
      "entropy": 5.938202905654907,
      "epoch": 3.007797270955166,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004108061336897032,
      "loss": 5.4823,
      "mean_token_accuracy": 0.18500783890485764,
      "num_tokens": 58863234.0,
      "step": 30090
    },
    {
      "entropy": 6.007377815246582,
      "epoch": 3.008297096016394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004107776770496116,
      "loss": 5.5087,
      "mean_token_accuracy": 0.18943593353033067,
      "num_tokens": 58871503.0,
      "step": 30095
    },
    {
      "entropy": 5.987948656082153,
      "epoch": 3.008796921077623,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004107492169933885,
      "loss": 5.4634,
      "mean_token_accuracy": 0.19837216883897782,
      "num_tokens": 58880660.0,
      "step": 30100
    },
    {
      "entropy": 6.006690645217896,
      "epoch": 3.0092967461388516,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041072075352175004,
      "loss": 5.4823,
      "mean_token_accuracy": 0.19671361595392228,
      "num_tokens": 58889563.0,
      "step": 30105
    },
    {
      "entropy": 5.921568870544434,
      "epoch": 3.00979657120008,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00041069228663541234,
      "loss": 5.4176,
      "mean_token_accuracy": 0.19413175582885742,
      "num_tokens": 58900139.0,
      "step": 30110
    },
    {
      "entropy": 5.869332504272461,
      "epoch": 3.0102963962613085,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004106638163350917,
      "loss": 5.3516,
      "mean_token_accuracy": 0.2056186765432358,
      "num_tokens": 58909644.0,
      "step": 30115
    },
    {
      "entropy": 5.988032674789428,
      "epoch": 3.010796221322537,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004106353426215043,
      "loss": 5.5128,
      "mean_token_accuracy": 0.1809345930814743,
      "num_tokens": 58919998.0,
      "step": 30120
    },
    {
      "entropy": 6.040743064880371,
      "epoch": 3.0112960463837655,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041060686549536656,
      "loss": 5.4907,
      "mean_token_accuracy": 0.18818971812725066,
      "num_tokens": 58928447.0,
      "step": 30125
    },
    {
      "entropy": 6.054534769058227,
      "epoch": 3.011795871444994,
      "grad_norm": 1.09375,
      "learning_rate": 0.000410578384957395,
      "loss": 5.5615,
      "mean_token_accuracy": 0.17826468944549562,
      "num_tokens": 58939410.0,
      "step": 30130
    },
    {
      "entropy": 6.0097884178161625,
      "epoch": 3.012295696506223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004105499010083062,
      "loss": 5.5398,
      "mean_token_accuracy": 0.1800971120595932,
      "num_tokens": 58949672.0,
      "step": 30135
    },
    {
      "entropy": 5.879815673828125,
      "epoch": 3.0127955215674516,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041052141364881665,
      "loss": 5.3511,
      "mean_token_accuracy": 0.20648044794797898,
      "num_tokens": 58960034.0,
      "step": 30140
    },
    {
      "entropy": 6.020365953445435,
      "epoch": 3.01329534662868,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041049292287964335,
      "loss": 5.4116,
      "mean_token_accuracy": 0.19209861606359482,
      "num_tokens": 58969770.0,
      "step": 30145
    },
    {
      "entropy": 5.968260812759399,
      "epoch": 3.0137951716899085,
      "grad_norm": 1.25,
      "learning_rate": 0.00041046442870150293,
      "loss": 5.4222,
      "mean_token_accuracy": 0.19586141705513,
      "num_tokens": 58979638.0,
      "step": 30150
    },
    {
      "entropy": 5.963473463058472,
      "epoch": 3.014294996751137,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041043593111511227,
      "loss": 5.4069,
      "mean_token_accuracy": 0.19369567781686783,
      "num_tokens": 58988566.0,
      "step": 30155
    },
    {
      "entropy": 5.983544158935547,
      "epoch": 3.0147948218123655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004104074301211885,
      "loss": 5.5016,
      "mean_token_accuracy": 0.18835275769233703,
      "num_tokens": 58998354.0,
      "step": 30160
    },
    {
      "entropy": 6.056784343719483,
      "epoch": 3.015294646873594,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041037892572044863,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1837203711271286,
      "num_tokens": 59007310.0,
      "step": 30165
    },
    {
      "entropy": 6.042945098876953,
      "epoch": 3.015794471934823,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041035041791360983,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18208677172660828,
      "num_tokens": 59017145.0,
      "step": 30170
    },
    {
      "entropy": 6.0080421447753904,
      "epoch": 3.0162942969960516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004103219067013893,
      "loss": 5.459,
      "mean_token_accuracy": 0.18673839569091796,
      "num_tokens": 59026941.0,
      "step": 30175
    },
    {
      "entropy": 5.967184972763062,
      "epoch": 3.01679412205728,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041029339208450445,
      "loss": 5.4462,
      "mean_token_accuracy": 0.19846125692129135,
      "num_tokens": 59036343.0,
      "step": 30180
    },
    {
      "entropy": 5.948143672943115,
      "epoch": 3.0172939471185085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004102648740636727,
      "loss": 5.4459,
      "mean_token_accuracy": 0.1905348241329193,
      "num_tokens": 59046601.0,
      "step": 30185
    },
    {
      "entropy": 6.003769826889038,
      "epoch": 3.017793772179737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004102363526396115,
      "loss": 5.5486,
      "mean_token_accuracy": 0.18116901367902755,
      "num_tokens": 59055819.0,
      "step": 30190
    },
    {
      "entropy": 5.924482583999634,
      "epoch": 3.0182935972409655,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004102078278130384,
      "loss": 5.4741,
      "mean_token_accuracy": 0.1946369156241417,
      "num_tokens": 59064956.0,
      "step": 30195
    },
    {
      "entropy": 5.943385934829712,
      "epoch": 3.018793422302194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041017929958467113,
      "loss": 5.3494,
      "mean_token_accuracy": 0.2005409777164459,
      "num_tokens": 59073842.0,
      "step": 30200
    },
    {
      "entropy": 6.039760303497315,
      "epoch": 3.019293247363423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004101507679552276,
      "loss": 5.5238,
      "mean_token_accuracy": 0.18748033195734023,
      "num_tokens": 59084946.0,
      "step": 30205
    },
    {
      "entropy": 6.057115745544434,
      "epoch": 3.0197930724246516,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004101222329254255,
      "loss": 5.5436,
      "mean_token_accuracy": 0.19005635380744934,
      "num_tokens": 59095142.0,
      "step": 30210
    },
    {
      "entropy": 5.989990901947022,
      "epoch": 3.02029289748588,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004100936944959828,
      "loss": 5.5062,
      "mean_token_accuracy": 0.18814261257648468,
      "num_tokens": 59104618.0,
      "step": 30215
    },
    {
      "entropy": 5.973377084732055,
      "epoch": 3.0207927225471085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004100651526676174,
      "loss": 5.5314,
      "mean_token_accuracy": 0.1902268961071968,
      "num_tokens": 59114485.0,
      "step": 30220
    },
    {
      "entropy": 6.067344427108765,
      "epoch": 3.021292547608337,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004100366074410476,
      "loss": 5.5884,
      "mean_token_accuracy": 0.17738519012928008,
      "num_tokens": 59124866.0,
      "step": 30225
    },
    {
      "entropy": 5.9446063995361325,
      "epoch": 3.0217923726695655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00041000805881699143,
      "loss": 5.3884,
      "mean_token_accuracy": 0.19874099791049957,
      "num_tokens": 59134123.0,
      "step": 30230
    },
    {
      "entropy": 6.002497863769531,
      "epoch": 3.022292197730794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004099795067961673,
      "loss": 5.539,
      "mean_token_accuracy": 0.1853841006755829,
      "num_tokens": 59144375.0,
      "step": 30235
    },
    {
      "entropy": 6.065521669387818,
      "epoch": 3.022792022792023,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004099509513792935,
      "loss": 5.5056,
      "mean_token_accuracy": 0.18841976821422576,
      "num_tokens": 59154185.0,
      "step": 30240
    },
    {
      "entropy": 6.083507013320923,
      "epoch": 3.0232918478532516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040992239256708844,
      "loss": 5.5841,
      "mean_token_accuracy": 0.17838609963655472,
      "num_tokens": 59164501.0,
      "step": 30245
    },
    {
      "entropy": 5.992962312698364,
      "epoch": 3.02379167291448,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004098938303602708,
      "loss": 5.4548,
      "mean_token_accuracy": 0.19509146362543106,
      "num_tokens": 59174349.0,
      "step": 30250
    },
    {
      "entropy": 6.012611627578735,
      "epoch": 3.0242914979757085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040986526475955904,
      "loss": 5.5085,
      "mean_token_accuracy": 0.19829973727464675,
      "num_tokens": 59183532.0,
      "step": 30255
    },
    {
      "entropy": 5.994039535522461,
      "epoch": 3.024791323036937,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000409836695765672,
      "loss": 5.4793,
      "mean_token_accuracy": 0.19091457426548003,
      "num_tokens": 59192492.0,
      "step": 30260
    },
    {
      "entropy": 6.0166768550872805,
      "epoch": 3.0252911480981655,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040980812337932835,
      "loss": 5.4942,
      "mean_token_accuracy": 0.1836957201361656,
      "num_tokens": 59203237.0,
      "step": 30265
    },
    {
      "entropy": 5.944023275375367,
      "epoch": 3.025790973159394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040977954760124693,
      "loss": 5.4213,
      "mean_token_accuracy": 0.1862822487950325,
      "num_tokens": 59213286.0,
      "step": 30270
    },
    {
      "entropy": 6.0402122974395756,
      "epoch": 3.026290798220623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040975096843214677,
      "loss": 5.5184,
      "mean_token_accuracy": 0.17668727040290833,
      "num_tokens": 59223211.0,
      "step": 30275
    },
    {
      "entropy": 5.987838172912598,
      "epoch": 3.0267906232818516,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040972238587274695,
      "loss": 5.5372,
      "mean_token_accuracy": 0.19037563353776932,
      "num_tokens": 59232702.0,
      "step": 30280
    },
    {
      "entropy": 6.013862991333008,
      "epoch": 3.02729044834308,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040969379992376664,
      "loss": 5.5735,
      "mean_token_accuracy": 0.18724547028541566,
      "num_tokens": 59242357.0,
      "step": 30285
    },
    {
      "entropy": 6.034306764602661,
      "epoch": 3.0277902734043085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004096652105859249,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1904923900961876,
      "num_tokens": 59250785.0,
      "step": 30290
    },
    {
      "entropy": 6.07140588760376,
      "epoch": 3.028290098465537,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004096366178599412,
      "loss": 5.5372,
      "mean_token_accuracy": 0.17956548929214478,
      "num_tokens": 59260264.0,
      "step": 30295
    },
    {
      "entropy": 6.007885551452636,
      "epoch": 3.0287899235267655,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004096080217465347,
      "loss": 5.5207,
      "mean_token_accuracy": 0.19402726590633393,
      "num_tokens": 59269603.0,
      "step": 30300
    },
    {
      "entropy": 6.046964120864868,
      "epoch": 3.029289748587994,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095794222464252,
      "loss": 5.5324,
      "mean_token_accuracy": 0.19080874025821687,
      "num_tokens": 59279255.0,
      "step": 30305
    },
    {
      "entropy": 6.066649580001831,
      "epoch": 3.029789573649223,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004095508193603319,
      "loss": 5.5807,
      "mean_token_accuracy": 0.18310682326555253,
      "num_tokens": 59288650.0,
      "step": 30310
    },
    {
      "entropy": 5.9321277141571045,
      "epoch": 3.0302893987104516,
      "grad_norm": 1.125,
      "learning_rate": 0.0004095222130889747,
      "loss": 5.504,
      "mean_token_accuracy": 0.18964519053697587,
      "num_tokens": 59298032.0,
      "step": 30315
    },
    {
      "entropy": 6.057075929641724,
      "epoch": 3.03078922377168,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040949360343307324,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18789344877004624,
      "num_tokens": 59307147.0,
      "step": 30320
    },
    {
      "entropy": 6.132550954818726,
      "epoch": 3.0312890488329085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004094649903933472,
      "loss": 5.5356,
      "mean_token_accuracy": 0.17598785012960433,
      "num_tokens": 59316602.0,
      "step": 30325
    },
    {
      "entropy": 5.95460467338562,
      "epoch": 3.031788873894137,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040943637397051683,
      "loss": 5.4771,
      "mean_token_accuracy": 0.19512465298175813,
      "num_tokens": 59326176.0,
      "step": 30330
    },
    {
      "entropy": 5.99467887878418,
      "epoch": 3.0322886989553655,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040940775416530164,
      "loss": 5.5497,
      "mean_token_accuracy": 0.18098105043172835,
      "num_tokens": 59335676.0,
      "step": 30335
    },
    {
      "entropy": 5.980919075012207,
      "epoch": 3.032788524016594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040937913097842204,
      "loss": 5.4068,
      "mean_token_accuracy": 0.20700278282165527,
      "num_tokens": 59345807.0,
      "step": 30340
    },
    {
      "entropy": 5.98985481262207,
      "epoch": 3.033288349077823,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004093505044105982,
      "loss": 5.4246,
      "mean_token_accuracy": 0.1930920213460922,
      "num_tokens": 59355197.0,
      "step": 30345
    },
    {
      "entropy": 6.024800968170166,
      "epoch": 3.033788174139051,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040932187446255004,
      "loss": 5.5897,
      "mean_token_accuracy": 0.19165802597999573,
      "num_tokens": 59364498.0,
      "step": 30350
    },
    {
      "entropy": 5.964720344543457,
      "epoch": 3.03428799920028,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004092932411349982,
      "loss": 5.4312,
      "mean_token_accuracy": 0.1938937783241272,
      "num_tokens": 59373541.0,
      "step": 30355
    },
    {
      "entropy": 6.050469636917114,
      "epoch": 3.0347878242615085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000409264604428663,
      "loss": 5.5966,
      "mean_token_accuracy": 0.1833668440580368,
      "num_tokens": 59384743.0,
      "step": 30360
    },
    {
      "entropy": 6.00672926902771,
      "epoch": 3.035287649322737,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004092359643442648,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18576920479536058,
      "num_tokens": 59394173.0,
      "step": 30365
    },
    {
      "entropy": 5.963884019851685,
      "epoch": 3.0357874743839655,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004092073208825243,
      "loss": 5.5378,
      "mean_token_accuracy": 0.18460131585597991,
      "num_tokens": 59403760.0,
      "step": 30370
    },
    {
      "entropy": 5.944233989715576,
      "epoch": 3.036287299445194,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004091786740441622,
      "loss": 5.4514,
      "mean_token_accuracy": 0.19174521714448928,
      "num_tokens": 59413008.0,
      "step": 30375
    },
    {
      "entropy": 6.035178995132446,
      "epoch": 3.036787124506423,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040915002382989915,
      "loss": 5.4772,
      "mean_token_accuracy": 0.19309130907058716,
      "num_tokens": 59423668.0,
      "step": 30380
    },
    {
      "entropy": 5.9955707550048825,
      "epoch": 3.037286949567651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000409121370240456,
      "loss": 5.4996,
      "mean_token_accuracy": 0.1920349717140198,
      "num_tokens": 59433317.0,
      "step": 30385
    },
    {
      "entropy": 6.0813109397888185,
      "epoch": 3.03778677462888,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004090927132765537,
      "loss": 5.5352,
      "mean_token_accuracy": 0.18642453402280806,
      "num_tokens": 59443608.0,
      "step": 30390
    },
    {
      "entropy": 5.952234601974487,
      "epoch": 3.0382865996901085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040906405293891323,
      "loss": 5.4513,
      "mean_token_accuracy": 0.1927007704973221,
      "num_tokens": 59453475.0,
      "step": 30395
    },
    {
      "entropy": 6.00013542175293,
      "epoch": 3.038786424751337,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004090353892282557,
      "loss": 5.5997,
      "mean_token_accuracy": 0.18141825944185258,
      "num_tokens": 59463415.0,
      "step": 30400
    },
    {
      "entropy": 5.9791642189025875,
      "epoch": 3.0392862498125655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004090067221453024,
      "loss": 5.4099,
      "mean_token_accuracy": 0.19573769271373748,
      "num_tokens": 59472787.0,
      "step": 30405
    },
    {
      "entropy": 6.0104701042175295,
      "epoch": 3.039786074873794,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004089780516907743,
      "loss": 5.4014,
      "mean_token_accuracy": 0.20049319565296173,
      "num_tokens": 59482698.0,
      "step": 30410
    },
    {
      "entropy": 6.03032660484314,
      "epoch": 3.040285899935023,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040894937786539287,
      "loss": 5.5444,
      "mean_token_accuracy": 0.18643111735582352,
      "num_tokens": 59492400.0,
      "step": 30415
    },
    {
      "entropy": 6.060033845901489,
      "epoch": 3.040785724996251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040892070066987966,
      "loss": 5.616,
      "mean_token_accuracy": 0.18465495407581328,
      "num_tokens": 59503635.0,
      "step": 30420
    },
    {
      "entropy": 5.940070629119873,
      "epoch": 3.04128555005748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040889202010495595,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20041150599718094,
      "num_tokens": 59513550.0,
      "step": 30425
    },
    {
      "entropy": 6.062500476837158,
      "epoch": 3.0417853751187085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004088633361713436,
      "loss": 5.5886,
      "mean_token_accuracy": 0.18753129690885545,
      "num_tokens": 59523286.0,
      "step": 30430
    },
    {
      "entropy": 5.958077001571655,
      "epoch": 3.042285200179937,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00040883464886976414,
      "loss": 5.5263,
      "mean_token_accuracy": 0.18338072896003724,
      "num_tokens": 59533976.0,
      "step": 30435
    },
    {
      "entropy": 5.971227836608887,
      "epoch": 3.0427850252411655,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004088059582009393,
      "loss": 5.456,
      "mean_token_accuracy": 0.20291794389486312,
      "num_tokens": 59542796.0,
      "step": 30440
    },
    {
      "entropy": 6.093722105026245,
      "epoch": 3.043284850302394,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000408777264165591,
      "loss": 5.6345,
      "mean_token_accuracy": 0.17625984996557237,
      "num_tokens": 59553213.0,
      "step": 30445
    },
    {
      "entropy": 5.972112083435059,
      "epoch": 3.043784675363623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040874856676444116,
      "loss": 5.4636,
      "mean_token_accuracy": 0.18932078778743744,
      "num_tokens": 59562136.0,
      "step": 30450
    },
    {
      "entropy": 6.037386417388916,
      "epoch": 3.044284500424851,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040871986599821177,
      "loss": 5.5927,
      "mean_token_accuracy": 0.18113817870616913,
      "num_tokens": 59572578.0,
      "step": 30455
    },
    {
      "entropy": 6.097305727005005,
      "epoch": 3.04478432548608,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004086911618676251,
      "loss": 5.5453,
      "mean_token_accuracy": 0.18488506823778153,
      "num_tokens": 59581976.0,
      "step": 30460
    },
    {
      "entropy": 6.005220651626587,
      "epoch": 3.0452841505473085,
      "grad_norm": 1.171875,
      "learning_rate": 0.000408662454373403,
      "loss": 5.4938,
      "mean_token_accuracy": 0.19133381098508834,
      "num_tokens": 59591879.0,
      "step": 30465
    },
    {
      "entropy": 6.02236647605896,
      "epoch": 3.045783975608537,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000408633743516268,
      "loss": 5.5669,
      "mean_token_accuracy": 0.18792663067579268,
      "num_tokens": 59601534.0,
      "step": 30470
    },
    {
      "entropy": 6.002684926986694,
      "epoch": 3.0462838006697655,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040860502929694246,
      "loss": 5.4916,
      "mean_token_accuracy": 0.1891309767961502,
      "num_tokens": 59611349.0,
      "step": 30475
    },
    {
      "entropy": 6.012878179550171,
      "epoch": 3.046783625730994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040857631171614873,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18597723990678788,
      "num_tokens": 59621743.0,
      "step": 30480
    },
    {
      "entropy": 6.064972066879273,
      "epoch": 3.047283450792223,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040854759077460934,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18933886140584946,
      "num_tokens": 59630967.0,
      "step": 30485
    },
    {
      "entropy": 6.0315381526947025,
      "epoch": 3.047783275853451,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040851886647304703,
      "loss": 5.4354,
      "mean_token_accuracy": 0.19036571234464644,
      "num_tokens": 59640388.0,
      "step": 30490
    },
    {
      "entropy": 5.933514261245728,
      "epoch": 3.04828310091468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004084901388121843,
      "loss": 5.4873,
      "mean_token_accuracy": 0.18643026947975158,
      "num_tokens": 59649720.0,
      "step": 30495
    },
    {
      "entropy": 5.901197338104248,
      "epoch": 3.0487829259759085,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040846140779274406,
      "loss": 5.4587,
      "mean_token_accuracy": 0.19283869117498398,
      "num_tokens": 59659336.0,
      "step": 30500
    },
    {
      "entropy": 6.060169887542725,
      "epoch": 3.049282751037137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040843267341544913,
      "loss": 5.5795,
      "mean_token_accuracy": 0.17926639467477798,
      "num_tokens": 59669376.0,
      "step": 30505
    },
    {
      "entropy": 6.019588327407837,
      "epoch": 3.0497825760983654,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040840393568102246,
      "loss": 5.4764,
      "mean_token_accuracy": 0.1872422769665718,
      "num_tokens": 59679180.0,
      "step": 30510
    },
    {
      "entropy": 5.965311479568482,
      "epoch": 3.050282401159594,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004083751945901871,
      "loss": 5.483,
      "mean_token_accuracy": 0.1956225663423538,
      "num_tokens": 59689427.0,
      "step": 30515
    },
    {
      "entropy": 5.956658220291137,
      "epoch": 3.050782226220823,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040834645014366623,
      "loss": 5.4384,
      "mean_token_accuracy": 0.19424837082624435,
      "num_tokens": 59698586.0,
      "step": 30520
    },
    {
      "entropy": 5.94865837097168,
      "epoch": 3.051282051282051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004083177023421829,
      "loss": 5.499,
      "mean_token_accuracy": 0.19051860123872758,
      "num_tokens": 59707927.0,
      "step": 30525
    },
    {
      "entropy": 5.968172073364258,
      "epoch": 3.05178187634328,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004082889511864606,
      "loss": 5.5574,
      "mean_token_accuracy": 0.18771989792585372,
      "num_tokens": 59718563.0,
      "step": 30530
    },
    {
      "entropy": 5.938368844985962,
      "epoch": 3.0522817014045085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040826019667722244,
      "loss": 5.2802,
      "mean_token_accuracy": 0.20332809239625932,
      "num_tokens": 59728718.0,
      "step": 30535
    },
    {
      "entropy": 5.97534589767456,
      "epoch": 3.052781526465737,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004082314388151921,
      "loss": 5.4603,
      "mean_token_accuracy": 0.19313037693500518,
      "num_tokens": 59738515.0,
      "step": 30540
    },
    {
      "entropy": 5.984713506698609,
      "epoch": 3.0532813515269654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004082026776010931,
      "loss": 5.5712,
      "mean_token_accuracy": 0.18485296070575713,
      "num_tokens": 59748606.0,
      "step": 30545
    },
    {
      "entropy": 5.96008267402649,
      "epoch": 3.053781176588194,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004081739130356489,
      "loss": 5.4788,
      "mean_token_accuracy": 0.19312864243984224,
      "num_tokens": 59757454.0,
      "step": 30550
    },
    {
      "entropy": 6.015268230438233,
      "epoch": 3.054281001649423,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040814514511958335,
      "loss": 5.4711,
      "mean_token_accuracy": 0.18797480762004853,
      "num_tokens": 59767500.0,
      "step": 30555
    },
    {
      "entropy": 6.004692459106446,
      "epoch": 3.054780826710651,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040811637385362024,
      "loss": 5.5056,
      "mean_token_accuracy": 0.1933903753757477,
      "num_tokens": 59776663.0,
      "step": 30560
    },
    {
      "entropy": 5.9625856399536135,
      "epoch": 3.05528065177188,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004080875992384834,
      "loss": 5.4491,
      "mean_token_accuracy": 0.18679282516241075,
      "num_tokens": 59787075.0,
      "step": 30565
    },
    {
      "entropy": 5.9242504119873045,
      "epoch": 3.0557804768331085,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004080588212748968,
      "loss": 5.3864,
      "mean_token_accuracy": 0.20360741913318633,
      "num_tokens": 59796385.0,
      "step": 30570
    },
    {
      "entropy": 5.917899227142334,
      "epoch": 3.056280301894337,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040803003996358445,
      "loss": 5.4304,
      "mean_token_accuracy": 0.20022509694099427,
      "num_tokens": 59805623.0,
      "step": 30575
    },
    {
      "entropy": 5.952084064483643,
      "epoch": 3.0567801269555654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040800125530527064,
      "loss": 5.5493,
      "mean_token_accuracy": 0.18147781640291213,
      "num_tokens": 59815041.0,
      "step": 30580
    },
    {
      "entropy": 5.94607138633728,
      "epoch": 3.057279952016794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040797246730067937,
      "loss": 5.4107,
      "mean_token_accuracy": 0.20236410796642304,
      "num_tokens": 59824699.0,
      "step": 30585
    },
    {
      "entropy": 6.0047770023345945,
      "epoch": 3.057779777078023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004079436759505351,
      "loss": 5.4662,
      "mean_token_accuracy": 0.19136884957551956,
      "num_tokens": 59834074.0,
      "step": 30590
    },
    {
      "entropy": 5.956073808670044,
      "epoch": 3.058279602139251,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004079148812555621,
      "loss": 5.5373,
      "mean_token_accuracy": 0.1917200818657875,
      "num_tokens": 59844681.0,
      "step": 30595
    },
    {
      "entropy": 5.959044981002807,
      "epoch": 3.05877942720048,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040788608321648496,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19912892132997512,
      "num_tokens": 59854033.0,
      "step": 30600
    },
    {
      "entropy": 6.023757648468018,
      "epoch": 3.0592792522617085,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004078572818340281,
      "loss": 5.5901,
      "mean_token_accuracy": 0.18483587056398393,
      "num_tokens": 59864191.0,
      "step": 30605
    },
    {
      "entropy": 5.975682544708252,
      "epoch": 3.059779077322937,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040782847710891623,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1879923701286316,
      "num_tokens": 59873266.0,
      "step": 30610
    },
    {
      "entropy": 6.00839376449585,
      "epoch": 3.0602789023841654,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040779966904187407,
      "loss": 5.5628,
      "mean_token_accuracy": 0.18052258640527724,
      "num_tokens": 59883050.0,
      "step": 30615
    },
    {
      "entropy": 6.027874374389649,
      "epoch": 3.060778727445394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040777085763362634,
      "loss": 5.4574,
      "mean_token_accuracy": 0.1947351410984993,
      "num_tokens": 59892897.0,
      "step": 30620
    },
    {
      "entropy": 6.033336448669433,
      "epoch": 3.061278552506623,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040774204288489805,
      "loss": 5.5443,
      "mean_token_accuracy": 0.18573326766490936,
      "num_tokens": 59902682.0,
      "step": 30625
    },
    {
      "entropy": 5.99347505569458,
      "epoch": 3.061778377567851,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004077132247964142,
      "loss": 5.5585,
      "mean_token_accuracy": 0.19053154736757277,
      "num_tokens": 59913176.0,
      "step": 30630
    },
    {
      "entropy": 5.973309898376465,
      "epoch": 3.06227820262908,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040768440336889965,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19138660579919814,
      "num_tokens": 59924238.0,
      "step": 30635
    },
    {
      "entropy": 6.037839031219482,
      "epoch": 3.0627780276903085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004076555786030797,
      "loss": 5.498,
      "mean_token_accuracy": 0.1960028365254402,
      "num_tokens": 59933766.0,
      "step": 30640
    },
    {
      "entropy": 6.085554742813111,
      "epoch": 3.063277852751537,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004076267504996795,
      "loss": 5.5399,
      "mean_token_accuracy": 0.19040786176919938,
      "num_tokens": 59944008.0,
      "step": 30645
    },
    {
      "entropy": 6.043449115753174,
      "epoch": 3.0637776778127654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004075979190594244,
      "loss": 5.5302,
      "mean_token_accuracy": 0.18134769052267075,
      "num_tokens": 59953882.0,
      "step": 30650
    },
    {
      "entropy": 5.989182806015014,
      "epoch": 3.064277502873994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040756908428303984,
      "loss": 5.6193,
      "mean_token_accuracy": 0.18469201177358627,
      "num_tokens": 59964271.0,
      "step": 30655
    },
    {
      "entropy": 6.034743738174439,
      "epoch": 3.064777327935223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040754024617125115,
      "loss": 5.4383,
      "mean_token_accuracy": 0.1995055302977562,
      "num_tokens": 59975376.0,
      "step": 30660
    },
    {
      "entropy": 5.97733588218689,
      "epoch": 3.065277152996451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000407511404724784,
      "loss": 5.4425,
      "mean_token_accuracy": 0.19399305135011674,
      "num_tokens": 59984200.0,
      "step": 30665
    },
    {
      "entropy": 5.944298267364502,
      "epoch": 3.06577697805768,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000407482559944364,
      "loss": 5.4286,
      "mean_token_accuracy": 0.19548420310020448,
      "num_tokens": 59994289.0,
      "step": 30670
    },
    {
      "entropy": 5.990743732452392,
      "epoch": 3.0662768031189085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004074537118307169,
      "loss": 5.5137,
      "mean_token_accuracy": 0.19128908067941666,
      "num_tokens": 60004218.0,
      "step": 30675
    },
    {
      "entropy": 5.943730449676513,
      "epoch": 3.0667766281801367,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004074248603845684,
      "loss": 5.4222,
      "mean_token_accuracy": 0.196084888279438,
      "num_tokens": 60012792.0,
      "step": 30680
    },
    {
      "entropy": 5.991735124588013,
      "epoch": 3.0672764532413654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040739600560664457,
      "loss": 5.4679,
      "mean_token_accuracy": 0.1927444279193878,
      "num_tokens": 60023022.0,
      "step": 30685
    },
    {
      "entropy": 5.9997793674469,
      "epoch": 3.067776278302594,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040736714749767136,
      "loss": 5.4927,
      "mean_token_accuracy": 0.1966174989938736,
      "num_tokens": 60032379.0,
      "step": 30690
    },
    {
      "entropy": 5.946637105941773,
      "epoch": 3.068276103363823,
      "grad_norm": 1.125,
      "learning_rate": 0.0004073382860583747,
      "loss": 5.4694,
      "mean_token_accuracy": 0.19652843624353408,
      "num_tokens": 60041851.0,
      "step": 30695
    },
    {
      "entropy": 6.0511720180511475,
      "epoch": 3.068775928425051,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004073094212894809,
      "loss": 5.6244,
      "mean_token_accuracy": 0.189349564909935,
      "num_tokens": 60051648.0,
      "step": 30700
    },
    {
      "entropy": 6.053693723678589,
      "epoch": 3.06927575348628,
      "grad_norm": 1.0625,
      "learning_rate": 0.000407280553191716,
      "loss": 5.5342,
      "mean_token_accuracy": 0.18934340626001359,
      "num_tokens": 60061479.0,
      "step": 30705
    },
    {
      "entropy": 5.99012565612793,
      "epoch": 3.0697755785475085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.4388,
      "mean_token_accuracy": 0.19986373633146287,
      "num_tokens": 60070378.0,
      "step": 30710
    },
    {
      "entropy": 5.974273061752319,
      "epoch": 3.0702754036087367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040722280701247865,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19616566449403763,
      "num_tokens": 60079911.0,
      "step": 30715
    },
    {
      "entropy": 6.094098043441773,
      "epoch": 3.0707752286699654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040719392893245903,
      "loss": 5.6292,
      "mean_token_accuracy": 0.17831578254699706,
      "num_tokens": 60089850.0,
      "step": 30720
    },
    {
      "entropy": 5.9936329364776615,
      "epoch": 3.071275053731194,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004071650475264742,
      "loss": 5.4283,
      "mean_token_accuracy": 0.19535650610923766,
      "num_tokens": 60100370.0,
      "step": 30725
    },
    {
      "entropy": 6.012389707565307,
      "epoch": 3.071774878792423,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040713616279525075,
      "loss": 5.5116,
      "mean_token_accuracy": 0.19230661541223526,
      "num_tokens": 60109331.0,
      "step": 30730
    },
    {
      "entropy": 6.063792276382446,
      "epoch": 3.072274703853651,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040710727473951546,
      "loss": 5.5018,
      "mean_token_accuracy": 0.18582753688097,
      "num_tokens": 60119053.0,
      "step": 30735
    },
    {
      "entropy": 6.042261266708374,
      "epoch": 3.07277452891488,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004070783833599951,
      "loss": 5.5722,
      "mean_token_accuracy": 0.19138296395540239,
      "num_tokens": 60129123.0,
      "step": 30740
    },
    {
      "entropy": 6.037561130523682,
      "epoch": 3.0732743539761085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004070494886574166,
      "loss": 5.5598,
      "mean_token_accuracy": 0.18595702648162843,
      "num_tokens": 60138848.0,
      "step": 30745
    },
    {
      "entropy": 5.961605548858643,
      "epoch": 3.0737741790373367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000407020590632507,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19823525696992875,
      "num_tokens": 60148604.0,
      "step": 30750
    },
    {
      "entropy": 5.959400081634522,
      "epoch": 3.0742740040985654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004069916892859933,
      "loss": 5.4542,
      "mean_token_accuracy": 0.19203483015298844,
      "num_tokens": 60159958.0,
      "step": 30755
    },
    {
      "entropy": 6.012706089019775,
      "epoch": 3.074773829159794,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004069627846186026,
      "loss": 5.483,
      "mean_token_accuracy": 0.1915075108408928,
      "num_tokens": 60169110.0,
      "step": 30760
    },
    {
      "entropy": 5.920055150985718,
      "epoch": 3.075273654221023,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040693387663106223,
      "loss": 5.406,
      "mean_token_accuracy": 0.2033870428800583,
      "num_tokens": 60177917.0,
      "step": 30765
    },
    {
      "entropy": 5.955008172988892,
      "epoch": 3.075773479282251,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004069049653240996,
      "loss": 5.5563,
      "mean_token_accuracy": 0.18773699402809144,
      "num_tokens": 60187086.0,
      "step": 30770
    },
    {
      "entropy": 6.032948398590088,
      "epoch": 3.07627330434348,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004068760506984418,
      "loss": 5.5573,
      "mean_token_accuracy": 0.18182040452957154,
      "num_tokens": 60196457.0,
      "step": 30775
    },
    {
      "entropy": 6.058506441116333,
      "epoch": 3.0767731294047085,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004068471327548166,
      "loss": 5.513,
      "mean_token_accuracy": 0.19098625779151918,
      "num_tokens": 60205160.0,
      "step": 30780
    },
    {
      "entropy": 6.101775169372559,
      "epoch": 3.0772729544659367,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040681821149395146,
      "loss": 5.503,
      "mean_token_accuracy": 0.18213044106960297,
      "num_tokens": 60214706.0,
      "step": 30785
    },
    {
      "entropy": 5.958443737030029,
      "epoch": 3.0777727795271654,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040678928691657395,
      "loss": 5.4473,
      "mean_token_accuracy": 0.18932726383209228,
      "num_tokens": 60224741.0,
      "step": 30790
    },
    {
      "entropy": 6.000067234039307,
      "epoch": 3.078272604588394,
      "grad_norm": 1.25,
      "learning_rate": 0.00040676035902341195,
      "loss": 5.5395,
      "mean_token_accuracy": 0.19170404970645905,
      "num_tokens": 60233666.0,
      "step": 30795
    },
    {
      "entropy": 6.015484380722046,
      "epoch": 3.078772429649623,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004067314278151933,
      "loss": 5.5489,
      "mean_token_accuracy": 0.1812763348221779,
      "num_tokens": 60244403.0,
      "step": 30800
    },
    {
      "entropy": 6.036729240417481,
      "epoch": 3.079272254710851,
      "grad_norm": 1.125,
      "learning_rate": 0.00040670249329264577,
      "loss": 5.5073,
      "mean_token_accuracy": 0.1896018534898758,
      "num_tokens": 60254339.0,
      "step": 30805
    },
    {
      "entropy": 6.054023504257202,
      "epoch": 3.07977207977208,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004066735554564973,
      "loss": 5.5635,
      "mean_token_accuracy": 0.19334927052259446,
      "num_tokens": 60264740.0,
      "step": 30810
    },
    {
      "entropy": 6.028745317459107,
      "epoch": 3.0802719048333085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004066446143074762,
      "loss": 5.515,
      "mean_token_accuracy": 0.19305855929851531,
      "num_tokens": 60273307.0,
      "step": 30815
    },
    {
      "entropy": 5.975570964813232,
      "epoch": 3.0807717298945367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004066156698463104,
      "loss": 5.5235,
      "mean_token_accuracy": 0.19003121703863143,
      "num_tokens": 60283734.0,
      "step": 30820
    },
    {
      "entropy": 6.008889389038086,
      "epoch": 3.0812715549557654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004065867220737282,
      "loss": 5.5131,
      "mean_token_accuracy": 0.19322737455368041,
      "num_tokens": 60294900.0,
      "step": 30825
    },
    {
      "entropy": 5.946542119979858,
      "epoch": 3.081771380016994,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040655777099045794,
      "loss": 5.491,
      "mean_token_accuracy": 0.18579414039850234,
      "num_tokens": 60305214.0,
      "step": 30830
    },
    {
      "entropy": 6.075993394851684,
      "epoch": 3.082271205078223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004065288165972281,
      "loss": 5.4924,
      "mean_token_accuracy": 0.19545225352048873,
      "num_tokens": 60314563.0,
      "step": 30835
    },
    {
      "entropy": 6.038094425201416,
      "epoch": 3.082771030139451,
      "grad_norm": 1.09375,
      "learning_rate": 0.000406499858894767,
      "loss": 5.5094,
      "mean_token_accuracy": 0.1927547350525856,
      "num_tokens": 60324510.0,
      "step": 30840
    },
    {
      "entropy": 6.081586074829102,
      "epoch": 3.08327085520068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004064708978838033,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1855264350771904,
      "num_tokens": 60335679.0,
      "step": 30845
    },
    {
      "entropy": 5.976683950424194,
      "epoch": 3.0837706802619085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040644193356506565,
      "loss": 5.3808,
      "mean_token_accuracy": 0.2045510306954384,
      "num_tokens": 60345006.0,
      "step": 30850
    },
    {
      "entropy": 6.014987659454346,
      "epoch": 3.0842705053231367,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004064129659392828,
      "loss": 5.5264,
      "mean_token_accuracy": 0.1865127921104431,
      "num_tokens": 60354411.0,
      "step": 30855
    },
    {
      "entropy": 5.949610471725464,
      "epoch": 3.0847703303843654,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004063839950071835,
      "loss": 5.4661,
      "mean_token_accuracy": 0.1905960813164711,
      "num_tokens": 60365347.0,
      "step": 30860
    },
    {
      "entropy": 5.92438702583313,
      "epoch": 3.085270155445594,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040635502076949665,
      "loss": 5.4692,
      "mean_token_accuracy": 0.19770287871360778,
      "num_tokens": 60375197.0,
      "step": 30865
    },
    {
      "entropy": 5.968564701080322,
      "epoch": 3.0857699805068224,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040632604322695135,
      "loss": 5.4598,
      "mean_token_accuracy": 0.19506329149007798,
      "num_tokens": 60384380.0,
      "step": 30870
    },
    {
      "entropy": 5.910825109481811,
      "epoch": 3.086269805568051,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004062970623802766,
      "loss": 5.4272,
      "mean_token_accuracy": 0.1984495386481285,
      "num_tokens": 60393990.0,
      "step": 30875
    },
    {
      "entropy": 5.992239141464234,
      "epoch": 3.08676963062928,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004062680782302015,
      "loss": 5.5062,
      "mean_token_accuracy": 0.18987037390470504,
      "num_tokens": 60402835.0,
      "step": 30880
    },
    {
      "entropy": 5.9644858837127686,
      "epoch": 3.0872694556905085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040623909077745535,
      "loss": 5.4509,
      "mean_token_accuracy": 0.20088296979665757,
      "num_tokens": 60412427.0,
      "step": 30885
    },
    {
      "entropy": 6.090335559844971,
      "epoch": 3.0877692807517367,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004062101000227674,
      "loss": 5.6374,
      "mean_token_accuracy": 0.18452135026454924,
      "num_tokens": 60424064.0,
      "step": 30890
    },
    {
      "entropy": 6.070415735244751,
      "epoch": 3.0882691058129654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004061811059668672,
      "loss": 5.6501,
      "mean_token_accuracy": 0.17930343002080917,
      "num_tokens": 60434182.0,
      "step": 30895
    },
    {
      "entropy": 6.012079572677612,
      "epoch": 3.088768930874194,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004061521086104841,
      "loss": 5.4901,
      "mean_token_accuracy": 0.1922512024641037,
      "num_tokens": 60444138.0,
      "step": 30900
    },
    {
      "entropy": 6.001358842849731,
      "epoch": 3.0892687559354224,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040612310795434765,
      "loss": 5.4785,
      "mean_token_accuracy": 0.18536950647830963,
      "num_tokens": 60452823.0,
      "step": 30905
    },
    {
      "entropy": 5.953254652023316,
      "epoch": 3.089768580996651,
      "grad_norm": 1.484375,
      "learning_rate": 0.00040609410399918754,
      "loss": 5.4714,
      "mean_token_accuracy": 0.19070627838373183,
      "num_tokens": 60462998.0,
      "step": 30910
    },
    {
      "entropy": 6.003339433670044,
      "epoch": 3.09026840605788,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040606509674573354,
      "loss": 5.5681,
      "mean_token_accuracy": 0.1868528828024864,
      "num_tokens": 60472204.0,
      "step": 30915
    },
    {
      "entropy": 5.937091922760009,
      "epoch": 3.0907682311191085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004060360861947155,
      "loss": 5.4631,
      "mean_token_accuracy": 0.19118652045726775,
      "num_tokens": 60481848.0,
      "step": 30920
    },
    {
      "entropy": 5.9721709251403805,
      "epoch": 3.0912680561803367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040600707234686315,
      "loss": 5.4903,
      "mean_token_accuracy": 0.18616895973682404,
      "num_tokens": 60491890.0,
      "step": 30925
    },
    {
      "entropy": 5.983632755279541,
      "epoch": 3.0917678812415654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040597805520290657,
      "loss": 5.4616,
      "mean_token_accuracy": 0.188823339343071,
      "num_tokens": 60501617.0,
      "step": 30930
    },
    {
      "entropy": 5.980251216888428,
      "epoch": 3.092267706302794,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004059490347635759,
      "loss": 5.5123,
      "mean_token_accuracy": 0.18695961087942123,
      "num_tokens": 60511699.0,
      "step": 30935
    },
    {
      "entropy": 6.1248026371002195,
      "epoch": 3.0927675313640224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040592001102960116,
      "loss": 5.6441,
      "mean_token_accuracy": 0.17628921568393707,
      "num_tokens": 60521631.0,
      "step": 30940
    },
    {
      "entropy": 5.990758466720581,
      "epoch": 3.093267356425251,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004058909840017126,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1862327978014946,
      "num_tokens": 60531760.0,
      "step": 30945
    },
    {
      "entropy": 6.016057062149048,
      "epoch": 3.09376718148648,
      "grad_norm": 1.25,
      "learning_rate": 0.0004058619536806407,
      "loss": 5.5673,
      "mean_token_accuracy": 0.18265179097652434,
      "num_tokens": 60542249.0,
      "step": 30950
    },
    {
      "entropy": 6.011344289779663,
      "epoch": 3.0942670065477085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004058329200671155,
      "loss": 5.5188,
      "mean_token_accuracy": 0.1848702311515808,
      "num_tokens": 60552638.0,
      "step": 30955
    },
    {
      "entropy": 6.06263370513916,
      "epoch": 3.0947668316089367,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040580388316186785,
      "loss": 5.5749,
      "mean_token_accuracy": 0.18398568630218506,
      "num_tokens": 60562928.0,
      "step": 30960
    },
    {
      "entropy": 6.059671545028687,
      "epoch": 3.0952666566701654,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004057748429656281,
      "loss": 5.5595,
      "mean_token_accuracy": 0.1929512470960617,
      "num_tokens": 60573398.0,
      "step": 30965
    },
    {
      "entropy": 6.004919767379761,
      "epoch": 3.095766481731394,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000405745799479127,
      "loss": 5.5183,
      "mean_token_accuracy": 0.19478958249092101,
      "num_tokens": 60582540.0,
      "step": 30970
    },
    {
      "entropy": 6.004763460159301,
      "epoch": 3.0962663067926224,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004057167527030952,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19770499914884568,
      "num_tokens": 60592111.0,
      "step": 30975
    },
    {
      "entropy": 5.978073406219482,
      "epoch": 3.096766131853851,
      "grad_norm": 1.328125,
      "learning_rate": 0.00040568770263826355,
      "loss": 5.5111,
      "mean_token_accuracy": 0.19028981328010558,
      "num_tokens": 60601687.0,
      "step": 30980
    },
    {
      "entropy": 6.017086315155029,
      "epoch": 3.0972659569150798,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004056586492853629,
      "loss": 5.5446,
      "mean_token_accuracy": 0.18935982137918472,
      "num_tokens": 60611021.0,
      "step": 30985
    },
    {
      "entropy": 6.003860998153686,
      "epoch": 3.0977657819763085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004056295926451243,
      "loss": 5.5762,
      "mean_token_accuracy": 0.18856922239065171,
      "num_tokens": 60620863.0,
      "step": 30990
    },
    {
      "entropy": 5.97135705947876,
      "epoch": 3.0982656070375367,
      "grad_norm": 1.25,
      "learning_rate": 0.00040560053271827866,
      "loss": 5.4587,
      "mean_token_accuracy": 0.18650951087474824,
      "num_tokens": 60630483.0,
      "step": 30995
    },
    {
      "entropy": 6.002308177947998,
      "epoch": 3.0987654320987654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040557146950555734,
      "loss": 5.5021,
      "mean_token_accuracy": 0.18897414356470107,
      "num_tokens": 60640594.0,
      "step": 31000
    },
    {
      "entropy": 5.936640024185181,
      "epoch": 3.099265257159994,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004055424030076914,
      "loss": 5.4359,
      "mean_token_accuracy": 0.19433670490980148,
      "num_tokens": 60651162.0,
      "step": 31005
    },
    {
      "entropy": 5.9918805122375485,
      "epoch": 3.0997650822212224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040551333322541225,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18761380016803741,
      "num_tokens": 60661467.0,
      "step": 31010
    },
    {
      "entropy": 6.021456480026245,
      "epoch": 3.100264907282451,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004054842601594512,
      "loss": 5.5487,
      "mean_token_accuracy": 0.1833578661084175,
      "num_tokens": 60672395.0,
      "step": 31015
    },
    {
      "entropy": 6.045238733291626,
      "epoch": 3.1007647323436798,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040545518381053964,
      "loss": 5.5575,
      "mean_token_accuracy": 0.18698793202638625,
      "num_tokens": 60680632.0,
      "step": 31020
    },
    {
      "entropy": 6.02172818183899,
      "epoch": 3.1012645574049085,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040542610417940926,
      "loss": 5.5632,
      "mean_token_accuracy": 0.18058133274316787,
      "num_tokens": 60692228.0,
      "step": 31025
    },
    {
      "entropy": 6.040108919143677,
      "epoch": 3.1017643824661367,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004053970212667917,
      "loss": 5.4832,
      "mean_token_accuracy": 0.1872377336025238,
      "num_tokens": 60701287.0,
      "step": 31030
    },
    {
      "entropy": 6.026408004760742,
      "epoch": 3.1022642075273654,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004053679350734186,
      "loss": 5.5653,
      "mean_token_accuracy": 0.19190991222858428,
      "num_tokens": 60713267.0,
      "step": 31035
    },
    {
      "entropy": 5.9434826374053955,
      "epoch": 3.102764032588594,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040533884560002196,
      "loss": 5.3973,
      "mean_token_accuracy": 0.19521703720092773,
      "num_tokens": 60722509.0,
      "step": 31040
    },
    {
      "entropy": 5.946299934387207,
      "epoch": 3.1032638576498224,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040530975284733327,
      "loss": 5.4986,
      "mean_token_accuracy": 0.1896148830652237,
      "num_tokens": 60732916.0,
      "step": 31045
    },
    {
      "entropy": 6.0922136306762695,
      "epoch": 3.103763682711051,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004052806568160849,
      "loss": 5.591,
      "mean_token_accuracy": 0.18731735795736312,
      "num_tokens": 60743257.0,
      "step": 31050
    },
    {
      "entropy": 5.896047973632813,
      "epoch": 3.1042635077722798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040525155750700863,
      "loss": 5.3282,
      "mean_token_accuracy": 0.2011484831571579,
      "num_tokens": 60752469.0,
      "step": 31055
    },
    {
      "entropy": 6.024909448623657,
      "epoch": 3.1047633328335085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040522245492083674,
      "loss": 5.5755,
      "mean_token_accuracy": 0.18874099850654602,
      "num_tokens": 60762986.0,
      "step": 31060
    },
    {
      "entropy": 6.028313493728637,
      "epoch": 3.1052631578947367,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040519334905830134,
      "loss": 5.4542,
      "mean_token_accuracy": 0.19262543618679046,
      "num_tokens": 60772557.0,
      "step": 31065
    },
    {
      "entropy": 6.077047538757324,
      "epoch": 3.1057629829559654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051642399201349,
      "loss": 5.654,
      "mean_token_accuracy": 0.17713721990585327,
      "num_tokens": 60782357.0,
      "step": 31070
    },
    {
      "entropy": 6.028645181655884,
      "epoch": 3.106262808017194,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040513512750706956,
      "loss": 5.4922,
      "mean_token_accuracy": 0.18916944414377213,
      "num_tokens": 60791821.0,
      "step": 31075
    },
    {
      "entropy": 6.080996131896972,
      "epoch": 3.1067626330784224,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000405106011819838,
      "loss": 5.5913,
      "mean_token_accuracy": 0.18256091922521592,
      "num_tokens": 60801750.0,
      "step": 31080
    },
    {
      "entropy": 6.0025159358978275,
      "epoch": 3.107262458139651,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040507689285917254,
      "loss": 5.4368,
      "mean_token_accuracy": 0.18911289870738984,
      "num_tokens": 60811177.0,
      "step": 31085
    },
    {
      "entropy": 5.961672782897949,
      "epoch": 3.1077622832008798,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040504777062580597,
      "loss": 5.4013,
      "mean_token_accuracy": 0.2002179890871048,
      "num_tokens": 60820817.0,
      "step": 31090
    },
    {
      "entropy": 6.084630298614502,
      "epoch": 3.1082621082621085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000405018645120471,
      "loss": 5.6504,
      "mean_token_accuracy": 0.179860233515501,
      "num_tokens": 60830827.0,
      "step": 31095
    },
    {
      "entropy": 6.100524044036865,
      "epoch": 3.1087619333233367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004049895163439003,
      "loss": 5.6513,
      "mean_token_accuracy": 0.17660912573337556,
      "num_tokens": 60841523.0,
      "step": 31100
    },
    {
      "entropy": 6.010053539276123,
      "epoch": 3.1092617583845654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040496038429682684,
      "loss": 5.4894,
      "mean_token_accuracy": 0.18691978007555007,
      "num_tokens": 60850179.0,
      "step": 31105
    },
    {
      "entropy": 6.002314043045044,
      "epoch": 3.109761583445794,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040493124897998363,
      "loss": 5.5126,
      "mean_token_accuracy": 0.201152403652668,
      "num_tokens": 60861571.0,
      "step": 31110
    },
    {
      "entropy": 6.058618259429932,
      "epoch": 3.1102614085070224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004049021103941036,
      "loss": 5.5184,
      "mean_token_accuracy": 0.1882848173379898,
      "num_tokens": 60871148.0,
      "step": 31115
    },
    {
      "entropy": 6.050678825378418,
      "epoch": 3.110761233568251,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040487296853991984,
      "loss": 5.5345,
      "mean_token_accuracy": 0.18332382291555405,
      "num_tokens": 60880869.0,
      "step": 31120
    },
    {
      "entropy": 6.055273151397705,
      "epoch": 3.1112610586294798,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004048438234181657,
      "loss": 5.5133,
      "mean_token_accuracy": 0.18677472025156022,
      "num_tokens": 60890770.0,
      "step": 31125
    },
    {
      "entropy": 5.94452919960022,
      "epoch": 3.1117608836907085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040481467502957426,
      "loss": 5.4537,
      "mean_token_accuracy": 0.19575224816799164,
      "num_tokens": 60900251.0,
      "step": 31130
    },
    {
      "entropy": 5.912837028503418,
      "epoch": 3.1122607087519367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000404785523374879,
      "loss": 5.4367,
      "mean_token_accuracy": 0.19870525747537612,
      "num_tokens": 60909053.0,
      "step": 31135
    },
    {
      "entropy": 5.971794128417969,
      "epoch": 3.1127605338131654,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040475636845481347,
      "loss": 5.4311,
      "mean_token_accuracy": 0.20110018253326417,
      "num_tokens": 60918583.0,
      "step": 31140
    },
    {
      "entropy": 6.028646326065063,
      "epoch": 3.113260358874394,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00040472721027011107,
      "loss": 5.5256,
      "mean_token_accuracy": 0.19061702489852905,
      "num_tokens": 60927635.0,
      "step": 31145
    },
    {
      "entropy": 6.056046867370606,
      "epoch": 3.1137601839356224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004046980488215054,
      "loss": 5.5339,
      "mean_token_accuracy": 0.18682736456394194,
      "num_tokens": 60937831.0,
      "step": 31150
    },
    {
      "entropy": 6.091821336746216,
      "epoch": 3.114260008996851,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040466888410973026,
      "loss": 5.605,
      "mean_token_accuracy": 0.1836787596344948,
      "num_tokens": 60947333.0,
      "step": 31155
    },
    {
      "entropy": 6.03050389289856,
      "epoch": 3.1147598340580798,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004046397161355193,
      "loss": 5.4722,
      "mean_token_accuracy": 0.18960429280996322,
      "num_tokens": 60956585.0,
      "step": 31160
    },
    {
      "entropy": 5.873997020721435,
      "epoch": 3.1152596591193085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004046105448996064,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20537732690572738,
      "num_tokens": 60966496.0,
      "step": 31165
    },
    {
      "entropy": 6.031907653808593,
      "epoch": 3.1157594841805367,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004045813704027257,
      "loss": 5.5325,
      "mean_token_accuracy": 0.18893091529607772,
      "num_tokens": 60976500.0,
      "step": 31170
    },
    {
      "entropy": 6.02504711151123,
      "epoch": 3.1162593092417654,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004045521926456109,
      "loss": 5.5434,
      "mean_token_accuracy": 0.1939183697104454,
      "num_tokens": 60985935.0,
      "step": 31175
    },
    {
      "entropy": 6.000186109542847,
      "epoch": 3.116759134302994,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004045230116289964,
      "loss": 5.5074,
      "mean_token_accuracy": 0.18856339752674103,
      "num_tokens": 60995349.0,
      "step": 31180
    },
    {
      "entropy": 6.045585489273071,
      "epoch": 3.1172589593642224,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040449382735361614,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18381789773702623,
      "num_tokens": 61006229.0,
      "step": 31185
    },
    {
      "entropy": 6.0347717761993405,
      "epoch": 3.117758784425451,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004044646398202046,
      "loss": 5.5521,
      "mean_token_accuracy": 0.18871208131313325,
      "num_tokens": 61016394.0,
      "step": 31190
    },
    {
      "entropy": 6.03511004447937,
      "epoch": 3.1182586094866798,
      "grad_norm": 1.078125,
      "learning_rate": 0.000404435449029496,
      "loss": 5.4735,
      "mean_token_accuracy": 0.19342928677797316,
      "num_tokens": 61026206.0,
      "step": 31195
    },
    {
      "entropy": 5.930072593688965,
      "epoch": 3.118758434547908,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004044062549822248,
      "loss": 5.4642,
      "mean_token_accuracy": 0.18711747378110885,
      "num_tokens": 61036081.0,
      "step": 31200
    },
    {
      "entropy": 5.969718503952026,
      "epoch": 3.1192582596091367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004043770576791255,
      "loss": 5.5187,
      "mean_token_accuracy": 0.19070078581571578,
      "num_tokens": 61045812.0,
      "step": 31205
    },
    {
      "entropy": 5.970858001708985,
      "epoch": 3.1197580846703654,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004043478571209328,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19875003546476364,
      "num_tokens": 61055122.0,
      "step": 31210
    },
    {
      "entropy": 5.907156229019165,
      "epoch": 3.120257909731594,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004043186533083812,
      "loss": 5.4425,
      "mean_token_accuracy": 0.19607459604740143,
      "num_tokens": 61064033.0,
      "step": 31215
    },
    {
      "entropy": 5.97075343132019,
      "epoch": 3.1207577347928224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040428944624220557,
      "loss": 5.4707,
      "mean_token_accuracy": 0.18857973366975783,
      "num_tokens": 61073881.0,
      "step": 31220
    },
    {
      "entropy": 6.004897689819336,
      "epoch": 3.121257559854051,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004042602359231409,
      "loss": 5.412,
      "mean_token_accuracy": 0.20072963386774062,
      "num_tokens": 61083651.0,
      "step": 31225
    },
    {
      "entropy": 6.0223150730133055,
      "epoch": 3.1217573849152798,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004042310223519218,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1773715943098068,
      "num_tokens": 61092860.0,
      "step": 31230
    },
    {
      "entropy": 5.922070837020874,
      "epoch": 3.122257209976508,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004042018055292835,
      "loss": 5.501,
      "mean_token_accuracy": 0.18957314789295196,
      "num_tokens": 61102937.0,
      "step": 31235
    },
    {
      "entropy": 5.983086967468262,
      "epoch": 3.1227570350377367,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004041725854559609,
      "loss": 5.3574,
      "mean_token_accuracy": 0.19673732817173004,
      "num_tokens": 61112422.0,
      "step": 31240
    },
    {
      "entropy": 6.077568626403808,
      "epoch": 3.1232568600989654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040414336213268943,
      "loss": 5.6594,
      "mean_token_accuracy": 0.17705402672290801,
      "num_tokens": 61122814.0,
      "step": 31245
    },
    {
      "entropy": 5.9918921947479244,
      "epoch": 3.123756685160194,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004041141355602041,
      "loss": 5.4973,
      "mean_token_accuracy": 0.18592311441898346,
      "num_tokens": 61133938.0,
      "step": 31250
    },
    {
      "entropy": 6.0164406299591064,
      "epoch": 3.1242565102214224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004040849057392405,
      "loss": 5.4996,
      "mean_token_accuracy": 0.191505728662014,
      "num_tokens": 61143941.0,
      "step": 31255
    },
    {
      "entropy": 6.0127040386199955,
      "epoch": 3.124756335282651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004040556726705337,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1920264407992363,
      "num_tokens": 61152674.0,
      "step": 31260
    },
    {
      "entropy": 5.979718017578125,
      "epoch": 3.1252561603438798,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004040264363548195,
      "loss": 5.5237,
      "mean_token_accuracy": 0.19408092200756072,
      "num_tokens": 61161825.0,
      "step": 31265
    },
    {
      "entropy": 6.009532928466797,
      "epoch": 3.125755985405108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004039971967928333,
      "loss": 5.5005,
      "mean_token_accuracy": 0.18769952952861785,
      "num_tokens": 61171283.0,
      "step": 31270
    },
    {
      "entropy": 6.006072998046875,
      "epoch": 3.1262558104663367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004039679539853108,
      "loss": 5.4715,
      "mean_token_accuracy": 0.19113269746303557,
      "num_tokens": 61180736.0,
      "step": 31275
    },
    {
      "entropy": 6.106741333007813,
      "epoch": 3.1267556355275654,
      "grad_norm": 1.125,
      "learning_rate": 0.0004039387079329878,
      "loss": 5.5788,
      "mean_token_accuracy": 0.18968260139226914,
      "num_tokens": 61190210.0,
      "step": 31280
    },
    {
      "entropy": 6.164996385574341,
      "epoch": 3.127255460588794,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004039094586366001,
      "loss": 5.6436,
      "mean_token_accuracy": 0.17551056444644927,
      "num_tokens": 61201000.0,
      "step": 31285
    },
    {
      "entropy": 5.970638465881348,
      "epoch": 3.1277552856500224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040388020609688366,
      "loss": 5.4623,
      "mean_token_accuracy": 0.19293225854635238,
      "num_tokens": 61209924.0,
      "step": 31290
    },
    {
      "entropy": 5.979393053054809,
      "epoch": 3.128255110711251,
      "grad_norm": 1.125,
      "learning_rate": 0.00040385095031457435,
      "loss": 5.5089,
      "mean_token_accuracy": 0.19036852121353148,
      "num_tokens": 61219472.0,
      "step": 31295
    },
    {
      "entropy": 5.9779106140136715,
      "epoch": 3.1287549357724798,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004038216912904082,
      "loss": 5.4263,
      "mean_token_accuracy": 0.2000921681523323,
      "num_tokens": 61229609.0,
      "step": 31300
    },
    {
      "entropy": 6.023142957687378,
      "epoch": 3.129254760833708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004037924290251215,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18996469676494598,
      "num_tokens": 61240352.0,
      "step": 31305
    },
    {
      "entropy": 5.995620441436768,
      "epoch": 3.1297545858949367,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004037631635194505,
      "loss": 5.4509,
      "mean_token_accuracy": 0.19116332679986953,
      "num_tokens": 61249789.0,
      "step": 31310
    },
    {
      "entropy": 5.879777336120606,
      "epoch": 3.1302544109561654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004037338947741313,
      "loss": 5.3346,
      "mean_token_accuracy": 0.20647925585508348,
      "num_tokens": 61259788.0,
      "step": 31315
    },
    {
      "entropy": 5.862354564666748,
      "epoch": 3.130754236017394,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040370462278990055,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21397629380226135,
      "num_tokens": 61269115.0,
      "step": 31320
    },
    {
      "entropy": 5.8686285495758055,
      "epoch": 3.1312540610786224,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004036753475674945,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19923417270183563,
      "num_tokens": 61279272.0,
      "step": 31325
    },
    {
      "entropy": 5.98060793876648,
      "epoch": 3.131753886139851,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040364606910764996,
      "loss": 5.5244,
      "mean_token_accuracy": 0.1856592819094658,
      "num_tokens": 61290451.0,
      "step": 31330
    },
    {
      "entropy": 5.980371809005737,
      "epoch": 3.1322537112010798,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004036167874111034,
      "loss": 5.4565,
      "mean_token_accuracy": 0.19006839245557786,
      "num_tokens": 61299666.0,
      "step": 31335
    },
    {
      "entropy": 6.078668594360352,
      "epoch": 3.132753536262308,
      "grad_norm": 1.125,
      "learning_rate": 0.00040358750247859146,
      "loss": 5.6119,
      "mean_token_accuracy": 0.18243710398674012,
      "num_tokens": 61308794.0,
      "step": 31340
    },
    {
      "entropy": 6.058984184265137,
      "epoch": 3.1332533613235367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004035582143108511,
      "loss": 5.5765,
      "mean_token_accuracy": 0.18400425761938094,
      "num_tokens": 61319532.0,
      "step": 31345
    },
    {
      "entropy": 6.01681661605835,
      "epoch": 3.1337531863847654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040352892290861916,
      "loss": 5.5301,
      "mean_token_accuracy": 0.18398915678262712,
      "num_tokens": 61329995.0,
      "step": 31350
    },
    {
      "entropy": 5.969514942169189,
      "epoch": 3.134253011445994,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004034996282726326,
      "loss": 5.4978,
      "mean_token_accuracy": 0.19002237170934677,
      "num_tokens": 61339552.0,
      "step": 31355
    },
    {
      "entropy": 5.897368907928467,
      "epoch": 3.1347528365072224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040347033040362835,
      "loss": 5.4473,
      "mean_token_accuracy": 0.1991187185049057,
      "num_tokens": 61349411.0,
      "step": 31360
    },
    {
      "entropy": 6.056374502182007,
      "epoch": 3.135252661568451,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040344102930234385,
      "loss": 5.4933,
      "mean_token_accuracy": 0.19023938924074174,
      "num_tokens": 61358538.0,
      "step": 31365
    },
    {
      "entropy": 6.015650081634521,
      "epoch": 3.1357524866296798,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040341172496951596,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1970134124159813,
      "num_tokens": 61368988.0,
      "step": 31370
    },
    {
      "entropy": 6.011067152023315,
      "epoch": 3.136252311690908,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004033824174058821,
      "loss": 5.5476,
      "mean_token_accuracy": 0.19093984514474868,
      "num_tokens": 61379643.0,
      "step": 31375
    },
    {
      "entropy": 6.0338465690612795,
      "epoch": 3.1367521367521367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004033531066121797,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1877192184329033,
      "num_tokens": 61389460.0,
      "step": 31380
    },
    {
      "entropy": 6.024660921096801,
      "epoch": 3.1372519618133654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040332379258914614,
      "loss": 5.5361,
      "mean_token_accuracy": 0.18488358855247497,
      "num_tokens": 61400034.0,
      "step": 31385
    },
    {
      "entropy": 6.001072072982788,
      "epoch": 3.1377517868745937,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000403294475337519,
      "loss": 5.5115,
      "mean_token_accuracy": 0.18045187443494798,
      "num_tokens": 61409957.0,
      "step": 31390
    },
    {
      "entropy": 6.080513048171997,
      "epoch": 3.1382516119358224,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004032651548580359,
      "loss": 5.5331,
      "mean_token_accuracy": 0.18065277487039566,
      "num_tokens": 61419265.0,
      "step": 31395
    },
    {
      "entropy": 5.982970714569092,
      "epoch": 3.138751436997051,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040323583115143446,
      "loss": 5.5054,
      "mean_token_accuracy": 0.18908618539571762,
      "num_tokens": 61429007.0,
      "step": 31400
    },
    {
      "entropy": 6.010506105422974,
      "epoch": 3.1392512620582798,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040320650421845256,
      "loss": 5.5355,
      "mean_token_accuracy": 0.19380047619342805,
      "num_tokens": 61439866.0,
      "step": 31405
    },
    {
      "entropy": 6.118713474273681,
      "epoch": 3.139751087119508,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040317717405982797,
      "loss": 5.6021,
      "mean_token_accuracy": 0.18351549059152603,
      "num_tokens": 61448953.0,
      "step": 31410
    },
    {
      "entropy": 6.0273253440856935,
      "epoch": 3.1402509121807367,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040314784067629867,
      "loss": 5.5257,
      "mean_token_accuracy": 0.1859845757484436,
      "num_tokens": 61458469.0,
      "step": 31415
    },
    {
      "entropy": 5.954194259643555,
      "epoch": 3.1407507372419654,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040311850406860267,
      "loss": 5.4089,
      "mean_token_accuracy": 0.1888916701078415,
      "num_tokens": 61468029.0,
      "step": 31420
    },
    {
      "entropy": 5.958087348937989,
      "epoch": 3.1412505623031937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040308916423747803,
      "loss": 5.5513,
      "mean_token_accuracy": 0.19233630299568177,
      "num_tokens": 61478112.0,
      "step": 31425
    },
    {
      "entropy": 6.002383375167847,
      "epoch": 3.1417503873644224,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000403059821183663,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1825003832578659,
      "num_tokens": 61487730.0,
      "step": 31430
    },
    {
      "entropy": 5.978605461120606,
      "epoch": 3.142250212425651,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004030304749078959,
      "loss": 5.4676,
      "mean_token_accuracy": 0.19207597970962526,
      "num_tokens": 61497603.0,
      "step": 31435
    },
    {
      "entropy": 5.930651473999023,
      "epoch": 3.1427500374868798,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040300112541091494,
      "loss": 5.5253,
      "mean_token_accuracy": 0.19612203985452653,
      "num_tokens": 61508090.0,
      "step": 31440
    },
    {
      "entropy": 5.917653894424438,
      "epoch": 3.143249862548108,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004029717726934586,
      "loss": 5.4018,
      "mean_token_accuracy": 0.1982937678694725,
      "num_tokens": 61516684.0,
      "step": 31445
    },
    {
      "entropy": 6.006748628616333,
      "epoch": 3.1437496876093367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040294241675626534,
      "loss": 5.527,
      "mean_token_accuracy": 0.1897008627653122,
      "num_tokens": 61525617.0,
      "step": 31450
    },
    {
      "entropy": 5.973118686676026,
      "epoch": 3.1442495126705654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004029130576000739,
      "loss": 5.5269,
      "mean_token_accuracy": 0.19222652018070222,
      "num_tokens": 61535365.0,
      "step": 31455
    },
    {
      "entropy": 5.948826503753662,
      "epoch": 3.1447493377317937,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004028836952256227,
      "loss": 5.4831,
      "mean_token_accuracy": 0.1940929651260376,
      "num_tokens": 61544355.0,
      "step": 31460
    },
    {
      "entropy": 6.045506143569947,
      "epoch": 3.1452491627930224,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004028543296336508,
      "loss": 5.4979,
      "mean_token_accuracy": 0.1958245411515236,
      "num_tokens": 61554291.0,
      "step": 31465
    },
    {
      "entropy": 6.050616598129272,
      "epoch": 3.145748987854251,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040282496082489684,
      "loss": 5.5447,
      "mean_token_accuracy": 0.1894158810377121,
      "num_tokens": 61563833.0,
      "step": 31470
    },
    {
      "entropy": 6.038721418380737,
      "epoch": 3.1462488129154798,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040279558880009967,
      "loss": 5.5115,
      "mean_token_accuracy": 0.18741511553525925,
      "num_tokens": 61573328.0,
      "step": 31475
    },
    {
      "entropy": 6.0381567001342775,
      "epoch": 3.146748637976708,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004027662135599984,
      "loss": 5.5388,
      "mean_token_accuracy": 0.18896886557340623,
      "num_tokens": 61583928.0,
      "step": 31480
    },
    {
      "entropy": 5.8695536136627195,
      "epoch": 3.1472484630379367,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004027368351053321,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19222893565893173,
      "num_tokens": 61593599.0,
      "step": 31485
    },
    {
      "entropy": 5.878364515304566,
      "epoch": 3.1477482880991654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004027074534368399,
      "loss": 5.3204,
      "mean_token_accuracy": 0.2110613092780113,
      "num_tokens": 61603284.0,
      "step": 31490
    },
    {
      "entropy": 6.0463613986969,
      "epoch": 3.1482481131603937,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040267806855526103,
      "loss": 5.4182,
      "mean_token_accuracy": 0.2018551781773567,
      "num_tokens": 61612027.0,
      "step": 31495
    },
    {
      "entropy": 6.009534454345703,
      "epoch": 3.1487479382216224,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004026486804613347,
      "loss": 5.5788,
      "mean_token_accuracy": 0.1839666560292244,
      "num_tokens": 61621665.0,
      "step": 31500
    },
    {
      "entropy": 5.941898536682129,
      "epoch": 3.149247763282851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040261928915580065,
      "loss": 5.5057,
      "mean_token_accuracy": 0.18851027637720108,
      "num_tokens": 61632684.0,
      "step": 31505
    },
    {
      "entropy": 6.0414955615997314,
      "epoch": 3.1497475883440798,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004025898946393979,
      "loss": 5.5307,
      "mean_token_accuracy": 0.193739652633667,
      "num_tokens": 61642401.0,
      "step": 31510
    },
    {
      "entropy": 5.9851212978363035,
      "epoch": 3.150247413405308,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004025604969128663,
      "loss": 5.4794,
      "mean_token_accuracy": 0.1933317616581917,
      "num_tokens": 61652644.0,
      "step": 31515
    },
    {
      "entropy": 5.934080648422241,
      "epoch": 3.1507472384665367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040253109597694544,
      "loss": 5.411,
      "mean_token_accuracy": 0.20206356197595596,
      "num_tokens": 61661951.0,
      "step": 31520
    },
    {
      "entropy": 5.915102815628051,
      "epoch": 3.1512470635277654,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040250169183237507,
      "loss": 5.4297,
      "mean_token_accuracy": 0.19276549965143203,
      "num_tokens": 61671637.0,
      "step": 31525
    },
    {
      "entropy": 5.935654258728027,
      "epoch": 3.1517468885889937,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040247228447989485,
      "loss": 5.4898,
      "mean_token_accuracy": 0.18900500237941742,
      "num_tokens": 61681867.0,
      "step": 31530
    },
    {
      "entropy": 6.00085940361023,
      "epoch": 3.1522467136502224,
      "grad_norm": 1.296875,
      "learning_rate": 0.00040244287392024485,
      "loss": 5.5227,
      "mean_token_accuracy": 0.18947719782590866,
      "num_tokens": 61691906.0,
      "step": 31535
    },
    {
      "entropy": 6.043622303009033,
      "epoch": 3.152746538711451,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004024134601541649,
      "loss": 5.5045,
      "mean_token_accuracy": 0.19745435416698456,
      "num_tokens": 61702927.0,
      "step": 31540
    },
    {
      "entropy": 5.986844778060913,
      "epoch": 3.1532463637726798,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004023840431823951,
      "loss": 5.4255,
      "mean_token_accuracy": 0.1995852679014206,
      "num_tokens": 61711585.0,
      "step": 31545
    },
    {
      "entropy": 5.904552555084228,
      "epoch": 3.153746188833908,
      "grad_norm": 1.125,
      "learning_rate": 0.0004023546230056754,
      "loss": 5.451,
      "mean_token_accuracy": 0.1979820042848587,
      "num_tokens": 61721806.0,
      "step": 31550
    },
    {
      "entropy": 5.957336950302124,
      "epoch": 3.1542460138951367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040232519962474635,
      "loss": 5.5257,
      "mean_token_accuracy": 0.18512628525495528,
      "num_tokens": 61731172.0,
      "step": 31555
    },
    {
      "entropy": 5.986332654953003,
      "epoch": 3.1547458389563654,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040229577304034793,
      "loss": 5.4744,
      "mean_token_accuracy": 0.1923870787024498,
      "num_tokens": 61740075.0,
      "step": 31560
    },
    {
      "entropy": 6.026744365692139,
      "epoch": 3.1552456640175937,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040226634325322063,
      "loss": 5.5266,
      "mean_token_accuracy": 0.18378583788871766,
      "num_tokens": 61750760.0,
      "step": 31565
    },
    {
      "entropy": 6.032047939300537,
      "epoch": 3.1557454890788224,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004022369102641049,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1888165444135666,
      "num_tokens": 61760256.0,
      "step": 31570
    },
    {
      "entropy": 5.979127025604248,
      "epoch": 3.156245314140051,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040220747407374133,
      "loss": 5.538,
      "mean_token_accuracy": 0.19101849645376207,
      "num_tokens": 61769769.0,
      "step": 31575
    },
    {
      "entropy": 6.0038733959198,
      "epoch": 3.1567451392012797,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040217803468287034,
      "loss": 5.53,
      "mean_token_accuracy": 0.18341259509325028,
      "num_tokens": 61779408.0,
      "step": 31580
    },
    {
      "entropy": 6.0454610824584964,
      "epoch": 3.157244964262508,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040214859209223273,
      "loss": 5.5168,
      "mean_token_accuracy": 0.1840352237224579,
      "num_tokens": 61789940.0,
      "step": 31585
    },
    {
      "entropy": 6.023625373840332,
      "epoch": 3.1577447893237367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004021191463025693,
      "loss": 5.5341,
      "mean_token_accuracy": 0.1820305749773979,
      "num_tokens": 61799837.0,
      "step": 31590
    },
    {
      "entropy": 5.915021896362305,
      "epoch": 3.1582446143849654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004020896973146208,
      "loss": 5.4749,
      "mean_token_accuracy": 0.191159288585186,
      "num_tokens": 61809861.0,
      "step": 31595
    },
    {
      "entropy": 5.967836666107178,
      "epoch": 3.1587444394461937,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040206024512912824,
      "loss": 5.4265,
      "mean_token_accuracy": 0.19555204957723618,
      "num_tokens": 61818711.0,
      "step": 31600
    },
    {
      "entropy": 5.91298360824585,
      "epoch": 3.1592442645074224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040203078974683265,
      "loss": 5.4264,
      "mean_token_accuracy": 0.19890432953834533,
      "num_tokens": 61829195.0,
      "step": 31605
    },
    {
      "entropy": 5.97921347618103,
      "epoch": 3.159744089568651,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004020013311684751,
      "loss": 5.5014,
      "mean_token_accuracy": 0.18832040429115296,
      "num_tokens": 61838473.0,
      "step": 31610
    },
    {
      "entropy": 6.052458238601685,
      "epoch": 3.1602439146298797,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00040197186939479665,
      "loss": 5.4909,
      "mean_token_accuracy": 0.18766199350357055,
      "num_tokens": 61848569.0,
      "step": 31615
    },
    {
      "entropy": 5.985138893127441,
      "epoch": 3.160743739691108,
      "grad_norm": 1.125,
      "learning_rate": 0.00040194240442653867,
      "loss": 5.567,
      "mean_token_accuracy": 0.18396290093660356,
      "num_tokens": 61858248.0,
      "step": 31620
    },
    {
      "entropy": 6.014611482620239,
      "epoch": 3.1612435647523367,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004019129362644425,
      "loss": 5.5259,
      "mean_token_accuracy": 0.19213992208242417,
      "num_tokens": 61867998.0,
      "step": 31625
    },
    {
      "entropy": 6.024351406097412,
      "epoch": 3.1617433898135654,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004018834649092494,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18881508111953735,
      "num_tokens": 61878436.0,
      "step": 31630
    },
    {
      "entropy": 6.018979787826538,
      "epoch": 3.1622432148747937,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004018539903617011,
      "loss": 5.5018,
      "mean_token_accuracy": 0.18646368086338044,
      "num_tokens": 61887660.0,
      "step": 31635
    },
    {
      "entropy": 6.014798641204834,
      "epoch": 3.1627430399360223,
      "grad_norm": 1.125,
      "learning_rate": 0.00040182451262253904,
      "loss": 5.5147,
      "mean_token_accuracy": 0.18398335725069045,
      "num_tokens": 61897280.0,
      "step": 31640
    },
    {
      "entropy": 5.924038696289062,
      "epoch": 3.163242864997251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004017950316925049,
      "loss": 5.438,
      "mean_token_accuracy": 0.19609144181013108,
      "num_tokens": 61907683.0,
      "step": 31645
    },
    {
      "entropy": 5.959536361694336,
      "epoch": 3.1637426900584797,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040176554757234017,
      "loss": 5.4613,
      "mean_token_accuracy": 0.18925249427556992,
      "num_tokens": 61917275.0,
      "step": 31650
    },
    {
      "entropy": 6.087010669708252,
      "epoch": 3.164242515119708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004017360602627871,
      "loss": 5.595,
      "mean_token_accuracy": 0.17699083983898162,
      "num_tokens": 61928754.0,
      "step": 31655
    },
    {
      "entropy": 6.07520546913147,
      "epoch": 3.1647423401809367,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040170656976458723,
      "loss": 5.5638,
      "mean_token_accuracy": 0.18223890364170076,
      "num_tokens": 61937569.0,
      "step": 31660
    },
    {
      "entropy": 5.947866821289063,
      "epoch": 3.1652421652421654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040167707607848273,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20256876796483994,
      "num_tokens": 61947058.0,
      "step": 31665
    },
    {
      "entropy": 5.925735998153686,
      "epoch": 3.1657419903033936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004016475792052157,
      "loss": 5.4051,
      "mean_token_accuracy": 0.19637378305196762,
      "num_tokens": 61956862.0,
      "step": 31670
    },
    {
      "entropy": 5.965548419952393,
      "epoch": 3.1662418153646223,
      "grad_norm": 1.1875,
      "learning_rate": 0.000401618079145528,
      "loss": 5.5542,
      "mean_token_accuracy": 0.19005659073591233,
      "num_tokens": 61966815.0,
      "step": 31675
    },
    {
      "entropy": 6.028543853759766,
      "epoch": 3.166741640425851,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004015885759001621,
      "loss": 5.5305,
      "mean_token_accuracy": 0.1910598397254944,
      "num_tokens": 61975872.0,
      "step": 31680
    },
    {
      "entropy": 6.03401255607605,
      "epoch": 3.1672414654870797,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040155906946986027,
      "loss": 5.556,
      "mean_token_accuracy": 0.18109699189662934,
      "num_tokens": 61986145.0,
      "step": 31685
    },
    {
      "entropy": 6.026480960845947,
      "epoch": 3.167741290548308,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004015295598553647,
      "loss": 5.5075,
      "mean_token_accuracy": 0.18667154908180236,
      "num_tokens": 61995305.0,
      "step": 31690
    },
    {
      "entropy": 5.916744756698608,
      "epoch": 3.1682411156095367,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004015000470574179,
      "loss": 5.4248,
      "mean_token_accuracy": 0.19464022517204285,
      "num_tokens": 62005364.0,
      "step": 31695
    },
    {
      "entropy": 5.9742683410644535,
      "epoch": 3.1687409406707654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040147053107676266,
      "loss": 5.5779,
      "mean_token_accuracy": 0.1854843407869339,
      "num_tokens": 62015323.0,
      "step": 31700
    },
    {
      "entropy": 6.106679964065552,
      "epoch": 3.1692407657319936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040144101191414127,
      "loss": 5.5876,
      "mean_token_accuracy": 0.18334802836179734,
      "num_tokens": 62024898.0,
      "step": 31705
    },
    {
      "entropy": 5.998483371734619,
      "epoch": 3.1697405907932223,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004014114895702966,
      "loss": 5.4103,
      "mean_token_accuracy": 0.20357892215251921,
      "num_tokens": 62033617.0,
      "step": 31710
    },
    {
      "entropy": 5.9217814922332765,
      "epoch": 3.170240415854451,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004013819640459714,
      "loss": 5.4227,
      "mean_token_accuracy": 0.19697626531124116,
      "num_tokens": 62042505.0,
      "step": 31715
    },
    {
      "entropy": 6.0093544006347654,
      "epoch": 3.1707402409156797,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040135243534190847,
      "loss": 5.5335,
      "mean_token_accuracy": 0.19033806174993514,
      "num_tokens": 62052787.0,
      "step": 31720
    },
    {
      "entropy": 6.0597435474395756,
      "epoch": 3.171240065976908,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040132290345885077,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19491894990205766,
      "num_tokens": 62062389.0,
      "step": 31725
    },
    {
      "entropy": 5.946049594879151,
      "epoch": 3.1717398910381367,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004012933683975413,
      "loss": 5.3366,
      "mean_token_accuracy": 0.20547687262296677,
      "num_tokens": 62071954.0,
      "step": 31730
    },
    {
      "entropy": 5.999673223495483,
      "epoch": 3.1722397160993654,
      "grad_norm": 1.3125,
      "learning_rate": 0.00040126383015872324,
      "loss": 5.5261,
      "mean_token_accuracy": 0.18718485534191132,
      "num_tokens": 62081667.0,
      "step": 31735
    },
    {
      "entropy": 6.026525259017944,
      "epoch": 3.1727395411605936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040123428874313974,
      "loss": 5.4746,
      "mean_token_accuracy": 0.19586081057786942,
      "num_tokens": 62091714.0,
      "step": 31740
    },
    {
      "entropy": 5.960441303253174,
      "epoch": 3.1732393662218223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040120474415153396,
      "loss": 5.383,
      "mean_token_accuracy": 0.19772748947143554,
      "num_tokens": 62100601.0,
      "step": 31745
    },
    {
      "entropy": 5.970952081680298,
      "epoch": 3.173739191283051,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040117519638464935,
      "loss": 5.4577,
      "mean_token_accuracy": 0.1974137842655182,
      "num_tokens": 62110051.0,
      "step": 31750
    },
    {
      "entropy": 5.9908891201019285,
      "epoch": 3.1742390163442793,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040114564544322913,
      "loss": 5.5172,
      "mean_token_accuracy": 0.18465505242347718,
      "num_tokens": 62119470.0,
      "step": 31755
    },
    {
      "entropy": 5.961732816696167,
      "epoch": 3.174738841405508,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000401116091328017,
      "loss": 5.4871,
      "mean_token_accuracy": 0.19538135528564454,
      "num_tokens": 62129763.0,
      "step": 31760
    },
    {
      "entropy": 6.004505586624146,
      "epoch": 3.1752386664667367,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040108653403975647,
      "loss": 5.516,
      "mean_token_accuracy": 0.18471003025770188,
      "num_tokens": 62139274.0,
      "step": 31765
    },
    {
      "entropy": 5.960278654098511,
      "epoch": 3.1757384915279654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004010569735791912,
      "loss": 5.5381,
      "mean_token_accuracy": 0.18694169223308563,
      "num_tokens": 62149128.0,
      "step": 31770
    },
    {
      "entropy": 5.976215648651123,
      "epoch": 3.1762383165891936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004010274099470649,
      "loss": 5.4898,
      "mean_token_accuracy": 0.1946890741586685,
      "num_tokens": 62158709.0,
      "step": 31775
    },
    {
      "entropy": 6.046113395690918,
      "epoch": 3.1767381416504223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004009978431441214,
      "loss": 5.5587,
      "mean_token_accuracy": 0.18604317903518677,
      "num_tokens": 62168935.0,
      "step": 31780
    },
    {
      "entropy": 5.982336854934692,
      "epoch": 3.177237966711651,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00040096827317110463,
      "loss": 5.4745,
      "mean_token_accuracy": 0.18710055947303772,
      "num_tokens": 62179511.0,
      "step": 31785
    },
    {
      "entropy": 6.008743095397949,
      "epoch": 3.1777377917728793,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004009387000287585,
      "loss": 5.5173,
      "mean_token_accuracy": 0.1831270232796669,
      "num_tokens": 62190223.0,
      "step": 31790
    },
    {
      "entropy": 5.978692197799683,
      "epoch": 3.178237616834108,
      "grad_norm": 1.09375,
      "learning_rate": 0.000400909123717827,
      "loss": 5.4978,
      "mean_token_accuracy": 0.19538167864084244,
      "num_tokens": 62200396.0,
      "step": 31795
    },
    {
      "entropy": 5.976021957397461,
      "epoch": 3.1787374418953367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004008795442390544,
      "loss": 5.4945,
      "mean_token_accuracy": 0.19142888486385345,
      "num_tokens": 62210893.0,
      "step": 31800
    },
    {
      "entropy": 5.91988959312439,
      "epoch": 3.1792372669565654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004008499615931849,
      "loss": 5.3864,
      "mean_token_accuracy": 0.2004285842180252,
      "num_tokens": 62221028.0,
      "step": 31805
    },
    {
      "entropy": 5.877358245849609,
      "epoch": 3.1797370920177936,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004008203757809627,
      "loss": 5.3444,
      "mean_token_accuracy": 0.20312277227640152,
      "num_tokens": 62230752.0,
      "step": 31810
    },
    {
      "entropy": 6.003968906402588,
      "epoch": 3.1802369170790223,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00040079078680313224,
      "loss": 5.5558,
      "mean_token_accuracy": 0.18254027664661407,
      "num_tokens": 62240910.0,
      "step": 31815
    },
    {
      "entropy": 6.009103679656983,
      "epoch": 3.180736742140251,
      "grad_norm": 1.203125,
      "learning_rate": 0.000400761194660438,
      "loss": 5.5032,
      "mean_token_accuracy": 0.18624015301465988,
      "num_tokens": 62249738.0,
      "step": 31820
    },
    {
      "entropy": 5.946263074874878,
      "epoch": 3.1812365672014793,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004007315993536244,
      "loss": 5.3406,
      "mean_token_accuracy": 0.21116994172334672,
      "num_tokens": 62259136.0,
      "step": 31825
    },
    {
      "entropy": 5.9413063526153564,
      "epoch": 3.181736392262708,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040070200088343613,
      "loss": 5.5104,
      "mean_token_accuracy": 0.19159673750400544,
      "num_tokens": 62269056.0,
      "step": 31830
    },
    {
      "entropy": 6.033565044403076,
      "epoch": 3.1822362173239367,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040067239925061784,
      "loss": 5.5575,
      "mean_token_accuracy": 0.190545853972435,
      "num_tokens": 62279164.0,
      "step": 31835
    },
    {
      "entropy": 6.047640705108643,
      "epoch": 3.1827360423851654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004006427944559144,
      "loss": 5.5661,
      "mean_token_accuracy": 0.1844073086977005,
      "num_tokens": 62289860.0,
      "step": 31840
    },
    {
      "entropy": 6.025404977798462,
      "epoch": 3.1832358674463936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040061318650007045,
      "loss": 5.5956,
      "mean_token_accuracy": 0.18245840817689896,
      "num_tokens": 62300543.0,
      "step": 31845
    },
    {
      "entropy": 5.983715391159057,
      "epoch": 3.1837356925076223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040058357538383117,
      "loss": 5.5334,
      "mean_token_accuracy": 0.19371493011713029,
      "num_tokens": 62310892.0,
      "step": 31850
    },
    {
      "entropy": 6.050567150115967,
      "epoch": 3.184235517568851,
      "grad_norm": 1.125,
      "learning_rate": 0.0004005539611079414,
      "loss": 5.5004,
      "mean_token_accuracy": 0.18854521811008454,
      "num_tokens": 62320139.0,
      "step": 31855
    },
    {
      "entropy": 5.955877161026001,
      "epoch": 3.1847353426300793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040052434367314634,
      "loss": 5.4987,
      "mean_token_accuracy": 0.19010785073041916,
      "num_tokens": 62330810.0,
      "step": 31860
    },
    {
      "entropy": 6.05691123008728,
      "epoch": 3.185235167691308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040049472308019104,
      "loss": 5.5667,
      "mean_token_accuracy": 0.1838898852467537,
      "num_tokens": 62340609.0,
      "step": 31865
    },
    {
      "entropy": 6.0132509708404545,
      "epoch": 3.1857349927525367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040046509932982076,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19061934649944307,
      "num_tokens": 62351714.0,
      "step": 31870
    },
    {
      "entropy": 6.05795464515686,
      "epoch": 3.1862348178137654,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004004354724227809,
      "loss": 5.4908,
      "mean_token_accuracy": 0.19789151847362518,
      "num_tokens": 62361965.0,
      "step": 31875
    },
    {
      "entropy": 6.0505249977111815,
      "epoch": 3.1867346428749936,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004004058423598168,
      "loss": 5.5652,
      "mean_token_accuracy": 0.18051865994930266,
      "num_tokens": 62371215.0,
      "step": 31880
    },
    {
      "entropy": 6.001713705062866,
      "epoch": 3.1872344679362223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040037620914167406,
      "loss": 5.5204,
      "mean_token_accuracy": 0.18925517201423644,
      "num_tokens": 62381497.0,
      "step": 31885
    },
    {
      "entropy": 5.963173294067383,
      "epoch": 3.187734292997451,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004003465727690981,
      "loss": 5.4492,
      "mean_token_accuracy": 0.1988639533519745,
      "num_tokens": 62390852.0,
      "step": 31890
    },
    {
      "entropy": 6.004897356033325,
      "epoch": 3.1882341180586793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040031693324283464,
      "loss": 5.5314,
      "mean_token_accuracy": 0.18202067762613297,
      "num_tokens": 62400444.0,
      "step": 31895
    },
    {
      "entropy": 6.069366550445556,
      "epoch": 3.188733943119908,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004002872905636294,
      "loss": 5.5949,
      "mean_token_accuracy": 0.17474806010723115,
      "num_tokens": 62410377.0,
      "step": 31900
    },
    {
      "entropy": 5.9684072971344,
      "epoch": 3.1892337681811367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040025764473222816,
      "loss": 5.4413,
      "mean_token_accuracy": 0.18688780814409256,
      "num_tokens": 62420943.0,
      "step": 31905
    },
    {
      "entropy": 5.904835653305054,
      "epoch": 3.189733593242365,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040022799574937674,
      "loss": 5.4018,
      "mean_token_accuracy": 0.2061727061867714,
      "num_tokens": 62430090.0,
      "step": 31910
    },
    {
      "entropy": 5.893175077438355,
      "epoch": 3.1902334183035936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040019834361582123,
      "loss": 5.406,
      "mean_token_accuracy": 0.20047574788331984,
      "num_tokens": 62439311.0,
      "step": 31915
    },
    {
      "entropy": 5.963905620574951,
      "epoch": 3.1907332433648223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040016868833230766,
      "loss": 5.5108,
      "mean_token_accuracy": 0.19143410474061967,
      "num_tokens": 62450600.0,
      "step": 31920
    },
    {
      "entropy": 5.987760353088379,
      "epoch": 3.191233068426051,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004001390298995821,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1849878638982773,
      "num_tokens": 62460483.0,
      "step": 31925
    },
    {
      "entropy": 6.029374027252198,
      "epoch": 3.1917328934872793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040010936831839065,
      "loss": 5.467,
      "mean_token_accuracy": 0.19338818341493608,
      "num_tokens": 62469903.0,
      "step": 31930
    },
    {
      "entropy": 5.940745115280151,
      "epoch": 3.192232718548508,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040007970358947977,
      "loss": 5.3737,
      "mean_token_accuracy": 0.19858913719654084,
      "num_tokens": 62479524.0,
      "step": 31935
    },
    {
      "entropy": 6.009343004226684,
      "epoch": 3.1927325436097367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040005003571359565,
      "loss": 5.6022,
      "mean_token_accuracy": 0.1812085211277008,
      "num_tokens": 62489925.0,
      "step": 31940
    },
    {
      "entropy": 6.001460266113281,
      "epoch": 3.193232368670965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040002036469148486,
      "loss": 5.471,
      "mean_token_accuracy": 0.19387203007936477,
      "num_tokens": 62499261.0,
      "step": 31945
    },
    {
      "entropy": 6.078736305236816,
      "epoch": 3.1937321937321936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039999069052389384,
      "loss": 5.6155,
      "mean_token_accuracy": 0.18424918055534362,
      "num_tokens": 62509513.0,
      "step": 31950
    },
    {
      "entropy": 6.020767307281494,
      "epoch": 3.1942320187934223,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039996101321156926,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1959834724664688,
      "num_tokens": 62518711.0,
      "step": 31955
    },
    {
      "entropy": 6.001788139343262,
      "epoch": 3.194731843854651,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039993133275525777,
      "loss": 5.4909,
      "mean_token_accuracy": 0.18603707551956178,
      "num_tokens": 62529495.0,
      "step": 31960
    },
    {
      "entropy": 6.012000799179077,
      "epoch": 3.1952316689158793,
      "grad_norm": 1.109375,
      "learning_rate": 0.000399901649155706,
      "loss": 5.587,
      "mean_token_accuracy": 0.18492999374866487,
      "num_tokens": 62539263.0,
      "step": 31965
    },
    {
      "entropy": 6.080004024505615,
      "epoch": 3.195731493977108,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039987196241366086,
      "loss": 5.6079,
      "mean_token_accuracy": 0.18579466342926027,
      "num_tokens": 62548708.0,
      "step": 31970
    },
    {
      "entropy": 6.0800049781799315,
      "epoch": 3.1962313190383367,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039984227252986936,
      "loss": 5.4601,
      "mean_token_accuracy": 0.1935839831829071,
      "num_tokens": 62558051.0,
      "step": 31975
    },
    {
      "entropy": 6.019784069061279,
      "epoch": 3.196731144099565,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039981257950507835,
      "loss": 5.5697,
      "mean_token_accuracy": 0.18945035934448243,
      "num_tokens": 62568741.0,
      "step": 31980
    },
    {
      "entropy": 5.8889007568359375,
      "epoch": 3.1972309691607936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000399782883340035,
      "loss": 5.3911,
      "mean_token_accuracy": 0.2052636221051216,
      "num_tokens": 62578695.0,
      "step": 31985
    },
    {
      "entropy": 5.964903831481934,
      "epoch": 3.1977307942220223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003997531840354865,
      "loss": 5.3249,
      "mean_token_accuracy": 0.20795344412326813,
      "num_tokens": 62588282.0,
      "step": 31990
    },
    {
      "entropy": 6.0280256271362305,
      "epoch": 3.198230619283251,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039972348159217987,
      "loss": 5.4922,
      "mean_token_accuracy": 0.19064632803201675,
      "num_tokens": 62597502.0,
      "step": 31995
    },
    {
      "entropy": 5.9432196617126465,
      "epoch": 3.1987304443444793,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039969377601086257,
      "loss": 5.5163,
      "mean_token_accuracy": 0.19158889651298522,
      "num_tokens": 62607782.0,
      "step": 32000
    },
    {
      "entropy": 5.960555791854858,
      "epoch": 3.199230269405708,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039966406729228187,
      "loss": 5.4579,
      "mean_token_accuracy": 0.19795658439397812,
      "num_tokens": 62617270.0,
      "step": 32005
    },
    {
      "entropy": 6.0226891994476315,
      "epoch": 3.1997300944669367,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039963435543718544,
      "loss": 5.4627,
      "mean_token_accuracy": 0.19963742196559905,
      "num_tokens": 62627060.0,
      "step": 32010
    },
    {
      "entropy": 6.084901905059814,
      "epoch": 3.200229919528165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039960464044632067,
      "loss": 5.6318,
      "mean_token_accuracy": 0.1780024826526642,
      "num_tokens": 62637510.0,
      "step": 32015
    },
    {
      "entropy": 6.034638166427612,
      "epoch": 3.2007297445893936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003995749223204351,
      "loss": 5.477,
      "mean_token_accuracy": 0.1903046578168869,
      "num_tokens": 62646666.0,
      "step": 32020
    },
    {
      "entropy": 5.9316857814788815,
      "epoch": 3.2012295696506223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003995452010602767,
      "loss": 5.515,
      "mean_token_accuracy": 0.19373775869607926,
      "num_tokens": 62655909.0,
      "step": 32025
    },
    {
      "entropy": 5.89943380355835,
      "epoch": 3.201729394711851,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039951547666659305,
      "loss": 5.418,
      "mean_token_accuracy": 0.19285477995872496,
      "num_tokens": 62664405.0,
      "step": 32030
    },
    {
      "entropy": 6.013440704345703,
      "epoch": 3.2022292197730793,
      "grad_norm": 1.046875,
      "learning_rate": 0.000399485749140132,
      "loss": 5.4831,
      "mean_token_accuracy": 0.18693702518939972,
      "num_tokens": 62673813.0,
      "step": 32035
    },
    {
      "entropy": 5.99118480682373,
      "epoch": 3.202729044834308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039945601848164153,
      "loss": 5.491,
      "mean_token_accuracy": 0.18628634214401246,
      "num_tokens": 62683711.0,
      "step": 32040
    },
    {
      "entropy": 5.950844621658325,
      "epoch": 3.2032288698955367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003994262846918697,
      "loss": 5.4554,
      "mean_token_accuracy": 0.1964090198278427,
      "num_tokens": 62693868.0,
      "step": 32045
    },
    {
      "entropy": 6.048493051528931,
      "epoch": 3.203728694956765,
      "grad_norm": 1.125,
      "learning_rate": 0.00039939654777156454,
      "loss": 5.5737,
      "mean_token_accuracy": 0.17662234157323836,
      "num_tokens": 62705459.0,
      "step": 32050
    },
    {
      "entropy": 6.076692152023315,
      "epoch": 3.2042285200179936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003993668077214742,
      "loss": 5.6063,
      "mean_token_accuracy": 0.18323782980442047,
      "num_tokens": 62715725.0,
      "step": 32055
    },
    {
      "entropy": 5.941365337371826,
      "epoch": 3.2047283450792223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003993370645423471,
      "loss": 5.3858,
      "mean_token_accuracy": 0.19089406430721284,
      "num_tokens": 62725161.0,
      "step": 32060
    },
    {
      "entropy": 5.948692035675049,
      "epoch": 3.205228170140451,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039930731823493136,
      "loss": 5.4784,
      "mean_token_accuracy": 0.18905194401741027,
      "num_tokens": 62733834.0,
      "step": 32065
    },
    {
      "entropy": 6.023223876953125,
      "epoch": 3.2057279952016793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039927756879997556,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18696535080671312,
      "num_tokens": 62743875.0,
      "step": 32070
    },
    {
      "entropy": 6.045221614837646,
      "epoch": 3.206227820262908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039924781623822803,
      "loss": 5.4846,
      "mean_token_accuracy": 0.1964882269501686,
      "num_tokens": 62753354.0,
      "step": 32075
    },
    {
      "entropy": 5.9891743659973145,
      "epoch": 3.2067276453241367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003992180605504374,
      "loss": 5.5111,
      "mean_token_accuracy": 0.19140659868717194,
      "num_tokens": 62763231.0,
      "step": 32080
    },
    {
      "entropy": 6.023884201049805,
      "epoch": 3.207227470385365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003991883017373523,
      "loss": 5.5705,
      "mean_token_accuracy": 0.18964013159275056,
      "num_tokens": 62773650.0,
      "step": 32085
    },
    {
      "entropy": 6.079916763305664,
      "epoch": 3.2077272954465936,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003991585397997216,
      "loss": 5.5183,
      "mean_token_accuracy": 0.19010755717754363,
      "num_tokens": 62783021.0,
      "step": 32090
    },
    {
      "entropy": 6.010465860366821,
      "epoch": 3.2082271205078223,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000399128774738294,
      "loss": 5.4975,
      "mean_token_accuracy": 0.19169914573431016,
      "num_tokens": 62793698.0,
      "step": 32095
    },
    {
      "entropy": 6.062621784210205,
      "epoch": 3.208726945569051,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003990990065538183,
      "loss": 5.5537,
      "mean_token_accuracy": 0.1905689761042595,
      "num_tokens": 62802742.0,
      "step": 32100
    },
    {
      "entropy": 6.021880769729615,
      "epoch": 3.2092267706302793,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003990692352470435,
      "loss": 5.503,
      "mean_token_accuracy": 0.18737844973802567,
      "num_tokens": 62812809.0,
      "step": 32105
    },
    {
      "entropy": 5.953884267807007,
      "epoch": 3.209726595691508,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039903946081871865,
      "loss": 5.4075,
      "mean_token_accuracy": 0.1973062515258789,
      "num_tokens": 62821577.0,
      "step": 32110
    },
    {
      "entropy": 5.85590238571167,
      "epoch": 3.2102264207527367,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039900968326959295,
      "loss": 5.3908,
      "mean_token_accuracy": 0.19679388850927354,
      "num_tokens": 62830319.0,
      "step": 32115
    },
    {
      "entropy": 6.0142134666442875,
      "epoch": 3.210726245813965,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039897990260041557,
      "loss": 5.5358,
      "mean_token_accuracy": 0.19037041813135147,
      "num_tokens": 62840571.0,
      "step": 32120
    },
    {
      "entropy": 6.0700150489807125,
      "epoch": 3.2112260708751936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003989501188119357,
      "loss": 5.4731,
      "mean_token_accuracy": 0.19494474083185195,
      "num_tokens": 62850916.0,
      "step": 32125
    },
    {
      "entropy": 5.963374471664428,
      "epoch": 3.2117258959364223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039892033190490274,
      "loss": 5.5171,
      "mean_token_accuracy": 0.19146075695753098,
      "num_tokens": 62860952.0,
      "step": 32130
    },
    {
      "entropy": 5.928944444656372,
      "epoch": 3.212225720997651,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039889054188006615,
      "loss": 5.3787,
      "mean_token_accuracy": 0.2047978863120079,
      "num_tokens": 62871097.0,
      "step": 32135
    },
    {
      "entropy": 5.958164548873901,
      "epoch": 3.2127255460588793,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003988607487381753,
      "loss": 5.4303,
      "mean_token_accuracy": 0.1959997221827507,
      "num_tokens": 62881542.0,
      "step": 32140
    },
    {
      "entropy": 6.015030860900879,
      "epoch": 3.213225371120108,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003988309524799799,
      "loss": 5.4158,
      "mean_token_accuracy": 0.19832359850406647,
      "num_tokens": 62891690.0,
      "step": 32145
    },
    {
      "entropy": 5.938912391662598,
      "epoch": 3.2137251961813367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039880115310622974,
      "loss": 5.488,
      "mean_token_accuracy": 0.18622808903455734,
      "num_tokens": 62901663.0,
      "step": 32150
    },
    {
      "entropy": 5.9921567916870115,
      "epoch": 3.214225021242565,
      "grad_norm": 1.125,
      "learning_rate": 0.00039877135061767424,
      "loss": 5.5382,
      "mean_token_accuracy": 0.1858345240354538,
      "num_tokens": 62911782.0,
      "step": 32155
    },
    {
      "entropy": 6.1053296566009525,
      "epoch": 3.2147248463037936,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003987415450150635,
      "loss": 5.5167,
      "mean_token_accuracy": 0.1953195869922638,
      "num_tokens": 62922803.0,
      "step": 32160
    },
    {
      "entropy": 6.060425424575806,
      "epoch": 3.2152246713650223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039871173629914736,
      "loss": 5.5184,
      "mean_token_accuracy": 0.19086491614580153,
      "num_tokens": 62932166.0,
      "step": 32165
    },
    {
      "entropy": 5.98786039352417,
      "epoch": 3.215724496426251,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039868192447067575,
      "loss": 5.561,
      "mean_token_accuracy": 0.18916563242673873,
      "num_tokens": 62941722.0,
      "step": 32170
    },
    {
      "entropy": 5.989275693893433,
      "epoch": 3.2162243214874793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039865210953039863,
      "loss": 5.5156,
      "mean_token_accuracy": 0.1853766217827797,
      "num_tokens": 62951148.0,
      "step": 32175
    },
    {
      "entropy": 5.9706333637237545,
      "epoch": 3.216724146548708,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003986222914790664,
      "loss": 5.4937,
      "mean_token_accuracy": 0.18994984030723572,
      "num_tokens": 62961156.0,
      "step": 32180
    },
    {
      "entropy": 5.957022571563721,
      "epoch": 3.2172239716099367,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00039859247031742907,
      "loss": 5.4434,
      "mean_token_accuracy": 0.1974493980407715,
      "num_tokens": 62970902.0,
      "step": 32185
    },
    {
      "entropy": 5.9225266456604,
      "epoch": 3.217723796671165,
      "grad_norm": 1.203125,
      "learning_rate": 0.000398562646046237,
      "loss": 5.4849,
      "mean_token_accuracy": 0.1864173009991646,
      "num_tokens": 62980137.0,
      "step": 32190
    },
    {
      "entropy": 6.03641791343689,
      "epoch": 3.2182236217323936,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039853281866624054,
      "loss": 5.5744,
      "mean_token_accuracy": 0.18732248842716218,
      "num_tokens": 62989957.0,
      "step": 32195
    },
    {
      "entropy": 6.054545497894287,
      "epoch": 3.2187234467936223,
      "grad_norm": 1.125,
      "learning_rate": 0.00039850298817819017,
      "loss": 5.4785,
      "mean_token_accuracy": 0.1943720191717148,
      "num_tokens": 62999043.0,
      "step": 32200
    },
    {
      "entropy": 5.984670352935791,
      "epoch": 3.219223271854851,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003984731545828364,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1913386255502701,
      "num_tokens": 63008858.0,
      "step": 32205
    },
    {
      "entropy": 5.972563695907593,
      "epoch": 3.2197230969160793,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039844331788092986,
      "loss": 5.514,
      "mean_token_accuracy": 0.18735232055187226,
      "num_tokens": 63017979.0,
      "step": 32210
    },
    {
      "entropy": 6.026068687438965,
      "epoch": 3.220222921977308,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003984134780732211,
      "loss": 5.5279,
      "mean_token_accuracy": 0.1798887550830841,
      "num_tokens": 63027226.0,
      "step": 32215
    },
    {
      "entropy": 6.048770523071289,
      "epoch": 3.2207227470385367,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003983836351604611,
      "loss": 5.5297,
      "mean_token_accuracy": 0.18332682698965072,
      "num_tokens": 63036822.0,
      "step": 32220
    },
    {
      "entropy": 6.007046222686768,
      "epoch": 3.221222572099765,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003983537891434005,
      "loss": 5.4984,
      "mean_token_accuracy": 0.18744549602270127,
      "num_tokens": 63048045.0,
      "step": 32225
    },
    {
      "entropy": 5.98017897605896,
      "epoch": 3.2217223971609936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003983239400227903,
      "loss": 5.4367,
      "mean_token_accuracy": 0.18921099305152894,
      "num_tokens": 63057528.0,
      "step": 32230
    },
    {
      "entropy": 5.992935562133789,
      "epoch": 3.2222222222222223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003982940877993816,
      "loss": 5.5468,
      "mean_token_accuracy": 0.18817624747753142,
      "num_tokens": 63067111.0,
      "step": 32235
    },
    {
      "entropy": 5.966059827804566,
      "epoch": 3.222722047283451,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039826423247392537,
      "loss": 5.4624,
      "mean_token_accuracy": 0.1916034385561943,
      "num_tokens": 63078350.0,
      "step": 32240
    },
    {
      "entropy": 5.989551830291748,
      "epoch": 3.2232218723446793,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003982343740471727,
      "loss": 5.5291,
      "mean_token_accuracy": 0.19114856272935868,
      "num_tokens": 63087459.0,
      "step": 32245
    },
    {
      "entropy": 5.999602842330932,
      "epoch": 3.223721697405908,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039820451251987494,
      "loss": 5.4624,
      "mean_token_accuracy": 0.18812587559223176,
      "num_tokens": 63097095.0,
      "step": 32250
    },
    {
      "entropy": 5.9898241519927975,
      "epoch": 3.2242215224671367,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039817464789278326,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19204442352056503,
      "num_tokens": 63106545.0,
      "step": 32255
    },
    {
      "entropy": 5.866599702835083,
      "epoch": 3.224721347528365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039814478016664923,
      "loss": 5.4003,
      "mean_token_accuracy": 0.20224519371986388,
      "num_tokens": 63115624.0,
      "step": 32260
    },
    {
      "entropy": 6.031519985198974,
      "epoch": 3.2252211725895936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039811490934222416,
      "loss": 5.585,
      "mean_token_accuracy": 0.18276869952678682,
      "num_tokens": 63126931.0,
      "step": 32265
    },
    {
      "entropy": 6.035090017318725,
      "epoch": 3.2257209976508223,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003980850354202598,
      "loss": 5.4998,
      "mean_token_accuracy": 0.19660528600215912,
      "num_tokens": 63137837.0,
      "step": 32270
    },
    {
      "entropy": 6.001530122756958,
      "epoch": 3.226220822712051,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039805515840150743,
      "loss": 5.4755,
      "mean_token_accuracy": 0.1905677706003189,
      "num_tokens": 63147361.0,
      "step": 32275
    },
    {
      "entropy": 5.803991365432739,
      "epoch": 3.2267206477732793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039802527828671896,
      "loss": 5.2607,
      "mean_token_accuracy": 0.2143144428730011,
      "num_tokens": 63158059.0,
      "step": 32280
    },
    {
      "entropy": 6.006458044052124,
      "epoch": 3.227220472834508,
      "grad_norm": 1.125,
      "learning_rate": 0.0003979953950766462,
      "loss": 5.5147,
      "mean_token_accuracy": 0.19278127253055571,
      "num_tokens": 63167970.0,
      "step": 32285
    },
    {
      "entropy": 5.972282409667969,
      "epoch": 3.2277202978957367,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039796550877204095,
      "loss": 5.4026,
      "mean_token_accuracy": 0.20632738918066024,
      "num_tokens": 63176532.0,
      "step": 32290
    },
    {
      "entropy": 6.0848020076751705,
      "epoch": 3.228220122956965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039793561937365515,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17221685498952866,
      "num_tokens": 63185734.0,
      "step": 32295
    },
    {
      "entropy": 5.968096971511841,
      "epoch": 3.2287199480181936,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039790572688224075,
      "loss": 5.4183,
      "mean_token_accuracy": 0.1946742296218872,
      "num_tokens": 63195458.0,
      "step": 32300
    },
    {
      "entropy": 6.036428785324096,
      "epoch": 3.2292197730794223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039787583129854997,
      "loss": 5.5342,
      "mean_token_accuracy": 0.18552290350198747,
      "num_tokens": 63205248.0,
      "step": 32305
    },
    {
      "entropy": 5.976945781707764,
      "epoch": 3.2297195981406506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003978459326233349,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19879602640867233,
      "num_tokens": 63215327.0,
      "step": 32310
    },
    {
      "entropy": 6.082912445068359,
      "epoch": 3.2302194232018793,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003978160308573477,
      "loss": 5.6375,
      "mean_token_accuracy": 0.18219879120588303,
      "num_tokens": 63225142.0,
      "step": 32315
    },
    {
      "entropy": 6.003087043762207,
      "epoch": 3.230719248263108,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003977861260013407,
      "loss": 5.5206,
      "mean_token_accuracy": 0.18975384831428527,
      "num_tokens": 63233825.0,
      "step": 32320
    },
    {
      "entropy": 6.024193572998047,
      "epoch": 3.2312190733243367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003977562180560665,
      "loss": 5.53,
      "mean_token_accuracy": 0.1859939157962799,
      "num_tokens": 63243389.0,
      "step": 32325
    },
    {
      "entropy": 5.986633825302124,
      "epoch": 3.231718898385565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003977263070222774,
      "loss": 5.4875,
      "mean_token_accuracy": 0.18611157387495042,
      "num_tokens": 63253684.0,
      "step": 32330
    },
    {
      "entropy": 5.997208929061889,
      "epoch": 3.2322187234467936,
      "grad_norm": 1.21875,
      "learning_rate": 0.000397696392900726,
      "loss": 5.5181,
      "mean_token_accuracy": 0.18613187074661255,
      "num_tokens": 63263463.0,
      "step": 32335
    },
    {
      "entropy": 5.910247325897217,
      "epoch": 3.2327185485080223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003976664756921649,
      "loss": 5.4104,
      "mean_token_accuracy": 0.2028303012251854,
      "num_tokens": 63273274.0,
      "step": 32340
    },
    {
      "entropy": 5.998226356506348,
      "epoch": 3.2332183735692506,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003976365553973469,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19398606121540068,
      "num_tokens": 63282429.0,
      "step": 32345
    },
    {
      "entropy": 5.9380542755126955,
      "epoch": 3.2337181986304793,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003976066320170247,
      "loss": 5.3507,
      "mean_token_accuracy": 0.2035409078001976,
      "num_tokens": 63292356.0,
      "step": 32350
    },
    {
      "entropy": 5.951413106918335,
      "epoch": 3.234218023691708,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039757670555195114,
      "loss": 5.5617,
      "mean_token_accuracy": 0.1855926439166069,
      "num_tokens": 63302852.0,
      "step": 32355
    },
    {
      "entropy": 5.929383850097656,
      "epoch": 3.2347178487529367,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039754677600287923,
      "loss": 5.4591,
      "mean_token_accuracy": 0.18881044834852218,
      "num_tokens": 63313557.0,
      "step": 32360
    },
    {
      "entropy": 6.033724737167359,
      "epoch": 3.235217673814165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039751684337056197,
      "loss": 5.4735,
      "mean_token_accuracy": 0.1898307427763939,
      "num_tokens": 63323248.0,
      "step": 32365
    },
    {
      "entropy": 6.076234149932861,
      "epoch": 3.2357174988753936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003974869076557525,
      "loss": 5.528,
      "mean_token_accuracy": 0.18643543422222136,
      "num_tokens": 63333698.0,
      "step": 32370
    },
    {
      "entropy": 5.9684028148651125,
      "epoch": 3.2362173239366223,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003974569688592038,
      "loss": 5.4703,
      "mean_token_accuracy": 0.18866670429706572,
      "num_tokens": 63343017.0,
      "step": 32375
    },
    {
      "entropy": 5.98458514213562,
      "epoch": 3.2367171489978506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003974270269816695,
      "loss": 5.4773,
      "mean_token_accuracy": 0.1887152835726738,
      "num_tokens": 63352165.0,
      "step": 32380
    },
    {
      "entropy": 6.052820301055908,
      "epoch": 3.2372169740590793,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039739708202390255,
      "loss": 5.5366,
      "mean_token_accuracy": 0.1870967850089073,
      "num_tokens": 63360724.0,
      "step": 32385
    },
    {
      "entropy": 5.953455018997192,
      "epoch": 3.237716799120308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003973671339866566,
      "loss": 5.5363,
      "mean_token_accuracy": 0.1856708586215973,
      "num_tokens": 63370352.0,
      "step": 32390
    },
    {
      "entropy": 5.955522251129151,
      "epoch": 3.2382166241815367,
      "grad_norm": 1.7734375,
      "learning_rate": 0.00039733718287068495,
      "loss": 5.3964,
      "mean_token_accuracy": 0.1983047232031822,
      "num_tokens": 63380155.0,
      "step": 32395
    },
    {
      "entropy": 5.89001612663269,
      "epoch": 3.238716449242765,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003973072286767413,
      "loss": 5.368,
      "mean_token_accuracy": 0.20648770928382873,
      "num_tokens": 63389472.0,
      "step": 32400
    },
    {
      "entropy": 5.946518754959106,
      "epoch": 3.2392162743039936,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003972772714055793,
      "loss": 5.495,
      "mean_token_accuracy": 0.19421089440584183,
      "num_tokens": 63400004.0,
      "step": 32405
    },
    {
      "entropy": 5.9514992237091064,
      "epoch": 3.2397160993652223,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039724731105795255,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19104635566473008,
      "num_tokens": 63410685.0,
      "step": 32410
    },
    {
      "entropy": 5.959885215759277,
      "epoch": 3.2402159244264506,
      "grad_norm": 1.171875,
      "learning_rate": 0.000397217347634615,
      "loss": 5.4277,
      "mean_token_accuracy": 0.1994063138961792,
      "num_tokens": 63419659.0,
      "step": 32415
    },
    {
      "entropy": 5.855419111251831,
      "epoch": 3.2407157494876793,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003971873811363203,
      "loss": 5.3157,
      "mean_token_accuracy": 0.20757546424865722,
      "num_tokens": 63430740.0,
      "step": 32420
    },
    {
      "entropy": 6.031111049652099,
      "epoch": 3.241215574548908,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039715741156382255,
      "loss": 5.5926,
      "mean_token_accuracy": 0.17750538438558577,
      "num_tokens": 63440129.0,
      "step": 32425
    },
    {
      "entropy": 5.9973217964172365,
      "epoch": 3.241715399610136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003971274389178759,
      "loss": 5.5672,
      "mean_token_accuracy": 0.18295410871505738,
      "num_tokens": 63449569.0,
      "step": 32430
    },
    {
      "entropy": 6.0383491039276125,
      "epoch": 3.242215224671365,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003970974631992342,
      "loss": 5.5126,
      "mean_token_accuracy": 0.1865051805973053,
      "num_tokens": 63458814.0,
      "step": 32435
    },
    {
      "entropy": 6.038397169113159,
      "epoch": 3.2427150497325936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003970674844086517,
      "loss": 5.5435,
      "mean_token_accuracy": 0.1862851560115814,
      "num_tokens": 63468507.0,
      "step": 32440
    },
    {
      "entropy": 6.019868326187134,
      "epoch": 3.2432148747938223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003970375025468829,
      "loss": 5.5513,
      "mean_token_accuracy": 0.1878555089235306,
      "num_tokens": 63477934.0,
      "step": 32445
    },
    {
      "entropy": 5.968447351455689,
      "epoch": 3.2437146998550506,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039700751761468175,
      "loss": 5.4745,
      "mean_token_accuracy": 0.1931324064731598,
      "num_tokens": 63488799.0,
      "step": 32450
    },
    {
      "entropy": 6.032505846023559,
      "epoch": 3.2442145249162793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003969775296128029,
      "loss": 5.5103,
      "mean_token_accuracy": 0.18681611567735673,
      "num_tokens": 63497664.0,
      "step": 32455
    },
    {
      "entropy": 5.961713123321533,
      "epoch": 3.244714349977508,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003969475385420008,
      "loss": 5.4727,
      "mean_token_accuracy": 0.18960687667131423,
      "num_tokens": 63507775.0,
      "step": 32460
    },
    {
      "entropy": 5.974391126632691,
      "epoch": 3.245214175038736,
      "grad_norm": 1.25,
      "learning_rate": 0.00039691754440303003,
      "loss": 5.4773,
      "mean_token_accuracy": 0.18564170151948928,
      "num_tokens": 63517379.0,
      "step": 32465
    },
    {
      "entropy": 6.051361465454102,
      "epoch": 3.245714000099965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039688754719664516,
      "loss": 5.5107,
      "mean_token_accuracy": 0.18585877120494843,
      "num_tokens": 63528012.0,
      "step": 32470
    },
    {
      "entropy": 6.056641101837158,
      "epoch": 3.2462138251611936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000396857546923601,
      "loss": 5.5555,
      "mean_token_accuracy": 0.19089751094579696,
      "num_tokens": 63537226.0,
      "step": 32475
    },
    {
      "entropy": 6.128174257278443,
      "epoch": 3.2467136502224223,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003968275435846523,
      "loss": 5.6274,
      "mean_token_accuracy": 0.17660630345344544,
      "num_tokens": 63548470.0,
      "step": 32480
    },
    {
      "entropy": 6.125441265106201,
      "epoch": 3.2472134752836506,
      "grad_norm": 1.125,
      "learning_rate": 0.000396797537180554,
      "loss": 5.6466,
      "mean_token_accuracy": 0.18191723227500917,
      "num_tokens": 63559254.0,
      "step": 32485
    },
    {
      "entropy": 5.9777124404907225,
      "epoch": 3.2477133003448793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000396767527712061,
      "loss": 5.4217,
      "mean_token_accuracy": 0.19688333123922347,
      "num_tokens": 63568772.0,
      "step": 32490
    },
    {
      "entropy": 5.948632287979126,
      "epoch": 3.248213125406108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039673751517992827,
      "loss": 5.515,
      "mean_token_accuracy": 0.18602924048900604,
      "num_tokens": 63579331.0,
      "step": 32495
    },
    {
      "entropy": 5.989554977416992,
      "epoch": 3.248712950467336,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039670749958491103,
      "loss": 5.4629,
      "mean_token_accuracy": 0.1952543705701828,
      "num_tokens": 63588616.0,
      "step": 32500
    },
    {
      "entropy": 5.982357597351074,
      "epoch": 3.249212775528565,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039667748092776453,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19770683199167252,
      "num_tokens": 63598989.0,
      "step": 32505
    },
    {
      "entropy": 5.957918548583985,
      "epoch": 3.2497126005897936,
      "grad_norm": 1.25,
      "learning_rate": 0.00039664745920924374,
      "loss": 5.4463,
      "mean_token_accuracy": 0.19346819669008256,
      "num_tokens": 63609568.0,
      "step": 32510
    },
    {
      "entropy": 5.991935920715332,
      "epoch": 3.2502124256510223,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003966174344301043,
      "loss": 5.5619,
      "mean_token_accuracy": 0.18604740649461746,
      "num_tokens": 63619561.0,
      "step": 32515
    },
    {
      "entropy": 6.057579898834229,
      "epoch": 3.2507122507122506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003965874065911014,
      "loss": 5.5754,
      "mean_token_accuracy": 0.18293206989765168,
      "num_tokens": 63629881.0,
      "step": 32520
    },
    {
      "entropy": 6.038230895996094,
      "epoch": 3.2512120757734793,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039655737569299073,
      "loss": 5.579,
      "mean_token_accuracy": 0.18781269192695618,
      "num_tokens": 63640897.0,
      "step": 32525
    },
    {
      "entropy": 6.001623725891113,
      "epoch": 3.251711900834708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003965273417365277,
      "loss": 5.5237,
      "mean_token_accuracy": 0.19141246676445006,
      "num_tokens": 63651031.0,
      "step": 32530
    },
    {
      "entropy": 6.00225830078125,
      "epoch": 3.252211725895936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039649730472246796,
      "loss": 5.3926,
      "mean_token_accuracy": 0.195442833006382,
      "num_tokens": 63661961.0,
      "step": 32535
    },
    {
      "entropy": 5.852300930023193,
      "epoch": 3.252711550957165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003964672646515674,
      "loss": 5.2587,
      "mean_token_accuracy": 0.2156531482934952,
      "num_tokens": 63672495.0,
      "step": 32540
    },
    {
      "entropy": 5.833238172531128,
      "epoch": 3.2532113760183936,
      "grad_norm": 1.578125,
      "learning_rate": 0.0003964372215245817,
      "loss": 5.369,
      "mean_token_accuracy": 0.20621465146541595,
      "num_tokens": 63682792.0,
      "step": 32545
    },
    {
      "entropy": 6.0130236625671385,
      "epoch": 3.2537112010796223,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039640717534226684,
      "loss": 5.56,
      "mean_token_accuracy": 0.18728429675102234,
      "num_tokens": 63692940.0,
      "step": 32550
    },
    {
      "entropy": 6.098140859603882,
      "epoch": 3.2542110261408506,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003963771261053786,
      "loss": 5.5744,
      "mean_token_accuracy": 0.18343122750520707,
      "num_tokens": 63703347.0,
      "step": 32555
    },
    {
      "entropy": 6.024584197998047,
      "epoch": 3.2547108512020793,
      "grad_norm": 1.125,
      "learning_rate": 0.00039634707381467303,
      "loss": 5.5715,
      "mean_token_accuracy": 0.18891411721706391,
      "num_tokens": 63713776.0,
      "step": 32560
    },
    {
      "entropy": 6.0263099193573,
      "epoch": 3.255210676263308,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003963170184709064,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18217258900403976,
      "num_tokens": 63723507.0,
      "step": 32565
    },
    {
      "entropy": 5.961766052246094,
      "epoch": 3.255710501324536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039628696007483477,
      "loss": 5.3999,
      "mean_token_accuracy": 0.19846896976232528,
      "num_tokens": 63733278.0,
      "step": 32570
    },
    {
      "entropy": 5.894601631164551,
      "epoch": 3.256210326385765,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039625689862721447,
      "loss": 5.3784,
      "mean_token_accuracy": 0.21175291538238525,
      "num_tokens": 63742205.0,
      "step": 32575
    },
    {
      "entropy": 5.983277273178101,
      "epoch": 3.2567101514469936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039622683412880174,
      "loss": 5.5451,
      "mean_token_accuracy": 0.18237304091453552,
      "num_tokens": 63752908.0,
      "step": 32580
    },
    {
      "entropy": 6.035596036911011,
      "epoch": 3.2572099765082223,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003961967665803532,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18463903814554214,
      "num_tokens": 63762682.0,
      "step": 32585
    },
    {
      "entropy": 5.993989944458008,
      "epoch": 3.2577098015694506,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039616669598262504,
      "loss": 5.4566,
      "mean_token_accuracy": 0.19428373724222184,
      "num_tokens": 63771757.0,
      "step": 32590
    },
    {
      "entropy": 5.985429859161377,
      "epoch": 3.2582096266306793,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039613662233637406,
      "loss": 5.5022,
      "mean_token_accuracy": 0.19522169828414918,
      "num_tokens": 63780805.0,
      "step": 32595
    },
    {
      "entropy": 5.989378213882446,
      "epoch": 3.258709451691908,
      "grad_norm": 1.125,
      "learning_rate": 0.00039610654564235684,
      "loss": 5.4878,
      "mean_token_accuracy": 0.1871519535779953,
      "num_tokens": 63790279.0,
      "step": 32600
    },
    {
      "entropy": 6.087039756774902,
      "epoch": 3.259209276753136,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003960764659013301,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18550067096948625,
      "num_tokens": 63799788.0,
      "step": 32605
    },
    {
      "entropy": 6.06182861328125,
      "epoch": 3.259709101814365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003960463831140505,
      "loss": 5.5441,
      "mean_token_accuracy": 0.19141582250595093,
      "num_tokens": 63810020.0,
      "step": 32610
    },
    {
      "entropy": 5.9647337913513185,
      "epoch": 3.2602089268755936,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039601629728127524,
      "loss": 5.5073,
      "mean_token_accuracy": 0.19573838263750076,
      "num_tokens": 63819174.0,
      "step": 32615
    },
    {
      "entropy": 5.9658256530761715,
      "epoch": 3.2607087519368223,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039598620840376096,
      "loss": 5.4535,
      "mean_token_accuracy": 0.1962643474340439,
      "num_tokens": 63830199.0,
      "step": 32620
    },
    {
      "entropy": 5.962610054016113,
      "epoch": 3.2612085769980506,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039595611648226485,
      "loss": 5.4656,
      "mean_token_accuracy": 0.1935232847929001,
      "num_tokens": 63839570.0,
      "step": 32625
    },
    {
      "entropy": 5.941076993942261,
      "epoch": 3.2617084020592793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000395926021517544,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1949406757950783,
      "num_tokens": 63848817.0,
      "step": 32630
    },
    {
      "entropy": 6.02610673904419,
      "epoch": 3.262208227120508,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039589592351035554,
      "loss": 5.5262,
      "mean_token_accuracy": 0.19057662338018416,
      "num_tokens": 63858057.0,
      "step": 32635
    },
    {
      "entropy": 6.005493116378784,
      "epoch": 3.262708052181736,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003958658224614567,
      "loss": 5.5026,
      "mean_token_accuracy": 0.1963416889309883,
      "num_tokens": 63868105.0,
      "step": 32640
    },
    {
      "entropy": 5.958808040618896,
      "epoch": 3.263207877242965,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039583571837160497,
      "loss": 5.5127,
      "mean_token_accuracy": 0.19089129120111464,
      "num_tokens": 63879126.0,
      "step": 32645
    },
    {
      "entropy": 5.983163976669312,
      "epoch": 3.2637077023041936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003958056112415576,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1919344797730446,
      "num_tokens": 63889045.0,
      "step": 32650
    },
    {
      "entropy": 6.0472495555877686,
      "epoch": 3.2642075273654223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003957755010720721,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18469949662685395,
      "num_tokens": 63899167.0,
      "step": 32655
    },
    {
      "entropy": 6.047087049484253,
      "epoch": 3.2647073524266506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039574538786390617,
      "loss": 5.4884,
      "mean_token_accuracy": 0.19025965929031372,
      "num_tokens": 63909591.0,
      "step": 32660
    },
    {
      "entropy": 5.984168434143067,
      "epoch": 3.2652071774878793,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003957152716178173,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1876394584774971,
      "num_tokens": 63918975.0,
      "step": 32665
    },
    {
      "entropy": 6.05690631866455,
      "epoch": 3.265707002549108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039568515233456326,
      "loss": 5.4878,
      "mean_token_accuracy": 0.18899298161268235,
      "num_tokens": 63928618.0,
      "step": 32670
    },
    {
      "entropy": 6.040109777450562,
      "epoch": 3.266206827610336,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003956550300149018,
      "loss": 5.4823,
      "mean_token_accuracy": 0.19150922298431397,
      "num_tokens": 63938581.0,
      "step": 32675
    },
    {
      "entropy": 6.052662944793701,
      "epoch": 3.266706652671565,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003956249046595909,
      "loss": 5.4474,
      "mean_token_accuracy": 0.19722742587327957,
      "num_tokens": 63947842.0,
      "step": 32680
    },
    {
      "entropy": 6.054673433303833,
      "epoch": 3.2672064777327936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003955947762693884,
      "loss": 5.5459,
      "mean_token_accuracy": 0.1805434212088585,
      "num_tokens": 63956975.0,
      "step": 32685
    },
    {
      "entropy": 6.01651406288147,
      "epoch": 3.2677063027940223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039556464484505235,
      "loss": 5.5781,
      "mean_token_accuracy": 0.18181786835193633,
      "num_tokens": 63966215.0,
      "step": 32690
    },
    {
      "entropy": 5.9864606857299805,
      "epoch": 3.2682061278552506,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003955345103873409,
      "loss": 5.448,
      "mean_token_accuracy": 0.19777975380420684,
      "num_tokens": 63975440.0,
      "step": 32695
    },
    {
      "entropy": 5.978647613525391,
      "epoch": 3.2687059529164793,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003955043728970121,
      "loss": 5.4855,
      "mean_token_accuracy": 0.1933702990412712,
      "num_tokens": 63984616.0,
      "step": 32700
    },
    {
      "entropy": 5.986061048507691,
      "epoch": 3.269205777977708,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003954742323748244,
      "loss": 5.4933,
      "mean_token_accuracy": 0.19315855205059052,
      "num_tokens": 63994792.0,
      "step": 32705
    },
    {
      "entropy": 6.043139266967773,
      "epoch": 3.269705603038936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039544408882153596,
      "loss": 5.6493,
      "mean_token_accuracy": 0.1778396725654602,
      "num_tokens": 64005068.0,
      "step": 32710
    },
    {
      "entropy": 6.0697825908660885,
      "epoch": 3.270205428100165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003954139422379052,
      "loss": 5.5792,
      "mean_token_accuracy": 0.18479794263839722,
      "num_tokens": 64014393.0,
      "step": 32715
    },
    {
      "entropy": 6.074567079544067,
      "epoch": 3.2707052531613936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003953837926246907,
      "loss": 5.4688,
      "mean_token_accuracy": 0.1887694478034973,
      "num_tokens": 64024310.0,
      "step": 32720
    },
    {
      "entropy": 6.025577449798584,
      "epoch": 3.2712050782226223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039535363998265095,
      "loss": 5.5871,
      "mean_token_accuracy": 0.18441994488239288,
      "num_tokens": 64033867.0,
      "step": 32725
    },
    {
      "entropy": 6.040678405761719,
      "epoch": 3.2717049032838506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039532348431254455,
      "loss": 5.5493,
      "mean_token_accuracy": 0.18431588709354402,
      "num_tokens": 64044548.0,
      "step": 32730
    },
    {
      "entropy": 6.125094556808472,
      "epoch": 3.2722047283450792,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003952933256151303,
      "loss": 5.6086,
      "mean_token_accuracy": 0.18137914985418319,
      "num_tokens": 64054360.0,
      "step": 32735
    },
    {
      "entropy": 5.925862979888916,
      "epoch": 3.272704553406308,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003952631638911669,
      "loss": 5.3783,
      "mean_token_accuracy": 0.19976481348276137,
      "num_tokens": 64065181.0,
      "step": 32740
    },
    {
      "entropy": 5.948255205154419,
      "epoch": 3.273204378467536,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003952329991414132,
      "loss": 5.4114,
      "mean_token_accuracy": 0.19570550918579102,
      "num_tokens": 64075052.0,
      "step": 32745
    },
    {
      "entropy": 5.983576011657715,
      "epoch": 3.273704203528765,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039520283136662825,
      "loss": 5.4797,
      "mean_token_accuracy": 0.1899937316775322,
      "num_tokens": 64083934.0,
      "step": 32750
    },
    {
      "entropy": 6.0411594867706295,
      "epoch": 3.2742040285899936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039517266056757097,
      "loss": 5.5558,
      "mean_token_accuracy": 0.17895030230283737,
      "num_tokens": 64094402.0,
      "step": 32755
    },
    {
      "entropy": 6.006614923477173,
      "epoch": 3.2747038536512223,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039514248674500055,
      "loss": 5.468,
      "mean_token_accuracy": 0.18641496002674102,
      "num_tokens": 64103681.0,
      "step": 32760
    },
    {
      "entropy": 6.013396406173706,
      "epoch": 3.2752036787124506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039511230989967594,
      "loss": 5.5031,
      "mean_token_accuracy": 0.18184670954942703,
      "num_tokens": 64113332.0,
      "step": 32765
    },
    {
      "entropy": 5.974966335296631,
      "epoch": 3.2757035037736792,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003950821300323566,
      "loss": 5.4989,
      "mean_token_accuracy": 0.19442080110311508,
      "num_tokens": 64123079.0,
      "step": 32770
    },
    {
      "entropy": 6.041613912582397,
      "epoch": 3.276203328834908,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003950519471438017,
      "loss": 5.5184,
      "mean_token_accuracy": 0.18995835781097412,
      "num_tokens": 64134034.0,
      "step": 32775
    },
    {
      "entropy": 6.032911205291748,
      "epoch": 3.276703153896136,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039502176123477076,
      "loss": 5.3184,
      "mean_token_accuracy": 0.20718421190977096,
      "num_tokens": 64144157.0,
      "step": 32780
    },
    {
      "entropy": 5.961499404907227,
      "epoch": 3.277202978957365,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039499157230602313,
      "loss": 5.5049,
      "mean_token_accuracy": 0.202034193277359,
      "num_tokens": 64154150.0,
      "step": 32785
    },
    {
      "entropy": 5.953851938247681,
      "epoch": 3.2777028040185936,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039496138035831845,
      "loss": 5.4531,
      "mean_token_accuracy": 0.1969784528017044,
      "num_tokens": 64163553.0,
      "step": 32790
    },
    {
      "entropy": 5.941063022613525,
      "epoch": 3.2782026290798223,
      "grad_norm": 1.375,
      "learning_rate": 0.00039493118539241643,
      "loss": 5.4266,
      "mean_token_accuracy": 0.19590681046247482,
      "num_tokens": 64173735.0,
      "step": 32795
    },
    {
      "entropy": 5.91037130355835,
      "epoch": 3.2787024541410505,
      "grad_norm": 1.125,
      "learning_rate": 0.00039490098740907654,
      "loss": 5.397,
      "mean_token_accuracy": 0.20096349716186523,
      "num_tokens": 64183501.0,
      "step": 32800
    },
    {
      "entropy": 5.930101537704468,
      "epoch": 3.2792022792022792,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039487078640905866,
      "loss": 5.4133,
      "mean_token_accuracy": 0.197498182952404,
      "num_tokens": 64192131.0,
      "step": 32805
    },
    {
      "entropy": 5.958723449707032,
      "epoch": 3.279702104263508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039484058239312264,
      "loss": 5.5355,
      "mean_token_accuracy": 0.19430094361305236,
      "num_tokens": 64201467.0,
      "step": 32810
    },
    {
      "entropy": 5.994084787368775,
      "epoch": 3.280201929324736,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039481037536202835,
      "loss": 5.5294,
      "mean_token_accuracy": 0.18694795966148375,
      "num_tokens": 64213159.0,
      "step": 32815
    },
    {
      "entropy": 6.045454454421997,
      "epoch": 3.280701754385965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039478016531653586,
      "loss": 5.5607,
      "mean_token_accuracy": 0.18231880962848662,
      "num_tokens": 64223627.0,
      "step": 32820
    },
    {
      "entropy": 6.074855041503906,
      "epoch": 3.2812015794471936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039474995225740526,
      "loss": 5.5239,
      "mean_token_accuracy": 0.18277339190244674,
      "num_tokens": 64232920.0,
      "step": 32825
    },
    {
      "entropy": 5.964660596847534,
      "epoch": 3.2817014045084223,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003947197361853968,
      "loss": 5.4619,
      "mean_token_accuracy": 0.19430216550827026,
      "num_tokens": 64242222.0,
      "step": 32830
    },
    {
      "entropy": 5.871334028244019,
      "epoch": 3.2822012295696505,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039468951710127043,
      "loss": 5.3334,
      "mean_token_accuracy": 0.21397134959697722,
      "num_tokens": 64252065.0,
      "step": 32835
    },
    {
      "entropy": 5.978372287750244,
      "epoch": 3.2827010546308792,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003946592950057867,
      "loss": 5.4762,
      "mean_token_accuracy": 0.19190580993890763,
      "num_tokens": 64262290.0,
      "step": 32840
    },
    {
      "entropy": 5.914823961257935,
      "epoch": 3.283200879692108,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003946290698997058,
      "loss": 5.3713,
      "mean_token_accuracy": 0.20307094752788543,
      "num_tokens": 64270828.0,
      "step": 32845
    },
    {
      "entropy": 5.994691801071167,
      "epoch": 3.283700704753336,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003945988417837883,
      "loss": 5.5868,
      "mean_token_accuracy": 0.1883083090186119,
      "num_tokens": 64281365.0,
      "step": 32850
    },
    {
      "entropy": 5.764198160171508,
      "epoch": 3.284200529814565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003945686106587948,
      "loss": 5.2251,
      "mean_token_accuracy": 0.21893788129091263,
      "num_tokens": 64290954.0,
      "step": 32855
    },
    {
      "entropy": 5.9771558284759525,
      "epoch": 3.2847003548757936,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039453837652548575,
      "loss": 5.4571,
      "mean_token_accuracy": 0.1962984323501587,
      "num_tokens": 64300720.0,
      "step": 32860
    },
    {
      "entropy": 5.999706363677978,
      "epoch": 3.2852001799370223,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039450813938462197,
      "loss": 5.5289,
      "mean_token_accuracy": 0.1884993940591812,
      "num_tokens": 64311164.0,
      "step": 32865
    },
    {
      "entropy": 6.027453947067261,
      "epoch": 3.2857000049982505,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039447789923696415,
      "loss": 5.4953,
      "mean_token_accuracy": 0.1910969838500023,
      "num_tokens": 64321792.0,
      "step": 32870
    },
    {
      "entropy": 5.994319105148316,
      "epoch": 3.2861998300594792,
      "grad_norm": 1.1875,
      "learning_rate": 0.000394447656083273,
      "loss": 5.587,
      "mean_token_accuracy": 0.17932286858558655,
      "num_tokens": 64331757.0,
      "step": 32875
    },
    {
      "entropy": 5.930206251144409,
      "epoch": 3.286699655120708,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039441740992430965,
      "loss": 5.3572,
      "mean_token_accuracy": 0.2082115963101387,
      "num_tokens": 64341923.0,
      "step": 32880
    },
    {
      "entropy": 6.044006776809693,
      "epoch": 3.287199480181936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039438716076083504,
      "loss": 5.4699,
      "mean_token_accuracy": 0.19815869480371476,
      "num_tokens": 64351729.0,
      "step": 32885
    },
    {
      "entropy": 5.938150358200073,
      "epoch": 3.287699305243165,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00039435690859361013,
      "loss": 5.4078,
      "mean_token_accuracy": 0.1976672366261482,
      "num_tokens": 64362357.0,
      "step": 32890
    },
    {
      "entropy": 5.911613655090332,
      "epoch": 3.2881991303043936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039432665342339614,
      "loss": 5.347,
      "mean_token_accuracy": 0.20591456294059754,
      "num_tokens": 64373800.0,
      "step": 32895
    },
    {
      "entropy": 6.0454426288604735,
      "epoch": 3.288698955365622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003942963952509543,
      "loss": 5.5607,
      "mean_token_accuracy": 0.18718247711658478,
      "num_tokens": 64383329.0,
      "step": 32900
    },
    {
      "entropy": 5.971540927886963,
      "epoch": 3.2891987804268505,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039426613407704574,
      "loss": 5.4453,
      "mean_token_accuracy": 0.18846573680639267,
      "num_tokens": 64391717.0,
      "step": 32905
    },
    {
      "entropy": 5.9371950149536135,
      "epoch": 3.2896986054880792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000394235869902432,
      "loss": 5.4517,
      "mean_token_accuracy": 0.19717291593551636,
      "num_tokens": 64401496.0,
      "step": 32910
    },
    {
      "entropy": 5.937746477127075,
      "epoch": 3.2901984305493075,
      "grad_norm": 1.265625,
      "learning_rate": 0.00039420560272787446,
      "loss": 5.4497,
      "mean_token_accuracy": 0.1943392649292946,
      "num_tokens": 64411280.0,
      "step": 32915
    },
    {
      "entropy": 5.986769104003907,
      "epoch": 3.290698255610536,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039417533255413465,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18741099238395692,
      "num_tokens": 64421560.0,
      "step": 32920
    },
    {
      "entropy": 5.953263187408448,
      "epoch": 3.291198080671765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039414505938197404,
      "loss": 5.4664,
      "mean_token_accuracy": 0.19562844783067704,
      "num_tokens": 64432580.0,
      "step": 32925
    },
    {
      "entropy": 6.01424617767334,
      "epoch": 3.2916979057329936,
      "grad_norm": 1.125,
      "learning_rate": 0.0003941147832121545,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19133878871798515,
      "num_tokens": 64442937.0,
      "step": 32930
    },
    {
      "entropy": 5.9379504203796385,
      "epoch": 3.292197730794222,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003940845040454376,
      "loss": 5.5167,
      "mean_token_accuracy": 0.18733414262533188,
      "num_tokens": 64452763.0,
      "step": 32935
    },
    {
      "entropy": 5.943646860122681,
      "epoch": 3.2926975558554505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003940542218825853,
      "loss": 5.4321,
      "mean_token_accuracy": 0.19110239893198014,
      "num_tokens": 64462036.0,
      "step": 32940
    },
    {
      "entropy": 6.051838684082031,
      "epoch": 3.2931973809166792,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039402393672435934,
      "loss": 5.5216,
      "mean_token_accuracy": 0.1873842343688011,
      "num_tokens": 64472258.0,
      "step": 32945
    },
    {
      "entropy": 5.951851463317871,
      "epoch": 3.2936972059779075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039399364857152177,
      "loss": 5.4802,
      "mean_token_accuracy": 0.1930294007062912,
      "num_tokens": 64481521.0,
      "step": 32950
    },
    {
      "entropy": 6.001679372787476,
      "epoch": 3.294197031039136,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003939633574248346,
      "loss": 5.5757,
      "mean_token_accuracy": 0.1832214504480362,
      "num_tokens": 64491561.0,
      "step": 32955
    },
    {
      "entropy": 6.019756555557251,
      "epoch": 3.294696856100365,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039393306328506004,
      "loss": 5.4737,
      "mean_token_accuracy": 0.18720970302820206,
      "num_tokens": 64501408.0,
      "step": 32960
    },
    {
      "entropy": 6.001588439941406,
      "epoch": 3.2951966811615936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039390276615296015,
      "loss": 5.4585,
      "mean_token_accuracy": 0.19076913893222808,
      "num_tokens": 64510788.0,
      "step": 32965
    },
    {
      "entropy": 5.936774349212646,
      "epoch": 3.295696506222822,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039387246602929726,
      "loss": 5.5592,
      "mean_token_accuracy": 0.18268159180879592,
      "num_tokens": 64521644.0,
      "step": 32970
    },
    {
      "entropy": 6.014894914627075,
      "epoch": 3.2961963312840505,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003938421629148338,
      "loss": 5.5157,
      "mean_token_accuracy": 0.1974494621157646,
      "num_tokens": 64531209.0,
      "step": 32975
    },
    {
      "entropy": 6.09172477722168,
      "epoch": 3.2966961563452792,
      "grad_norm": 1.09375,
      "learning_rate": 0.000393811856810332,
      "loss": 5.6324,
      "mean_token_accuracy": 0.18369712233543395,
      "num_tokens": 64541280.0,
      "step": 32980
    },
    {
      "entropy": 5.977735614776611,
      "epoch": 3.2971959814065075,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003937815477165545,
      "loss": 5.5025,
      "mean_token_accuracy": 0.19096154272556304,
      "num_tokens": 64552440.0,
      "step": 32985
    },
    {
      "entropy": 5.90072112083435,
      "epoch": 3.297695806467736,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003937512356342638,
      "loss": 5.4482,
      "mean_token_accuracy": 0.19255473464727402,
      "num_tokens": 64561994.0,
      "step": 32990
    },
    {
      "entropy": 5.930997705459594,
      "epoch": 3.298195631528965,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003937209205642226,
      "loss": 5.4094,
      "mean_token_accuracy": 0.20146719366312027,
      "num_tokens": 64571022.0,
      "step": 32995
    },
    {
      "entropy": 6.022281360626221,
      "epoch": 3.2986954565901936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003936906025071936,
      "loss": 5.5916,
      "mean_token_accuracy": 0.17718040496110915,
      "num_tokens": 64581376.0,
      "step": 33000
    },
    {
      "epoch": 3.2986954565901936,
      "eval_entropy": 5.7874773452344,
      "eval_loss": 5.634433746337891,
      "eval_mean_token_accuracy": 0.19289452125695342,
      "eval_num_tokens": 64581376.0,
      "eval_runtime": 24.1276,
      "eval_samples_per_second": 1286.698,
      "eval_steps_per_second": 160.853,
      "step": 33000
    },
    {
      "entropy": 6.032714986801148,
      "epoch": 3.299195281651422,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039366028146393955,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1934676617383957,
      "num_tokens": 64590656.0,
      "step": 33005
    },
    {
      "entropy": 5.958738183975219,
      "epoch": 3.2996951067126505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003936299574352233,
      "loss": 5.4495,
      "mean_token_accuracy": 0.19716468453407288,
      "num_tokens": 64601276.0,
      "step": 33010
    },
    {
      "entropy": 5.976379251480102,
      "epoch": 3.3001949317738792,
      "grad_norm": 1.203125,
      "learning_rate": 0.000393599630421808,
      "loss": 5.4945,
      "mean_token_accuracy": 0.19179236590862275,
      "num_tokens": 64611251.0,
      "step": 33015
    },
    {
      "entropy": 6.038413572311401,
      "epoch": 3.3006947568351075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003935693004244564,
      "loss": 5.5264,
      "mean_token_accuracy": 0.18909091353416443,
      "num_tokens": 64619721.0,
      "step": 33020
    },
    {
      "entropy": 5.946103811264038,
      "epoch": 3.301194581896336,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039353896744393163,
      "loss": 5.459,
      "mean_token_accuracy": 0.19057969152927398,
      "num_tokens": 64629456.0,
      "step": 33025
    },
    {
      "entropy": 5.968410968780518,
      "epoch": 3.301694406957565,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003935086314809971,
      "loss": 5.44,
      "mean_token_accuracy": 0.19157896488904952,
      "num_tokens": 64639513.0,
      "step": 33030
    },
    {
      "entropy": 5.93816909790039,
      "epoch": 3.3021942320187936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039347829253641583,
      "loss": 5.392,
      "mean_token_accuracy": 0.19367333501577377,
      "num_tokens": 64648653.0,
      "step": 33035
    },
    {
      "entropy": 6.011275100708008,
      "epoch": 3.302694057080022,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039344795061095123,
      "loss": 5.5616,
      "mean_token_accuracy": 0.19085598438978196,
      "num_tokens": 64659566.0,
      "step": 33040
    },
    {
      "entropy": 5.904532814025879,
      "epoch": 3.3031938821412505,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003934176057053667,
      "loss": 5.4011,
      "mean_token_accuracy": 0.20200401544570923,
      "num_tokens": 64669041.0,
      "step": 33045
    },
    {
      "entropy": 6.0525695323944095,
      "epoch": 3.3036937072024792,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039338725782042563,
      "loss": 5.5754,
      "mean_token_accuracy": 0.18764784783124924,
      "num_tokens": 64678262.0,
      "step": 33050
    },
    {
      "entropy": 5.9226563453674315,
      "epoch": 3.3041935322637075,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039335690695689157,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20279361307621002,
      "num_tokens": 64687530.0,
      "step": 33055
    },
    {
      "entropy": 6.061674308776856,
      "epoch": 3.304693357324936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003933265531155283,
      "loss": 5.6158,
      "mean_token_accuracy": 0.1845640316605568,
      "num_tokens": 64697034.0,
      "step": 33060
    },
    {
      "entropy": 6.032867479324341,
      "epoch": 3.305193182386165,
      "grad_norm": 1.125,
      "learning_rate": 0.00039329619629709935,
      "loss": 5.5517,
      "mean_token_accuracy": 0.18578650653362275,
      "num_tokens": 64706047.0,
      "step": 33065
    },
    {
      "entropy": 6.030163431167603,
      "epoch": 3.3056930074473936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039326583650236847,
      "loss": 5.5635,
      "mean_token_accuracy": 0.19308728128671646,
      "num_tokens": 64715225.0,
      "step": 33070
    },
    {
      "entropy": 5.936201667785644,
      "epoch": 3.306192832508622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039323547373209974,
      "loss": 5.453,
      "mean_token_accuracy": 0.19565550237894058,
      "num_tokens": 64724332.0,
      "step": 33075
    },
    {
      "entropy": 5.973576927185059,
      "epoch": 3.3066926575698505,
      "grad_norm": 1.328125,
      "learning_rate": 0.00039320510798705684,
      "loss": 5.4679,
      "mean_token_accuracy": 0.19380795657634736,
      "num_tokens": 64734022.0,
      "step": 33080
    },
    {
      "entropy": 5.957065153121948,
      "epoch": 3.3071924826310792,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003931747392680038,
      "loss": 5.5033,
      "mean_token_accuracy": 0.19108137339353562,
      "num_tokens": 64742677.0,
      "step": 33085
    },
    {
      "entropy": 6.00252776145935,
      "epoch": 3.3076923076923075,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003931443675757047,
      "loss": 5.5296,
      "mean_token_accuracy": 0.1892552837729454,
      "num_tokens": 64752052.0,
      "step": 33090
    },
    {
      "entropy": 6.062704801559448,
      "epoch": 3.308192132753536,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039311399291092384,
      "loss": 5.5283,
      "mean_token_accuracy": 0.19007256031036376,
      "num_tokens": 64761465.0,
      "step": 33095
    },
    {
      "entropy": 5.992096757888794,
      "epoch": 3.308691957814765,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003930836152744251,
      "loss": 5.4765,
      "mean_token_accuracy": 0.19185051023960115,
      "num_tokens": 64772233.0,
      "step": 33100
    },
    {
      "entropy": 5.978529167175293,
      "epoch": 3.3091917828759936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039305323466697317,
      "loss": 5.4812,
      "mean_token_accuracy": 0.19166397750377656,
      "num_tokens": 64782006.0,
      "step": 33105
    },
    {
      "entropy": 5.985164928436279,
      "epoch": 3.309691607937222,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00039302285108933224,
      "loss": 5.4585,
      "mean_token_accuracy": 0.1908344492316246,
      "num_tokens": 64790755.0,
      "step": 33110
    },
    {
      "entropy": 5.917209243774414,
      "epoch": 3.3101914329984505,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003929924645422667,
      "loss": 5.4056,
      "mean_token_accuracy": 0.20044118911027908,
      "num_tokens": 64800970.0,
      "step": 33115
    },
    {
      "entropy": 5.99032039642334,
      "epoch": 3.3106912580596792,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039296207502654103,
      "loss": 5.5565,
      "mean_token_accuracy": 0.18477713763713838,
      "num_tokens": 64810679.0,
      "step": 33120
    },
    {
      "entropy": 5.966642284393311,
      "epoch": 3.3111910831209075,
      "grad_norm": 1.25,
      "learning_rate": 0.00039293168254292,
      "loss": 5.4407,
      "mean_token_accuracy": 0.19984062016010284,
      "num_tokens": 64820469.0,
      "step": 33125
    },
    {
      "entropy": 5.967524671554566,
      "epoch": 3.311690908182136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003929012870921681,
      "loss": 5.4179,
      "mean_token_accuracy": 0.19915213882923127,
      "num_tokens": 64830839.0,
      "step": 33130
    },
    {
      "entropy": 6.02615442276001,
      "epoch": 3.312190733243365,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003928708886750503,
      "loss": 5.5617,
      "mean_token_accuracy": 0.19169730842113494,
      "num_tokens": 64840985.0,
      "step": 33135
    },
    {
      "entropy": 6.029364347457886,
      "epoch": 3.3126905583045936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039284048729233115,
      "loss": 5.5294,
      "mean_token_accuracy": 0.19217559844255447,
      "num_tokens": 64850874.0,
      "step": 33140
    },
    {
      "entropy": 5.984996128082275,
      "epoch": 3.313190383365822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003928100829447757,
      "loss": 5.5128,
      "mean_token_accuracy": 0.1921686664223671,
      "num_tokens": 64860374.0,
      "step": 33145
    },
    {
      "entropy": 6.006088447570801,
      "epoch": 3.3136902084270505,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003927796756331489,
      "loss": 5.4952,
      "mean_token_accuracy": 0.18071205615997316,
      "num_tokens": 64869441.0,
      "step": 33150
    },
    {
      "entropy": 5.98968505859375,
      "epoch": 3.3141900334882792,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003927492653582157,
      "loss": 5.5569,
      "mean_token_accuracy": 0.18142445534467697,
      "num_tokens": 64879981.0,
      "step": 33155
    },
    {
      "entropy": 5.973738527297973,
      "epoch": 3.3146898585495075,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039271885212074134,
      "loss": 5.4671,
      "mean_token_accuracy": 0.19331486225128175,
      "num_tokens": 64889124.0,
      "step": 33160
    },
    {
      "entropy": 6.045901203155518,
      "epoch": 3.315189683610736,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039268843592149094,
      "loss": 5.5224,
      "mean_token_accuracy": 0.18811685144901275,
      "num_tokens": 64897931.0,
      "step": 33165
    },
    {
      "entropy": 6.066919565200806,
      "epoch": 3.315689508671965,
      "grad_norm": 1.125,
      "learning_rate": 0.00039265801676122977,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18910743445158004,
      "num_tokens": 64908253.0,
      "step": 33170
    },
    {
      "entropy": 5.989940738677978,
      "epoch": 3.3161893337331936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039262759464072316,
      "loss": 5.4611,
      "mean_token_accuracy": 0.19777586162090302,
      "num_tokens": 64917950.0,
      "step": 33175
    },
    {
      "entropy": 5.939294481277466,
      "epoch": 3.316689158794422,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003925971695607366,
      "loss": 5.4165,
      "mean_token_accuracy": 0.1995576500892639,
      "num_tokens": 64927614.0,
      "step": 33180
    },
    {
      "entropy": 6.014116048812866,
      "epoch": 3.3171889838556505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039256674152203546,
      "loss": 5.5265,
      "mean_token_accuracy": 0.1838432878255844,
      "num_tokens": 64938011.0,
      "step": 33185
    },
    {
      "entropy": 5.990730905532837,
      "epoch": 3.3176888089168792,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003925363105253854,
      "loss": 5.4481,
      "mean_token_accuracy": 0.19192507714033127,
      "num_tokens": 64947048.0,
      "step": 33190
    },
    {
      "entropy": 6.013100862503052,
      "epoch": 3.3181886339781075,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039250587657155197,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19891363978385926,
      "num_tokens": 64956997.0,
      "step": 33195
    },
    {
      "entropy": 5.963138246536255,
      "epoch": 3.318688459039336,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003924754396613009,
      "loss": 5.5131,
      "mean_token_accuracy": 0.1892402321100235,
      "num_tokens": 64967921.0,
      "step": 33200
    },
    {
      "entropy": 5.9198401927947994,
      "epoch": 3.319188284100565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003924449997953981,
      "loss": 5.4218,
      "mean_token_accuracy": 0.19323173761367798,
      "num_tokens": 64977641.0,
      "step": 33205
    },
    {
      "entropy": 5.918396425247193,
      "epoch": 3.3196881091617936,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039241455697460923,
      "loss": 5.4436,
      "mean_token_accuracy": 0.1936696484684944,
      "num_tokens": 64987576.0,
      "step": 33210
    },
    {
      "entropy": 5.9778436660766605,
      "epoch": 3.320187934223022,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039238411119970034,
      "loss": 5.4732,
      "mean_token_accuracy": 0.1873411774635315,
      "num_tokens": 64997072.0,
      "step": 33215
    },
    {
      "entropy": 5.984380006790161,
      "epoch": 3.3206877592842505,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003923536624714374,
      "loss": 5.4817,
      "mean_token_accuracy": 0.18994379192590713,
      "num_tokens": 65007072.0,
      "step": 33220
    },
    {
      "entropy": 5.994820594787598,
      "epoch": 3.3211875843454792,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039232321079058657,
      "loss": 5.5052,
      "mean_token_accuracy": 0.1944347620010376,
      "num_tokens": 65017067.0,
      "step": 33225
    },
    {
      "entropy": 6.058481502532959,
      "epoch": 3.3216874094067075,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039229275615791393,
      "loss": 5.5688,
      "mean_token_accuracy": 0.18868968784809112,
      "num_tokens": 65026302.0,
      "step": 33230
    },
    {
      "entropy": 5.9708123207092285,
      "epoch": 3.322187234467936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003922622985741857,
      "loss": 5.5149,
      "mean_token_accuracy": 0.19819526821374894,
      "num_tokens": 65036485.0,
      "step": 33235
    },
    {
      "entropy": 5.966877508163452,
      "epoch": 3.322687059529165,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039223183804016823,
      "loss": 5.6114,
      "mean_token_accuracy": 0.17976024746894836,
      "num_tokens": 65045704.0,
      "step": 33240
    },
    {
      "entropy": 6.022281551361084,
      "epoch": 3.3231868845903936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039220137455662785,
      "loss": 5.4793,
      "mean_token_accuracy": 0.19459908157587053,
      "num_tokens": 65054997.0,
      "step": 33245
    },
    {
      "entropy": 6.137752723693848,
      "epoch": 3.323686709651622,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039217090812433106,
      "loss": 5.6491,
      "mean_token_accuracy": 0.18000176697969436,
      "num_tokens": 65065381.0,
      "step": 33250
    },
    {
      "entropy": 5.940825414657593,
      "epoch": 3.3241865347128505,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003921404387440444,
      "loss": 5.4322,
      "mean_token_accuracy": 0.19716406613588333,
      "num_tokens": 65075348.0,
      "step": 33255
    },
    {
      "entropy": 5.968258666992187,
      "epoch": 3.3246863597740792,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003921099664165344,
      "loss": 5.5275,
      "mean_token_accuracy": 0.19086549282073975,
      "num_tokens": 65084566.0,
      "step": 33260
    },
    {
      "entropy": 5.973997497558594,
      "epoch": 3.3251861848353075,
      "grad_norm": 1.125,
      "learning_rate": 0.0003920794911425678,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1920412391424179,
      "num_tokens": 65094585.0,
      "step": 33265
    },
    {
      "entropy": 6.030121898651123,
      "epoch": 3.325686009896536,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039204901292291133,
      "loss": 5.5031,
      "mean_token_accuracy": 0.19376032948493957,
      "num_tokens": 65105288.0,
      "step": 33270
    },
    {
      "entropy": 6.03459677696228,
      "epoch": 3.326185834957765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003920185317583318,
      "loss": 5.4626,
      "mean_token_accuracy": 0.1936059847474098,
      "num_tokens": 65115815.0,
      "step": 33275
    },
    {
      "entropy": 6.091108942031861,
      "epoch": 3.3266856600189936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003919880476495962,
      "loss": 5.5783,
      "mean_token_accuracy": 0.18259162902832032,
      "num_tokens": 65126654.0,
      "step": 33280
    },
    {
      "entropy": 5.981639385223389,
      "epoch": 3.327185485080222,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003919575605974713,
      "loss": 5.5474,
      "mean_token_accuracy": 0.18423958867788315,
      "num_tokens": 65137061.0,
      "step": 33285
    },
    {
      "entropy": 6.079230880737304,
      "epoch": 3.3276853101414505,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003919270706027244,
      "loss": 5.5825,
      "mean_token_accuracy": 0.18776353746652602,
      "num_tokens": 65146991.0,
      "step": 33290
    },
    {
      "entropy": 6.060645675659179,
      "epoch": 3.3281851352026792,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003918965776661224,
      "loss": 5.5141,
      "mean_token_accuracy": 0.18757054954767227,
      "num_tokens": 65156146.0,
      "step": 33295
    },
    {
      "entropy": 5.993308019638062,
      "epoch": 3.3286849602639075,
      "grad_norm": 1.125,
      "learning_rate": 0.00039186608178843257,
      "loss": 5.5258,
      "mean_token_accuracy": 0.1930167406797409,
      "num_tokens": 65166005.0,
      "step": 33300
    },
    {
      "entropy": 5.987994384765625,
      "epoch": 3.329184785325136,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039183558297042225,
      "loss": 5.4919,
      "mean_token_accuracy": 0.19532340317964553,
      "num_tokens": 65177311.0,
      "step": 33305
    },
    {
      "entropy": 5.993380784988403,
      "epoch": 3.329684610386365,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00039180508121285866,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1911967933177948,
      "num_tokens": 65186371.0,
      "step": 33310
    },
    {
      "entropy": 6.037397861480713,
      "epoch": 3.3301844354475936,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003917745765165094,
      "loss": 5.4784,
      "mean_token_accuracy": 0.19345286637544631,
      "num_tokens": 65194861.0,
      "step": 33315
    },
    {
      "entropy": 5.9898021697998045,
      "epoch": 3.330684260508822,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003917440688821417,
      "loss": 5.5676,
      "mean_token_accuracy": 0.18297934234142305,
      "num_tokens": 65204479.0,
      "step": 33320
    },
    {
      "entropy": 5.975824451446533,
      "epoch": 3.3311840855700505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003917135583105235,
      "loss": 5.4685,
      "mean_token_accuracy": 0.18821781128644943,
      "num_tokens": 65213756.0,
      "step": 33325
    },
    {
      "entropy": 6.017086982727051,
      "epoch": 3.331683910631279,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039168304480242196,
      "loss": 5.501,
      "mean_token_accuracy": 0.19076623916625976,
      "num_tokens": 65223061.0,
      "step": 33330
    },
    {
      "entropy": 5.9193634510040285,
      "epoch": 3.3321837356925075,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003916525283586051,
      "loss": 5.3886,
      "mean_token_accuracy": 0.19884000569581986,
      "num_tokens": 65231743.0,
      "step": 33335
    },
    {
      "entropy": 6.033977794647217,
      "epoch": 3.332683560753736,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003916220089798407,
      "loss": 5.5464,
      "mean_token_accuracy": 0.19406975358724593,
      "num_tokens": 65240629.0,
      "step": 33340
    },
    {
      "entropy": 5.995237827301025,
      "epoch": 3.333183385814965,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039159148666689656,
      "loss": 5.5319,
      "mean_token_accuracy": 0.19184719622135163,
      "num_tokens": 65249766.0,
      "step": 33345
    },
    {
      "entropy": 5.962952423095703,
      "epoch": 3.3336832108761936,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003915609614205406,
      "loss": 5.4499,
      "mean_token_accuracy": 0.20099124163389206,
      "num_tokens": 65258621.0,
      "step": 33350
    },
    {
      "entropy": 6.0230494976043705,
      "epoch": 3.334183035937422,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003915304332415408,
      "loss": 5.5015,
      "mean_token_accuracy": 0.1934247389435768,
      "num_tokens": 65268036.0,
      "step": 33355
    },
    {
      "entropy": 5.997662973403931,
      "epoch": 3.3346828609986505,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003914999021306653,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1889515459537506,
      "num_tokens": 65278377.0,
      "step": 33360
    },
    {
      "entropy": 5.973831844329834,
      "epoch": 3.335182686059879,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003914693680886823,
      "loss": 5.5104,
      "mean_token_accuracy": 0.19140854626893997,
      "num_tokens": 65288778.0,
      "step": 33365
    },
    {
      "entropy": 5.949272727966308,
      "epoch": 3.3356825111211075,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039143883111635995,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19884912967681884,
      "num_tokens": 65297993.0,
      "step": 33370
    },
    {
      "entropy": 6.052727603912354,
      "epoch": 3.336182336182336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003914082912144666,
      "loss": 5.553,
      "mean_token_accuracy": 0.18137103766202928,
      "num_tokens": 65307551.0,
      "step": 33375
    },
    {
      "entropy": 5.946649169921875,
      "epoch": 3.336682161243565,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039137774838377047,
      "loss": 5.4464,
      "mean_token_accuracy": 0.19450520128011703,
      "num_tokens": 65317747.0,
      "step": 33380
    },
    {
      "entropy": 6.027565574645996,
      "epoch": 3.3371819863047936,
      "grad_norm": 1.125,
      "learning_rate": 0.0003913472026250402,
      "loss": 5.5951,
      "mean_token_accuracy": 0.18224784582853318,
      "num_tokens": 65328383.0,
      "step": 33385
    },
    {
      "entropy": 6.066370677947998,
      "epoch": 3.337681811366022,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003913166539390443,
      "loss": 5.6055,
      "mean_token_accuracy": 0.18567105233669282,
      "num_tokens": 65338134.0,
      "step": 33390
    },
    {
      "entropy": 6.0257415771484375,
      "epoch": 3.3381816364272505,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0003912861023265513,
      "loss": 5.4922,
      "mean_token_accuracy": 0.19233597964048385,
      "num_tokens": 65348520.0,
      "step": 33395
    },
    {
      "entropy": 6.000319576263427,
      "epoch": 3.338681461488479,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003912555477883299,
      "loss": 5.4588,
      "mean_token_accuracy": 0.1931587874889374,
      "num_tokens": 65357383.0,
      "step": 33400
    },
    {
      "entropy": 5.930278539657593,
      "epoch": 3.3391812865497075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003912249903251488,
      "loss": 5.4804,
      "mean_token_accuracy": 0.19394194483757018,
      "num_tokens": 65367463.0,
      "step": 33405
    },
    {
      "entropy": 6.034960985183716,
      "epoch": 3.339681111610936,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003911944299377768,
      "loss": 5.5216,
      "mean_token_accuracy": 0.18816245794296266,
      "num_tokens": 65377276.0,
      "step": 33410
    },
    {
      "entropy": 5.94196310043335,
      "epoch": 3.340180936672165,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003911638666269829,
      "loss": 5.4725,
      "mean_token_accuracy": 0.19690613746643065,
      "num_tokens": 65386716.0,
      "step": 33415
    },
    {
      "entropy": 5.909934043884277,
      "epoch": 3.340680761733393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039113330039353605,
      "loss": 5.4223,
      "mean_token_accuracy": 0.20203209072351455,
      "num_tokens": 65396900.0,
      "step": 33420
    },
    {
      "entropy": 5.979069662094116,
      "epoch": 3.341180586794622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039110273123820524,
      "loss": 5.4208,
      "mean_token_accuracy": 0.19794777035713196,
      "num_tokens": 65406036.0,
      "step": 33425
    },
    {
      "entropy": 5.9939000606536865,
      "epoch": 3.3416804118558505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039107215916175947,
      "loss": 5.4621,
      "mean_token_accuracy": 0.1915482237935066,
      "num_tokens": 65415374.0,
      "step": 33430
    },
    {
      "entropy": 5.991790771484375,
      "epoch": 3.3421802369170788,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003910415841649681,
      "loss": 5.4164,
      "mean_token_accuracy": 0.1953519120812416,
      "num_tokens": 65424969.0,
      "step": 33435
    },
    {
      "entropy": 5.928611850738525,
      "epoch": 3.3426800619783075,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003910110062486004,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19752547442913054,
      "num_tokens": 65434068.0,
      "step": 33440
    },
    {
      "entropy": 6.033936834335327,
      "epoch": 3.343179887039536,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00039098042541342554,
      "loss": 5.6134,
      "mean_token_accuracy": 0.18716612160205842,
      "num_tokens": 65445169.0,
      "step": 33445
    },
    {
      "entropy": 5.974177360534668,
      "epoch": 3.343679712100765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039094984166021307,
      "loss": 5.4218,
      "mean_token_accuracy": 0.19845286011695862,
      "num_tokens": 65454409.0,
      "step": 33450
    },
    {
      "entropy": 6.007299900054932,
      "epoch": 3.344179537161993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003909192549897325,
      "loss": 5.5228,
      "mean_token_accuracy": 0.19281553030014037,
      "num_tokens": 65464101.0,
      "step": 33455
    },
    {
      "entropy": 6.08740382194519,
      "epoch": 3.344679362223222,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003908886654027532,
      "loss": 5.6634,
      "mean_token_accuracy": 0.1764666721224785,
      "num_tokens": 65473774.0,
      "step": 33460
    },
    {
      "entropy": 5.961142253875733,
      "epoch": 3.3451791872844505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039085807290004495,
      "loss": 5.4322,
      "mean_token_accuracy": 0.19835235178470612,
      "num_tokens": 65483414.0,
      "step": 33465
    },
    {
      "entropy": 5.9253003120422365,
      "epoch": 3.3456790123456788,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039082747748237737,
      "loss": 5.4574,
      "mean_token_accuracy": 0.19466268718242646,
      "num_tokens": 65493875.0,
      "step": 33470
    },
    {
      "entropy": 6.0120398044586185,
      "epoch": 3.3461788374069075,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003907968791505202,
      "loss": 5.5085,
      "mean_token_accuracy": 0.1871066614985466,
      "num_tokens": 65503801.0,
      "step": 33475
    },
    {
      "entropy": 6.003343200683593,
      "epoch": 3.346678662468136,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039076627790524343,
      "loss": 5.4643,
      "mean_token_accuracy": 0.19809413254261016,
      "num_tokens": 65514151.0,
      "step": 33480
    },
    {
      "entropy": 5.982717418670655,
      "epoch": 3.347178487529365,
      "grad_norm": 0.96875,
      "learning_rate": 0.00039073567374731685,
      "loss": 5.4722,
      "mean_token_accuracy": 0.19402533769607544,
      "num_tokens": 65525246.0,
      "step": 33485
    },
    {
      "entropy": 5.903830480575562,
      "epoch": 3.347678312590593,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039070506667751066,
      "loss": 5.4017,
      "mean_token_accuracy": 0.19860177338123322,
      "num_tokens": 65534766.0,
      "step": 33490
    },
    {
      "entropy": 6.038247299194336,
      "epoch": 3.348178137651822,
      "grad_norm": 1.125,
      "learning_rate": 0.0003906744566965946,
      "loss": 5.5394,
      "mean_token_accuracy": 0.18905537128448485,
      "num_tokens": 65545058.0,
      "step": 33495
    },
    {
      "entropy": 6.020566320419311,
      "epoch": 3.3486779627130505,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000390643843805339,
      "loss": 5.4609,
      "mean_token_accuracy": 0.19898003488779067,
      "num_tokens": 65554234.0,
      "step": 33500
    },
    {
      "entropy": 6.042502927780151,
      "epoch": 3.3491777877742788,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000390613228004514,
      "loss": 5.6251,
      "mean_token_accuracy": 0.18782709538936615,
      "num_tokens": 65565115.0,
      "step": 33505
    },
    {
      "entropy": 5.948621463775635,
      "epoch": 3.3496776128355075,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00039058260929489,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19784953743219375,
      "num_tokens": 65575291.0,
      "step": 33510
    },
    {
      "entropy": 6.01177830696106,
      "epoch": 3.350177437896736,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039055198767723725,
      "loss": 5.5007,
      "mean_token_accuracy": 0.19470321536064147,
      "num_tokens": 65585105.0,
      "step": 33515
    },
    {
      "entropy": 6.029422855377197,
      "epoch": 3.350677262957965,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039052136315232625,
      "loss": 5.4987,
      "mean_token_accuracy": 0.19504923075437547,
      "num_tokens": 65594786.0,
      "step": 33520
    },
    {
      "entropy": 5.911636018753052,
      "epoch": 3.351177088019193,
      "grad_norm": 1.125,
      "learning_rate": 0.00039049073572092755,
      "loss": 5.3726,
      "mean_token_accuracy": 0.19897006154060365,
      "num_tokens": 65604553.0,
      "step": 33525
    },
    {
      "entropy": 5.952718496322632,
      "epoch": 3.351676913080422,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003904601053838116,
      "loss": 5.479,
      "mean_token_accuracy": 0.19484733194112777,
      "num_tokens": 65613808.0,
      "step": 33530
    },
    {
      "entropy": 6.0127355575561525,
      "epoch": 3.3521767381416505,
      "grad_norm": 1.21875,
      "learning_rate": 0.00039042947214174906,
      "loss": 5.5046,
      "mean_token_accuracy": 0.19457001239061356,
      "num_tokens": 65622741.0,
      "step": 33535
    },
    {
      "entropy": 5.972920751571655,
      "epoch": 3.3526765632028788,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039039883599551074,
      "loss": 5.4352,
      "mean_token_accuracy": 0.19903043508529664,
      "num_tokens": 65632325.0,
      "step": 33540
    },
    {
      "entropy": 6.026919984817505,
      "epoch": 3.3531763882641075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003903681969458674,
      "loss": 5.4563,
      "mean_token_accuracy": 0.1947249189019203,
      "num_tokens": 65642636.0,
      "step": 33545
    },
    {
      "entropy": 5.978336238861084,
      "epoch": 3.353676213325336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003903375549935899,
      "loss": 5.5038,
      "mean_token_accuracy": 0.19114208221435547,
      "num_tokens": 65652058.0,
      "step": 33550
    },
    {
      "entropy": 5.983210563659668,
      "epoch": 3.354176038386565,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003903069101394492,
      "loss": 5.4553,
      "mean_token_accuracy": 0.20022477954626083,
      "num_tokens": 65662213.0,
      "step": 33555
    },
    {
      "entropy": 6.024156808853149,
      "epoch": 3.354675863447793,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003902762623842164,
      "loss": 5.6083,
      "mean_token_accuracy": 0.18101637363433837,
      "num_tokens": 65672302.0,
      "step": 33560
    },
    {
      "entropy": 5.875590991973877,
      "epoch": 3.355175688509022,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039024561172866236,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20502525717020034,
      "num_tokens": 65681642.0,
      "step": 33565
    },
    {
      "entropy": 5.979424619674683,
      "epoch": 3.3556755135702505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003902149581735585,
      "loss": 5.3889,
      "mean_token_accuracy": 0.1982231169939041,
      "num_tokens": 65691165.0,
      "step": 33570
    },
    {
      "entropy": 5.936292123794556,
      "epoch": 3.3561753386314788,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039018430171967584,
      "loss": 5.4844,
      "mean_token_accuracy": 0.19442086815834045,
      "num_tokens": 65701556.0,
      "step": 33575
    },
    {
      "entropy": 5.924810934066772,
      "epoch": 3.3566751636927075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003901536423677859,
      "loss": 5.4395,
      "mean_token_accuracy": 0.19905002415180206,
      "num_tokens": 65711481.0,
      "step": 33580
    },
    {
      "entropy": 5.986252498626709,
      "epoch": 3.357174988753936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003901229801186598,
      "loss": 5.4364,
      "mean_token_accuracy": 0.19397483319044112,
      "num_tokens": 65720731.0,
      "step": 33585
    },
    {
      "entropy": 5.9231650829315186,
      "epoch": 3.357674813815165,
      "grad_norm": 1.375,
      "learning_rate": 0.0003900923149730693,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19792777448892593,
      "num_tokens": 65730219.0,
      "step": 33590
    },
    {
      "entropy": 6.003418064117431,
      "epoch": 3.358174638876393,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039006164693178573,
      "loss": 5.4973,
      "mean_token_accuracy": 0.19850286096334457,
      "num_tokens": 65739893.0,
      "step": 33595
    },
    {
      "entropy": 6.051329660415649,
      "epoch": 3.358674463937622,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003900309759955806,
      "loss": 5.554,
      "mean_token_accuracy": 0.1874864086508751,
      "num_tokens": 65748925.0,
      "step": 33600
    },
    {
      "entropy": 6.031333637237549,
      "epoch": 3.3591742889988505,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003900003021652259,
      "loss": 5.5443,
      "mean_token_accuracy": 0.1854175418615341,
      "num_tokens": 65759355.0,
      "step": 33605
    },
    {
      "entropy": 6.005221939086914,
      "epoch": 3.3596741140600788,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038996962544149313,
      "loss": 5.4745,
      "mean_token_accuracy": 0.19559985995292664,
      "num_tokens": 65767853.0,
      "step": 33610
    },
    {
      "entropy": 5.949774122238159,
      "epoch": 3.3601739391213075,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003899389458251541,
      "loss": 5.4796,
      "mean_token_accuracy": 0.19787080883979796,
      "num_tokens": 65776672.0,
      "step": 33615
    },
    {
      "entropy": 5.961060333251953,
      "epoch": 3.360673764182536,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003899082633169808,
      "loss": 5.4824,
      "mean_token_accuracy": 0.18729114383459092,
      "num_tokens": 65785839.0,
      "step": 33620
    },
    {
      "entropy": 6.033926630020142,
      "epoch": 3.361173589243765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038987757791774513,
      "loss": 5.631,
      "mean_token_accuracy": 0.18250963985919952,
      "num_tokens": 65796648.0,
      "step": 33625
    },
    {
      "entropy": 5.950387382507325,
      "epoch": 3.361673414304993,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003898468896282193,
      "loss": 5.4177,
      "mean_token_accuracy": 0.1965209200978279,
      "num_tokens": 65806170.0,
      "step": 33630
    },
    {
      "entropy": 6.086455011367798,
      "epoch": 3.362173239366222,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003898161984491752,
      "loss": 5.5814,
      "mean_token_accuracy": 0.18400689214468002,
      "num_tokens": 65816319.0,
      "step": 33635
    },
    {
      "entropy": 5.8919188499450685,
      "epoch": 3.3626730644274505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038978550438138504,
      "loss": 5.458,
      "mean_token_accuracy": 0.19800342321395875,
      "num_tokens": 65827334.0,
      "step": 33640
    },
    {
      "entropy": 5.88982081413269,
      "epoch": 3.3631728894886788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003897548074256211,
      "loss": 5.4378,
      "mean_token_accuracy": 0.20146023482084274,
      "num_tokens": 65835948.0,
      "step": 33645
    },
    {
      "entropy": 5.96736707687378,
      "epoch": 3.3636727145499075,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003897241075826559,
      "loss": 5.4965,
      "mean_token_accuracy": 0.1997979536652565,
      "num_tokens": 65845026.0,
      "step": 33650
    },
    {
      "entropy": 5.927409696578979,
      "epoch": 3.364172539611136,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00038969340485326147,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1876918062567711,
      "num_tokens": 65854820.0,
      "step": 33655
    },
    {
      "entropy": 5.964037895202637,
      "epoch": 3.364672364672365,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003896626992382106,
      "loss": 5.4969,
      "mean_token_accuracy": 0.19035959988832474,
      "num_tokens": 65864557.0,
      "step": 33660
    },
    {
      "entropy": 6.022389650344849,
      "epoch": 3.365172189733593,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038963199073827574,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1940610095858574,
      "num_tokens": 65874534.0,
      "step": 33665
    },
    {
      "entropy": 5.917154932022095,
      "epoch": 3.365672014794822,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038960127935422947,
      "loss": 5.3574,
      "mean_token_accuracy": 0.20704445689916612,
      "num_tokens": 65883722.0,
      "step": 33670
    },
    {
      "entropy": 5.940880298614502,
      "epoch": 3.3661718398560505,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003895705650868445,
      "loss": 5.4583,
      "mean_token_accuracy": 0.19039404690265654,
      "num_tokens": 65894061.0,
      "step": 33675
    },
    {
      "entropy": 5.906609678268433,
      "epoch": 3.3666716649172788,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038953984793689355,
      "loss": 5.4794,
      "mean_token_accuracy": 0.19161394834518433,
      "num_tokens": 65904495.0,
      "step": 33680
    },
    {
      "entropy": 6.039932632446289,
      "epoch": 3.3671714899785075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038950912790514954,
      "loss": 5.5608,
      "mean_token_accuracy": 0.1846650004386902,
      "num_tokens": 65915958.0,
      "step": 33685
    },
    {
      "entropy": 5.9597978591918945,
      "epoch": 3.367671315039736,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003894784049923854,
      "loss": 5.4388,
      "mean_token_accuracy": 0.20117318630218506,
      "num_tokens": 65925689.0,
      "step": 33690
    },
    {
      "entropy": 5.935555362701416,
      "epoch": 3.368171140100965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003894476791993739,
      "loss": 5.4369,
      "mean_token_accuracy": 0.1964769408106804,
      "num_tokens": 65935024.0,
      "step": 33695
    },
    {
      "entropy": 5.936588621139526,
      "epoch": 3.368670965162193,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038941695052688833,
      "loss": 5.48,
      "mean_token_accuracy": 0.19370829164981843,
      "num_tokens": 65944646.0,
      "step": 33700
    },
    {
      "entropy": 5.986376476287842,
      "epoch": 3.369170790223422,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038938621897570175,
      "loss": 5.5674,
      "mean_token_accuracy": 0.18685307651758193,
      "num_tokens": 65954670.0,
      "step": 33705
    },
    {
      "entropy": 6.0265007495880125,
      "epoch": 3.3696706152846505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003893554845465873,
      "loss": 5.6146,
      "mean_token_accuracy": 0.1784324049949646,
      "num_tokens": 65964844.0,
      "step": 33710
    },
    {
      "entropy": 6.109501552581787,
      "epoch": 3.3701704403458788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003893247472403183,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18341607302427293,
      "num_tokens": 65974625.0,
      "step": 33715
    },
    {
      "entropy": 5.962427711486816,
      "epoch": 3.3706702654071075,
      "grad_norm": 1.171875,
      "learning_rate": 0.000389294007057668,
      "loss": 5.5185,
      "mean_token_accuracy": 0.1878181904554367,
      "num_tokens": 65983266.0,
      "step": 33720
    },
    {
      "entropy": 6.050905323028564,
      "epoch": 3.371170090468336,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038926326399941,
      "loss": 5.7051,
      "mean_token_accuracy": 0.17475820779800416,
      "num_tokens": 65992611.0,
      "step": 33725
    },
    {
      "entropy": 6.002885961532593,
      "epoch": 3.371669915529565,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003892325180663176,
      "loss": 5.4914,
      "mean_token_accuracy": 0.18720174282789231,
      "num_tokens": 66003387.0,
      "step": 33730
    },
    {
      "entropy": 6.046358346939087,
      "epoch": 3.372169740590793,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003892017692591645,
      "loss": 5.6215,
      "mean_token_accuracy": 0.18392744660377502,
      "num_tokens": 66013574.0,
      "step": 33735
    },
    {
      "entropy": 6.047493743896484,
      "epoch": 3.372669565652022,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038917101757872434,
      "loss": 5.4943,
      "mean_token_accuracy": 0.18926388919353485,
      "num_tokens": 66024312.0,
      "step": 33740
    },
    {
      "entropy": 6.022190284729004,
      "epoch": 3.3731693907132505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003891402630257706,
      "loss": 5.5032,
      "mean_token_accuracy": 0.1874435603618622,
      "num_tokens": 66034444.0,
      "step": 33745
    },
    {
      "entropy": 5.893392133712768,
      "epoch": 3.3736692157744788,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038910950560107734,
      "loss": 5.4753,
      "mean_token_accuracy": 0.19553083926439285,
      "num_tokens": 66043524.0,
      "step": 33750
    },
    {
      "entropy": 5.916419887542725,
      "epoch": 3.3741690408357075,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038907874530541824,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19097287207841873,
      "num_tokens": 66053223.0,
      "step": 33755
    },
    {
      "entropy": 5.91071629524231,
      "epoch": 3.374668865896936,
      "grad_norm": 1.125,
      "learning_rate": 0.00038904798213956716,
      "loss": 5.4324,
      "mean_token_accuracy": 0.20166666060686111,
      "num_tokens": 66062042.0,
      "step": 33760
    },
    {
      "entropy": 5.892795610427856,
      "epoch": 3.375168690958165,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003890172161042983,
      "loss": 5.5238,
      "mean_token_accuracy": 0.1937461942434311,
      "num_tokens": 66071685.0,
      "step": 33765
    },
    {
      "entropy": 5.949362897872925,
      "epoch": 3.375668516019393,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038898644720038563,
      "loss": 5.5196,
      "mean_token_accuracy": 0.19214480221271515,
      "num_tokens": 66080677.0,
      "step": 33770
    },
    {
      "entropy": 5.974997282028198,
      "epoch": 3.376168341080622,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003889556754286032,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1960396483540535,
      "num_tokens": 66089310.0,
      "step": 33775
    },
    {
      "entropy": 5.915404987335205,
      "epoch": 3.3766681661418505,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038892490078972544,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19823605716228485,
      "num_tokens": 66099818.0,
      "step": 33780
    },
    {
      "entropy": 5.919692039489746,
      "epoch": 3.3771679912030788,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003888941232845264,
      "loss": 5.4085,
      "mean_token_accuracy": 0.19918284714221954,
      "num_tokens": 66109622.0,
      "step": 33785
    },
    {
      "entropy": 5.905157089233398,
      "epoch": 3.3776678162643075,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038886334291378047,
      "loss": 5.4274,
      "mean_token_accuracy": 0.19796027839183808,
      "num_tokens": 66118677.0,
      "step": 33790
    },
    {
      "entropy": 5.898518466949463,
      "epoch": 3.378167641325536,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003888325596782622,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1964363306760788,
      "num_tokens": 66127818.0,
      "step": 33795
    },
    {
      "entropy": 6.0082920551300045,
      "epoch": 3.378667466386765,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000388801773578746,
      "loss": 5.5185,
      "mean_token_accuracy": 0.19207246154546737,
      "num_tokens": 66138972.0,
      "step": 33800
    },
    {
      "entropy": 5.962406063079834,
      "epoch": 3.379167291447993,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003887709846160065,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19481825977563857,
      "num_tokens": 66147900.0,
      "step": 33805
    },
    {
      "entropy": 5.7441949367523195,
      "epoch": 3.379667116509222,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003887401927908183,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21823657751083375,
      "num_tokens": 66158302.0,
      "step": 33810
    },
    {
      "entropy": 6.026461219787597,
      "epoch": 3.3801669415704505,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038870939810395605,
      "loss": 5.5472,
      "mean_token_accuracy": 0.1855640634894371,
      "num_tokens": 66168069.0,
      "step": 33815
    },
    {
      "entropy": 6.039138412475586,
      "epoch": 3.3806667666316788,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038867860055619456,
      "loss": 5.5586,
      "mean_token_accuracy": 0.1863809794187546,
      "num_tokens": 66176702.0,
      "step": 33820
    },
    {
      "entropy": 5.942537260055542,
      "epoch": 3.3811665916929075,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003886478001483089,
      "loss": 5.443,
      "mean_token_accuracy": 0.19888330698013307,
      "num_tokens": 66185293.0,
      "step": 33825
    },
    {
      "entropy": 5.87330846786499,
      "epoch": 3.381666416754136,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003886169968810737,
      "loss": 5.2918,
      "mean_token_accuracy": 0.21505151987075805,
      "num_tokens": 66195289.0,
      "step": 33830
    },
    {
      "entropy": 5.861311531066894,
      "epoch": 3.382166241815365,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003885861907552642,
      "loss": 5.4037,
      "mean_token_accuracy": 0.20100258141756058,
      "num_tokens": 66205778.0,
      "step": 33835
    },
    {
      "entropy": 6.0037397861480715,
      "epoch": 3.382666066876593,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003885553817716553,
      "loss": 5.4661,
      "mean_token_accuracy": 0.18888431191444396,
      "num_tokens": 66214861.0,
      "step": 33840
    },
    {
      "entropy": 6.040483665466309,
      "epoch": 3.383165891937822,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003885245699310223,
      "loss": 5.5351,
      "mean_token_accuracy": 0.1926989287137985,
      "num_tokens": 66224721.0,
      "step": 33845
    },
    {
      "entropy": 5.949206924438476,
      "epoch": 3.3836657169990505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003884937552341402,
      "loss": 5.4932,
      "mean_token_accuracy": 0.19085606038570405,
      "num_tokens": 66234132.0,
      "step": 33850
    },
    {
      "entropy": 6.023981809616089,
      "epoch": 3.3841655420602788,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038846293768178444,
      "loss": 5.619,
      "mean_token_accuracy": 0.18539710193872452,
      "num_tokens": 66244098.0,
      "step": 33855
    },
    {
      "entropy": 5.9447380065917965,
      "epoch": 3.3846653671215075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003884321172747304,
      "loss": 5.419,
      "mean_token_accuracy": 0.19935036897659303,
      "num_tokens": 66253813.0,
      "step": 33860
    },
    {
      "entropy": 5.979374027252197,
      "epoch": 3.385165192182736,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003884012940137536,
      "loss": 5.4459,
      "mean_token_accuracy": 0.1957242637872696,
      "num_tokens": 66263699.0,
      "step": 33865
    },
    {
      "entropy": 5.947338247299195,
      "epoch": 3.385665017243965,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003883704678996293,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19824420809745788,
      "num_tokens": 66273493.0,
      "step": 33870
    },
    {
      "entropy": 5.91345911026001,
      "epoch": 3.386164842305193,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038833963893313326,
      "loss": 5.4399,
      "mean_token_accuracy": 0.19228278696537018,
      "num_tokens": 66282372.0,
      "step": 33875
    },
    {
      "entropy": 5.973013305664063,
      "epoch": 3.386664667366422,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003883088071150409,
      "loss": 5.4707,
      "mean_token_accuracy": 0.19430087357759476,
      "num_tokens": 66291799.0,
      "step": 33880
    },
    {
      "entropy": 5.983700466156006,
      "epoch": 3.3871644924276505,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003882779724461282,
      "loss": 5.4746,
      "mean_token_accuracy": 0.18652842044830323,
      "num_tokens": 66301973.0,
      "step": 33885
    },
    {
      "entropy": 5.914132308959961,
      "epoch": 3.3876643174888788,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038824713492717093,
      "loss": 5.4225,
      "mean_token_accuracy": 0.19723087251186372,
      "num_tokens": 66311590.0,
      "step": 33890
    },
    {
      "entropy": 6.006572580337524,
      "epoch": 3.3881641425501074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003882162945589449,
      "loss": 5.4526,
      "mean_token_accuracy": 0.19671905487775804,
      "num_tokens": 66321331.0,
      "step": 33895
    },
    {
      "entropy": 6.007521343231201,
      "epoch": 3.388663967611336,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003881854513422259,
      "loss": 5.4772,
      "mean_token_accuracy": 0.19240449219942093,
      "num_tokens": 66331012.0,
      "step": 33900
    },
    {
      "entropy": 5.9049266338348385,
      "epoch": 3.389163792672565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038815460527779017,
      "loss": 5.4527,
      "mean_token_accuracy": 0.198721045255661,
      "num_tokens": 66340160.0,
      "step": 33905
    },
    {
      "entropy": 5.881530046463013,
      "epoch": 3.389663617733793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038812375636641366,
      "loss": 5.4077,
      "mean_token_accuracy": 0.19819531887769698,
      "num_tokens": 66350943.0,
      "step": 33910
    },
    {
      "entropy": 5.909305381774902,
      "epoch": 3.390163442795022,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038809290460887255,
      "loss": 5.3219,
      "mean_token_accuracy": 0.20594122558832167,
      "num_tokens": 66360538.0,
      "step": 33915
    },
    {
      "entropy": 5.9916328430175785,
      "epoch": 3.3906632678562505,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038806205000594307,
      "loss": 5.4672,
      "mean_token_accuracy": 0.19330343455076218,
      "num_tokens": 66369357.0,
      "step": 33920
    },
    {
      "entropy": 5.990479326248169,
      "epoch": 3.3911630929174787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003880311925584015,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1878092333674431,
      "num_tokens": 66378863.0,
      "step": 33925
    },
    {
      "entropy": 5.98684868812561,
      "epoch": 3.3916629179787074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038800033226702417,
      "loss": 5.4626,
      "mean_token_accuracy": 0.2013371467590332,
      "num_tokens": 66388461.0,
      "step": 33930
    },
    {
      "entropy": 5.915777778625488,
      "epoch": 3.392162743039936,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003879694691325877,
      "loss": 5.3389,
      "mean_token_accuracy": 0.2087661698460579,
      "num_tokens": 66398083.0,
      "step": 33935
    },
    {
      "entropy": 5.993072843551635,
      "epoch": 3.3926625681011644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038793860315586824,
      "loss": 5.5651,
      "mean_token_accuracy": 0.18985484689474105,
      "num_tokens": 66407411.0,
      "step": 33940
    },
    {
      "entropy": 5.985146141052246,
      "epoch": 3.393162393162393,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003879077343376428,
      "loss": 5.4075,
      "mean_token_accuracy": 0.2004818558692932,
      "num_tokens": 66416996.0,
      "step": 33945
    },
    {
      "entropy": 5.9954126358032225,
      "epoch": 3.393662218223622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038787686267868765,
      "loss": 5.4592,
      "mean_token_accuracy": 0.19226199090480806,
      "num_tokens": 66426855.0,
      "step": 33950
    },
    {
      "entropy": 5.9211297035217285,
      "epoch": 3.3941620432848505,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003878459881797798,
      "loss": 5.4869,
      "mean_token_accuracy": 0.1883169174194336,
      "num_tokens": 66437003.0,
      "step": 33955
    },
    {
      "entropy": 5.867582654953003,
      "epoch": 3.3946618683460787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038781511084169586,
      "loss": 5.3643,
      "mean_token_accuracy": 0.20473100394010543,
      "num_tokens": 66447013.0,
      "step": 33960
    },
    {
      "entropy": 5.932321834564209,
      "epoch": 3.3951616934073074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038778423066521275,
      "loss": 5.4777,
      "mean_token_accuracy": 0.19107092767953873,
      "num_tokens": 66456240.0,
      "step": 33965
    },
    {
      "entropy": 5.979287433624267,
      "epoch": 3.395661518468536,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038775334765110746,
      "loss": 5.4212,
      "mean_token_accuracy": 0.20534548610448838,
      "num_tokens": 66466284.0,
      "step": 33970
    },
    {
      "entropy": 6.064418077468872,
      "epoch": 3.3961613435297644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038772246180015694,
      "loss": 5.5657,
      "mean_token_accuracy": 0.18897740095853804,
      "num_tokens": 66476983.0,
      "step": 33975
    },
    {
      "entropy": 5.918005323410034,
      "epoch": 3.396661168590993,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003876915731131383,
      "loss": 5.4738,
      "mean_token_accuracy": 0.19131845533847808,
      "num_tokens": 66488143.0,
      "step": 33980
    },
    {
      "entropy": 5.932526397705078,
      "epoch": 3.397160993652222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003876606815908286,
      "loss": 5.4082,
      "mean_token_accuracy": 0.1991519033908844,
      "num_tokens": 66498354.0,
      "step": 33985
    },
    {
      "entropy": 5.884347152709961,
      "epoch": 3.39766081871345,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003876297872340052,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20780438035726548,
      "num_tokens": 66508751.0,
      "step": 33990
    },
    {
      "entropy": 5.95495572090149,
      "epoch": 3.3981606437746787,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038759889004344537,
      "loss": 5.4967,
      "mean_token_accuracy": 0.19533025473356247,
      "num_tokens": 66518414.0,
      "step": 33995
    },
    {
      "entropy": 5.958155393600464,
      "epoch": 3.3986604688359074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038756799001992646,
      "loss": 5.4634,
      "mean_token_accuracy": 0.19492805004119873,
      "num_tokens": 66527876.0,
      "step": 34000
    },
    {
      "entropy": 6.017147445678711,
      "epoch": 3.399160293897136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003875370871642258,
      "loss": 5.4789,
      "mean_token_accuracy": 0.1949700027704239,
      "num_tokens": 66536964.0,
      "step": 34005
    },
    {
      "entropy": 6.0361121654510494,
      "epoch": 3.3996601189583644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038750618147712106,
      "loss": 5.5554,
      "mean_token_accuracy": 0.18587732315063477,
      "num_tokens": 66546271.0,
      "step": 34010
    },
    {
      "entropy": 5.914024734497071,
      "epoch": 3.400159944019593,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00038747527295938966,
      "loss": 5.4544,
      "mean_token_accuracy": 0.20046927332878112,
      "num_tokens": 66556056.0,
      "step": 34015
    },
    {
      "entropy": 5.901616907119751,
      "epoch": 3.400659769080822,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003874443616118095,
      "loss": 5.4941,
      "mean_token_accuracy": 0.19477889835834503,
      "num_tokens": 66565588.0,
      "step": 34020
    },
    {
      "entropy": 6.023278141021729,
      "epoch": 3.40115959414205,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00038741344743515796,
      "loss": 5.4803,
      "mean_token_accuracy": 0.18674306124448775,
      "num_tokens": 66575814.0,
      "step": 34025
    },
    {
      "entropy": 6.041391754150391,
      "epoch": 3.4016594192032787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000387382530430213,
      "loss": 5.4582,
      "mean_token_accuracy": 0.19169710278511048,
      "num_tokens": 66586290.0,
      "step": 34030
    },
    {
      "entropy": 6.009279823303222,
      "epoch": 3.4021592442645074,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003873516105977526,
      "loss": 5.4842,
      "mean_token_accuracy": 0.1941351994872093,
      "num_tokens": 66595203.0,
      "step": 34035
    },
    {
      "entropy": 5.952558279037476,
      "epoch": 3.402659069325736,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038732068793855456,
      "loss": 5.457,
      "mean_token_accuracy": 0.19559229165315628,
      "num_tokens": 66604945.0,
      "step": 34040
    },
    {
      "entropy": 5.949074745178223,
      "epoch": 3.4031588943869644,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003872897624533969,
      "loss": 5.5382,
      "mean_token_accuracy": 0.18837041407823563,
      "num_tokens": 66614730.0,
      "step": 34045
    },
    {
      "entropy": 6.033965969085694,
      "epoch": 3.403658719448193,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003872588341430577,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1919575721025467,
      "num_tokens": 66625567.0,
      "step": 34050
    },
    {
      "entropy": 5.916818904876709,
      "epoch": 3.404158544509422,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003872279030083151,
      "loss": 5.4257,
      "mean_token_accuracy": 0.20212469398975372,
      "num_tokens": 66635517.0,
      "step": 34055
    },
    {
      "entropy": 6.005887413024903,
      "epoch": 3.40465836957065,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038719696904994746,
      "loss": 5.5384,
      "mean_token_accuracy": 0.18940234631299974,
      "num_tokens": 66645497.0,
      "step": 34060
    },
    {
      "entropy": 6.014895582199097,
      "epoch": 3.4051581946318787,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038716603226873286,
      "loss": 5.5322,
      "mean_token_accuracy": 0.19040439873933793,
      "num_tokens": 66655283.0,
      "step": 34065
    },
    {
      "entropy": 5.978982019424438,
      "epoch": 3.4056580196931074,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038713509266544973,
      "loss": 5.4838,
      "mean_token_accuracy": 0.19324944466352462,
      "num_tokens": 66664925.0,
      "step": 34070
    },
    {
      "entropy": 5.953125095367431,
      "epoch": 3.406157844754336,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003871041502408765,
      "loss": 5.5128,
      "mean_token_accuracy": 0.1938397094607353,
      "num_tokens": 66674273.0,
      "step": 34075
    },
    {
      "entropy": 5.980450773239136,
      "epoch": 3.4066576698155644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038707320499579176,
      "loss": 5.4044,
      "mean_token_accuracy": 0.1967630922794342,
      "num_tokens": 66683225.0,
      "step": 34080
    },
    {
      "entropy": 5.9391590595245365,
      "epoch": 3.407157494876793,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038704225693097404,
      "loss": 5.411,
      "mean_token_accuracy": 0.19832914918661118,
      "num_tokens": 66693614.0,
      "step": 34085
    },
    {
      "entropy": 6.012066888809204,
      "epoch": 3.407657319938022,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038701130604720194,
      "loss": 5.4937,
      "mean_token_accuracy": 0.1956542670726776,
      "num_tokens": 66703877.0,
      "step": 34090
    },
    {
      "entropy": 5.9999524593353275,
      "epoch": 3.40815714499925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003869803523452542,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19932015240192413,
      "num_tokens": 66713800.0,
      "step": 34095
    },
    {
      "entropy": 6.012040090560913,
      "epoch": 3.4086569700604787,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003869493958259096,
      "loss": 5.533,
      "mean_token_accuracy": 0.1854144513607025,
      "num_tokens": 66724387.0,
      "step": 34100
    },
    {
      "entropy": 6.034504842758179,
      "epoch": 3.4091567951217074,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000386918436489947,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1982484370470047,
      "num_tokens": 66733435.0,
      "step": 34105
    },
    {
      "entropy": 6.011856651306152,
      "epoch": 3.409656620182936,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003868874743381453,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1844882771372795,
      "num_tokens": 66742799.0,
      "step": 34110
    },
    {
      "entropy": 6.0086853981018065,
      "epoch": 3.4101564452441644,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003868565093712836,
      "loss": 5.5304,
      "mean_token_accuracy": 0.1864339143037796,
      "num_tokens": 66753417.0,
      "step": 34115
    },
    {
      "entropy": 5.9778295993804935,
      "epoch": 3.410656270305393,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038682554159014094,
      "loss": 5.4633,
      "mean_token_accuracy": 0.1960441753268242,
      "num_tokens": 66763141.0,
      "step": 34120
    },
    {
      "entropy": 6.0126396179199215,
      "epoch": 3.411156095366622,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038679457099549637,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18755683302879333,
      "num_tokens": 66772929.0,
      "step": 34125
    },
    {
      "entropy": 5.9568483352661135,
      "epoch": 3.41165592042785,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003867635975881292,
      "loss": 5.4972,
      "mean_token_accuracy": 0.1907669112086296,
      "num_tokens": 66783343.0,
      "step": 34130
    },
    {
      "entropy": 5.979887342453003,
      "epoch": 3.4121557454890787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003867326213688186,
      "loss": 5.4773,
      "mean_token_accuracy": 0.19341920763254167,
      "num_tokens": 66792069.0,
      "step": 34135
    },
    {
      "entropy": 6.038926124572754,
      "epoch": 3.4126555705503074,
      "grad_norm": 1.046875,
      "learning_rate": 0.00038670164233834406,
      "loss": 5.5549,
      "mean_token_accuracy": 0.18496417254209518,
      "num_tokens": 66803736.0,
      "step": 34140
    },
    {
      "entropy": 5.954047203063965,
      "epoch": 3.413155395611536,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00038667066049748487,
      "loss": 5.5085,
      "mean_token_accuracy": 0.19078054577112197,
      "num_tokens": 66813007.0,
      "step": 34145
    },
    {
      "entropy": 6.060598373413086,
      "epoch": 3.4136552206727644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038663967584702064,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18379650563001632,
      "num_tokens": 66823305.0,
      "step": 34150
    },
    {
      "entropy": 5.97602481842041,
      "epoch": 3.414155045733993,
      "grad_norm": 1.21875,
      "learning_rate": 0.000386608688387731,
      "loss": 5.4696,
      "mean_token_accuracy": 0.19679942727088928,
      "num_tokens": 66833051.0,
      "step": 34155
    },
    {
      "entropy": 5.958174991607666,
      "epoch": 3.414654870795222,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038657769812039535,
      "loss": 5.468,
      "mean_token_accuracy": 0.19386890083551406,
      "num_tokens": 66842992.0,
      "step": 34160
    },
    {
      "entropy": 5.9039184093475345,
      "epoch": 3.41515469585645,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003865467050457935,
      "loss": 5.3768,
      "mean_token_accuracy": 0.20541460067033768,
      "num_tokens": 66852290.0,
      "step": 34165
    },
    {
      "entropy": 5.943043565750122,
      "epoch": 3.4156545209176787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038651570916470535,
      "loss": 5.4262,
      "mean_token_accuracy": 0.1921062245965004,
      "num_tokens": 66862597.0,
      "step": 34170
    },
    {
      "entropy": 5.946780443191528,
      "epoch": 3.4161543459789074,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038648471047791064,
      "loss": 5.4344,
      "mean_token_accuracy": 0.19622379243373872,
      "num_tokens": 66873815.0,
      "step": 34175
    },
    {
      "entropy": 6.051115703582764,
      "epoch": 3.416654171040136,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00038645370898618924,
      "loss": 5.5339,
      "mean_token_accuracy": 0.1915270358324051,
      "num_tokens": 66884855.0,
      "step": 34180
    },
    {
      "entropy": 5.896151828765869,
      "epoch": 3.4171539961013644,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038642270469032127,
      "loss": 5.4172,
      "mean_token_accuracy": 0.20635372251272202,
      "num_tokens": 66893850.0,
      "step": 34185
    },
    {
      "entropy": 5.973970985412597,
      "epoch": 3.417653821162593,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003863916975910867,
      "loss": 5.5217,
      "mean_token_accuracy": 0.1925130307674408,
      "num_tokens": 66903587.0,
      "step": 34190
    },
    {
      "entropy": 6.01916275024414,
      "epoch": 3.418153646223822,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003863606876892657,
      "loss": 5.4625,
      "mean_token_accuracy": 0.18553622364997863,
      "num_tokens": 66912520.0,
      "step": 34195
    },
    {
      "entropy": 5.926861238479614,
      "epoch": 3.41865347128505,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00038632967498563843,
      "loss": 5.4287,
      "mean_token_accuracy": 0.20064572393894195,
      "num_tokens": 66923115.0,
      "step": 34200
    },
    {
      "entropy": 6.019441699981689,
      "epoch": 3.4191532963462787,
      "grad_norm": 1.125,
      "learning_rate": 0.00038629865948098515,
      "loss": 5.5338,
      "mean_token_accuracy": 0.18596696108579636,
      "num_tokens": 66932688.0,
      "step": 34205
    },
    {
      "entropy": 5.979830408096314,
      "epoch": 3.4196531214075074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038626764117608627,
      "loss": 5.5073,
      "mean_token_accuracy": 0.19171400517225265,
      "num_tokens": 66944013.0,
      "step": 34210
    },
    {
      "entropy": 5.980470848083496,
      "epoch": 3.420152946468736,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003862366200717222,
      "loss": 5.6107,
      "mean_token_accuracy": 0.18569727540016173,
      "num_tokens": 66953924.0,
      "step": 34215
    },
    {
      "entropy": 6.03786187171936,
      "epoch": 3.4206527715299644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003862055961686733,
      "loss": 5.6179,
      "mean_token_accuracy": 0.18342809081077577,
      "num_tokens": 66963749.0,
      "step": 34220
    },
    {
      "entropy": 5.985335063934326,
      "epoch": 3.421152596591193,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00038617456946772035,
      "loss": 5.3983,
      "mean_token_accuracy": 0.20183158218860625,
      "num_tokens": 66973370.0,
      "step": 34225
    },
    {
      "entropy": 5.994614171981811,
      "epoch": 3.421652421652422,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003861435399696437,
      "loss": 5.5297,
      "mean_token_accuracy": 0.19699530005455018,
      "num_tokens": 66983556.0,
      "step": 34230
    },
    {
      "entropy": 5.897173929214477,
      "epoch": 3.42215224671365,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00038611250767522427,
      "loss": 5.3919,
      "mean_token_accuracy": 0.20580139458179475,
      "num_tokens": 66993002.0,
      "step": 34235
    },
    {
      "entropy": 5.934391498565674,
      "epoch": 3.4226520717748787,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038608147258524273,
      "loss": 5.4136,
      "mean_token_accuracy": 0.19362969398498536,
      "num_tokens": 67002111.0,
      "step": 34240
    },
    {
      "entropy": 6.006271123886108,
      "epoch": 3.4231518968361074,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003860504347004799,
      "loss": 5.4405,
      "mean_token_accuracy": 0.19227734059095383,
      "num_tokens": 67011236.0,
      "step": 34245
    },
    {
      "entropy": 6.080173969268799,
      "epoch": 3.423651721897336,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038601939402171676,
      "loss": 5.5458,
      "mean_token_accuracy": 0.1948930725455284,
      "num_tokens": 67021636.0,
      "step": 34250
    },
    {
      "entropy": 6.0028135776519775,
      "epoch": 3.4241515469585644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038598835054973413,
      "loss": 5.533,
      "mean_token_accuracy": 0.1901842325925827,
      "num_tokens": 67032099.0,
      "step": 34255
    },
    {
      "entropy": 5.96284852027893,
      "epoch": 3.424651372019793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038595730428531323,
      "loss": 5.4734,
      "mean_token_accuracy": 0.18797180205583572,
      "num_tokens": 67041714.0,
      "step": 34260
    },
    {
      "entropy": 5.9652925491333,
      "epoch": 3.425151197081022,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038592625522923513,
      "loss": 5.4565,
      "mean_token_accuracy": 0.19494511187076569,
      "num_tokens": 67051500.0,
      "step": 34265
    },
    {
      "entropy": 6.009584426879883,
      "epoch": 3.42565102214225,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038589520338228094,
      "loss": 5.5063,
      "mean_token_accuracy": 0.19517099112272263,
      "num_tokens": 67060706.0,
      "step": 34270
    },
    {
      "entropy": 5.894058704376221,
      "epoch": 3.4261508472034787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038586414874523204,
      "loss": 5.3936,
      "mean_token_accuracy": 0.21490250676870346,
      "num_tokens": 67070351.0,
      "step": 34275
    },
    {
      "entropy": 5.834274959564209,
      "epoch": 3.4266506722647074,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003858330913188696,
      "loss": 5.3728,
      "mean_token_accuracy": 0.2013929381966591,
      "num_tokens": 67081058.0,
      "step": 34280
    },
    {
      "entropy": 5.909427881240845,
      "epoch": 3.427150497325936,
      "grad_norm": 1.171875,
      "learning_rate": 0.00038580203110397517,
      "loss": 5.431,
      "mean_token_accuracy": 0.1983833834528923,
      "num_tokens": 67090789.0,
      "step": 34285
    },
    {
      "entropy": 5.927088832855224,
      "epoch": 3.4276503223871644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038577096810133013,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19741872549057007,
      "num_tokens": 67099277.0,
      "step": 34290
    },
    {
      "entropy": 6.050859117507935,
      "epoch": 3.428150147448393,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038573990231171603,
      "loss": 5.5661,
      "mean_token_accuracy": 0.18236276358366013,
      "num_tokens": 67109219.0,
      "step": 34295
    },
    {
      "entropy": 5.944821691513061,
      "epoch": 3.428649972509622,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00038570883373591447,
      "loss": 5.4203,
      "mean_token_accuracy": 0.19389599859714507,
      "num_tokens": 67119656.0,
      "step": 34300
    },
    {
      "entropy": 5.95143461227417,
      "epoch": 3.42914979757085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038567776237470716,
      "loss": 5.4311,
      "mean_token_accuracy": 0.19676543027162552,
      "num_tokens": 67128810.0,
      "step": 34305
    },
    {
      "entropy": 5.911523771286011,
      "epoch": 3.4296496226320787,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038564668822887576,
      "loss": 5.3886,
      "mean_token_accuracy": 0.2017642617225647,
      "num_tokens": 67137748.0,
      "step": 34310
    },
    {
      "entropy": 5.96105170249939,
      "epoch": 3.4301494476933074,
      "grad_norm": 1.25,
      "learning_rate": 0.00038561561129920224,
      "loss": 5.513,
      "mean_token_accuracy": 0.19419171065092086,
      "num_tokens": 67147787.0,
      "step": 34315
    },
    {
      "entropy": 6.035624647140503,
      "epoch": 3.430649272754536,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003855845315864684,
      "loss": 5.5893,
      "mean_token_accuracy": 0.1883111521601677,
      "num_tokens": 67158356.0,
      "step": 34320
    },
    {
      "entropy": 5.978832387924195,
      "epoch": 3.4311490978157644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003855534490914562,
      "loss": 5.4783,
      "mean_token_accuracy": 0.19325657635927201,
      "num_tokens": 67167583.0,
      "step": 34325
    },
    {
      "entropy": 5.937611484527588,
      "epoch": 3.431648922876993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003855223638149476,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19901565164327623,
      "num_tokens": 67179126.0,
      "step": 34330
    },
    {
      "entropy": 6.0075218200683596,
      "epoch": 3.4321487479382218,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003854912757577249,
      "loss": 5.4949,
      "mean_token_accuracy": 0.19754274189472198,
      "num_tokens": 67188947.0,
      "step": 34335
    },
    {
      "entropy": 5.974856948852539,
      "epoch": 3.43264857299945,
      "grad_norm": 1.109375,
      "learning_rate": 0.00038546018492057,
      "loss": 5.4353,
      "mean_token_accuracy": 0.2010850265622139,
      "num_tokens": 67198084.0,
      "step": 34340
    },
    {
      "entropy": 6.010065078735352,
      "epoch": 3.4331483980606787,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00038542909130426527,
      "loss": 5.5222,
      "mean_token_accuracy": 0.19671258628368377,
      "num_tokens": 67207207.0,
      "step": 34345
    },
    {
      "entropy": 6.018010139465332,
      "epoch": 3.4336482231219074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003853979949095931,
      "loss": 5.5133,
      "mean_token_accuracy": 0.1957763224840164,
      "num_tokens": 67217568.0,
      "step": 34350
    },
    {
      "entropy": 5.989614629745484,
      "epoch": 3.434148048183136,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038536689573733575,
      "loss": 5.4216,
      "mean_token_accuracy": 0.19711461365222932,
      "num_tokens": 67228086.0,
      "step": 34355
    },
    {
      "entropy": 6.046931362152099,
      "epoch": 3.4346478732443644,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038533579378827567,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18860358744859695,
      "num_tokens": 67238123.0,
      "step": 34360
    },
    {
      "entropy": 5.9722999095916744,
      "epoch": 3.435147698305593,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003853046890631955,
      "loss": 5.5034,
      "mean_token_accuracy": 0.19387096911668777,
      "num_tokens": 67247570.0,
      "step": 34365
    },
    {
      "entropy": 5.939824438095092,
      "epoch": 3.4356475233668218,
      "grad_norm": 1.125,
      "learning_rate": 0.00038527358156287766,
      "loss": 5.4693,
      "mean_token_accuracy": 0.20060993432998658,
      "num_tokens": 67257298.0,
      "step": 34370
    },
    {
      "entropy": 6.022543954849243,
      "epoch": 3.43614734842805,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003852424712881049,
      "loss": 5.5757,
      "mean_token_accuracy": 0.18720244914293288,
      "num_tokens": 67266219.0,
      "step": 34375
    },
    {
      "entropy": 5.9512717723846436,
      "epoch": 3.4366471734892787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038521135823965994,
      "loss": 5.4011,
      "mean_token_accuracy": 0.20465911328792571,
      "num_tokens": 67276436.0,
      "step": 34380
    },
    {
      "entropy": 5.956686115264892,
      "epoch": 3.4371469985505074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038518024241832553,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19267508387565613,
      "num_tokens": 67286661.0,
      "step": 34385
    },
    {
      "entropy": 5.996560287475586,
      "epoch": 3.437646823611736,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003851491238248846,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1903931975364685,
      "num_tokens": 67295586.0,
      "step": 34390
    },
    {
      "entropy": 5.972581243515014,
      "epoch": 3.4381466486729644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003851180024601201,
      "loss": 5.4652,
      "mean_token_accuracy": 0.19171782582998276,
      "num_tokens": 67304701.0,
      "step": 34395
    },
    {
      "entropy": 5.946770238876343,
      "epoch": 3.438646473734193,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000385086878324815,
      "loss": 5.501,
      "mean_token_accuracy": 0.19819259643554688,
      "num_tokens": 67313825.0,
      "step": 34400
    },
    {
      "entropy": 5.9731330394744875,
      "epoch": 3.4391462987954218,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003850557514197523,
      "loss": 5.441,
      "mean_token_accuracy": 0.19284245669841765,
      "num_tokens": 67323306.0,
      "step": 34405
    },
    {
      "entropy": 6.023801231384278,
      "epoch": 3.43964612385665,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003850246217457152,
      "loss": 5.5184,
      "mean_token_accuracy": 0.18061012923717498,
      "num_tokens": 67333579.0,
      "step": 34410
    },
    {
      "entropy": 6.004775762557983,
      "epoch": 3.4401459489178787,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000384993489303487,
      "loss": 5.4816,
      "mean_token_accuracy": 0.1925164207816124,
      "num_tokens": 67343242.0,
      "step": 34415
    },
    {
      "entropy": 6.049419689178467,
      "epoch": 3.4406457739791074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038496235409385084,
      "loss": 5.476,
      "mean_token_accuracy": 0.19076974093914031,
      "num_tokens": 67352466.0,
      "step": 34420
    },
    {
      "entropy": 5.936966419219971,
      "epoch": 3.441145599040336,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038493121611759016,
      "loss": 5.446,
      "mean_token_accuracy": 0.19668568223714827,
      "num_tokens": 67361298.0,
      "step": 34425
    },
    {
      "entropy": 5.965119791030884,
      "epoch": 3.4416454241015644,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038490007537548843,
      "loss": 5.5159,
      "mean_token_accuracy": 0.19011884182691574,
      "num_tokens": 67370007.0,
      "step": 34430
    },
    {
      "entropy": 6.001757192611694,
      "epoch": 3.442145249162793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003848689318683291,
      "loss": 5.4466,
      "mean_token_accuracy": 0.20215579867362976,
      "num_tokens": 67381047.0,
      "step": 34435
    },
    {
      "entropy": 5.9712762355804445,
      "epoch": 3.4426450742240218,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003848377855968957,
      "loss": 5.4559,
      "mean_token_accuracy": 0.1961929440498352,
      "num_tokens": 67390021.0,
      "step": 34440
    },
    {
      "entropy": 5.919237852096558,
      "epoch": 3.44314489928525,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003848066365619718,
      "loss": 5.4475,
      "mean_token_accuracy": 0.19392339140176773,
      "num_tokens": 67399723.0,
      "step": 34445
    },
    {
      "entropy": 5.89318585395813,
      "epoch": 3.4436447243464787,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003847754847643412,
      "loss": 5.3564,
      "mean_token_accuracy": 0.21195252388715743,
      "num_tokens": 67410296.0,
      "step": 34450
    },
    {
      "entropy": 5.928204727172852,
      "epoch": 3.4441445494077074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003847443302047877,
      "loss": 5.4805,
      "mean_token_accuracy": 0.19912767112255098,
      "num_tokens": 67420000.0,
      "step": 34455
    },
    {
      "entropy": 5.922889328002929,
      "epoch": 3.444644374468936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000384713172884095,
      "loss": 5.4044,
      "mean_token_accuracy": 0.2021465077996254,
      "num_tokens": 67428599.0,
      "step": 34460
    },
    {
      "entropy": 5.912845468521118,
      "epoch": 3.4451441995301644,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003846820128030472,
      "loss": 5.4468,
      "mean_token_accuracy": 0.19970516115427017,
      "num_tokens": 67438390.0,
      "step": 34465
    },
    {
      "entropy": 5.959617233276367,
      "epoch": 3.445644024591393,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038465084996242824,
      "loss": 5.4404,
      "mean_token_accuracy": 0.19194228947162628,
      "num_tokens": 67448189.0,
      "step": 34470
    },
    {
      "entropy": 5.926181221008301,
      "epoch": 3.4461438496526218,
      "grad_norm": 1.046875,
      "learning_rate": 0.000384619684363022,
      "loss": 5.4785,
      "mean_token_accuracy": 0.19536871314048768,
      "num_tokens": 67458650.0,
      "step": 34475
    },
    {
      "entropy": 5.9634088516235355,
      "epoch": 3.44664367471385,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845885160056127,
      "loss": 5.4443,
      "mean_token_accuracy": 0.1947993040084839,
      "num_tokens": 67468316.0,
      "step": 34480
    },
    {
      "entropy": 5.929837942123413,
      "epoch": 3.4471434997750787,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038455734489098453,
      "loss": 5.4286,
      "mean_token_accuracy": 0.1957886904478073,
      "num_tokens": 67477507.0,
      "step": 34485
    },
    {
      "entropy": 5.981268835067749,
      "epoch": 3.4476433248363074,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003845261710199218,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18655556291341782,
      "num_tokens": 67487310.0,
      "step": 34490
    },
    {
      "entropy": 5.9447594165802,
      "epoch": 3.4481431498975357,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038449499439320876,
      "loss": 5.4053,
      "mean_token_accuracy": 0.20148599296808242,
      "num_tokens": 67496930.0,
      "step": 34495
    },
    {
      "entropy": 6.0566963195800785,
      "epoch": 3.4486429749587644,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038446381501162983,
      "loss": 5.5709,
      "mean_token_accuracy": 0.18679034262895583,
      "num_tokens": 67506105.0,
      "step": 34500
    },
    {
      "entropy": 5.948022842407227,
      "epoch": 3.449142800019993,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038443263287596966,
      "loss": 5.3855,
      "mean_token_accuracy": 0.20427834391593933,
      "num_tokens": 67516326.0,
      "step": 34505
    },
    {
      "entropy": 5.940356016159058,
      "epoch": 3.4496426250812213,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003844014479870124,
      "loss": 5.4552,
      "mean_token_accuracy": 0.2011139214038849,
      "num_tokens": 67525709.0,
      "step": 34510
    },
    {
      "entropy": 5.896925592422486,
      "epoch": 3.45014245014245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00038437026034554285,
      "loss": 5.3453,
      "mean_token_accuracy": 0.20347571671009063,
      "num_tokens": 67534790.0,
      "step": 34515
    },
    {
      "entropy": 5.786301755905152,
      "epoch": 3.4506422752036787,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003843390699523458,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2119930773973465,
      "num_tokens": 67544446.0,
      "step": 34520
    },
    {
      "entropy": 5.966781806945801,
      "epoch": 3.4511421002649074,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003843078768082058,
      "loss": 5.4991,
      "mean_token_accuracy": 0.18515886664390563,
      "num_tokens": 67554029.0,
      "step": 34525
    },
    {
      "entropy": 5.9961995601654055,
      "epoch": 3.4516419253261357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038427668091390785,
      "loss": 5.4779,
      "mean_token_accuracy": 0.18810901939868926,
      "num_tokens": 67563224.0,
      "step": 34530
    },
    {
      "entropy": 5.986273860931396,
      "epoch": 3.4521417503873644,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038424548227023657,
      "loss": 5.4809,
      "mean_token_accuracy": 0.1898829460144043,
      "num_tokens": 67572228.0,
      "step": 34535
    },
    {
      "entropy": 6.0247128963470455,
      "epoch": 3.452641575448593,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038421428087797725,
      "loss": 5.5243,
      "mean_token_accuracy": 0.1924059972167015,
      "num_tokens": 67582222.0,
      "step": 34540
    },
    {
      "entropy": 5.959749698638916,
      "epoch": 3.4531414005098213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038418307673791465,
      "loss": 5.428,
      "mean_token_accuracy": 0.20064736008644105,
      "num_tokens": 67591692.0,
      "step": 34545
    },
    {
      "entropy": 6.016943073272705,
      "epoch": 3.45364122557105,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003841518698508339,
      "loss": 5.5334,
      "mean_token_accuracy": 0.19320822060108184,
      "num_tokens": 67600972.0,
      "step": 34550
    },
    {
      "entropy": 5.951087093353271,
      "epoch": 3.4541410506322787,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003841206602175202,
      "loss": 5.3856,
      "mean_token_accuracy": 0.20264792889356614,
      "num_tokens": 67610697.0,
      "step": 34555
    },
    {
      "entropy": 5.977321672439575,
      "epoch": 3.4546408756935074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003840894478387588,
      "loss": 5.5007,
      "mean_token_accuracy": 0.19351483434438704,
      "num_tokens": 67620268.0,
      "step": 34560
    },
    {
      "entropy": 5.944418478012085,
      "epoch": 3.4551407007547357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038405823271533496,
      "loss": 5.4286,
      "mean_token_accuracy": 0.20050645023584365,
      "num_tokens": 67629100.0,
      "step": 34565
    },
    {
      "entropy": 5.9626226902008055,
      "epoch": 3.4556405258159644,
      "grad_norm": 1.078125,
      "learning_rate": 0.00038402701484803404,
      "loss": 5.5352,
      "mean_token_accuracy": 0.1894117921590805,
      "num_tokens": 67639339.0,
      "step": 34570
    },
    {
      "entropy": 5.979023075103759,
      "epoch": 3.456140350877193,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038399579423764146,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1924199193716049,
      "num_tokens": 67649310.0,
      "step": 34575
    },
    {
      "entropy": 6.014988660812378,
      "epoch": 3.4566401759384213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003839645708849428,
      "loss": 5.4937,
      "mean_token_accuracy": 0.19958399534225463,
      "num_tokens": 67659272.0,
      "step": 34580
    },
    {
      "entropy": 5.944198131561279,
      "epoch": 3.45714000099965,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003839333447907235,
      "loss": 5.4671,
      "mean_token_accuracy": 0.19564908891916274,
      "num_tokens": 67668906.0,
      "step": 34585
    },
    {
      "entropy": 5.983950662612915,
      "epoch": 3.4576398260608787,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003839021159557693,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18932231813669204,
      "num_tokens": 67679030.0,
      "step": 34590
    },
    {
      "entropy": 5.930665349960327,
      "epoch": 3.4581396511221074,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003838708843808659,
      "loss": 5.4155,
      "mean_token_accuracy": 0.2001168340444565,
      "num_tokens": 67688347.0,
      "step": 34595
    },
    {
      "entropy": 6.016345071792602,
      "epoch": 3.4586394761833357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038383965006679913,
      "loss": 5.5719,
      "mean_token_accuracy": 0.18584331423044204,
      "num_tokens": 67697681.0,
      "step": 34600
    },
    {
      "entropy": 5.943961668014526,
      "epoch": 3.4591393012445644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038380841301435467,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19263962507247925,
      "num_tokens": 67707068.0,
      "step": 34605
    },
    {
      "entropy": 6.051000022888184,
      "epoch": 3.459639126305793,
      "grad_norm": 1.125,
      "learning_rate": 0.0003837771732243186,
      "loss": 5.5089,
      "mean_token_accuracy": 0.18949636667966843,
      "num_tokens": 67716201.0,
      "step": 34610
    },
    {
      "entropy": 5.995186901092529,
      "epoch": 3.4601389513670213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003837459306974768,
      "loss": 5.5632,
      "mean_token_accuracy": 0.18344490230083466,
      "num_tokens": 67726251.0,
      "step": 34615
    },
    {
      "entropy": 5.972705698013305,
      "epoch": 3.46063877642825,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003837146854346154,
      "loss": 5.4833,
      "mean_token_accuracy": 0.20180203914642333,
      "num_tokens": 67736158.0,
      "step": 34620
    },
    {
      "entropy": 5.944543838500977,
      "epoch": 3.4611386014894787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038368343743652044,
      "loss": 5.3865,
      "mean_token_accuracy": 0.20081903636455536,
      "num_tokens": 67744686.0,
      "step": 34625
    },
    {
      "entropy": 5.961348056793213,
      "epoch": 3.4616384265507074,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003836521867039782,
      "loss": 5.5035,
      "mean_token_accuracy": 0.18968516290187837,
      "num_tokens": 67754977.0,
      "step": 34630
    },
    {
      "entropy": 6.0660545349121096,
      "epoch": 3.4621382516119357,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038362093323777486,
      "loss": 5.6565,
      "mean_token_accuracy": 0.17819753885269166,
      "num_tokens": 67765801.0,
      "step": 34635
    },
    {
      "entropy": 6.031957530975342,
      "epoch": 3.4626380766731644,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003835896770386968,
      "loss": 5.5096,
      "mean_token_accuracy": 0.19889622330665588,
      "num_tokens": 67775733.0,
      "step": 34640
    },
    {
      "entropy": 5.976533079147339,
      "epoch": 3.463137901734393,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038355841810753047,
      "loss": 5.3838,
      "mean_token_accuracy": 0.2038208395242691,
      "num_tokens": 67785379.0,
      "step": 34645
    },
    {
      "entropy": 5.942957258224487,
      "epoch": 3.4636377267956213,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003835271564450622,
      "loss": 5.4161,
      "mean_token_accuracy": 0.19372160434722902,
      "num_tokens": 67794561.0,
      "step": 34650
    },
    {
      "entropy": 5.880633783340454,
      "epoch": 3.46413755185685,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003834958920520786,
      "loss": 5.4353,
      "mean_token_accuracy": 0.20270622074604033,
      "num_tokens": 67803398.0,
      "step": 34655
    },
    {
      "entropy": 5.952063369750976,
      "epoch": 3.4646373769180787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003834646249293664,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1973013237118721,
      "num_tokens": 67813073.0,
      "step": 34660
    },
    {
      "entropy": 5.897465991973877,
      "epoch": 3.4651372019793074,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038343335507771205,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19805702269077302,
      "num_tokens": 67823191.0,
      "step": 34665
    },
    {
      "entropy": 5.888796997070313,
      "epoch": 3.4656370270405357,
      "grad_norm": 1.15625,
      "learning_rate": 0.00038340208249790236,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19902877807617186,
      "num_tokens": 67833426.0,
      "step": 34670
    },
    {
      "entropy": 5.927476930618286,
      "epoch": 3.4661368521017644,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003833708071907242,
      "loss": 5.5054,
      "mean_token_accuracy": 0.18950989693403245,
      "num_tokens": 67842225.0,
      "step": 34675
    },
    {
      "entropy": 5.943680667877198,
      "epoch": 3.466636677162993,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003833395291569644,
      "loss": 5.4655,
      "mean_token_accuracy": 0.19825413525104524,
      "num_tokens": 67850772.0,
      "step": 34680
    },
    {
      "entropy": 6.01129150390625,
      "epoch": 3.4671365022242213,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038330824839740996,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1836872473359108,
      "num_tokens": 67860811.0,
      "step": 34685
    },
    {
      "entropy": 6.073420238494873,
      "epoch": 3.46763632728545,
      "grad_norm": 1.296875,
      "learning_rate": 0.00038327696491284786,
      "loss": 5.5576,
      "mean_token_accuracy": 0.19049286842346191,
      "num_tokens": 67871297.0,
      "step": 34690
    },
    {
      "entropy": 5.911839246749878,
      "epoch": 3.4681361523466787,
      "grad_norm": 1.25,
      "learning_rate": 0.00038324567870406517,
      "loss": 5.4093,
      "mean_token_accuracy": 0.19715202152729033,
      "num_tokens": 67881530.0,
      "step": 34695
    },
    {
      "entropy": 6.003573703765869,
      "epoch": 3.4686359774079074,
      "grad_norm": 1.09375,
      "learning_rate": 0.000383214389771849,
      "loss": 5.5702,
      "mean_token_accuracy": 0.18445385098457337,
      "num_tokens": 67891655.0,
      "step": 34700
    },
    {
      "entropy": 5.931922483444214,
      "epoch": 3.4691358024691357,
      "grad_norm": 1.125,
      "learning_rate": 0.0003831830981169867,
      "loss": 5.4085,
      "mean_token_accuracy": 0.20564027726650239,
      "num_tokens": 67900941.0,
      "step": 34705
    },
    {
      "entropy": 5.926672124862671,
      "epoch": 3.4696356275303644,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003831518037402655,
      "loss": 5.4646,
      "mean_token_accuracy": 0.20171046108007432,
      "num_tokens": 67910713.0,
      "step": 34710
    },
    {
      "entropy": 5.964695739746094,
      "epoch": 3.470135452591593,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003831205066424728,
      "loss": 5.481,
      "mean_token_accuracy": 0.1971085175871849,
      "num_tokens": 67920274.0,
      "step": 34715
    },
    {
      "entropy": 5.968218803405762,
      "epoch": 3.4706352776528213,
      "grad_norm": 1.125,
      "learning_rate": 0.00038308920682439586,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18699869513511658,
      "num_tokens": 67930357.0,
      "step": 34720
    },
    {
      "entropy": 5.955906105041504,
      "epoch": 3.47113510271405,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003830579042868223,
      "loss": 5.4639,
      "mean_token_accuracy": 0.20363645106554032,
      "num_tokens": 67939375.0,
      "step": 34725
    },
    {
      "entropy": 5.958024787902832,
      "epoch": 3.4716349277752787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003830265990305397,
      "loss": 5.4447,
      "mean_token_accuracy": 0.19753199368715285,
      "num_tokens": 67947901.0,
      "step": 34730
    },
    {
      "entropy": 5.96468300819397,
      "epoch": 3.4721347528365074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003829952910563356,
      "loss": 5.4853,
      "mean_token_accuracy": 0.1996941924095154,
      "num_tokens": 67958801.0,
      "step": 34735
    },
    {
      "entropy": 6.036898183822632,
      "epoch": 3.4726345778977357,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038296398036499784,
      "loss": 5.5272,
      "mean_token_accuracy": 0.18907613009214402,
      "num_tokens": 67967291.0,
      "step": 34740
    },
    {
      "entropy": 5.944076824188232,
      "epoch": 3.4731344029589644,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038293266695731414,
      "loss": 5.4066,
      "mean_token_accuracy": 0.2034500628709793,
      "num_tokens": 67978304.0,
      "step": 34745
    },
    {
      "entropy": 5.982691907882691,
      "epoch": 3.473634228020193,
      "grad_norm": 1.21875,
      "learning_rate": 0.00038290135083407235,
      "loss": 5.4774,
      "mean_token_accuracy": 0.19270886480808258,
      "num_tokens": 67987931.0,
      "step": 34750
    },
    {
      "entropy": 5.969304418563842,
      "epoch": 3.4741340530814213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038287003199606026,
      "loss": 5.53,
      "mean_token_accuracy": 0.18820810168981553,
      "num_tokens": 67998199.0,
      "step": 34755
    },
    {
      "entropy": 5.915692329406738,
      "epoch": 3.47463387814265,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003828387104440659,
      "loss": 5.4744,
      "mean_token_accuracy": 0.2053638443350792,
      "num_tokens": 68010650.0,
      "step": 34760
    },
    {
      "entropy": 5.994839668273926,
      "epoch": 3.4751337032038787,
      "grad_norm": 1.125,
      "learning_rate": 0.0003828073861788773,
      "loss": 5.5053,
      "mean_token_accuracy": 0.19292525202035904,
      "num_tokens": 68019803.0,
      "step": 34765
    },
    {
      "entropy": 6.000560474395752,
      "epoch": 3.4756335282651074,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038277605920128275,
      "loss": 5.5217,
      "mean_token_accuracy": 0.19146130084991456,
      "num_tokens": 68029809.0,
      "step": 34770
    },
    {
      "entropy": 5.950550174713134,
      "epoch": 3.4761333533263357,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003827447295120702,
      "loss": 5.3862,
      "mean_token_accuracy": 0.20256172567605973,
      "num_tokens": 68039644.0,
      "step": 34775
    },
    {
      "entropy": 5.932251930236816,
      "epoch": 3.4766331783875644,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038271339711202796,
      "loss": 5.418,
      "mean_token_accuracy": 0.1909886881709099,
      "num_tokens": 68049610.0,
      "step": 34780
    },
    {
      "entropy": 6.041212272644043,
      "epoch": 3.477133003448793,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003826820620019444,
      "loss": 5.5379,
      "mean_token_accuracy": 0.1869180366396904,
      "num_tokens": 68059959.0,
      "step": 34785
    },
    {
      "entropy": 6.00243182182312,
      "epoch": 3.4776328285100213,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003826507241826079,
      "loss": 5.504,
      "mean_token_accuracy": 0.18684251010417938,
      "num_tokens": 68070174.0,
      "step": 34790
    },
    {
      "entropy": 5.897769927978516,
      "epoch": 3.47813265357125,
      "grad_norm": 1.125,
      "learning_rate": 0.00038261938365480677,
      "loss": 5.4368,
      "mean_token_accuracy": 0.20212691128253937,
      "num_tokens": 68081080.0,
      "step": 34795
    },
    {
      "entropy": 5.9178466796875,
      "epoch": 3.4786324786324787,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003825880404193297,
      "loss": 5.4527,
      "mean_token_accuracy": 0.19425536841154098,
      "num_tokens": 68090595.0,
      "step": 34800
    },
    {
      "entropy": 5.981298780441284,
      "epoch": 3.4791323036937074,
      "grad_norm": 1.234375,
      "learning_rate": 0.00038255669447696523,
      "loss": 5.4692,
      "mean_token_accuracy": 0.19942354559898376,
      "num_tokens": 68100187.0,
      "step": 34805
    },
    {
      "entropy": 5.960832786560059,
      "epoch": 3.4796321287549357,
      "grad_norm": 1.09375,
      "learning_rate": 0.000382525345828502,
      "loss": 5.4517,
      "mean_token_accuracy": 0.190430748462677,
      "num_tokens": 68109793.0,
      "step": 34810
    },
    {
      "entropy": 5.961051940917969,
      "epoch": 3.4801319538161644,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003824939944747288,
      "loss": 5.5301,
      "mean_token_accuracy": 0.19027723520994186,
      "num_tokens": 68119773.0,
      "step": 34815
    },
    {
      "entropy": 6.015424108505249,
      "epoch": 3.480631778877393,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003824626404164343,
      "loss": 5.441,
      "mean_token_accuracy": 0.19524428993463516,
      "num_tokens": 68129839.0,
      "step": 34820
    },
    {
      "entropy": 5.97985053062439,
      "epoch": 3.4811316039386213,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038243128365440744,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19782850295305252,
      "num_tokens": 68139168.0,
      "step": 34825
    },
    {
      "entropy": 5.881820249557495,
      "epoch": 3.48163142899985,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000382399924189437,
      "loss": 5.4213,
      "mean_token_accuracy": 0.20597147494554519,
      "num_tokens": 68148062.0,
      "step": 34830
    },
    {
      "entropy": 5.9155433177948,
      "epoch": 3.4821312540610787,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003823685620223122,
      "loss": 5.4772,
      "mean_token_accuracy": 0.1973780408501625,
      "num_tokens": 68157266.0,
      "step": 34835
    },
    {
      "entropy": 5.946260929107666,
      "epoch": 3.4826310791223074,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003823371971538221,
      "loss": 5.4342,
      "mean_token_accuracy": 0.19901068806648253,
      "num_tokens": 68167156.0,
      "step": 34840
    },
    {
      "entropy": 6.088600301742554,
      "epoch": 3.4831309041835357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038230582958475565,
      "loss": 5.5472,
      "mean_token_accuracy": 0.18551700562238693,
      "num_tokens": 68176900.0,
      "step": 34845
    },
    {
      "entropy": 5.9332483291625975,
      "epoch": 3.4836307292447644,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038227445931590215,
      "loss": 5.437,
      "mean_token_accuracy": 0.2033606693148613,
      "num_tokens": 68186193.0,
      "step": 34850
    },
    {
      "entropy": 5.926942777633667,
      "epoch": 3.484130554305993,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003822430863480508,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19668649584054948,
      "num_tokens": 68195962.0,
      "step": 34855
    },
    {
      "entropy": 5.998989868164062,
      "epoch": 3.4846303793672213,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038221171068199104,
      "loss": 5.5469,
      "mean_token_accuracy": 0.18671034127473832,
      "num_tokens": 68205002.0,
      "step": 34860
    },
    {
      "entropy": 5.965097332000733,
      "epoch": 3.48513020442845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003821803323185123,
      "loss": 5.398,
      "mean_token_accuracy": 0.20205953866243362,
      "num_tokens": 68213229.0,
      "step": 34865
    },
    {
      "entropy": 6.021445560455322,
      "epoch": 3.4856300294896787,
      "grad_norm": 1.125,
      "learning_rate": 0.0003821489512584039,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19135491847991942,
      "num_tokens": 68223390.0,
      "step": 34870
    },
    {
      "entropy": 5.983006143569947,
      "epoch": 3.4861298545509074,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038211756750245547,
      "loss": 5.5665,
      "mean_token_accuracy": 0.18429654091596603,
      "num_tokens": 68233363.0,
      "step": 34875
    },
    {
      "entropy": 6.045814561843872,
      "epoch": 3.4866296796121357,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003820861810514566,
      "loss": 5.5149,
      "mean_token_accuracy": 0.19166799932718276,
      "num_tokens": 68245139.0,
      "step": 34880
    },
    {
      "entropy": 6.024434185028076,
      "epoch": 3.4871295046733644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038205479190619703,
      "loss": 5.5432,
      "mean_token_accuracy": 0.1918042778968811,
      "num_tokens": 68254347.0,
      "step": 34885
    },
    {
      "entropy": 6.010404205322265,
      "epoch": 3.487629329734593,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00038202340006746635,
      "loss": 5.5377,
      "mean_token_accuracy": 0.19054551422595978,
      "num_tokens": 68263900.0,
      "step": 34890
    },
    {
      "entropy": 5.974705839157105,
      "epoch": 3.4881291547958213,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003819920055360545,
      "loss": 5.3745,
      "mean_token_accuracy": 0.19999324828386306,
      "num_tokens": 68273348.0,
      "step": 34895
    },
    {
      "entropy": 5.891230916976928,
      "epoch": 3.48862897985705,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038196060831275134,
      "loss": 5.399,
      "mean_token_accuracy": 0.19872559159994124,
      "num_tokens": 68282448.0,
      "step": 34900
    },
    {
      "entropy": 5.903797292709351,
      "epoch": 3.4891288049182787,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003819292083983468,
      "loss": 5.4032,
      "mean_token_accuracy": 0.1996441438794136,
      "num_tokens": 68291945.0,
      "step": 34905
    },
    {
      "entropy": 5.877175140380859,
      "epoch": 3.4896286299795074,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003818978057936308,
      "loss": 5.3722,
      "mean_token_accuracy": 0.200311017036438,
      "num_tokens": 68302072.0,
      "step": 34910
    },
    {
      "entropy": 5.953031539916992,
      "epoch": 3.4901284550407357,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00038186640049939365,
      "loss": 5.4959,
      "mean_token_accuracy": 0.1909414201974869,
      "num_tokens": 68310968.0,
      "step": 34915
    },
    {
      "entropy": 5.9791693687438965,
      "epoch": 3.4906282801019644,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003818349925164253,
      "loss": 5.5113,
      "mean_token_accuracy": 0.19166507720947265,
      "num_tokens": 68320925.0,
      "step": 34920
    },
    {
      "entropy": 6.008647775650024,
      "epoch": 3.491128105163193,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000381803581845516,
      "loss": 5.4732,
      "mean_token_accuracy": 0.19687847197055816,
      "num_tokens": 68331440.0,
      "step": 34925
    },
    {
      "entropy": 5.963428258895874,
      "epoch": 3.4916279302244213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00038177216848745614,
      "loss": 5.4821,
      "mean_token_accuracy": 0.18992066532373428,
      "num_tokens": 68340726.0,
      "step": 34930
    },
    {
      "entropy": 5.908040761947632,
      "epoch": 3.49212775528565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003817407524430359,
      "loss": 5.4089,
      "mean_token_accuracy": 0.201625294983387,
      "num_tokens": 68349835.0,
      "step": 34935
    },
    {
      "entropy": 5.895816040039063,
      "epoch": 3.4926275803468787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038170933371304585,
      "loss": 5.4847,
      "mean_token_accuracy": 0.2017952799797058,
      "num_tokens": 68360270.0,
      "step": 34940
    },
    {
      "entropy": 5.982848453521728,
      "epoch": 3.4931274054081074,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003816779122982765,
      "loss": 5.3784,
      "mean_token_accuracy": 0.20236151665449142,
      "num_tokens": 68370268.0,
      "step": 34945
    },
    {
      "entropy": 5.9399330615997314,
      "epoch": 3.4936272304693357,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003816464881995182,
      "loss": 5.4062,
      "mean_token_accuracy": 0.1994234099984169,
      "num_tokens": 68379698.0,
      "step": 34950
    },
    {
      "entropy": 5.992791080474854,
      "epoch": 3.4941270555305644,
      "grad_norm": 1.125,
      "learning_rate": 0.0003816150614175617,
      "loss": 5.4978,
      "mean_token_accuracy": 0.19231075793504715,
      "num_tokens": 68388825.0,
      "step": 34955
    },
    {
      "entropy": 6.0221028327941895,
      "epoch": 3.494626880591793,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003815836319531978,
      "loss": 5.5547,
      "mean_token_accuracy": 0.1845039203763008,
      "num_tokens": 68398553.0,
      "step": 34960
    },
    {
      "entropy": 5.865931510925293,
      "epoch": 3.4951267056530213,
      "grad_norm": 1.078125,
      "learning_rate": 0.000381552199807217,
      "loss": 5.3525,
      "mean_token_accuracy": 0.20461723506450652,
      "num_tokens": 68407933.0,
      "step": 34965
    },
    {
      "entropy": 5.947754812240601,
      "epoch": 3.49562653071425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003815207649804104,
      "loss": 5.5112,
      "mean_token_accuracy": 0.19127586781978606,
      "num_tokens": 68417224.0,
      "step": 34970
    },
    {
      "entropy": 5.951096439361573,
      "epoch": 3.4961263557754787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003814893274735687,
      "loss": 5.4411,
      "mean_token_accuracy": 0.19685624837875365,
      "num_tokens": 68426614.0,
      "step": 34975
    },
    {
      "entropy": 5.9537595272064205,
      "epoch": 3.4966261808367074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00038145788728748294,
      "loss": 5.4369,
      "mean_token_accuracy": 0.19770211577415467,
      "num_tokens": 68435681.0,
      "step": 34980
    },
    {
      "entropy": 5.990784597396851,
      "epoch": 3.4971260058979357,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003814264444229441,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18715455532073974,
      "num_tokens": 68444900.0,
      "step": 34985
    },
    {
      "entropy": 5.973998928070069,
      "epoch": 3.4976258309591644,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003813949988807434,
      "loss": 5.477,
      "mean_token_accuracy": 0.19569265395402907,
      "num_tokens": 68453810.0,
      "step": 34990
    },
    {
      "entropy": 5.953943347930908,
      "epoch": 3.498125656020393,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038136355066167177,
      "loss": 5.4262,
      "mean_token_accuracy": 0.19820523709058763,
      "num_tokens": 68462850.0,
      "step": 34995
    },
    {
      "entropy": 6.052598524093628,
      "epoch": 3.4986254810816213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003813320997665207,
      "loss": 5.6126,
      "mean_token_accuracy": 0.18894586861133575,
      "num_tokens": 68472900.0,
      "step": 35000
    },
    {
      "entropy": 5.956964206695557,
      "epoch": 3.49912530614285,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00038130064619608123,
      "loss": 5.4273,
      "mean_token_accuracy": 0.20568559914827347,
      "num_tokens": 68482620.0,
      "step": 35005
    },
    {
      "entropy": 5.976876831054687,
      "epoch": 3.4996251312040787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000381269189951145,
      "loss": 5.4392,
      "mean_token_accuracy": 0.19713898450136186,
      "num_tokens": 68492145.0,
      "step": 35010
    },
    {
      "entropy": 6.003053045272827,
      "epoch": 3.5001249562653074,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003812377310325032,
      "loss": 5.499,
      "mean_token_accuracy": 0.19013407826423645,
      "num_tokens": 68501193.0,
      "step": 35015
    },
    {
      "entropy": 5.969914150238037,
      "epoch": 3.5006247813265357,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003812062694409474,
      "loss": 5.4807,
      "mean_token_accuracy": 0.19109124541282654,
      "num_tokens": 68511525.0,
      "step": 35020
    },
    {
      "entropy": 5.953226566314697,
      "epoch": 3.5011246063877643,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003811748051772693,
      "loss": 5.5227,
      "mean_token_accuracy": 0.19420771449804305,
      "num_tokens": 68521612.0,
      "step": 35025
    },
    {
      "entropy": 6.024365329742432,
      "epoch": 3.5016244314489926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00038114333824226034,
      "loss": 5.5999,
      "mean_token_accuracy": 0.18246715366840363,
      "num_tokens": 68531986.0,
      "step": 35030
    },
    {
      "entropy": 6.0187310695648195,
      "epoch": 3.5021242565102213,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003811118686367123,
      "loss": 5.5141,
      "mean_token_accuracy": 0.19322337955236435,
      "num_tokens": 68541008.0,
      "step": 35035
    },
    {
      "entropy": 6.032729768753052,
      "epoch": 3.50262408157145,
      "grad_norm": 1.171875,
      "learning_rate": 0.000381080396361417,
      "loss": 5.5246,
      "mean_token_accuracy": 0.19534792602062226,
      "num_tokens": 68550805.0,
      "step": 35040
    },
    {
      "entropy": 5.974455976486206,
      "epoch": 3.5031239066326787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003810489214171662,
      "loss": 5.5058,
      "mean_token_accuracy": 0.19315775036811828,
      "num_tokens": 68561428.0,
      "step": 35045
    },
    {
      "entropy": 5.919817638397217,
      "epoch": 3.5036237316939074,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003810174438047517,
      "loss": 5.469,
      "mean_token_accuracy": 0.19294420182704924,
      "num_tokens": 68569624.0,
      "step": 35050
    },
    {
      "entropy": 5.99331431388855,
      "epoch": 3.5041235567551356,
      "grad_norm": 1.125,
      "learning_rate": 0.0003809859635249658,
      "loss": 5.4396,
      "mean_token_accuracy": 0.1928870275616646,
      "num_tokens": 68579464.0,
      "step": 35055
    },
    {
      "entropy": 5.995477199554443,
      "epoch": 3.5046233818163643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003809544805786003,
      "loss": 5.5192,
      "mean_token_accuracy": 0.1922709196805954,
      "num_tokens": 68589440.0,
      "step": 35060
    },
    {
      "entropy": 5.973752689361572,
      "epoch": 3.5051232068775926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003809229949664473,
      "loss": 5.4744,
      "mean_token_accuracy": 0.1889815390110016,
      "num_tokens": 68599267.0,
      "step": 35065
    },
    {
      "entropy": 5.918180131912232,
      "epoch": 3.5056230319388213,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003808915066892989,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19942651987075805,
      "num_tokens": 68608976.0,
      "step": 35070
    },
    {
      "entropy": 6.083821487426758,
      "epoch": 3.50612285700005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00038086001574794755,
      "loss": 5.5729,
      "mean_token_accuracy": 0.18355290293693544,
      "num_tokens": 68618342.0,
      "step": 35075
    },
    {
      "entropy": 5.962140989303589,
      "epoch": 3.5066226820612787,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003808285221431854,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20146614015102388,
      "num_tokens": 68628519.0,
      "step": 35080
    },
    {
      "entropy": 6.009766912460327,
      "epoch": 3.5071225071225074,
      "grad_norm": 1.359375,
      "learning_rate": 0.00038079702587580486,
      "loss": 5.6091,
      "mean_token_accuracy": 0.1801582396030426,
      "num_tokens": 68639408.0,
      "step": 35085
    },
    {
      "entropy": 5.940073776245117,
      "epoch": 3.5076223321837356,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00038076552694659837,
      "loss": 5.4757,
      "mean_token_accuracy": 0.1942267656326294,
      "num_tokens": 68650484.0,
      "step": 35090
    },
    {
      "entropy": 6.008402729034424,
      "epoch": 3.5081221572449643,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00038073402535635854,
      "loss": 5.5352,
      "mean_token_accuracy": 0.18884797394275665,
      "num_tokens": 68660567.0,
      "step": 35095
    },
    {
      "entropy": 5.94819188117981,
      "epoch": 3.5086219823061926,
      "grad_norm": 0.96875,
      "learning_rate": 0.00038070252110587776,
      "loss": 5.5065,
      "mean_token_accuracy": 0.19635921269655227,
      "num_tokens": 68671340.0,
      "step": 35100
    },
    {
      "entropy": 6.00738787651062,
      "epoch": 3.5091218073674213,
      "grad_norm": 1.265625,
      "learning_rate": 0.00038067101419594874,
      "loss": 5.4828,
      "mean_token_accuracy": 0.19762916415929793,
      "num_tokens": 68680866.0,
      "step": 35105
    },
    {
      "entropy": 5.774124050140381,
      "epoch": 3.50962163242865,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038063950462736425,
      "loss": 5.275,
      "mean_token_accuracy": 0.2207723543047905,
      "num_tokens": 68690682.0,
      "step": 35110
    },
    {
      "entropy": 5.996037244796753,
      "epoch": 3.5101214574898787,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003806079924009169,
      "loss": 5.5163,
      "mean_token_accuracy": 0.19267436861991882,
      "num_tokens": 68700586.0,
      "step": 35115
    },
    {
      "entropy": 5.968429660797119,
      "epoch": 3.510621282551107,
      "grad_norm": 1.0625,
      "learning_rate": 0.00038057647751739963,
      "loss": 5.4615,
      "mean_token_accuracy": 0.19349278658628463,
      "num_tokens": 68710878.0,
      "step": 35120
    },
    {
      "entropy": 5.992898035049438,
      "epoch": 3.5111211076123356,
      "grad_norm": 1.125,
      "learning_rate": 0.0003805449599776055,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1834139958024025,
      "num_tokens": 68720611.0,
      "step": 35125
    },
    {
      "entropy": 5.906705665588379,
      "epoch": 3.5116209326735643,
      "grad_norm": 1.390625,
      "learning_rate": 0.0003805134397823273,
      "loss": 5.4244,
      "mean_token_accuracy": 0.20380700826644899,
      "num_tokens": 68730871.0,
      "step": 35130
    },
    {
      "entropy": 5.946548795700073,
      "epoch": 3.5121207577347926,
      "grad_norm": 1.046875,
      "learning_rate": 0.000380481916932358,
      "loss": 5.3519,
      "mean_token_accuracy": 0.2038868173956871,
      "num_tokens": 68740981.0,
      "step": 35135
    },
    {
      "entropy": 5.9735167026519775,
      "epoch": 3.5126205827960213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003804503914284909,
      "loss": 5.4364,
      "mean_token_accuracy": 0.20040358752012252,
      "num_tokens": 68751323.0,
      "step": 35140
    },
    {
      "entropy": 6.001402568817139,
      "epoch": 3.51312040785725,
      "grad_norm": 1.125,
      "learning_rate": 0.00038041886327151904,
      "loss": 5.5688,
      "mean_token_accuracy": 0.19304869771003724,
      "num_tokens": 68761803.0,
      "step": 35145
    },
    {
      "entropy": 5.938369560241699,
      "epoch": 3.5136202329184787,
      "grad_norm": 1.140625,
      "learning_rate": 0.00038038733246223574,
      "loss": 5.4553,
      "mean_token_accuracy": 0.1979713574051857,
      "num_tokens": 68770455.0,
      "step": 35150
    },
    {
      "entropy": 5.94798264503479,
      "epoch": 3.514120057979707,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003803557990014343,
      "loss": 5.5074,
      "mean_token_accuracy": 0.19169859290122987,
      "num_tokens": 68779826.0,
      "step": 35155
    },
    {
      "entropy": 5.946558237075806,
      "epoch": 3.5146198830409356,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000380324262889908,
      "loss": 5.4762,
      "mean_token_accuracy": 0.19114730656147003,
      "num_tokens": 68789323.0,
      "step": 35160
    },
    {
      "entropy": 5.980030918121338,
      "epoch": 3.5151197081021643,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003802927241284504,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19730884134769439,
      "num_tokens": 68799967.0,
      "step": 35165
    },
    {
      "entropy": 5.8883990287780765,
      "epoch": 3.5156195331633926,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003802611827178549,
      "loss": 5.3649,
      "mean_token_accuracy": 0.21003871262073517,
      "num_tokens": 68808997.0,
      "step": 35170
    },
    {
      "entropy": 5.987416696548462,
      "epoch": 3.5161193582246213,
      "grad_norm": 1.09375,
      "learning_rate": 0.00038022963865891514,
      "loss": 5.4641,
      "mean_token_accuracy": 0.1957197442650795,
      "num_tokens": 68818597.0,
      "step": 35175
    },
    {
      "entropy": 5.9527668952941895,
      "epoch": 3.51661918328585,
      "grad_norm": 1.28125,
      "learning_rate": 0.00038019809195242484,
      "loss": 5.4119,
      "mean_token_accuracy": 0.19919167906045915,
      "num_tokens": 68827813.0,
      "step": 35180
    },
    {
      "entropy": 5.909467124938965,
      "epoch": 3.5171190083470787,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003801665425991775,
      "loss": 5.457,
      "mean_token_accuracy": 0.19638947248458863,
      "num_tokens": 68836359.0,
      "step": 35185
    },
    {
      "entropy": 5.906164216995239,
      "epoch": 3.517618833408307,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003801349905999671,
      "loss": 5.4956,
      "mean_token_accuracy": 0.19149542301893235,
      "num_tokens": 68845063.0,
      "step": 35190
    },
    {
      "entropy": 6.0193463325500485,
      "epoch": 3.5181186584695356,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003801034359555873,
      "loss": 5.5401,
      "mean_token_accuracy": 0.18392157852649688,
      "num_tokens": 68855262.0,
      "step": 35195
    },
    {
      "entropy": 6.028322172164917,
      "epoch": 3.5186184835307643,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00038007187866683217,
      "loss": 5.4605,
      "mean_token_accuracy": 0.19701593667268752,
      "num_tokens": 68864768.0,
      "step": 35200
    },
    {
      "entropy": 5.9928045749664305,
      "epoch": 3.5191183085919926,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003800403187344955,
      "loss": 5.4413,
      "mean_token_accuracy": 0.19474657922983168,
      "num_tokens": 68874950.0,
      "step": 35205
    },
    {
      "entropy": 5.940837287902832,
      "epoch": 3.5196181336532213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003800087561593716,
      "loss": 5.5122,
      "mean_token_accuracy": 0.19430647492408754,
      "num_tokens": 68884331.0,
      "step": 35210
    },
    {
      "entropy": 5.8204484462738035,
      "epoch": 3.52011795871445,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003799771909422543,
      "loss": 5.3355,
      "mean_token_accuracy": 0.21305233538150786,
      "num_tokens": 68894610.0,
      "step": 35215
    },
    {
      "entropy": 5.969574356079102,
      "epoch": 3.5206177837756787,
      "grad_norm": 2.5625,
      "learning_rate": 0.00037994562308393793,
      "loss": 5.5119,
      "mean_token_accuracy": 0.19916877299547195,
      "num_tokens": 68904482.0,
      "step": 35220
    },
    {
      "entropy": 6.0218000411987305,
      "epoch": 3.521117608836907,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003799140525852167,
      "loss": 5.4619,
      "mean_token_accuracy": 0.19283913373947142,
      "num_tokens": 68914219.0,
      "step": 35225
    },
    {
      "entropy": 5.935389041900635,
      "epoch": 3.5216174338981356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003798824794468849,
      "loss": 5.3958,
      "mean_token_accuracy": 0.19517311751842498,
      "num_tokens": 68922980.0,
      "step": 35230
    },
    {
      "entropy": 5.930998992919922,
      "epoch": 3.5221172589593643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003798509036697369,
      "loss": 5.4587,
      "mean_token_accuracy": 0.19832732379436493,
      "num_tokens": 68933014.0,
      "step": 35235
    },
    {
      "entropy": 5.93671555519104,
      "epoch": 3.5226170840205926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037981932525456714,
      "loss": 5.3956,
      "mean_token_accuracy": 0.19217992573976517,
      "num_tokens": 68942942.0,
      "step": 35240
    },
    {
      "entropy": 5.903719234466553,
      "epoch": 3.5231169090818213,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003797877442021701,
      "loss": 5.3925,
      "mean_token_accuracy": 0.20363486111164092,
      "num_tokens": 68953575.0,
      "step": 35245
    },
    {
      "entropy": 5.944115161895752,
      "epoch": 3.52361673414305,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003797561605133404,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19229235351085663,
      "num_tokens": 68963549.0,
      "step": 35250
    },
    {
      "entropy": 6.015992212295532,
      "epoch": 3.5241165592042787,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003797245741888727,
      "loss": 5.5435,
      "mean_token_accuracy": 0.1935306578874588,
      "num_tokens": 68973616.0,
      "step": 35255
    },
    {
      "entropy": 5.987265777587891,
      "epoch": 3.524616384265507,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037969298522956156,
      "loss": 5.413,
      "mean_token_accuracy": 0.1929441958665848,
      "num_tokens": 68983786.0,
      "step": 35260
    },
    {
      "entropy": 5.93670802116394,
      "epoch": 3.5251162093267356,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037966139363620197,
      "loss": 5.4253,
      "mean_token_accuracy": 0.2044527232646942,
      "num_tokens": 68993900.0,
      "step": 35265
    },
    {
      "entropy": 6.028407621383667,
      "epoch": 3.5256160343879643,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003796297994095886,
      "loss": 5.5775,
      "mean_token_accuracy": 0.18919315934181213,
      "num_tokens": 69003062.0,
      "step": 35270
    },
    {
      "entropy": 5.80401291847229,
      "epoch": 3.5261158594491926,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003795982025505164,
      "loss": 5.2498,
      "mean_token_accuracy": 0.2108839675784111,
      "num_tokens": 69011466.0,
      "step": 35275
    },
    {
      "entropy": 5.911711931228638,
      "epoch": 3.5266156845104213,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003795666030597803,
      "loss": 5.4506,
      "mean_token_accuracy": 0.1973807618021965,
      "num_tokens": 69020821.0,
      "step": 35280
    },
    {
      "entropy": 6.0337423324584964,
      "epoch": 3.52711550957165,
      "grad_norm": 1.015625,
      "learning_rate": 0.00037953500093817544,
      "loss": 5.5278,
      "mean_token_accuracy": 0.19172074049711227,
      "num_tokens": 69031788.0,
      "step": 35285
    },
    {
      "entropy": 5.895536613464356,
      "epoch": 3.5276153346328787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037950339618649675,
      "loss": 5.4094,
      "mean_token_accuracy": 0.1966879650950432,
      "num_tokens": 69041520.0,
      "step": 35290
    },
    {
      "entropy": 5.956196308135986,
      "epoch": 3.528115159694107,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003794717888055397,
      "loss": 5.5331,
      "mean_token_accuracy": 0.19159616976976396,
      "num_tokens": 69050781.0,
      "step": 35295
    },
    {
      "entropy": 5.963995313644409,
      "epoch": 3.5286149847553356,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037944017879609917,
      "loss": 5.4877,
      "mean_token_accuracy": 0.20020514279603957,
      "num_tokens": 69062108.0,
      "step": 35300
    },
    {
      "entropy": 6.038738155364991,
      "epoch": 3.5291148098165643,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003794085661589707,
      "loss": 5.4405,
      "mean_token_accuracy": 0.19633643329143524,
      "num_tokens": 69071695.0,
      "step": 35305
    },
    {
      "entropy": 6.067940902709961,
      "epoch": 3.5296146348777926,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037937695089494943,
      "loss": 5.4754,
      "mean_token_accuracy": 0.1952625259757042,
      "num_tokens": 69081431.0,
      "step": 35310
    },
    {
      "entropy": 5.88950138092041,
      "epoch": 3.5301144599390213,
      "grad_norm": 1.15625,
      "learning_rate": 0.000379345333004831,
      "loss": 5.4778,
      "mean_token_accuracy": 0.20432850420475007,
      "num_tokens": 69091907.0,
      "step": 35315
    },
    {
      "entropy": 5.965456104278564,
      "epoch": 3.53061428500025,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003793137124894109,
      "loss": 5.4647,
      "mean_token_accuracy": 0.19498918056488038,
      "num_tokens": 69100655.0,
      "step": 35320
    },
    {
      "entropy": 5.960495328903198,
      "epoch": 3.5311141100614787,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00037928208934948467,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20457542538642884,
      "num_tokens": 69111253.0,
      "step": 35325
    },
    {
      "entropy": 5.960485219955444,
      "epoch": 3.531613935122707,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037925046358584775,
      "loss": 5.4261,
      "mean_token_accuracy": 0.19546428471803665,
      "num_tokens": 69121658.0,
      "step": 35330
    },
    {
      "entropy": 6.039993619918823,
      "epoch": 3.5321137601839356,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003792188351992961,
      "loss": 5.5447,
      "mean_token_accuracy": 0.19380109012126923,
      "num_tokens": 69131572.0,
      "step": 35335
    },
    {
      "entropy": 5.9369899272918705,
      "epoch": 3.5326135852451643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037918720419062536,
      "loss": 5.4292,
      "mean_token_accuracy": 0.190799717605114,
      "num_tokens": 69143265.0,
      "step": 35340
    },
    {
      "entropy": 5.972865629196167,
      "epoch": 3.5331134103063926,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003791555705606314,
      "loss": 5.5264,
      "mean_token_accuracy": 0.19413284808397294,
      "num_tokens": 69152660.0,
      "step": 35345
    },
    {
      "entropy": 5.975875759124756,
      "epoch": 3.5336132353676213,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037912393431011005,
      "loss": 5.4469,
      "mean_token_accuracy": 0.19996673315763475,
      "num_tokens": 69162118.0,
      "step": 35350
    },
    {
      "entropy": 6.0916636943817135,
      "epoch": 3.53411306042885,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003790922954398573,
      "loss": 5.5422,
      "mean_token_accuracy": 0.18509762287139891,
      "num_tokens": 69172394.0,
      "step": 35355
    },
    {
      "entropy": 5.929638242721557,
      "epoch": 3.5346128854900787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003790606539506692,
      "loss": 5.422,
      "mean_token_accuracy": 0.19666023403406144,
      "num_tokens": 69181656.0,
      "step": 35360
    },
    {
      "entropy": 5.88312873840332,
      "epoch": 3.535112710551307,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037902900984334176,
      "loss": 5.3495,
      "mean_token_accuracy": 0.21098336726427078,
      "num_tokens": 69191309.0,
      "step": 35365
    },
    {
      "entropy": 5.99410309791565,
      "epoch": 3.5356125356125356,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037899736311867126,
      "loss": 5.5272,
      "mean_token_accuracy": 0.19165489077568054,
      "num_tokens": 69200835.0,
      "step": 35370
    },
    {
      "entropy": 5.93045015335083,
      "epoch": 3.5361123606737643,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037896571377745385,
      "loss": 5.4052,
      "mean_token_accuracy": 0.19850625097751617,
      "num_tokens": 69210810.0,
      "step": 35375
    },
    {
      "entropy": 5.986789321899414,
      "epoch": 3.5366121857349926,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003789340618204859,
      "loss": 5.4879,
      "mean_token_accuracy": 0.1954991787672043,
      "num_tokens": 69220411.0,
      "step": 35380
    },
    {
      "entropy": 5.985239505767822,
      "epoch": 3.5371120107962213,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003789024072485637,
      "loss": 5.5319,
      "mean_token_accuracy": 0.1955115705728531,
      "num_tokens": 69230304.0,
      "step": 35385
    },
    {
      "entropy": 6.042995882034302,
      "epoch": 3.53761183585745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037887075006248354,
      "loss": 5.5084,
      "mean_token_accuracy": 0.19230694621801375,
      "num_tokens": 69241199.0,
      "step": 35390
    },
    {
      "entropy": 6.021862602233886,
      "epoch": 3.5381116609186787,
      "grad_norm": 0.94921875,
      "learning_rate": 0.00037883909026304206,
      "loss": 5.5232,
      "mean_token_accuracy": 0.19686050117015838,
      "num_tokens": 69251869.0,
      "step": 35395
    },
    {
      "entropy": 5.9442305088043215,
      "epoch": 3.538611485979907,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037880742785103584,
      "loss": 5.3825,
      "mean_token_accuracy": 0.20407621413469315,
      "num_tokens": 69261940.0,
      "step": 35400
    },
    {
      "entropy": 5.90971941947937,
      "epoch": 3.5391113110411356,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037877576282726146,
      "loss": 5.4168,
      "mean_token_accuracy": 0.19509959667921067,
      "num_tokens": 69271471.0,
      "step": 35405
    },
    {
      "entropy": 5.938017463684082,
      "epoch": 3.5396111361023643,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037874409519251563,
      "loss": 5.4634,
      "mean_token_accuracy": 0.20337602347135544,
      "num_tokens": 69280370.0,
      "step": 35410
    },
    {
      "entropy": 5.944207763671875,
      "epoch": 3.5401109611635926,
      "grad_norm": 1.171875,
      "learning_rate": 0.000378712424947595,
      "loss": 5.459,
      "mean_token_accuracy": 0.19581639766693115,
      "num_tokens": 69290621.0,
      "step": 35415
    },
    {
      "entropy": 6.116810035705567,
      "epoch": 3.5406107862248213,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00037868075209329644,
      "loss": 5.5679,
      "mean_token_accuracy": 0.19442861676216125,
      "num_tokens": 69300613.0,
      "step": 35420
    },
    {
      "entropy": 5.965278148651123,
      "epoch": 3.54111061128605,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037864907663041686,
      "loss": 5.4129,
      "mean_token_accuracy": 0.2015095680952072,
      "num_tokens": 69310974.0,
      "step": 35425
    },
    {
      "entropy": 5.917321062088012,
      "epoch": 3.5416104363472787,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037861739855975316,
      "loss": 5.426,
      "mean_token_accuracy": 0.1984239861369133,
      "num_tokens": 69321155.0,
      "step": 35430
    },
    {
      "entropy": 5.887596797943115,
      "epoch": 3.542110261408507,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037858571788210236,
      "loss": 5.4299,
      "mean_token_accuracy": 0.1996580719947815,
      "num_tokens": 69331280.0,
      "step": 35435
    },
    {
      "entropy": 5.979969310760498,
      "epoch": 3.5426100864697356,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003785540345982616,
      "loss": 5.483,
      "mean_token_accuracy": 0.1912370428442955,
      "num_tokens": 69341350.0,
      "step": 35440
    },
    {
      "entropy": 5.98038535118103,
      "epoch": 3.5431099115309643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000378522348709028,
      "loss": 5.5015,
      "mean_token_accuracy": 0.19670553803443908,
      "num_tokens": 69352133.0,
      "step": 35445
    },
    {
      "entropy": 5.959635400772095,
      "epoch": 3.5436097365921926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003784906602151987,
      "loss": 5.5006,
      "mean_token_accuracy": 0.19385444670915603,
      "num_tokens": 69362141.0,
      "step": 35450
    },
    {
      "entropy": 5.89735221862793,
      "epoch": 3.5441095616534213,
      "grad_norm": 1.203125,
      "learning_rate": 0.000378458969117571,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19559195786714553,
      "num_tokens": 69371149.0,
      "step": 35455
    },
    {
      "entropy": 5.8925559520721436,
      "epoch": 3.54460938671465,
      "grad_norm": 1.375,
      "learning_rate": 0.00037842727541694226,
      "loss": 5.3632,
      "mean_token_accuracy": 0.21003697514533998,
      "num_tokens": 69381576.0,
      "step": 35460
    },
    {
      "entropy": 6.012968301773071,
      "epoch": 3.5451092117758787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003783955791141099,
      "loss": 5.5282,
      "mean_token_accuracy": 0.19139776080846788,
      "num_tokens": 69392917.0,
      "step": 35465
    },
    {
      "entropy": 6.011476039886475,
      "epoch": 3.545609036837107,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003783638802098714,
      "loss": 5.4764,
      "mean_token_accuracy": 0.19670365750789642,
      "num_tokens": 69402203.0,
      "step": 35470
    },
    {
      "entropy": 6.005444860458374,
      "epoch": 3.5461088618983356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003783321787050243,
      "loss": 5.5548,
      "mean_token_accuracy": 0.19188275337219238,
      "num_tokens": 69412910.0,
      "step": 35475
    },
    {
      "entropy": 5.910686016082764,
      "epoch": 3.546608686959564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003783004746003661,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19609311372041702,
      "num_tokens": 69422412.0,
      "step": 35480
    },
    {
      "entropy": 5.904728126525879,
      "epoch": 3.5471085120207926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003782687678966946,
      "loss": 5.4362,
      "mean_token_accuracy": 0.1967225506901741,
      "num_tokens": 69432123.0,
      "step": 35485
    },
    {
      "entropy": 5.94927487373352,
      "epoch": 3.5476083370820213,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003782370585948074,
      "loss": 5.3728,
      "mean_token_accuracy": 0.20117879360914231,
      "num_tokens": 69440844.0,
      "step": 35490
    },
    {
      "entropy": 5.92869348526001,
      "epoch": 3.54810816214325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003782053466955024,
      "loss": 5.4451,
      "mean_token_accuracy": 0.19597396850585938,
      "num_tokens": 69451466.0,
      "step": 35495
    },
    {
      "entropy": 5.9747633934021,
      "epoch": 3.5486079872044787,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037817363219957745,
      "loss": 5.5923,
      "mean_token_accuracy": 0.18340528011322021,
      "num_tokens": 69461322.0,
      "step": 35500
    },
    {
      "entropy": 5.991430044174194,
      "epoch": 3.549107812265707,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037814191510783047,
      "loss": 5.4135,
      "mean_token_accuracy": 0.20101094692945481,
      "num_tokens": 69469875.0,
      "step": 35505
    },
    {
      "entropy": 5.975218772888184,
      "epoch": 3.5496076373269356,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003781101954210594,
      "loss": 5.477,
      "mean_token_accuracy": 0.19830033630132676,
      "num_tokens": 69479774.0,
      "step": 35510
    },
    {
      "entropy": 5.972976446151733,
      "epoch": 3.550107462388164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037807847314006246,
      "loss": 5.3905,
      "mean_token_accuracy": 0.2053739070892334,
      "num_tokens": 69490259.0,
      "step": 35515
    },
    {
      "entropy": 5.904043483734131,
      "epoch": 3.5506072874493926,
      "grad_norm": 1.125,
      "learning_rate": 0.0003780467482656375,
      "loss": 5.4497,
      "mean_token_accuracy": 0.20167117565870285,
      "num_tokens": 69501050.0,
      "step": 35520
    },
    {
      "entropy": 5.920540809631348,
      "epoch": 3.5511071125106213,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037801502079858295,
      "loss": 5.5382,
      "mean_token_accuracy": 0.19120516180992125,
      "num_tokens": 69511475.0,
      "step": 35525
    },
    {
      "entropy": 5.820021438598633,
      "epoch": 3.55160693757185,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000377983290739697,
      "loss": 5.2281,
      "mean_token_accuracy": 0.22038473486900328,
      "num_tokens": 69521105.0,
      "step": 35530
    },
    {
      "entropy": 6.008819055557251,
      "epoch": 3.5521067626330787,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003779515580897779,
      "loss": 5.5542,
      "mean_token_accuracy": 0.18785459399223328,
      "num_tokens": 69530616.0,
      "step": 35535
    },
    {
      "entropy": 5.90145788192749,
      "epoch": 3.552606587694307,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037791982284962416,
      "loss": 5.408,
      "mean_token_accuracy": 0.2050342082977295,
      "num_tokens": 69540349.0,
      "step": 35540
    },
    {
      "entropy": 5.997347831726074,
      "epoch": 3.5531064127555356,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037788808502003414,
      "loss": 5.5275,
      "mean_token_accuracy": 0.185726597905159,
      "num_tokens": 69550134.0,
      "step": 35545
    },
    {
      "entropy": 5.991408967971802,
      "epoch": 3.553606237816764,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037785634460180633,
      "loss": 5.4928,
      "mean_token_accuracy": 0.19569769501686096,
      "num_tokens": 69560335.0,
      "step": 35550
    },
    {
      "entropy": 5.948108816146851,
      "epoch": 3.5541060628779926,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037782460159573944,
      "loss": 5.4389,
      "mean_token_accuracy": 0.18936973363161086,
      "num_tokens": 69569828.0,
      "step": 35555
    },
    {
      "entropy": 5.853258466720581,
      "epoch": 3.5546058879392213,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037779285600263195,
      "loss": 5.3299,
      "mean_token_accuracy": 0.21497069597244262,
      "num_tokens": 69579840.0,
      "step": 35560
    },
    {
      "entropy": 6.0629607200622555,
      "epoch": 3.55510571300045,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003777611078232827,
      "loss": 5.5898,
      "mean_token_accuracy": 0.19047681242227554,
      "num_tokens": 69589513.0,
      "step": 35565
    },
    {
      "entropy": 5.976849365234375,
      "epoch": 3.5556055380616787,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037772935705849046,
      "loss": 5.4589,
      "mean_token_accuracy": 0.19724978059530257,
      "num_tokens": 69598505.0,
      "step": 35570
    },
    {
      "entropy": 5.857768964767456,
      "epoch": 3.556105363122907,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003776976037090539,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20794083029031754,
      "num_tokens": 69608955.0,
      "step": 35575
    },
    {
      "entropy": 5.905356931686401,
      "epoch": 3.5566051881841356,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037766584777577223,
      "loss": 5.3788,
      "mean_token_accuracy": 0.20696185976266862,
      "num_tokens": 69618130.0,
      "step": 35580
    },
    {
      "entropy": 5.964679861068726,
      "epoch": 3.557105013245364,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003776340892594441,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19361442029476167,
      "num_tokens": 69627605.0,
      "step": 35585
    },
    {
      "entropy": 5.974945545196533,
      "epoch": 3.5576048383065926,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037760232816086875,
      "loss": 5.4421,
      "mean_token_accuracy": 0.19708698242902756,
      "num_tokens": 69637631.0,
      "step": 35590
    },
    {
      "entropy": 5.930099105834961,
      "epoch": 3.5581046633678213,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003775705644808452,
      "loss": 5.5455,
      "mean_token_accuracy": 0.18700893968343735,
      "num_tokens": 69647568.0,
      "step": 35595
    },
    {
      "entropy": 5.997428703308105,
      "epoch": 3.55860448842905,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037753879822017265,
      "loss": 5.4769,
      "mean_token_accuracy": 0.19992662519216536,
      "num_tokens": 69656768.0,
      "step": 35600
    },
    {
      "entropy": 6.045084238052368,
      "epoch": 3.5591043134902787,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003775070293796504,
      "loss": 5.5259,
      "mean_token_accuracy": 0.19376151859760285,
      "num_tokens": 69665758.0,
      "step": 35605
    },
    {
      "entropy": 6.002308464050293,
      "epoch": 3.559604138551507,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003774752579600776,
      "loss": 5.5806,
      "mean_token_accuracy": 0.19178764522075653,
      "num_tokens": 69675281.0,
      "step": 35610
    },
    {
      "entropy": 6.051943778991699,
      "epoch": 3.5601039636127356,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003774434839622537,
      "loss": 5.4792,
      "mean_token_accuracy": 0.18485872000455855,
      "num_tokens": 69685354.0,
      "step": 35615
    },
    {
      "entropy": 6.065635728836059,
      "epoch": 3.560603788673964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037741170738697806,
      "loss": 5.5737,
      "mean_token_accuracy": 0.18403851389884948,
      "num_tokens": 69695516.0,
      "step": 35620
    },
    {
      "entropy": 6.001170015335083,
      "epoch": 3.5611036137351926,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003773799282350503,
      "loss": 5.4749,
      "mean_token_accuracy": 0.1947905406355858,
      "num_tokens": 69704096.0,
      "step": 35625
    },
    {
      "entropy": 5.959004974365234,
      "epoch": 3.5616034387964213,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003773481465072698,
      "loss": 5.4468,
      "mean_token_accuracy": 0.2027274876832962,
      "num_tokens": 69714175.0,
      "step": 35630
    },
    {
      "entropy": 5.858661317825318,
      "epoch": 3.56210326385765,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037731636220443625,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20954469591379166,
      "num_tokens": 69723481.0,
      "step": 35635
    },
    {
      "entropy": 5.933830690383911,
      "epoch": 3.5626030889188787,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037728457532734943,
      "loss": 5.4288,
      "mean_token_accuracy": 0.1992910921573639,
      "num_tokens": 69733976.0,
      "step": 35640
    },
    {
      "entropy": 5.957467842102051,
      "epoch": 3.563102913980107,
      "grad_norm": 1.125,
      "learning_rate": 0.000377252785876809,
      "loss": 5.4467,
      "mean_token_accuracy": 0.19644124954938888,
      "num_tokens": 69744111.0,
      "step": 35645
    },
    {
      "entropy": 5.956220722198486,
      "epoch": 3.5636027390413356,
      "grad_norm": 1.265625,
      "learning_rate": 0.00037722099385361475,
      "loss": 5.4885,
      "mean_token_accuracy": 0.18952691107988356,
      "num_tokens": 69753629.0,
      "step": 35650
    },
    {
      "entropy": 5.950561285018921,
      "epoch": 3.564102564102564,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003771891992585667,
      "loss": 5.4399,
      "mean_token_accuracy": 0.20055257081985473,
      "num_tokens": 69762406.0,
      "step": 35655
    },
    {
      "entropy": 5.944299364089966,
      "epoch": 3.5646023891637926,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003771574020924646,
      "loss": 5.3709,
      "mean_token_accuracy": 0.19865902960300447,
      "num_tokens": 69771976.0,
      "step": 35660
    },
    {
      "entropy": 5.945953798294068,
      "epoch": 3.5651022142250213,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003771256023561086,
      "loss": 5.4517,
      "mean_token_accuracy": 0.1979595333337784,
      "num_tokens": 69781779.0,
      "step": 35665
    },
    {
      "entropy": 5.932467126846314,
      "epoch": 3.56560203928625,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003770938000502987,
      "loss": 5.4441,
      "mean_token_accuracy": 0.20260272771120072,
      "num_tokens": 69792254.0,
      "step": 35670
    },
    {
      "entropy": 5.998671960830689,
      "epoch": 3.5661018643474782,
      "grad_norm": 1.125,
      "learning_rate": 0.000377061995175835,
      "loss": 5.4861,
      "mean_token_accuracy": 0.19432442784309387,
      "num_tokens": 69801394.0,
      "step": 35675
    },
    {
      "entropy": 5.966099786758423,
      "epoch": 3.566601689408707,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003770301877335178,
      "loss": 5.4848,
      "mean_token_accuracy": 0.1909452646970749,
      "num_tokens": 69811273.0,
      "step": 35680
    },
    {
      "entropy": 5.966381788253784,
      "epoch": 3.5671015144699356,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037699837772414734,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20659606009721757,
      "num_tokens": 69820241.0,
      "step": 35685
    },
    {
      "entropy": 5.967731142044068,
      "epoch": 3.567601339531164,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000376966565148524,
      "loss": 5.4491,
      "mean_token_accuracy": 0.2021606296300888,
      "num_tokens": 69829650.0,
      "step": 35690
    },
    {
      "entropy": 6.078899335861206,
      "epoch": 3.5681011645923926,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037693475000744807,
      "loss": 5.5717,
      "mean_token_accuracy": 0.1795274794101715,
      "num_tokens": 69839934.0,
      "step": 35695
    },
    {
      "entropy": 6.018221807479859,
      "epoch": 3.5686009896536213,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037690293230172007,
      "loss": 5.5496,
      "mean_token_accuracy": 0.18612394481897354,
      "num_tokens": 69849406.0,
      "step": 35700
    },
    {
      "entropy": 5.930010175704956,
      "epoch": 3.56910081471485,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037687111203214054,
      "loss": 5.4189,
      "mean_token_accuracy": 0.19464795142412186,
      "num_tokens": 69858546.0,
      "step": 35705
    },
    {
      "entropy": 5.979265880584717,
      "epoch": 3.569600639776078,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037683928919951006,
      "loss": 5.4424,
      "mean_token_accuracy": 0.19805121123790742,
      "num_tokens": 69868752.0,
      "step": 35710
    },
    {
      "entropy": 5.907741641998291,
      "epoch": 3.570100464837307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037680746380462914,
      "loss": 5.3919,
      "mean_token_accuracy": 0.20038724541664124,
      "num_tokens": 69878601.0,
      "step": 35715
    },
    {
      "entropy": 6.000824069976806,
      "epoch": 3.5706002898985356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003767756358482987,
      "loss": 5.4465,
      "mean_token_accuracy": 0.19278993904590608,
      "num_tokens": 69888976.0,
      "step": 35720
    },
    {
      "entropy": 6.044857120513916,
      "epoch": 3.571100114959764,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003767438053313195,
      "loss": 5.5567,
      "mean_token_accuracy": 0.18517379611730575,
      "num_tokens": 69898532.0,
      "step": 35725
    },
    {
      "entropy": 5.921230125427246,
      "epoch": 3.5715999400209926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003767119722544923,
      "loss": 5.4805,
      "mean_token_accuracy": 0.20058141201734542,
      "num_tokens": 69907260.0,
      "step": 35730
    },
    {
      "entropy": 5.966745758056641,
      "epoch": 3.5720997650822213,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003766801366186181,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18678858280181884,
      "num_tokens": 69917347.0,
      "step": 35735
    },
    {
      "entropy": 5.931243133544922,
      "epoch": 3.57259959014345,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003766482984244978,
      "loss": 5.4299,
      "mean_token_accuracy": 0.19948402494192125,
      "num_tokens": 69926900.0,
      "step": 35740
    },
    {
      "entropy": 5.982673168182373,
      "epoch": 3.573099415204678,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003766164576729324,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1967747688293457,
      "num_tokens": 69935838.0,
      "step": 35745
    },
    {
      "entropy": 5.852257633209229,
      "epoch": 3.573599240265907,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037658461436472306,
      "loss": 5.3437,
      "mean_token_accuracy": 0.2095990374684334,
      "num_tokens": 69946310.0,
      "step": 35750
    },
    {
      "entropy": 5.857971286773681,
      "epoch": 3.5740990653271356,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00037655276850067103,
      "loss": 5.3174,
      "mean_token_accuracy": 0.2045376181602478,
      "num_tokens": 69955179.0,
      "step": 35755
    },
    {
      "entropy": 5.9960329055786135,
      "epoch": 3.574598890388364,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003765209200815776,
      "loss": 5.5321,
      "mean_token_accuracy": 0.18484777063131333,
      "num_tokens": 69965408.0,
      "step": 35760
    },
    {
      "entropy": 5.895020771026611,
      "epoch": 3.5750987154495926,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037648906910824374,
      "loss": 5.3534,
      "mean_token_accuracy": 0.21240172684192657,
      "num_tokens": 69973751.0,
      "step": 35765
    },
    {
      "entropy": 5.951747989654541,
      "epoch": 3.5755985405108213,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037645721558147104,
      "loss": 5.5124,
      "mean_token_accuracy": 0.19192036092281342,
      "num_tokens": 69983857.0,
      "step": 35770
    },
    {
      "entropy": 5.929650020599365,
      "epoch": 3.57609836557205,
      "grad_norm": 1.109375,
      "learning_rate": 0.000376425359502061,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20100677758455276,
      "num_tokens": 69993000.0,
      "step": 35775
    },
    {
      "entropy": 5.976902627944947,
      "epoch": 3.576598190633278,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037639350087081497,
      "loss": 5.4444,
      "mean_token_accuracy": 0.19238300770521163,
      "num_tokens": 70003009.0,
      "step": 35780
    },
    {
      "entropy": 5.959492301940918,
      "epoch": 3.577098015694507,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037636163968853453,
      "loss": 5.456,
      "mean_token_accuracy": 0.1933118686079979,
      "num_tokens": 70012246.0,
      "step": 35785
    },
    {
      "entropy": 5.916239881515503,
      "epoch": 3.5775978407557356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003763297759560213,
      "loss": 5.3947,
      "mean_token_accuracy": 0.19571837782859802,
      "num_tokens": 70022822.0,
      "step": 35790
    },
    {
      "entropy": 6.000546026229858,
      "epoch": 3.578097665816964,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003762979096740771,
      "loss": 5.569,
      "mean_token_accuracy": 0.18800393491983414,
      "num_tokens": 70032852.0,
      "step": 35795
    },
    {
      "entropy": 5.951161575317383,
      "epoch": 3.5785974908781926,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003762660408435035,
      "loss": 5.4445,
      "mean_token_accuracy": 0.192120523750782,
      "num_tokens": 70043367.0,
      "step": 35800
    },
    {
      "entropy": 6.00063009262085,
      "epoch": 3.5790973159394213,
      "grad_norm": 1.25,
      "learning_rate": 0.0003762341694651023,
      "loss": 5.4245,
      "mean_token_accuracy": 0.2063809335231781,
      "num_tokens": 70052935.0,
      "step": 35805
    },
    {
      "entropy": 5.947547483444214,
      "epoch": 3.57959714100065,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037620229553967543,
      "loss": 5.4909,
      "mean_token_accuracy": 0.1935177832841873,
      "num_tokens": 70064361.0,
      "step": 35810
    },
    {
      "entropy": 5.993914413452148,
      "epoch": 3.580096966061878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00037617041906802496,
      "loss": 5.4848,
      "mean_token_accuracy": 0.19329200685024261,
      "num_tokens": 70074183.0,
      "step": 35815
    },
    {
      "entropy": 5.9311998844146725,
      "epoch": 3.580596791123107,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003761385400509527,
      "loss": 5.4223,
      "mean_token_accuracy": 0.19622453004121782,
      "num_tokens": 70083929.0,
      "step": 35820
    },
    {
      "entropy": 5.948178100585937,
      "epoch": 3.5810966161843356,
      "grad_norm": 1.03125,
      "learning_rate": 0.00037610665848926086,
      "loss": 5.475,
      "mean_token_accuracy": 0.19715356081724167,
      "num_tokens": 70094956.0,
      "step": 35825
    },
    {
      "entropy": 5.9742053031921385,
      "epoch": 3.581596441245564,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003760747743837515,
      "loss": 5.5274,
      "mean_token_accuracy": 0.19731252491474152,
      "num_tokens": 70103814.0,
      "step": 35830
    },
    {
      "entropy": 5.917807388305664,
      "epoch": 3.5820962663067926,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037604288773522673,
      "loss": 5.3656,
      "mean_token_accuracy": 0.2075920134782791,
      "num_tokens": 70113715.0,
      "step": 35835
    },
    {
      "entropy": 5.9098375797271725,
      "epoch": 3.5825960913680213,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037601099854448893,
      "loss": 5.4198,
      "mean_token_accuracy": 0.19942557215690612,
      "num_tokens": 70123034.0,
      "step": 35840
    },
    {
      "entropy": 6.023796033859253,
      "epoch": 3.58309591642925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037597910681234047,
      "loss": 5.4943,
      "mean_token_accuracy": 0.196606408059597,
      "num_tokens": 70132592.0,
      "step": 35845
    },
    {
      "entropy": 5.9830334186553955,
      "epoch": 3.583595741490478,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003759472125395837,
      "loss": 5.5167,
      "mean_token_accuracy": 0.1928295224905014,
      "num_tokens": 70142288.0,
      "step": 35850
    },
    {
      "entropy": 5.930526065826416,
      "epoch": 3.584095566551707,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003759153157270209,
      "loss": 5.4784,
      "mean_token_accuracy": 0.19191448986530305,
      "num_tokens": 70152063.0,
      "step": 35855
    },
    {
      "entropy": 5.987587881088257,
      "epoch": 3.5845953916129356,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037588341637545487,
      "loss": 5.4716,
      "mean_token_accuracy": 0.19649477154016495,
      "num_tokens": 70162153.0,
      "step": 35860
    },
    {
      "entropy": 5.947652435302734,
      "epoch": 3.585095216674164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037585151448568796,
      "loss": 5.4261,
      "mean_token_accuracy": 0.1932845324277878,
      "num_tokens": 70171615.0,
      "step": 35865
    },
    {
      "entropy": 5.927873277664185,
      "epoch": 3.5855950417353926,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037581961005852296,
      "loss": 5.4429,
      "mean_token_accuracy": 0.19697312861680985,
      "num_tokens": 70181056.0,
      "step": 35870
    },
    {
      "entropy": 5.959801244735718,
      "epoch": 3.5860948667966213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003757877030947624,
      "loss": 5.4993,
      "mean_token_accuracy": 0.18803146928548814,
      "num_tokens": 70191336.0,
      "step": 35875
    },
    {
      "entropy": 6.013561201095581,
      "epoch": 3.58659469185785,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003757557935952092,
      "loss": 5.5692,
      "mean_token_accuracy": 0.17850475907325744,
      "num_tokens": 70200935.0,
      "step": 35880
    },
    {
      "entropy": 6.027308702468872,
      "epoch": 3.587094516919078,
      "grad_norm": 1.25,
      "learning_rate": 0.00037572388156066627,
      "loss": 5.4678,
      "mean_token_accuracy": 0.19577056765556336,
      "num_tokens": 70210111.0,
      "step": 35885
    },
    {
      "entropy": 5.84450397491455,
      "epoch": 3.587594341980307,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037569196699193633,
      "loss": 5.3105,
      "mean_token_accuracy": 0.20924149453639984,
      "num_tokens": 70219946.0,
      "step": 35890
    },
    {
      "entropy": 5.941036748886108,
      "epoch": 3.5880941670415356,
      "grad_norm": 1.234375,
      "learning_rate": 0.00037566004988982235,
      "loss": 5.5107,
      "mean_token_accuracy": 0.1902075454592705,
      "num_tokens": 70230145.0,
      "step": 35895
    },
    {
      "entropy": 6.024781274795532,
      "epoch": 3.588593992102764,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003756281302551274,
      "loss": 5.5223,
      "mean_token_accuracy": 0.19189031571149825,
      "num_tokens": 70239080.0,
      "step": 35900
    },
    {
      "entropy": 5.976364088058472,
      "epoch": 3.5890938171639926,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037559620808865467,
      "loss": 5.461,
      "mean_token_accuracy": 0.2036716803908348,
      "num_tokens": 70248936.0,
      "step": 35905
    },
    {
      "entropy": 5.907679796218872,
      "epoch": 3.5895936422252213,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037556428339120725,
      "loss": 5.3987,
      "mean_token_accuracy": 0.21232109218835832,
      "num_tokens": 70258355.0,
      "step": 35910
    },
    {
      "entropy": 5.920311498641968,
      "epoch": 3.59009346728645,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037553235616358835,
      "loss": 5.4538,
      "mean_token_accuracy": 0.20175427943468094,
      "num_tokens": 70267787.0,
      "step": 35915
    },
    {
      "entropy": 6.061601448059082,
      "epoch": 3.590593292347678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037550042640660113,
      "loss": 5.613,
      "mean_token_accuracy": 0.18679176419973373,
      "num_tokens": 70278695.0,
      "step": 35920
    },
    {
      "entropy": 5.935562944412231,
      "epoch": 3.591093117408907,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003754684941210491,
      "loss": 5.3484,
      "mean_token_accuracy": 0.21582671403884887,
      "num_tokens": 70288033.0,
      "step": 35925
    },
    {
      "entropy": 5.955764579772949,
      "epoch": 3.5915929424701356,
      "grad_norm": 1.1875,
      "learning_rate": 0.00037543655930773564,
      "loss": 5.5128,
      "mean_token_accuracy": 0.19318868815898896,
      "num_tokens": 70297431.0,
      "step": 35930
    },
    {
      "entropy": 6.003672409057617,
      "epoch": 3.592092767531364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003754046219674641,
      "loss": 5.5055,
      "mean_token_accuracy": 0.19085861444473268,
      "num_tokens": 70307693.0,
      "step": 35935
    },
    {
      "entropy": 5.955934047698975,
      "epoch": 3.5925925925925926,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037537268210103826,
      "loss": 5.4788,
      "mean_token_accuracy": 0.19917910695075988,
      "num_tokens": 70317121.0,
      "step": 35940
    },
    {
      "entropy": 5.95882248878479,
      "epoch": 3.5930924176538213,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003753407397092614,
      "loss": 5.4261,
      "mean_token_accuracy": 0.19478468149900435,
      "num_tokens": 70326564.0,
      "step": 35945
    },
    {
      "entropy": 5.976425361633301,
      "epoch": 3.59359224271505,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0003753087947929375,
      "loss": 5.5336,
      "mean_token_accuracy": 0.1866498351097107,
      "num_tokens": 70337668.0,
      "step": 35950
    },
    {
      "entropy": 5.974970769882202,
      "epoch": 3.594092067776278,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037527684735287007,
      "loss": 5.5028,
      "mean_token_accuracy": 0.19213390201330185,
      "num_tokens": 70347096.0,
      "step": 35955
    },
    {
      "entropy": 5.937854194641114,
      "epoch": 3.594591892837507,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003752448973898629,
      "loss": 5.3871,
      "mean_token_accuracy": 0.1994688853621483,
      "num_tokens": 70356814.0,
      "step": 35960
    },
    {
      "entropy": 5.922655916213989,
      "epoch": 3.5950917178987356,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037521294490472,
      "loss": 5.4019,
      "mean_token_accuracy": 0.2061278775334358,
      "num_tokens": 70365796.0,
      "step": 35965
    },
    {
      "entropy": 6.005922889709472,
      "epoch": 3.595591542959964,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003751809898982451,
      "loss": 5.5048,
      "mean_token_accuracy": 0.1911402612924576,
      "num_tokens": 70376411.0,
      "step": 35970
    },
    {
      "entropy": 5.86298131942749,
      "epoch": 3.5960913680211926,
      "grad_norm": 1.125,
      "learning_rate": 0.0003751490323712424,
      "loss": 5.3711,
      "mean_token_accuracy": 0.2004739448428154,
      "num_tokens": 70385758.0,
      "step": 35975
    },
    {
      "entropy": 5.922159481048584,
      "epoch": 3.5965911930824213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003751170723245157,
      "loss": 5.4029,
      "mean_token_accuracy": 0.2010003462433815,
      "num_tokens": 70395035.0,
      "step": 35980
    },
    {
      "entropy": 5.946831083297729,
      "epoch": 3.59709101814365,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037508510975886925,
      "loss": 5.4622,
      "mean_token_accuracy": 0.19572520554065703,
      "num_tokens": 70405620.0,
      "step": 35985
    },
    {
      "entropy": 5.983683490753174,
      "epoch": 3.597590843204878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00037505314467510726,
      "loss": 5.4997,
      "mean_token_accuracy": 0.1946408122777939,
      "num_tokens": 70415697.0,
      "step": 35990
    },
    {
      "entropy": 5.955062389373779,
      "epoch": 3.598090668266107,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003750211770740338,
      "loss": 5.3968,
      "mean_token_accuracy": 0.19864419847726822,
      "num_tokens": 70425121.0,
      "step": 35995
    },
    {
      "entropy": 5.955512475967407,
      "epoch": 3.598590493327335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003749892069564533,
      "loss": 5.4124,
      "mean_token_accuracy": 0.20334694534540176,
      "num_tokens": 70435140.0,
      "step": 36000
    },
    {
      "epoch": 3.598590493327335,
      "eval_entropy": 5.8067518071826045,
      "eval_loss": 5.585202217102051,
      "eval_mean_token_accuracy": 0.19816949023752137,
      "eval_num_tokens": 70435140.0,
      "eval_runtime": 24.0194,
      "eval_samples_per_second": 1292.497,
      "eval_steps_per_second": 161.578,
      "step": 36000
    },
    {
      "entropy": 6.06345157623291,
      "epoch": 3.599090318388564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003749572343231701,
      "loss": 5.5098,
      "mean_token_accuracy": 0.18940526247024536,
      "num_tokens": 70445058.0,
      "step": 36005
    },
    {
      "entropy": 5.952438640594482,
      "epoch": 3.5995901434497926,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003749252591749887,
      "loss": 5.4766,
      "mean_token_accuracy": 0.19809864163398744,
      "num_tokens": 70456343.0,
      "step": 36010
    },
    {
      "entropy": 5.888872909545898,
      "epoch": 3.6000899685110213,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003748932815127133,
      "loss": 5.4165,
      "mean_token_accuracy": 0.20706526637077333,
      "num_tokens": 70465521.0,
      "step": 36015
    },
    {
      "entropy": 5.997795581817627,
      "epoch": 3.60058979357225,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037486130133714874,
      "loss": 5.5023,
      "mean_token_accuracy": 0.19184548705816268,
      "num_tokens": 70476174.0,
      "step": 36020
    },
    {
      "entropy": 5.995801067352295,
      "epoch": 3.601089618633478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003748293186490995,
      "loss": 5.4864,
      "mean_token_accuracy": 0.19700284600257872,
      "num_tokens": 70485121.0,
      "step": 36025
    },
    {
      "entropy": 5.945806884765625,
      "epoch": 3.601589443694707,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003747973334493703,
      "loss": 5.495,
      "mean_token_accuracy": 0.1949640393257141,
      "num_tokens": 70494714.0,
      "step": 36030
    },
    {
      "entropy": 5.8331300735473635,
      "epoch": 3.602089268755935,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0003747653457387659,
      "loss": 5.2449,
      "mean_token_accuracy": 0.21995995193719864,
      "num_tokens": 70504371.0,
      "step": 36035
    },
    {
      "entropy": 5.989732456207276,
      "epoch": 3.602589093817164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037473335551809107,
      "loss": 5.4909,
      "mean_token_accuracy": 0.19195983409881592,
      "num_tokens": 70514634.0,
      "step": 36040
    },
    {
      "entropy": 5.903343677520752,
      "epoch": 3.6030889188783926,
      "grad_norm": 1.453125,
      "learning_rate": 0.00037470136278815067,
      "loss": 5.404,
      "mean_token_accuracy": 0.20476682633161544,
      "num_tokens": 70524407.0,
      "step": 36045
    },
    {
      "entropy": 5.948181009292602,
      "epoch": 3.6035887439396213,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00037466936754974964,
      "loss": 5.3891,
      "mean_token_accuracy": 0.1996911108493805,
      "num_tokens": 70534180.0,
      "step": 36050
    },
    {
      "entropy": 5.960618686676026,
      "epoch": 3.60408856900085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037463736980369297,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19903535544872283,
      "num_tokens": 70544183.0,
      "step": 36055
    },
    {
      "entropy": 5.979894685745239,
      "epoch": 3.604588394062078,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003746053695507856,
      "loss": 5.497,
      "mean_token_accuracy": 0.19093688130378722,
      "num_tokens": 70554427.0,
      "step": 36060
    },
    {
      "entropy": 6.0153553009033205,
      "epoch": 3.605088219123307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003745733667918329,
      "loss": 5.5963,
      "mean_token_accuracy": 0.18768952488899232,
      "num_tokens": 70564969.0,
      "step": 36065
    },
    {
      "entropy": 5.972879219055176,
      "epoch": 3.605588044184535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003745413615276399,
      "loss": 5.4515,
      "mean_token_accuracy": 0.19581044763326644,
      "num_tokens": 70574141.0,
      "step": 36070
    },
    {
      "entropy": 5.9482110977172855,
      "epoch": 3.606087869245764,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037450935375901186,
      "loss": 5.4491,
      "mean_token_accuracy": 0.1898746222257614,
      "num_tokens": 70584009.0,
      "step": 36075
    },
    {
      "entropy": 5.901648616790771,
      "epoch": 3.6065876943069926,
      "grad_norm": 1.109375,
      "learning_rate": 0.000374477343486754,
      "loss": 5.4071,
      "mean_token_accuracy": 0.20079761296510695,
      "num_tokens": 70595276.0,
      "step": 36080
    },
    {
      "entropy": 5.989128971099854,
      "epoch": 3.6070875193682213,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003744453307116719,
      "loss": 5.5383,
      "mean_token_accuracy": 0.18799040019512175,
      "num_tokens": 70604901.0,
      "step": 36085
    },
    {
      "entropy": 5.859037590026856,
      "epoch": 3.60758734442945,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037441331543457075,
      "loss": 5.4505,
      "mean_token_accuracy": 0.20043877959251405,
      "num_tokens": 70614832.0,
      "step": 36090
    },
    {
      "entropy": 6.012501335144043,
      "epoch": 3.608087169490678,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00037438129765625627,
      "loss": 5.466,
      "mean_token_accuracy": 0.18982666730880737,
      "num_tokens": 70623674.0,
      "step": 36095
    },
    {
      "entropy": 6.090039396286011,
      "epoch": 3.608586994551907,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037434927737753387,
      "loss": 5.5873,
      "mean_token_accuracy": 0.18435032218694686,
      "num_tokens": 70633197.0,
      "step": 36100
    },
    {
      "entropy": 6.004482555389404,
      "epoch": 3.609086819613135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037431725459920927,
      "loss": 5.5624,
      "mean_token_accuracy": 0.19108985215425492,
      "num_tokens": 70644053.0,
      "step": 36105
    },
    {
      "entropy": 6.0053136348724365,
      "epoch": 3.609586644674364,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00037428522932208804,
      "loss": 5.5502,
      "mean_token_accuracy": 0.1900079593062401,
      "num_tokens": 70656229.0,
      "step": 36110
    },
    {
      "entropy": 5.952552843093872,
      "epoch": 3.6100864697355926,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037425320154697595,
      "loss": 5.436,
      "mean_token_accuracy": 0.2009458377957344,
      "num_tokens": 70666262.0,
      "step": 36115
    },
    {
      "entropy": 6.040557813644409,
      "epoch": 3.6105862947968212,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037422117127467884,
      "loss": 5.5255,
      "mean_token_accuracy": 0.18234846740961075,
      "num_tokens": 70675941.0,
      "step": 36120
    },
    {
      "entropy": 5.874759912490845,
      "epoch": 3.61108611985805,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003741891385060027,
      "loss": 5.3677,
      "mean_token_accuracy": 0.21222431659698487,
      "num_tokens": 70685174.0,
      "step": 36125
    },
    {
      "entropy": 5.887393188476563,
      "epoch": 3.611585944919278,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003741571032417533,
      "loss": 5.4676,
      "mean_token_accuracy": 0.19577489346265792,
      "num_tokens": 70694090.0,
      "step": 36130
    },
    {
      "entropy": 5.908625221252441,
      "epoch": 3.612085769980507,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003741250654827367,
      "loss": 5.3775,
      "mean_token_accuracy": 0.2053065523505211,
      "num_tokens": 70703230.0,
      "step": 36135
    },
    {
      "entropy": 6.047343826293945,
      "epoch": 3.612585595041735,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037409302522975897,
      "loss": 5.4431,
      "mean_token_accuracy": 0.20067551881074905,
      "num_tokens": 70715029.0,
      "step": 36140
    },
    {
      "entropy": 5.990359544754028,
      "epoch": 3.613085420102964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037406098248362625,
      "loss": 5.5261,
      "mean_token_accuracy": 0.19571435898542405,
      "num_tokens": 70724515.0,
      "step": 36145
    },
    {
      "entropy": 5.849160146713257,
      "epoch": 3.6135852451641925,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037402893724514466,
      "loss": 5.3985,
      "mean_token_accuracy": 0.19677513539791108,
      "num_tokens": 70734129.0,
      "step": 36150
    },
    {
      "entropy": 5.896769666671753,
      "epoch": 3.6140850702254212,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037399688951512044,
      "loss": 5.3923,
      "mean_token_accuracy": 0.19724063724279403,
      "num_tokens": 70743525.0,
      "step": 36155
    },
    {
      "entropy": 5.954914665222168,
      "epoch": 3.61458489528665,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003739648392943601,
      "loss": 5.4634,
      "mean_token_accuracy": 0.19062265157699584,
      "num_tokens": 70753219.0,
      "step": 36160
    },
    {
      "entropy": 5.9294873714447025,
      "epoch": 3.615084720347878,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003739327865836697,
      "loss": 5.3702,
      "mean_token_accuracy": 0.2047821745276451,
      "num_tokens": 70761876.0,
      "step": 36165
    },
    {
      "entropy": 5.937766599655151,
      "epoch": 3.615584545409107,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037390073138385597,
      "loss": 5.45,
      "mean_token_accuracy": 0.20404200553894042,
      "num_tokens": 70771710.0,
      "step": 36170
    },
    {
      "entropy": 5.985383987426758,
      "epoch": 3.616084370470335,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003738686736957253,
      "loss": 5.547,
      "mean_token_accuracy": 0.1859624683856964,
      "num_tokens": 70781670.0,
      "step": 36175
    },
    {
      "entropy": 5.974583625793457,
      "epoch": 3.616584195531564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037383661352008413,
      "loss": 5.4142,
      "mean_token_accuracy": 0.19564757943153382,
      "num_tokens": 70790376.0,
      "step": 36180
    },
    {
      "entropy": 5.958635854721069,
      "epoch": 3.6170840205927925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037380455085773924,
      "loss": 5.5462,
      "mean_token_accuracy": 0.19024048894643783,
      "num_tokens": 70800659.0,
      "step": 36185
    },
    {
      "entropy": 5.989128971099854,
      "epoch": 3.6175838456540212,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037377248570949733,
      "loss": 5.4013,
      "mean_token_accuracy": 0.19720576852560043,
      "num_tokens": 70810614.0,
      "step": 36190
    },
    {
      "entropy": 5.93278546333313,
      "epoch": 3.6180836707152495,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037374041807616506,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19969121366739273,
      "num_tokens": 70820097.0,
      "step": 36195
    },
    {
      "entropy": 5.919229793548584,
      "epoch": 3.618583495776478,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037370834795854923,
      "loss": 5.4649,
      "mean_token_accuracy": 0.19499887377023697,
      "num_tokens": 70829864.0,
      "step": 36200
    },
    {
      "entropy": 5.920054912567139,
      "epoch": 3.619083320837707,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003736762753574568,
      "loss": 5.3636,
      "mean_token_accuracy": 0.20686652809381484,
      "num_tokens": 70838999.0,
      "step": 36205
    },
    {
      "entropy": 5.873578119277954,
      "epoch": 3.619583145898935,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037364420027369467,
      "loss": 5.3815,
      "mean_token_accuracy": 0.21170964390039443,
      "num_tokens": 70848473.0,
      "step": 36210
    },
    {
      "entropy": 5.961076354980468,
      "epoch": 3.620082970960164,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003736121227080699,
      "loss": 5.4051,
      "mean_token_accuracy": 0.20046701431274414,
      "num_tokens": 70859067.0,
      "step": 36215
    },
    {
      "entropy": 5.910855436325074,
      "epoch": 3.6205827960213925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037358004266138943,
      "loss": 5.4409,
      "mean_token_accuracy": 0.1965669184923172,
      "num_tokens": 70868028.0,
      "step": 36220
    },
    {
      "entropy": 5.916614627838134,
      "epoch": 3.6210826210826212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00037354796013446044,
      "loss": 5.4086,
      "mean_token_accuracy": 0.19533161967992782,
      "num_tokens": 70878326.0,
      "step": 36225
    },
    {
      "entropy": 5.984881591796875,
      "epoch": 3.6215824461438495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003735158751280902,
      "loss": 5.4477,
      "mean_token_accuracy": 0.19581355303525924,
      "num_tokens": 70887634.0,
      "step": 36230
    },
    {
      "entropy": 5.958936643600464,
      "epoch": 3.622082271205078,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003734837876430859,
      "loss": 5.4408,
      "mean_token_accuracy": 0.20095958113670348,
      "num_tokens": 70898154.0,
      "step": 36235
    },
    {
      "entropy": 6.020298099517822,
      "epoch": 3.622582096266307,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003734516976802548,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18749165087938308,
      "num_tokens": 70907247.0,
      "step": 36240
    },
    {
      "entropy": 5.994506120681763,
      "epoch": 3.623081921327535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003734196052404043,
      "loss": 5.4639,
      "mean_token_accuracy": 0.18994495570659636,
      "num_tokens": 70916665.0,
      "step": 36245
    },
    {
      "entropy": 6.0349424362182615,
      "epoch": 3.623581746388764,
      "grad_norm": 1.125,
      "learning_rate": 0.0003733875103243419,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1850900024175644,
      "num_tokens": 70928166.0,
      "step": 36250
    },
    {
      "entropy": 5.962709045410156,
      "epoch": 3.6240815714499925,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003733554129328751,
      "loss": 5.4764,
      "mean_token_accuracy": 0.20025517642498017,
      "num_tokens": 70936809.0,
      "step": 36255
    },
    {
      "entropy": 6.015788507461548,
      "epoch": 3.6245813965112212,
      "grad_norm": 1.203125,
      "learning_rate": 0.00037332331306681136,
      "loss": 5.5751,
      "mean_token_accuracy": 0.19023239314556123,
      "num_tokens": 70947131.0,
      "step": 36260
    },
    {
      "entropy": 5.960133457183838,
      "epoch": 3.6250812215724495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00037329121072695833,
      "loss": 5.4417,
      "mean_token_accuracy": 0.19798551201820375,
      "num_tokens": 70956728.0,
      "step": 36265
    },
    {
      "entropy": 5.917379570007324,
      "epoch": 3.625581046633678,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037325910591412374,
      "loss": 5.4006,
      "mean_token_accuracy": 0.20161187797784805,
      "num_tokens": 70966422.0,
      "step": 36270
    },
    {
      "entropy": 5.978040742874145,
      "epoch": 3.626080871694907,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003732269986291154,
      "loss": 5.4234,
      "mean_token_accuracy": 0.20244252532720566,
      "num_tokens": 70976063.0,
      "step": 36275
    },
    {
      "entropy": 5.976516437530518,
      "epoch": 3.626580696756135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037319488887274096,
      "loss": 5.5128,
      "mean_token_accuracy": 0.19662631750106813,
      "num_tokens": 70986772.0,
      "step": 36280
    },
    {
      "entropy": 5.9741926193237305,
      "epoch": 3.627080521817364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003731627766458084,
      "loss": 5.4021,
      "mean_token_accuracy": 0.20041827112436295,
      "num_tokens": 70996458.0,
      "step": 36285
    },
    {
      "entropy": 6.055977535247803,
      "epoch": 3.6275803468785925,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003731306619491257,
      "loss": 5.5884,
      "mean_token_accuracy": 0.19383085519075394,
      "num_tokens": 71006612.0,
      "step": 36290
    },
    {
      "entropy": 5.894438600540161,
      "epoch": 3.6280801719398212,
      "grad_norm": 1.125,
      "learning_rate": 0.0003730985447835008,
      "loss": 5.3365,
      "mean_token_accuracy": 0.20561756193637848,
      "num_tokens": 71016058.0,
      "step": 36295
    },
    {
      "entropy": 6.016855382919312,
      "epoch": 3.6285799970010495,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003730664251497417,
      "loss": 5.5227,
      "mean_token_accuracy": 0.1958644926548004,
      "num_tokens": 71025267.0,
      "step": 36300
    },
    {
      "entropy": 6.000559997558594,
      "epoch": 3.629079822062278,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003730343030486566,
      "loss": 5.4784,
      "mean_token_accuracy": 0.20030955523252486,
      "num_tokens": 71033972.0,
      "step": 36305
    },
    {
      "entropy": 6.083905172348023,
      "epoch": 3.629579647123507,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003730021784810536,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1888029083609581,
      "num_tokens": 71045014.0,
      "step": 36310
    },
    {
      "entropy": 5.952948045730591,
      "epoch": 3.630079472184735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003729700514477411,
      "loss": 5.3853,
      "mean_token_accuracy": 0.19375546872615815,
      "num_tokens": 71053559.0,
      "step": 36315
    },
    {
      "entropy": 5.974283742904663,
      "epoch": 3.630579297245964,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003729379219495273,
      "loss": 5.4911,
      "mean_token_accuracy": 0.19448275566101075,
      "num_tokens": 71063495.0,
      "step": 36320
    },
    {
      "entropy": 5.932019996643066,
      "epoch": 3.6310791223071925,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003729057899872206,
      "loss": 5.446,
      "mean_token_accuracy": 0.19461567103862762,
      "num_tokens": 71072998.0,
      "step": 36325
    },
    {
      "entropy": 5.976826095581055,
      "epoch": 3.6315789473684212,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003728736555616293,
      "loss": 5.3842,
      "mean_token_accuracy": 0.20064849108457566,
      "num_tokens": 71083517.0,
      "step": 36330
    },
    {
      "entropy": 6.030563831329346,
      "epoch": 3.6320787724296495,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037284151867356217,
      "loss": 5.5311,
      "mean_token_accuracy": 0.19307831227779387,
      "num_tokens": 71094789.0,
      "step": 36335
    },
    {
      "entropy": 5.96163420677185,
      "epoch": 3.632578597490878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037280937932382754,
      "loss": 5.4939,
      "mean_token_accuracy": 0.19653484374284744,
      "num_tokens": 71103933.0,
      "step": 36340
    },
    {
      "entropy": 5.9830797672271725,
      "epoch": 3.633078422552107,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000372777237513234,
      "loss": 5.3986,
      "mean_token_accuracy": 0.19685533493757248,
      "num_tokens": 71112650.0,
      "step": 36345
    },
    {
      "entropy": 5.975281524658203,
      "epoch": 3.633578247613335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037274509324259047,
      "loss": 5.5151,
      "mean_token_accuracy": 0.19902223646640776,
      "num_tokens": 71122415.0,
      "step": 36350
    },
    {
      "entropy": 5.952491760253906,
      "epoch": 3.634078072674564,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003727129465127055,
      "loss": 5.4082,
      "mean_token_accuracy": 0.2033350557088852,
      "num_tokens": 71133092.0,
      "step": 36355
    },
    {
      "entropy": 5.910762214660645,
      "epoch": 3.6345778977357925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003726807973243879,
      "loss": 5.4072,
      "mean_token_accuracy": 0.20832220762968062,
      "num_tokens": 71143924.0,
      "step": 36360
    },
    {
      "entropy": 5.845979976654053,
      "epoch": 3.6350777227970212,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037264864567844664,
      "loss": 5.3842,
      "mean_token_accuracy": 0.20438672006130218,
      "num_tokens": 71154109.0,
      "step": 36365
    },
    {
      "entropy": 5.87294921875,
      "epoch": 3.6355775478582495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003726164915756905,
      "loss": 5.3071,
      "mean_token_accuracy": 0.21877274811267852,
      "num_tokens": 71163621.0,
      "step": 36370
    },
    {
      "entropy": 6.047135496139527,
      "epoch": 3.636077372919478,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00037258433501692856,
      "loss": 5.4608,
      "mean_token_accuracy": 0.19270727634429932,
      "num_tokens": 71173525.0,
      "step": 36375
    },
    {
      "entropy": 5.944543647766113,
      "epoch": 3.636577197980707,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003725521760029699,
      "loss": 5.3324,
      "mean_token_accuracy": 0.20532321482896804,
      "num_tokens": 71182224.0,
      "step": 36380
    },
    {
      "entropy": 5.791663885116577,
      "epoch": 3.637077023041935,
      "grad_norm": 1.25,
      "learning_rate": 0.00037252001453462366,
      "loss": 5.3623,
      "mean_token_accuracy": 0.19810970574617387,
      "num_tokens": 71191104.0,
      "step": 36385
    },
    {
      "entropy": 5.974976348876953,
      "epoch": 3.637576848103164,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00037248785061269883,
      "loss": 5.5238,
      "mean_token_accuracy": 0.19714518934488295,
      "num_tokens": 71201193.0,
      "step": 36390
    },
    {
      "entropy": 6.071406316757202,
      "epoch": 3.6380766731643925,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003724556842380048,
      "loss": 5.4799,
      "mean_token_accuracy": 0.1910958468914032,
      "num_tokens": 71210679.0,
      "step": 36395
    },
    {
      "entropy": 6.0182366371154785,
      "epoch": 3.6385764982256212,
      "grad_norm": 1.125,
      "learning_rate": 0.00037242351541135084,
      "loss": 5.5475,
      "mean_token_accuracy": 0.19346291720867156,
      "num_tokens": 71221615.0,
      "step": 36400
    },
    {
      "entropy": 5.9201642036437985,
      "epoch": 3.6390763232868495,
      "grad_norm": 1.078125,
      "learning_rate": 0.00037239134413354633,
      "loss": 5.4927,
      "mean_token_accuracy": 0.18974112570285798,
      "num_tokens": 71231728.0,
      "step": 36405
    },
    {
      "entropy": 5.993667650222778,
      "epoch": 3.639576148348078,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003723591704054006,
      "loss": 5.5105,
      "mean_token_accuracy": 0.20080572217702866,
      "num_tokens": 71241754.0,
      "step": 36410
    },
    {
      "entropy": 5.995787143707275,
      "epoch": 3.640075973409307,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037232699422772323,
      "loss": 5.4591,
      "mean_token_accuracy": 0.18795789480209352,
      "num_tokens": 71251232.0,
      "step": 36415
    },
    {
      "entropy": 5.936248302459717,
      "epoch": 3.640575798470535,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003722948156013238,
      "loss": 5.3968,
      "mean_token_accuracy": 0.20272252708673477,
      "num_tokens": 71261158.0,
      "step": 36420
    },
    {
      "entropy": 5.9258746147155765,
      "epoch": 3.641075623531764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037226263452701177,
      "loss": 5.4444,
      "mean_token_accuracy": 0.1964477479457855,
      "num_tokens": 71270204.0,
      "step": 36425
    },
    {
      "entropy": 5.91667332649231,
      "epoch": 3.6415754485929925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003722304510055969,
      "loss": 5.3948,
      "mean_token_accuracy": 0.19781232327222825,
      "num_tokens": 71280305.0,
      "step": 36430
    },
    {
      "entropy": 5.956279468536377,
      "epoch": 3.6420752736542212,
      "grad_norm": 1.25,
      "learning_rate": 0.00037219826503788884,
      "loss": 5.4337,
      "mean_token_accuracy": 0.1962420105934143,
      "num_tokens": 71288724.0,
      "step": 36435
    },
    {
      "entropy": 5.861773586273193,
      "epoch": 3.6425750987154495,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003721660766246975,
      "loss": 5.4199,
      "mean_token_accuracy": 0.1975342348217964,
      "num_tokens": 71297662.0,
      "step": 36440
    },
    {
      "entropy": 5.964395475387573,
      "epoch": 3.643074923776678,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003721338857668327,
      "loss": 5.4569,
      "mean_token_accuracy": 0.1923978492617607,
      "num_tokens": 71307892.0,
      "step": 36445
    },
    {
      "entropy": 6.001063299179077,
      "epoch": 3.643574748837907,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003721016924651042,
      "loss": 5.4651,
      "mean_token_accuracy": 0.19503865838050843,
      "num_tokens": 71318380.0,
      "step": 36450
    },
    {
      "entropy": 5.944518136978149,
      "epoch": 3.644074573899135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003720694967203223,
      "loss": 5.3652,
      "mean_token_accuracy": 0.20296450555324555,
      "num_tokens": 71327172.0,
      "step": 36455
    },
    {
      "entropy": 5.879050064086914,
      "epoch": 3.644574398960364,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037203729853329675,
      "loss": 5.3943,
      "mean_token_accuracy": 0.2040441021323204,
      "num_tokens": 71337039.0,
      "step": 36460
    },
    {
      "entropy": 5.928999137878418,
      "epoch": 3.6450742240215925,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037200509790483773,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19306766539812087,
      "num_tokens": 71347305.0,
      "step": 36465
    },
    {
      "entropy": 5.984585237503052,
      "epoch": 3.6455740490828212,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003719728948357554,
      "loss": 5.545,
      "mean_token_accuracy": 0.1915456920862198,
      "num_tokens": 71356975.0,
      "step": 36470
    },
    {
      "entropy": 5.991099119186401,
      "epoch": 3.6460738741440495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037194068932686,
      "loss": 5.526,
      "mean_token_accuracy": 0.19063329994678496,
      "num_tokens": 71366489.0,
      "step": 36475
    },
    {
      "entropy": 5.848983097076416,
      "epoch": 3.646573699205278,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037190848137896176,
      "loss": 5.3955,
      "mean_token_accuracy": 0.2063763454556465,
      "num_tokens": 71376364.0,
      "step": 36480
    },
    {
      "entropy": 5.9238447666168215,
      "epoch": 3.647073524266507,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00037187627099287114,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19633592069149017,
      "num_tokens": 71385679.0,
      "step": 36485
    },
    {
      "entropy": 5.957970333099365,
      "epoch": 3.647573349327735,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00037184405816939847,
      "loss": 5.3606,
      "mean_token_accuracy": 0.20645343959331514,
      "num_tokens": 71395778.0,
      "step": 36490
    },
    {
      "entropy": 5.9433002948760985,
      "epoch": 3.648073174388964,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003718118429093542,
      "loss": 5.4484,
      "mean_token_accuracy": 0.20457788407802582,
      "num_tokens": 71407547.0,
      "step": 36495
    },
    {
      "entropy": 5.942843580245972,
      "epoch": 3.6485729994501925,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003717796252135488,
      "loss": 5.4172,
      "mean_token_accuracy": 0.2019705131649971,
      "num_tokens": 71417013.0,
      "step": 36500
    },
    {
      "entropy": 5.941914224624634,
      "epoch": 3.6490728245114212,
      "grad_norm": 1.21875,
      "learning_rate": 0.00037174740508279296,
      "loss": 5.3998,
      "mean_token_accuracy": 0.1983698785305023,
      "num_tokens": 71426768.0,
      "step": 36505
    },
    {
      "entropy": 5.868246173858642,
      "epoch": 3.6495726495726495,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003717151825178974,
      "loss": 5.3691,
      "mean_token_accuracy": 0.21176693141460418,
      "num_tokens": 71437099.0,
      "step": 36510
    },
    {
      "entropy": 5.950830125808716,
      "epoch": 3.650072474633878,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003716829575196726,
      "loss": 5.4144,
      "mean_token_accuracy": 0.19362177848815917,
      "num_tokens": 71447740.0,
      "step": 36515
    },
    {
      "entropy": 6.030160427093506,
      "epoch": 3.6505722996951064,
      "grad_norm": 1.125,
      "learning_rate": 0.0003716507300889296,
      "loss": 5.463,
      "mean_token_accuracy": 0.19885798394680024,
      "num_tokens": 71457994.0,
      "step": 36520
    },
    {
      "entropy": 5.950605726242065,
      "epoch": 3.651072124756335,
      "grad_norm": 1.0625,
      "learning_rate": 0.00037161850022647905,
      "loss": 5.4487,
      "mean_token_accuracy": 0.2003240928053856,
      "num_tokens": 71467800.0,
      "step": 36525
    },
    {
      "entropy": 5.934073734283447,
      "epoch": 3.651571949817564,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003715862679331318,
      "loss": 5.4427,
      "mean_token_accuracy": 0.19721879810094833,
      "num_tokens": 71477275.0,
      "step": 36530
    },
    {
      "entropy": 6.0711469650268555,
      "epoch": 3.6520717748787925,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00037155403320969906,
      "loss": 5.5394,
      "mean_token_accuracy": 0.1944073483347893,
      "num_tokens": 71487868.0,
      "step": 36535
    },
    {
      "entropy": 5.966585159301758,
      "epoch": 3.6525715999400212,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003715217960569915,
      "loss": 5.4724,
      "mean_token_accuracy": 0.19720102548599244,
      "num_tokens": 71497315.0,
      "step": 36540
    },
    {
      "entropy": 5.873092889785767,
      "epoch": 3.6530714250012495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00037148955647582063,
      "loss": 5.3806,
      "mean_token_accuracy": 0.20024988502264024,
      "num_tokens": 71507800.0,
      "step": 36545
    },
    {
      "entropy": 5.956368684768677,
      "epoch": 3.653571250062478,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037145731446699715,
      "loss": 5.5221,
      "mean_token_accuracy": 0.19272327721118926,
      "num_tokens": 71517996.0,
      "step": 36550
    },
    {
      "entropy": 5.986179399490356,
      "epoch": 3.6540710751237064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003714250700313325,
      "loss": 5.5231,
      "mean_token_accuracy": 0.19944812208414078,
      "num_tokens": 71528162.0,
      "step": 36555
    },
    {
      "entropy": 5.977230262756348,
      "epoch": 3.654570900184935,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037139282316963804,
      "loss": 5.4486,
      "mean_token_accuracy": 0.19551653563976287,
      "num_tokens": 71536641.0,
      "step": 36560
    },
    {
      "entropy": 5.879087734222412,
      "epoch": 3.655070725246164,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003713605738827248,
      "loss": 5.3531,
      "mean_token_accuracy": 0.19596351385116578,
      "num_tokens": 71547182.0,
      "step": 36565
    },
    {
      "entropy": 5.9057597637176515,
      "epoch": 3.6555705503073925,
      "grad_norm": 1.0,
      "learning_rate": 0.0003713283221714044,
      "loss": 5.3634,
      "mean_token_accuracy": 0.20339488387107849,
      "num_tokens": 71556220.0,
      "step": 36570
    },
    {
      "entropy": 5.960431528091431,
      "epoch": 3.6560703753686212,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00037129606803648814,
      "loss": 5.4403,
      "mean_token_accuracy": 0.1962999939918518,
      "num_tokens": 71565488.0,
      "step": 36575
    },
    {
      "entropy": 5.932176017761231,
      "epoch": 3.6565702004298495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037126381147878763,
      "loss": 5.4243,
      "mean_token_accuracy": 0.1982654795050621,
      "num_tokens": 71575396.0,
      "step": 36580
    },
    {
      "entropy": 5.9642328262329105,
      "epoch": 3.657070025491078,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037123155249911435,
      "loss": 5.4154,
      "mean_token_accuracy": 0.2013577252626419,
      "num_tokens": 71585380.0,
      "step": 36585
    },
    {
      "entropy": 5.900813865661621,
      "epoch": 3.6575698505523064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037119929109828,
      "loss": 5.3408,
      "mean_token_accuracy": 0.2005098655819893,
      "num_tokens": 71594936.0,
      "step": 36590
    },
    {
      "entropy": 5.90737738609314,
      "epoch": 3.658069675613535,
      "grad_norm": 1.125,
      "learning_rate": 0.0003711670272770962,
      "loss": 5.3055,
      "mean_token_accuracy": 0.21711107939481736,
      "num_tokens": 71604418.0,
      "step": 36595
    },
    {
      "entropy": 5.94374794960022,
      "epoch": 3.658569500674764,
      "grad_norm": 1.15625,
      "learning_rate": 0.00037113476103637483,
      "loss": 5.5175,
      "mean_token_accuracy": 0.18939049988985063,
      "num_tokens": 71614780.0,
      "step": 36600
    },
    {
      "entropy": 5.9479328155517575,
      "epoch": 3.6590693257359925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003711024923769275,
      "loss": 5.4444,
      "mean_token_accuracy": 0.19910719990730286,
      "num_tokens": 71623868.0,
      "step": 36605
    },
    {
      "entropy": 5.998028087615967,
      "epoch": 3.6595691507972212,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003710702212995662,
      "loss": 5.557,
      "mean_token_accuracy": 0.18608714640140533,
      "num_tokens": 71634481.0,
      "step": 36610
    },
    {
      "entropy": 5.9720543384552,
      "epoch": 3.6600689758584495,
      "grad_norm": 1.125,
      "learning_rate": 0.00037103794780510296,
      "loss": 5.4659,
      "mean_token_accuracy": 0.19357605427503585,
      "num_tokens": 71643180.0,
      "step": 36615
    },
    {
      "entropy": 5.937610101699829,
      "epoch": 3.660568800919678,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037100567189434966,
      "loss": 5.3699,
      "mean_token_accuracy": 0.20048279762268068,
      "num_tokens": 71652370.0,
      "step": 36620
    },
    {
      "entropy": 5.847789573669433,
      "epoch": 3.6610686259809064,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003709733935681183,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21013870388269423,
      "num_tokens": 71663047.0,
      "step": 36625
    },
    {
      "entropy": 5.955120706558228,
      "epoch": 3.661568451042135,
      "grad_norm": 1.265625,
      "learning_rate": 0.000370941112827221,
      "loss": 5.4592,
      "mean_token_accuracy": 0.1900647610425949,
      "num_tokens": 71672879.0,
      "step": 36630
    },
    {
      "entropy": 6.047144746780395,
      "epoch": 3.662068276103364,
      "grad_norm": 1.09375,
      "learning_rate": 0.00037090882967247003,
      "loss": 5.5599,
      "mean_token_accuracy": 0.18958876579999923,
      "num_tokens": 71682773.0,
      "step": 36635
    },
    {
      "entropy": 6.00274019241333,
      "epoch": 3.6625681011645925,
      "grad_norm": 1.125,
      "learning_rate": 0.0003708765441046776,
      "loss": 5.4657,
      "mean_token_accuracy": 0.20290157794952393,
      "num_tokens": 71692374.0,
      "step": 36640
    },
    {
      "entropy": 6.035192632675171,
      "epoch": 3.6630679262258212,
      "grad_norm": 1.125,
      "learning_rate": 0.00037084425612465605,
      "loss": 5.5914,
      "mean_token_accuracy": 0.18474362492561341,
      "num_tokens": 71701703.0,
      "step": 36645
    },
    {
      "entropy": 5.9298415184021,
      "epoch": 3.6635677512870495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003708119657332176,
      "loss": 5.4827,
      "mean_token_accuracy": 0.19800171703100206,
      "num_tokens": 71711783.0,
      "step": 36650
    },
    {
      "entropy": 5.949930715560913,
      "epoch": 3.664067576348278,
      "grad_norm": 1.140625,
      "learning_rate": 0.00037077967293117477,
      "loss": 5.4565,
      "mean_token_accuracy": 0.20787809938192367,
      "num_tokens": 71721374.0,
      "step": 36655
    },
    {
      "entropy": 6.039707517623901,
      "epoch": 3.6645674014095064,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037074737771933996,
      "loss": 5.4395,
      "mean_token_accuracy": 0.19965266734361647,
      "num_tokens": 71729742.0,
      "step": 36660
    },
    {
      "entropy": 5.982461166381836,
      "epoch": 3.665067226470735,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003707150800985259,
      "loss": 5.596,
      "mean_token_accuracy": 0.18389514535665513,
      "num_tokens": 71739759.0,
      "step": 36665
    },
    {
      "entropy": 5.918521070480347,
      "epoch": 3.665567051531964,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00037068278006954493,
      "loss": 5.4001,
      "mean_token_accuracy": 0.20141125470399857,
      "num_tokens": 71749882.0,
      "step": 36670
    },
    {
      "entropy": 5.993468284606934,
      "epoch": 3.6660668765931925,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00037065047763320993,
      "loss": 5.5073,
      "mean_token_accuracy": 0.18794069588184356,
      "num_tokens": 71760212.0,
      "step": 36675
    },
    {
      "entropy": 5.986481952667236,
      "epoch": 3.6665667016544212,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003706181727903334,
      "loss": 5.5182,
      "mean_token_accuracy": 0.19304834306240082,
      "num_tokens": 71770151.0,
      "step": 36680
    },
    {
      "entropy": 5.910734748840332,
      "epoch": 3.6670665267156495,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00037058586554172834,
      "loss": 5.4244,
      "mean_token_accuracy": 0.19926592707633972,
      "num_tokens": 71780435.0,
      "step": 36685
    },
    {
      "entropy": 5.966846752166748,
      "epoch": 3.667566351776878,
      "grad_norm": 1.046875,
      "learning_rate": 0.00037055355588820746,
      "loss": 5.5072,
      "mean_token_accuracy": 0.1958868160843849,
      "num_tokens": 71790782.0,
      "step": 36690
    },
    {
      "entropy": 5.987583875656128,
      "epoch": 3.6680661768381064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003705212438305837,
      "loss": 5.4668,
      "mean_token_accuracy": 0.19657485634088517,
      "num_tokens": 71800166.0,
      "step": 36695
    },
    {
      "entropy": 5.993458986282349,
      "epoch": 3.668566001899335,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003704889293696701,
      "loss": 5.5046,
      "mean_token_accuracy": 0.19168413281440735,
      "num_tokens": 71810344.0,
      "step": 36700
    },
    {
      "entropy": 5.945382165908813,
      "epoch": 3.669065826960564,
      "grad_norm": 1.171875,
      "learning_rate": 0.00037045661250627955,
      "loss": 5.486,
      "mean_token_accuracy": 0.19638395309448242,
      "num_tokens": 71819289.0,
      "step": 36705
    },
    {
      "entropy": 6.007130861282349,
      "epoch": 3.6695656520217925,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003704242932412252,
      "loss": 5.4977,
      "mean_token_accuracy": 0.1990058496594429,
      "num_tokens": 71829287.0,
      "step": 36710
    },
    {
      "entropy": 5.99685435295105,
      "epoch": 3.6700654770830208,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00037039197157532017,
      "loss": 5.4865,
      "mean_token_accuracy": 0.1917545258998871,
      "num_tokens": 71838598.0,
      "step": 36715
    },
    {
      "entropy": 5.959127950668335,
      "epoch": 3.6705653021442495,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003703596475093776,
      "loss": 5.4278,
      "mean_token_accuracy": 0.19585584998130798,
      "num_tokens": 71848655.0,
      "step": 36720
    },
    {
      "entropy": 5.933819341659546,
      "epoch": 3.671065127205478,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00037032732104421096,
      "loss": 5.4326,
      "mean_token_accuracy": 0.20475054979324342,
      "num_tokens": 71858328.0,
      "step": 36725
    },
    {
      "entropy": 5.84304027557373,
      "epoch": 3.6715649522667064,
      "grad_norm": 1.109375,
      "learning_rate": 0.00037029499218063346,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20550208687782287,
      "num_tokens": 71868119.0,
      "step": 36730
    },
    {
      "entropy": 5.877623128890991,
      "epoch": 3.672064777327935,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00037026266091945843,
      "loss": 5.349,
      "mean_token_accuracy": 0.21069214195013047,
      "num_tokens": 71877486.0,
      "step": 36735
    },
    {
      "entropy": 5.90480751991272,
      "epoch": 3.672564602389164,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003702303272614994,
      "loss": 5.3775,
      "mean_token_accuracy": 0.20703172385692598,
      "num_tokens": 71888078.0,
      "step": 36740
    },
    {
      "entropy": 5.99316177368164,
      "epoch": 3.6730644274503925,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003701979912075698,
      "loss": 5.462,
      "mean_token_accuracy": 0.19910405874252318,
      "num_tokens": 71897217.0,
      "step": 36745
    },
    {
      "entropy": 5.9143757820129395,
      "epoch": 3.6735642525116208,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003701656527584832,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20244749933481215,
      "num_tokens": 71905619.0,
      "step": 36750
    },
    {
      "entropy": 5.923930883407593,
      "epoch": 3.6740640775728495,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003701333119150534,
      "loss": 5.5248,
      "mean_token_accuracy": 0.18157870471477508,
      "num_tokens": 71917197.0,
      "step": 36755
    },
    {
      "entropy": 5.938189744949341,
      "epoch": 3.674563902634078,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00037010096867809384,
      "loss": 5.3642,
      "mean_token_accuracy": 0.21006759256124496,
      "num_tokens": 71928601.0,
      "step": 36760
    },
    {
      "entropy": 5.958063220977783,
      "epoch": 3.6750637276953064,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00037006862304841845,
      "loss": 5.4146,
      "mean_token_accuracy": 0.20463523268699646,
      "num_tokens": 71938490.0,
      "step": 36765
    },
    {
      "entropy": 5.925861930847168,
      "epoch": 3.675563552756535,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003700362750268409,
      "loss": 5.3943,
      "mean_token_accuracy": 0.19981783330440522,
      "num_tokens": 71947998.0,
      "step": 36770
    },
    {
      "entropy": 5.961448717117309,
      "epoch": 3.676063377817764,
      "grad_norm": 1.125,
      "learning_rate": 0.0003700039246141752,
      "loss": 5.5127,
      "mean_token_accuracy": 0.1970476523041725,
      "num_tokens": 71958839.0,
      "step": 36775
    },
    {
      "entropy": 5.8864305973052975,
      "epoch": 3.6765632028789925,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003699715718112351,
      "loss": 5.3879,
      "mean_token_accuracy": 0.20020072609186174,
      "num_tokens": 71968296.0,
      "step": 36780
    },
    {
      "entropy": 6.069977188110352,
      "epoch": 3.6770630279402208,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036993921661883475,
      "loss": 5.513,
      "mean_token_accuracy": 0.190928915143013,
      "num_tokens": 71978246.0,
      "step": 36785
    },
    {
      "entropy": 6.049114513397217,
      "epoch": 3.6775628530014495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036990685903778815,
      "loss": 5.5479,
      "mean_token_accuracy": 0.19332844018936157,
      "num_tokens": 71988101.0,
      "step": 36790
    },
    {
      "entropy": 5.979723167419434,
      "epoch": 3.678062678062678,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036987449906890937,
      "loss": 5.4073,
      "mean_token_accuracy": 0.20276427417993545,
      "num_tokens": 71998000.0,
      "step": 36795
    },
    {
      "entropy": 5.869611024856567,
      "epoch": 3.6785625031239064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036984213671301266,
      "loss": 5.3046,
      "mean_token_accuracy": 0.20548151433467865,
      "num_tokens": 72007438.0,
      "step": 36800
    },
    {
      "entropy": 5.974175024032593,
      "epoch": 3.679062328185135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036980977197091213,
      "loss": 5.5984,
      "mean_token_accuracy": 0.1903516724705696,
      "num_tokens": 72017995.0,
      "step": 36805
    },
    {
      "entropy": 5.9129414558410645,
      "epoch": 3.679562153246364,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036977740484342214,
      "loss": 5.4518,
      "mean_token_accuracy": 0.1966363161802292,
      "num_tokens": 72027289.0,
      "step": 36810
    },
    {
      "entropy": 5.874715137481689,
      "epoch": 3.6800619783075925,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003697450353313571,
      "loss": 5.323,
      "mean_token_accuracy": 0.21162489503622056,
      "num_tokens": 72036721.0,
      "step": 36815
    },
    {
      "entropy": 5.96316876411438,
      "epoch": 3.6805618033688208,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003697126634355313,
      "loss": 5.4459,
      "mean_token_accuracy": 0.19532612562179566,
      "num_tokens": 72046760.0,
      "step": 36820
    },
    {
      "entropy": 5.998377513885498,
      "epoch": 3.6810616284300495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036968028915675924,
      "loss": 5.5028,
      "mean_token_accuracy": 0.18964026123285294,
      "num_tokens": 72056285.0,
      "step": 36825
    },
    {
      "entropy": 6.005773401260376,
      "epoch": 3.681561453491278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003696479124958555,
      "loss": 5.5233,
      "mean_token_accuracy": 0.18629183024168014,
      "num_tokens": 72067082.0,
      "step": 36830
    },
    {
      "entropy": 5.9660727977752686,
      "epoch": 3.6820612785525064,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003696155334536346,
      "loss": 5.3975,
      "mean_token_accuracy": 0.19752222597599028,
      "num_tokens": 72077203.0,
      "step": 36835
    },
    {
      "entropy": 5.994343948364258,
      "epoch": 3.682561103613735,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003695831520309113,
      "loss": 5.6129,
      "mean_token_accuracy": 0.1863567888736725,
      "num_tokens": 72086669.0,
      "step": 36840
    },
    {
      "entropy": 5.958217573165894,
      "epoch": 3.683060928674964,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036955076822850014,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19322068691253663,
      "num_tokens": 72097752.0,
      "step": 36845
    },
    {
      "entropy": 6.01839690208435,
      "epoch": 3.6835607537361925,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036951838204721607,
      "loss": 5.4519,
      "mean_token_accuracy": 0.19745438992977143,
      "num_tokens": 72107446.0,
      "step": 36850
    },
    {
      "entropy": 5.960170555114746,
      "epoch": 3.6840605787974208,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036948599348787375,
      "loss": 5.4107,
      "mean_token_accuracy": 0.2020252302289009,
      "num_tokens": 72117099.0,
      "step": 36855
    },
    {
      "entropy": 5.92310848236084,
      "epoch": 3.6845604038586495,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036945360255128816,
      "loss": 5.4328,
      "mean_token_accuracy": 0.1919179618358612,
      "num_tokens": 72126472.0,
      "step": 36860
    },
    {
      "entropy": 5.985779047012329,
      "epoch": 3.685060228919878,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036942120923827426,
      "loss": 5.5083,
      "mean_token_accuracy": 0.19615776985883712,
      "num_tokens": 72135910.0,
      "step": 36865
    },
    {
      "entropy": 5.956652498245239,
      "epoch": 3.6855600539811064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036938881354964704,
      "loss": 5.4581,
      "mean_token_accuracy": 0.19911131560802459,
      "num_tokens": 72145173.0,
      "step": 36870
    },
    {
      "entropy": 5.912838935852051,
      "epoch": 3.686059879042335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036935641548622154,
      "loss": 5.4357,
      "mean_token_accuracy": 0.19522885382175445,
      "num_tokens": 72154930.0,
      "step": 36875
    },
    {
      "entropy": 5.872433614730835,
      "epoch": 3.686559704103564,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003693240150488129,
      "loss": 5.2991,
      "mean_token_accuracy": 0.20632757544517516,
      "num_tokens": 72165221.0,
      "step": 36880
    },
    {
      "entropy": 5.926385688781738,
      "epoch": 3.6870595291647925,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003692916122382364,
      "loss": 5.4126,
      "mean_token_accuracy": 0.1943084716796875,
      "num_tokens": 72175532.0,
      "step": 36885
    },
    {
      "entropy": 5.994768953323364,
      "epoch": 3.6875593542260208,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036925920705530713,
      "loss": 5.51,
      "mean_token_accuracy": 0.1878288358449936,
      "num_tokens": 72185588.0,
      "step": 36890
    },
    {
      "entropy": 5.922608470916748,
      "epoch": 3.6880591792872495,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036922679950084043,
      "loss": 5.4624,
      "mean_token_accuracy": 0.19878366440534592,
      "num_tokens": 72196051.0,
      "step": 36895
    },
    {
      "entropy": 6.004730129241944,
      "epoch": 3.688559004348478,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003691943895756518,
      "loss": 5.5302,
      "mean_token_accuracy": 0.19594482183456421,
      "num_tokens": 72207530.0,
      "step": 36900
    },
    {
      "entropy": 5.990080165863037,
      "epoch": 3.6890588294097064,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036916197728055646,
      "loss": 5.4731,
      "mean_token_accuracy": 0.19171450436115264,
      "num_tokens": 72217764.0,
      "step": 36905
    },
    {
      "entropy": 5.911848831176758,
      "epoch": 3.689558654470935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036912956261637005,
      "loss": 5.3529,
      "mean_token_accuracy": 0.21006039828062056,
      "num_tokens": 72226971.0,
      "step": 36910
    },
    {
      "entropy": 6.0083380222320555,
      "epoch": 3.690058479532164,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003690971455839081,
      "loss": 5.5332,
      "mean_token_accuracy": 0.19394144117832185,
      "num_tokens": 72236045.0,
      "step": 36915
    },
    {
      "entropy": 6.034781789779663,
      "epoch": 3.6905583045933925,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036906472618398605,
      "loss": 5.546,
      "mean_token_accuracy": 0.18721033334732057,
      "num_tokens": 72246150.0,
      "step": 36920
    },
    {
      "entropy": 5.951325225830078,
      "epoch": 3.6910581296546208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036903230441741986,
      "loss": 5.4035,
      "mean_token_accuracy": 0.2014303207397461,
      "num_tokens": 72254916.0,
      "step": 36925
    },
    {
      "entropy": 5.973876285552978,
      "epoch": 3.6915579547158495,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036899988028502495,
      "loss": 5.5006,
      "mean_token_accuracy": 0.19489087760448456,
      "num_tokens": 72265104.0,
      "step": 36930
    },
    {
      "entropy": 5.993616342544556,
      "epoch": 3.692057779777078,
      "grad_norm": 1.234375,
      "learning_rate": 0.00036896745378761734,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1985207676887512,
      "num_tokens": 72274899.0,
      "step": 36935
    },
    {
      "entropy": 5.960777854919433,
      "epoch": 3.6925576048383064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036893502492601273,
      "loss": 5.4166,
      "mean_token_accuracy": 0.19464750736951827,
      "num_tokens": 72285618.0,
      "step": 36940
    },
    {
      "entropy": 5.989142560958863,
      "epoch": 3.693057429899535,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003689025937010271,
      "loss": 5.4194,
      "mean_token_accuracy": 0.20198968797922134,
      "num_tokens": 72294943.0,
      "step": 36945
    },
    {
      "entropy": 5.943863248825073,
      "epoch": 3.693557254960764,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036887016011347633,
      "loss": 5.3897,
      "mean_token_accuracy": 0.1991766571998596,
      "num_tokens": 72304908.0,
      "step": 36950
    },
    {
      "entropy": 5.905210638046265,
      "epoch": 3.6940570800219925,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003688377241641765,
      "loss": 5.4162,
      "mean_token_accuracy": 0.199274505674839,
      "num_tokens": 72314646.0,
      "step": 36955
    },
    {
      "entropy": 6.012728834152222,
      "epoch": 3.6945569050832208,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036880528585394364,
      "loss": 5.4837,
      "mean_token_accuracy": 0.20346966087818147,
      "num_tokens": 72324329.0,
      "step": 36960
    },
    {
      "entropy": 6.028608655929565,
      "epoch": 3.6950567301444495,
      "grad_norm": 1.28125,
      "learning_rate": 0.000368772845183594,
      "loss": 5.5128,
      "mean_token_accuracy": 0.19326688796281816,
      "num_tokens": 72333201.0,
      "step": 36965
    },
    {
      "entropy": 5.916664266586304,
      "epoch": 3.695556555205678,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036874040215394366,
      "loss": 5.3965,
      "mean_token_accuracy": 0.2067709743976593,
      "num_tokens": 72343922.0,
      "step": 36970
    },
    {
      "entropy": 6.050643491744995,
      "epoch": 3.6960563802669064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036870795676580885,
      "loss": 5.545,
      "mean_token_accuracy": 0.18567026704549788,
      "num_tokens": 72354081.0,
      "step": 36975
    },
    {
      "entropy": 5.95833911895752,
      "epoch": 3.696556205328135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003686755090200061,
      "loss": 5.4426,
      "mean_token_accuracy": 0.19264349788427354,
      "num_tokens": 72363646.0,
      "step": 36980
    },
    {
      "entropy": 5.948434972763062,
      "epoch": 3.697056030389364,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003686430589173516,
      "loss": 5.4849,
      "mean_token_accuracy": 0.18915074467658996,
      "num_tokens": 72373847.0,
      "step": 36985
    },
    {
      "entropy": 5.939994764328003,
      "epoch": 3.6975558554505925,
      "grad_norm": 1.125,
      "learning_rate": 0.00036861060645866184,
      "loss": 5.3075,
      "mean_token_accuracy": 0.20485715866088866,
      "num_tokens": 72382991.0,
      "step": 36990
    },
    {
      "entropy": 5.905100774765015,
      "epoch": 3.6980556805118208,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003685781516447533,
      "loss": 5.3953,
      "mean_token_accuracy": 0.19678889811038972,
      "num_tokens": 72392260.0,
      "step": 36995
    },
    {
      "entropy": 5.993224477767944,
      "epoch": 3.6985555055730495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003685456944764425,
      "loss": 5.5145,
      "mean_token_accuracy": 0.185841865837574,
      "num_tokens": 72401542.0,
      "step": 37000
    },
    {
      "entropy": 6.032862901687622,
      "epoch": 3.699055330634278,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036851323495454616,
      "loss": 5.5705,
      "mean_token_accuracy": 0.19228719770908356,
      "num_tokens": 72412690.0,
      "step": 37005
    },
    {
      "entropy": 5.955319547653199,
      "epoch": 3.6995551556955064,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036848077307988084,
      "loss": 5.4065,
      "mean_token_accuracy": 0.20464820861816407,
      "num_tokens": 72421925.0,
      "step": 37010
    },
    {
      "entropy": 5.912825679779052,
      "epoch": 3.700054980756735,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036844830885326334,
      "loss": 5.4151,
      "mean_token_accuracy": 0.1954771786928177,
      "num_tokens": 72432252.0,
      "step": 37015
    },
    {
      "entropy": 5.999054861068726,
      "epoch": 3.700554805817964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036841584227551043,
      "loss": 5.4842,
      "mean_token_accuracy": 0.19577295035123826,
      "num_tokens": 72441210.0,
      "step": 37020
    },
    {
      "entropy": 5.942434072494507,
      "epoch": 3.7010546308791925,
      "grad_norm": 1.109375,
      "learning_rate": 0.000368383373347439,
      "loss": 5.4378,
      "mean_token_accuracy": 0.19716989398002624,
      "num_tokens": 72450570.0,
      "step": 37025
    },
    {
      "entropy": 5.970706653594971,
      "epoch": 3.7015544559404208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036835090206986594,
      "loss": 5.4824,
      "mean_token_accuracy": 0.1927194267511368,
      "num_tokens": 72460597.0,
      "step": 37030
    },
    {
      "entropy": 5.982940769195556,
      "epoch": 3.7020542810016495,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036831842844360814,
      "loss": 5.4918,
      "mean_token_accuracy": 0.1903828963637352,
      "num_tokens": 72471232.0,
      "step": 37035
    },
    {
      "entropy": 5.932367658615112,
      "epoch": 3.702554106062878,
      "grad_norm": 1.125,
      "learning_rate": 0.0003682859524694827,
      "loss": 5.3838,
      "mean_token_accuracy": 0.1986141696572304,
      "num_tokens": 72480547.0,
      "step": 37040
    },
    {
      "entropy": 5.889956760406494,
      "epoch": 3.7030539311241064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036825347414830674,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20631335824728012,
      "num_tokens": 72490507.0,
      "step": 37045
    },
    {
      "entropy": 5.94412636756897,
      "epoch": 3.703553756185335,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00036822099348089737,
      "loss": 5.4609,
      "mean_token_accuracy": 0.19699058830738067,
      "num_tokens": 72499651.0,
      "step": 37050
    },
    {
      "entropy": 5.910927724838257,
      "epoch": 3.704053581246564,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036818851046807176,
      "loss": 5.4323,
      "mean_token_accuracy": 0.1986427903175354,
      "num_tokens": 72509580.0,
      "step": 37055
    },
    {
      "entropy": 5.936616563796997,
      "epoch": 3.7045534063077925,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036815602511064725,
      "loss": 5.3977,
      "mean_token_accuracy": 0.19489024877548217,
      "num_tokens": 72519184.0,
      "step": 37060
    },
    {
      "entropy": 5.922580146789551,
      "epoch": 3.7050532313690208,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036812353740944104,
      "loss": 5.4071,
      "mean_token_accuracy": 0.20063260495662688,
      "num_tokens": 72528112.0,
      "step": 37065
    },
    {
      "entropy": 5.997852945327759,
      "epoch": 3.7055530564302495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036809104736527064,
      "loss": 5.5236,
      "mean_token_accuracy": 0.1897483840584755,
      "num_tokens": 72537197.0,
      "step": 37070
    },
    {
      "entropy": 5.998776531219482,
      "epoch": 3.7060528814914777,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036805855497895346,
      "loss": 5.5415,
      "mean_token_accuracy": 0.1896999016404152,
      "num_tokens": 72547443.0,
      "step": 37075
    },
    {
      "entropy": 5.981916856765747,
      "epoch": 3.7065527065527064,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036802606025130696,
      "loss": 5.4282,
      "mean_token_accuracy": 0.19332086741924287,
      "num_tokens": 72556741.0,
      "step": 37080
    },
    {
      "entropy": 5.937097501754761,
      "epoch": 3.707052531613935,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003679935631831488,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19747294038534163,
      "num_tokens": 72567448.0,
      "step": 37085
    },
    {
      "entropy": 5.916181325912476,
      "epoch": 3.707552356675164,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036796106377529644,
      "loss": 5.4532,
      "mean_token_accuracy": 0.19380633533000946,
      "num_tokens": 72577433.0,
      "step": 37090
    },
    {
      "entropy": 5.989292097091675,
      "epoch": 3.7080521817363925,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003679285620285676,
      "loss": 5.4529,
      "mean_token_accuracy": 0.20088311433792114,
      "num_tokens": 72587754.0,
      "step": 37095
    },
    {
      "entropy": 5.970761346817016,
      "epoch": 3.7085520067976208,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00036789605794378013,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20024549216032028,
      "num_tokens": 72596333.0,
      "step": 37100
    },
    {
      "entropy": 5.993132305145264,
      "epoch": 3.7090518318588495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036786355152175173,
      "loss": 5.5088,
      "mean_token_accuracy": 0.19350391030311584,
      "num_tokens": 72605813.0,
      "step": 37105
    },
    {
      "entropy": 5.912662696838379,
      "epoch": 3.7095516569200777,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003678310427633003,
      "loss": 5.4789,
      "mean_token_accuracy": 0.19863303005695343,
      "num_tokens": 72615435.0,
      "step": 37110
    },
    {
      "entropy": 6.002118158340454,
      "epoch": 3.7100514819813064,
      "grad_norm": 1.125,
      "learning_rate": 0.00036779853166924353,
      "loss": 5.4833,
      "mean_token_accuracy": 0.1974848061800003,
      "num_tokens": 72625276.0,
      "step": 37115
    },
    {
      "entropy": 5.92646279335022,
      "epoch": 3.710551307042535,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003677660182403998,
      "loss": 5.3254,
      "mean_token_accuracy": 0.21151579171419144,
      "num_tokens": 72634225.0,
      "step": 37120
    },
    {
      "entropy": 5.907098293304443,
      "epoch": 3.711051132103764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036773350247758685,
      "loss": 5.3953,
      "mean_token_accuracy": 0.203899484872818,
      "num_tokens": 72644133.0,
      "step": 37125
    },
    {
      "entropy": 5.975671291351318,
      "epoch": 3.7115509571649925,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003677009843816228,
      "loss": 5.4429,
      "mean_token_accuracy": 0.1979668602347374,
      "num_tokens": 72653758.0,
      "step": 37130
    },
    {
      "entropy": 5.966531085968017,
      "epoch": 3.7120507822262208,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003676684639533259,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19321514219045638,
      "num_tokens": 72664049.0,
      "step": 37135
    },
    {
      "entropy": 6.026922130584717,
      "epoch": 3.7125506072874495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003676359411935143,
      "loss": 5.5188,
      "mean_token_accuracy": 0.19443004429340363,
      "num_tokens": 72673419.0,
      "step": 37140
    },
    {
      "entropy": 5.930480670928955,
      "epoch": 3.7130504323486777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003676034161030062,
      "loss": 5.383,
      "mean_token_accuracy": 0.1975005805492401,
      "num_tokens": 72682392.0,
      "step": 37145
    },
    {
      "entropy": 5.953829431533814,
      "epoch": 3.7135502574099064,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036757088868262,
      "loss": 5.5097,
      "mean_token_accuracy": 0.1953076407313347,
      "num_tokens": 72692092.0,
      "step": 37150
    },
    {
      "entropy": 5.993164587020874,
      "epoch": 3.714050082471135,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036753835893317413,
      "loss": 5.5164,
      "mean_token_accuracy": 0.19548808187246322,
      "num_tokens": 72700098.0,
      "step": 37155
    },
    {
      "entropy": 5.975512504577637,
      "epoch": 3.714549907532364,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003675058268554868,
      "loss": 5.4,
      "mean_token_accuracy": 0.1988615557551384,
      "num_tokens": 72709443.0,
      "step": 37160
    },
    {
      "entropy": 6.057206487655639,
      "epoch": 3.7150497325935925,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003674732924503768,
      "loss": 5.6251,
      "mean_token_accuracy": 0.1896086171269417,
      "num_tokens": 72718849.0,
      "step": 37165
    },
    {
      "entropy": 6.086708688735962,
      "epoch": 3.7155495576548208,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003674407557186625,
      "loss": 5.4957,
      "mean_token_accuracy": 0.18770601749420165,
      "num_tokens": 72728414.0,
      "step": 37170
    },
    {
      "entropy": 5.9181159973144535,
      "epoch": 3.7160493827160495,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036740821666116265,
      "loss": 5.3733,
      "mean_token_accuracy": 0.2121296226978302,
      "num_tokens": 72736432.0,
      "step": 37175
    },
    {
      "entropy": 5.731105136871338,
      "epoch": 3.7165492077772777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003673756752786958,
      "loss": 5.3003,
      "mean_token_accuracy": 0.20912740826606752,
      "num_tokens": 72745630.0,
      "step": 37180
    },
    {
      "entropy": 5.916123390197754,
      "epoch": 3.7170490328385064,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036734313157208066,
      "loss": 5.4795,
      "mean_token_accuracy": 0.19636629223823548,
      "num_tokens": 72755590.0,
      "step": 37185
    },
    {
      "entropy": 5.976729917526245,
      "epoch": 3.717548857899735,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003673105855421362,
      "loss": 5.3755,
      "mean_token_accuracy": 0.19857311397790908,
      "num_tokens": 72765187.0,
      "step": 37190
    },
    {
      "entropy": 5.932783603668213,
      "epoch": 3.718048682960964,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003672780371896811,
      "loss": 5.3269,
      "mean_token_accuracy": 0.2066359043121338,
      "num_tokens": 72775306.0,
      "step": 37195
    },
    {
      "entropy": 5.969524192810058,
      "epoch": 3.7185485080221925,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036724548651553433,
      "loss": 5.478,
      "mean_token_accuracy": 0.19728603810071946,
      "num_tokens": 72784616.0,
      "step": 37200
    },
    {
      "entropy": 5.941134643554688,
      "epoch": 3.7190483330834208,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036721293352051486,
      "loss": 5.3982,
      "mean_token_accuracy": 0.2005496859550476,
      "num_tokens": 72794220.0,
      "step": 37205
    },
    {
      "entropy": 6.021936368942261,
      "epoch": 3.7195481581446495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003671803782054418,
      "loss": 5.6055,
      "mean_token_accuracy": 0.18160310238599778,
      "num_tokens": 72804364.0,
      "step": 37210
    },
    {
      "entropy": 5.914173793792725,
      "epoch": 3.7200479832058777,
      "grad_norm": 1.03125,
      "learning_rate": 0.000367147820571134,
      "loss": 5.4187,
      "mean_token_accuracy": 0.20006176978349685,
      "num_tokens": 72814572.0,
      "step": 37215
    },
    {
      "entropy": 5.848346614837647,
      "epoch": 3.7205478082671064,
      "grad_norm": 1.125,
      "learning_rate": 0.00036711526061841086,
      "loss": 5.3024,
      "mean_token_accuracy": 0.20741209834814073,
      "num_tokens": 72825249.0,
      "step": 37220
    },
    {
      "entropy": 6.062492179870605,
      "epoch": 3.721047633328335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003670826983480914,
      "loss": 5.5911,
      "mean_token_accuracy": 0.18181876242160797,
      "num_tokens": 72835525.0,
      "step": 37225
    },
    {
      "entropy": 6.027635717391968,
      "epoch": 3.721547458389564,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036705013376099496,
      "loss": 5.5013,
      "mean_token_accuracy": 0.19287504404783248,
      "num_tokens": 72845973.0,
      "step": 37230
    },
    {
      "entropy": 5.976816558837891,
      "epoch": 3.7220472834507925,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003670175668579409,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1960530936717987,
      "num_tokens": 72855650.0,
      "step": 37235
    },
    {
      "entropy": 5.919238424301147,
      "epoch": 3.7225471085120208,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003669849976397484,
      "loss": 5.4503,
      "mean_token_accuracy": 0.1958076298236847,
      "num_tokens": 72865798.0,
      "step": 37240
    },
    {
      "entropy": 5.931877613067627,
      "epoch": 3.7230469335732495,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003669524261072371,
      "loss": 5.4297,
      "mean_token_accuracy": 0.20583972781896592,
      "num_tokens": 72875536.0,
      "step": 37245
    },
    {
      "entropy": 5.916248941421509,
      "epoch": 3.7235467586344777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003669198522612265,
      "loss": 5.3802,
      "mean_token_accuracy": 0.19952306747436524,
      "num_tokens": 72886134.0,
      "step": 37250
    },
    {
      "entropy": 6.03784556388855,
      "epoch": 3.7240465836957064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036688727610253595,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1924194723367691,
      "num_tokens": 72896269.0,
      "step": 37255
    },
    {
      "entropy": 5.960064792633057,
      "epoch": 3.724546408756935,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003668546976319853,
      "loss": 5.3589,
      "mean_token_accuracy": 0.19875985383987427,
      "num_tokens": 72906528.0,
      "step": 37260
    },
    {
      "entropy": 5.917395782470703,
      "epoch": 3.725046233818164,
      "grad_norm": 1.140625,
      "learning_rate": 0.000366822116850394,
      "loss": 5.4778,
      "mean_token_accuracy": 0.19847125709056854,
      "num_tokens": 72915390.0,
      "step": 37265
    },
    {
      "entropy": 5.892303991317749,
      "epoch": 3.725546058879392,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003667895337585819,
      "loss": 5.4441,
      "mean_token_accuracy": 0.196845942735672,
      "num_tokens": 72926194.0,
      "step": 37270
    },
    {
      "entropy": 5.887879276275635,
      "epoch": 3.7260458839406208,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036675694835736875,
      "loss": 5.3713,
      "mean_token_accuracy": 0.19868656396865844,
      "num_tokens": 72935434.0,
      "step": 37275
    },
    {
      "entropy": 5.978984737396241,
      "epoch": 3.7265457090018494,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003667243606475744,
      "loss": 5.4666,
      "mean_token_accuracy": 0.1953526556491852,
      "num_tokens": 72944664.0,
      "step": 37280
    },
    {
      "entropy": 5.95403642654419,
      "epoch": 3.7270455340630777,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00036669177063001866,
      "loss": 5.5026,
      "mean_token_accuracy": 0.19778708517551422,
      "num_tokens": 72954213.0,
      "step": 37285
    },
    {
      "entropy": 5.927277708053589,
      "epoch": 3.7275453591243064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003666591783055216,
      "loss": 5.3702,
      "mean_token_accuracy": 0.198016694188118,
      "num_tokens": 72964097.0,
      "step": 37290
    },
    {
      "entropy": 5.852818870544434,
      "epoch": 3.728045184185535,
      "grad_norm": 1.125,
      "learning_rate": 0.0003666265836749033,
      "loss": 5.3631,
      "mean_token_accuracy": 0.1988002523779869,
      "num_tokens": 72973807.0,
      "step": 37295
    },
    {
      "entropy": 6.0253458499908445,
      "epoch": 3.728545009246764,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036659398673898356,
      "loss": 5.5044,
      "mean_token_accuracy": 0.18844879120588304,
      "num_tokens": 72983466.0,
      "step": 37300
    },
    {
      "entropy": 5.8940176486969,
      "epoch": 3.729044834307992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036656138749858287,
      "loss": 5.3549,
      "mean_token_accuracy": 0.20272477418184282,
      "num_tokens": 72992377.0,
      "step": 37305
    },
    {
      "entropy": 5.859115028381348,
      "epoch": 3.7295446593692207,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036652878595452105,
      "loss": 5.4174,
      "mean_token_accuracy": 0.1961490035057068,
      "num_tokens": 73001699.0,
      "step": 37310
    },
    {
      "entropy": 5.860056066513062,
      "epoch": 3.7300444844304494,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003664961821076187,
      "loss": 5.3901,
      "mean_token_accuracy": 0.208195298910141,
      "num_tokens": 73011011.0,
      "step": 37315
    },
    {
      "entropy": 5.909420490264893,
      "epoch": 3.7305443094916777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036646357595869586,
      "loss": 5.3898,
      "mean_token_accuracy": 0.21258852630853653,
      "num_tokens": 73019868.0,
      "step": 37320
    },
    {
      "entropy": 5.925349235534668,
      "epoch": 3.7310441345529064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000366430967508573,
      "loss": 5.3895,
      "mean_token_accuracy": 0.20169070214033127,
      "num_tokens": 73028812.0,
      "step": 37325
    },
    {
      "entropy": 5.879197120666504,
      "epoch": 3.731543959614135,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003663983567580705,
      "loss": 5.322,
      "mean_token_accuracy": 0.21256615221500397,
      "num_tokens": 73038138.0,
      "step": 37330
    },
    {
      "entropy": 5.963660383224488,
      "epoch": 3.732043784675364,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036636574370800887,
      "loss": 5.4575,
      "mean_token_accuracy": 0.19735547304153442,
      "num_tokens": 73048705.0,
      "step": 37335
    },
    {
      "entropy": 5.955954456329346,
      "epoch": 3.732543609736592,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003663331283592086,
      "loss": 5.4448,
      "mean_token_accuracy": 0.2009085714817047,
      "num_tokens": 73058107.0,
      "step": 37340
    },
    {
      "entropy": 5.927512550354004,
      "epoch": 3.7330434347978207,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00036630051071249043,
      "loss": 5.4055,
      "mean_token_accuracy": 0.200644414126873,
      "num_tokens": 73067869.0,
      "step": 37345
    },
    {
      "entropy": 5.902604818344116,
      "epoch": 3.7335432598590494,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036626789076867484,
      "loss": 5.4376,
      "mean_token_accuracy": 0.1958249405026436,
      "num_tokens": 73078020.0,
      "step": 37350
    },
    {
      "entropy": 5.958486032485962,
      "epoch": 3.7340430849202777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003662352685285826,
      "loss": 5.433,
      "mean_token_accuracy": 0.19403730630874633,
      "num_tokens": 73088451.0,
      "step": 37355
    },
    {
      "entropy": 5.965102958679199,
      "epoch": 3.7345429099815064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003662026439930345,
      "loss": 5.413,
      "mean_token_accuracy": 0.1998227372765541,
      "num_tokens": 73098409.0,
      "step": 37360
    },
    {
      "entropy": 6.058593606948852,
      "epoch": 3.735042735042735,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036617001716285135,
      "loss": 5.5551,
      "mean_token_accuracy": 0.18619562834501266,
      "num_tokens": 73109022.0,
      "step": 37365
    },
    {
      "entropy": 5.934650135040283,
      "epoch": 3.735542560103964,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000366137388038854,
      "loss": 5.4566,
      "mean_token_accuracy": 0.19957093745470048,
      "num_tokens": 73118585.0,
      "step": 37370
    },
    {
      "entropy": 5.991137599945068,
      "epoch": 3.736042385165192,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036610475662186343,
      "loss": 5.5434,
      "mean_token_accuracy": 0.1983332633972168,
      "num_tokens": 73128152.0,
      "step": 37375
    },
    {
      "entropy": 6.003345680236817,
      "epoch": 3.7365422102264207,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003660721229127006,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19967614561319352,
      "num_tokens": 73137103.0,
      "step": 37380
    },
    {
      "entropy": 5.9580583572387695,
      "epoch": 3.7370420352876494,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036603948691218665,
      "loss": 5.4266,
      "mean_token_accuracy": 0.2003611832857132,
      "num_tokens": 73145867.0,
      "step": 37385
    },
    {
      "entropy": 5.9604637145996096,
      "epoch": 3.7375418603488777,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003660068486211426,
      "loss": 5.4621,
      "mean_token_accuracy": 0.1999151661992073,
      "num_tokens": 73154381.0,
      "step": 37390
    },
    {
      "entropy": 5.868138551712036,
      "epoch": 3.7380416854101064,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003659742080403897,
      "loss": 5.4424,
      "mean_token_accuracy": 0.2000004917383194,
      "num_tokens": 73164493.0,
      "step": 37395
    },
    {
      "entropy": 5.9321881294250485,
      "epoch": 3.738541510471335,
      "grad_norm": 1.25,
      "learning_rate": 0.0003659415651707491,
      "loss": 5.4266,
      "mean_token_accuracy": 0.2004410982131958,
      "num_tokens": 73174230.0,
      "step": 37400
    },
    {
      "entropy": 5.972543239593506,
      "epoch": 3.739041335532564,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003659089200130421,
      "loss": 5.4081,
      "mean_token_accuracy": 0.19910161942243576,
      "num_tokens": 73184858.0,
      "step": 37405
    },
    {
      "entropy": 6.005197286605835,
      "epoch": 3.739541160593792,
      "grad_norm": 1.125,
      "learning_rate": 0.00036587627256809,
      "loss": 5.4809,
      "mean_token_accuracy": 0.20230561792850493,
      "num_tokens": 73194770.0,
      "step": 37410
    },
    {
      "entropy": 5.989680624008178,
      "epoch": 3.7400409856550207,
      "grad_norm": 1.125,
      "learning_rate": 0.0003658436228367144,
      "loss": 5.384,
      "mean_token_accuracy": 0.20443608462810517,
      "num_tokens": 73203836.0,
      "step": 37415
    },
    {
      "entropy": 5.893261957168579,
      "epoch": 3.7405408107162494,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036581097081973655,
      "loss": 5.4486,
      "mean_token_accuracy": 0.19470659494400025,
      "num_tokens": 73213359.0,
      "step": 37420
    },
    {
      "entropy": 5.887816429138184,
      "epoch": 3.7410406357774777,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003657783165179781,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2113081470131874,
      "num_tokens": 73223064.0,
      "step": 37425
    },
    {
      "entropy": 5.963651084899903,
      "epoch": 3.7415404608387064,
      "grad_norm": 1.125,
      "learning_rate": 0.0003657456599322606,
      "loss": 5.4021,
      "mean_token_accuracy": 0.20513152778148652,
      "num_tokens": 73232036.0,
      "step": 37430
    },
    {
      "entropy": 5.905822134017944,
      "epoch": 3.742040285899935,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036571300106340553,
      "loss": 5.431,
      "mean_token_accuracy": 0.19586022496223449,
      "num_tokens": 73242346.0,
      "step": 37435
    },
    {
      "entropy": 5.904023694992065,
      "epoch": 3.742540110961164,
      "grad_norm": 1.25,
      "learning_rate": 0.00036568033991223476,
      "loss": 5.3936,
      "mean_token_accuracy": 0.19484868943691253,
      "num_tokens": 73249471.0,
      "step": 37440
    },
    {
      "entropy": 5.927030372619629,
      "epoch": 3.743039936022392,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036564767647956993,
      "loss": 5.496,
      "mean_token_accuracy": 0.19260770529508592,
      "num_tokens": 73258709.0,
      "step": 37445
    },
    {
      "entropy": 5.958307552337646,
      "epoch": 3.7435397610836207,
      "grad_norm": 1.25,
      "learning_rate": 0.0003656150107662329,
      "loss": 5.3907,
      "mean_token_accuracy": 0.2058065041899681,
      "num_tokens": 73267432.0,
      "step": 37450
    },
    {
      "entropy": 5.972232913970947,
      "epoch": 3.7440395861448494,
      "grad_norm": 1.125,
      "learning_rate": 0.00036558234277304555,
      "loss": 5.5122,
      "mean_token_accuracy": 0.1926780089735985,
      "num_tokens": 73277071.0,
      "step": 37455
    },
    {
      "entropy": 5.957896900177002,
      "epoch": 3.7445394112060777,
      "grad_norm": 1.125,
      "learning_rate": 0.00036554967250082977,
      "loss": 5.4132,
      "mean_token_accuracy": 0.19508455842733383,
      "num_tokens": 73285528.0,
      "step": 37460
    },
    {
      "entropy": 5.917503595352173,
      "epoch": 3.7450392362673064,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036551699995040755,
      "loss": 5.3729,
      "mean_token_accuracy": 0.20305712521076202,
      "num_tokens": 73294530.0,
      "step": 37465
    },
    {
      "entropy": 6.002968168258667,
      "epoch": 3.745539061328535,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036548432512260084,
      "loss": 5.5135,
      "mean_token_accuracy": 0.1936844617128372,
      "num_tokens": 73304960.0,
      "step": 37470
    },
    {
      "entropy": 5.940178871154785,
      "epoch": 3.746038886389764,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003654516480182318,
      "loss": 5.4154,
      "mean_token_accuracy": 0.20198030173778533,
      "num_tokens": 73314523.0,
      "step": 37475
    },
    {
      "entropy": 5.921142196655273,
      "epoch": 3.746538711450992,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003654189686381225,
      "loss": 5.4462,
      "mean_token_accuracy": 0.19897663295269014,
      "num_tokens": 73323618.0,
      "step": 37480
    },
    {
      "entropy": 6.00618543624878,
      "epoch": 3.7470385365122207,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003653862869830953,
      "loss": 5.5323,
      "mean_token_accuracy": 0.1902869775891304,
      "num_tokens": 73334443.0,
      "step": 37485
    },
    {
      "entropy": 5.92495551109314,
      "epoch": 3.7475383615734494,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003653536030539725,
      "loss": 5.4348,
      "mean_token_accuracy": 0.1977334588766098,
      "num_tokens": 73344902.0,
      "step": 37490
    },
    {
      "entropy": 5.900302362442017,
      "epoch": 3.7480381866346777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036532091685157614,
      "loss": 5.3297,
      "mean_token_accuracy": 0.20708636045455933,
      "num_tokens": 73354129.0,
      "step": 37495
    },
    {
      "entropy": 5.923274087905884,
      "epoch": 3.7485380116959064,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036528822837672884,
      "loss": 5.3553,
      "mean_token_accuracy": 0.20636734813451768,
      "num_tokens": 73363711.0,
      "step": 37500
    },
    {
      "entropy": 5.90574688911438,
      "epoch": 3.749037836757135,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003652555376302528,
      "loss": 5.4409,
      "mean_token_accuracy": 0.20490527898073196,
      "num_tokens": 73372746.0,
      "step": 37505
    },
    {
      "entropy": 5.907392883300782,
      "epoch": 3.749537661818364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036522284461297077,
      "loss": 5.4398,
      "mean_token_accuracy": 0.19293965697288512,
      "num_tokens": 73383767.0,
      "step": 37510
    },
    {
      "entropy": 6.0765033721923825,
      "epoch": 3.750037486879592,
      "grad_norm": 1.125,
      "learning_rate": 0.00036519014932570517,
      "loss": 5.5249,
      "mean_token_accuracy": 0.1926767572760582,
      "num_tokens": 73392592.0,
      "step": 37515
    },
    {
      "entropy": 5.930399417877197,
      "epoch": 3.7505373119408207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036515745176927863,
      "loss": 5.4187,
      "mean_token_accuracy": 0.2011762633919716,
      "num_tokens": 73401634.0,
      "step": 37520
    },
    {
      "entropy": 5.939880132675171,
      "epoch": 3.7510371370020494,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003651247519445138,
      "loss": 5.3834,
      "mean_token_accuracy": 0.20495097935199738,
      "num_tokens": 73411086.0,
      "step": 37525
    },
    {
      "entropy": 5.958910036087036,
      "epoch": 3.7515369620632777,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036509204985223336,
      "loss": 5.4996,
      "mean_token_accuracy": 0.19317017793655394,
      "num_tokens": 73420300.0,
      "step": 37530
    },
    {
      "entropy": 5.911330366134644,
      "epoch": 3.7520367871245064,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003650593454932602,
      "loss": 5.3624,
      "mean_token_accuracy": 0.20773118883371353,
      "num_tokens": 73429983.0,
      "step": 37535
    },
    {
      "entropy": 5.939871740341187,
      "epoch": 3.752536612185735,
      "grad_norm": 1.265625,
      "learning_rate": 0.000365026638868417,
      "loss": 5.3805,
      "mean_token_accuracy": 0.20241614431142807,
      "num_tokens": 73439075.0,
      "step": 37540
    },
    {
      "entropy": 5.97621717453003,
      "epoch": 3.753036437246964,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036499392997852684,
      "loss": 5.4267,
      "mean_token_accuracy": 0.2002446934580803,
      "num_tokens": 73448708.0,
      "step": 37545
    },
    {
      "entropy": 5.9583885192871096,
      "epoch": 3.753536262308192,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003649612188244125,
      "loss": 5.5148,
      "mean_token_accuracy": 0.194077005982399,
      "num_tokens": 73457331.0,
      "step": 37550
    },
    {
      "entropy": 5.873609972000122,
      "epoch": 3.7540360873694207,
      "grad_norm": 1.234375,
      "learning_rate": 0.000364928505406897,
      "loss": 5.2718,
      "mean_token_accuracy": 0.20328764468431473,
      "num_tokens": 73466061.0,
      "step": 37555
    },
    {
      "entropy": 5.877326726913452,
      "epoch": 3.7545359124306494,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003648957897268035,
      "loss": 5.4178,
      "mean_token_accuracy": 0.2051236018538475,
      "num_tokens": 73477598.0,
      "step": 37560
    },
    {
      "entropy": 5.84876389503479,
      "epoch": 3.7550357374918777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003648630717849551,
      "loss": 5.2449,
      "mean_token_accuracy": 0.2187777951359749,
      "num_tokens": 73487453.0,
      "step": 37565
    },
    {
      "entropy": 5.9822605609893795,
      "epoch": 3.7555355625531064,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036483035158217485,
      "loss": 5.4957,
      "mean_token_accuracy": 0.195467609167099,
      "num_tokens": 73496239.0,
      "step": 37570
    },
    {
      "entropy": 5.970463848114013,
      "epoch": 3.756035387614335,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036479762911928606,
      "loss": 5.454,
      "mean_token_accuracy": 0.19237158000469207,
      "num_tokens": 73506295.0,
      "step": 37575
    },
    {
      "entropy": 5.8704695224761965,
      "epoch": 3.756535212675564,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036476490439711206,
      "loss": 5.3818,
      "mean_token_accuracy": 0.20979726910591126,
      "num_tokens": 73516474.0,
      "step": 37580
    },
    {
      "entropy": 5.903824281692505,
      "epoch": 3.757035037736792,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003647321774164762,
      "loss": 5.3113,
      "mean_token_accuracy": 0.21198756098747254,
      "num_tokens": 73526141.0,
      "step": 37585
    },
    {
      "entropy": 6.004900026321411,
      "epoch": 3.7575348627980207,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003646994481782018,
      "loss": 5.5711,
      "mean_token_accuracy": 0.18615491241216658,
      "num_tokens": 73536203.0,
      "step": 37590
    },
    {
      "entropy": 5.846355342864991,
      "epoch": 3.758034687859249,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003646667166831124,
      "loss": 5.2769,
      "mean_token_accuracy": 0.2150360271334648,
      "num_tokens": 73545785.0,
      "step": 37595
    },
    {
      "entropy": 5.945638751983642,
      "epoch": 3.7585345129204777,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003646339829320315,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20077814012765885,
      "num_tokens": 73554736.0,
      "step": 37600
    },
    {
      "entropy": 5.950120258331299,
      "epoch": 3.7590343379817064,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003646012469257826,
      "loss": 5.4051,
      "mean_token_accuracy": 0.20036603659391403,
      "num_tokens": 73564507.0,
      "step": 37605
    },
    {
      "entropy": 5.985602569580078,
      "epoch": 3.759534163042935,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003645685086651894,
      "loss": 5.4768,
      "mean_token_accuracy": 0.1934155896306038,
      "num_tokens": 73574120.0,
      "step": 37610
    },
    {
      "entropy": 5.9842535018920895,
      "epoch": 3.760033988104164,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003645357681510756,
      "loss": 5.482,
      "mean_token_accuracy": 0.20030355900526048,
      "num_tokens": 73584191.0,
      "step": 37615
    },
    {
      "entropy": 5.957414436340332,
      "epoch": 3.760533813165392,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003645030253842649,
      "loss": 5.4383,
      "mean_token_accuracy": 0.19430186599493027,
      "num_tokens": 73593964.0,
      "step": 37620
    },
    {
      "entropy": 6.032321500778198,
      "epoch": 3.7610336382266207,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036447028036558105,
      "loss": 5.5478,
      "mean_token_accuracy": 0.19030876606702804,
      "num_tokens": 73603952.0,
      "step": 37625
    },
    {
      "entropy": 6.025291109085083,
      "epoch": 3.761533463287849,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000364437533095848,
      "loss": 5.4823,
      "mean_token_accuracy": 0.19491055607795715,
      "num_tokens": 73612868.0,
      "step": 37630
    },
    {
      "entropy": 5.991901588439942,
      "epoch": 3.7620332883490777,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036440478357588964,
      "loss": 5.5286,
      "mean_token_accuracy": 0.19137470424175262,
      "num_tokens": 73623272.0,
      "step": 37635
    },
    {
      "entropy": 5.9284515380859375,
      "epoch": 3.7625331134103064,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00036437203180653,
      "loss": 5.3863,
      "mean_token_accuracy": 0.19700649082660676,
      "num_tokens": 73634050.0,
      "step": 37640
    },
    {
      "entropy": 5.941171646118164,
      "epoch": 3.763032938471535,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000364339277788593,
      "loss": 5.3949,
      "mean_token_accuracy": 0.2024761766195297,
      "num_tokens": 73643327.0,
      "step": 37645
    },
    {
      "entropy": 5.955933427810669,
      "epoch": 3.763532763532764,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003643065215229027,
      "loss": 5.418,
      "mean_token_accuracy": 0.20046029537916182,
      "num_tokens": 73653644.0,
      "step": 37650
    },
    {
      "entropy": 5.958263540267945,
      "epoch": 3.764032588593992,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036427376301028335,
      "loss": 5.4091,
      "mean_token_accuracy": 0.200792396068573,
      "num_tokens": 73663711.0,
      "step": 37655
    },
    {
      "entropy": 5.943794584274292,
      "epoch": 3.7645324136552207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003642410022515591,
      "loss": 5.3634,
      "mean_token_accuracy": 0.20560530573129654,
      "num_tokens": 73673078.0,
      "step": 37660
    },
    {
      "entropy": 5.861472940444946,
      "epoch": 3.765032238716449,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003642082392475541,
      "loss": 5.271,
      "mean_token_accuracy": 0.21531863659620284,
      "num_tokens": 73682144.0,
      "step": 37665
    },
    {
      "entropy": 5.854277324676514,
      "epoch": 3.7655320637776777,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036417547399909285,
      "loss": 5.263,
      "mean_token_accuracy": 0.21331105679273604,
      "num_tokens": 73691502.0,
      "step": 37670
    },
    {
      "entropy": 5.9999659061431885,
      "epoch": 3.7660318888389064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036414270650699963,
      "loss": 5.5957,
      "mean_token_accuracy": 0.18647281676530839,
      "num_tokens": 73701584.0,
      "step": 37675
    },
    {
      "entropy": 5.983278894424439,
      "epoch": 3.766531713900135,
      "grad_norm": 1.5,
      "learning_rate": 0.00036410993677209885,
      "loss": 5.4664,
      "mean_token_accuracy": 0.19933225512504577,
      "num_tokens": 73711495.0,
      "step": 37680
    },
    {
      "entropy": 5.9589649677276615,
      "epoch": 3.767031538961364,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000364077164795215,
      "loss": 5.382,
      "mean_token_accuracy": 0.20143394470214843,
      "num_tokens": 73721531.0,
      "step": 37685
    },
    {
      "entropy": 5.932640457153321,
      "epoch": 3.767531364022592,
      "grad_norm": 1.125,
      "learning_rate": 0.0003640443905771726,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20175531953573228,
      "num_tokens": 73731900.0,
      "step": 37690
    },
    {
      "entropy": 5.960603141784668,
      "epoch": 3.7680311890838207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036401161411879615,
      "loss": 5.4247,
      "mean_token_accuracy": 0.20115264654159545,
      "num_tokens": 73741861.0,
      "step": 37695
    },
    {
      "entropy": 5.899132347106933,
      "epoch": 3.768531014145049,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036397883542091045,
      "loss": 5.3723,
      "mean_token_accuracy": 0.2016010031104088,
      "num_tokens": 73751672.0,
      "step": 37700
    },
    {
      "entropy": 5.877369594573975,
      "epoch": 3.7690308392062777,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036394605448434014,
      "loss": 5.2925,
      "mean_token_accuracy": 0.2203026831150055,
      "num_tokens": 73761287.0,
      "step": 37705
    },
    {
      "entropy": 5.871726846694946,
      "epoch": 3.7695306642675064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003639132713099101,
      "loss": 5.4203,
      "mean_token_accuracy": 0.2068535417318344,
      "num_tokens": 73771377.0,
      "step": 37710
    },
    {
      "entropy": 6.137187099456787,
      "epoch": 3.770030489328735,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003638804858984449,
      "loss": 5.5959,
      "mean_token_accuracy": 0.18558543920516968,
      "num_tokens": 73781995.0,
      "step": 37715
    },
    {
      "entropy": 6.049393987655639,
      "epoch": 3.7705303143899638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036384769825076964,
      "loss": 5.4769,
      "mean_token_accuracy": 0.19333163946866988,
      "num_tokens": 73792372.0,
      "step": 37720
    },
    {
      "entropy": 5.946305751800537,
      "epoch": 3.771030139451192,
      "grad_norm": 1.28125,
      "learning_rate": 0.000363814908367709,
      "loss": 5.4103,
      "mean_token_accuracy": 0.19898100793361664,
      "num_tokens": 73801298.0,
      "step": 37725
    },
    {
      "entropy": 5.8995490074157715,
      "epoch": 3.7715299645124207,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00036378211625008827,
      "loss": 5.4197,
      "mean_token_accuracy": 0.20101588517427443,
      "num_tokens": 73810913.0,
      "step": 37730
    },
    {
      "entropy": 5.937164783477783,
      "epoch": 3.772029789573649,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003637493218987323,
      "loss": 5.3954,
      "mean_token_accuracy": 0.2013123005628586,
      "num_tokens": 73821558.0,
      "step": 37735
    },
    {
      "entropy": 5.927765941619873,
      "epoch": 3.7725296146348777,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003637165253144662,
      "loss": 5.4026,
      "mean_token_accuracy": 0.19750298410654069,
      "num_tokens": 73830820.0,
      "step": 37740
    },
    {
      "entropy": 5.911085033416748,
      "epoch": 3.7730294396961064,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003636837264981152,
      "loss": 5.3898,
      "mean_token_accuracy": 0.20576549023389817,
      "num_tokens": 73840877.0,
      "step": 37745
    },
    {
      "entropy": 5.9774034976959225,
      "epoch": 3.773529264757335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003636509254505045,
      "loss": 5.5361,
      "mean_token_accuracy": 0.1954856336116791,
      "num_tokens": 73850054.0,
      "step": 37750
    },
    {
      "entropy": 6.015618181228637,
      "epoch": 3.7740290898185638,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003636181221724592,
      "loss": 5.4448,
      "mean_token_accuracy": 0.19447749704122544,
      "num_tokens": 73860478.0,
      "step": 37755
    },
    {
      "entropy": 5.996906995773315,
      "epoch": 3.774528914879792,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036358531666480477,
      "loss": 5.4988,
      "mean_token_accuracy": 0.19881636053323745,
      "num_tokens": 73870793.0,
      "step": 37760
    },
    {
      "entropy": 5.9226432800292965,
      "epoch": 3.7750287399410207,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036355250892836663,
      "loss": 5.3361,
      "mean_token_accuracy": 0.20304545015096664,
      "num_tokens": 73880411.0,
      "step": 37765
    },
    {
      "entropy": 5.862315368652344,
      "epoch": 3.775528565002249,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036351969896397006,
      "loss": 5.4434,
      "mean_token_accuracy": 0.20985670387744904,
      "num_tokens": 73890626.0,
      "step": 37770
    },
    {
      "entropy": 6.016927719116211,
      "epoch": 3.7760283900634777,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036348688677244074,
      "loss": 5.4913,
      "mean_token_accuracy": 0.19210301488637924,
      "num_tokens": 73900131.0,
      "step": 37775
    },
    {
      "entropy": 5.845275783538819,
      "epoch": 3.7765282151247064,
      "grad_norm": 1.125,
      "learning_rate": 0.00036345407235460417,
      "loss": 5.2765,
      "mean_token_accuracy": 0.22265084236860275,
      "num_tokens": 73909905.0,
      "step": 37780
    },
    {
      "entropy": 5.9615888595581055,
      "epoch": 3.777028040185935,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036342125571128575,
      "loss": 5.4937,
      "mean_token_accuracy": 0.20105593055486679,
      "num_tokens": 73920064.0,
      "step": 37785
    },
    {
      "entropy": 5.963688468933105,
      "epoch": 3.7775278652471633,
      "grad_norm": 1.21875,
      "learning_rate": 0.00036338843684331133,
      "loss": 5.5063,
      "mean_token_accuracy": 0.19064111411571502,
      "num_tokens": 73929698.0,
      "step": 37790
    },
    {
      "entropy": 5.931861877441406,
      "epoch": 3.778027690308392,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036335561575150663,
      "loss": 5.5229,
      "mean_token_accuracy": 0.1980072170495987,
      "num_tokens": 73939465.0,
      "step": 37795
    },
    {
      "entropy": 5.989009523391724,
      "epoch": 3.7785275153696207,
      "grad_norm": 1.125,
      "learning_rate": 0.00036332279243669733,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19110678434371947,
      "num_tokens": 73949071.0,
      "step": 37800
    },
    {
      "entropy": 5.986454153060913,
      "epoch": 3.779027340430849,
      "grad_norm": 1.1875,
      "learning_rate": 0.00036328996689970937,
      "loss": 5.382,
      "mean_token_accuracy": 0.21034340858459472,
      "num_tokens": 73957979.0,
      "step": 37805
    },
    {
      "entropy": 5.962321996688843,
      "epoch": 3.7795271654920777,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003632571391413685,
      "loss": 5.3597,
      "mean_token_accuracy": 0.2041837215423584,
      "num_tokens": 73967312.0,
      "step": 37810
    },
    {
      "entropy": 5.966868495941162,
      "epoch": 3.7800269905533064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036322430916250077,
      "loss": 5.5034,
      "mean_token_accuracy": 0.19090303629636765,
      "num_tokens": 73976933.0,
      "step": 37815
    },
    {
      "entropy": 6.006643295288086,
      "epoch": 3.780526815614535,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036319147696393207,
      "loss": 5.5595,
      "mean_token_accuracy": 0.19027288556098937,
      "num_tokens": 73986527.0,
      "step": 37820
    },
    {
      "entropy": 5.991017532348633,
      "epoch": 3.7810266406757633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003631586425464885,
      "loss": 5.5502,
      "mean_token_accuracy": 0.1865057662129402,
      "num_tokens": 73996225.0,
      "step": 37825
    },
    {
      "entropy": 6.008220148086548,
      "epoch": 3.781526465736992,
      "grad_norm": 1.046875,
      "learning_rate": 0.00036312580591099625,
      "loss": 5.4881,
      "mean_token_accuracy": 0.19106250256299973,
      "num_tokens": 74006509.0,
      "step": 37830
    },
    {
      "entropy": 5.930658197402954,
      "epoch": 3.7820262907982207,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003630929670582813,
      "loss": 5.3586,
      "mean_token_accuracy": 0.20679402202367783,
      "num_tokens": 74014924.0,
      "step": 37835
    },
    {
      "entropy": 5.929020023345947,
      "epoch": 3.782526115859449,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036306012598917,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1959429770708084,
      "num_tokens": 74024880.0,
      "step": 37840
    },
    {
      "entropy": 5.965630912780762,
      "epoch": 3.7830259409206777,
      "grad_norm": 1.125,
      "learning_rate": 0.0003630272827044886,
      "loss": 5.4535,
      "mean_token_accuracy": 0.19822342395782472,
      "num_tokens": 74033695.0,
      "step": 37845
    },
    {
      "entropy": 5.995543432235718,
      "epoch": 3.7835257659819064,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036299443720506344,
      "loss": 5.4273,
      "mean_token_accuracy": 0.19965787529945372,
      "num_tokens": 74044044.0,
      "step": 37850
    },
    {
      "entropy": 6.003092336654663,
      "epoch": 3.784025591043135,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036296158949172087,
      "loss": 5.5192,
      "mean_token_accuracy": 0.18727033585309982,
      "num_tokens": 74052961.0,
      "step": 37855
    },
    {
      "entropy": 5.930329418182373,
      "epoch": 3.7845254161043633,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003629287395652873,
      "loss": 5.4377,
      "mean_token_accuracy": 0.19757704436779022,
      "num_tokens": 74063116.0,
      "step": 37860
    },
    {
      "entropy": 5.943247079849243,
      "epoch": 3.785025241165592,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003628958874265893,
      "loss": 5.4342,
      "mean_token_accuracy": 0.19752874225378036,
      "num_tokens": 74073191.0,
      "step": 37865
    },
    {
      "entropy": 5.942607259750366,
      "epoch": 3.7855250662268207,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003628630330764534,
      "loss": 5.419,
      "mean_token_accuracy": 0.19725254625082017,
      "num_tokens": 74082974.0,
      "step": 37870
    },
    {
      "entropy": 5.997967481613159,
      "epoch": 3.786024891288049,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003628301765157062,
      "loss": 5.4667,
      "mean_token_accuracy": 0.1966873213648796,
      "num_tokens": 74092287.0,
      "step": 37875
    },
    {
      "entropy": 5.983355140686035,
      "epoch": 3.7865247163492777,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036279731774517435,
      "loss": 5.5059,
      "mean_token_accuracy": 0.19576913267374038,
      "num_tokens": 74101618.0,
      "step": 37880
    },
    {
      "entropy": 5.933871269226074,
      "epoch": 3.7870245414105064,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00036276445676568453,
      "loss": 5.4143,
      "mean_token_accuracy": 0.20160800814628602,
      "num_tokens": 74111412.0,
      "step": 37885
    },
    {
      "entropy": 5.942109870910644,
      "epoch": 3.787524366471735,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00036273159357806363,
      "loss": 5.4577,
      "mean_token_accuracy": 0.1986621603369713,
      "num_tokens": 74121435.0,
      "step": 37890
    },
    {
      "entropy": 5.947222948074341,
      "epoch": 3.7880241915329633,
      "grad_norm": 1.03125,
      "learning_rate": 0.00036269872818313836,
      "loss": 5.4189,
      "mean_token_accuracy": 0.2024250328540802,
      "num_tokens": 74131763.0,
      "step": 37895
    },
    {
      "entropy": 5.9657808303833,
      "epoch": 3.788524016594192,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003626658605817357,
      "loss": 5.4597,
      "mean_token_accuracy": 0.19667388647794723,
      "num_tokens": 74141671.0,
      "step": 37900
    },
    {
      "entropy": 5.87021656036377,
      "epoch": 3.7890238416554207,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036263299077468256,
      "loss": 5.3159,
      "mean_token_accuracy": 0.20370438247919082,
      "num_tokens": 74151554.0,
      "step": 37905
    },
    {
      "entropy": 5.836766576766967,
      "epoch": 3.789523666716649,
      "grad_norm": 1.0,
      "learning_rate": 0.0003626001187628059,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21358111947774888,
      "num_tokens": 74160420.0,
      "step": 37910
    },
    {
      "entropy": 5.861282873153686,
      "epoch": 3.7900234917778777,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003625672445469327,
      "loss": 5.3772,
      "mean_token_accuracy": 0.21284514367580415,
      "num_tokens": 74171966.0,
      "step": 37915
    },
    {
      "entropy": 5.8570598602294925,
      "epoch": 3.7905233168391064,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00036253436812789034,
      "loss": 5.3361,
      "mean_token_accuracy": 0.21016574501991273,
      "num_tokens": 74181938.0,
      "step": 37920
    },
    {
      "entropy": 5.9922913074493405,
      "epoch": 3.791023141900335,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003625014895065057,
      "loss": 5.4709,
      "mean_token_accuracy": 0.20021916180849075,
      "num_tokens": 74191557.0,
      "step": 37925
    },
    {
      "entropy": 5.979022264480591,
      "epoch": 3.7915229669615633,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003624686086836061,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20405324697494506,
      "num_tokens": 74200807.0,
      "step": 37930
    },
    {
      "entropy": 5.947378301620484,
      "epoch": 3.792022792022792,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003624357256600188,
      "loss": 5.4681,
      "mean_token_accuracy": 0.20092708021402358,
      "num_tokens": 74210635.0,
      "step": 37935
    },
    {
      "entropy": 5.96271858215332,
      "epoch": 3.7925226170840207,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003624028404365712,
      "loss": 5.4845,
      "mean_token_accuracy": 0.19223793596029282,
      "num_tokens": 74220189.0,
      "step": 37940
    },
    {
      "entropy": 6.014693927764893,
      "epoch": 3.793022442145249,
      "grad_norm": 1.25,
      "learning_rate": 0.0003623699530140906,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19955286979675294,
      "num_tokens": 74229916.0,
      "step": 37945
    },
    {
      "entropy": 6.02412257194519,
      "epoch": 3.7935222672064777,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003623370633934043,
      "loss": 5.5082,
      "mean_token_accuracy": 0.19104171097278594,
      "num_tokens": 74239829.0,
      "step": 37950
    },
    {
      "entropy": 6.004844713211059,
      "epoch": 3.7940220922677064,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003623041715753402,
      "loss": 5.4856,
      "mean_token_accuracy": 0.19114019870758056,
      "num_tokens": 74248395.0,
      "step": 37955
    },
    {
      "entropy": 5.933782863616943,
      "epoch": 3.794521917328935,
      "grad_norm": 1.0625,
      "learning_rate": 0.00036227127756072554,
      "loss": 5.3796,
      "mean_token_accuracy": 0.20275249630212783,
      "num_tokens": 74258122.0,
      "step": 37960
    },
    {
      "entropy": 6.002561044692993,
      "epoch": 3.7950217423901633,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003622383813503879,
      "loss": 5.509,
      "mean_token_accuracy": 0.1938689738512039,
      "num_tokens": 74268100.0,
      "step": 37965
    },
    {
      "entropy": 6.072827434539795,
      "epoch": 3.795521567451392,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00036220548294515516,
      "loss": 5.555,
      "mean_token_accuracy": 0.1908547818660736,
      "num_tokens": 74278251.0,
      "step": 37970
    },
    {
      "entropy": 5.939408206939698,
      "epoch": 3.7960213925126207,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003621725823458549,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19890183806419373,
      "num_tokens": 74288379.0,
      "step": 37975
    },
    {
      "entropy": 5.991646003723145,
      "epoch": 3.796521217573849,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036213967955331475,
      "loss": 5.4291,
      "mean_token_accuracy": 0.20731855928897858,
      "num_tokens": 74297867.0,
      "step": 37980
    },
    {
      "entropy": 5.9355944156646725,
      "epoch": 3.7970210426350777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036210677456836285,
      "loss": 5.4326,
      "mean_token_accuracy": 0.2085006430745125,
      "num_tokens": 74308305.0,
      "step": 37985
    },
    {
      "entropy": 5.912218141555786,
      "epoch": 3.7975208676963064,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036207386739182685,
      "loss": 5.4406,
      "mean_token_accuracy": 0.20642007291316986,
      "num_tokens": 74318217.0,
      "step": 37990
    },
    {
      "entropy": 5.882007026672364,
      "epoch": 3.798020692757535,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003620409580245348,
      "loss": 5.3782,
      "mean_token_accuracy": 0.20941568166017532,
      "num_tokens": 74327420.0,
      "step": 37995
    },
    {
      "entropy": 5.985098886489868,
      "epoch": 3.7985205178187633,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036200804646731457,
      "loss": 5.4964,
      "mean_token_accuracy": 0.1928093448281288,
      "num_tokens": 74336805.0,
      "step": 38000
    },
    {
      "entropy": 6.0030592918396,
      "epoch": 3.799020342879992,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003619751327209943,
      "loss": 5.4685,
      "mean_token_accuracy": 0.19906605333089827,
      "num_tokens": 74346989.0,
      "step": 38005
    },
    {
      "entropy": 5.9364338397979735,
      "epoch": 3.7995201679412207,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036194221678640206,
      "loss": 5.3989,
      "mean_token_accuracy": 0.20158365964889527,
      "num_tokens": 74356637.0,
      "step": 38010
    },
    {
      "entropy": 5.944382667541504,
      "epoch": 3.800019993002449,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000361909298664366,
      "loss": 5.4293,
      "mean_token_accuracy": 0.19726411551237105,
      "num_tokens": 74366024.0,
      "step": 38015
    },
    {
      "entropy": 6.010348320007324,
      "epoch": 3.8005198180636777,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003618763783557144,
      "loss": 5.386,
      "mean_token_accuracy": 0.20554914623498916,
      "num_tokens": 74375630.0,
      "step": 38020
    },
    {
      "entropy": 5.8916847705841064,
      "epoch": 3.8010196431249064,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003618434558612755,
      "loss": 5.3798,
      "mean_token_accuracy": 0.20161200761795045,
      "num_tokens": 74385473.0,
      "step": 38025
    },
    {
      "entropy": 5.911258602142334,
      "epoch": 3.801519468186135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00036181053118187757,
      "loss": 5.4349,
      "mean_token_accuracy": 0.19609220176935196,
      "num_tokens": 74395516.0,
      "step": 38030
    },
    {
      "entropy": 5.862637042999268,
      "epoch": 3.8020192932473633,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036177760431834897,
      "loss": 5.3668,
      "mean_token_accuracy": 0.20690566599369048,
      "num_tokens": 74405421.0,
      "step": 38035
    },
    {
      "entropy": 6.031329154968262,
      "epoch": 3.802519118308592,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003617446752715182,
      "loss": 5.5674,
      "mean_token_accuracy": 0.19042708426713945,
      "num_tokens": 74416142.0,
      "step": 38040
    },
    {
      "entropy": 6.003783941268921,
      "epoch": 3.8030189433698207,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036171174404221377,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19637117087841033,
      "num_tokens": 74425943.0,
      "step": 38045
    },
    {
      "entropy": 5.996371936798096,
      "epoch": 3.803518768431049,
      "grad_norm": 1.171875,
      "learning_rate": 0.00036167881063126414,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19713517725467683,
      "num_tokens": 74435088.0,
      "step": 38050
    },
    {
      "entropy": 5.956855821609497,
      "epoch": 3.8040185934922777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003616458750394979,
      "loss": 5.5164,
      "mean_token_accuracy": 0.19077016413211823,
      "num_tokens": 74444922.0,
      "step": 38055
    },
    {
      "entropy": 5.9807350635528564,
      "epoch": 3.8045184185535064,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003616129372677438,
      "loss": 5.4365,
      "mean_token_accuracy": 0.20079980790615082,
      "num_tokens": 74455243.0,
      "step": 38060
    },
    {
      "entropy": 5.994822311401367,
      "epoch": 3.805018243614735,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003615799973168305,
      "loss": 5.5311,
      "mean_token_accuracy": 0.19644793570041658,
      "num_tokens": 74464925.0,
      "step": 38065
    },
    {
      "entropy": 5.957571363449096,
      "epoch": 3.8055180686759633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003615470551875867,
      "loss": 5.4548,
      "mean_token_accuracy": 0.2022285431623459,
      "num_tokens": 74474438.0,
      "step": 38070
    },
    {
      "entropy": 6.032702827453614,
      "epoch": 3.806017893737192,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003615141108808412,
      "loss": 5.4286,
      "mean_token_accuracy": 0.2084379479289055,
      "num_tokens": 74484311.0,
      "step": 38075
    },
    {
      "entropy": 6.0764782428741455,
      "epoch": 3.8065177187984207,
      "grad_norm": 1.109375,
      "learning_rate": 0.000361481164397423,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18805021196603774,
      "num_tokens": 74494841.0,
      "step": 38080
    },
    {
      "entropy": 5.857574510574341,
      "epoch": 3.807017543859649,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036144821573816095,
      "loss": 5.4278,
      "mean_token_accuracy": 0.20219024568796157,
      "num_tokens": 74505354.0,
      "step": 38085
    },
    {
      "entropy": 5.93500337600708,
      "epoch": 3.8075173689208777,
      "grad_norm": 1.28125,
      "learning_rate": 0.00036141526490388406,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20285316705703735,
      "num_tokens": 74514182.0,
      "step": 38090
    },
    {
      "entropy": 6.073058652877807,
      "epoch": 3.8080171939821064,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00036138231189542137,
      "loss": 5.4725,
      "mean_token_accuracy": 0.19572115540504456,
      "num_tokens": 74523824.0,
      "step": 38095
    },
    {
      "entropy": 6.008641099929809,
      "epoch": 3.808517019043335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003613493567136019,
      "loss": 5.5185,
      "mean_token_accuracy": 0.1860068067908287,
      "num_tokens": 74534016.0,
      "step": 38100
    },
    {
      "entropy": 5.887848806381226,
      "epoch": 3.8090168441045633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003613163993592548,
      "loss": 5.3242,
      "mean_token_accuracy": 0.20972476452589034,
      "num_tokens": 74542935.0,
      "step": 38105
    },
    {
      "entropy": 5.892220830917358,
      "epoch": 3.809516669165792,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003612834398332093,
      "loss": 5.4366,
      "mean_token_accuracy": 0.20385946184396744,
      "num_tokens": 74553556.0,
      "step": 38110
    },
    {
      "entropy": 5.92061185836792,
      "epoch": 3.8100164942270203,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003612504781362947,
      "loss": 5.3403,
      "mean_token_accuracy": 0.21431064158678054,
      "num_tokens": 74563691.0,
      "step": 38115
    },
    {
      "entropy": 6.008881092071533,
      "epoch": 3.810516319288249,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003612175142693402,
      "loss": 5.3991,
      "mean_token_accuracy": 0.20427086651325227,
      "num_tokens": 74572512.0,
      "step": 38120
    },
    {
      "entropy": 5.842020463943482,
      "epoch": 3.8110161443494777,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003611845482331753,
      "loss": 5.3099,
      "mean_token_accuracy": 0.21523078829050063,
      "num_tokens": 74582542.0,
      "step": 38125
    },
    {
      "entropy": 5.944562816619873,
      "epoch": 3.8115159694107064,
      "grad_norm": 1.09375,
      "learning_rate": 0.00036115158002862933,
      "loss": 5.396,
      "mean_token_accuracy": 0.19965545684099198,
      "num_tokens": 74591367.0,
      "step": 38130
    },
    {
      "entropy": 5.959423589706421,
      "epoch": 3.812015794471935,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003611186096565318,
      "loss": 5.4685,
      "mean_token_accuracy": 0.20011759847402572,
      "num_tokens": 74600732.0,
      "step": 38135
    },
    {
      "entropy": 5.995504140853882,
      "epoch": 3.8125156195331633,
      "grad_norm": 1.125,
      "learning_rate": 0.0003610856371177122,
      "loss": 5.4519,
      "mean_token_accuracy": 0.1977409988641739,
      "num_tokens": 74610220.0,
      "step": 38140
    },
    {
      "entropy": 5.983244895935059,
      "epoch": 3.813015444594392,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00036105266241300005,
      "loss": 5.4117,
      "mean_token_accuracy": 0.20393800884485244,
      "num_tokens": 74619545.0,
      "step": 38145
    },
    {
      "entropy": 5.974001550674439,
      "epoch": 3.8135152696556203,
      "grad_norm": 1.125,
      "learning_rate": 0.0003610196855432251,
      "loss": 5.4327,
      "mean_token_accuracy": 0.20452870279550553,
      "num_tokens": 74628942.0,
      "step": 38150
    },
    {
      "entropy": 5.947923088073731,
      "epoch": 3.814015094716849,
      "grad_norm": 1.109375,
      "learning_rate": 0.00036098670650921706,
      "loss": 5.4267,
      "mean_token_accuracy": 0.201309996843338,
      "num_tokens": 74639774.0,
      "step": 38155
    },
    {
      "entropy": 5.927144193649292,
      "epoch": 3.8145149197780777,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003609537253118056,
      "loss": 5.4126,
      "mean_token_accuracy": 0.205323426425457,
      "num_tokens": 74649824.0,
      "step": 38160
    },
    {
      "entropy": 5.998037910461425,
      "epoch": 3.8150147448393064,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003609207419518204,
      "loss": 5.4444,
      "mean_token_accuracy": 0.20018993616104125,
      "num_tokens": 74660021.0,
      "step": 38165
    },
    {
      "entropy": 5.934172916412353,
      "epoch": 3.815514569900535,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003608877564300915,
      "loss": 5.4661,
      "mean_token_accuracy": 0.20404136776924134,
      "num_tokens": 74669545.0,
      "step": 38170
    },
    {
      "entropy": 5.965950679779053,
      "epoch": 3.8160143949617633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003608547687474488,
      "loss": 5.4061,
      "mean_token_accuracy": 0.19343761205673218,
      "num_tokens": 74678353.0,
      "step": 38175
    },
    {
      "entropy": 5.898930835723877,
      "epoch": 3.816514220022992,
      "grad_norm": 1.125,
      "learning_rate": 0.00036082177890472224,
      "loss": 5.3649,
      "mean_token_accuracy": 0.21076193749904631,
      "num_tokens": 74687559.0,
      "step": 38180
    },
    {
      "entropy": 5.933311033248901,
      "epoch": 3.8170140450842203,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036078878690274177,
      "loss": 5.5203,
      "mean_token_accuracy": 0.19398004710674285,
      "num_tokens": 74698288.0,
      "step": 38185
    },
    {
      "entropy": 5.958037376403809,
      "epoch": 3.817513870145449,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003607557927423374,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20618181228637694,
      "num_tokens": 74707671.0,
      "step": 38190
    },
    {
      "entropy": 6.033723449707031,
      "epoch": 3.8180136952066777,
      "grad_norm": 1.015625,
      "learning_rate": 0.00036072279642433946,
      "loss": 5.5157,
      "mean_token_accuracy": 0.19446359425783158,
      "num_tokens": 74717492.0,
      "step": 38195
    },
    {
      "entropy": 5.96299376487732,
      "epoch": 3.8185135202679064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036068979794957797,
      "loss": 5.3907,
      "mean_token_accuracy": 0.20786735117435456,
      "num_tokens": 74727711.0,
      "step": 38200
    },
    {
      "entropy": 5.929358243942261,
      "epoch": 3.819013345329135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003606567973188833,
      "loss": 5.4179,
      "mean_token_accuracy": 0.19878582656383514,
      "num_tokens": 74738126.0,
      "step": 38205
    },
    {
      "entropy": 6.0475812435150145,
      "epoch": 3.8195131703903633,
      "grad_norm": 1.203125,
      "learning_rate": 0.00036062379453308556,
      "loss": 5.5291,
      "mean_token_accuracy": 0.18911541998386383,
      "num_tokens": 74748262.0,
      "step": 38210
    },
    {
      "entropy": 5.946753740310669,
      "epoch": 3.820012995451592,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00036059078959301515,
      "loss": 5.3676,
      "mean_token_accuracy": 0.20717428177595137,
      "num_tokens": 74758161.0,
      "step": 38215
    },
    {
      "entropy": 5.886636590957641,
      "epoch": 3.8205128205128203,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003605577824995026,
      "loss": 5.3511,
      "mean_token_accuracy": 0.20993249565362931,
      "num_tokens": 74767970.0,
      "step": 38220
    },
    {
      "entropy": 6.0056596279144285,
      "epoch": 3.821012645574049,
      "grad_norm": 1.140625,
      "learning_rate": 0.00036052477325337824,
      "loss": 5.5111,
      "mean_token_accuracy": 0.19141466170549393,
      "num_tokens": 74779410.0,
      "step": 38225
    },
    {
      "entropy": 6.001379871368409,
      "epoch": 3.8215124706352777,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00036049176185547254,
      "loss": 5.5187,
      "mean_token_accuracy": 0.18876935094594954,
      "num_tokens": 74788200.0,
      "step": 38230
    },
    {
      "entropy": 5.957904243469239,
      "epoch": 3.8220122956965064,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00036045874830661613,
      "loss": 5.4972,
      "mean_token_accuracy": 0.1994644120335579,
      "num_tokens": 74797776.0,
      "step": 38235
    },
    {
      "entropy": 5.977708959579468,
      "epoch": 3.822512120757735,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003604257326076397,
      "loss": 5.463,
      "mean_token_accuracy": 0.19659383445978165,
      "num_tokens": 74807103.0,
      "step": 38240
    },
    {
      "entropy": 5.933729076385498,
      "epoch": 3.8230119458189633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00036039271475937366,
      "loss": 5.3717,
      "mean_token_accuracy": 0.20768076032400132,
      "num_tokens": 74816003.0,
      "step": 38245
    },
    {
      "entropy": 5.949706411361694,
      "epoch": 3.823511770880192,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000360359694762649,
      "loss": 5.3984,
      "mean_token_accuracy": 0.19996250718832015,
      "num_tokens": 74825387.0,
      "step": 38250
    },
    {
      "entropy": 5.960859775543213,
      "epoch": 3.8240115959414203,
      "grad_norm": 1.125,
      "learning_rate": 0.0003603266726182963,
      "loss": 5.4663,
      "mean_token_accuracy": 0.1996582880616188,
      "num_tokens": 74833909.0,
      "step": 38255
    },
    {
      "entropy": 6.041090774536133,
      "epoch": 3.824511421002649,
      "grad_norm": 1.125,
      "learning_rate": 0.0003602936483271465,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18408622741699218,
      "num_tokens": 74843685.0,
      "step": 38260
    },
    {
      "entropy": 5.952756929397583,
      "epoch": 3.8250112460638777,
      "grad_norm": 1.15625,
      "learning_rate": 0.00036026062189003046,
      "loss": 5.3909,
      "mean_token_accuracy": 0.20834011137485503,
      "num_tokens": 74852376.0,
      "step": 38265
    },
    {
      "entropy": 5.980066871643066,
      "epoch": 3.8255110711251064,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003602275933077792,
      "loss": 5.4497,
      "mean_token_accuracy": 0.2020851194858551,
      "num_tokens": 74863057.0,
      "step": 38270
    },
    {
      "entropy": 6.009368276596069,
      "epoch": 3.826010896186335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00036019456258122346,
      "loss": 5.4869,
      "mean_token_accuracy": 0.19621894359588624,
      "num_tokens": 74873337.0,
      "step": 38275
    },
    {
      "entropy": 5.901997232437134,
      "epoch": 3.8265107212475633,
      "grad_norm": 1.078125,
      "learning_rate": 0.00036016152971119457,
      "loss": 5.3114,
      "mean_token_accuracy": 0.20184279680252076,
      "num_tokens": 74882994.0,
      "step": 38280
    },
    {
      "entropy": 5.946200037002564,
      "epoch": 3.827010546308792,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003601284946985235,
      "loss": 5.4917,
      "mean_token_accuracy": 0.1914817675948143,
      "num_tokens": 74893490.0,
      "step": 38285
    },
    {
      "entropy": 5.952967405319214,
      "epoch": 3.8275103713700203,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003600954575440414,
      "loss": 5.4129,
      "mean_token_accuracy": 0.20609336644411086,
      "num_tokens": 74903152.0,
      "step": 38290
    },
    {
      "entropy": 5.944463109970092,
      "epoch": 3.828010196431249,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003600624182485794,
      "loss": 5.379,
      "mean_token_accuracy": 0.2034478396177292,
      "num_tokens": 74913146.0,
      "step": 38295
    },
    {
      "entropy": 5.967310810089112,
      "epoch": 3.8285100214924777,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003600293768129689,
      "loss": 5.5809,
      "mean_token_accuracy": 0.1868455857038498,
      "num_tokens": 74923313.0,
      "step": 38300
    },
    {
      "entropy": 5.9606428146362305,
      "epoch": 3.8290098465537064,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035999633323804123,
      "loss": 5.491,
      "mean_token_accuracy": 0.19532238245010375,
      "num_tokens": 74933339.0,
      "step": 38305
    },
    {
      "entropy": 6.046218824386597,
      "epoch": 3.8295096716149346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035996328752462757,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1937365159392357,
      "num_tokens": 74943047.0,
      "step": 38310
    },
    {
      "entropy": 6.098005628585815,
      "epoch": 3.8300094966761633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035993023967355954,
      "loss": 5.5932,
      "mean_token_accuracy": 0.19222526848316193,
      "num_tokens": 74952040.0,
      "step": 38315
    },
    {
      "entropy": 5.9568479537963865,
      "epoch": 3.830509321737392,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035989718968566846,
      "loss": 5.4368,
      "mean_token_accuracy": 0.19647896587848662,
      "num_tokens": 74960848.0,
      "step": 38320
    },
    {
      "entropy": 5.869131422042846,
      "epoch": 3.8310091467986203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000359864137561786,
      "loss": 5.3649,
      "mean_token_accuracy": 0.1988675519824028,
      "num_tokens": 74970357.0,
      "step": 38325
    },
    {
      "entropy": 6.0585020065307615,
      "epoch": 3.831508971859849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003598310833027437,
      "loss": 5.4905,
      "mean_token_accuracy": 0.1867958128452301,
      "num_tokens": 74979433.0,
      "step": 38330
    },
    {
      "entropy": 5.95999813079834,
      "epoch": 3.8320087969210777,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035979802690937315,
      "loss": 5.3293,
      "mean_token_accuracy": 0.20496200621128083,
      "num_tokens": 74989733.0,
      "step": 38335
    },
    {
      "entropy": 5.904738712310791,
      "epoch": 3.8325086219823064,
      "grad_norm": 1.125,
      "learning_rate": 0.00035976496838250607,
      "loss": 5.3694,
      "mean_token_accuracy": 0.199457286298275,
      "num_tokens": 75000139.0,
      "step": 38340
    },
    {
      "entropy": 6.047831726074219,
      "epoch": 3.8330084470435346,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035973190772297417,
      "loss": 5.5759,
      "mean_token_accuracy": 0.1882543981075287,
      "num_tokens": 75009165.0,
      "step": 38345
    },
    {
      "entropy": 5.960987138748169,
      "epoch": 3.8335082721047633,
      "grad_norm": 1.125,
      "learning_rate": 0.00035969884493160925,
      "loss": 5.4584,
      "mean_token_accuracy": 0.20283973962068558,
      "num_tokens": 75019077.0,
      "step": 38350
    },
    {
      "entropy": 5.882628536224365,
      "epoch": 3.834008097165992,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035966578000924324,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20661706179380418,
      "num_tokens": 75029351.0,
      "step": 38355
    },
    {
      "entropy": 6.000039052963257,
      "epoch": 3.8345079222272203,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003596327129567079,
      "loss": 5.4272,
      "mean_token_accuracy": 0.2034045308828354,
      "num_tokens": 75038542.0,
      "step": 38360
    },
    {
      "entropy": 6.002588987350464,
      "epoch": 3.835007747288449,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035959964377483536,
      "loss": 5.4656,
      "mean_token_accuracy": 0.19892458021640777,
      "num_tokens": 75049288.0,
      "step": 38365
    },
    {
      "entropy": 5.941528701782227,
      "epoch": 3.8355075723496777,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003595665724644576,
      "loss": 5.4869,
      "mean_token_accuracy": 0.19531770795583725,
      "num_tokens": 75059595.0,
      "step": 38370
    },
    {
      "entropy": 5.96993236541748,
      "epoch": 3.8360073974109063,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003595334990264065,
      "loss": 5.4504,
      "mean_token_accuracy": 0.1985209345817566,
      "num_tokens": 75069566.0,
      "step": 38375
    },
    {
      "entropy": 6.044158029556274,
      "epoch": 3.8365072224721346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003595004234615144,
      "loss": 5.5366,
      "mean_token_accuracy": 0.1844862386584282,
      "num_tokens": 75079555.0,
      "step": 38380
    },
    {
      "entropy": 6.033324432373047,
      "epoch": 3.8370070475333633,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035946734577061345,
      "loss": 5.4125,
      "mean_token_accuracy": 0.19789723306894302,
      "num_tokens": 75088703.0,
      "step": 38385
    },
    {
      "entropy": 5.9714960098266605,
      "epoch": 3.837506872594592,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003594342659545357,
      "loss": 5.5135,
      "mean_token_accuracy": 0.19524967670440674,
      "num_tokens": 75098328.0,
      "step": 38390
    },
    {
      "entropy": 5.9188148975372314,
      "epoch": 3.8380066976558203,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003594011840141136,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20020497888326644,
      "num_tokens": 75108248.0,
      "step": 38395
    },
    {
      "entropy": 6.002490854263305,
      "epoch": 3.838506522717049,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035936809995017936,
      "loss": 5.5027,
      "mean_token_accuracy": 0.19415270239114762,
      "num_tokens": 75116876.0,
      "step": 38400
    },
    {
      "entropy": 5.936320114135742,
      "epoch": 3.8390063477782777,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035933501376356555,
      "loss": 5.4182,
      "mean_token_accuracy": 0.20387574285268784,
      "num_tokens": 75126977.0,
      "step": 38405
    },
    {
      "entropy": 5.9967621803283695,
      "epoch": 3.8395061728395063,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003593019254551045,
      "loss": 5.3745,
      "mean_token_accuracy": 0.2007889524102211,
      "num_tokens": 75136565.0,
      "step": 38410
    },
    {
      "entropy": 5.989413166046143,
      "epoch": 3.8400059979007346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035926883502562855,
      "loss": 5.4975,
      "mean_token_accuracy": 0.1869767650961876,
      "num_tokens": 75146438.0,
      "step": 38415
    },
    {
      "entropy": 5.996684122085571,
      "epoch": 3.8405058229619633,
      "grad_norm": 1.125,
      "learning_rate": 0.00035923574247597053,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1955714002251625,
      "num_tokens": 75155875.0,
      "step": 38420
    },
    {
      "entropy": 5.880424928665161,
      "epoch": 3.841005648023192,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035920264780696284,
      "loss": 5.3063,
      "mean_token_accuracy": 0.2008490115404129,
      "num_tokens": 75164875.0,
      "step": 38425
    },
    {
      "entropy": 5.940051507949829,
      "epoch": 3.8415054730844203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003591695510194382,
      "loss": 5.3575,
      "mean_token_accuracy": 0.20751779675483703,
      "num_tokens": 75174384.0,
      "step": 38430
    },
    {
      "entropy": 5.934393978118896,
      "epoch": 3.842005298145649,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003591364521142293,
      "loss": 5.336,
      "mean_token_accuracy": 0.20246125459671022,
      "num_tokens": 75183766.0,
      "step": 38435
    },
    {
      "entropy": 5.914346742630005,
      "epoch": 3.8425051232068776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00035910335109216895,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19887708127498627,
      "num_tokens": 75193453.0,
      "step": 38440
    },
    {
      "entropy": 5.9687034606933596,
      "epoch": 3.8430049482681063,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035907024795409,
      "loss": 5.4442,
      "mean_token_accuracy": 0.20340796560049057,
      "num_tokens": 75203193.0,
      "step": 38445
    },
    {
      "entropy": 5.923344707489013,
      "epoch": 3.8435047733293346,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035903714270082515,
      "loss": 5.3805,
      "mean_token_accuracy": 0.20838464498519899,
      "num_tokens": 75212465.0,
      "step": 38450
    },
    {
      "entropy": 6.043428087234497,
      "epoch": 3.8440045983905633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003590040353332074,
      "loss": 5.5637,
      "mean_token_accuracy": 0.19517429769039155,
      "num_tokens": 75221529.0,
      "step": 38455
    },
    {
      "entropy": 5.9255860328674315,
      "epoch": 3.844504423451792,
      "grad_norm": 1.40625,
      "learning_rate": 0.0003589709258520698,
      "loss": 5.3916,
      "mean_token_accuracy": 0.20889589786529542,
      "num_tokens": 75231676.0,
      "step": 38460
    },
    {
      "entropy": 5.922569799423218,
      "epoch": 3.8450042485130203,
      "grad_norm": 1.0,
      "learning_rate": 0.0003589378142582453,
      "loss": 5.4378,
      "mean_token_accuracy": 0.2045730471611023,
      "num_tokens": 75241420.0,
      "step": 38465
    },
    {
      "entropy": 5.9569860935211185,
      "epoch": 3.845504073574249,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035890470055256697,
      "loss": 5.4776,
      "mean_token_accuracy": 0.19273646473884581,
      "num_tokens": 75250725.0,
      "step": 38470
    },
    {
      "entropy": 5.9195472240448,
      "epoch": 3.8460038986354776,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035887158473586793,
      "loss": 5.2536,
      "mean_token_accuracy": 0.21012057065963746,
      "num_tokens": 75260915.0,
      "step": 38475
    },
    {
      "entropy": 5.878107738494873,
      "epoch": 3.8465037236967063,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003588384668089815,
      "loss": 5.3582,
      "mean_token_accuracy": 0.20789642184972762,
      "num_tokens": 75270803.0,
      "step": 38480
    },
    {
      "entropy": 5.922365188598633,
      "epoch": 3.8470035487579346,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00035880534677274075,
      "loss": 5.3722,
      "mean_token_accuracy": 0.19717672616243362,
      "num_tokens": 75280323.0,
      "step": 38485
    },
    {
      "entropy": 5.92101526260376,
      "epoch": 3.8475033738191633,
      "grad_norm": 1.078125,
      "learning_rate": 0.000358772224627979,
      "loss": 5.3974,
      "mean_token_accuracy": 0.19929681420326234,
      "num_tokens": 75290115.0,
      "step": 38490
    },
    {
      "entropy": 5.910966300964356,
      "epoch": 3.848003198880392,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035873910037552973,
      "loss": 5.4204,
      "mean_token_accuracy": 0.20620449036359786,
      "num_tokens": 75300868.0,
      "step": 38495
    },
    {
      "entropy": 5.981737041473389,
      "epoch": 3.8485030239416202,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003587059740162262,
      "loss": 5.4496,
      "mean_token_accuracy": 0.19407575875520705,
      "num_tokens": 75310859.0,
      "step": 38500
    },
    {
      "entropy": 6.058179950714111,
      "epoch": 3.849002849002849,
      "grad_norm": 1.125,
      "learning_rate": 0.0003586728455509019,
      "loss": 5.4524,
      "mean_token_accuracy": 0.19905563890933992,
      "num_tokens": 75320682.0,
      "step": 38505
    },
    {
      "entropy": 6.096875429153442,
      "epoch": 3.8495026740640776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035863971498039033,
      "loss": 5.5111,
      "mean_token_accuracy": 0.20070581436157225,
      "num_tokens": 75330466.0,
      "step": 38510
    },
    {
      "entropy": 5.9178431034088135,
      "epoch": 3.8500024991253063,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035860658230552506,
      "loss": 5.3948,
      "mean_token_accuracy": 0.205663001537323,
      "num_tokens": 75339650.0,
      "step": 38515
    },
    {
      "entropy": 5.9313990592956545,
      "epoch": 3.8505023241865346,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003585734475271397,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19396144449710845,
      "num_tokens": 75350093.0,
      "step": 38520
    },
    {
      "entropy": 6.051080226898193,
      "epoch": 3.8510021492477633,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003585403106460679,
      "loss": 5.5069,
      "mean_token_accuracy": 0.1900491163134575,
      "num_tokens": 75358433.0,
      "step": 38525
    },
    {
      "entropy": 5.926530742645264,
      "epoch": 3.851501974308992,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0003585071716631433,
      "loss": 5.2643,
      "mean_token_accuracy": 0.22046216875314711,
      "num_tokens": 75366538.0,
      "step": 38530
    },
    {
      "entropy": 5.938153076171875,
      "epoch": 3.8520017993702202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003584740305791999,
      "loss": 5.3984,
      "mean_token_accuracy": 0.2042742669582367,
      "num_tokens": 75376314.0,
      "step": 38535
    },
    {
      "entropy": 6.016844129562378,
      "epoch": 3.852501624431449,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003584408873950713,
      "loss": 5.4302,
      "mean_token_accuracy": 0.20204218924045564,
      "num_tokens": 75386097.0,
      "step": 38540
    },
    {
      "entropy": 5.953931665420532,
      "epoch": 3.8530014494926776,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035840774211159146,
      "loss": 5.3923,
      "mean_token_accuracy": 0.2050534263253212,
      "num_tokens": 75396407.0,
      "step": 38545
    },
    {
      "entropy": 5.946669816970825,
      "epoch": 3.8535012745539063,
      "grad_norm": 1.125,
      "learning_rate": 0.00035837459472959425,
      "loss": 5.4329,
      "mean_token_accuracy": 0.19453460723161697,
      "num_tokens": 75406119.0,
      "step": 38550
    },
    {
      "entropy": 5.888563585281372,
      "epoch": 3.8540010996151346,
      "grad_norm": 1.125,
      "learning_rate": 0.0003583414452499137,
      "loss": 5.3469,
      "mean_token_accuracy": 0.2087235078215599,
      "num_tokens": 75415642.0,
      "step": 38555
    },
    {
      "entropy": 6.0588150978088375,
      "epoch": 3.8545009246763633,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035830829367338384,
      "loss": 5.6169,
      "mean_token_accuracy": 0.1876517117023468,
      "num_tokens": 75426005.0,
      "step": 38560
    },
    {
      "entropy": 6.031765460968018,
      "epoch": 3.855000749737592,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003582751400008388,
      "loss": 5.3967,
      "mean_token_accuracy": 0.20136492401361467,
      "num_tokens": 75435641.0,
      "step": 38565
    },
    {
      "entropy": 5.860448932647705,
      "epoch": 3.8555005747988202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035824198423311263,
      "loss": 5.304,
      "mean_token_accuracy": 0.21251945495605468,
      "num_tokens": 75445329.0,
      "step": 38570
    },
    {
      "entropy": 5.951228952407837,
      "epoch": 3.856000399860049,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003582088263710396,
      "loss": 5.4641,
      "mean_token_accuracy": 0.20463106781244278,
      "num_tokens": 75455247.0,
      "step": 38575
    },
    {
      "entropy": 5.961295175552368,
      "epoch": 3.8565002249212776,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035817566641545376,
      "loss": 5.3987,
      "mean_token_accuracy": 0.20371610373258592,
      "num_tokens": 75465986.0,
      "step": 38580
    },
    {
      "entropy": 5.987462568283081,
      "epoch": 3.8570000499825063,
      "grad_norm": 1.25,
      "learning_rate": 0.0003581425043671897,
      "loss": 5.4374,
      "mean_token_accuracy": 0.1975521668791771,
      "num_tokens": 75475367.0,
      "step": 38585
    },
    {
      "entropy": 6.004096460342407,
      "epoch": 3.8574998750437346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003581093402270816,
      "loss": 5.3781,
      "mean_token_accuracy": 0.203182552754879,
      "num_tokens": 75484531.0,
      "step": 38590
    },
    {
      "entropy": 5.9783164978027346,
      "epoch": 3.8579997001049633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035807617399596383,
      "loss": 5.4493,
      "mean_token_accuracy": 0.19737277925014496,
      "num_tokens": 75493740.0,
      "step": 38595
    },
    {
      "entropy": 5.9786803245544435,
      "epoch": 3.858499525166192,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003580430056746709,
      "loss": 5.477,
      "mean_token_accuracy": 0.20043650269508362,
      "num_tokens": 75503225.0,
      "step": 38600
    },
    {
      "entropy": 6.0849385261535645,
      "epoch": 3.8589993502274202,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003580098352640374,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18872447460889816,
      "num_tokens": 75513229.0,
      "step": 38605
    },
    {
      "entropy": 5.988102197647095,
      "epoch": 3.859499175288649,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035797666276489773,
      "loss": 5.3772,
      "mean_token_accuracy": 0.20201382786035538,
      "num_tokens": 75523299.0,
      "step": 38610
    },
    {
      "entropy": 5.984303188323975,
      "epoch": 3.8599990003498776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035794348817808666,
      "loss": 5.5445,
      "mean_token_accuracy": 0.18907174468040466,
      "num_tokens": 75532740.0,
      "step": 38615
    },
    {
      "entropy": 5.944781112670898,
      "epoch": 3.8604988254111063,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035791031150443864,
      "loss": 5.4166,
      "mean_token_accuracy": 0.20262534618377687,
      "num_tokens": 75543140.0,
      "step": 38620
    },
    {
      "entropy": 5.964598321914673,
      "epoch": 3.8609986504723346,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003578771327447886,
      "loss": 5.3804,
      "mean_token_accuracy": 0.20643334537744523,
      "num_tokens": 75552482.0,
      "step": 38625
    },
    {
      "entropy": 5.981219482421875,
      "epoch": 3.8614984755335633,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003578439518999712,
      "loss": 5.4287,
      "mean_token_accuracy": 0.20115817040205003,
      "num_tokens": 75561564.0,
      "step": 38630
    },
    {
      "entropy": 5.968373584747314,
      "epoch": 3.861998300594792,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035781076897082134,
      "loss": 5.492,
      "mean_token_accuracy": 0.19356848746538163,
      "num_tokens": 75570352.0,
      "step": 38635
    },
    {
      "entropy": 6.018442106246948,
      "epoch": 3.8624981256560202,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035777758395817374,
      "loss": 5.5321,
      "mean_token_accuracy": 0.19247304946184157,
      "num_tokens": 75579977.0,
      "step": 38640
    },
    {
      "entropy": 5.912220001220703,
      "epoch": 3.862997950717249,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003577443968628635,
      "loss": 5.3307,
      "mean_token_accuracy": 0.20915856659412385,
      "num_tokens": 75589926.0,
      "step": 38645
    },
    {
      "entropy": 5.968596458435059,
      "epoch": 3.8634977757784776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003577112076857256,
      "loss": 5.3181,
      "mean_token_accuracy": 0.20964466631412507,
      "num_tokens": 75600432.0,
      "step": 38650
    },
    {
      "entropy": 5.928662967681885,
      "epoch": 3.8639976008397063,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035767801642759485,
      "loss": 5.382,
      "mean_token_accuracy": 0.2005261778831482,
      "num_tokens": 75610545.0,
      "step": 38655
    },
    {
      "entropy": 5.953776073455811,
      "epoch": 3.8644974259009346,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035764482308930657,
      "loss": 5.462,
      "mean_token_accuracy": 0.19559051096439362,
      "num_tokens": 75620430.0,
      "step": 38660
    },
    {
      "entropy": 5.731444215774536,
      "epoch": 3.8649972509621633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003576116276716958,
      "loss": 5.1911,
      "mean_token_accuracy": 0.24075427204370498,
      "num_tokens": 75631597.0,
      "step": 38665
    },
    {
      "entropy": 6.062215566635132,
      "epoch": 3.8654970760233915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003575784301755977,
      "loss": 5.5013,
      "mean_token_accuracy": 0.19016134440898896,
      "num_tokens": 75641201.0,
      "step": 38670
    },
    {
      "entropy": 6.08081636428833,
      "epoch": 3.8659969010846202,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035754523060184753,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19471445679664612,
      "num_tokens": 75650465.0,
      "step": 38675
    },
    {
      "entropy": 5.970568418502808,
      "epoch": 3.866496726145849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035751202895128065,
      "loss": 5.4298,
      "mean_token_accuracy": 0.19535891711711884,
      "num_tokens": 75659258.0,
      "step": 38680
    },
    {
      "entropy": 5.972808408737182,
      "epoch": 3.8669965512070776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003574788252247324,
      "loss": 5.4613,
      "mean_token_accuracy": 0.19644760191440583,
      "num_tokens": 75669103.0,
      "step": 38685
    },
    {
      "entropy": 6.0562450885772705,
      "epoch": 3.8674963762683063,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035744561942303805,
      "loss": 5.5445,
      "mean_token_accuracy": 0.18840801864862441,
      "num_tokens": 75679651.0,
      "step": 38690
    },
    {
      "entropy": 5.957889652252197,
      "epoch": 3.8679962013295346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003574124115470332,
      "loss": 5.3224,
      "mean_token_accuracy": 0.20373619347810745,
      "num_tokens": 75689668.0,
      "step": 38695
    },
    {
      "entropy": 5.929937648773193,
      "epoch": 3.8684960263907633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035737920159755317,
      "loss": 5.3888,
      "mean_token_accuracy": 0.2010420471429825,
      "num_tokens": 75698617.0,
      "step": 38700
    },
    {
      "entropy": 5.937395429611206,
      "epoch": 3.8689958514519915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003573459895754337,
      "loss": 5.3913,
      "mean_token_accuracy": 0.19844850301742553,
      "num_tokens": 75707526.0,
      "step": 38705
    },
    {
      "entropy": 5.981868076324463,
      "epoch": 3.8694956765132202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003573127754815103,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19063973426818848,
      "num_tokens": 75718251.0,
      "step": 38710
    },
    {
      "entropy": 5.9318764209747314,
      "epoch": 3.869995501574449,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035727955931661863,
      "loss": 5.3312,
      "mean_token_accuracy": 0.2080619141459465,
      "num_tokens": 75729567.0,
      "step": 38715
    },
    {
      "entropy": 6.032916069030762,
      "epoch": 3.8704953266356776,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035724634108159453,
      "loss": 5.5191,
      "mean_token_accuracy": 0.19116322249174117,
      "num_tokens": 75738422.0,
      "step": 38720
    },
    {
      "entropy": 5.957482385635376,
      "epoch": 3.8709951516969063,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003572131207772736,
      "loss": 5.468,
      "mean_token_accuracy": 0.1996527135372162,
      "num_tokens": 75748247.0,
      "step": 38725
    },
    {
      "entropy": 6.003903484344482,
      "epoch": 3.8714949767581346,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003571798984044916,
      "loss": 5.4987,
      "mean_token_accuracy": 0.19457464069128036,
      "num_tokens": 75757169.0,
      "step": 38730
    },
    {
      "entropy": 6.065610599517822,
      "epoch": 3.8719948018193633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035714667396408455,
      "loss": 5.5221,
      "mean_token_accuracy": 0.1945802316069603,
      "num_tokens": 75766062.0,
      "step": 38735
    },
    {
      "entropy": 5.954581832885742,
      "epoch": 3.8724946268805915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003571134474568883,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20898675918579102,
      "num_tokens": 75776049.0,
      "step": 38740
    },
    {
      "entropy": 5.969726085662842,
      "epoch": 3.8729944519418202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000357080218883739,
      "loss": 5.4441,
      "mean_token_accuracy": 0.19873103350400925,
      "num_tokens": 75786917.0,
      "step": 38745
    },
    {
      "entropy": 5.978623104095459,
      "epoch": 3.873494277003049,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003570469882454724,
      "loss": 5.3751,
      "mean_token_accuracy": 0.2074020951986313,
      "num_tokens": 75796393.0,
      "step": 38750
    },
    {
      "entropy": 6.009975099563599,
      "epoch": 3.8739941020642776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035701375554292477,
      "loss": 5.4396,
      "mean_token_accuracy": 0.20304720252752304,
      "num_tokens": 75806522.0,
      "step": 38755
    },
    {
      "entropy": 6.075493049621582,
      "epoch": 3.8744939271255063,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00035698052077693206,
      "loss": 5.5022,
      "mean_token_accuracy": 0.19295008182525636,
      "num_tokens": 75817072.0,
      "step": 38760
    },
    {
      "entropy": 6.041600513458252,
      "epoch": 3.8749937521867346,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003569472839483305,
      "loss": 5.489,
      "mean_token_accuracy": 0.1950263872742653,
      "num_tokens": 75827858.0,
      "step": 38765
    },
    {
      "entropy": 6.005127716064453,
      "epoch": 3.8754935772479633,
      "grad_norm": 1.3125,
      "learning_rate": 0.00035691404505795645,
      "loss": 5.4427,
      "mean_token_accuracy": 0.20161572396755217,
      "num_tokens": 75837256.0,
      "step": 38770
    },
    {
      "entropy": 5.962048435211182,
      "epoch": 3.8759934023091915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003568808041066461,
      "loss": 5.3951,
      "mean_token_accuracy": 0.20073340237140655,
      "num_tokens": 75846853.0,
      "step": 38775
    },
    {
      "entropy": 5.932744073867798,
      "epoch": 3.8764932273704202,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003568475610952357,
      "loss": 5.3996,
      "mean_token_accuracy": 0.19305374324321747,
      "num_tokens": 75856843.0,
      "step": 38780
    },
    {
      "entropy": 6.00246114730835,
      "epoch": 3.876993052431649,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035681431602456183,
      "loss": 5.4496,
      "mean_token_accuracy": 0.20025357753038406,
      "num_tokens": 75866682.0,
      "step": 38785
    },
    {
      "entropy": 5.989636373519898,
      "epoch": 3.8774928774928776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035678106889546087,
      "loss": 5.4837,
      "mean_token_accuracy": 0.1931480050086975,
      "num_tokens": 75876211.0,
      "step": 38790
    },
    {
      "entropy": 6.073602819442749,
      "epoch": 3.8779927025541063,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035674781970876905,
      "loss": 5.5113,
      "mean_token_accuracy": 0.19166806042194368,
      "num_tokens": 75885580.0,
      "step": 38795
    },
    {
      "entropy": 5.970753002166748,
      "epoch": 3.8784925276153346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003567145684653233,
      "loss": 5.3915,
      "mean_token_accuracy": 0.200725981593132,
      "num_tokens": 75894831.0,
      "step": 38800
    },
    {
      "entropy": 5.942077589035034,
      "epoch": 3.8789923526765633,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035668131516596,
      "loss": 5.449,
      "mean_token_accuracy": 0.19148667752742768,
      "num_tokens": 75905773.0,
      "step": 38805
    },
    {
      "entropy": 6.0158298969268795,
      "epoch": 3.8794921777377915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035664805981151576,
      "loss": 5.4645,
      "mean_token_accuracy": 0.19757121354341506,
      "num_tokens": 75916264.0,
      "step": 38810
    },
    {
      "entropy": 6.0389947414398195,
      "epoch": 3.8799920027990202,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003566148024028274,
      "loss": 5.5347,
      "mean_token_accuracy": 0.1941233441233635,
      "num_tokens": 75925602.0,
      "step": 38815
    },
    {
      "entropy": 5.929768848419189,
      "epoch": 3.880491827860249,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035658154294073157,
      "loss": 5.3725,
      "mean_token_accuracy": 0.20647752285003662,
      "num_tokens": 75934027.0,
      "step": 38820
    },
    {
      "entropy": 5.929457902908325,
      "epoch": 3.8809916529214776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035654828142606517,
      "loss": 5.4231,
      "mean_token_accuracy": 0.20309561491012573,
      "num_tokens": 75943076.0,
      "step": 38825
    },
    {
      "entropy": 6.032755422592163,
      "epoch": 3.8814914779827063,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000356515017859665,
      "loss": 5.5389,
      "mean_token_accuracy": 0.1859097495675087,
      "num_tokens": 75953569.0,
      "step": 38830
    },
    {
      "entropy": 5.9392718315124515,
      "epoch": 3.8819913030439346,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00035648175224236797,
      "loss": 5.4087,
      "mean_token_accuracy": 0.19570448249578476,
      "num_tokens": 75963289.0,
      "step": 38835
    },
    {
      "entropy": 5.9179931640625,
      "epoch": 3.8824911281051633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035644848457501096,
      "loss": 5.3359,
      "mean_token_accuracy": 0.20788282603025438,
      "num_tokens": 75973521.0,
      "step": 38840
    },
    {
      "entropy": 6.044767999649048,
      "epoch": 3.8829909531663915,
      "grad_norm": 1.125,
      "learning_rate": 0.000356415214858431,
      "loss": 5.5741,
      "mean_token_accuracy": 0.18524640798568726,
      "num_tokens": 75984598.0,
      "step": 38845
    },
    {
      "entropy": 6.005588245391846,
      "epoch": 3.8834907782276202,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003563819430934652,
      "loss": 5.4351,
      "mean_token_accuracy": 0.20130767673254013,
      "num_tokens": 75994178.0,
      "step": 38850
    },
    {
      "entropy": 5.996833372116089,
      "epoch": 3.883990603288849,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003563486692809508,
      "loss": 5.4378,
      "mean_token_accuracy": 0.20473249554634093,
      "num_tokens": 76003848.0,
      "step": 38855
    },
    {
      "entropy": 6.0312402725219725,
      "epoch": 3.8844904283500776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003563153934217246,
      "loss": 5.5498,
      "mean_token_accuracy": 0.18659426718950273,
      "num_tokens": 76014385.0,
      "step": 38860
    },
    {
      "entropy": 5.919492864608765,
      "epoch": 3.884990253411306,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003562821155166242,
      "loss": 5.3384,
      "mean_token_accuracy": 0.1988813042640686,
      "num_tokens": 76022997.0,
      "step": 38865
    },
    {
      "entropy": 5.8743866920471195,
      "epoch": 3.8854900784725346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035624883556648664,
      "loss": 5.3253,
      "mean_token_accuracy": 0.1975140944123268,
      "num_tokens": 76034463.0,
      "step": 38870
    },
    {
      "entropy": 5.960879278182984,
      "epoch": 3.8859899035337633,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003562155535721492,
      "loss": 5.3778,
      "mean_token_accuracy": 0.21057115197181703,
      "num_tokens": 76042909.0,
      "step": 38875
    },
    {
      "entropy": 6.013467407226562,
      "epoch": 3.8864897285949915,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003561822695344494,
      "loss": 5.5591,
      "mean_token_accuracy": 0.1857211172580719,
      "num_tokens": 76052665.0,
      "step": 38880
    },
    {
      "entropy": 6.010235691070557,
      "epoch": 3.8869895536562202,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003561489834542246,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19846698343753816,
      "num_tokens": 76062628.0,
      "step": 38885
    },
    {
      "entropy": 5.978723382949829,
      "epoch": 3.887489378717449,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035611569533231224,
      "loss": 5.4836,
      "mean_token_accuracy": 0.1989881679415703,
      "num_tokens": 76072847.0,
      "step": 38890
    },
    {
      "entropy": 6.01358995437622,
      "epoch": 3.8879892037786776,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00035608240516954995,
      "loss": 5.3786,
      "mean_token_accuracy": 0.20476105362176894,
      "num_tokens": 76081077.0,
      "step": 38895
    },
    {
      "entropy": 5.941545629501343,
      "epoch": 3.888489028839906,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035604911296677503,
      "loss": 5.3992,
      "mean_token_accuracy": 0.20316108614206313,
      "num_tokens": 76090919.0,
      "step": 38900
    },
    {
      "entropy": 5.975410270690918,
      "epoch": 3.8889888539011346,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003560158187248254,
      "loss": 5.4408,
      "mean_token_accuracy": 0.2033294379711151,
      "num_tokens": 76101529.0,
      "step": 38905
    },
    {
      "entropy": 6.010900402069092,
      "epoch": 3.8894886789623633,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003559825224445386,
      "loss": 5.4187,
      "mean_token_accuracy": 0.20336339473724366,
      "num_tokens": 76110332.0,
      "step": 38910
    },
    {
      "entropy": 5.972135734558106,
      "epoch": 3.8899885040235915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003559492241267524,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19491073340177537,
      "num_tokens": 76119951.0,
      "step": 38915
    },
    {
      "entropy": 5.970221424102784,
      "epoch": 3.8904883290848202,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035591592377230447,
      "loss": 5.4033,
      "mean_token_accuracy": 0.20258427560329437,
      "num_tokens": 76129316.0,
      "step": 38920
    },
    {
      "entropy": 5.955249261856079,
      "epoch": 3.890988154146049,
      "grad_norm": 1.125,
      "learning_rate": 0.00035588262138203276,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20116233378648757,
      "num_tokens": 76139103.0,
      "step": 38925
    },
    {
      "entropy": 5.950492477416992,
      "epoch": 3.8914879792072776,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035584931695677516,
      "loss": 5.4504,
      "mean_token_accuracy": 0.19974664151668547,
      "num_tokens": 76148310.0,
      "step": 38930
    },
    {
      "entropy": 5.855969333648682,
      "epoch": 3.891987804268506,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003558160104973695,
      "loss": 5.3207,
      "mean_token_accuracy": 0.21189628690481185,
      "num_tokens": 76158594.0,
      "step": 38935
    },
    {
      "entropy": 5.948531770706177,
      "epoch": 3.8924876293297346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003557827020046538,
      "loss": 5.4823,
      "mean_token_accuracy": 0.195450995862484,
      "num_tokens": 76168111.0,
      "step": 38940
    },
    {
      "entropy": 5.978985691070557,
      "epoch": 3.8929874543909633,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003557493914794661,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19084497392177582,
      "num_tokens": 76177616.0,
      "step": 38945
    },
    {
      "entropy": 6.004848384857178,
      "epoch": 3.8934872794521915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003557160789226445,
      "loss": 5.4876,
      "mean_token_accuracy": 0.19386011958122254,
      "num_tokens": 76186969.0,
      "step": 38950
    },
    {
      "entropy": 5.9639078140258786,
      "epoch": 3.8939871045134202,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003556827643350271,
      "loss": 5.3667,
      "mean_token_accuracy": 0.20408596843481064,
      "num_tokens": 76196206.0,
      "step": 38955
    },
    {
      "entropy": 5.994195508956909,
      "epoch": 3.894486929574649,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003556494477174522,
      "loss": 5.5093,
      "mean_token_accuracy": 0.19965025335550307,
      "num_tokens": 76205446.0,
      "step": 38960
    },
    {
      "entropy": 6.11598424911499,
      "epoch": 3.8949867546358776,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035561612907075794,
      "loss": 5.7159,
      "mean_token_accuracy": 0.17651729807257652,
      "num_tokens": 76214880.0,
      "step": 38965
    },
    {
      "entropy": 5.942145824432373,
      "epoch": 3.895486579697106,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035558280839578256,
      "loss": 5.3603,
      "mean_token_accuracy": 0.20833688527345656,
      "num_tokens": 76224928.0,
      "step": 38970
    },
    {
      "entropy": 5.984712553024292,
      "epoch": 3.8959864047583346,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003555494856933644,
      "loss": 5.4376,
      "mean_token_accuracy": 0.19407194554805757,
      "num_tokens": 76233802.0,
      "step": 38975
    },
    {
      "entropy": 5.997788190841675,
      "epoch": 3.8964862298195633,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000355516160964342,
      "loss": 5.4537,
      "mean_token_accuracy": 0.1976654052734375,
      "num_tokens": 76243303.0,
      "step": 38980
    },
    {
      "entropy": 5.961005783081054,
      "epoch": 3.8969860548807915,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035548283420955377,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19319913685321807,
      "num_tokens": 76253154.0,
      "step": 38985
    },
    {
      "entropy": 6.045575380325317,
      "epoch": 3.8974858799420202,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035544950542983793,
      "loss": 5.5149,
      "mean_token_accuracy": 0.19344032406806946,
      "num_tokens": 76262007.0,
      "step": 38990
    },
    {
      "entropy": 6.045155811309814,
      "epoch": 3.897985705003249,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035541617462603336,
      "loss": 5.5077,
      "mean_token_accuracy": 0.19141681343317032,
      "num_tokens": 76271104.0,
      "step": 38995
    },
    {
      "entropy": 5.937533521652222,
      "epoch": 3.8984855300644776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003553828417989786,
      "loss": 5.4386,
      "mean_token_accuracy": 0.19998737871646882,
      "num_tokens": 76282285.0,
      "step": 39000
    },
    {
      "epoch": 3.8984855300644776,
      "eval_entropy": 5.683745968682777,
      "eval_loss": 5.546858310699463,
      "eval_mean_token_accuracy": 0.20280931971600547,
      "eval_num_tokens": 76282285.0,
      "eval_runtime": 24.0282,
      "eval_samples_per_second": 1292.024,
      "eval_steps_per_second": 161.519,
      "step": 39000
    },
    {
      "entropy": 5.863832998275757,
      "epoch": 3.898985355125706,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035534950694951213,
      "loss": 5.3079,
      "mean_token_accuracy": 0.20302294045686722,
      "num_tokens": 76291465.0,
      "step": 39005
    },
    {
      "entropy": 5.832617330551147,
      "epoch": 3.8994851801869346,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0003553161700784728,
      "loss": 5.3603,
      "mean_token_accuracy": 0.2112121671438217,
      "num_tokens": 76300641.0,
      "step": 39010
    },
    {
      "entropy": 5.967563533782959,
      "epoch": 3.8999850052481633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003552828311866992,
      "loss": 5.4206,
      "mean_token_accuracy": 0.20477023720741272,
      "num_tokens": 76310747.0,
      "step": 39015
    },
    {
      "entropy": 6.0049357414245605,
      "epoch": 3.9004848303093915,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003552494902750303,
      "loss": 5.406,
      "mean_token_accuracy": 0.20232659131288527,
      "num_tokens": 76321307.0,
      "step": 39020
    },
    {
      "entropy": 5.946672487258911,
      "epoch": 3.9009846553706202,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035521614734430483,
      "loss": 5.451,
      "mean_token_accuracy": 0.1989259749650955,
      "num_tokens": 76330807.0,
      "step": 39025
    },
    {
      "entropy": 5.905821943283081,
      "epoch": 3.901484480431849,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035518280239536174,
      "loss": 5.3356,
      "mean_token_accuracy": 0.20921533554792404,
      "num_tokens": 76340683.0,
      "step": 39030
    },
    {
      "entropy": 5.871835947036743,
      "epoch": 3.9019843054930776,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003551494554290401,
      "loss": 5.2928,
      "mean_token_accuracy": 0.21204397827386856,
      "num_tokens": 76350793.0,
      "step": 39035
    },
    {
      "entropy": 6.0278795719146725,
      "epoch": 3.902484130554306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00035511610644617863,
      "loss": 5.5001,
      "mean_token_accuracy": 0.1991288423538208,
      "num_tokens": 76360221.0,
      "step": 39040
    },
    {
      "entropy": 5.924808931350708,
      "epoch": 3.9029839556155346,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003550827554476166,
      "loss": 5.4163,
      "mean_token_accuracy": 0.20140321552753448,
      "num_tokens": 76369094.0,
      "step": 39045
    },
    {
      "entropy": 5.887404060363769,
      "epoch": 3.9034837806767633,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035504940243419303,
      "loss": 5.4226,
      "mean_token_accuracy": 0.19698790162801744,
      "num_tokens": 76379044.0,
      "step": 39050
    },
    {
      "entropy": 5.9323728561401365,
      "epoch": 3.9039836057379915,
      "grad_norm": 1.125,
      "learning_rate": 0.0003550160474067471,
      "loss": 5.4384,
      "mean_token_accuracy": 0.19546860009431838,
      "num_tokens": 76388473.0,
      "step": 39055
    },
    {
      "entropy": 6.046583223342895,
      "epoch": 3.9044834307992202,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000354982690366118,
      "loss": 5.5631,
      "mean_token_accuracy": 0.18598492294549943,
      "num_tokens": 76400490.0,
      "step": 39060
    },
    {
      "entropy": 6.030742597579956,
      "epoch": 3.904983255860449,
      "grad_norm": 1.15625,
      "learning_rate": 0.000354949331313145,
      "loss": 5.5031,
      "mean_token_accuracy": 0.19439710974693297,
      "num_tokens": 76410418.0,
      "step": 39065
    },
    {
      "entropy": 6.034343290328979,
      "epoch": 3.9054830809216776,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035491597024866746,
      "loss": 5.5904,
      "mean_token_accuracy": 0.19229853749275208,
      "num_tokens": 76419878.0,
      "step": 39070
    },
    {
      "entropy": 5.941678142547607,
      "epoch": 3.905982905982906,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003548826071735246,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19628131687641143,
      "num_tokens": 76430078.0,
      "step": 39075
    },
    {
      "entropy": 5.91738920211792,
      "epoch": 3.9064827310441346,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003548492420885559,
      "loss": 5.4542,
      "mean_token_accuracy": 0.20534256845712662,
      "num_tokens": 76439259.0,
      "step": 39080
    },
    {
      "entropy": 5.988408184051513,
      "epoch": 3.9069825561053633,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035481587499460083,
      "loss": 5.4177,
      "mean_token_accuracy": 0.19986580163240433,
      "num_tokens": 76448622.0,
      "step": 39085
    },
    {
      "entropy": 5.833770370483398,
      "epoch": 3.9074823811665915,
      "grad_norm": 1.125,
      "learning_rate": 0.0003547825058924989,
      "loss": 5.3314,
      "mean_token_accuracy": 0.2031079798936844,
      "num_tokens": 76458197.0,
      "step": 39090
    },
    {
      "entropy": 5.964679145812989,
      "epoch": 3.90798220622782,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003547491347830896,
      "loss": 5.4441,
      "mean_token_accuracy": 0.20182772129774093,
      "num_tokens": 76466787.0,
      "step": 39095
    },
    {
      "entropy": 6.028105068206787,
      "epoch": 3.908482031289049,
      "grad_norm": 1.21875,
      "learning_rate": 0.00035471576166721267,
      "loss": 5.5045,
      "mean_token_accuracy": 0.19419795870780945,
      "num_tokens": 76475610.0,
      "step": 39100
    },
    {
      "entropy": 5.931717729568481,
      "epoch": 3.9089818563502776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035468238654570765,
      "loss": 5.3811,
      "mean_token_accuracy": 0.19875770062208176,
      "num_tokens": 76484903.0,
      "step": 39105
    },
    {
      "entropy": 6.014105987548828,
      "epoch": 3.909481681411506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003546490094194142,
      "loss": 5.481,
      "mean_token_accuracy": 0.1954732209444046,
      "num_tokens": 76495908.0,
      "step": 39110
    },
    {
      "entropy": 6.066151666641235,
      "epoch": 3.9099815064727346,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035461563028917233,
      "loss": 5.5146,
      "mean_token_accuracy": 0.19432215988636017,
      "num_tokens": 76505851.0,
      "step": 39115
    },
    {
      "entropy": 5.985326337814331,
      "epoch": 3.9104813315339633,
      "grad_norm": 1.125,
      "learning_rate": 0.00035458224915582156,
      "loss": 5.4062,
      "mean_token_accuracy": 0.19450160413980483,
      "num_tokens": 76516207.0,
      "step": 39120
    },
    {
      "entropy": 5.956322240829468,
      "epoch": 3.9109811565951915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035454886602020194,
      "loss": 5.4023,
      "mean_token_accuracy": 0.20280639678239823,
      "num_tokens": 76526882.0,
      "step": 39125
    },
    {
      "entropy": 5.966113424301147,
      "epoch": 3.91148098165642,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035451548088315325,
      "loss": 5.4586,
      "mean_token_accuracy": 0.18939703404903413,
      "num_tokens": 76536865.0,
      "step": 39130
    },
    {
      "entropy": 5.928976964950562,
      "epoch": 3.911980806717649,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003544820937455156,
      "loss": 5.4007,
      "mean_token_accuracy": 0.1998457208275795,
      "num_tokens": 76547046.0,
      "step": 39135
    },
    {
      "entropy": 5.8940346240997314,
      "epoch": 3.9124806317788776,
      "grad_norm": 1.03125,
      "learning_rate": 0.00035444870460812894,
      "loss": 5.3542,
      "mean_token_accuracy": 0.20295654237270355,
      "num_tokens": 76556941.0,
      "step": 39140
    },
    {
      "entropy": 6.012131595611573,
      "epoch": 3.912980456840106,
      "grad_norm": 1.09375,
      "learning_rate": 0.00035441531347183325,
      "loss": 5.5024,
      "mean_token_accuracy": 0.19311755001544953,
      "num_tokens": 76566843.0,
      "step": 39145
    },
    {
      "entropy": 6.005764102935791,
      "epoch": 3.9134802819013346,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003543819203374688,
      "loss": 5.4583,
      "mean_token_accuracy": 0.1971929967403412,
      "num_tokens": 76576069.0,
      "step": 39150
    },
    {
      "entropy": 6.013684368133545,
      "epoch": 3.9139801069625633,
      "grad_norm": 1.125,
      "learning_rate": 0.0003543485252058756,
      "loss": 5.4287,
      "mean_token_accuracy": 0.20229420065879822,
      "num_tokens": 76587000.0,
      "step": 39155
    },
    {
      "entropy": 5.958639097213745,
      "epoch": 3.9144799320237915,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003543151280778939,
      "loss": 5.4134,
      "mean_token_accuracy": 0.20152945220470428,
      "num_tokens": 76596133.0,
      "step": 39160
    },
    {
      "entropy": 5.958941555023193,
      "epoch": 3.91497975708502,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000354281728954364,
      "loss": 5.4506,
      "mean_token_accuracy": 0.20007920563220977,
      "num_tokens": 76606249.0,
      "step": 39165
    },
    {
      "entropy": 5.963868904113769,
      "epoch": 3.915479582146249,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035424832783612623,
      "loss": 5.4707,
      "mean_token_accuracy": 0.19617170393466948,
      "num_tokens": 76615822.0,
      "step": 39170
    },
    {
      "entropy": 5.922160387039185,
      "epoch": 3.9159794072074776,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003542149247240209,
      "loss": 5.38,
      "mean_token_accuracy": 0.20199327170848846,
      "num_tokens": 76625677.0,
      "step": 39175
    },
    {
      "entropy": 6.050334739685058,
      "epoch": 3.916479232268706,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035418151961888846,
      "loss": 5.5092,
      "mean_token_accuracy": 0.19721291661262513,
      "num_tokens": 76634783.0,
      "step": 39180
    },
    {
      "entropy": 5.967521142959595,
      "epoch": 3.9169790573299346,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00035414811252156936,
      "loss": 5.4488,
      "mean_token_accuracy": 0.1962417483329773,
      "num_tokens": 76644299.0,
      "step": 39185
    },
    {
      "entropy": 5.997948455810547,
      "epoch": 3.917478882391163,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003541147034329041,
      "loss": 5.4729,
      "mean_token_accuracy": 0.19423086792230607,
      "num_tokens": 76653992.0,
      "step": 39190
    },
    {
      "entropy": 5.9791687488555905,
      "epoch": 3.9179787074523915,
      "grad_norm": 1.25,
      "learning_rate": 0.00035408129235373325,
      "loss": 5.4123,
      "mean_token_accuracy": 0.2042781412601471,
      "num_tokens": 76663723.0,
      "step": 39195
    },
    {
      "entropy": 5.927764511108398,
      "epoch": 3.91847853251362,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035404787928489746,
      "loss": 5.3728,
      "mean_token_accuracy": 0.20157629549503325,
      "num_tokens": 76673299.0,
      "step": 39200
    },
    {
      "entropy": 5.929790019989014,
      "epoch": 3.918978357574849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035401446422723737,
      "loss": 5.3575,
      "mean_token_accuracy": 0.20918487757444382,
      "num_tokens": 76683264.0,
      "step": 39205
    },
    {
      "entropy": 5.965509510040283,
      "epoch": 3.9194781826360776,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035398104718159366,
      "loss": 5.4141,
      "mean_token_accuracy": 0.20341272354125978,
      "num_tokens": 76693004.0,
      "step": 39210
    },
    {
      "entropy": 5.833124876022339,
      "epoch": 3.919978007697306,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003539476281488072,
      "loss": 5.3463,
      "mean_token_accuracy": 0.2046533480286598,
      "num_tokens": 76701885.0,
      "step": 39215
    },
    {
      "entropy": 5.890773868560791,
      "epoch": 3.9204778327585346,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035391420712971864,
      "loss": 5.273,
      "mean_token_accuracy": 0.20942777395248413,
      "num_tokens": 76711224.0,
      "step": 39220
    },
    {
      "entropy": 6.039401531219482,
      "epoch": 3.920977657819763,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000353880784125169,
      "loss": 5.5483,
      "mean_token_accuracy": 0.19518547654151916,
      "num_tokens": 76721258.0,
      "step": 39225
    },
    {
      "entropy": 6.018417644500732,
      "epoch": 3.9214774828809915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035384735913599906,
      "loss": 5.4698,
      "mean_token_accuracy": 0.19530568718910218,
      "num_tokens": 76731510.0,
      "step": 39230
    },
    {
      "entropy": 5.915244102478027,
      "epoch": 3.92197730794222,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003538139321630499,
      "loss": 5.3656,
      "mean_token_accuracy": 0.20706900358200073,
      "num_tokens": 76741221.0,
      "step": 39235
    },
    {
      "entropy": 5.916113042831421,
      "epoch": 3.922477133003449,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003537805032071625,
      "loss": 5.3961,
      "mean_token_accuracy": 0.20022635757923127,
      "num_tokens": 76751246.0,
      "step": 39240
    },
    {
      "entropy": 5.902905511856079,
      "epoch": 3.9229769580646776,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003537470722691779,
      "loss": 5.3733,
      "mean_token_accuracy": 0.2127324253320694,
      "num_tokens": 76761346.0,
      "step": 39245
    },
    {
      "entropy": 5.892908334732056,
      "epoch": 3.923476783125906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003537136393499372,
      "loss": 5.4166,
      "mean_token_accuracy": 0.20270735174417495,
      "num_tokens": 76771463.0,
      "step": 39250
    },
    {
      "entropy": 5.928045415878296,
      "epoch": 3.9239766081871346,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003536802044502816,
      "loss": 5.4282,
      "mean_token_accuracy": 0.20077121257781982,
      "num_tokens": 76781079.0,
      "step": 39255
    },
    {
      "entropy": 5.943382406234742,
      "epoch": 3.924476433248363,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035364676757105234,
      "loss": 5.3794,
      "mean_token_accuracy": 0.20973521173000337,
      "num_tokens": 76790582.0,
      "step": 39260
    },
    {
      "entropy": 5.99786410331726,
      "epoch": 3.9249762583095915,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003536133287130906,
      "loss": 5.4577,
      "mean_token_accuracy": 0.19664919823408128,
      "num_tokens": 76800119.0,
      "step": 39265
    },
    {
      "entropy": 5.9851385116577145,
      "epoch": 3.92547608337082,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00035357988787723774,
      "loss": 5.4544,
      "mean_token_accuracy": 0.2017691984772682,
      "num_tokens": 76810190.0,
      "step": 39270
    },
    {
      "entropy": 6.007297801971435,
      "epoch": 3.925975908432049,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035354644506433515,
      "loss": 5.4785,
      "mean_token_accuracy": 0.19765117317438125,
      "num_tokens": 76819130.0,
      "step": 39275
    },
    {
      "entropy": 5.960615158081055,
      "epoch": 3.9264757334932776,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003535130002752242,
      "loss": 5.3414,
      "mean_token_accuracy": 0.20465812534093858,
      "num_tokens": 76828375.0,
      "step": 39280
    },
    {
      "entropy": 5.759033441543579,
      "epoch": 3.926975558554506,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003534795535107464,
      "loss": 5.2089,
      "mean_token_accuracy": 0.22311539202928543,
      "num_tokens": 76838423.0,
      "step": 39285
    },
    {
      "entropy": 5.955652523040771,
      "epoch": 3.9274753836157346,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035344610477174314,
      "loss": 5.3997,
      "mean_token_accuracy": 0.20083175152540206,
      "num_tokens": 76848558.0,
      "step": 39290
    },
    {
      "entropy": 6.048390626907349,
      "epoch": 3.927975208676963,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035341265405905615,
      "loss": 5.485,
      "mean_token_accuracy": 0.1941492661833763,
      "num_tokens": 76857251.0,
      "step": 39295
    },
    {
      "entropy": 5.912201356887818,
      "epoch": 3.9284750337381915,
      "grad_norm": 1.09375,
      "learning_rate": 0.000353379201373527,
      "loss": 5.2762,
      "mean_token_accuracy": 0.20999111831188202,
      "num_tokens": 76866244.0,
      "step": 39300
    },
    {
      "entropy": 5.8827245235443115,
      "epoch": 3.92897485879942,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035334574671599717,
      "loss": 5.392,
      "mean_token_accuracy": 0.19685044884681702,
      "num_tokens": 76875679.0,
      "step": 39305
    },
    {
      "entropy": 5.932296895980835,
      "epoch": 3.929474683860649,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003533122900873086,
      "loss": 5.4556,
      "mean_token_accuracy": 0.19057689905166625,
      "num_tokens": 76886440.0,
      "step": 39310
    },
    {
      "entropy": 6.020578336715698,
      "epoch": 3.9299745089218776,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003532788314883029,
      "loss": 5.4982,
      "mean_token_accuracy": 0.1940762594342232,
      "num_tokens": 76896541.0,
      "step": 39315
    },
    {
      "entropy": 5.976851892471314,
      "epoch": 3.930474333983106,
      "grad_norm": 1.109375,
      "learning_rate": 0.00035324537091982206,
      "loss": 5.4134,
      "mean_token_accuracy": 0.20133240669965743,
      "num_tokens": 76905339.0,
      "step": 39320
    },
    {
      "entropy": 5.934492778778076,
      "epoch": 3.9309741590443346,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035321190838270773,
      "loss": 5.3269,
      "mean_token_accuracy": 0.21081198751926422,
      "num_tokens": 76916004.0,
      "step": 39325
    },
    {
      "entropy": 5.9124597072601315,
      "epoch": 3.931473984105563,
      "grad_norm": 1.15625,
      "learning_rate": 0.000353178443877802,
      "loss": 5.3186,
      "mean_token_accuracy": 0.20867275893688203,
      "num_tokens": 76924885.0,
      "step": 39330
    },
    {
      "entropy": 5.89966049194336,
      "epoch": 3.9319738091667915,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035314497740594666,
      "loss": 5.4182,
      "mean_token_accuracy": 0.19904227703809738,
      "num_tokens": 76934874.0,
      "step": 39335
    },
    {
      "entropy": 5.9815733432769775,
      "epoch": 3.93247363422802,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003531115089679837,
      "loss": 5.4164,
      "mean_token_accuracy": 0.20275276750326157,
      "num_tokens": 76945831.0,
      "step": 39340
    },
    {
      "entropy": 5.9624463558197025,
      "epoch": 3.932973459289249,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035307803856475544,
      "loss": 5.3927,
      "mean_token_accuracy": 0.20403096377849578,
      "num_tokens": 76955972.0,
      "step": 39345
    },
    {
      "entropy": 6.0543938159942625,
      "epoch": 3.9334732843504776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003530445661971038,
      "loss": 5.5677,
      "mean_token_accuracy": 0.18860728591680526,
      "num_tokens": 76965940.0,
      "step": 39350
    },
    {
      "entropy": 5.9227701187133786,
      "epoch": 3.933973109411706,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035301109186587095,
      "loss": 5.2585,
      "mean_token_accuracy": 0.21189347207546233,
      "num_tokens": 76975302.0,
      "step": 39355
    },
    {
      "entropy": 5.936888647079468,
      "epoch": 3.9344729344729346,
      "grad_norm": 1.171875,
      "learning_rate": 0.00035297761557189904,
      "loss": 5.456,
      "mean_token_accuracy": 0.19619981050491334,
      "num_tokens": 76984788.0,
      "step": 39360
    },
    {
      "entropy": 5.859767913818359,
      "epoch": 3.934972759534163,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035294413731603035,
      "loss": 5.3907,
      "mean_token_accuracy": 0.2050231158733368,
      "num_tokens": 76994608.0,
      "step": 39365
    },
    {
      "entropy": 5.9647284030914305,
      "epoch": 3.9354725845953915,
      "grad_norm": 1.1875,
      "learning_rate": 0.00035291065709910734,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19537832140922545,
      "num_tokens": 77004070.0,
      "step": 39370
    },
    {
      "entropy": 6.048437547683716,
      "epoch": 3.93597240965662,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003528771749219722,
      "loss": 5.448,
      "mean_token_accuracy": 0.20022472441196443,
      "num_tokens": 77013807.0,
      "step": 39375
    },
    {
      "entropy": 5.890052223205567,
      "epoch": 3.936472234717849,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003528436907854674,
      "loss": 5.3698,
      "mean_token_accuracy": 0.1996056467294693,
      "num_tokens": 77022967.0,
      "step": 39380
    },
    {
      "entropy": 5.872053527832032,
      "epoch": 3.936972059779077,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003528102046904354,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20070850402116774,
      "num_tokens": 77032629.0,
      "step": 39385
    },
    {
      "entropy": 5.930863761901856,
      "epoch": 3.937471884840306,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003527767166377186,
      "loss": 5.346,
      "mean_token_accuracy": 0.2005032941699028,
      "num_tokens": 77042449.0,
      "step": 39390
    },
    {
      "entropy": 5.909991312026977,
      "epoch": 3.9379717099015346,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003527432266281597,
      "loss": 5.4009,
      "mean_token_accuracy": 0.1978142410516739,
      "num_tokens": 77052048.0,
      "step": 39395
    },
    {
      "entropy": 6.050573635101318,
      "epoch": 3.938471534962763,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003527097346626011,
      "loss": 5.5887,
      "mean_token_accuracy": 0.18334147483110427,
      "num_tokens": 77062438.0,
      "step": 39400
    },
    {
      "entropy": 5.924137258529663,
      "epoch": 3.9389713600239915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035267624074188564,
      "loss": 5.3153,
      "mean_token_accuracy": 0.20805560946464538,
      "num_tokens": 77071584.0,
      "step": 39405
    },
    {
      "entropy": 5.869039392471313,
      "epoch": 3.93947118508522,
      "grad_norm": 1.296875,
      "learning_rate": 0.00035264274486685593,
      "loss": 5.4125,
      "mean_token_accuracy": 0.2131005510687828,
      "num_tokens": 77082099.0,
      "step": 39410
    },
    {
      "entropy": 5.801083040237427,
      "epoch": 3.939971010146449,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003526092470383548,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21082894057035445,
      "num_tokens": 77091260.0,
      "step": 39415
    },
    {
      "entropy": 6.038938856124878,
      "epoch": 3.940470835207677,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035257574725722495,
      "loss": 5.5254,
      "mean_token_accuracy": 0.1928686872124672,
      "num_tokens": 77101467.0,
      "step": 39420
    },
    {
      "entropy": 6.057336616516113,
      "epoch": 3.940970660268906,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035254224552430923,
      "loss": 5.5997,
      "mean_token_accuracy": 0.1878153771162033,
      "num_tokens": 77111881.0,
      "step": 39425
    },
    {
      "entropy": 5.977145481109619,
      "epoch": 3.9414704853301346,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035250874184045063,
      "loss": 5.3618,
      "mean_token_accuracy": 0.20420619249343872,
      "num_tokens": 77121944.0,
      "step": 39430
    },
    {
      "entropy": 5.956996297836303,
      "epoch": 3.941970310391363,
      "grad_norm": 1.125,
      "learning_rate": 0.00035247523620649196,
      "loss": 5.3987,
      "mean_token_accuracy": 0.1995293229818344,
      "num_tokens": 77131576.0,
      "step": 39435
    },
    {
      "entropy": 5.955951499938965,
      "epoch": 3.9424701354525915,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035244172862327635,
      "loss": 5.4639,
      "mean_token_accuracy": 0.19582952558994293,
      "num_tokens": 77141013.0,
      "step": 39440
    },
    {
      "entropy": 5.951698112487793,
      "epoch": 3.94296996051382,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035240821909164675,
      "loss": 5.4904,
      "mean_token_accuracy": 0.19945783913135529,
      "num_tokens": 77151386.0,
      "step": 39445
    },
    {
      "entropy": 6.029550218582154,
      "epoch": 3.943469785575049,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003523747076124462,
      "loss": 5.5055,
      "mean_token_accuracy": 0.19071188122034072,
      "num_tokens": 77161382.0,
      "step": 39450
    },
    {
      "entropy": 5.888212585449219,
      "epoch": 3.943969610636277,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035234119418651804,
      "loss": 5.3147,
      "mean_token_accuracy": 0.21350348591804505,
      "num_tokens": 77173000.0,
      "step": 39455
    },
    {
      "entropy": 5.927023792266846,
      "epoch": 3.944469435697506,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003523076788147053,
      "loss": 5.4072,
      "mean_token_accuracy": 0.19501935094594955,
      "num_tokens": 77183149.0,
      "step": 39460
    },
    {
      "entropy": 5.959052228927613,
      "epoch": 3.9449692607587346,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003522741614978512,
      "loss": 5.428,
      "mean_token_accuracy": 0.1998415097594261,
      "num_tokens": 77192667.0,
      "step": 39465
    },
    {
      "entropy": 5.9983217239379885,
      "epoch": 3.945469085819963,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00035224064223679914,
      "loss": 5.4918,
      "mean_token_accuracy": 0.19418750256299971,
      "num_tokens": 77202670.0,
      "step": 39470
    },
    {
      "entropy": 6.003360843658447,
      "epoch": 3.9459689108811915,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003522071210323924,
      "loss": 5.4948,
      "mean_token_accuracy": 0.19606138467788697,
      "num_tokens": 77213060.0,
      "step": 39475
    },
    {
      "entropy": 5.949881935119629,
      "epoch": 3.94646873594242,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003521735978854743,
      "loss": 5.4806,
      "mean_token_accuracy": 0.19425065517425538,
      "num_tokens": 77223272.0,
      "step": 39480
    },
    {
      "entropy": 5.860894870758057,
      "epoch": 3.946968561003649,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003521400727968884,
      "loss": 5.3829,
      "mean_token_accuracy": 0.20727160423994065,
      "num_tokens": 77233350.0,
      "step": 39485
    },
    {
      "entropy": 5.979928684234619,
      "epoch": 3.947468386064877,
      "grad_norm": 1.203125,
      "learning_rate": 0.00035210654576747806,
      "loss": 5.4213,
      "mean_token_accuracy": 0.1935130923986435,
      "num_tokens": 77242614.0,
      "step": 39490
    },
    {
      "entropy": 6.019276857376099,
      "epoch": 3.947968211126106,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00035207301679808685,
      "loss": 5.4773,
      "mean_token_accuracy": 0.19679417163133622,
      "num_tokens": 77253579.0,
      "step": 39495
    },
    {
      "entropy": 6.009422397613525,
      "epoch": 3.9484680361873346,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003520394858895584,
      "loss": 5.4954,
      "mean_token_accuracy": 0.19719769209623336,
      "num_tokens": 77262921.0,
      "step": 39500
    },
    {
      "entropy": 5.884223175048828,
      "epoch": 3.948967861248563,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003520059530427364,
      "loss": 5.303,
      "mean_token_accuracy": 0.20747730284929275,
      "num_tokens": 77272261.0,
      "step": 39505
    },
    {
      "entropy": 5.9203653812408445,
      "epoch": 3.9494676863097915,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003519724182584643,
      "loss": 5.507,
      "mean_token_accuracy": 0.19597334712743758,
      "num_tokens": 77282295.0,
      "step": 39510
    },
    {
      "entropy": 5.874135541915893,
      "epoch": 3.94996751137102,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003519388815375861,
      "loss": 5.3132,
      "mean_token_accuracy": 0.20751135498285295,
      "num_tokens": 77291300.0,
      "step": 39515
    },
    {
      "entropy": 5.959390544891358,
      "epoch": 3.950467336432249,
      "grad_norm": 1.046875,
      "learning_rate": 0.00035190534288094533,
      "loss": 5.4472,
      "mean_token_accuracy": 0.20096753537654877,
      "num_tokens": 77303195.0,
      "step": 39520
    },
    {
      "entropy": 5.9249364852905275,
      "epoch": 3.950967161493477,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00035187180228938593,
      "loss": 5.3131,
      "mean_token_accuracy": 0.21092970222234725,
      "num_tokens": 77312550.0,
      "step": 39525
    },
    {
      "entropy": 5.994776964187622,
      "epoch": 3.951466986554706,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003518382597637518,
      "loss": 5.4015,
      "mean_token_accuracy": 0.20631795823574067,
      "num_tokens": 77323668.0,
      "step": 39530
    },
    {
      "entropy": 5.928653812408447,
      "epoch": 3.9519668116159345,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003518047153048868,
      "loss": 5.3932,
      "mean_token_accuracy": 0.2075226292014122,
      "num_tokens": 77334080.0,
      "step": 39535
    },
    {
      "entropy": 6.019042348861694,
      "epoch": 3.952466636677163,
      "grad_norm": 1.1875,
      "learning_rate": 0.000351771168913635,
      "loss": 5.5207,
      "mean_token_accuracy": 0.1948639288544655,
      "num_tokens": 77343849.0,
      "step": 39540
    },
    {
      "entropy": 5.9046674251556395,
      "epoch": 3.9529664617383915,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003517376205908403,
      "loss": 5.3968,
      "mean_token_accuracy": 0.20205700844526292,
      "num_tokens": 77353037.0,
      "step": 39545
    },
    {
      "entropy": 5.884303426742553,
      "epoch": 3.95346628679962,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003517040703373467,
      "loss": 5.4234,
      "mean_token_accuracy": 0.19543200135231018,
      "num_tokens": 77362758.0,
      "step": 39550
    },
    {
      "entropy": 5.906667232513428,
      "epoch": 3.953966111860849,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003516705181539986,
      "loss": 5.3851,
      "mean_token_accuracy": 0.2071445718407631,
      "num_tokens": 77372099.0,
      "step": 39555
    },
    {
      "entropy": 6.0321635723114015,
      "epoch": 3.954465936922077,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003516369640416399,
      "loss": 5.4936,
      "mean_token_accuracy": 0.19101513922214508,
      "num_tokens": 77382126.0,
      "step": 39560
    },
    {
      "entropy": 5.9584388256073,
      "epoch": 3.954965761983306,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035160340800111496,
      "loss": 5.2926,
      "mean_token_accuracy": 0.2114017128944397,
      "num_tokens": 77391674.0,
      "step": 39565
    },
    {
      "entropy": 5.984993648529053,
      "epoch": 3.9554655870445345,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000351569850033268,
      "loss": 5.5196,
      "mean_token_accuracy": 0.19415921568870545,
      "num_tokens": 77401193.0,
      "step": 39570
    },
    {
      "entropy": 5.976235628128052,
      "epoch": 3.955965412105763,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003515362901389432,
      "loss": 5.467,
      "mean_token_accuracy": 0.197413070499897,
      "num_tokens": 77410876.0,
      "step": 39575
    },
    {
      "entropy": 5.99087872505188,
      "epoch": 3.9564652371669915,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003515027283189851,
      "loss": 5.4139,
      "mean_token_accuracy": 0.2077787235379219,
      "num_tokens": 77420095.0,
      "step": 39580
    },
    {
      "entropy": 5.949046802520752,
      "epoch": 3.95696506222822,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003514691645742381,
      "loss": 5.4243,
      "mean_token_accuracy": 0.2054653286933899,
      "num_tokens": 77430526.0,
      "step": 39585
    },
    {
      "entropy": 5.940197849273682,
      "epoch": 3.957464887289449,
      "grad_norm": 1.125,
      "learning_rate": 0.0003514355989055465,
      "loss": 5.3844,
      "mean_token_accuracy": 0.20508961677551268,
      "num_tokens": 77440042.0,
      "step": 39590
    },
    {
      "entropy": 5.934018468856811,
      "epoch": 3.957964712350677,
      "grad_norm": 1.28125,
      "learning_rate": 0.000351402031313755,
      "loss": 5.4219,
      "mean_token_accuracy": 0.201909539103508,
      "num_tokens": 77450237.0,
      "step": 39595
    },
    {
      "entropy": 5.969886541366577,
      "epoch": 3.958464537411906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035136846179970797,
      "loss": 5.3613,
      "mean_token_accuracy": 0.2029539316892624,
      "num_tokens": 77459726.0,
      "step": 39600
    },
    {
      "entropy": 5.839463806152343,
      "epoch": 3.9589643624731345,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00035133489036425007,
      "loss": 5.3508,
      "mean_token_accuracy": 0.20499856472015382,
      "num_tokens": 77469477.0,
      "step": 39605
    },
    {
      "entropy": 5.9515125274658205,
      "epoch": 3.959464187534363,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035130131700822603,
      "loss": 5.475,
      "mean_token_accuracy": 0.19609327465295792,
      "num_tokens": 77479474.0,
      "step": 39610
    },
    {
      "entropy": 5.932398843765259,
      "epoch": 3.9599640125955915,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003512677417324805,
      "loss": 5.3523,
      "mean_token_accuracy": 0.19901374131441116,
      "num_tokens": 77489149.0,
      "step": 39615
    },
    {
      "entropy": 5.920387935638428,
      "epoch": 3.96046383765682,
      "grad_norm": 1.125,
      "learning_rate": 0.0003512341645378582,
      "loss": 5.3747,
      "mean_token_accuracy": 0.20490497797727586,
      "num_tokens": 77499267.0,
      "step": 39620
    },
    {
      "entropy": 5.8615882396698,
      "epoch": 3.960963662718049,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003512005854252039,
      "loss": 5.302,
      "mean_token_accuracy": 0.2109546482563019,
      "num_tokens": 77509225.0,
      "step": 39625
    },
    {
      "entropy": 5.878637075424194,
      "epoch": 3.961463487779277,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00035116700439536254,
      "loss": 5.4496,
      "mean_token_accuracy": 0.19979067593812944,
      "num_tokens": 77518380.0,
      "step": 39630
    },
    {
      "entropy": 5.958262872695923,
      "epoch": 3.961963312840506,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00035113342144917894,
      "loss": 5.4936,
      "mean_token_accuracy": 0.1870016112923622,
      "num_tokens": 77527689.0,
      "step": 39635
    },
    {
      "entropy": 6.057064867019653,
      "epoch": 3.9624631379017345,
      "grad_norm": 1.0625,
      "learning_rate": 0.00035109983658749796,
      "loss": 5.511,
      "mean_token_accuracy": 0.18705556243658067,
      "num_tokens": 77538265.0,
      "step": 39640
    },
    {
      "entropy": 5.9494939804077145,
      "epoch": 3.962962962962963,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003510662498111648,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20740821659564973,
      "num_tokens": 77548771.0,
      "step": 39645
    },
    {
      "entropy": 5.908082962036133,
      "epoch": 3.9634627880241915,
      "grad_norm": 1.078125,
      "learning_rate": 0.00035103266112102423,
      "loss": 5.4505,
      "mean_token_accuracy": 0.2023815020918846,
      "num_tokens": 77559224.0,
      "step": 39650
    },
    {
      "entropy": 6.013995027542114,
      "epoch": 3.96396261308542,
      "grad_norm": 1.15625,
      "learning_rate": 0.00035099907051792153,
      "loss": 5.4968,
      "mean_token_accuracy": 0.19497900009155272,
      "num_tokens": 77568593.0,
      "step": 39655
    },
    {
      "entropy": 5.9046276092529295,
      "epoch": 3.964462438146649,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003509654780027018,
      "loss": 5.3365,
      "mean_token_accuracy": 0.2114981010556221,
      "num_tokens": 77577867.0,
      "step": 39660
    },
    {
      "entropy": 5.934952878952027,
      "epoch": 3.964962263207877,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003509318835762102,
      "loss": 5.3994,
      "mean_token_accuracy": 0.2031873032450676,
      "num_tokens": 77588207.0,
      "step": 39665
    },
    {
      "entropy": 5.9727277755737305,
      "epoch": 3.965462088269106,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003508982872392919,
      "loss": 5.3594,
      "mean_token_accuracy": 0.20001390427350998,
      "num_tokens": 77597538.0,
      "step": 39670
    },
    {
      "entropy": 5.960658597946167,
      "epoch": 3.9659619133303345,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003508646889927923,
      "loss": 5.4848,
      "mean_token_accuracy": 0.1934969425201416,
      "num_tokens": 77607538.0,
      "step": 39675
    },
    {
      "entropy": 5.883718633651734,
      "epoch": 3.966461738391563,
      "grad_norm": 1.140625,
      "learning_rate": 0.00035083108883755647,
      "loss": 5.3373,
      "mean_token_accuracy": 0.20232246220111846,
      "num_tokens": 77618339.0,
      "step": 39680
    },
    {
      "entropy": 5.9071986198425295,
      "epoch": 3.9669615634527915,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035079748677443014,
      "loss": 5.3468,
      "mean_token_accuracy": 0.20130581706762313,
      "num_tokens": 77627022.0,
      "step": 39685
    },
    {
      "entropy": 5.977650547027588,
      "epoch": 3.96746138851402,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003507638828042585,
      "loss": 5.505,
      "mean_token_accuracy": 0.19222853034734727,
      "num_tokens": 77637129.0,
      "step": 39690
    },
    {
      "entropy": 5.9850929260253904,
      "epoch": 3.967961213575249,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0003507302769278871,
      "loss": 5.4306,
      "mean_token_accuracy": 0.20600355118513108,
      "num_tokens": 77647325.0,
      "step": 39695
    },
    {
      "entropy": 5.944225263595581,
      "epoch": 3.968461038636477,
      "grad_norm": 1.125,
      "learning_rate": 0.0003506966691461614,
      "loss": 5.3857,
      "mean_token_accuracy": 0.20599175244569778,
      "num_tokens": 77657673.0,
      "step": 39700
    },
    {
      "entropy": 5.901599407196045,
      "epoch": 3.968960863697706,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003506630594599269,
      "loss": 5.3497,
      "mean_token_accuracy": 0.21190887689590454,
      "num_tokens": 77667168.0,
      "step": 39705
    },
    {
      "entropy": 5.957151412963867,
      "epoch": 3.969460688758934,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003506294478700293,
      "loss": 5.4031,
      "mean_token_accuracy": 0.20273101776838304,
      "num_tokens": 77675125.0,
      "step": 39710
    },
    {
      "entropy": 5.905121755599976,
      "epoch": 3.969960513820163,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00035059583437731435,
      "loss": 5.4189,
      "mean_token_accuracy": 0.2007981851696968,
      "num_tokens": 77685256.0,
      "step": 39715
    },
    {
      "entropy": 5.976315593719482,
      "epoch": 3.9704603388813915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035056221898262756,
      "loss": 5.477,
      "mean_token_accuracy": 0.19155273586511612,
      "num_tokens": 77696278.0,
      "step": 39720
    },
    {
      "entropy": 5.971701431274414,
      "epoch": 3.97096016394262,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00035052860168681486,
      "loss": 5.3936,
      "mean_token_accuracy": 0.1970364958047867,
      "num_tokens": 77705493.0,
      "step": 39725
    },
    {
      "entropy": 5.963785552978516,
      "epoch": 3.971459989003849,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000350494982490722,
      "loss": 5.415,
      "mean_token_accuracy": 0.20580146014690398,
      "num_tokens": 77715429.0,
      "step": 39730
    },
    {
      "entropy": 5.949447345733643,
      "epoch": 3.971959814065077,
      "grad_norm": 1.125,
      "learning_rate": 0.00035046136139519474,
      "loss": 5.4334,
      "mean_token_accuracy": 0.19879972487688063,
      "num_tokens": 77725529.0,
      "step": 39735
    },
    {
      "entropy": 5.920466375350952,
      "epoch": 3.972459639126306,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000350427738401079,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21155526340007783,
      "num_tokens": 77734191.0,
      "step": 39740
    },
    {
      "entropy": 5.898494815826416,
      "epoch": 3.972959464187534,
      "grad_norm": 1.25,
      "learning_rate": 0.0003503941135092209,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20462250858545303,
      "num_tokens": 77744145.0,
      "step": 39745
    },
    {
      "entropy": 6.028903150558472,
      "epoch": 3.973459289248763,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003503604867204662,
      "loss": 5.5091,
      "mean_token_accuracy": 0.19108510613441468,
      "num_tokens": 77754027.0,
      "step": 39750
    },
    {
      "entropy": 6.007675552368164,
      "epoch": 3.9739591143099915,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003503268580356611,
      "loss": 5.4927,
      "mean_token_accuracy": 0.19534268826246262,
      "num_tokens": 77766002.0,
      "step": 39755
    },
    {
      "entropy": 5.961576986312866,
      "epoch": 3.97445893937122,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00035029322745565165,
      "loss": 5.4359,
      "mean_token_accuracy": 0.2022378996014595,
      "num_tokens": 77775995.0,
      "step": 39760
    },
    {
      "entropy": 6.0808248043060305,
      "epoch": 3.974958764432449,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003502595949812839,
      "loss": 5.6032,
      "mean_token_accuracy": 0.1904863551259041,
      "num_tokens": 77786890.0,
      "step": 39765
    },
    {
      "entropy": 5.865001916885376,
      "epoch": 3.975458589493677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00035022596061340416,
      "loss": 5.3027,
      "mean_token_accuracy": 0.20604859739542009,
      "num_tokens": 77797236.0,
      "step": 39770
    },
    {
      "entropy": 5.931922388076782,
      "epoch": 3.975958414554906,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00035019232435285863,
      "loss": 5.3932,
      "mean_token_accuracy": 0.2048187166452408,
      "num_tokens": 77807564.0,
      "step": 39775
    },
    {
      "entropy": 5.950830030441284,
      "epoch": 3.976458239616134,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003501586862004935,
      "loss": 5.3905,
      "mean_token_accuracy": 0.1978711649775505,
      "num_tokens": 77817297.0,
      "step": 39780
    },
    {
      "entropy": 5.889858675003052,
      "epoch": 3.976958064677363,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003501250461571552,
      "loss": 5.3807,
      "mean_token_accuracy": 0.20105933398008347,
      "num_tokens": 77826209.0,
      "step": 39785
    },
    {
      "entropy": 5.993305492401123,
      "epoch": 3.9774578897385915,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003500914042236901,
      "loss": 5.5199,
      "mean_token_accuracy": 0.20178601145744324,
      "num_tokens": 77836076.0,
      "step": 39790
    },
    {
      "entropy": 5.960287094116211,
      "epoch": 3.97795771479982,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003500577604009445,
      "loss": 5.4052,
      "mean_token_accuracy": 0.19709648936986923,
      "num_tokens": 77845032.0,
      "step": 39795
    },
    {
      "entropy": 5.930457019805909,
      "epoch": 3.978457539861049,
      "grad_norm": 1.234375,
      "learning_rate": 0.00035002411468976506,
      "loss": 5.4114,
      "mean_token_accuracy": 0.2012923002243042,
      "num_tokens": 77856069.0,
      "step": 39800
    },
    {
      "entropy": 5.858736991882324,
      "epoch": 3.978957364922277,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034999046709099815,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20705099850893022,
      "num_tokens": 77865917.0,
      "step": 39805
    },
    {
      "entropy": 5.8993617534637455,
      "epoch": 3.979457189983506,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003499568176054904,
      "loss": 5.3343,
      "mean_token_accuracy": 0.21024831086397172,
      "num_tokens": 77874597.0,
      "step": 39810
    },
    {
      "entropy": 5.943438625335693,
      "epoch": 3.979957015044734,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034992316623408843,
      "loss": 5.4209,
      "mean_token_accuracy": 0.20318839997053145,
      "num_tokens": 77885014.0,
      "step": 39815
    },
    {
      "entropy": 5.94946870803833,
      "epoch": 3.980456840105963,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034988951297763887,
      "loss": 5.3946,
      "mean_token_accuracy": 0.20099509805440902,
      "num_tokens": 77894901.0,
      "step": 39820
    },
    {
      "entropy": 6.022850751876831,
      "epoch": 3.9809566651671915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034985585783698845,
      "loss": 5.4363,
      "mean_token_accuracy": 0.1983616203069687,
      "num_tokens": 77904466.0,
      "step": 39825
    },
    {
      "entropy": 5.940260744094848,
      "epoch": 3.98145649022842,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003498222008129838,
      "loss": 5.4449,
      "mean_token_accuracy": 0.20436283051967621,
      "num_tokens": 77913700.0,
      "step": 39830
    },
    {
      "entropy": 5.935576009750366,
      "epoch": 3.981956315289649,
      "grad_norm": 1.25,
      "learning_rate": 0.00034978854190647197,
      "loss": 5.4452,
      "mean_token_accuracy": 0.19385509341955184,
      "num_tokens": 77923385.0,
      "step": 39835
    },
    {
      "entropy": 5.923566102981567,
      "epoch": 3.982456140350877,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034975488111829955,
      "loss": 5.4027,
      "mean_token_accuracy": 0.1997555360198021,
      "num_tokens": 77933193.0,
      "step": 39840
    },
    {
      "entropy": 5.888824892044068,
      "epoch": 3.982955965412106,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034972121844931365,
      "loss": 5.3927,
      "mean_token_accuracy": 0.20267246663570404,
      "num_tokens": 77942878.0,
      "step": 39845
    },
    {
      "entropy": 5.951868009567261,
      "epoch": 3.983455790473334,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034968755390036107,
      "loss": 5.4712,
      "mean_token_accuracy": 0.20186376124620437,
      "num_tokens": 77953499.0,
      "step": 39850
    },
    {
      "entropy": 5.970104932785034,
      "epoch": 3.983955615534563,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034965388747228886,
      "loss": 5.3871,
      "mean_token_accuracy": 0.19760524332523347,
      "num_tokens": 77963716.0,
      "step": 39855
    },
    {
      "entropy": 5.9172686576843265,
      "epoch": 3.9844554405957915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034962021916594405,
      "loss": 5.4044,
      "mean_token_accuracy": 0.19856030791997908,
      "num_tokens": 77973118.0,
      "step": 39860
    },
    {
      "entropy": 5.929386568069458,
      "epoch": 3.98495526565702,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003495865489821737,
      "loss": 5.4186,
      "mean_token_accuracy": 0.20166170597076416,
      "num_tokens": 77983658.0,
      "step": 39865
    },
    {
      "entropy": 5.9180882453918455,
      "epoch": 3.985455090718249,
      "grad_norm": 1.140625,
      "learning_rate": 0.000349552876921825,
      "loss": 5.4112,
      "mean_token_accuracy": 0.20290945172309877,
      "num_tokens": 77993661.0,
      "step": 39870
    },
    {
      "entropy": 5.995750141143799,
      "epoch": 3.985954915779477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003495192029857452,
      "loss": 5.4911,
      "mean_token_accuracy": 0.19708727449178695,
      "num_tokens": 78003900.0,
      "step": 39875
    },
    {
      "entropy": 5.982664918899536,
      "epoch": 3.986454740840706,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003494855271747813,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19160526543855666,
      "num_tokens": 78013936.0,
      "step": 39880
    },
    {
      "entropy": 5.9855818271636965,
      "epoch": 3.986954565901934,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034945184948978074,
      "loss": 5.4401,
      "mean_token_accuracy": 0.20141021013259888,
      "num_tokens": 78025190.0,
      "step": 39885
    },
    {
      "entropy": 5.931356763839721,
      "epoch": 3.987454390963163,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003494181699315908,
      "loss": 5.42,
      "mean_token_accuracy": 0.19835955798625945,
      "num_tokens": 78034562.0,
      "step": 39890
    },
    {
      "entropy": 6.013452768325806,
      "epoch": 3.9879542160243915,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003493844885010589,
      "loss": 5.4993,
      "mean_token_accuracy": 0.2003009021282196,
      "num_tokens": 78044793.0,
      "step": 39895
    },
    {
      "entropy": 5.943293046951294,
      "epoch": 3.98845404108562,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034935080519903235,
      "loss": 5.4711,
      "mean_token_accuracy": 0.20348672121763228,
      "num_tokens": 78054099.0,
      "step": 39900
    },
    {
      "entropy": 5.989203310012817,
      "epoch": 3.9889538661468484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003493171200263586,
      "loss": 5.5556,
      "mean_token_accuracy": 0.19732647985219956,
      "num_tokens": 78064647.0,
      "step": 39905
    },
    {
      "entropy": 5.9911401748657225,
      "epoch": 3.989453691208077,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003492834329838854,
      "loss": 5.4158,
      "mean_token_accuracy": 0.20434311628341675,
      "num_tokens": 78073787.0,
      "step": 39910
    },
    {
      "entropy": 5.950287342071533,
      "epoch": 3.989953516269306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034924974407246005,
      "loss": 5.4625,
      "mean_token_accuracy": 0.20313582718372344,
      "num_tokens": 78082780.0,
      "step": 39915
    },
    {
      "entropy": 5.933841276168823,
      "epoch": 3.990453341330534,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034921605329293025,
      "loss": 5.4349,
      "mean_token_accuracy": 0.1978723645210266,
      "num_tokens": 78092379.0,
      "step": 39920
    },
    {
      "entropy": 5.927737045288086,
      "epoch": 3.990953166391763,
      "grad_norm": 1.015625,
      "learning_rate": 0.00034918236064614366,
      "loss": 5.4557,
      "mean_token_accuracy": 0.20355177223682402,
      "num_tokens": 78101311.0,
      "step": 39925
    },
    {
      "entropy": 5.940766906738281,
      "epoch": 3.9914529914529915,
      "grad_norm": 1.078125,
      "learning_rate": 0.000349148666132948,
      "loss": 5.3951,
      "mean_token_accuracy": 0.2058054193854332,
      "num_tokens": 78111646.0,
      "step": 39930
    },
    {
      "entropy": 6.023172283172608,
      "epoch": 3.99195281651422,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003491149697541909,
      "loss": 5.4653,
      "mean_token_accuracy": 0.19913459718227386,
      "num_tokens": 78122350.0,
      "step": 39935
    },
    {
      "entropy": 5.978123044967651,
      "epoch": 3.9924526415754484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034908127151072027,
      "loss": 5.3793,
      "mean_token_accuracy": 0.20301311165094377,
      "num_tokens": 78133724.0,
      "step": 39940
    },
    {
      "entropy": 5.90438323020935,
      "epoch": 3.992952466636677,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003490475714033839,
      "loss": 5.4149,
      "mean_token_accuracy": 0.20607622116804122,
      "num_tokens": 78143001.0,
      "step": 39945
    },
    {
      "entropy": 6.046658229827881,
      "epoch": 3.993452291697906,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034901386943302976,
      "loss": 5.4166,
      "mean_token_accuracy": 0.2042853429913521,
      "num_tokens": 78152882.0,
      "step": 39950
    },
    {
      "entropy": 5.985553455352783,
      "epoch": 3.993952116759134,
      "grad_norm": 1.125,
      "learning_rate": 0.0003489801656005057,
      "loss": 5.393,
      "mean_token_accuracy": 0.20385841727256776,
      "num_tokens": 78162011.0,
      "step": 39955
    },
    {
      "entropy": 5.913377046585083,
      "epoch": 3.994451941820363,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034894645990665957,
      "loss": 5.3383,
      "mean_token_accuracy": 0.21104785054922104,
      "num_tokens": 78172142.0,
      "step": 39960
    },
    {
      "entropy": 5.917280626296997,
      "epoch": 3.9949517668815915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034891275235233966,
      "loss": 5.3774,
      "mean_token_accuracy": 0.2054872140288353,
      "num_tokens": 78181429.0,
      "step": 39965
    },
    {
      "entropy": 5.872385883331299,
      "epoch": 3.99545159194282,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034887904293839385,
      "loss": 5.3462,
      "mean_token_accuracy": 0.2016613781452179,
      "num_tokens": 78190314.0,
      "step": 39970
    },
    {
      "entropy": 5.965636587142944,
      "epoch": 3.9959514170040484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003488453316656704,
      "loss": 5.3348,
      "mean_token_accuracy": 0.20644858628511428,
      "num_tokens": 78200256.0,
      "step": 39975
    },
    {
      "entropy": 6.0080901145935055,
      "epoch": 3.996451242065277,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003488116185350174,
      "loss": 5.4785,
      "mean_token_accuracy": 0.2015056475996971,
      "num_tokens": 78210596.0,
      "step": 39980
    },
    {
      "entropy": 5.967173004150391,
      "epoch": 3.996951067126506,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034877790354728304,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20306208729743958,
      "num_tokens": 78220254.0,
      "step": 39985
    },
    {
      "entropy": 5.960416603088379,
      "epoch": 3.997450892187734,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003487441867033156,
      "loss": 5.4235,
      "mean_token_accuracy": 0.20016899406909944,
      "num_tokens": 78229243.0,
      "step": 39990
    },
    {
      "entropy": 5.895805835723877,
      "epoch": 3.997950717248963,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003487104680039634,
      "loss": 5.3505,
      "mean_token_accuracy": 0.20272949039936067,
      "num_tokens": 78239312.0,
      "step": 39995
    },
    {
      "entropy": 5.910320615768432,
      "epoch": 3.9984505423101915,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003486767474500748,
      "loss": 5.4029,
      "mean_token_accuracy": 0.19523339569568635,
      "num_tokens": 78249306.0,
      "step": 40000
    },
    {
      "entropy": 5.939142656326294,
      "epoch": 3.99895036737142,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034864302504249823,
      "loss": 5.4085,
      "mean_token_accuracy": 0.20553729087114334,
      "num_tokens": 78258186.0,
      "step": 40005
    },
    {
      "entropy": 5.93809232711792,
      "epoch": 3.9994501924326484,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034860930078208207,
      "loss": 5.4376,
      "mean_token_accuracy": 0.20523661822080613,
      "num_tokens": 78268051.0,
      "step": 40010
    },
    {
      "entropy": 5.97911229133606,
      "epoch": 3.999950017493877,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034857557466967486,
      "loss": 5.3961,
      "mean_token_accuracy": 0.2027093529701233,
      "num_tokens": 78277574.0,
      "step": 40015
    },
    {
      "entropy": 5.92851644092136,
      "epoch": 4.000399860048983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034854184670612514,
      "loss": 5.3978,
      "mean_token_accuracy": 0.2115433696243498,
      "num_tokens": 78286849.0,
      "step": 40020
    },
    {
      "entropy": 5.9663629055023195,
      "epoch": 4.000899685110212,
      "grad_norm": 1.125,
      "learning_rate": 0.0003485081168922814,
      "loss": 5.355,
      "mean_token_accuracy": 0.21200315207242965,
      "num_tokens": 78297180.0,
      "step": 40025
    },
    {
      "entropy": 6.062990379333496,
      "epoch": 4.00139951017144,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034847438522899246,
      "loss": 5.5355,
      "mean_token_accuracy": 0.192893385887146,
      "num_tokens": 78307613.0,
      "step": 40030
    },
    {
      "entropy": 5.94445834159851,
      "epoch": 4.001899335232668,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003484406517171068,
      "loss": 5.2659,
      "mean_token_accuracy": 0.21230709552764893,
      "num_tokens": 78316955.0,
      "step": 40035
    },
    {
      "entropy": 5.967542409896851,
      "epoch": 4.002399160293897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003484069163574733,
      "loss": 5.3459,
      "mean_token_accuracy": 0.2080559954047203,
      "num_tokens": 78327458.0,
      "step": 40040
    },
    {
      "entropy": 5.984727573394776,
      "epoch": 4.002898985355126,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034837317915094063,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20854541957378386,
      "num_tokens": 78337636.0,
      "step": 40045
    },
    {
      "entropy": 5.916361284255982,
      "epoch": 4.003398810416354,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034833944009835774,
      "loss": 5.2165,
      "mean_token_accuracy": 0.2108582317829132,
      "num_tokens": 78346845.0,
      "step": 40050
    },
    {
      "entropy": 5.9736048698425295,
      "epoch": 4.003898635477583,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034830569920057323,
      "loss": 5.356,
      "mean_token_accuracy": 0.2026491194963455,
      "num_tokens": 78357797.0,
      "step": 40055
    },
    {
      "entropy": 5.967108297348022,
      "epoch": 4.004398460538812,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034827195645843636,
      "loss": 5.4372,
      "mean_token_accuracy": 0.19986324459314347,
      "num_tokens": 78368194.0,
      "step": 40060
    },
    {
      "entropy": 5.882105350494385,
      "epoch": 4.00489828560004,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003482382118727959,
      "loss": 5.2851,
      "mean_token_accuracy": 0.20479073375463486,
      "num_tokens": 78378394.0,
      "step": 40065
    },
    {
      "entropy": 5.927546453475952,
      "epoch": 4.005398110661268,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003482044654445008,
      "loss": 5.3268,
      "mean_token_accuracy": 0.21239462047815322,
      "num_tokens": 78387549.0,
      "step": 40070
    },
    {
      "entropy": 5.997396421432495,
      "epoch": 4.005897935722497,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034817071717440024,
      "loss": 5.3827,
      "mean_token_accuracy": 0.1997874066233635,
      "num_tokens": 78396896.0,
      "step": 40075
    },
    {
      "entropy": 6.004905939102173,
      "epoch": 4.006397760783726,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003481369670633433,
      "loss": 5.3314,
      "mean_token_accuracy": 0.21176677942276,
      "num_tokens": 78404754.0,
      "step": 40080
    },
    {
      "entropy": 5.734345531463623,
      "epoch": 4.006897585844954,
      "grad_norm": 0.9375,
      "learning_rate": 0.00034810321511217894,
      "loss": 5.1688,
      "mean_token_accuracy": 0.22712947577238082,
      "num_tokens": 78415645.0,
      "step": 40085
    },
    {
      "entropy": 5.925586271286011,
      "epoch": 4.007397410906183,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003480694613217567,
      "loss": 5.2742,
      "mean_token_accuracy": 0.22031389772892,
      "num_tokens": 78425821.0,
      "step": 40090
    },
    {
      "entropy": 5.9688694953918455,
      "epoch": 4.007897235967412,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003480357056929255,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2170162484049797,
      "num_tokens": 78436260.0,
      "step": 40095
    },
    {
      "entropy": 5.9745080947875975,
      "epoch": 4.00839706102864,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00034800194822653474,
      "loss": 5.4169,
      "mean_token_accuracy": 0.19853576868772507,
      "num_tokens": 78446042.0,
      "step": 40100
    },
    {
      "entropy": 5.946594095230102,
      "epoch": 4.008896886089868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034796818892343374,
      "loss": 5.3865,
      "mean_token_accuracy": 0.20017210394144058,
      "num_tokens": 78455120.0,
      "step": 40105
    },
    {
      "entropy": 5.980182218551636,
      "epoch": 4.009396711151097,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034793442778447196,
      "loss": 5.336,
      "mean_token_accuracy": 0.2107594132423401,
      "num_tokens": 78465173.0,
      "step": 40110
    },
    {
      "entropy": 5.906267499923706,
      "epoch": 4.009896536212326,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003479006648104987,
      "loss": 5.3905,
      "mean_token_accuracy": 0.19685445129871368,
      "num_tokens": 78475154.0,
      "step": 40115
    },
    {
      "entropy": 5.930439901351929,
      "epoch": 4.010396361273554,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003478669000023634,
      "loss": 5.3904,
      "mean_token_accuracy": 0.2033759042620659,
      "num_tokens": 78485270.0,
      "step": 40120
    },
    {
      "entropy": 5.969518852233887,
      "epoch": 4.010896186334783,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00034783313336091576,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20577598512172698,
      "num_tokens": 78495576.0,
      "step": 40125
    },
    {
      "entropy": 6.012857103347779,
      "epoch": 4.011396011396012,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034779936488700514,
      "loss": 5.4633,
      "mean_token_accuracy": 0.20054974257946015,
      "num_tokens": 78505444.0,
      "step": 40130
    },
    {
      "entropy": 5.947394466400146,
      "epoch": 4.01189583645724,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003477655945814812,
      "loss": 5.4215,
      "mean_token_accuracy": 0.20225146859884263,
      "num_tokens": 78514476.0,
      "step": 40135
    },
    {
      "entropy": 5.8708470344543455,
      "epoch": 4.012395661518468,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034773182244519376,
      "loss": 5.3504,
      "mean_token_accuracy": 0.20764851123094558,
      "num_tokens": 78523864.0,
      "step": 40140
    },
    {
      "entropy": 5.92213807106018,
      "epoch": 4.012895486579697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003476980484789923,
      "loss": 5.3497,
      "mean_token_accuracy": 0.20389092564582825,
      "num_tokens": 78533402.0,
      "step": 40145
    },
    {
      "entropy": 5.91709885597229,
      "epoch": 4.0133953116409256,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003476642726837267,
      "loss": 5.2788,
      "mean_token_accuracy": 0.20921474695205688,
      "num_tokens": 78542919.0,
      "step": 40150
    },
    {
      "entropy": 5.931597661972046,
      "epoch": 4.013895136702154,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034763049506024657,
      "loss": 5.3424,
      "mean_token_accuracy": 0.20814954489469528,
      "num_tokens": 78552360.0,
      "step": 40155
    },
    {
      "entropy": 5.853214883804322,
      "epoch": 4.014394961763383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000347596715609402,
      "loss": 5.293,
      "mean_token_accuracy": 0.22007670998573303,
      "num_tokens": 78561943.0,
      "step": 40160
    },
    {
      "entropy": 5.954863882064819,
      "epoch": 4.014894786824612,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003475629343320427,
      "loss": 5.2972,
      "mean_token_accuracy": 0.21458668559789656,
      "num_tokens": 78571538.0,
      "step": 40165
    },
    {
      "entropy": 5.970846033096313,
      "epoch": 4.01539461188584,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034752915122901863,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19976669996976854,
      "num_tokens": 78581369.0,
      "step": 40170
    },
    {
      "entropy": 5.872440576553345,
      "epoch": 4.015894436947068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034749536630117984,
      "loss": 5.3193,
      "mean_token_accuracy": 0.20922008901834488,
      "num_tokens": 78592412.0,
      "step": 40175
    },
    {
      "entropy": 5.8721435546875,
      "epoch": 4.016394262008297,
      "grad_norm": 1.125,
      "learning_rate": 0.0003474615795493763,
      "loss": 5.3577,
      "mean_token_accuracy": 0.20515966564416885,
      "num_tokens": 78601527.0,
      "step": 40180
    },
    {
      "entropy": 5.920341348648071,
      "epoch": 4.0168940870695256,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034742779097445795,
      "loss": 5.3497,
      "mean_token_accuracy": 0.20060724914073944,
      "num_tokens": 78611906.0,
      "step": 40185
    },
    {
      "entropy": 5.981380271911621,
      "epoch": 4.017393912130754,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003473940005772751,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20313855409622192,
      "num_tokens": 78621153.0,
      "step": 40190
    },
    {
      "entropy": 6.05165286064148,
      "epoch": 4.017893737191983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034736020835867776,
      "loss": 5.487,
      "mean_token_accuracy": 0.1969892993569374,
      "num_tokens": 78629851.0,
      "step": 40195
    },
    {
      "entropy": 6.046611309051514,
      "epoch": 4.018393562253212,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034732641431951626,
      "loss": 5.4385,
      "mean_token_accuracy": 0.2013244703412056,
      "num_tokens": 78639837.0,
      "step": 40200
    },
    {
      "entropy": 5.969648694992065,
      "epoch": 4.01889338731444,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003472926184606408,
      "loss": 5.3582,
      "mean_token_accuracy": 0.20227339565753938,
      "num_tokens": 78649341.0,
      "step": 40205
    },
    {
      "entropy": 5.970524644851684,
      "epoch": 4.019393212375668,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034725882078290163,
      "loss": 5.3015,
      "mean_token_accuracy": 0.20503397136926652,
      "num_tokens": 78657360.0,
      "step": 40210
    },
    {
      "entropy": 5.942204999923706,
      "epoch": 4.019893037436897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003472250212871491,
      "loss": 5.3513,
      "mean_token_accuracy": 0.20551547259092331,
      "num_tokens": 78667787.0,
      "step": 40215
    },
    {
      "entropy": 5.853451108932495,
      "epoch": 4.0203928624981256,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034719121997423366,
      "loss": 5.2412,
      "mean_token_accuracy": 0.2164212465286255,
      "num_tokens": 78677158.0,
      "step": 40220
    },
    {
      "entropy": 5.859960317611694,
      "epoch": 4.020892687559354,
      "grad_norm": 1.125,
      "learning_rate": 0.0003471574168450057,
      "loss": 5.2836,
      "mean_token_accuracy": 0.22269098460674286,
      "num_tokens": 78687805.0,
      "step": 40225
    },
    {
      "entropy": 6.014342021942139,
      "epoch": 4.021392512620583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034712361190031564,
      "loss": 5.4341,
      "mean_token_accuracy": 0.19306478202342986,
      "num_tokens": 78697937.0,
      "step": 40230
    },
    {
      "entropy": 5.8781345844268795,
      "epoch": 4.021892337681812,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003470898051410141,
      "loss": 5.2943,
      "mean_token_accuracy": 0.2171505019068718,
      "num_tokens": 78708591.0,
      "step": 40235
    },
    {
      "entropy": 5.952771949768066,
      "epoch": 4.02239216274304,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003470559965679516,
      "loss": 5.3609,
      "mean_token_accuracy": 0.2027575433254242,
      "num_tokens": 78718559.0,
      "step": 40240
    },
    {
      "entropy": 5.852596712112427,
      "epoch": 4.022891987804268,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034702218618197885,
      "loss": 5.2457,
      "mean_token_accuracy": 0.2172536224126816,
      "num_tokens": 78727412.0,
      "step": 40245
    },
    {
      "entropy": 5.795462036132813,
      "epoch": 4.023391812865497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003469883739839463,
      "loss": 5.1972,
      "mean_token_accuracy": 0.21701402664184571,
      "num_tokens": 78737298.0,
      "step": 40250
    },
    {
      "entropy": 5.907680940628052,
      "epoch": 4.0238916379267256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034695455997470486,
      "loss": 5.3088,
      "mean_token_accuracy": 0.20863093733787536,
      "num_tokens": 78747301.0,
      "step": 40255
    },
    {
      "entropy": 5.96299295425415,
      "epoch": 4.024391462987954,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034692074415510523,
      "loss": 5.3342,
      "mean_token_accuracy": 0.20570573210716248,
      "num_tokens": 78756615.0,
      "step": 40260
    },
    {
      "entropy": 6.002839136123657,
      "epoch": 4.024891288049183,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034688692652599815,
      "loss": 5.3705,
      "mean_token_accuracy": 0.21081386208534242,
      "num_tokens": 78766277.0,
      "step": 40265
    },
    {
      "entropy": 5.943508815765381,
      "epoch": 4.025391113110412,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003468531070882345,
      "loss": 5.3609,
      "mean_token_accuracy": 0.19985802471637726,
      "num_tokens": 78775899.0,
      "step": 40270
    },
    {
      "entropy": 5.906406497955322,
      "epoch": 4.02589093817164,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034681928584266513,
      "loss": 5.2528,
      "mean_token_accuracy": 0.2165019005537033,
      "num_tokens": 78785472.0,
      "step": 40275
    },
    {
      "entropy": 5.908063554763794,
      "epoch": 4.026390763232868,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000346785462790141,
      "loss": 5.3054,
      "mean_token_accuracy": 0.2137579247355461,
      "num_tokens": 78795312.0,
      "step": 40280
    },
    {
      "entropy": 5.872123050689697,
      "epoch": 4.026890588294097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034675163793151313,
      "loss": 5.309,
      "mean_token_accuracy": 0.2125829666852951,
      "num_tokens": 78805767.0,
      "step": 40285
    },
    {
      "entropy": 5.939899492263794,
      "epoch": 4.0273904133553255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034671781126763263,
      "loss": 5.3605,
      "mean_token_accuracy": 0.2104242265224457,
      "num_tokens": 78815138.0,
      "step": 40290
    },
    {
      "entropy": 5.9383635997772215,
      "epoch": 4.027890238416554,
      "grad_norm": 1.0625,
      "learning_rate": 0.00034668398279935034,
      "loss": 5.3384,
      "mean_token_accuracy": 0.20427903085947036,
      "num_tokens": 78823958.0,
      "step": 40295
    },
    {
      "entropy": 5.960575675964355,
      "epoch": 4.028390063477783,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034665015252751747,
      "loss": 5.3884,
      "mean_token_accuracy": 0.19683534055948257,
      "num_tokens": 78833921.0,
      "step": 40300
    },
    {
      "entropy": 6.022540712356568,
      "epoch": 4.028889888539012,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003466163204529852,
      "loss": 5.4191,
      "mean_token_accuracy": 0.19812141954898835,
      "num_tokens": 78843640.0,
      "step": 40305
    },
    {
      "entropy": 5.866796159744263,
      "epoch": 4.02938971360024,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034658248657660476,
      "loss": 5.19,
      "mean_token_accuracy": 0.22634413540363313,
      "num_tokens": 78854917.0,
      "step": 40310
    },
    {
      "entropy": 5.944691562652588,
      "epoch": 4.029889538661468,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003465486508992274,
      "loss": 5.3613,
      "mean_token_accuracy": 0.21080266386270524,
      "num_tokens": 78865487.0,
      "step": 40315
    },
    {
      "entropy": 5.952260446548462,
      "epoch": 4.030389363722697,
      "grad_norm": 1.125,
      "learning_rate": 0.0003465148134217043,
      "loss": 5.3603,
      "mean_token_accuracy": 0.19959144592285155,
      "num_tokens": 78874759.0,
      "step": 40320
    },
    {
      "entropy": 5.818061971664429,
      "epoch": 4.0308891887839255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003464809741448871,
      "loss": 5.2347,
      "mean_token_accuracy": 0.2170904517173767,
      "num_tokens": 78884132.0,
      "step": 40325
    },
    {
      "entropy": 5.9578532695770265,
      "epoch": 4.031389013845154,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0003464471330696268,
      "loss": 5.3822,
      "mean_token_accuracy": 0.2012786462903023,
      "num_tokens": 78893730.0,
      "step": 40330
    },
    {
      "entropy": 5.899780893325806,
      "epoch": 4.031888838906383,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003464132901967751,
      "loss": 5.325,
      "mean_token_accuracy": 0.21410801112651826,
      "num_tokens": 78903769.0,
      "step": 40335
    },
    {
      "entropy": 5.845198392868042,
      "epoch": 4.032388663967612,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003463794455271833,
      "loss": 5.2753,
      "mean_token_accuracy": 0.21651267260313034,
      "num_tokens": 78913442.0,
      "step": 40340
    },
    {
      "entropy": 5.96634430885315,
      "epoch": 4.03288848902884,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034634559906170304,
      "loss": 5.3814,
      "mean_token_accuracy": 0.19981021136045457,
      "num_tokens": 78923292.0,
      "step": 40345
    },
    {
      "entropy": 6.018067455291748,
      "epoch": 4.033388314090068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000346311750801186,
      "loss": 5.347,
      "mean_token_accuracy": 0.2030012145638466,
      "num_tokens": 78933393.0,
      "step": 40350
    },
    {
      "entropy": 6.00525393486023,
      "epoch": 4.033888139151297,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003462779007464835,
      "loss": 5.3915,
      "mean_token_accuracy": 0.20600425451993942,
      "num_tokens": 78942319.0,
      "step": 40355
    },
    {
      "entropy": 5.910784101486206,
      "epoch": 4.0343879642125255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003462440488984475,
      "loss": 5.3096,
      "mean_token_accuracy": 0.2084883466362953,
      "num_tokens": 78952666.0,
      "step": 40360
    },
    {
      "entropy": 5.902736806869507,
      "epoch": 4.034887789273754,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034621019525792946,
      "loss": 5.3366,
      "mean_token_accuracy": 0.21162895113229752,
      "num_tokens": 78962443.0,
      "step": 40365
    },
    {
      "entropy": 5.953275156021118,
      "epoch": 4.035387614334983,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034617633982578124,
      "loss": 5.3901,
      "mean_token_accuracy": 0.2124612808227539,
      "num_tokens": 78972384.0,
      "step": 40370
    },
    {
      "entropy": 5.954825210571289,
      "epoch": 4.035887439396212,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003461424826028546,
      "loss": 5.3859,
      "mean_token_accuracy": 0.20425349473953247,
      "num_tokens": 78980799.0,
      "step": 40375
    },
    {
      "entropy": 5.867015409469604,
      "epoch": 4.03638726445744,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034610862359000144,
      "loss": 5.316,
      "mean_token_accuracy": 0.20376093536615372,
      "num_tokens": 78990753.0,
      "step": 40380
    },
    {
      "entropy": 5.909558439254761,
      "epoch": 4.036887089518668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003460747627880736,
      "loss": 5.3164,
      "mean_token_accuracy": 0.2109616592526436,
      "num_tokens": 79000160.0,
      "step": 40385
    },
    {
      "entropy": 5.913118839263916,
      "epoch": 4.037386914579897,
      "grad_norm": 0.984375,
      "learning_rate": 0.0003460409001979231,
      "loss": 5.3082,
      "mean_token_accuracy": 0.2071054145693779,
      "num_tokens": 79009602.0,
      "step": 40390
    },
    {
      "entropy": 5.9292699813842775,
      "epoch": 4.0378867396411255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034600703582040183,
      "loss": 5.398,
      "mean_token_accuracy": 0.2097340613603592,
      "num_tokens": 79019196.0,
      "step": 40395
    },
    {
      "entropy": 5.843757581710816,
      "epoch": 4.038386564702354,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034597316965636173,
      "loss": 5.2577,
      "mean_token_accuracy": 0.2197154939174652,
      "num_tokens": 79028142.0,
      "step": 40400
    },
    {
      "entropy": 5.987527894973755,
      "epoch": 4.038886389763583,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034593930170665496,
      "loss": 5.4357,
      "mean_token_accuracy": 0.2093260407447815,
      "num_tokens": 79038044.0,
      "step": 40405
    },
    {
      "entropy": 5.984679651260376,
      "epoch": 4.039386214824812,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034590543197213355,
      "loss": 5.4035,
      "mean_token_accuracy": 0.19948399364948272,
      "num_tokens": 79047462.0,
      "step": 40410
    },
    {
      "entropy": 5.890209054946899,
      "epoch": 4.0398860398860394,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034587156045364983,
      "loss": 5.2983,
      "mean_token_accuracy": 0.21161562502384185,
      "num_tokens": 79057454.0,
      "step": 40415
    },
    {
      "entropy": 5.9737237930297855,
      "epoch": 4.040385864947268,
      "grad_norm": 1.21875,
      "learning_rate": 0.00034583768715205584,
      "loss": 5.3631,
      "mean_token_accuracy": 0.20391895323991777,
      "num_tokens": 79066665.0,
      "step": 40420
    },
    {
      "entropy": 5.968359088897705,
      "epoch": 4.040885690008497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003458038120682038,
      "loss": 5.3485,
      "mean_token_accuracy": 0.20805672109127044,
      "num_tokens": 79076138.0,
      "step": 40425
    },
    {
      "entropy": 5.940025043487549,
      "epoch": 4.0413855150697255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034576993520294623,
      "loss": 5.2841,
      "mean_token_accuracy": 0.20452869534492493,
      "num_tokens": 79086378.0,
      "step": 40430
    },
    {
      "entropy": 5.988382625579834,
      "epoch": 4.041885340130954,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034573605655713514,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20143476873636246,
      "num_tokens": 79096751.0,
      "step": 40435
    },
    {
      "entropy": 5.9090766429901125,
      "epoch": 4.042385165192183,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034570217613162315,
      "loss": 5.2865,
      "mean_token_accuracy": 0.204510597884655,
      "num_tokens": 79106469.0,
      "step": 40440
    },
    {
      "entropy": 5.862514066696167,
      "epoch": 4.042884990253412,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034566829392726256,
      "loss": 5.3074,
      "mean_token_accuracy": 0.2074306532740593,
      "num_tokens": 79116411.0,
      "step": 40445
    },
    {
      "entropy": 5.929564189910889,
      "epoch": 4.0433848153146394,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003456344099449059,
      "loss": 5.4004,
      "mean_token_accuracy": 0.21007805764675141,
      "num_tokens": 79125595.0,
      "step": 40450
    },
    {
      "entropy": 5.886483001708984,
      "epoch": 4.043884640375868,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034560052418540564,
      "loss": 5.3189,
      "mean_token_accuracy": 0.2062122941017151,
      "num_tokens": 79134250.0,
      "step": 40455
    },
    {
      "entropy": 6.073524618148804,
      "epoch": 4.044384465437097,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003455666366496145,
      "loss": 5.5012,
      "mean_token_accuracy": 0.18846641778945922,
      "num_tokens": 79146249.0,
      "step": 40460
    },
    {
      "entropy": 5.91035532951355,
      "epoch": 4.0448842904983255,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034553274733838476,
      "loss": 5.2508,
      "mean_token_accuracy": 0.2163828894495964,
      "num_tokens": 79156128.0,
      "step": 40465
    },
    {
      "entropy": 5.942204093933105,
      "epoch": 4.045384115559554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003454988562525693,
      "loss": 5.3134,
      "mean_token_accuracy": 0.20887143760919571,
      "num_tokens": 79166399.0,
      "step": 40470
    },
    {
      "entropy": 5.960860347747802,
      "epoch": 4.045883940620783,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034546496339302077,
      "loss": 5.3377,
      "mean_token_accuracy": 0.21103584617376328,
      "num_tokens": 79177084.0,
      "step": 40475
    },
    {
      "entropy": 5.943429708480835,
      "epoch": 4.046383765682012,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003454310687605919,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20245476812124252,
      "num_tokens": 79185506.0,
      "step": 40480
    },
    {
      "entropy": 5.871845245361328,
      "epoch": 4.0468835907432394,
      "grad_norm": 1.25,
      "learning_rate": 0.00034539717235613544,
      "loss": 5.2378,
      "mean_token_accuracy": 0.2191609889268875,
      "num_tokens": 79195175.0,
      "step": 40485
    },
    {
      "entropy": 5.975759649276734,
      "epoch": 4.047383415804468,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003453632741805042,
      "loss": 5.4704,
      "mean_token_accuracy": 0.20083824843168258,
      "num_tokens": 79205022.0,
      "step": 40490
    },
    {
      "entropy": 5.9088624000549315,
      "epoch": 4.047883240865697,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003453293742345512,
      "loss": 5.2343,
      "mean_token_accuracy": 0.2220226854085922,
      "num_tokens": 79214219.0,
      "step": 40495
    },
    {
      "entropy": 5.9075957298278805,
      "epoch": 4.0483830659269255,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034529547251912926,
      "loss": 5.3737,
      "mean_token_accuracy": 0.20752676725387573,
      "num_tokens": 79223538.0,
      "step": 40500
    },
    {
      "entropy": 5.965269565582275,
      "epoch": 4.048882890988154,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034526156903509116,
      "loss": 5.3579,
      "mean_token_accuracy": 0.2063616096973419,
      "num_tokens": 79233056.0,
      "step": 40505
    },
    {
      "entropy": 5.929641962051392,
      "epoch": 4.049382716049383,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003452276637832902,
      "loss": 5.4089,
      "mean_token_accuracy": 0.20530281513929366,
      "num_tokens": 79242292.0,
      "step": 40510
    },
    {
      "entropy": 5.863769292831421,
      "epoch": 4.049882541110612,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003451937567645792,
      "loss": 5.2316,
      "mean_token_accuracy": 0.2173752561211586,
      "num_tokens": 79252273.0,
      "step": 40515
    },
    {
      "entropy": 5.9411444664001465,
      "epoch": 4.050382366171839,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034515984797981136,
      "loss": 5.3713,
      "mean_token_accuracy": 0.20793272107839583,
      "num_tokens": 79261086.0,
      "step": 40520
    },
    {
      "entropy": 5.965070199966431,
      "epoch": 4.050882191233068,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003451259374298398,
      "loss": 5.3779,
      "mean_token_accuracy": 0.20494178533554078,
      "num_tokens": 79270233.0,
      "step": 40525
    },
    {
      "entropy": 5.998249292373657,
      "epoch": 4.051382016294297,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034509202511551776,
      "loss": 5.4072,
      "mean_token_accuracy": 0.2030648782849312,
      "num_tokens": 79279905.0,
      "step": 40530
    },
    {
      "entropy": 5.976556873321533,
      "epoch": 4.0518818413555255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003450581110376984,
      "loss": 5.3978,
      "mean_token_accuracy": 0.20458509922027587,
      "num_tokens": 79289744.0,
      "step": 40535
    },
    {
      "entropy": 5.935463809967041,
      "epoch": 4.052381666416754,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003450241951972349,
      "loss": 5.4483,
      "mean_token_accuracy": 0.2058778554201126,
      "num_tokens": 79298190.0,
      "step": 40540
    },
    {
      "entropy": 5.939845895767212,
      "epoch": 4.052881491477983,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003449902775949807,
      "loss": 5.3591,
      "mean_token_accuracy": 0.2045484557747841,
      "num_tokens": 79307261.0,
      "step": 40545
    },
    {
      "entropy": 5.932804441452026,
      "epoch": 4.053381316539212,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034495635823178917,
      "loss": 5.329,
      "mean_token_accuracy": 0.2078299641609192,
      "num_tokens": 79316148.0,
      "step": 40550
    },
    {
      "entropy": 5.991153717041016,
      "epoch": 4.053881141600439,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003449224371085136,
      "loss": 5.4323,
      "mean_token_accuracy": 0.19452447295188904,
      "num_tokens": 79326218.0,
      "step": 40555
    },
    {
      "entropy": 6.0209136486053465,
      "epoch": 4.054380966661668,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003448885142260076,
      "loss": 5.4695,
      "mean_token_accuracy": 0.19556648284196854,
      "num_tokens": 79336045.0,
      "step": 40560
    },
    {
      "entropy": 5.845690488815308,
      "epoch": 4.054880791722897,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00034485458958512444,
      "loss": 5.2191,
      "mean_token_accuracy": 0.2232319712638855,
      "num_tokens": 79348579.0,
      "step": 40565
    },
    {
      "entropy": 5.897698545455933,
      "epoch": 4.0553806167841255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003448206631867179,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2069247379899025,
      "num_tokens": 79359358.0,
      "step": 40570
    },
    {
      "entropy": 5.94537844657898,
      "epoch": 4.055880441845354,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034478673503164143,
      "loss": 5.383,
      "mean_token_accuracy": 0.21251351833343507,
      "num_tokens": 79369376.0,
      "step": 40575
    },
    {
      "entropy": 5.927747583389282,
      "epoch": 4.056380266906583,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003447528051207486,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21077249348163604,
      "num_tokens": 79378260.0,
      "step": 40580
    },
    {
      "entropy": 5.923851060867309,
      "epoch": 4.056880091967812,
      "grad_norm": 1.125,
      "learning_rate": 0.0003447188734548932,
      "loss": 5.366,
      "mean_token_accuracy": 0.20893165618181228,
      "num_tokens": 79387681.0,
      "step": 40585
    },
    {
      "entropy": 5.964718961715699,
      "epoch": 4.057379917029039,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003446849400349289,
      "loss": 5.3654,
      "mean_token_accuracy": 0.20434602349996567,
      "num_tokens": 79396733.0,
      "step": 40590
    },
    {
      "entropy": 5.853693866729737,
      "epoch": 4.057879742090268,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003446510048617093,
      "loss": 5.2958,
      "mean_token_accuracy": 0.2208547443151474,
      "num_tokens": 79405664.0,
      "step": 40595
    },
    {
      "entropy": 5.759365177154541,
      "epoch": 4.058379567151497,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003446170679360884,
      "loss": 5.1571,
      "mean_token_accuracy": 0.23331172168254852,
      "num_tokens": 79415046.0,
      "step": 40600
    },
    {
      "entropy": 5.911055994033814,
      "epoch": 4.0588793922127255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003445831292589201,
      "loss": 5.4411,
      "mean_token_accuracy": 0.19494868367910384,
      "num_tokens": 79425139.0,
      "step": 40605
    },
    {
      "entropy": 5.999856996536255,
      "epoch": 4.059379217273954,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003445491888310581,
      "loss": 5.3843,
      "mean_token_accuracy": 0.20846618562936783,
      "num_tokens": 79434973.0,
      "step": 40610
    },
    {
      "entropy": 6.018252944946289,
      "epoch": 4.059879042335183,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003445152466533564,
      "loss": 5.4255,
      "mean_token_accuracy": 0.19980820268392563,
      "num_tokens": 79443880.0,
      "step": 40615
    },
    {
      "entropy": 5.844971084594727,
      "epoch": 4.060378867396412,
      "grad_norm": 1.15625,
      "learning_rate": 0.000344481302726669,
      "loss": 5.2408,
      "mean_token_accuracy": 0.21788232624530793,
      "num_tokens": 79454300.0,
      "step": 40620
    },
    {
      "entropy": 5.850299882888794,
      "epoch": 4.060878692457639,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034444735705184984,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21470244824886323,
      "num_tokens": 79463771.0,
      "step": 40625
    },
    {
      "entropy": 5.914721298217773,
      "epoch": 4.061378517518868,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003444134096297531,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20911143273115157,
      "num_tokens": 79473227.0,
      "step": 40630
    },
    {
      "entropy": 5.9612266540527346,
      "epoch": 4.061878342580097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003443794604612328,
      "loss": 5.3042,
      "mean_token_accuracy": 0.20677004158496856,
      "num_tokens": 79483797.0,
      "step": 40635
    },
    {
      "entropy": 5.955880069732666,
      "epoch": 4.0623781676413255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003443455095471432,
      "loss": 5.3006,
      "mean_token_accuracy": 0.21269181221723557,
      "num_tokens": 79492773.0,
      "step": 40640
    },
    {
      "entropy": 5.948481130599975,
      "epoch": 4.062877992702554,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00034431155688833837,
      "loss": 5.4132,
      "mean_token_accuracy": 0.19806791096925735,
      "num_tokens": 79502795.0,
      "step": 40645
    },
    {
      "entropy": 5.943785572052002,
      "epoch": 4.063377817763783,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003442776024856727,
      "loss": 5.4128,
      "mean_token_accuracy": 0.19458921253681183,
      "num_tokens": 79513374.0,
      "step": 40650
    },
    {
      "entropy": 5.9521277904510494,
      "epoch": 4.063877642825012,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034424364634000023,
      "loss": 5.3874,
      "mean_token_accuracy": 0.21215200424194336,
      "num_tokens": 79523542.0,
      "step": 40655
    },
    {
      "entropy": 6.018523645401001,
      "epoch": 4.064377467886239,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003442096884521754,
      "loss": 5.4313,
      "mean_token_accuracy": 0.19862000197172164,
      "num_tokens": 79533143.0,
      "step": 40660
    },
    {
      "entropy": 5.933532571792602,
      "epoch": 4.064877292947468,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034417572882305275,
      "loss": 5.362,
      "mean_token_accuracy": 0.2057307779788971,
      "num_tokens": 79542683.0,
      "step": 40665
    },
    {
      "entropy": 5.956983280181885,
      "epoch": 4.065377118008697,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003441417674534865,
      "loss": 5.3644,
      "mean_token_accuracy": 0.21388774514198303,
      "num_tokens": 79552521.0,
      "step": 40670
    },
    {
      "entropy": 5.988804626464844,
      "epoch": 4.0658769430699255,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003441078043443313,
      "loss": 5.3618,
      "mean_token_accuracy": 0.20080869197845458,
      "num_tokens": 79561843.0,
      "step": 40675
    },
    {
      "entropy": 5.8964423656463625,
      "epoch": 4.066376768131154,
      "grad_norm": 1.234375,
      "learning_rate": 0.00034407383949644146,
      "loss": 5.2554,
      "mean_token_accuracy": 0.2095391109585762,
      "num_tokens": 79571153.0,
      "step": 40680
    },
    {
      "entropy": 5.910944652557373,
      "epoch": 4.066876593192383,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003440398729106715,
      "loss": 5.332,
      "mean_token_accuracy": 0.21104592084884644,
      "num_tokens": 79581126.0,
      "step": 40685
    },
    {
      "entropy": 5.9820885181427,
      "epoch": 4.067376418253612,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003440059045878761,
      "loss": 5.4645,
      "mean_token_accuracy": 0.1945442244410515,
      "num_tokens": 79590647.0,
      "step": 40690
    },
    {
      "entropy": 5.879202842712402,
      "epoch": 4.067876243314839,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034397193452891003,
      "loss": 5.2583,
      "mean_token_accuracy": 0.21667467504739762,
      "num_tokens": 79599723.0,
      "step": 40695
    },
    {
      "entropy": 5.946272277832032,
      "epoch": 4.068376068376068,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003439379627346277,
      "loss": 5.4062,
      "mean_token_accuracy": 0.20856606364250183,
      "num_tokens": 79609699.0,
      "step": 40700
    },
    {
      "entropy": 5.829761171340943,
      "epoch": 4.068875893437297,
      "grad_norm": 1.03125,
      "learning_rate": 0.00034390398920588404,
      "loss": 5.2121,
      "mean_token_accuracy": 0.21568648666143417,
      "num_tokens": 79619173.0,
      "step": 40705
    },
    {
      "entropy": 5.880647325515747,
      "epoch": 4.0693757184985255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003438700139435338,
      "loss": 5.2509,
      "mean_token_accuracy": 0.21603213548660277,
      "num_tokens": 79628685.0,
      "step": 40710
    },
    {
      "entropy": 6.020151042938233,
      "epoch": 4.069875543559754,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003438360369484317,
      "loss": 5.354,
      "mean_token_accuracy": 0.20610783249139786,
      "num_tokens": 79639117.0,
      "step": 40715
    },
    {
      "entropy": 5.995475339889526,
      "epoch": 4.070375368620983,
      "grad_norm": 1.125,
      "learning_rate": 0.0003438020582214325,
      "loss": 5.44,
      "mean_token_accuracy": 0.19955674409866334,
      "num_tokens": 79649502.0,
      "step": 40720
    },
    {
      "entropy": 5.986223602294922,
      "epoch": 4.070875193682212,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003437680777633913,
      "loss": 5.4209,
      "mean_token_accuracy": 0.1977478325366974,
      "num_tokens": 79658787.0,
      "step": 40725
    },
    {
      "entropy": 5.966217708587647,
      "epoch": 4.071375018743439,
      "grad_norm": 1.171875,
      "learning_rate": 0.000343734095575163,
      "loss": 5.4061,
      "mean_token_accuracy": 0.20138639360666274,
      "num_tokens": 79667979.0,
      "step": 40730
    },
    {
      "entropy": 5.9863533020019535,
      "epoch": 4.071874843804668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003437001116576025,
      "loss": 5.4521,
      "mean_token_accuracy": 0.199602472782135,
      "num_tokens": 79677997.0,
      "step": 40735
    },
    {
      "entropy": 5.931427907943726,
      "epoch": 4.072374668865897,
      "grad_norm": 1.078125,
      "learning_rate": 0.00034366612601156486,
      "loss": 5.406,
      "mean_token_accuracy": 0.20527503043413162,
      "num_tokens": 79687205.0,
      "step": 40740
    },
    {
      "entropy": 5.933200836181641,
      "epoch": 4.0728744939271255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034363213863790527,
      "loss": 5.3497,
      "mean_token_accuracy": 0.20568083971738815,
      "num_tokens": 79698425.0,
      "step": 40745
    },
    {
      "entropy": 5.910302543640137,
      "epoch": 4.073374318988354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034359814953747856,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21339871287345885,
      "num_tokens": 79707923.0,
      "step": 40750
    },
    {
      "entropy": 5.939322376251221,
      "epoch": 4.073874144049583,
      "grad_norm": 1.125,
      "learning_rate": 0.00034356415871114013,
      "loss": 5.3485,
      "mean_token_accuracy": 0.19985028803348542,
      "num_tokens": 79719294.0,
      "step": 40755
    },
    {
      "entropy": 5.989749956130981,
      "epoch": 4.074373969110812,
      "grad_norm": 1.203125,
      "learning_rate": 0.00034353016615974513,
      "loss": 5.5072,
      "mean_token_accuracy": 0.1988230064511299,
      "num_tokens": 79728400.0,
      "step": 40760
    },
    {
      "entropy": 5.9466729164123535,
      "epoch": 4.074873794172039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003434961718841488,
      "loss": 5.2982,
      "mean_token_accuracy": 0.21145112067461014,
      "num_tokens": 79737749.0,
      "step": 40765
    },
    {
      "entropy": 5.953240489959716,
      "epoch": 4.075373619233268,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034346217588520636,
      "loss": 5.398,
      "mean_token_accuracy": 0.19724612534046174,
      "num_tokens": 79747438.0,
      "step": 40770
    },
    {
      "entropy": 5.904358386993408,
      "epoch": 4.075873444294497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003434281781637732,
      "loss": 5.3182,
      "mean_token_accuracy": 0.21202570348978042,
      "num_tokens": 79757820.0,
      "step": 40775
    },
    {
      "entropy": 5.98792085647583,
      "epoch": 4.0763732693557255,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00034339417872070464,
      "loss": 5.4504,
      "mean_token_accuracy": 0.1993762344121933,
      "num_tokens": 79767827.0,
      "step": 40780
    },
    {
      "entropy": 6.030006551742554,
      "epoch": 4.076873094416954,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034336017755685623,
      "loss": 5.441,
      "mean_token_accuracy": 0.1988022118806839,
      "num_tokens": 79777687.0,
      "step": 40785
    },
    {
      "entropy": 5.857165193557739,
      "epoch": 4.077372919478183,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003433261746730833,
      "loss": 5.3095,
      "mean_token_accuracy": 0.21236782222986222,
      "num_tokens": 79788111.0,
      "step": 40790
    },
    {
      "entropy": 5.867184638977051,
      "epoch": 4.077872744539412,
      "grad_norm": 1.125,
      "learning_rate": 0.00034329217007024145,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21341889053583146,
      "num_tokens": 79796897.0,
      "step": 40795
    },
    {
      "entropy": 5.848789882659912,
      "epoch": 4.078372569600639,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003432581637491862,
      "loss": 5.2057,
      "mean_token_accuracy": 0.2270244225859642,
      "num_tokens": 79806759.0,
      "step": 40800
    },
    {
      "entropy": 6.00701675415039,
      "epoch": 4.078872394661868,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000343224155710773,
      "loss": 5.4096,
      "mean_token_accuracy": 0.20176855474710464,
      "num_tokens": 79817125.0,
      "step": 40805
    },
    {
      "entropy": 6.003981018066407,
      "epoch": 4.079372219723097,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034319014595585764,
      "loss": 5.4598,
      "mean_token_accuracy": 0.19425539672374725,
      "num_tokens": 79828114.0,
      "step": 40810
    },
    {
      "entropy": 5.975851106643677,
      "epoch": 4.0798720447843255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034315613448529574,
      "loss": 5.3441,
      "mean_token_accuracy": 0.2040207639336586,
      "num_tokens": 79838427.0,
      "step": 40815
    },
    {
      "entropy": 5.967152547836304,
      "epoch": 4.080371869845554,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003431221212999431,
      "loss": 5.3379,
      "mean_token_accuracy": 0.21280634105205537,
      "num_tokens": 79847620.0,
      "step": 40820
    },
    {
      "entropy": 5.845602321624756,
      "epoch": 4.080871694906783,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034308810640065534,
      "loss": 5.2687,
      "mean_token_accuracy": 0.22004587352275848,
      "num_tokens": 79858061.0,
      "step": 40825
    },
    {
      "entropy": 5.9569995403289795,
      "epoch": 4.081371519968012,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003430540897882884,
      "loss": 5.3991,
      "mean_token_accuracy": 0.2034058004617691,
      "num_tokens": 79868902.0,
      "step": 40830
    },
    {
      "entropy": 5.97858567237854,
      "epoch": 4.081871345029239,
      "grad_norm": 1.34375,
      "learning_rate": 0.000343020071463698,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20800871104002,
      "num_tokens": 79879763.0,
      "step": 40835
    },
    {
      "entropy": 5.902143669128418,
      "epoch": 4.082371170090468,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034298605142774014,
      "loss": 5.324,
      "mean_token_accuracy": 0.20630838274955748,
      "num_tokens": 79890014.0,
      "step": 40840
    },
    {
      "entropy": 5.742364382743835,
      "epoch": 4.082870995151697,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003429520296812708,
      "loss": 5.1646,
      "mean_token_accuracy": 0.22970276474952697,
      "num_tokens": 79898978.0,
      "step": 40845
    },
    {
      "entropy": 5.954917907714844,
      "epoch": 4.0833708202129255,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003429180062251458,
      "loss": 5.44,
      "mean_token_accuracy": 0.19987183064222336,
      "num_tokens": 79908706.0,
      "step": 40850
    },
    {
      "entropy": 5.936946439743042,
      "epoch": 4.083870645274154,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003428839810602212,
      "loss": 5.3027,
      "mean_token_accuracy": 0.21023423671722413,
      "num_tokens": 79918916.0,
      "step": 40855
    },
    {
      "entropy": 6.052512454986572,
      "epoch": 4.084370470335383,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003428499541873532,
      "loss": 5.416,
      "mean_token_accuracy": 0.20650901049375534,
      "num_tokens": 79927289.0,
      "step": 40860
    },
    {
      "entropy": 5.924692440032959,
      "epoch": 4.084870295396612,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034281592560739775,
      "loss": 5.2671,
      "mean_token_accuracy": 0.21557008922100068,
      "num_tokens": 79936141.0,
      "step": 40865
    },
    {
      "entropy": 5.937809753417969,
      "epoch": 4.085370120457839,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003427818953212111,
      "loss": 5.3444,
      "mean_token_accuracy": 0.2146711602807045,
      "num_tokens": 79946136.0,
      "step": 40870
    },
    {
      "entropy": 5.898311805725098,
      "epoch": 4.085869945519068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034274786332964943,
      "loss": 5.2282,
      "mean_token_accuracy": 0.21777033507823945,
      "num_tokens": 79956747.0,
      "step": 40875
    },
    {
      "entropy": 5.789913845062256,
      "epoch": 4.086369770580297,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003427138296335688,
      "loss": 5.2809,
      "mean_token_accuracy": 0.21754397004842757,
      "num_tokens": 79966694.0,
      "step": 40880
    },
    {
      "entropy": 5.940693283081055,
      "epoch": 4.0868695956415255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003426797942338258,
      "loss": 5.3695,
      "mean_token_accuracy": 0.20061712861061096,
      "num_tokens": 79976334.0,
      "step": 40885
    },
    {
      "entropy": 5.940596342086792,
      "epoch": 4.087369420702754,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003426457571312766,
      "loss": 5.3419,
      "mean_token_accuracy": 0.20846354961395264,
      "num_tokens": 79985520.0,
      "step": 40890
    },
    {
      "entropy": 5.871496248245239,
      "epoch": 4.087869245763983,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034261171832677744,
      "loss": 5.278,
      "mean_token_accuracy": 0.20871687382459642,
      "num_tokens": 79995023.0,
      "step": 40895
    },
    {
      "entropy": 5.924239587783814,
      "epoch": 4.088369070825211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003425776778211849,
      "loss": 5.422,
      "mean_token_accuracy": 0.20100004374980926,
      "num_tokens": 80005298.0,
      "step": 40900
    },
    {
      "entropy": 5.993998050689697,
      "epoch": 4.088868895886439,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003425436356153554,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20314888209104537,
      "num_tokens": 80015057.0,
      "step": 40905
    },
    {
      "entropy": 5.99723596572876,
      "epoch": 4.089368720947668,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034250959171014536,
      "loss": 5.3577,
      "mean_token_accuracy": 0.2092099517583847,
      "num_tokens": 80024419.0,
      "step": 40910
    },
    {
      "entropy": 5.9625848770141605,
      "epoch": 4.089868546008897,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003424755461064114,
      "loss": 5.3887,
      "mean_token_accuracy": 0.20518414080142974,
      "num_tokens": 80034845.0,
      "step": 40915
    },
    {
      "entropy": 6.016858243942261,
      "epoch": 4.0903683710701255,
      "grad_norm": 1.28125,
      "learning_rate": 0.00034244149880501007,
      "loss": 5.3923,
      "mean_token_accuracy": 0.1986510306596756,
      "num_tokens": 80044373.0,
      "step": 40920
    },
    {
      "entropy": 6.03705472946167,
      "epoch": 4.090868196131354,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000342407449806798,
      "loss": 5.4257,
      "mean_token_accuracy": 0.20001767724752426,
      "num_tokens": 80055454.0,
      "step": 40925
    },
    {
      "entropy": 5.864574241638183,
      "epoch": 4.091368021192583,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00034237339911263176,
      "loss": 5.2653,
      "mean_token_accuracy": 0.21728247851133348,
      "num_tokens": 80065554.0,
      "step": 40930
    },
    {
      "entropy": 5.961031293869018,
      "epoch": 4.091867846253811,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034233934672336827,
      "loss": 5.4381,
      "mean_token_accuracy": 0.20052247643470764,
      "num_tokens": 80074962.0,
      "step": 40935
    },
    {
      "entropy": 6.085455846786499,
      "epoch": 4.092367671315039,
      "grad_norm": 1.109375,
      "learning_rate": 0.000342305292639864,
      "loss": 5.4342,
      "mean_token_accuracy": 0.1933562234044075,
      "num_tokens": 80084105.0,
      "step": 40940
    },
    {
      "entropy": 5.9834003925323485,
      "epoch": 4.092867496376268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003422712368629759,
      "loss": 5.4557,
      "mean_token_accuracy": 0.20147402286529542,
      "num_tokens": 80095032.0,
      "step": 40945
    },
    {
      "entropy": 5.913961553573609,
      "epoch": 4.093367321437497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00034223717939356083,
      "loss": 5.3848,
      "mean_token_accuracy": 0.20626039057970047,
      "num_tokens": 80104667.0,
      "step": 40950
    },
    {
      "entropy": 5.89668231010437,
      "epoch": 4.0938671464987255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003422031202324757,
      "loss": 5.3232,
      "mean_token_accuracy": 0.20666970163583756,
      "num_tokens": 80114566.0,
      "step": 40955
    },
    {
      "entropy": 5.9530120372772215,
      "epoch": 4.094366971559954,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003421690593805773,
      "loss": 5.3614,
      "mean_token_accuracy": 0.20238589346408845,
      "num_tokens": 80126228.0,
      "step": 40960
    },
    {
      "entropy": 5.99973087310791,
      "epoch": 4.094866796621183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034213499683872264,
      "loss": 5.4061,
      "mean_token_accuracy": 0.19882909059524537,
      "num_tokens": 80136034.0,
      "step": 40965
    },
    {
      "entropy": 5.980393409729004,
      "epoch": 4.095366621682411,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034210093260776865,
      "loss": 5.3809,
      "mean_token_accuracy": 0.20779024362564086,
      "num_tokens": 80146784.0,
      "step": 40970
    },
    {
      "entropy": 5.868754577636719,
      "epoch": 4.095866446743639,
      "grad_norm": 1.109375,
      "learning_rate": 0.00034206686668857255,
      "loss": 5.2048,
      "mean_token_accuracy": 0.21960227638483049,
      "num_tokens": 80156777.0,
      "step": 40975
    },
    {
      "entropy": 5.774506855010986,
      "epoch": 4.096366271804868,
      "grad_norm": 1.296875,
      "learning_rate": 0.00034203279908199137,
      "loss": 5.1664,
      "mean_token_accuracy": 0.22568369209766387,
      "num_tokens": 80167486.0,
      "step": 40980
    },
    {
      "entropy": 5.931567621231079,
      "epoch": 4.096866096866097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003419987297888822,
      "loss": 5.3463,
      "mean_token_accuracy": 0.2118603527545929,
      "num_tokens": 80176618.0,
      "step": 40985
    },
    {
      "entropy": 5.921817445755005,
      "epoch": 4.0973659219273255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003419646588101022,
      "loss": 5.3036,
      "mean_token_accuracy": 0.2178333207964897,
      "num_tokens": 80186334.0,
      "step": 40990
    },
    {
      "entropy": 5.886446380615235,
      "epoch": 4.097865746988554,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034193058614650866,
      "loss": 5.2568,
      "mean_token_accuracy": 0.21438699662685395,
      "num_tokens": 80195871.0,
      "step": 40995
    },
    {
      "entropy": 5.93344144821167,
      "epoch": 4.098365572049783,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003418965117989587,
      "loss": 5.3067,
      "mean_token_accuracy": 0.21177975535392762,
      "num_tokens": 80205855.0,
      "step": 41000
    },
    {
      "entropy": 5.898735904693604,
      "epoch": 4.098865397111011,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034186243576830985,
      "loss": 5.2755,
      "mean_token_accuracy": 0.21610288619995116,
      "num_tokens": 80215010.0,
      "step": 41005
    },
    {
      "entropy": 5.812387418746948,
      "epoch": 4.099365222172239,
      "grad_norm": 1.40625,
      "learning_rate": 0.00034182835805541916,
      "loss": 5.1377,
      "mean_token_accuracy": 0.22420572787523269,
      "num_tokens": 80223710.0,
      "step": 41010
    },
    {
      "entropy": 5.930747222900391,
      "epoch": 4.099865047233468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003417942786611443,
      "loss": 5.3577,
      "mean_token_accuracy": 0.2029957264661789,
      "num_tokens": 80232821.0,
      "step": 41015
    },
    {
      "entropy": 5.918143320083618,
      "epoch": 4.100364872294697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003417601975863425,
      "loss": 5.3841,
      "mean_token_accuracy": 0.20281955152750014,
      "num_tokens": 80242339.0,
      "step": 41020
    },
    {
      "entropy": 6.017640542984009,
      "epoch": 4.1008646973559255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034172611483187143,
      "loss": 5.4275,
      "mean_token_accuracy": 0.20312604159116746,
      "num_tokens": 80251965.0,
      "step": 41025
    },
    {
      "entropy": 5.931193542480469,
      "epoch": 4.101364522417154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003416920303985884,
      "loss": 5.3528,
      "mean_token_accuracy": 0.20933321714401246,
      "num_tokens": 80260859.0,
      "step": 41030
    },
    {
      "entropy": 5.84721736907959,
      "epoch": 4.101864347478383,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034165794428735104,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21384650468826294,
      "num_tokens": 80270056.0,
      "step": 41035
    },
    {
      "entropy": 5.948648118972779,
      "epoch": 4.102364172539611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00034162385649901694,
      "loss": 5.4667,
      "mean_token_accuracy": 0.19687893986701965,
      "num_tokens": 80279574.0,
      "step": 41040
    },
    {
      "entropy": 6.01546082496643,
      "epoch": 4.102863997600839,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034158976703444377,
      "loss": 5.4483,
      "mean_token_accuracy": 0.1984674870967865,
      "num_tokens": 80288510.0,
      "step": 41045
    },
    {
      "entropy": 5.9493029594421385,
      "epoch": 4.103363822662068,
      "grad_norm": 1.15625,
      "learning_rate": 0.00034155567589448925,
      "loss": 5.2694,
      "mean_token_accuracy": 0.21197342574596406,
      "num_tokens": 80298292.0,
      "step": 41050
    },
    {
      "entropy": 6.012387323379516,
      "epoch": 4.103863647723297,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000341521583080011,
      "loss": 5.5337,
      "mean_token_accuracy": 0.18909737318754197,
      "num_tokens": 80308756.0,
      "step": 41055
    },
    {
      "entropy": 5.927402782440185,
      "epoch": 4.1043634727845255,
      "grad_norm": 1.25,
      "learning_rate": 0.0003414874885918669,
      "loss": 5.3332,
      "mean_token_accuracy": 0.20778064727783202,
      "num_tokens": 80318553.0,
      "step": 41060
    },
    {
      "entropy": 5.899279260635376,
      "epoch": 4.104863297845754,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00034145339243091456,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21151027977466583,
      "num_tokens": 80327952.0,
      "step": 41065
    },
    {
      "entropy": 5.926576423645019,
      "epoch": 4.105363122906983,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003414192945980121,
      "loss": 5.3853,
      "mean_token_accuracy": 0.1965849816799164,
      "num_tokens": 80338712.0,
      "step": 41070
    },
    {
      "entropy": 5.94453911781311,
      "epoch": 4.105862947968211,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003413851950940172,
      "loss": 5.3874,
      "mean_token_accuracy": 0.2002808317542076,
      "num_tokens": 80349300.0,
      "step": 41075
    },
    {
      "entropy": 5.945654678344726,
      "epoch": 4.106362773029439,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034135109391978785,
      "loss": 5.371,
      "mean_token_accuracy": 0.20884331464767455,
      "num_tokens": 80358746.0,
      "step": 41080
    },
    {
      "entropy": 5.925637817382812,
      "epoch": 4.106862598090668,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034131699107618215,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20830946266651154,
      "num_tokens": 80367505.0,
      "step": 41085
    },
    {
      "entropy": 5.952963924407959,
      "epoch": 4.107362423151897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00034128288656405794,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19970128536224366,
      "num_tokens": 80377875.0,
      "step": 41090
    },
    {
      "entropy": 6.081288766860962,
      "epoch": 4.1078622482131255,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034124878038427337,
      "loss": 5.4995,
      "mean_token_accuracy": 0.19674088954925537,
      "num_tokens": 80386143.0,
      "step": 41095
    },
    {
      "entropy": 6.005099105834961,
      "epoch": 4.108362073274354,
      "grad_norm": 1.09375,
      "learning_rate": 0.00034121467253768657,
      "loss": 5.4215,
      "mean_token_accuracy": 0.19929047226905822,
      "num_tokens": 80396658.0,
      "step": 41100
    },
    {
      "entropy": 5.964631462097168,
      "epoch": 4.108861898335583,
      "grad_norm": 1.4375,
      "learning_rate": 0.0003411805630251556,
      "loss": 5.3283,
      "mean_token_accuracy": 0.21272628456354142,
      "num_tokens": 80406063.0,
      "step": 41105
    },
    {
      "entropy": 6.012062072753906,
      "epoch": 4.109361723396811,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00034114645184753873,
      "loss": 5.4878,
      "mean_token_accuracy": 0.20237526744604112,
      "num_tokens": 80415298.0,
      "step": 41110
    },
    {
      "entropy": 5.938869667053223,
      "epoch": 4.109861548458039,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034111233900569415,
      "loss": 5.3675,
      "mean_token_accuracy": 0.19624172449111937,
      "num_tokens": 80425689.0,
      "step": 41115
    },
    {
      "entropy": 5.969009733200073,
      "epoch": 4.110361373519268,
      "grad_norm": 1.328125,
      "learning_rate": 0.00034107822450048014,
      "loss": 5.3808,
      "mean_token_accuracy": 0.20075520724058152,
      "num_tokens": 80435999.0,
      "step": 41120
    },
    {
      "entropy": 5.903009700775146,
      "epoch": 4.110861198580497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000341044108332755,
      "loss": 5.2618,
      "mean_token_accuracy": 0.20952228605747222,
      "num_tokens": 80445025.0,
      "step": 41125
    },
    {
      "entropy": 5.934450435638428,
      "epoch": 4.1113610236417255,
      "grad_norm": 1.140625,
      "learning_rate": 0.00034100999050337705,
      "loss": 5.4194,
      "mean_token_accuracy": 0.19539570063352585,
      "num_tokens": 80455375.0,
      "step": 41130
    },
    {
      "entropy": 5.888372945785522,
      "epoch": 4.111860848702954,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034097587101320474,
      "loss": 5.27,
      "mean_token_accuracy": 0.20910421162843704,
      "num_tokens": 80464202.0,
      "step": 41135
    },
    {
      "entropy": 5.9119758129119875,
      "epoch": 4.112360673764183,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034094174986309657,
      "loss": 5.3543,
      "mean_token_accuracy": 0.20011022984981536,
      "num_tokens": 80473216.0,
      "step": 41140
    },
    {
      "entropy": 5.9880396842956545,
      "epoch": 4.112860498825411,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003409076270539108,
      "loss": 5.4148,
      "mean_token_accuracy": 0.2044037327170372,
      "num_tokens": 80482455.0,
      "step": 41145
    },
    {
      "entropy": 5.946517133712769,
      "epoch": 4.113360323886639,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003408735025865062,
      "loss": 5.3427,
      "mean_token_accuracy": 0.20726565569639205,
      "num_tokens": 80492319.0,
      "step": 41150
    },
    {
      "entropy": 5.982586765289307,
      "epoch": 4.113860148947868,
      "grad_norm": 1.125,
      "learning_rate": 0.0003408393764617412,
      "loss": 5.429,
      "mean_token_accuracy": 0.1965298429131508,
      "num_tokens": 80502798.0,
      "step": 41155
    },
    {
      "entropy": 5.9261887550354,
      "epoch": 4.114359974009097,
      "grad_norm": 1.265625,
      "learning_rate": 0.00034080524868047445,
      "loss": 5.3544,
      "mean_token_accuracy": 0.2051659420132637,
      "num_tokens": 80511766.0,
      "step": 41160
    },
    {
      "entropy": 5.901782655715943,
      "epoch": 4.1148597990703255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003407711192435646,
      "loss": 5.276,
      "mean_token_accuracy": 0.21831444948911666,
      "num_tokens": 80521728.0,
      "step": 41165
    },
    {
      "entropy": 5.920960712432861,
      "epoch": 4.115359624131554,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003407369881518702,
      "loss": 5.3192,
      "mean_token_accuracy": 0.20900093764066696,
      "num_tokens": 80530543.0,
      "step": 41170
    },
    {
      "entropy": 5.900981283187866,
      "epoch": 4.115859449192783,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003407028554062502,
      "loss": 5.3018,
      "mean_token_accuracy": 0.2074249267578125,
      "num_tokens": 80540402.0,
      "step": 41175
    },
    {
      "entropy": 5.903803920745849,
      "epoch": 4.116359274254011,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003406687210075632,
      "loss": 5.3163,
      "mean_token_accuracy": 0.20533705949783326,
      "num_tokens": 80550777.0,
      "step": 41180
    },
    {
      "entropy": 5.938193607330322,
      "epoch": 4.116859099315239,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003406345849566681,
      "loss": 5.3893,
      "mean_token_accuracy": 0.20941306203603743,
      "num_tokens": 80560236.0,
      "step": 41185
    },
    {
      "entropy": 5.908311939239502,
      "epoch": 4.117358924376468,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003406004472544237,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20549913197755815,
      "num_tokens": 80569537.0,
      "step": 41190
    },
    {
      "entropy": 6.019424915313721,
      "epoch": 4.117858749437697,
      "grad_norm": 1.375,
      "learning_rate": 0.0003405663079016889,
      "loss": 5.4066,
      "mean_token_accuracy": 0.2009757161140442,
      "num_tokens": 80578825.0,
      "step": 41195
    },
    {
      "entropy": 5.92310848236084,
      "epoch": 4.1183585744989255,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003405321668993228,
      "loss": 5.3902,
      "mean_token_accuracy": 0.20970139354467393,
      "num_tokens": 80589620.0,
      "step": 41200
    },
    {
      "entropy": 5.880673599243164,
      "epoch": 4.118858399560154,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003404980242481842,
      "loss": 5.2113,
      "mean_token_accuracy": 0.22377046048641205,
      "num_tokens": 80599489.0,
      "step": 41205
    },
    {
      "entropy": 5.926363372802735,
      "epoch": 4.119358224621383,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00034046387994913216,
      "loss": 5.3679,
      "mean_token_accuracy": 0.2043956473469734,
      "num_tokens": 80610100.0,
      "step": 41210
    },
    {
      "entropy": 6.011868953704834,
      "epoch": 4.119858049682611,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003404297340030257,
      "loss": 5.3962,
      "mean_token_accuracy": 0.19984215945005418,
      "num_tokens": 80619949.0,
      "step": 41215
    },
    {
      "entropy": 6.017270851135254,
      "epoch": 4.120357874743839,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00034039558641072397,
      "loss": 5.4325,
      "mean_token_accuracy": 0.19816361516714096,
      "num_tokens": 80630194.0,
      "step": 41220
    },
    {
      "entropy": 5.968650817871094,
      "epoch": 4.120857699805068,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003403614371730861,
      "loss": 5.4007,
      "mean_token_accuracy": 0.20221034586429595,
      "num_tokens": 80640394.0,
      "step": 41225
    },
    {
      "entropy": 5.941448640823364,
      "epoch": 4.121357524866297,
      "grad_norm": 1.1875,
      "learning_rate": 0.00034032728629097137,
      "loss": 5.325,
      "mean_token_accuracy": 0.20583885610103608,
      "num_tokens": 80651056.0,
      "step": 41230
    },
    {
      "entropy": 5.955168008804321,
      "epoch": 4.1218573499275255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003402931337652389,
      "loss": 5.4066,
      "mean_token_accuracy": 0.2060915306210518,
      "num_tokens": 80661370.0,
      "step": 41235
    },
    {
      "entropy": 5.952804136276245,
      "epoch": 4.122357174988754,
      "grad_norm": 1.234375,
      "learning_rate": 0.000340258979596748,
      "loss": 5.4026,
      "mean_token_accuracy": 0.20637848377227783,
      "num_tokens": 80670627.0,
      "step": 41240
    },
    {
      "entropy": 5.926789569854736,
      "epoch": 4.122857000049983,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003402248237863579,
      "loss": 5.3444,
      "mean_token_accuracy": 0.21273787766695024,
      "num_tokens": 80680535.0,
      "step": 41245
    },
    {
      "entropy": 5.857923698425293,
      "epoch": 4.123356825111211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003401906663349281,
      "loss": 5.2333,
      "mean_token_accuracy": 0.21956729739904404,
      "num_tokens": 80689808.0,
      "step": 41250
    },
    {
      "entropy": 5.908638334274292,
      "epoch": 4.123856650172439,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00034015650724331776,
      "loss": 5.339,
      "mean_token_accuracy": 0.21885989010334014,
      "num_tokens": 80698537.0,
      "step": 41255
    },
    {
      "entropy": 5.9716168403625485,
      "epoch": 4.124356475233668,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00034012234651238654,
      "loss": 5.4194,
      "mean_token_accuracy": 0.20769248008728028,
      "num_tokens": 80709044.0,
      "step": 41260
    },
    {
      "entropy": 5.936302328109742,
      "epoch": 4.124856300294897,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003400881841429939,
      "loss": 5.2683,
      "mean_token_accuracy": 0.2230658322572708,
      "num_tokens": 80718644.0,
      "step": 41265
    },
    {
      "entropy": 5.90023250579834,
      "epoch": 4.1253561253561255,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003400540201359992,
      "loss": 5.2904,
      "mean_token_accuracy": 0.21239131838083267,
      "num_tokens": 80728410.0,
      "step": 41270
    },
    {
      "entropy": 5.936369705200195,
      "epoch": 4.125855950417354,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003400198544922622,
      "loss": 5.4239,
      "mean_token_accuracy": 0.19715702384710312,
      "num_tokens": 80739075.0,
      "step": 41275
    },
    {
      "entropy": 5.927850294113159,
      "epoch": 4.126355775478583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003399856872126424,
      "loss": 5.2832,
      "mean_token_accuracy": 0.2107976868748665,
      "num_tokens": 80748663.0,
      "step": 41280
    },
    {
      "entropy": 5.937838792800903,
      "epoch": 4.126855600539811,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033995151829799926,
      "loss": 5.2971,
      "mean_token_accuracy": 0.20908122956752778,
      "num_tokens": 80758164.0,
      "step": 41285
    },
    {
      "entropy": 5.8449688911437985,
      "epoch": 4.127355425601039,
      "grad_norm": 1.296875,
      "learning_rate": 0.00033991734774919274,
      "loss": 5.3519,
      "mean_token_accuracy": 0.21590555608272552,
      "num_tokens": 80768098.0,
      "step": 41290
    },
    {
      "entropy": 5.912838745117187,
      "epoch": 4.127855250662268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003398831755670825,
      "loss": 5.3392,
      "mean_token_accuracy": 0.20468257814645768,
      "num_tokens": 80776716.0,
      "step": 41295
    },
    {
      "entropy": 6.00807523727417,
      "epoch": 4.128355075723497,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003398490017525283,
      "loss": 5.4034,
      "mean_token_accuracy": 0.20045696943998337,
      "num_tokens": 80787000.0,
      "step": 41300
    },
    {
      "entropy": 5.972836875915528,
      "epoch": 4.1288549007847255,
      "grad_norm": 1.125,
      "learning_rate": 0.00033981482630638976,
      "loss": 5.3762,
      "mean_token_accuracy": 0.20475365668535234,
      "num_tokens": 80796872.0,
      "step": 41305
    },
    {
      "entropy": 5.994618988037109,
      "epoch": 4.129354725845954,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033978064922952694,
      "loss": 5.4056,
      "mean_token_accuracy": 0.1999476283788681,
      "num_tokens": 80807057.0,
      "step": 41310
    },
    {
      "entropy": 5.852697706222534,
      "epoch": 4.129854550907183,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033974647052279963,
      "loss": 5.2621,
      "mean_token_accuracy": 0.20958958566188812,
      "num_tokens": 80816710.0,
      "step": 41315
    },
    {
      "entropy": 5.981629753112793,
      "epoch": 4.130354375968411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003397122901870678,
      "loss": 5.4578,
      "mean_token_accuracy": 0.20156090706586838,
      "num_tokens": 80826381.0,
      "step": 41320
    },
    {
      "entropy": 5.8717005252838135,
      "epoch": 4.130854201029639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003396781082231914,
      "loss": 5.2482,
      "mean_token_accuracy": 0.21574874073266984,
      "num_tokens": 80835131.0,
      "step": 41325
    },
    {
      "entropy": 5.915409278869629,
      "epoch": 4.131354026090868,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033964392463203043,
      "loss": 5.3459,
      "mean_token_accuracy": 0.21467838138341905,
      "num_tokens": 80844179.0,
      "step": 41330
    },
    {
      "entropy": 5.925680923461914,
      "epoch": 4.131853851152097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033960973941444503,
      "loss": 5.2705,
      "mean_token_accuracy": 0.20808534771203996,
      "num_tokens": 80853666.0,
      "step": 41335
    },
    {
      "entropy": 5.949933671951294,
      "epoch": 4.1323536762133255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033957555257129514,
      "loss": 5.278,
      "mean_token_accuracy": 0.2149443119764328,
      "num_tokens": 80862931.0,
      "step": 41340
    },
    {
      "entropy": 5.96659107208252,
      "epoch": 4.132853501274554,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003395413641034409,
      "loss": 5.3868,
      "mean_token_accuracy": 0.19947737604379653,
      "num_tokens": 80871742.0,
      "step": 41345
    },
    {
      "entropy": 5.9132849216461185,
      "epoch": 4.133353326335783,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003395071740117426,
      "loss": 5.3423,
      "mean_token_accuracy": 0.2129166379570961,
      "num_tokens": 80880421.0,
      "step": 41350
    },
    {
      "entropy": 5.889363050460815,
      "epoch": 4.133853151397011,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033947298229706043,
      "loss": 5.2198,
      "mean_token_accuracy": 0.219336661696434,
      "num_tokens": 80889663.0,
      "step": 41355
    },
    {
      "entropy": 6.009418058395386,
      "epoch": 4.134352976458239,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003394387889602546,
      "loss": 5.3727,
      "mean_token_accuracy": 0.2001928523182869,
      "num_tokens": 80898406.0,
      "step": 41360
    },
    {
      "entropy": 5.931454515457153,
      "epoch": 4.134852801519468,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003394045940021854,
      "loss": 5.3264,
      "mean_token_accuracy": 0.20609643012285234,
      "num_tokens": 80908634.0,
      "step": 41365
    },
    {
      "entropy": 5.945607471466064,
      "epoch": 4.135352626580697,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033937039742371327,
      "loss": 5.4197,
      "mean_token_accuracy": 0.1969640463590622,
      "num_tokens": 80919014.0,
      "step": 41370
    },
    {
      "entropy": 5.924429273605346,
      "epoch": 4.1358524516419255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033933619922569847,
      "loss": 5.3634,
      "mean_token_accuracy": 0.20518173277378082,
      "num_tokens": 80929714.0,
      "step": 41375
    },
    {
      "entropy": 5.930080461502075,
      "epoch": 4.136352276703154,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003393019994090014,
      "loss": 5.2941,
      "mean_token_accuracy": 0.2125010147690773,
      "num_tokens": 80939590.0,
      "step": 41380
    },
    {
      "entropy": 5.900098562240601,
      "epoch": 4.136852101764383,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033926779797448253,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20930035710334777,
      "num_tokens": 80949569.0,
      "step": 41385
    },
    {
      "entropy": 5.951712274551392,
      "epoch": 4.137351926825611,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003392335949230025,
      "loss": 5.3819,
      "mean_token_accuracy": 0.2078543335199356,
      "num_tokens": 80960288.0,
      "step": 41390
    },
    {
      "entropy": 5.927412748336792,
      "epoch": 4.137851751886839,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033919939025542166,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20465531349182128,
      "num_tokens": 80969680.0,
      "step": 41395
    },
    {
      "entropy": 5.93670129776001,
      "epoch": 4.138351576948068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003391651839726008,
      "loss": 5.3812,
      "mean_token_accuracy": 0.20473868995904923,
      "num_tokens": 80978935.0,
      "step": 41400
    },
    {
      "entropy": 5.953960800170899,
      "epoch": 4.138851402009297,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003391309760754004,
      "loss": 5.4086,
      "mean_token_accuracy": 0.19836397767066954,
      "num_tokens": 80988206.0,
      "step": 41405
    },
    {
      "entropy": 5.9455101013183596,
      "epoch": 4.1393512270705255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003390967665646811,
      "loss": 5.3974,
      "mean_token_accuracy": 0.20792005360126495,
      "num_tokens": 80997550.0,
      "step": 41410
    },
    {
      "entropy": 5.8640735149383545,
      "epoch": 4.139851052131754,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003390625554413036,
      "loss": 5.297,
      "mean_token_accuracy": 0.20688647776842117,
      "num_tokens": 81007401.0,
      "step": 41415
    },
    {
      "entropy": 5.997671270370484,
      "epoch": 4.140350877192983,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003390283427061288,
      "loss": 5.4437,
      "mean_token_accuracy": 0.2019524171948433,
      "num_tokens": 81017132.0,
      "step": 41420
    },
    {
      "entropy": 6.00748872756958,
      "epoch": 4.140850702254211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003389941283600173,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20027556717395784,
      "num_tokens": 81026365.0,
      "step": 41425
    },
    {
      "entropy": 5.959452342987061,
      "epoch": 4.141350527315439,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033895991240383,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2031283438205719,
      "num_tokens": 81036888.0,
      "step": 41430
    },
    {
      "entropy": 5.929624080657959,
      "epoch": 4.141850352376668,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003389256948384278,
      "loss": 5.369,
      "mean_token_accuracy": 0.21333786994218826,
      "num_tokens": 81047461.0,
      "step": 41435
    },
    {
      "entropy": 5.9603509426116945,
      "epoch": 4.142350177437897,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003388914756646715,
      "loss": 5.3666,
      "mean_token_accuracy": 0.20130023062229158,
      "num_tokens": 81057631.0,
      "step": 41440
    },
    {
      "entropy": 6.039345598220825,
      "epoch": 4.1428500024991255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003388572548834222,
      "loss": 5.4113,
      "mean_token_accuracy": 0.19922517538070678,
      "num_tokens": 81068047.0,
      "step": 41445
    },
    {
      "entropy": 5.95827693939209,
      "epoch": 4.143349827560354,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003388230324955407,
      "loss": 5.3368,
      "mean_token_accuracy": 0.20633876770734788,
      "num_tokens": 81077382.0,
      "step": 41450
    },
    {
      "entropy": 5.925182628631592,
      "epoch": 4.143849652621583,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00033878880850188813,
      "loss": 5.2912,
      "mean_token_accuracy": 0.21155433505773544,
      "num_tokens": 81086985.0,
      "step": 41455
    },
    {
      "entropy": 5.973010301589966,
      "epoch": 4.144349477682811,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003387545829033257,
      "loss": 5.3605,
      "mean_token_accuracy": 0.2034759134054184,
      "num_tokens": 81096796.0,
      "step": 41460
    },
    {
      "entropy": 5.9328040599823,
      "epoch": 4.144849302744039,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00033872035570071425,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21260802447795868,
      "num_tokens": 81105611.0,
      "step": 41465
    },
    {
      "entropy": 5.9352454662323,
      "epoch": 4.145349127805268,
      "grad_norm": 1.125,
      "learning_rate": 0.00033868612689491503,
      "loss": 5.3313,
      "mean_token_accuracy": 0.20503551810979842,
      "num_tokens": 81115079.0,
      "step": 41470
    },
    {
      "entropy": 5.888482141494751,
      "epoch": 4.145848952866497,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033865189648678935,
      "loss": 5.3384,
      "mean_token_accuracy": 0.20768121182918547,
      "num_tokens": 81124180.0,
      "step": 41475
    },
    {
      "entropy": 5.9611810684204105,
      "epoch": 4.1463487779277255,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033861766447719824,
      "loss": 5.4666,
      "mean_token_accuracy": 0.20175649970769882,
      "num_tokens": 81132827.0,
      "step": 41480
    },
    {
      "entropy": 6.018667507171631,
      "epoch": 4.146848602988954,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033858343086700304,
      "loss": 5.4033,
      "mean_token_accuracy": 0.19985881447792053,
      "num_tokens": 81142704.0,
      "step": 41485
    },
    {
      "entropy": 5.9421154975891115,
      "epoch": 4.147348428050183,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033854919565706515,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2053966522216797,
      "num_tokens": 81151936.0,
      "step": 41490
    },
    {
      "entropy": 5.958634853363037,
      "epoch": 4.147848253111411,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033851495884824586,
      "loss": 5.39,
      "mean_token_accuracy": 0.20069396495819092,
      "num_tokens": 81161880.0,
      "step": 41495
    },
    {
      "entropy": 5.955117464065552,
      "epoch": 4.148348078172639,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003384807204414065,
      "loss": 5.3218,
      "mean_token_accuracy": 0.2069438561797142,
      "num_tokens": 81171481.0,
      "step": 41500
    },
    {
      "entropy": 5.923992538452149,
      "epoch": 4.148847903233868,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003384464804374086,
      "loss": 5.3462,
      "mean_token_accuracy": 0.20141062885522842,
      "num_tokens": 81181479.0,
      "step": 41505
    },
    {
      "entropy": 5.917004823684692,
      "epoch": 4.149347728295097,
      "grad_norm": 1.25,
      "learning_rate": 0.0003384122388371135,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20935108959674836,
      "num_tokens": 81190143.0,
      "step": 41510
    },
    {
      "entropy": 5.911036396026612,
      "epoch": 4.1498475533563255,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003383779956413829,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20835882127285005,
      "num_tokens": 81199475.0,
      "step": 41515
    },
    {
      "entropy": 5.9418230056762695,
      "epoch": 4.150347378417554,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003383437508510781,
      "loss": 5.3423,
      "mean_token_accuracy": 0.21199232041835786,
      "num_tokens": 81208344.0,
      "step": 41520
    },
    {
      "entropy": 5.919611024856567,
      "epoch": 4.150847203478782,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0003383095044670609,
      "loss": 5.2297,
      "mean_token_accuracy": 0.22391003221273423,
      "num_tokens": 81217939.0,
      "step": 41525
    },
    {
      "entropy": 5.842412948608398,
      "epoch": 4.151347028540011,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033827525649019293,
      "loss": 5.2329,
      "mean_token_accuracy": 0.216540364921093,
      "num_tokens": 81227672.0,
      "step": 41530
    },
    {
      "entropy": 5.903822565078736,
      "epoch": 4.151846853601239,
      "grad_norm": 1.328125,
      "learning_rate": 0.00033824100692133566,
      "loss": 5.2984,
      "mean_token_accuracy": 0.21691416054964066,
      "num_tokens": 81237061.0,
      "step": 41535
    },
    {
      "entropy": 5.97203311920166,
      "epoch": 4.152346678662468,
      "grad_norm": 1.1875,
      "learning_rate": 0.000338206755761351,
      "loss": 5.3171,
      "mean_token_accuracy": 0.212379290163517,
      "num_tokens": 81247546.0,
      "step": 41540
    },
    {
      "entropy": 5.909684658050537,
      "epoch": 4.152846503723697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003381725030111006,
      "loss": 5.3631,
      "mean_token_accuracy": 0.19914334416389465,
      "num_tokens": 81257874.0,
      "step": 41545
    },
    {
      "entropy": 5.769840240478516,
      "epoch": 4.1533463287849255,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003381382486714463,
      "loss": 5.1944,
      "mean_token_accuracy": 0.2281726524233818,
      "num_tokens": 81267788.0,
      "step": 41550
    },
    {
      "entropy": 5.983113527297974,
      "epoch": 4.153846153846154,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00033810399274324994,
      "loss": 5.3879,
      "mean_token_accuracy": 0.20052670538425446,
      "num_tokens": 81277623.0,
      "step": 41555
    },
    {
      "entropy": 5.960531520843506,
      "epoch": 4.154345978907382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033806973522737326,
      "loss": 5.3461,
      "mean_token_accuracy": 0.2069755896925926,
      "num_tokens": 81287888.0,
      "step": 41560
    },
    {
      "entropy": 5.847367668151856,
      "epoch": 4.154845803968611,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003380354761246784,
      "loss": 5.247,
      "mean_token_accuracy": 0.21949817091226578,
      "num_tokens": 81296595.0,
      "step": 41565
    },
    {
      "entropy": 5.835123682022095,
      "epoch": 4.155345629029839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003380012154360271,
      "loss": 5.3445,
      "mean_token_accuracy": 0.20705133080482482,
      "num_tokens": 81306910.0,
      "step": 41570
    },
    {
      "entropy": 5.904671716690063,
      "epoch": 4.155845454091068,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003379669531622814,
      "loss": 5.3491,
      "mean_token_accuracy": 0.2060815304517746,
      "num_tokens": 81315971.0,
      "step": 41575
    },
    {
      "entropy": 5.983008861541748,
      "epoch": 4.156345279152297,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003379326893043034,
      "loss": 5.4116,
      "mean_token_accuracy": 0.1968699425458908,
      "num_tokens": 81326229.0,
      "step": 41580
    },
    {
      "entropy": 5.94983401298523,
      "epoch": 4.1568451042135255,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003378984238629552,
      "loss": 5.3489,
      "mean_token_accuracy": 0.21293849796056746,
      "num_tokens": 81336256.0,
      "step": 41585
    },
    {
      "entropy": 5.910355806350708,
      "epoch": 4.157344929274754,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003378641568390988,
      "loss": 5.3032,
      "mean_token_accuracy": 0.21223272234201432,
      "num_tokens": 81345535.0,
      "step": 41590
    },
    {
      "entropy": 5.932229137420654,
      "epoch": 4.157844754335982,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003378298882335963,
      "loss": 5.3306,
      "mean_token_accuracy": 0.21307721585035325,
      "num_tokens": 81354418.0,
      "step": 41595
    },
    {
      "entropy": 6.022017240524292,
      "epoch": 4.158344579397211,
      "grad_norm": 1.0,
      "learning_rate": 0.00033779561804731005,
      "loss": 5.4903,
      "mean_token_accuracy": 0.19312811344861985,
      "num_tokens": 81365390.0,
      "step": 41600
    },
    {
      "entropy": 5.933361291885376,
      "epoch": 4.158844404458439,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003377613462811022,
      "loss": 5.3282,
      "mean_token_accuracy": 0.20754481106996536,
      "num_tokens": 81376662.0,
      "step": 41605
    },
    {
      "entropy": 6.013990211486816,
      "epoch": 4.159344229519668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003377270729358351,
      "loss": 5.3939,
      "mean_token_accuracy": 0.19592822045087815,
      "num_tokens": 81387071.0,
      "step": 41610
    },
    {
      "entropy": 5.972828197479248,
      "epoch": 4.159844054580897,
      "grad_norm": 1.125,
      "learning_rate": 0.0003376927980123709,
      "loss": 5.4565,
      "mean_token_accuracy": 0.20129198879003524,
      "num_tokens": 81397202.0,
      "step": 41615
    },
    {
      "entropy": 5.889663553237915,
      "epoch": 4.1603438796421255,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003376585215115721,
      "loss": 5.2784,
      "mean_token_accuracy": 0.21264119148254396,
      "num_tokens": 81406757.0,
      "step": 41620
    },
    {
      "entropy": 5.966645860671997,
      "epoch": 4.160843704703354,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033762424343430106,
      "loss": 5.4014,
      "mean_token_accuracy": 0.19970934838056564,
      "num_tokens": 81416493.0,
      "step": 41625
    },
    {
      "entropy": 5.941965627670288,
      "epoch": 4.161343529764582,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033758996378142,
      "loss": 5.3957,
      "mean_token_accuracy": 0.20862885564565659,
      "num_tokens": 81427108.0,
      "step": 41630
    },
    {
      "entropy": 5.931457138061523,
      "epoch": 4.161843354825811,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033755568255379166,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20675024539232253,
      "num_tokens": 81436112.0,
      "step": 41635
    },
    {
      "entropy": 6.002251434326172,
      "epoch": 4.162343179887039,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003375213997522784,
      "loss": 5.3479,
      "mean_token_accuracy": 0.20846948623657227,
      "num_tokens": 81447198.0,
      "step": 41640
    },
    {
      "entropy": 5.9215494155883786,
      "epoch": 4.162843004948268,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033748711537774287,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20846324115991594,
      "num_tokens": 81457203.0,
      "step": 41645
    },
    {
      "entropy": 6.021171140670776,
      "epoch": 4.163342830009497,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003374528294310476,
      "loss": 5.427,
      "mean_token_accuracy": 0.1984502926468849,
      "num_tokens": 81467264.0,
      "step": 41650
    },
    {
      "entropy": 5.97119927406311,
      "epoch": 4.1638426550707255,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033741854191305515,
      "loss": 5.3443,
      "mean_token_accuracy": 0.2071132093667984,
      "num_tokens": 81476590.0,
      "step": 41655
    },
    {
      "entropy": 5.932653570175171,
      "epoch": 4.164342480131954,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003373842528246283,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19930412024259567,
      "num_tokens": 81487224.0,
      "step": 41660
    },
    {
      "entropy": 5.943230819702149,
      "epoch": 4.164842305193182,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003373499621666296,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21324344128370284,
      "num_tokens": 81496498.0,
      "step": 41665
    },
    {
      "entropy": 5.923976039886474,
      "epoch": 4.165342130254411,
      "grad_norm": 1.234375,
      "learning_rate": 0.00033731566993992204,
      "loss": 5.2878,
      "mean_token_accuracy": 0.2101055532693863,
      "num_tokens": 81505323.0,
      "step": 41670
    },
    {
      "entropy": 5.977946424484253,
      "epoch": 4.165841955315639,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033728137614536815,
      "loss": 5.3605,
      "mean_token_accuracy": 0.213010510802269,
      "num_tokens": 81515609.0,
      "step": 41675
    },
    {
      "entropy": 5.961260032653809,
      "epoch": 4.166341780376868,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003372470807838309,
      "loss": 5.4144,
      "mean_token_accuracy": 0.21040434390306473,
      "num_tokens": 81525199.0,
      "step": 41680
    },
    {
      "entropy": 5.92318377494812,
      "epoch": 4.166841605438097,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033721278385617315,
      "loss": 5.3045,
      "mean_token_accuracy": 0.20959978997707368,
      "num_tokens": 81536283.0,
      "step": 41685
    },
    {
      "entropy": 5.98917965888977,
      "epoch": 4.1673414304993255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003371784853632577,
      "loss": 5.414,
      "mean_token_accuracy": 0.19904717057943344,
      "num_tokens": 81545711.0,
      "step": 41690
    },
    {
      "entropy": 5.876005601882935,
      "epoch": 4.167841255560554,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033714418530594767,
      "loss": 5.2649,
      "mean_token_accuracy": 0.20645889788866043,
      "num_tokens": 81556032.0,
      "step": 41695
    },
    {
      "entropy": 5.982289743423462,
      "epoch": 4.168341080621782,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003371098836851058,
      "loss": 5.4505,
      "mean_token_accuracy": 0.19970735013484955,
      "num_tokens": 81567939.0,
      "step": 41700
    },
    {
      "entropy": 5.9201240062713625,
      "epoch": 4.168840905683011,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003370755805015953,
      "loss": 5.2819,
      "mean_token_accuracy": 0.21406870037317277,
      "num_tokens": 81576900.0,
      "step": 41705
    },
    {
      "entropy": 6.013292551040649,
      "epoch": 4.169340730744239,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003370412757562792,
      "loss": 5.383,
      "mean_token_accuracy": 0.20420292615890503,
      "num_tokens": 81586776.0,
      "step": 41710
    },
    {
      "entropy": 5.859568119049072,
      "epoch": 4.169840555805468,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003370069694500205,
      "loss": 5.2533,
      "mean_token_accuracy": 0.21557969599962234,
      "num_tokens": 81598043.0,
      "step": 41715
    },
    {
      "entropy": 5.9156506061553955,
      "epoch": 4.170340380866697,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003369726615836825,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21066380590200423,
      "num_tokens": 81607522.0,
      "step": 41720
    },
    {
      "entropy": 5.984977579116821,
      "epoch": 4.1708402059279255,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003369383521581282,
      "loss": 5.4672,
      "mean_token_accuracy": 0.1999448612332344,
      "num_tokens": 81616668.0,
      "step": 41725
    },
    {
      "entropy": 5.917151880264282,
      "epoch": 4.171340030989154,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033690404117422084,
      "loss": 5.3601,
      "mean_token_accuracy": 0.20972791016101838,
      "num_tokens": 81626967.0,
      "step": 41730
    },
    {
      "entropy": 6.006461954116821,
      "epoch": 4.171839856050382,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033686972863282384,
      "loss": 5.4039,
      "mean_token_accuracy": 0.2064061641693115,
      "num_tokens": 81637399.0,
      "step": 41735
    },
    {
      "entropy": 6.00159010887146,
      "epoch": 4.172339681111611,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003368354145348003,
      "loss": 5.3838,
      "mean_token_accuracy": 0.204112708568573,
      "num_tokens": 81646959.0,
      "step": 41740
    },
    {
      "entropy": 5.910500574111938,
      "epoch": 4.172839506172839,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033680109888101375,
      "loss": 5.347,
      "mean_token_accuracy": 0.20697734653949737,
      "num_tokens": 81656110.0,
      "step": 41745
    },
    {
      "entropy": 5.8965918064117435,
      "epoch": 4.173339331234068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003367667816723274,
      "loss": 5.3139,
      "mean_token_accuracy": 0.21019722670316696,
      "num_tokens": 81666077.0,
      "step": 41750
    },
    {
      "entropy": 5.980490350723267,
      "epoch": 4.173839156295297,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003367324629096047,
      "loss": 5.4139,
      "mean_token_accuracy": 0.2050003081560135,
      "num_tokens": 81675000.0,
      "step": 41755
    },
    {
      "entropy": 5.939600467681885,
      "epoch": 4.1743389813565255,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003366981425937091,
      "loss": 5.3126,
      "mean_token_accuracy": 0.21088829785585403,
      "num_tokens": 81685347.0,
      "step": 41760
    },
    {
      "entropy": 5.9698412895202635,
      "epoch": 4.174838806417754,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033666382072550417,
      "loss": 5.4113,
      "mean_token_accuracy": 0.20331325232982636,
      "num_tokens": 81694953.0,
      "step": 41765
    },
    {
      "entropy": 5.966524219512939,
      "epoch": 4.175338631478982,
      "grad_norm": 1.125,
      "learning_rate": 0.00033662949730585336,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20165566653013228,
      "num_tokens": 81704883.0,
      "step": 41770
    },
    {
      "entropy": 5.96060848236084,
      "epoch": 4.175838456540211,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003365951723356202,
      "loss": 5.3081,
      "mean_token_accuracy": 0.2134371131658554,
      "num_tokens": 81713404.0,
      "step": 41775
    },
    {
      "entropy": 5.92298641204834,
      "epoch": 4.176338281601439,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033656084581566844,
      "loss": 5.3493,
      "mean_token_accuracy": 0.20861700922250748,
      "num_tokens": 81722656.0,
      "step": 41780
    },
    {
      "entropy": 5.876442289352417,
      "epoch": 4.176838106662668,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033652651774686163,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20847167819738388,
      "num_tokens": 81732862.0,
      "step": 41785
    },
    {
      "entropy": 5.988072538375855,
      "epoch": 4.177337931723897,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033649218813006337,
      "loss": 5.4303,
      "mean_token_accuracy": 0.20563313961029053,
      "num_tokens": 81742730.0,
      "step": 41790
    },
    {
      "entropy": 5.938894605636596,
      "epoch": 4.1778377567851255,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003364578569661376,
      "loss": 5.3683,
      "mean_token_accuracy": 0.20300418883562088,
      "num_tokens": 81752177.0,
      "step": 41795
    },
    {
      "entropy": 5.943050813674927,
      "epoch": 4.178337581846354,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033642352425594787,
      "loss": 5.3432,
      "mean_token_accuracy": 0.20022958368062974,
      "num_tokens": 81761377.0,
      "step": 41800
    },
    {
      "entropy": 5.9414294242858885,
      "epoch": 4.178837406907582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033638919000035815,
      "loss": 5.3529,
      "mean_token_accuracy": 0.20779592394828797,
      "num_tokens": 81771417.0,
      "step": 41805
    },
    {
      "entropy": 5.553664064407348,
      "epoch": 4.179337231968811,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033635485420023217,
      "loss": 4.9575,
      "mean_token_accuracy": 0.2527132362127304,
      "num_tokens": 81780626.0,
      "step": 41810
    },
    {
      "entropy": 5.961479902267456,
      "epoch": 4.179837057030039,
      "grad_norm": 1.125,
      "learning_rate": 0.0003363205168564338,
      "loss": 5.3965,
      "mean_token_accuracy": 0.20161438286304473,
      "num_tokens": 81790954.0,
      "step": 41815
    },
    {
      "entropy": 5.962175130844116,
      "epoch": 4.180336882091268,
      "grad_norm": 1.03125,
      "learning_rate": 0.00033628617796982707,
      "loss": 5.4188,
      "mean_token_accuracy": 0.20395028442144394,
      "num_tokens": 81801343.0,
      "step": 41820
    },
    {
      "entropy": 5.980080318450928,
      "epoch": 4.180836707152497,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033625183754127584,
      "loss": 5.4048,
      "mean_token_accuracy": 0.2028943732380867,
      "num_tokens": 81810615.0,
      "step": 41825
    },
    {
      "entropy": 6.004215145111084,
      "epoch": 4.1813365322137255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003362174955716442,
      "loss": 5.4597,
      "mean_token_accuracy": 0.20217142701148988,
      "num_tokens": 81821104.0,
      "step": 41830
    },
    {
      "entropy": 5.936572170257568,
      "epoch": 4.181836357274954,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033618315206179614,
      "loss": 5.3461,
      "mean_token_accuracy": 0.2021329566836357,
      "num_tokens": 81830841.0,
      "step": 41835
    },
    {
      "entropy": 5.914356327056884,
      "epoch": 4.182336182336182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003361488070125956,
      "loss": 5.3389,
      "mean_token_accuracy": 0.20778726637363434,
      "num_tokens": 81840356.0,
      "step": 41840
    },
    {
      "entropy": 5.956907081604004,
      "epoch": 4.182836007397411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003361144604249069,
      "loss": 5.3687,
      "mean_token_accuracy": 0.2083652585744858,
      "num_tokens": 81850768.0,
      "step": 41845
    },
    {
      "entropy": 5.886463308334351,
      "epoch": 4.183335832458639,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033608011229959404,
      "loss": 5.3484,
      "mean_token_accuracy": 0.20340894311666488,
      "num_tokens": 81862285.0,
      "step": 41850
    },
    {
      "entropy": 5.988341331481934,
      "epoch": 4.183835657519868,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003360457626375213,
      "loss": 5.3781,
      "mean_token_accuracy": 0.20606345385313035,
      "num_tokens": 81871662.0,
      "step": 41855
    },
    {
      "entropy": 5.932784557342529,
      "epoch": 4.184335482581097,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003360114114395529,
      "loss": 5.3725,
      "mean_token_accuracy": 0.21067556440830232,
      "num_tokens": 81881170.0,
      "step": 41860
    },
    {
      "entropy": 5.885516214370727,
      "epoch": 4.1848353076423255,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000335977058706553,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20912066996097564,
      "num_tokens": 81891782.0,
      "step": 41865
    },
    {
      "entropy": 5.995335054397583,
      "epoch": 4.185335132703554,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033594270443938604,
      "loss": 5.409,
      "mean_token_accuracy": 0.2024126961827278,
      "num_tokens": 81900945.0,
      "step": 41870
    },
    {
      "entropy": 5.992324733734131,
      "epoch": 4.185834957764782,
      "grad_norm": 1.125,
      "learning_rate": 0.00033590834863891645,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20574240535497665,
      "num_tokens": 81910114.0,
      "step": 41875
    },
    {
      "entropy": 5.924401807785034,
      "epoch": 4.186334782826011,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003358739913060083,
      "loss": 5.3493,
      "mean_token_accuracy": 0.21250832974910736,
      "num_tokens": 81919605.0,
      "step": 41880
    },
    {
      "entropy": 5.971412706375122,
      "epoch": 4.186834607887239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003358396324415263,
      "loss": 5.4557,
      "mean_token_accuracy": 0.1967099726200104,
      "num_tokens": 81929585.0,
      "step": 41885
    },
    {
      "entropy": 5.995824956893921,
      "epoch": 4.187334432948468,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033580527204633473,
      "loss": 5.4139,
      "mean_token_accuracy": 0.19936602413654328,
      "num_tokens": 81938666.0,
      "step": 41890
    },
    {
      "entropy": 5.939746713638305,
      "epoch": 4.187834258009697,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003357709101212983,
      "loss": 5.3975,
      "mean_token_accuracy": 0.2016312912106514,
      "num_tokens": 81949517.0,
      "step": 41895
    },
    {
      "entropy": 5.859354829788208,
      "epoch": 4.1883340830709255,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033573654666728127,
      "loss": 5.3313,
      "mean_token_accuracy": 0.21134763360023498,
      "num_tokens": 81960997.0,
      "step": 41900
    },
    {
      "entropy": 5.892227697372436,
      "epoch": 4.188833908132154,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003357021816851484,
      "loss": 5.334,
      "mean_token_accuracy": 0.20827700942754745,
      "num_tokens": 81971613.0,
      "step": 41905
    },
    {
      "entropy": 5.898682308197022,
      "epoch": 4.189333733193382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003356678151757643,
      "loss": 5.272,
      "mean_token_accuracy": 0.20846756249666215,
      "num_tokens": 81980030.0,
      "step": 41910
    },
    {
      "entropy": 5.935645914077758,
      "epoch": 4.189833558254611,
      "grad_norm": 1.09375,
      "learning_rate": 0.00033563344713999344,
      "loss": 5.365,
      "mean_token_accuracy": 0.21205391436815263,
      "num_tokens": 81989610.0,
      "step": 41915
    },
    {
      "entropy": 5.866631364822387,
      "epoch": 4.190333383315839,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033559907757870074,
      "loss": 5.3328,
      "mean_token_accuracy": 0.2120044395327568,
      "num_tokens": 81999808.0,
      "step": 41920
    },
    {
      "entropy": 5.924660968780517,
      "epoch": 4.190833208377068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033556470649275086,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2159358501434326,
      "num_tokens": 82009520.0,
      "step": 41925
    },
    {
      "entropy": 5.816134452819824,
      "epoch": 4.191333033438297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003355303338830086,
      "loss": 5.2116,
      "mean_token_accuracy": 0.2178143799304962,
      "num_tokens": 82019920.0,
      "step": 41930
    },
    {
      "entropy": 5.996751594543457,
      "epoch": 4.1918328584995255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003354959597503386,
      "loss": 5.3921,
      "mean_token_accuracy": 0.19951252341270448,
      "num_tokens": 82030098.0,
      "step": 41935
    },
    {
      "entropy": 5.970304489135742,
      "epoch": 4.192332683560753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033546158409560587,
      "loss": 5.4003,
      "mean_token_accuracy": 0.2045649141073227,
      "num_tokens": 82039830.0,
      "step": 41940
    },
    {
      "entropy": 5.971395063400268,
      "epoch": 4.192832508621982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003354272069196752,
      "loss": 5.376,
      "mean_token_accuracy": 0.2071874499320984,
      "num_tokens": 82049428.0,
      "step": 41945
    },
    {
      "entropy": 5.865433216094971,
      "epoch": 4.193332333683211,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003353928282234116,
      "loss": 5.3889,
      "mean_token_accuracy": 0.2135101616382599,
      "num_tokens": 82059970.0,
      "step": 41950
    },
    {
      "entropy": 5.850697612762451,
      "epoch": 4.193832158744439,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033535844800768,
      "loss": 5.1709,
      "mean_token_accuracy": 0.23064377903938293,
      "num_tokens": 82069944.0,
      "step": 41955
    },
    {
      "entropy": 5.990732860565186,
      "epoch": 4.194331983805668,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033532406627334534,
      "loss": 5.3866,
      "mean_token_accuracy": 0.20261933654546738,
      "num_tokens": 82079061.0,
      "step": 41960
    },
    {
      "entropy": 5.9116113662719725,
      "epoch": 4.194831808866897,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003352896830212728,
      "loss": 5.3395,
      "mean_token_accuracy": 0.2048830956220627,
      "num_tokens": 82088304.0,
      "step": 41965
    },
    {
      "entropy": 5.895105695724487,
      "epoch": 4.1953316339281255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003352552982523272,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21360128819942475,
      "num_tokens": 82098721.0,
      "step": 41970
    },
    {
      "entropy": 5.886775255203247,
      "epoch": 4.195831458989353,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003352209119673739,
      "loss": 5.4162,
      "mean_token_accuracy": 0.19515071362257003,
      "num_tokens": 82108845.0,
      "step": 41975
    },
    {
      "entropy": 5.935017538070679,
      "epoch": 4.196331284050582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033518652416727785,
      "loss": 5.282,
      "mean_token_accuracy": 0.2073858141899109,
      "num_tokens": 82118327.0,
      "step": 41980
    },
    {
      "entropy": 5.909512615203857,
      "epoch": 4.196831109111811,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003351521348529044,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21166408360004424,
      "num_tokens": 82127601.0,
      "step": 41985
    },
    {
      "entropy": 5.843794345855713,
      "epoch": 4.197330934173039,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033511774402511876,
      "loss": 5.2392,
      "mean_token_accuracy": 0.22020660787820817,
      "num_tokens": 82138311.0,
      "step": 41990
    },
    {
      "entropy": 5.8369855880737305,
      "epoch": 4.197830759234268,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003350833516847861,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21582858711481095,
      "num_tokens": 82148430.0,
      "step": 41995
    },
    {
      "entropy": 5.884143590927124,
      "epoch": 4.198330584295497,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033504895783277176,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20346311628818511,
      "num_tokens": 82158032.0,
      "step": 42000
    },
    {
      "epoch": 4.198330584295497,
      "eval_entropy": 5.687623953211095,
      "eval_loss": 5.51631498336792,
      "eval_mean_token_accuracy": 0.20674249747347137,
      "eval_num_tokens": 82158032.0,
      "eval_runtime": 24.079,
      "eval_samples_per_second": 1289.298,
      "eval_steps_per_second": 161.178,
      "step": 42000
    },
    {
      "entropy": 5.9468036651611325,
      "epoch": 4.1988304093567255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003350145624699412,
      "loss": 5.3728,
      "mean_token_accuracy": 0.20606600046157836,
      "num_tokens": 82167689.0,
      "step": 42005
    },
    {
      "entropy": 5.885760164260864,
      "epoch": 4.199330234417953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033498016559715954,
      "loss": 5.2638,
      "mean_token_accuracy": 0.21171732991933823,
      "num_tokens": 82178053.0,
      "step": 42010
    },
    {
      "entropy": 5.886480188369751,
      "epoch": 4.199830059479182,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033494576721529234,
      "loss": 5.3051,
      "mean_token_accuracy": 0.20795165300369262,
      "num_tokens": 82187182.0,
      "step": 42015
    },
    {
      "entropy": 5.95084319114685,
      "epoch": 4.200329884540411,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00033491136732520516,
      "loss": 5.4155,
      "mean_token_accuracy": 0.19530435651540756,
      "num_tokens": 82198532.0,
      "step": 42020
    },
    {
      "entropy": 5.925795269012451,
      "epoch": 4.200829709601639,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003348769659277634,
      "loss": 5.2644,
      "mean_token_accuracy": 0.22195200324058534,
      "num_tokens": 82208913.0,
      "step": 42025
    },
    {
      "entropy": 5.917975759506225,
      "epoch": 4.201329534662868,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033484256302383254,
      "loss": 5.2563,
      "mean_token_accuracy": 0.2124785527586937,
      "num_tokens": 82218746.0,
      "step": 42030
    },
    {
      "entropy": 5.951236820220947,
      "epoch": 4.201829359724097,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033480815861427826,
      "loss": 5.3663,
      "mean_token_accuracy": 0.21027446836233138,
      "num_tokens": 82228117.0,
      "step": 42035
    },
    {
      "entropy": 5.977363061904907,
      "epoch": 4.2023291847853255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000334773752699966,
      "loss": 5.4965,
      "mean_token_accuracy": 0.1901382178068161,
      "num_tokens": 82237656.0,
      "step": 42040
    },
    {
      "entropy": 5.846442079544067,
      "epoch": 4.202829009846553,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003347393452817616,
      "loss": 5.267,
      "mean_token_accuracy": 0.2079084634780884,
      "num_tokens": 82247209.0,
      "step": 42045
    },
    {
      "entropy": 5.956150150299072,
      "epoch": 4.203328834907782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003347049363605306,
      "loss": 5.4267,
      "mean_token_accuracy": 0.19980484247207642,
      "num_tokens": 82256473.0,
      "step": 42050
    },
    {
      "entropy": 5.996184635162353,
      "epoch": 4.203828659969011,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003346705259371387,
      "loss": 5.4511,
      "mean_token_accuracy": 0.2003634661436081,
      "num_tokens": 82267158.0,
      "step": 42055
    },
    {
      "entropy": 5.919577598571777,
      "epoch": 4.204328485030239,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003346361140124518,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2137128308415413,
      "num_tokens": 82276556.0,
      "step": 42060
    },
    {
      "entropy": 5.89529447555542,
      "epoch": 4.204828310091468,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003346017005873356,
      "loss": 5.3052,
      "mean_token_accuracy": 0.21001219600439072,
      "num_tokens": 82286079.0,
      "step": 42065
    },
    {
      "entropy": 5.883049249649048,
      "epoch": 4.205328135152697,
      "grad_norm": 1.109375,
      "learning_rate": 0.000334567285662656,
      "loss": 5.3805,
      "mean_token_accuracy": 0.21009393632411957,
      "num_tokens": 82296374.0,
      "step": 42070
    },
    {
      "entropy": 6.001918840408325,
      "epoch": 4.2058279602139255,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003345328692392787,
      "loss": 5.426,
      "mean_token_accuracy": 0.20049092620611192,
      "num_tokens": 82306406.0,
      "step": 42075
    },
    {
      "entropy": 6.031495809555054,
      "epoch": 4.206327785275153,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033449845131806985,
      "loss": 5.3733,
      "mean_token_accuracy": 0.20570916086435317,
      "num_tokens": 82315376.0,
      "step": 42080
    },
    {
      "entropy": 5.868555641174316,
      "epoch": 4.206827610336382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003344640318998952,
      "loss": 5.2732,
      "mean_token_accuracy": 0.22035528868436813,
      "num_tokens": 82325137.0,
      "step": 42085
    },
    {
      "entropy": 5.931623840332032,
      "epoch": 4.207327435397611,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033442961098562084,
      "loss": 5.3258,
      "mean_token_accuracy": 0.2076758086681366,
      "num_tokens": 82334525.0,
      "step": 42090
    },
    {
      "entropy": 5.885899019241333,
      "epoch": 4.207827260458839,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003343951885761127,
      "loss": 5.2775,
      "mean_token_accuracy": 0.20889659523963927,
      "num_tokens": 82343549.0,
      "step": 42095
    },
    {
      "entropy": 5.902189159393311,
      "epoch": 4.208327085520068,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033436076467223695,
      "loss": 5.3256,
      "mean_token_accuracy": 0.20927158296108245,
      "num_tokens": 82352589.0,
      "step": 42100
    },
    {
      "entropy": 5.9558991432189945,
      "epoch": 4.208826910581297,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003343263392748597,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20475890040397643,
      "num_tokens": 82363225.0,
      "step": 42105
    },
    {
      "entropy": 5.979271125793457,
      "epoch": 4.2093267356425255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003342919123848469,
      "loss": 5.4249,
      "mean_token_accuracy": 0.20073592513799668,
      "num_tokens": 82373495.0,
      "step": 42110
    },
    {
      "entropy": 5.93928599357605,
      "epoch": 4.209826560703753,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003342574840030649,
      "loss": 5.3466,
      "mean_token_accuracy": 0.2065521076321602,
      "num_tokens": 82384539.0,
      "step": 42115
    },
    {
      "entropy": 5.964757633209229,
      "epoch": 4.210326385764982,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003342230541303799,
      "loss": 5.3781,
      "mean_token_accuracy": 0.2103997826576233,
      "num_tokens": 82394524.0,
      "step": 42120
    },
    {
      "entropy": 5.924071168899536,
      "epoch": 4.210826210826211,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000334188622767658,
      "loss": 5.3311,
      "mean_token_accuracy": 0.2051895335316658,
      "num_tokens": 82403975.0,
      "step": 42125
    },
    {
      "entropy": 5.974514102935791,
      "epoch": 4.211326035887439,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003341541899157657,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19833489060401915,
      "num_tokens": 82413813.0,
      "step": 42130
    },
    {
      "entropy": 5.881864023208618,
      "epoch": 4.211825860948668,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003341197555755691,
      "loss": 5.2657,
      "mean_token_accuracy": 0.2054738163948059,
      "num_tokens": 82423530.0,
      "step": 42135
    },
    {
      "entropy": 5.945368242263794,
      "epoch": 4.212325686009897,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033408531974793464,
      "loss": 5.3943,
      "mean_token_accuracy": 0.2018829643726349,
      "num_tokens": 82433313.0,
      "step": 42140
    },
    {
      "entropy": 5.9232430934906,
      "epoch": 4.2128255110711255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003340508824337288,
      "loss": 5.3545,
      "mean_token_accuracy": 0.20769564658403397,
      "num_tokens": 82443441.0,
      "step": 42145
    },
    {
      "entropy": 5.959226799011231,
      "epoch": 4.213325336132353,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033401644363381796,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21774295717477798,
      "num_tokens": 82453274.0,
      "step": 42150
    },
    {
      "entropy": 6.026158809661865,
      "epoch": 4.213825161193582,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003339820033490686,
      "loss": 5.5922,
      "mean_token_accuracy": 0.1917801693081856,
      "num_tokens": 82463028.0,
      "step": 42155
    },
    {
      "entropy": 5.926073455810547,
      "epoch": 4.214324986254811,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033394756158034724,
      "loss": 5.3244,
      "mean_token_accuracy": 0.21464277803897858,
      "num_tokens": 82473145.0,
      "step": 42160
    },
    {
      "entropy": 5.855519580841064,
      "epoch": 4.214824811316039,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003339131183285203,
      "loss": 5.3394,
      "mean_token_accuracy": 0.2125682294368744,
      "num_tokens": 82483588.0,
      "step": 42165
    },
    {
      "entropy": 5.941143798828125,
      "epoch": 4.215324636377268,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033387867359445443,
      "loss": 5.333,
      "mean_token_accuracy": 0.21128010600805283,
      "num_tokens": 82492807.0,
      "step": 42170
    },
    {
      "entropy": 5.930727529525757,
      "epoch": 4.215824461438497,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003338442273790164,
      "loss": 5.2596,
      "mean_token_accuracy": 0.2174778088927269,
      "num_tokens": 82502165.0,
      "step": 42175
    },
    {
      "entropy": 5.902155256271362,
      "epoch": 4.2163242864997255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033380977968307273,
      "loss": 5.28,
      "mean_token_accuracy": 0.2118616059422493,
      "num_tokens": 82512408.0,
      "step": 42180
    },
    {
      "entropy": 5.89588680267334,
      "epoch": 4.216824111560953,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033377533050749016,
      "loss": 5.3436,
      "mean_token_accuracy": 0.21247533559799195,
      "num_tokens": 82521729.0,
      "step": 42185
    },
    {
      "entropy": 6.021088457107544,
      "epoch": 4.217323936622182,
      "grad_norm": 1.125,
      "learning_rate": 0.0003337408798531353,
      "loss": 5.4591,
      "mean_token_accuracy": 0.190402814745903,
      "num_tokens": 82531167.0,
      "step": 42190
    },
    {
      "entropy": 5.900983142852783,
      "epoch": 4.217823761683411,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033370642772087507,
      "loss": 5.2899,
      "mean_token_accuracy": 0.21145959347486495,
      "num_tokens": 82541140.0,
      "step": 42195
    },
    {
      "entropy": 5.980422687530518,
      "epoch": 4.218323586744639,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003336719741115762,
      "loss": 5.507,
      "mean_token_accuracy": 0.1973233103752136,
      "num_tokens": 82551347.0,
      "step": 42200
    },
    {
      "entropy": 6.00376148223877,
      "epoch": 4.218823411805868,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003336375190261056,
      "loss": 5.4417,
      "mean_token_accuracy": 0.2072818398475647,
      "num_tokens": 82561140.0,
      "step": 42205
    },
    {
      "entropy": 5.947339391708374,
      "epoch": 4.219323236867097,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033360306246533013,
      "loss": 5.3344,
      "mean_token_accuracy": 0.2009625792503357,
      "num_tokens": 82570496.0,
      "step": 42210
    },
    {
      "entropy": 5.905517148971557,
      "epoch": 4.2198230619283255,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00033356860443011665,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21234001219272614,
      "num_tokens": 82579974.0,
      "step": 42215
    },
    {
      "entropy": 5.8627598762512205,
      "epoch": 4.220322886989553,
      "grad_norm": 1.25,
      "learning_rate": 0.00033353414492133217,
      "loss": 5.1961,
      "mean_token_accuracy": 0.22465941309928894,
      "num_tokens": 82590090.0,
      "step": 42220
    },
    {
      "entropy": 5.9077721118927,
      "epoch": 4.220822712050782,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003334996839398436,
      "loss": 5.3388,
      "mean_token_accuracy": 0.20372971892356873,
      "num_tokens": 82599793.0,
      "step": 42225
    },
    {
      "entropy": 5.9304726123809814,
      "epoch": 4.221322537112011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033346522148651817,
      "loss": 5.452,
      "mean_token_accuracy": 0.19847740679979325,
      "num_tokens": 82610284.0,
      "step": 42230
    },
    {
      "entropy": 5.95780577659607,
      "epoch": 4.221822362173239,
      "grad_norm": 1.546875,
      "learning_rate": 0.00033343075756222265,
      "loss": 5.3091,
      "mean_token_accuracy": 0.2144986554980278,
      "num_tokens": 82621119.0,
      "step": 42235
    },
    {
      "entropy": 5.929177188873291,
      "epoch": 4.222322187234468,
      "grad_norm": 1.125,
      "learning_rate": 0.00033339629216782435,
      "loss": 5.2801,
      "mean_token_accuracy": 0.20474023669958114,
      "num_tokens": 82631254.0,
      "step": 42240
    },
    {
      "entropy": 5.953227567672729,
      "epoch": 4.222822012295697,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033336182530419045,
      "loss": 5.3561,
      "mean_token_accuracy": 0.20340363532304764,
      "num_tokens": 82640766.0,
      "step": 42245
    },
    {
      "entropy": 5.8742368698120115,
      "epoch": 4.2233218373569255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033332735697218804,
      "loss": 5.3023,
      "mean_token_accuracy": 0.20995782762765886,
      "num_tokens": 82651076.0,
      "step": 42250
    },
    {
      "entropy": 5.899253082275391,
      "epoch": 4.223821662418153,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003332928871726842,
      "loss": 5.2842,
      "mean_token_accuracy": 0.2191697582602501,
      "num_tokens": 82660331.0,
      "step": 42255
    },
    {
      "entropy": 5.967183208465576,
      "epoch": 4.224321487479382,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033325841590654647,
      "loss": 5.4321,
      "mean_token_accuracy": 0.20237600207328796,
      "num_tokens": 82671099.0,
      "step": 42260
    },
    {
      "entropy": 5.9188055992126465,
      "epoch": 4.224821312540611,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003332239431746419,
      "loss": 5.3163,
      "mean_token_accuracy": 0.20651976764202118,
      "num_tokens": 82680103.0,
      "step": 42265
    },
    {
      "entropy": 5.997226095199585,
      "epoch": 4.225321137601839,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003331894689778379,
      "loss": 5.3908,
      "mean_token_accuracy": 0.20068781524896623,
      "num_tokens": 82689206.0,
      "step": 42270
    },
    {
      "entropy": 5.904348134994507,
      "epoch": 4.225820962663068,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003331549933170019,
      "loss": 5.316,
      "mean_token_accuracy": 0.21062091886997222,
      "num_tokens": 82698555.0,
      "step": 42275
    },
    {
      "entropy": 5.875715923309326,
      "epoch": 4.226320787724297,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003331205161930012,
      "loss": 5.2947,
      "mean_token_accuracy": 0.21684521436691284,
      "num_tokens": 82708446.0,
      "step": 42280
    },
    {
      "entropy": 5.8901725769042965,
      "epoch": 4.2268206127855255,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003330860376067033,
      "loss": 5.3161,
      "mean_token_accuracy": 0.21111924350261688,
      "num_tokens": 82718431.0,
      "step": 42285
    },
    {
      "entropy": 5.892410087585449,
      "epoch": 4.227320437846753,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003330515575589756,
      "loss": 5.3419,
      "mean_token_accuracy": 0.19961674511432648,
      "num_tokens": 82728400.0,
      "step": 42290
    },
    {
      "entropy": 5.995226240158081,
      "epoch": 4.227820262907982,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003330170760506857,
      "loss": 5.4399,
      "mean_token_accuracy": 0.19447892755270005,
      "num_tokens": 82737637.0,
      "step": 42295
    },
    {
      "entropy": 6.00431342124939,
      "epoch": 4.228320087969211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033298259308270096,
      "loss": 5.4007,
      "mean_token_accuracy": 0.20573314279317856,
      "num_tokens": 82748375.0,
      "step": 42300
    },
    {
      "entropy": 5.874521398544312,
      "epoch": 4.228819913030439,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033294810865588915,
      "loss": 5.3035,
      "mean_token_accuracy": 0.20888233333826065,
      "num_tokens": 82758364.0,
      "step": 42305
    },
    {
      "entropy": 5.920189952850341,
      "epoch": 4.229319738091668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033291362277111793,
      "loss": 5.3377,
      "mean_token_accuracy": 0.20955329537391662,
      "num_tokens": 82767937.0,
      "step": 42310
    },
    {
      "entropy": 5.899498748779297,
      "epoch": 4.229819563152897,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033287913542925484,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21246770322322844,
      "num_tokens": 82778110.0,
      "step": 42315
    },
    {
      "entropy": 5.921962547302246,
      "epoch": 4.2303193882141255,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003328446466311676,
      "loss": 5.3892,
      "mean_token_accuracy": 0.19925851225852967,
      "num_tokens": 82788181.0,
      "step": 42320
    },
    {
      "entropy": 5.93304967880249,
      "epoch": 4.230819213275353,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003328101563777238,
      "loss": 5.3111,
      "mean_token_accuracy": 0.2159376099705696,
      "num_tokens": 82796997.0,
      "step": 42325
    },
    {
      "entropy": 5.9235083103179935,
      "epoch": 4.231319038336582,
      "grad_norm": 1.125,
      "learning_rate": 0.0003327756646697915,
      "loss": 5.31,
      "mean_token_accuracy": 0.21401509046554565,
      "num_tokens": 82806308.0,
      "step": 42330
    },
    {
      "entropy": 5.888381481170654,
      "epoch": 4.231818863397811,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003327411715082382,
      "loss": 5.2362,
      "mean_token_accuracy": 0.21754882633686065,
      "num_tokens": 82816229.0,
      "step": 42335
    },
    {
      "entropy": 5.890602159500122,
      "epoch": 4.232318688459039,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000332706676893932,
      "loss": 5.328,
      "mean_token_accuracy": 0.2146204426884651,
      "num_tokens": 82826180.0,
      "step": 42340
    },
    {
      "entropy": 5.896803188323974,
      "epoch": 4.232818513520268,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033267218082774064,
      "loss": 5.3231,
      "mean_token_accuracy": 0.21393367052078247,
      "num_tokens": 82836148.0,
      "step": 42345
    },
    {
      "entropy": 5.902700996398925,
      "epoch": 4.233318338581497,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003326376833105321,
      "loss": 5.383,
      "mean_token_accuracy": 0.2047794684767723,
      "num_tokens": 82846289.0,
      "step": 42350
    },
    {
      "entropy": 5.945898056030273,
      "epoch": 4.2338181636427255,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033260318434317414,
      "loss": 5.3226,
      "mean_token_accuracy": 0.2135414406657219,
      "num_tokens": 82856283.0,
      "step": 42355
    },
    {
      "entropy": 5.904102420806884,
      "epoch": 4.234317988703953,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00033256868392653497,
      "loss": 5.3673,
      "mean_token_accuracy": 0.2091175839304924,
      "num_tokens": 82867539.0,
      "step": 42360
    },
    {
      "entropy": 6.013313055038452,
      "epoch": 4.234817813765182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003325341820614825,
      "loss": 5.423,
      "mean_token_accuracy": 0.20675648450851442,
      "num_tokens": 82876926.0,
      "step": 42365
    },
    {
      "entropy": 5.913746547698975,
      "epoch": 4.235317638826411,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003324996787488849,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20491236001253127,
      "num_tokens": 82887369.0,
      "step": 42370
    },
    {
      "entropy": 5.856533861160278,
      "epoch": 4.235817463887639,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003324651739896101,
      "loss": 5.2828,
      "mean_token_accuracy": 0.21180928349494935,
      "num_tokens": 82896553.0,
      "step": 42375
    },
    {
      "entropy": 5.8761543273925785,
      "epoch": 4.236317288948868,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003324306677845263,
      "loss": 5.2609,
      "mean_token_accuracy": 0.22014174610376358,
      "num_tokens": 82906687.0,
      "step": 42380
    },
    {
      "entropy": 5.94670877456665,
      "epoch": 4.236817114010097,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003323961601345017,
      "loss": 5.3537,
      "mean_token_accuracy": 0.20942272245883942,
      "num_tokens": 82916616.0,
      "step": 42385
    },
    {
      "entropy": 5.998528861999512,
      "epoch": 4.2373169390713255,
      "grad_norm": 1.609375,
      "learning_rate": 0.0003323616510404045,
      "loss": 5.417,
      "mean_token_accuracy": 0.20432834029197694,
      "num_tokens": 82925966.0,
      "step": 42390
    },
    {
      "entropy": 5.78526725769043,
      "epoch": 4.237816764132553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00033232714050310285,
      "loss": 5.1566,
      "mean_token_accuracy": 0.2252880573272705,
      "num_tokens": 82935504.0,
      "step": 42395
    },
    {
      "entropy": 5.896826601028442,
      "epoch": 4.238316589193782,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033229262852346514,
      "loss": 5.3193,
      "mean_token_accuracy": 0.21912603527307511,
      "num_tokens": 82945489.0,
      "step": 42400
    },
    {
      "entropy": 5.918535614013672,
      "epoch": 4.238816414255011,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00033225811510235965,
      "loss": 5.338,
      "mean_token_accuracy": 0.21313151568174363,
      "num_tokens": 82955138.0,
      "step": 42405
    },
    {
      "entropy": 5.927353477478027,
      "epoch": 4.239316239316239,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033222360024065474,
      "loss": 5.4095,
      "mean_token_accuracy": 0.20272316187620162,
      "num_tokens": 82965334.0,
      "step": 42410
    },
    {
      "entropy": 5.975210762023925,
      "epoch": 4.239816064377468,
      "grad_norm": 1.171875,
      "learning_rate": 0.00033218908393921874,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19550225585699083,
      "num_tokens": 82976114.0,
      "step": 42415
    },
    {
      "entropy": 5.918724203109742,
      "epoch": 4.240315889438697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003321545661989201,
      "loss": 5.3084,
      "mean_token_accuracy": 0.21048786044120787,
      "num_tokens": 82985533.0,
      "step": 42420
    },
    {
      "entropy": 5.982336139678955,
      "epoch": 4.2408157144999254,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033212004702062734,
      "loss": 5.3813,
      "mean_token_accuracy": 0.20835826694965362,
      "num_tokens": 82995576.0,
      "step": 42425
    },
    {
      "entropy": 5.9330390930175785,
      "epoch": 4.241315539561153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003320855264052088,
      "loss": 5.2991,
      "mean_token_accuracy": 0.2092570185661316,
      "num_tokens": 83005602.0,
      "step": 42430
    },
    {
      "entropy": 5.995650863647461,
      "epoch": 4.241815364622382,
      "grad_norm": 1.28125,
      "learning_rate": 0.00033205100435353314,
      "loss": 5.3936,
      "mean_token_accuracy": 0.2026470869779587,
      "num_tokens": 83013801.0,
      "step": 42435
    },
    {
      "entropy": 5.981161308288574,
      "epoch": 4.242315189683611,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003320164808664689,
      "loss": 5.4113,
      "mean_token_accuracy": 0.20334526896476746,
      "num_tokens": 83023198.0,
      "step": 42440
    },
    {
      "entropy": 5.920407915115357,
      "epoch": 4.242815014744839,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003319819559448847,
      "loss": 5.2946,
      "mean_token_accuracy": 0.22019240409135818,
      "num_tokens": 83032071.0,
      "step": 42445
    },
    {
      "entropy": 5.9718070983886715,
      "epoch": 4.243314839806068,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033194742958964914,
      "loss": 5.3754,
      "mean_token_accuracy": 0.19721204340457915,
      "num_tokens": 83041235.0,
      "step": 42450
    },
    {
      "entropy": 6.039362907409668,
      "epoch": 4.243814664867297,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00033191290180163084,
      "loss": 5.4265,
      "mean_token_accuracy": 0.19569335281848907,
      "num_tokens": 83050871.0,
      "step": 42455
    },
    {
      "entropy": 5.961792469024658,
      "epoch": 4.2443144899285254,
      "grad_norm": 1.265625,
      "learning_rate": 0.00033187837258169864,
      "loss": 5.3381,
      "mean_token_accuracy": 0.21476491987705232,
      "num_tokens": 83058549.0,
      "step": 42460
    },
    {
      "entropy": 5.840431165695191,
      "epoch": 4.244814314989753,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0003318438419307212,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21060795783996583,
      "num_tokens": 83069236.0,
      "step": 42465
    },
    {
      "entropy": 5.871422815322876,
      "epoch": 4.245314140050982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003318093098495673,
      "loss": 5.3235,
      "mean_token_accuracy": 0.20536428093910217,
      "num_tokens": 83078245.0,
      "step": 42470
    },
    {
      "entropy": 5.904888725280761,
      "epoch": 4.245813965112211,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033177477633910586,
      "loss": 5.2791,
      "mean_token_accuracy": 0.22295595556497574,
      "num_tokens": 83088703.0,
      "step": 42475
    },
    {
      "entropy": 5.948534870147705,
      "epoch": 4.246313790173439,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00033174024140020556,
      "loss": 5.3269,
      "mean_token_accuracy": 0.21090080142021178,
      "num_tokens": 83097999.0,
      "step": 42480
    },
    {
      "entropy": 6.0336321830749515,
      "epoch": 4.246813615234668,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00033170570503373545,
      "loss": 5.4209,
      "mean_token_accuracy": 0.20563495606184007,
      "num_tokens": 83107107.0,
      "step": 42485
    },
    {
      "entropy": 5.980328607559204,
      "epoch": 4.247313440295897,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003316711672405644,
      "loss": 5.3715,
      "mean_token_accuracy": 0.20370817333459854,
      "num_tokens": 83116342.0,
      "step": 42490
    },
    {
      "entropy": 5.895579242706299,
      "epoch": 4.2478132653571254,
      "grad_norm": 1.21875,
      "learning_rate": 0.00033163662802156136,
      "loss": 5.2848,
      "mean_token_accuracy": 0.2145911142230034,
      "num_tokens": 83125073.0,
      "step": 42495
    },
    {
      "entropy": 5.97892918586731,
      "epoch": 4.248313090418353,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033160208737759535,
      "loss": 5.4035,
      "mean_token_accuracy": 0.20347760617733002,
      "num_tokens": 83135307.0,
      "step": 42500
    },
    {
      "entropy": 5.885803508758545,
      "epoch": 4.248812915479582,
      "grad_norm": 1.25,
      "learning_rate": 0.0003315675453095353,
      "loss": 5.2568,
      "mean_token_accuracy": 0.2154754415154457,
      "num_tokens": 83145727.0,
      "step": 42505
    },
    {
      "entropy": 5.8793586730957035,
      "epoch": 4.249312740540811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003315330018182504,
      "loss": 5.243,
      "mean_token_accuracy": 0.21637309491634368,
      "num_tokens": 83155786.0,
      "step": 42510
    },
    {
      "entropy": 5.948736047744751,
      "epoch": 4.249812565602039,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033149845690460974,
      "loss": 5.3419,
      "mean_token_accuracy": 0.2028066858649254,
      "num_tokens": 83165045.0,
      "step": 42515
    },
    {
      "entropy": 5.893549489974975,
      "epoch": 4.250312390663268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00033146391056948244,
      "loss": 5.3579,
      "mean_token_accuracy": 0.2131126642227173,
      "num_tokens": 83173784.0,
      "step": 42520
    },
    {
      "entropy": 5.941919660568237,
      "epoch": 4.250812215724497,
      "grad_norm": 1.375,
      "learning_rate": 0.00033142936281373777,
      "loss": 5.3169,
      "mean_token_accuracy": 0.2091740295290947,
      "num_tokens": 83182768.0,
      "step": 42525
    },
    {
      "entropy": 6.000212717056274,
      "epoch": 4.251312040785725,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00033139481363824477,
      "loss": 5.4488,
      "mean_token_accuracy": 0.19848431497812272,
      "num_tokens": 83191871.0,
      "step": 42530
    },
    {
      "entropy": 5.983804655075073,
      "epoch": 4.251811865846953,
      "grad_norm": 1.1875,
      "learning_rate": 0.00033136026304387276,
      "loss": 5.3839,
      "mean_token_accuracy": 0.20407738387584687,
      "num_tokens": 83200359.0,
      "step": 42535
    },
    {
      "entropy": 5.903144836425781,
      "epoch": 4.252311690908182,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000331325711031491,
      "loss": 5.3115,
      "mean_token_accuracy": 0.20840407609939576,
      "num_tokens": 83209008.0,
      "step": 42540
    },
    {
      "entropy": 5.909700918197632,
      "epoch": 4.252811515969411,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033129115760196885,
      "loss": 5.3873,
      "mean_token_accuracy": 0.20418146699666978,
      "num_tokens": 83218567.0,
      "step": 42545
    },
    {
      "entropy": 5.77313084602356,
      "epoch": 4.253311341030639,
      "grad_norm": 1.25,
      "learning_rate": 0.00033125660275617563,
      "loss": 5.1336,
      "mean_token_accuracy": 0.23466939479112625,
      "num_tokens": 83228455.0,
      "step": 42550
    },
    {
      "entropy": 5.952725219726562,
      "epoch": 4.253811166091868,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003312220464949809,
      "loss": 5.2925,
      "mean_token_accuracy": 0.20113265216350557,
      "num_tokens": 83237679.0,
      "step": 42555
    },
    {
      "entropy": 5.92109866142273,
      "epoch": 4.254310991153097,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003311874888192539,
      "loss": 5.3698,
      "mean_token_accuracy": 0.2075188785791397,
      "num_tokens": 83246685.0,
      "step": 42560
    },
    {
      "entropy": 5.929313611984253,
      "epoch": 4.254810816214325,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003311529297298641,
      "loss": 5.3459,
      "mean_token_accuracy": 0.20903182923793792,
      "num_tokens": 83256154.0,
      "step": 42565
    },
    {
      "entropy": 5.91396050453186,
      "epoch": 4.255310641275553,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003311183692276811,
      "loss": 5.2673,
      "mean_token_accuracy": 0.2195390582084656,
      "num_tokens": 83266533.0,
      "step": 42570
    },
    {
      "entropy": 5.931393957138061,
      "epoch": 4.255810466336782,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033108380731357426,
      "loss": 5.3018,
      "mean_token_accuracy": 0.21646783351898194,
      "num_tokens": 83277397.0,
      "step": 42575
    },
    {
      "entropy": 6.002592658996582,
      "epoch": 4.256310291398011,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003310492439884132,
      "loss": 5.3792,
      "mean_token_accuracy": 0.20436572879552842,
      "num_tokens": 83287931.0,
      "step": 42580
    },
    {
      "entropy": 5.964669847488404,
      "epoch": 4.256810116459239,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033101467925306777,
      "loss": 5.3957,
      "mean_token_accuracy": 0.19693127125501633,
      "num_tokens": 83297157.0,
      "step": 42585
    },
    {
      "entropy": 5.95057487487793,
      "epoch": 4.257309941520468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003309801131084074,
      "loss": 5.3801,
      "mean_token_accuracy": 0.2040078952908516,
      "num_tokens": 83306560.0,
      "step": 42590
    },
    {
      "entropy": 5.94780421257019,
      "epoch": 4.257809766581697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033094554555530175,
      "loss": 5.3338,
      "mean_token_accuracy": 0.2025921255350113,
      "num_tokens": 83316187.0,
      "step": 42595
    },
    {
      "entropy": 5.9673370838165285,
      "epoch": 4.258309591642925,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003309109765946206,
      "loss": 5.3947,
      "mean_token_accuracy": 0.20534368306398393,
      "num_tokens": 83326812.0,
      "step": 42600
    },
    {
      "entropy": 5.96913709640503,
      "epoch": 4.258809416704153,
      "grad_norm": 1.140625,
      "learning_rate": 0.00033087640622723357,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20992923080921172,
      "num_tokens": 83336729.0,
      "step": 42605
    },
    {
      "entropy": 5.862431955337525,
      "epoch": 4.259309241765382,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003308418344540106,
      "loss": 5.2866,
      "mean_token_accuracy": 0.21465162187814713,
      "num_tokens": 83346806.0,
      "step": 42610
    },
    {
      "entropy": 5.895345973968506,
      "epoch": 4.259809066826611,
      "grad_norm": 1.0625,
      "learning_rate": 0.00033080726127582147,
      "loss": 5.4181,
      "mean_token_accuracy": 0.1941641554236412,
      "num_tokens": 83356932.0,
      "step": 42615
    },
    {
      "entropy": 5.960794401168823,
      "epoch": 4.260308891887839,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003307726866935359,
      "loss": 5.3323,
      "mean_token_accuracy": 0.20957248508930207,
      "num_tokens": 83366918.0,
      "step": 42620
    },
    {
      "entropy": 5.94733681678772,
      "epoch": 4.260808716949068,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003307381107080241,
      "loss": 5.2613,
      "mean_token_accuracy": 0.20845486372709274,
      "num_tokens": 83376146.0,
      "step": 42625
    },
    {
      "entropy": 5.87862377166748,
      "epoch": 4.261308542010297,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003307035333201556,
      "loss": 5.3331,
      "mean_token_accuracy": 0.21277815848588943,
      "num_tokens": 83386415.0,
      "step": 42630
    },
    {
      "entropy": 5.870230102539063,
      "epoch": 4.2618083670715246,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003306689545308006,
      "loss": 5.306,
      "mean_token_accuracy": 0.22040654420852662,
      "num_tokens": 83396917.0,
      "step": 42635
    },
    {
      "entropy": 5.935095691680909,
      "epoch": 4.262308192132753,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00033063437434082896,
      "loss": 5.3336,
      "mean_token_accuracy": 0.20610687881708145,
      "num_tokens": 83407366.0,
      "step": 42640
    },
    {
      "entropy": 5.934714460372925,
      "epoch": 4.262808017193982,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003305997927511108,
      "loss": 5.3943,
      "mean_token_accuracy": 0.2090449720621109,
      "num_tokens": 83417173.0,
      "step": 42645
    },
    {
      "entropy": 5.875620031356812,
      "epoch": 4.263307842255211,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00033056520976251617,
      "loss": 5.2829,
      "mean_token_accuracy": 0.21255708932876588,
      "num_tokens": 83426740.0,
      "step": 42650
    },
    {
      "entropy": 5.996032571792602,
      "epoch": 4.263807667316439,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033053062537591516,
      "loss": 5.3977,
      "mean_token_accuracy": 0.19875403493642807,
      "num_tokens": 83437461.0,
      "step": 42655
    },
    {
      "entropy": 5.969452285766602,
      "epoch": 4.264307492377668,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000330496039592178,
      "loss": 5.3563,
      "mean_token_accuracy": 0.20440228879451752,
      "num_tokens": 83446360.0,
      "step": 42660
    },
    {
      "entropy": 5.9111559867858885,
      "epoch": 4.264807317438897,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033046145241217466,
      "loss": 5.3137,
      "mean_token_accuracy": 0.21139181405305862,
      "num_tokens": 83455531.0,
      "step": 42665
    },
    {
      "entropy": 5.988850212097168,
      "epoch": 4.2653071425001245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003304268638367754,
      "loss": 5.4162,
      "mean_token_accuracy": 0.19826823472976685,
      "num_tokens": 83465544.0,
      "step": 42670
    },
    {
      "entropy": 5.941551733016968,
      "epoch": 4.265806967561353,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00033039227386685063,
      "loss": 5.3848,
      "mean_token_accuracy": 0.20271400660276412,
      "num_tokens": 83474855.0,
      "step": 42675
    },
    {
      "entropy": 5.948987102508545,
      "epoch": 4.266306792622582,
      "grad_norm": 1.15625,
      "learning_rate": 0.00033035768250327046,
      "loss": 5.3559,
      "mean_token_accuracy": 0.2103319987654686,
      "num_tokens": 83483843.0,
      "step": 42680
    },
    {
      "entropy": 6.013801002502442,
      "epoch": 4.266806617683811,
      "grad_norm": 1.125,
      "learning_rate": 0.0003303230897469052,
      "loss": 5.4549,
      "mean_token_accuracy": 0.1994909942150116,
      "num_tokens": 83492976.0,
      "step": 42685
    },
    {
      "entropy": 5.891788625717163,
      "epoch": 4.267306442745039,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003302884955986253,
      "loss": 5.3567,
      "mean_token_accuracy": 0.20663716197013854,
      "num_tokens": 83503114.0,
      "step": 42690
    },
    {
      "entropy": 5.983260440826416,
      "epoch": 4.267806267806268,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003302539000593011,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20355470925569535,
      "num_tokens": 83512343.0,
      "step": 42695
    },
    {
      "entropy": 5.980424165725708,
      "epoch": 4.268306092867497,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003302193031298029,
      "loss": 5.3626,
      "mean_token_accuracy": 0.20836845189332961,
      "num_tokens": 83522572.0,
      "step": 42700
    },
    {
      "entropy": 5.863089609146118,
      "epoch": 4.2688059179287245,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033018470481100125,
      "loss": 5.2426,
      "mean_token_accuracy": 0.21362507939338685,
      "num_tokens": 83532026.0,
      "step": 42705
    },
    {
      "entropy": 5.939913177490235,
      "epoch": 4.269305742989953,
      "grad_norm": 1.109375,
      "learning_rate": 0.00033015010510376676,
      "loss": 5.3818,
      "mean_token_accuracy": 0.198563751578331,
      "num_tokens": 83541301.0,
      "step": 42710
    },
    {
      "entropy": 6.000452995300293,
      "epoch": 4.269805568051182,
      "grad_norm": 1.046875,
      "learning_rate": 0.00033011550400896967,
      "loss": 5.4345,
      "mean_token_accuracy": 0.19452550560235976,
      "num_tokens": 83550629.0,
      "step": 42715
    },
    {
      "entropy": 5.981116437911988,
      "epoch": 4.270305393112411,
      "grad_norm": 1.078125,
      "learning_rate": 0.00033008090152748077,
      "loss": 5.3776,
      "mean_token_accuracy": 0.20540752559900283,
      "num_tokens": 83560631.0,
      "step": 42720
    },
    {
      "entropy": 5.935222816467285,
      "epoch": 4.270805218173639,
      "grad_norm": 1.25,
      "learning_rate": 0.00033004629766017046,
      "loss": 5.3576,
      "mean_token_accuracy": 0.20747154653072358,
      "num_tokens": 83569637.0,
      "step": 42725
    },
    {
      "entropy": 5.817612171173096,
      "epoch": 4.271305043234868,
      "grad_norm": 1.203125,
      "learning_rate": 0.00033001169240790955,
      "loss": 5.2048,
      "mean_token_accuracy": 0.23082145005464555,
      "num_tokens": 83579171.0,
      "step": 42730
    },
    {
      "entropy": 5.798088836669922,
      "epoch": 4.271804868296097,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003299770857715686,
      "loss": 5.2422,
      "mean_token_accuracy": 0.2205585613846779,
      "num_tokens": 83587870.0,
      "step": 42735
    },
    {
      "entropy": 5.892365741729736,
      "epoch": 4.2723046933573245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003299424777520183,
      "loss": 5.287,
      "mean_token_accuracy": 0.21120430231094361,
      "num_tokens": 83598028.0,
      "step": 42740
    },
    {
      "entropy": 5.859552097320557,
      "epoch": 4.272804518418553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032990786835012943,
      "loss": 5.3284,
      "mean_token_accuracy": 0.20709843933582306,
      "num_tokens": 83607715.0,
      "step": 42745
    },
    {
      "entropy": 5.872103977203369,
      "epoch": 4.273304343479782,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032987325756677266,
      "loss": 5.3434,
      "mean_token_accuracy": 0.19936052411794664,
      "num_tokens": 83617514.0,
      "step": 42750
    },
    {
      "entropy": 5.92534990310669,
      "epoch": 4.273804168541011,
      "grad_norm": 1.328125,
      "learning_rate": 0.0003298386454028189,
      "loss": 5.264,
      "mean_token_accuracy": 0.21285820454359056,
      "num_tokens": 83627072.0,
      "step": 42755
    },
    {
      "entropy": 5.979276180267334,
      "epoch": 4.274303993602239,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003298040318591389,
      "loss": 5.3766,
      "mean_token_accuracy": 0.2117307499051094,
      "num_tokens": 83637138.0,
      "step": 42760
    },
    {
      "entropy": 5.973269987106323,
      "epoch": 4.274803818663468,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003297694169366036,
      "loss": 5.4201,
      "mean_token_accuracy": 0.20010316520929336,
      "num_tokens": 83646586.0,
      "step": 42765
    },
    {
      "entropy": 5.930415439605713,
      "epoch": 4.275303643724697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003297348006360839,
      "loss": 5.3241,
      "mean_token_accuracy": 0.21008390933275223,
      "num_tokens": 83656080.0,
      "step": 42770
    },
    {
      "entropy": 5.885944271087647,
      "epoch": 4.2758034687859245,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003297001829584506,
      "loss": 5.2883,
      "mean_token_accuracy": 0.20966732650995254,
      "num_tokens": 83665848.0,
      "step": 42775
    },
    {
      "entropy": 5.914840364456177,
      "epoch": 4.276303293847153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032966556390457487,
      "loss": 5.293,
      "mean_token_accuracy": 0.20275438129901885,
      "num_tokens": 83676178.0,
      "step": 42780
    },
    {
      "entropy": 6.008195495605468,
      "epoch": 4.276803118908382,
      "grad_norm": 1.359375,
      "learning_rate": 0.00032963094347532757,
      "loss": 5.4203,
      "mean_token_accuracy": 0.19973266273736953,
      "num_tokens": 83685849.0,
      "step": 42785
    },
    {
      "entropy": 5.873637056350708,
      "epoch": 4.277302943969611,
      "grad_norm": 1.125,
      "learning_rate": 0.00032959632167157976,
      "loss": 5.3372,
      "mean_token_accuracy": 0.20368496030569078,
      "num_tokens": 83696095.0,
      "step": 42790
    },
    {
      "entropy": 5.8976939678192135,
      "epoch": 4.277802769030839,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003295616984942026,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21848683506250383,
      "num_tokens": 83706507.0,
      "step": 42795
    },
    {
      "entropy": 5.972263956069947,
      "epoch": 4.278302594092068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003295270739440671,
      "loss": 5.3765,
      "mean_token_accuracy": 0.20191690325737,
      "num_tokens": 83716354.0,
      "step": 42800
    },
    {
      "entropy": 5.927693462371826,
      "epoch": 4.278802419153297,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032949244802204445,
      "loss": 5.3726,
      "mean_token_accuracy": 0.20945853888988494,
      "num_tokens": 83726384.0,
      "step": 42805
    },
    {
      "entropy": 5.899393796920776,
      "epoch": 4.2793022442145245,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032945782072900593,
      "loss": 5.3553,
      "mean_token_accuracy": 0.21508315503597258,
      "num_tokens": 83735455.0,
      "step": 42810
    },
    {
      "entropy": 5.9012082576751705,
      "epoch": 4.279802069275753,
      "grad_norm": 1.25,
      "learning_rate": 0.0003294231920658225,
      "loss": 5.3284,
      "mean_token_accuracy": 0.2071952998638153,
      "num_tokens": 83744197.0,
      "step": 42815
    },
    {
      "entropy": 6.024853992462158,
      "epoch": 4.280301894336982,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003293885620333657,
      "loss": 5.4256,
      "mean_token_accuracy": 0.20085898339748381,
      "num_tokens": 83753709.0,
      "step": 42820
    },
    {
      "entropy": 6.009854555130005,
      "epoch": 4.280801719398211,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032935393063250655,
      "loss": 5.3712,
      "mean_token_accuracy": 0.21094940751791,
      "num_tokens": 83763274.0,
      "step": 42825
    },
    {
      "entropy": 5.885459470748901,
      "epoch": 4.281301544459439,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003293192978641166,
      "loss": 5.286,
      "mean_token_accuracy": 0.2138707920908928,
      "num_tokens": 83771521.0,
      "step": 42830
    },
    {
      "entropy": 5.877171277999878,
      "epoch": 4.281801369520668,
      "grad_norm": 1.1875,
      "learning_rate": 0.000329284663729067,
      "loss": 5.3539,
      "mean_token_accuracy": 0.2012176677584648,
      "num_tokens": 83781683.0,
      "step": 42835
    },
    {
      "entropy": 5.848325634002686,
      "epoch": 4.282301194581897,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003292500282282294,
      "loss": 5.2218,
      "mean_token_accuracy": 0.21677405387163162,
      "num_tokens": 83790964.0,
      "step": 42840
    },
    {
      "entropy": 5.92686505317688,
      "epoch": 4.2828010196431245,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032921539136247487,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20757606476545334,
      "num_tokens": 83799108.0,
      "step": 42845
    },
    {
      "entropy": 5.95250391960144,
      "epoch": 4.283300844704353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032918075313267503,
      "loss": 5.3424,
      "mean_token_accuracy": 0.21674895882606507,
      "num_tokens": 83808354.0,
      "step": 42850
    },
    {
      "entropy": 5.946483612060547,
      "epoch": 4.283800669765582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032914611353970145,
      "loss": 5.3736,
      "mean_token_accuracy": 0.2062017634510994,
      "num_tokens": 83818857.0,
      "step": 42855
    },
    {
      "entropy": 5.926468801498413,
      "epoch": 4.284300494826811,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003291114725844256,
      "loss": 5.2474,
      "mean_token_accuracy": 0.2254989981651306,
      "num_tokens": 83828071.0,
      "step": 42860
    },
    {
      "entropy": 5.920537424087525,
      "epoch": 4.284800319888039,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000329076830267719,
      "loss": 5.3782,
      "mean_token_accuracy": 0.2136052131652832,
      "num_tokens": 83837827.0,
      "step": 42865
    },
    {
      "entropy": 5.917628288269043,
      "epoch": 4.285300144949268,
      "grad_norm": 1.34375,
      "learning_rate": 0.0003290421865904533,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21198486685752868,
      "num_tokens": 83847723.0,
      "step": 42870
    },
    {
      "entropy": 5.903476858139038,
      "epoch": 4.285799970010497,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003290075415535001,
      "loss": 5.3135,
      "mean_token_accuracy": 0.206315016746521,
      "num_tokens": 83856997.0,
      "step": 42875
    },
    {
      "entropy": 5.871958065032959,
      "epoch": 4.2862997950717245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000328972895157731,
      "loss": 5.2342,
      "mean_token_accuracy": 0.21506549417972565,
      "num_tokens": 83867137.0,
      "step": 42880
    },
    {
      "entropy": 5.8667370796203615,
      "epoch": 4.286799620132953,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032893824740401776,
      "loss": 5.2708,
      "mean_token_accuracy": 0.21107998192310334,
      "num_tokens": 83876444.0,
      "step": 42885
    },
    {
      "entropy": 5.937594366073609,
      "epoch": 4.287299445194182,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032890359829323214,
      "loss": 5.3952,
      "mean_token_accuracy": 0.20413703918457032,
      "num_tokens": 83886335.0,
      "step": 42890
    },
    {
      "entropy": 5.962606906890869,
      "epoch": 4.287799270255411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003288689478262458,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20495793521404265,
      "num_tokens": 83896377.0,
      "step": 42895
    },
    {
      "entropy": 5.961015510559082,
      "epoch": 4.288299095316639,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032883429600393053,
      "loss": 5.37,
      "mean_token_accuracy": 0.20260681360960006,
      "num_tokens": 83906067.0,
      "step": 42900
    },
    {
      "entropy": 5.694618606567383,
      "epoch": 4.288798920377868,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032879964282715843,
      "loss": 5.0732,
      "mean_token_accuracy": 0.24132467657327653,
      "num_tokens": 83916642.0,
      "step": 42905
    },
    {
      "entropy": 5.760056400299073,
      "epoch": 4.289298745439097,
      "grad_norm": 1.0625,
      "learning_rate": 0.000328764988296801,
      "loss": 5.1991,
      "mean_token_accuracy": 0.2153922736644745,
      "num_tokens": 83927164.0,
      "step": 42910
    },
    {
      "entropy": 5.9791162490844725,
      "epoch": 4.2897985705003245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032873033241373034,
      "loss": 5.3446,
      "mean_token_accuracy": 0.2046997681260109,
      "num_tokens": 83937206.0,
      "step": 42915
    },
    {
      "entropy": 5.94559907913208,
      "epoch": 4.290298395561553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003286956751788184,
      "loss": 5.3629,
      "mean_token_accuracy": 0.2069172292947769,
      "num_tokens": 83947017.0,
      "step": 42920
    },
    {
      "entropy": 5.949058246612549,
      "epoch": 4.290798220622782,
      "grad_norm": 1.125,
      "learning_rate": 0.00032866101659293693,
      "loss": 5.4048,
      "mean_token_accuracy": 0.20964595526456833,
      "num_tokens": 83956430.0,
      "step": 42925
    },
    {
      "entropy": 5.914544773101807,
      "epoch": 4.291298045684011,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032862635665695825,
      "loss": 5.3608,
      "mean_token_accuracy": 0.2051038548350334,
      "num_tokens": 83965229.0,
      "step": 42930
    },
    {
      "entropy": 5.947645378112793,
      "epoch": 4.291797870745239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032859169537175414,
      "loss": 5.3121,
      "mean_token_accuracy": 0.21227284669876098,
      "num_tokens": 83974330.0,
      "step": 42935
    },
    {
      "entropy": 5.961039400100708,
      "epoch": 4.292297695806468,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032855703273819686,
      "loss": 5.3502,
      "mean_token_accuracy": 0.20242003202438355,
      "num_tokens": 83984623.0,
      "step": 42940
    },
    {
      "entropy": 5.8971062183380125,
      "epoch": 4.292797520867696,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032852236875715836,
      "loss": 5.2169,
      "mean_token_accuracy": 0.21771706640720367,
      "num_tokens": 83994977.0,
      "step": 42945
    },
    {
      "entropy": 5.801345109939575,
      "epoch": 4.2932973459289245,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032848770342951085,
      "loss": 5.2423,
      "mean_token_accuracy": 0.22457996606826783,
      "num_tokens": 84006022.0,
      "step": 42950
    },
    {
      "entropy": 5.871743297576904,
      "epoch": 4.293797170990153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003284530367561264,
      "loss": 5.3681,
      "mean_token_accuracy": 0.20323301553726197,
      "num_tokens": 84015569.0,
      "step": 42955
    },
    {
      "entropy": 5.9142426490783695,
      "epoch": 4.294296996051382,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003284183687378774,
      "loss": 5.2733,
      "mean_token_accuracy": 0.20930234342813492,
      "num_tokens": 84024798.0,
      "step": 42960
    },
    {
      "entropy": 5.932222318649292,
      "epoch": 4.294796821112611,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00032838369937563587,
      "loss": 5.2892,
      "mean_token_accuracy": 0.20809615701436995,
      "num_tokens": 84034117.0,
      "step": 42965
    },
    {
      "entropy": 5.858179664611816,
      "epoch": 4.295296646173839,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032834902867027435,
      "loss": 5.3085,
      "mean_token_accuracy": 0.2084166333079338,
      "num_tokens": 84043690.0,
      "step": 42970
    },
    {
      "entropy": 5.933700799942017,
      "epoch": 4.295796471235068,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032831435662266495,
      "loss": 5.4399,
      "mean_token_accuracy": 0.1970370039343834,
      "num_tokens": 84053664.0,
      "step": 42975
    },
    {
      "entropy": 5.897338819503784,
      "epoch": 4.296296296296296,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003282796832336801,
      "loss": 5.3149,
      "mean_token_accuracy": 0.2125254526734352,
      "num_tokens": 84064444.0,
      "step": 42980
    },
    {
      "entropy": 5.953618144989013,
      "epoch": 4.2967961213575245,
      "grad_norm": 1.140625,
      "learning_rate": 0.000328245008504192,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20692121088504792,
      "num_tokens": 84074433.0,
      "step": 42985
    },
    {
      "entropy": 5.949477815628052,
      "epoch": 4.297295946418753,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032821033243507323,
      "loss": 5.4285,
      "mean_token_accuracy": 0.1937796726822853,
      "num_tokens": 84084798.0,
      "step": 42990
    },
    {
      "entropy": 5.846411323547363,
      "epoch": 4.297795771479982,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003281756550271962,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21693576425313948,
      "num_tokens": 84095562.0,
      "step": 42995
    },
    {
      "entropy": 5.905516338348389,
      "epoch": 4.298295596541211,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003281409762814334,
      "loss": 5.3012,
      "mean_token_accuracy": 0.20954070687294007,
      "num_tokens": 84104629.0,
      "step": 43000
    },
    {
      "entropy": 5.889457654953003,
      "epoch": 4.298795421602439,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003281062961986573,
      "loss": 5.2787,
      "mean_token_accuracy": 0.21054794043302535,
      "num_tokens": 84113000.0,
      "step": 43005
    },
    {
      "entropy": 5.927538442611694,
      "epoch": 4.299295246663668,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003280716147797405,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20345885455608367,
      "num_tokens": 84122510.0,
      "step": 43010
    },
    {
      "entropy": 5.906912326812744,
      "epoch": 4.299795071724896,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032803693202555545,
      "loss": 5.3464,
      "mean_token_accuracy": 0.20823132395744323,
      "num_tokens": 84131605.0,
      "step": 43015
    },
    {
      "entropy": 5.8552481174469,
      "epoch": 4.3002948967861245,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00032800224793697486,
      "loss": 5.289,
      "mean_token_accuracy": 0.2175571873784065,
      "num_tokens": 84141985.0,
      "step": 43020
    },
    {
      "entropy": 5.86442084312439,
      "epoch": 4.300794721847353,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00032796756251487135,
      "loss": 5.2785,
      "mean_token_accuracy": 0.2144756108522415,
      "num_tokens": 84152944.0,
      "step": 43025
    },
    {
      "entropy": 5.973126029968261,
      "epoch": 4.301294546908582,
      "grad_norm": 1.125,
      "learning_rate": 0.0003279328757601176,
      "loss": 5.3826,
      "mean_token_accuracy": 0.20222160071134568,
      "num_tokens": 84162455.0,
      "step": 43030
    },
    {
      "entropy": 5.8783022403717045,
      "epoch": 4.301794371969811,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003278981876735864,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20893764197826387,
      "num_tokens": 84172484.0,
      "step": 43035
    },
    {
      "entropy": 5.986411619186401,
      "epoch": 4.302294197031039,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032786349825615033,
      "loss": 5.4603,
      "mean_token_accuracy": 0.1925027310848236,
      "num_tokens": 84181748.0,
      "step": 43040
    },
    {
      "entropy": 5.997806692123413,
      "epoch": 4.302794022092268,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003278288075086822,
      "loss": 5.3518,
      "mean_token_accuracy": 0.20944347530603408,
      "num_tokens": 84191332.0,
      "step": 43045
    },
    {
      "entropy": 5.963162136077881,
      "epoch": 4.303293847153496,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003277941154320549,
      "loss": 5.3675,
      "mean_token_accuracy": 0.2107326164841652,
      "num_tokens": 84200752.0,
      "step": 43050
    },
    {
      "entropy": 5.970373201370239,
      "epoch": 4.3037936722147245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032775942202714126,
      "loss": 5.3882,
      "mean_token_accuracy": 0.20112041532993316,
      "num_tokens": 84209913.0,
      "step": 43055
    },
    {
      "entropy": 5.993549251556397,
      "epoch": 4.304293497275953,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032772472729481416,
      "loss": 5.4739,
      "mean_token_accuracy": 0.19950639754533767,
      "num_tokens": 84219194.0,
      "step": 43060
    },
    {
      "entropy": 5.900812339782715,
      "epoch": 4.304793322337182,
      "grad_norm": 1.125,
      "learning_rate": 0.00032769003123594645,
      "loss": 5.2846,
      "mean_token_accuracy": 0.21103867292404174,
      "num_tokens": 84227885.0,
      "step": 43065
    },
    {
      "entropy": 6.0181409358978275,
      "epoch": 4.305293147398411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003276553338514111,
      "loss": 5.4568,
      "mean_token_accuracy": 0.19701223224401473,
      "num_tokens": 84237730.0,
      "step": 43070
    },
    {
      "entropy": 5.9400848865509035,
      "epoch": 4.305792972459639,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003276206351420811,
      "loss": 5.3124,
      "mean_token_accuracy": 0.21162826865911483,
      "num_tokens": 84246918.0,
      "step": 43075
    },
    {
      "entropy": 5.89477767944336,
      "epoch": 4.306292797520868,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032758593510882954,
      "loss": 5.3148,
      "mean_token_accuracy": 0.2141265705227852,
      "num_tokens": 84256584.0,
      "step": 43080
    },
    {
      "entropy": 5.89746732711792,
      "epoch": 4.306792622582096,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003275512337525293,
      "loss": 5.3279,
      "mean_token_accuracy": 0.20753991156816481,
      "num_tokens": 84266370.0,
      "step": 43085
    },
    {
      "entropy": 5.935218381881714,
      "epoch": 4.3072924476433245,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032751653107405363,
      "loss": 5.3559,
      "mean_token_accuracy": 0.2005547985434532,
      "num_tokens": 84276315.0,
      "step": 43090
    },
    {
      "entropy": 5.910463476181031,
      "epoch": 4.307792272704553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032748182707427546,
      "loss": 5.369,
      "mean_token_accuracy": 0.21162097305059432,
      "num_tokens": 84285936.0,
      "step": 43095
    },
    {
      "entropy": 5.961101055145264,
      "epoch": 4.308292097765782,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032744712175406806,
      "loss": 5.3376,
      "mean_token_accuracy": 0.20892414152622224,
      "num_tokens": 84295533.0,
      "step": 43100
    },
    {
      "entropy": 5.895693063735962,
      "epoch": 4.308791922827011,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003274124151143046,
      "loss": 5.3271,
      "mean_token_accuracy": 0.21109672635793686,
      "num_tokens": 84303936.0,
      "step": 43105
    },
    {
      "entropy": 5.865316438674927,
      "epoch": 4.309291747888239,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003273777071558582,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21380651146173477,
      "num_tokens": 84313094.0,
      "step": 43110
    },
    {
      "entropy": 5.928299903869629,
      "epoch": 4.309791572949468,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003273429978796022,
      "loss": 5.3396,
      "mean_token_accuracy": 0.21028394848108292,
      "num_tokens": 84323447.0,
      "step": 43115
    },
    {
      "entropy": 5.927888870239258,
      "epoch": 4.310291398010696,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032730828728640983,
      "loss": 5.4138,
      "mean_token_accuracy": 0.20422422140836716,
      "num_tokens": 84333661.0,
      "step": 43120
    },
    {
      "entropy": 5.898119401931763,
      "epoch": 4.3107912230719245,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032727357537715445,
      "loss": 5.2564,
      "mean_token_accuracy": 0.2152776837348938,
      "num_tokens": 84343103.0,
      "step": 43125
    },
    {
      "entropy": 5.8881110668182375,
      "epoch": 4.311291048133153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003272388621527093,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21466884911060333,
      "num_tokens": 84352407.0,
      "step": 43130
    },
    {
      "entropy": 5.955860662460327,
      "epoch": 4.311790873194382,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003272041476139479,
      "loss": 5.4592,
      "mean_token_accuracy": 0.19599487483501435,
      "num_tokens": 84362449.0,
      "step": 43135
    },
    {
      "entropy": 5.924659252166748,
      "epoch": 4.312290698255611,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032716943176174353,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20432025492191314,
      "num_tokens": 84371540.0,
      "step": 43140
    },
    {
      "entropy": 5.8950988292694095,
      "epoch": 4.312790523316839,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003271347145969696,
      "loss": 5.2614,
      "mean_token_accuracy": 0.21784522831439973,
      "num_tokens": 84381688.0,
      "step": 43145
    },
    {
      "entropy": 5.957766199111939,
      "epoch": 4.313290348378068,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032709999612049985,
      "loss": 5.4255,
      "mean_token_accuracy": 0.20204849243164064,
      "num_tokens": 84392895.0,
      "step": 43150
    },
    {
      "entropy": 5.89048433303833,
      "epoch": 4.313790173439296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003270652763332075,
      "loss": 5.2723,
      "mean_token_accuracy": 0.21602960526943207,
      "num_tokens": 84401724.0,
      "step": 43155
    },
    {
      "entropy": 5.965531921386718,
      "epoch": 4.3142899985005245,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003270305552359662,
      "loss": 5.4483,
      "mean_token_accuracy": 0.19898761808872223,
      "num_tokens": 84411669.0,
      "step": 43160
    },
    {
      "entropy": 5.9222092628479,
      "epoch": 4.314789823561753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032699583282964955,
      "loss": 5.3269,
      "mean_token_accuracy": 0.21061337888240814,
      "num_tokens": 84422338.0,
      "step": 43165
    },
    {
      "entropy": 5.851854276657105,
      "epoch": 4.315289648622982,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032696110911513106,
      "loss": 5.2757,
      "mean_token_accuracy": 0.2246418982744217,
      "num_tokens": 84432263.0,
      "step": 43170
    },
    {
      "entropy": 5.908148145675659,
      "epoch": 4.315789473684211,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003269263840932845,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21128714829683304,
      "num_tokens": 84440481.0,
      "step": 43175
    },
    {
      "entropy": 5.927217817306518,
      "epoch": 4.316289298745439,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032689165776498343,
      "loss": 5.4082,
      "mean_token_accuracy": 0.20597002208232879,
      "num_tokens": 84451592.0,
      "step": 43180
    },
    {
      "entropy": 5.974880504608154,
      "epoch": 4.316789123806668,
      "grad_norm": 1.390625,
      "learning_rate": 0.00032685693013110163,
      "loss": 5.3674,
      "mean_token_accuracy": 0.20019892752170562,
      "num_tokens": 84460757.0,
      "step": 43185
    },
    {
      "entropy": 5.68567008972168,
      "epoch": 4.317288948867896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032682220119251284,
      "loss": 4.978,
      "mean_token_accuracy": 0.24377271682024002,
      "num_tokens": 84471219.0,
      "step": 43190
    },
    {
      "entropy": 5.902726745605468,
      "epoch": 4.3177887739291245,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032678747095009076,
      "loss": 5.4117,
      "mean_token_accuracy": 0.20270862728357314,
      "num_tokens": 84481603.0,
      "step": 43195
    },
    {
      "entropy": 5.8779199600219725,
      "epoch": 4.318288598990353,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032675273940470926,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21482599675655364,
      "num_tokens": 84491731.0,
      "step": 43200
    },
    {
      "entropy": 5.987922811508179,
      "epoch": 4.318788424051582,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032671800655724215,
      "loss": 5.3968,
      "mean_token_accuracy": 0.20613829493522645,
      "num_tokens": 84500824.0,
      "step": 43205
    },
    {
      "entropy": 5.9560857772827145,
      "epoch": 4.319288249112811,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032668327240856324,
      "loss": 5.3518,
      "mean_token_accuracy": 0.20626856982707978,
      "num_tokens": 84509977.0,
      "step": 43210
    },
    {
      "entropy": 5.839247417449951,
      "epoch": 4.319788074174039,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003266485369595465,
      "loss": 5.2859,
      "mean_token_accuracy": 0.20878345519304276,
      "num_tokens": 84520767.0,
      "step": 43215
    },
    {
      "entropy": 5.84418420791626,
      "epoch": 4.320287899235268,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032661380021106593,
      "loss": 5.1844,
      "mean_token_accuracy": 0.2219425454735756,
      "num_tokens": 84531190.0,
      "step": 43220
    },
    {
      "entropy": 6.010970783233643,
      "epoch": 4.320787724296496,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003265790621639955,
      "loss": 5.4547,
      "mean_token_accuracy": 0.19893655329942703,
      "num_tokens": 84541619.0,
      "step": 43225
    },
    {
      "entropy": 6.006933641433716,
      "epoch": 4.3212875493577245,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032654432281920904,
      "loss": 5.4167,
      "mean_token_accuracy": 0.2040778636932373,
      "num_tokens": 84551649.0,
      "step": 43230
    },
    {
      "entropy": 5.828367853164673,
      "epoch": 4.321787374418953,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032650958217758067,
      "loss": 5.2271,
      "mean_token_accuracy": 0.22078922539949417,
      "num_tokens": 84563224.0,
      "step": 43235
    },
    {
      "entropy": 6.06985182762146,
      "epoch": 4.322287199480182,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003264748402399844,
      "loss": 5.5306,
      "mean_token_accuracy": 0.19063864797353744,
      "num_tokens": 84573145.0,
      "step": 43240
    },
    {
      "entropy": 5.972728061676025,
      "epoch": 4.322787024541411,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003264400970072945,
      "loss": 5.3899,
      "mean_token_accuracy": 0.205687952041626,
      "num_tokens": 84582077.0,
      "step": 43245
    },
    {
      "entropy": 5.888902854919434,
      "epoch": 4.323286849602639,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032640535248038493,
      "loss": 5.3348,
      "mean_token_accuracy": 0.21638042479753494,
      "num_tokens": 84590743.0,
      "step": 43250
    },
    {
      "entropy": 5.913424444198609,
      "epoch": 4.323786674663868,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032637060666013004,
      "loss": 5.3462,
      "mean_token_accuracy": 0.2138850659132004,
      "num_tokens": 84600836.0,
      "step": 43255
    },
    {
      "entropy": 5.871462440490722,
      "epoch": 4.324286499725096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003263358595474039,
      "loss": 5.3768,
      "mean_token_accuracy": 0.20175715088844298,
      "num_tokens": 84611415.0,
      "step": 43260
    },
    {
      "entropy": 6.0394556522369385,
      "epoch": 4.3247863247863245,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003263011111430806,
      "loss": 5.4299,
      "mean_token_accuracy": 0.19694327116012572,
      "num_tokens": 84621723.0,
      "step": 43265
    },
    {
      "entropy": 5.896545696258545,
      "epoch": 4.325286149847553,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032626636144803456,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21669901311397552,
      "num_tokens": 84631724.0,
      "step": 43270
    },
    {
      "entropy": 5.900948333740234,
      "epoch": 4.325785974908782,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003262316104631402,
      "loss": 5.2762,
      "mean_token_accuracy": 0.20663199424743653,
      "num_tokens": 84641048.0,
      "step": 43275
    },
    {
      "entropy": 5.862585735321045,
      "epoch": 4.326285799970011,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003261968581892716,
      "loss": 5.3325,
      "mean_token_accuracy": 0.20823441296815873,
      "num_tokens": 84651269.0,
      "step": 43280
    },
    {
      "entropy": 5.979424905776978,
      "epoch": 4.326785625031239,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003261621046273032,
      "loss": 5.418,
      "mean_token_accuracy": 0.20166222751140594,
      "num_tokens": 84660823.0,
      "step": 43285
    },
    {
      "entropy": 5.928414154052734,
      "epoch": 4.327285450092468,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003261273497781095,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21514962166547774,
      "num_tokens": 84670135.0,
      "step": 43290
    },
    {
      "entropy": 5.8966450691223145,
      "epoch": 4.327785275153696,
      "grad_norm": 1.125,
      "learning_rate": 0.0003260925936425648,
      "loss": 5.2788,
      "mean_token_accuracy": 0.22409244924783706,
      "num_tokens": 84680896.0,
      "step": 43295
    },
    {
      "entropy": 5.8526740074157715,
      "epoch": 4.3282851002149245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003260578362215435,
      "loss": 5.2913,
      "mean_token_accuracy": 0.20639881491661072,
      "num_tokens": 84690969.0,
      "step": 43300
    },
    {
      "entropy": 5.919255352020263,
      "epoch": 4.328784925276153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003260230775159203,
      "loss": 5.335,
      "mean_token_accuracy": 0.2081049397587776,
      "num_tokens": 84700229.0,
      "step": 43305
    },
    {
      "entropy": 5.957380914688111,
      "epoch": 4.329284750337382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032598831752656953,
      "loss": 5.3627,
      "mean_token_accuracy": 0.20353077799081803,
      "num_tokens": 84710651.0,
      "step": 43310
    },
    {
      "entropy": 5.914744520187378,
      "epoch": 4.329784575398611,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00032595355625436584,
      "loss": 5.3123,
      "mean_token_accuracy": 0.2081456243991852,
      "num_tokens": 84721155.0,
      "step": 43315
    },
    {
      "entropy": 5.848350143432617,
      "epoch": 4.330284400459839,
      "grad_norm": 1.046875,
      "learning_rate": 0.00032591879370018383,
      "loss": 5.1476,
      "mean_token_accuracy": 0.22584341317415238,
      "num_tokens": 84729955.0,
      "step": 43320
    },
    {
      "entropy": 5.898440980911255,
      "epoch": 4.330784225521068,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032588402986489807,
      "loss": 5.3349,
      "mean_token_accuracy": 0.22095074951648713,
      "num_tokens": 84741870.0,
      "step": 43325
    },
    {
      "entropy": 5.931753921508789,
      "epoch": 4.331284050582296,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003258492647493831,
      "loss": 5.3632,
      "mean_token_accuracy": 0.19946187138557434,
      "num_tokens": 84752367.0,
      "step": 43330
    },
    {
      "entropy": 5.9251847743988035,
      "epoch": 4.3317838756435245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003258144983545138,
      "loss": 5.3346,
      "mean_token_accuracy": 0.204889415204525,
      "num_tokens": 84761997.0,
      "step": 43335
    },
    {
      "entropy": 5.931079483032226,
      "epoch": 4.332283700704753,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032577973068116477,
      "loss": 5.345,
      "mean_token_accuracy": 0.20115093737840653,
      "num_tokens": 84771424.0,
      "step": 43340
    },
    {
      "entropy": 5.875628566741943,
      "epoch": 4.332783525765982,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032574496173021087,
      "loss": 5.3147,
      "mean_token_accuracy": 0.20245178341865538,
      "num_tokens": 84781012.0,
      "step": 43345
    },
    {
      "entropy": 5.983359861373901,
      "epoch": 4.333283350827211,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032571019150252667,
      "loss": 5.4197,
      "mean_token_accuracy": 0.20067842602729796,
      "num_tokens": 84791801.0,
      "step": 43350
    },
    {
      "entropy": 5.968421411514282,
      "epoch": 4.333783175888439,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032567541999898713,
      "loss": 5.3904,
      "mean_token_accuracy": 0.2066498205065727,
      "num_tokens": 84801423.0,
      "step": 43355
    },
    {
      "entropy": 5.912809991836548,
      "epoch": 4.334283000949668,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003256406472204671,
      "loss": 5.2402,
      "mean_token_accuracy": 0.21667885333299636,
      "num_tokens": 84811084.0,
      "step": 43360
    },
    {
      "entropy": 5.872257280349731,
      "epoch": 4.334782826010896,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032560587316784133,
      "loss": 5.2968,
      "mean_token_accuracy": 0.21583369821310044,
      "num_tokens": 84820985.0,
      "step": 43365
    },
    {
      "entropy": 5.877821445465088,
      "epoch": 4.3352826510721245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003255710978419849,
      "loss": 5.3063,
      "mean_token_accuracy": 0.2087443232536316,
      "num_tokens": 84830561.0,
      "step": 43370
    },
    {
      "entropy": 5.87128324508667,
      "epoch": 4.335782476133353,
      "grad_norm": 2.21875,
      "learning_rate": 0.0003255363212437727,
      "loss": 5.4058,
      "mean_token_accuracy": 0.20709093660116196,
      "num_tokens": 84840480.0,
      "step": 43375
    },
    {
      "entropy": 6.003480005264282,
      "epoch": 4.336282301194582,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032550154337407963,
      "loss": 5.4759,
      "mean_token_accuracy": 0.1929636150598526,
      "num_tokens": 84850529.0,
      "step": 43380
    },
    {
      "entropy": 5.914376449584961,
      "epoch": 4.336782126255811,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032546676423378065,
      "loss": 5.3439,
      "mean_token_accuracy": 0.21098688691854478,
      "num_tokens": 84859335.0,
      "step": 43385
    },
    {
      "entropy": 5.9030109405517575,
      "epoch": 4.337281951317039,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003254319838237509,
      "loss": 5.3647,
      "mean_token_accuracy": 0.20324338376522064,
      "num_tokens": 84869092.0,
      "step": 43390
    },
    {
      "entropy": 5.948614025115967,
      "epoch": 4.337781776378268,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003253972021448654,
      "loss": 5.3609,
      "mean_token_accuracy": 0.2009579986333847,
      "num_tokens": 84877825.0,
      "step": 43395
    },
    {
      "entropy": 5.9779585838317875,
      "epoch": 4.338281601439496,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032536241919799933,
      "loss": 5.3285,
      "mean_token_accuracy": 0.2183151736855507,
      "num_tokens": 84886928.0,
      "step": 43400
    },
    {
      "entropy": 5.921256685256958,
      "epoch": 4.3387814265007245,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032532763498402774,
      "loss": 5.3732,
      "mean_token_accuracy": 0.2117169424891472,
      "num_tokens": 84896558.0,
      "step": 43405
    },
    {
      "entropy": 5.905552959442138,
      "epoch": 4.339281251561953,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003252928495038257,
      "loss": 5.3044,
      "mean_token_accuracy": 0.2113443821668625,
      "num_tokens": 84905465.0,
      "step": 43410
    },
    {
      "entropy": 5.931244802474976,
      "epoch": 4.339781076623182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032525806275826856,
      "loss": 5.3276,
      "mean_token_accuracy": 0.21445937901735307,
      "num_tokens": 84914891.0,
      "step": 43415
    },
    {
      "entropy": 5.920033311843872,
      "epoch": 4.340280901684411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003252232747482316,
      "loss": 5.3128,
      "mean_token_accuracy": 0.20855290144681932,
      "num_tokens": 84924482.0,
      "step": 43420
    },
    {
      "entropy": 5.949988842010498,
      "epoch": 4.340780726745639,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003251884854745899,
      "loss": 5.3318,
      "mean_token_accuracy": 0.21012841165065765,
      "num_tokens": 84934016.0,
      "step": 43425
    },
    {
      "entropy": 5.7668787956237795,
      "epoch": 4.341280551806868,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0003251536949382188,
      "loss": 5.2262,
      "mean_token_accuracy": 0.22409867197275163,
      "num_tokens": 84944601.0,
      "step": 43430
    },
    {
      "entropy": 5.889566564559937,
      "epoch": 4.341780376868096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032511890313999367,
      "loss": 5.3126,
      "mean_token_accuracy": 0.20730935782194138,
      "num_tokens": 84954386.0,
      "step": 43435
    },
    {
      "entropy": 5.830828142166138,
      "epoch": 4.3422802019293245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032508411008078976,
      "loss": 5.2347,
      "mean_token_accuracy": 0.22193704396486283,
      "num_tokens": 84964660.0,
      "step": 43440
    },
    {
      "entropy": 5.949969959259033,
      "epoch": 4.342780026990553,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003250493157614826,
      "loss": 5.342,
      "mean_token_accuracy": 0.20581674426794053,
      "num_tokens": 84974813.0,
      "step": 43445
    },
    {
      "entropy": 5.9184661388397215,
      "epoch": 4.343279852051782,
      "grad_norm": 1.0625,
      "learning_rate": 0.00032501452018294754,
      "loss": 5.3862,
      "mean_token_accuracy": 0.20321767181158065,
      "num_tokens": 84985110.0,
      "step": 43450
    },
    {
      "entropy": 5.9183086395263675,
      "epoch": 4.343779677113011,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032497972334605995,
      "loss": 5.3739,
      "mean_token_accuracy": 0.20316651910543443,
      "num_tokens": 84994798.0,
      "step": 43455
    },
    {
      "entropy": 5.940277862548828,
      "epoch": 4.344279502174239,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032494492525169544,
      "loss": 5.304,
      "mean_token_accuracy": 0.21401805728673934,
      "num_tokens": 85003957.0,
      "step": 43460
    },
    {
      "entropy": 5.932051467895508,
      "epoch": 4.344779327235468,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003249101259007295,
      "loss": 5.335,
      "mean_token_accuracy": 0.20693042427301406,
      "num_tokens": 85013785.0,
      "step": 43465
    },
    {
      "entropy": 5.907449102401733,
      "epoch": 4.345279152296696,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032487532529403755,
      "loss": 5.3163,
      "mean_token_accuracy": 0.20969195663928986,
      "num_tokens": 85023277.0,
      "step": 43470
    },
    {
      "entropy": 5.912177562713623,
      "epoch": 4.3457789773579245,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003248405234324953,
      "loss": 5.3072,
      "mean_token_accuracy": 0.21849604845046997,
      "num_tokens": 85033586.0,
      "step": 43475
    },
    {
      "entropy": 5.969133377075195,
      "epoch": 4.346278802419153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032480572031697827,
      "loss": 5.4089,
      "mean_token_accuracy": 0.19948757588863372,
      "num_tokens": 85042803.0,
      "step": 43480
    },
    {
      "entropy": 5.897007894515991,
      "epoch": 4.346778627480382,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003247709159483622,
      "loss": 5.3245,
      "mean_token_accuracy": 0.2062579572200775,
      "num_tokens": 85051921.0,
      "step": 43485
    },
    {
      "entropy": 5.841880989074707,
      "epoch": 4.347278452541611,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003247361103275226,
      "loss": 5.3426,
      "mean_token_accuracy": 0.20426492691040038,
      "num_tokens": 85061300.0,
      "step": 43490
    },
    {
      "entropy": 5.909214830398559,
      "epoch": 4.347778277602839,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003247013034553353,
      "loss": 5.3761,
      "mean_token_accuracy": 0.20886121690273285,
      "num_tokens": 85071438.0,
      "step": 43495
    },
    {
      "entropy": 5.947154140472412,
      "epoch": 4.348278102664068,
      "grad_norm": 1.234375,
      "learning_rate": 0.000324666495332676,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19777599722146988,
      "num_tokens": 85080290.0,
      "step": 43500
    },
    {
      "entropy": 5.93761305809021,
      "epoch": 4.348777927725296,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003246316859604205,
      "loss": 5.2321,
      "mean_token_accuracy": 0.2163703292608261,
      "num_tokens": 85089581.0,
      "step": 43505
    },
    {
      "entropy": 5.881261968612671,
      "epoch": 4.3492777527865245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003245968753394445,
      "loss": 5.3438,
      "mean_token_accuracy": 0.21264994591474534,
      "num_tokens": 85099233.0,
      "step": 43510
    },
    {
      "entropy": 5.908130741119384,
      "epoch": 4.349777577847753,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003245620634706239,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19588148593902588,
      "num_tokens": 85108876.0,
      "step": 43515
    },
    {
      "entropy": 5.89360237121582,
      "epoch": 4.350277402908982,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032452725035483454,
      "loss": 5.3296,
      "mean_token_accuracy": 0.2102837085723877,
      "num_tokens": 85118845.0,
      "step": 43520
    },
    {
      "entropy": 5.942954969406128,
      "epoch": 4.350777227970211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032449243599295225,
      "loss": 5.3603,
      "mean_token_accuracy": 0.21022357642650605,
      "num_tokens": 85129234.0,
      "step": 43525
    },
    {
      "entropy": 5.999971914291382,
      "epoch": 4.351277053031439,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003244576203858531,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19858209490776063,
      "num_tokens": 85138634.0,
      "step": 43530
    },
    {
      "entropy": 5.999076795578003,
      "epoch": 4.351776878092668,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003244228035344129,
      "loss": 5.3997,
      "mean_token_accuracy": 0.19938480854034424,
      "num_tokens": 85148560.0,
      "step": 43535
    },
    {
      "entropy": 5.983792495727539,
      "epoch": 4.352276703153896,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032438798543950765,
      "loss": 5.3768,
      "mean_token_accuracy": 0.2079692855477333,
      "num_tokens": 85157843.0,
      "step": 43540
    },
    {
      "entropy": 5.90248818397522,
      "epoch": 4.3527765282151245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003243531661020134,
      "loss": 5.3878,
      "mean_token_accuracy": 0.20561209321022034,
      "num_tokens": 85169254.0,
      "step": 43545
    },
    {
      "entropy": 5.813577461242676,
      "epoch": 4.353276353276353,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003243183455228062,
      "loss": 5.2488,
      "mean_token_accuracy": 0.21829016059637069,
      "num_tokens": 85178443.0,
      "step": 43550
    },
    {
      "entropy": 5.918198585510254,
      "epoch": 4.353776178337582,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032428352370276217,
      "loss": 5.3507,
      "mean_token_accuracy": 0.206256502866745,
      "num_tokens": 85188596.0,
      "step": 43555
    },
    {
      "entropy": 6.018516635894775,
      "epoch": 4.354276003398811,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003242487006427573,
      "loss": 5.4988,
      "mean_token_accuracy": 0.1916905701160431,
      "num_tokens": 85197610.0,
      "step": 43560
    },
    {
      "entropy": 5.858258247375488,
      "epoch": 4.354775828460039,
      "grad_norm": 1.125,
      "learning_rate": 0.0003242138763436678,
      "loss": 5.2898,
      "mean_token_accuracy": 0.21712450832128524,
      "num_tokens": 85207271.0,
      "step": 43565
    },
    {
      "entropy": 5.8429454326629635,
      "epoch": 4.355275653521268,
      "grad_norm": 1.28125,
      "learning_rate": 0.00032417905080636987,
      "loss": 5.2642,
      "mean_token_accuracy": 0.21804148107767105,
      "num_tokens": 85216803.0,
      "step": 43570
    },
    {
      "entropy": 5.873576068878174,
      "epoch": 4.355775478582496,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003241442240317397,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21214254200458527,
      "num_tokens": 85226586.0,
      "step": 43575
    },
    {
      "entropy": 5.757416200637818,
      "epoch": 4.3562753036437245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003241093960206535,
      "loss": 5.2139,
      "mean_token_accuracy": 0.2302005797624588,
      "num_tokens": 85236877.0,
      "step": 43580
    },
    {
      "entropy": 5.8117632389068605,
      "epoch": 4.356775128704953,
      "grad_norm": 1.125,
      "learning_rate": 0.0003240745667739874,
      "loss": 5.23,
      "mean_token_accuracy": 0.22221189737319946,
      "num_tokens": 85247408.0,
      "step": 43585
    },
    {
      "entropy": 5.965312719345093,
      "epoch": 4.357274953766182,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032403973629261795,
      "loss": 5.3538,
      "mean_token_accuracy": 0.20770427733659744,
      "num_tokens": 85257383.0,
      "step": 43590
    },
    {
      "entropy": 5.978850173950195,
      "epoch": 4.357774778827411,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032400490457742135,
      "loss": 5.4667,
      "mean_token_accuracy": 0.19517406672239304,
      "num_tokens": 85266460.0,
      "step": 43595
    },
    {
      "entropy": 5.888784456253052,
      "epoch": 4.358274603888639,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003239700716292739,
      "loss": 5.3011,
      "mean_token_accuracy": 0.20921433418989183,
      "num_tokens": 85275355.0,
      "step": 43600
    },
    {
      "entropy": 5.884625673294067,
      "epoch": 4.358774428949868,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032393523744905214,
      "loss": 5.2979,
      "mean_token_accuracy": 0.2153917968273163,
      "num_tokens": 85285509.0,
      "step": 43605
    },
    {
      "entropy": 5.933052062988281,
      "epoch": 4.359274254011096,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003239004020376323,
      "loss": 5.3548,
      "mean_token_accuracy": 0.20233195871114731,
      "num_tokens": 85295815.0,
      "step": 43610
    },
    {
      "entropy": 6.032077264785767,
      "epoch": 4.3597740790723245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032386556539589097,
      "loss": 5.5233,
      "mean_token_accuracy": 0.19304767102003098,
      "num_tokens": 85305225.0,
      "step": 43615
    },
    {
      "entropy": 5.951277399063111,
      "epoch": 4.360273904133553,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003238307275247046,
      "loss": 5.3663,
      "mean_token_accuracy": 0.20798682868480683,
      "num_tokens": 85316091.0,
      "step": 43620
    },
    {
      "entropy": 5.909072303771973,
      "epoch": 4.360773729194782,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003237958884249497,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20988404750823975,
      "num_tokens": 85325524.0,
      "step": 43625
    },
    {
      "entropy": 5.964887046813965,
      "epoch": 4.361273554256011,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032376104809750275,
      "loss": 5.4681,
      "mean_token_accuracy": 0.20427092164754868,
      "num_tokens": 85335641.0,
      "step": 43630
    },
    {
      "entropy": 5.884477758407593,
      "epoch": 4.361773379317239,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032372620654324034,
      "loss": 5.313,
      "mean_token_accuracy": 0.21046171635389327,
      "num_tokens": 85345699.0,
      "step": 43635
    },
    {
      "entropy": 5.975217676162719,
      "epoch": 4.362273204378468,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032369136376303925,
      "loss": 5.412,
      "mean_token_accuracy": 0.20141126811504365,
      "num_tokens": 85355309.0,
      "step": 43640
    },
    {
      "entropy": 5.924478960037232,
      "epoch": 4.362773029439696,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032365651975777595,
      "loss": 5.3033,
      "mean_token_accuracy": 0.21387562155723572,
      "num_tokens": 85365012.0,
      "step": 43645
    },
    {
      "entropy": 5.955682849884033,
      "epoch": 4.3632728545009245,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003236216745283271,
      "loss": 5.4321,
      "mean_token_accuracy": 0.2025787800550461,
      "num_tokens": 85374061.0,
      "step": 43650
    },
    {
      "entropy": 5.880499935150146,
      "epoch": 4.363772679562153,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003235868280755694,
      "loss": 5.3092,
      "mean_token_accuracy": 0.21658923625946044,
      "num_tokens": 85383019.0,
      "step": 43655
    },
    {
      "entropy": 5.944695377349854,
      "epoch": 4.364272504623382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003235519804003796,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20766069293022155,
      "num_tokens": 85393044.0,
      "step": 43660
    },
    {
      "entropy": 5.968473052978515,
      "epoch": 4.364772329684611,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032351713150363453,
      "loss": 5.4471,
      "mean_token_accuracy": 0.20075337290763856,
      "num_tokens": 85403287.0,
      "step": 43665
    },
    {
      "entropy": 5.953539657592773,
      "epoch": 4.365272154745839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003234822813862108,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20678330957889557,
      "num_tokens": 85412898.0,
      "step": 43670
    },
    {
      "entropy": 5.919124126434326,
      "epoch": 4.365771979807068,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003234474300489856,
      "loss": 5.3527,
      "mean_token_accuracy": 0.21129605621099473,
      "num_tokens": 85421894.0,
      "step": 43675
    },
    {
      "entropy": 6.00896692276001,
      "epoch": 4.366271804868296,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032341257749283526,
      "loss": 5.3423,
      "mean_token_accuracy": 0.21512164026498795,
      "num_tokens": 85432442.0,
      "step": 43680
    },
    {
      "entropy": 5.904562139511109,
      "epoch": 4.3667716299295245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000323377723718637,
      "loss": 5.3582,
      "mean_token_accuracy": 0.21269619166851045,
      "num_tokens": 85441840.0,
      "step": 43685
    },
    {
      "entropy": 5.897079753875732,
      "epoch": 4.367271454990753,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003233428687272677,
      "loss": 5.3871,
      "mean_token_accuracy": 0.2005034178495407,
      "num_tokens": 85452830.0,
      "step": 43690
    },
    {
      "entropy": 5.980501747131347,
      "epoch": 4.367771280051982,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032330801251960424,
      "loss": 5.3865,
      "mean_token_accuracy": 0.19561844319105148,
      "num_tokens": 85462528.0,
      "step": 43695
    },
    {
      "entropy": 5.742366981506348,
      "epoch": 4.368271105113211,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032327315509652363,
      "loss": 5.1068,
      "mean_token_accuracy": 0.227889846265316,
      "num_tokens": 85472633.0,
      "step": 43700
    },
    {
      "entropy": 5.938648271560669,
      "epoch": 4.368770930174439,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032323829645890283,
      "loss": 5.2775,
      "mean_token_accuracy": 0.2148847311735153,
      "num_tokens": 85481556.0,
      "step": 43705
    },
    {
      "entropy": 5.897350692749024,
      "epoch": 4.369270755235667,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00032320343660761897,
      "loss": 5.2636,
      "mean_token_accuracy": 0.20762865543365477,
      "num_tokens": 85491553.0,
      "step": 43710
    },
    {
      "entropy": 5.849291133880615,
      "epoch": 4.369770580296896,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000323168575543549,
      "loss": 5.3976,
      "mean_token_accuracy": 0.2112387463450432,
      "num_tokens": 85501735.0,
      "step": 43715
    },
    {
      "entropy": 5.877937459945679,
      "epoch": 4.3702704053581245,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003231337132675701,
      "loss": 5.2353,
      "mean_token_accuracy": 0.21137638241052628,
      "num_tokens": 85511539.0,
      "step": 43720
    },
    {
      "entropy": 5.894645833969117,
      "epoch": 4.370770230419353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00032309884978055925,
      "loss": 5.2914,
      "mean_token_accuracy": 0.2165353700518608,
      "num_tokens": 85522392.0,
      "step": 43725
    },
    {
      "entropy": 6.011235523223877,
      "epoch": 4.371270055480582,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003230639850833939,
      "loss": 5.3412,
      "mean_token_accuracy": 0.20833624750375748,
      "num_tokens": 85532846.0,
      "step": 43730
    },
    {
      "entropy": 6.00254635810852,
      "epoch": 4.371769880541811,
      "grad_norm": 1.4375,
      "learning_rate": 0.00032302911917695093,
      "loss": 5.3979,
      "mean_token_accuracy": 0.20620027780532837,
      "num_tokens": 85542116.0,
      "step": 43735
    },
    {
      "entropy": 5.924696016311645,
      "epoch": 4.372269705603039,
      "grad_norm": 1.09375,
      "learning_rate": 0.00032299425206210763,
      "loss": 5.2874,
      "mean_token_accuracy": 0.20972689837217331,
      "num_tokens": 85551951.0,
      "step": 43740
    },
    {
      "entropy": 5.909104824066162,
      "epoch": 4.372769530664267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032295938373974137,
      "loss": 5.3465,
      "mean_token_accuracy": 0.2109329491853714,
      "num_tokens": 85561441.0,
      "step": 43745
    },
    {
      "entropy": 5.89897780418396,
      "epoch": 4.373269355725496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003229245142107294,
      "loss": 5.3942,
      "mean_token_accuracy": 0.2029876008629799,
      "num_tokens": 85571271.0,
      "step": 43750
    },
    {
      "entropy": 5.862679576873779,
      "epoch": 4.3737691807867245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003228896434759489,
      "loss": 5.2301,
      "mean_token_accuracy": 0.21745408773422242,
      "num_tokens": 85581687.0,
      "step": 43755
    },
    {
      "entropy": 5.8819899559021,
      "epoch": 4.374269005847953,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003228547715362773,
      "loss": 5.2821,
      "mean_token_accuracy": 0.21249290853738784,
      "num_tokens": 85591434.0,
      "step": 43760
    },
    {
      "entropy": 5.963987970352173,
      "epoch": 4.374768830909182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032281989839259205,
      "loss": 5.4677,
      "mean_token_accuracy": 0.20206053256988527,
      "num_tokens": 85600949.0,
      "step": 43765
    },
    {
      "entropy": 5.996231174468994,
      "epoch": 4.375268655970411,
      "grad_norm": 1.3125,
      "learning_rate": 0.0003227850240457704,
      "loss": 5.3115,
      "mean_token_accuracy": 0.21590442210435867,
      "num_tokens": 85610495.0,
      "step": 43770
    },
    {
      "entropy": 6.036848020553589,
      "epoch": 4.375768481031639,
      "grad_norm": 1.078125,
      "learning_rate": 0.00032275014849668984,
      "loss": 5.4778,
      "mean_token_accuracy": 0.2038942277431488,
      "num_tokens": 85620912.0,
      "step": 43775
    },
    {
      "entropy": 5.995353078842163,
      "epoch": 4.376268306092867,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003227152717462279,
      "loss": 5.4415,
      "mean_token_accuracy": 0.20659095346927642,
      "num_tokens": 85630166.0,
      "step": 43780
    },
    {
      "entropy": 5.922981548309326,
      "epoch": 4.376768131154096,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032268039379526186,
      "loss": 5.3269,
      "mean_token_accuracy": 0.20815746784210204,
      "num_tokens": 85641313.0,
      "step": 43785
    },
    {
      "entropy": 5.920622158050537,
      "epoch": 4.3772679562153245,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003226455146446695,
      "loss": 5.3183,
      "mean_token_accuracy": 0.2067755103111267,
      "num_tokens": 85649454.0,
      "step": 43790
    },
    {
      "entropy": 5.951517343521118,
      "epoch": 4.377767781276553,
      "grad_norm": 1.125,
      "learning_rate": 0.00032261063429532815,
      "loss": 5.3658,
      "mean_token_accuracy": 0.20472036451101303,
      "num_tokens": 85658534.0,
      "step": 43795
    },
    {
      "entropy": 5.927905941009522,
      "epoch": 4.378267606337782,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032257575274811554,
      "loss": 5.4219,
      "mean_token_accuracy": 0.20050668716430664,
      "num_tokens": 85667501.0,
      "step": 43800
    },
    {
      "entropy": 5.9867981433868405,
      "epoch": 4.378767431399011,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003225408700039093,
      "loss": 5.3696,
      "mean_token_accuracy": 0.1996404320001602,
      "num_tokens": 85675866.0,
      "step": 43805
    },
    {
      "entropy": 5.984622764587402,
      "epoch": 4.379267256460239,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003225059860635869,
      "loss": 5.3013,
      "mean_token_accuracy": 0.20574388206005095,
      "num_tokens": 85685098.0,
      "step": 43810
    },
    {
      "entropy": 5.895149230957031,
      "epoch": 4.379767081521467,
      "grad_norm": 1.1875,
      "learning_rate": 0.00032247110092802625,
      "loss": 5.3577,
      "mean_token_accuracy": 0.2168194830417633,
      "num_tokens": 85695046.0,
      "step": 43815
    },
    {
      "entropy": 5.867125463485718,
      "epoch": 4.380266906582696,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003224362145981048,
      "loss": 5.2683,
      "mean_token_accuracy": 0.21833175122737886,
      "num_tokens": 85703850.0,
      "step": 43820
    },
    {
      "entropy": 5.919351530075073,
      "epoch": 4.3807667316439245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00032240132707470043,
      "loss": 5.3082,
      "mean_token_accuracy": 0.20977627635002136,
      "num_tokens": 85713013.0,
      "step": 43825
    },
    {
      "entropy": 5.905646181106567,
      "epoch": 4.381266556705153,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032236643835869085,
      "loss": 5.3299,
      "mean_token_accuracy": 0.2089082956314087,
      "num_tokens": 85724186.0,
      "step": 43830
    },
    {
      "entropy": 5.909748983383179,
      "epoch": 4.381766381766382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032233154845095384,
      "loss": 5.2674,
      "mean_token_accuracy": 0.2140958234667778,
      "num_tokens": 85733711.0,
      "step": 43835
    },
    {
      "entropy": 5.915919351577759,
      "epoch": 4.382266206827611,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003222966573523673,
      "loss": 5.2911,
      "mean_token_accuracy": 0.20951049774885178,
      "num_tokens": 85743667.0,
      "step": 43840
    },
    {
      "entropy": 5.913338327407837,
      "epoch": 4.382766031888839,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000322261765063809,
      "loss": 5.2609,
      "mean_token_accuracy": 0.20747953355312349,
      "num_tokens": 85753784.0,
      "step": 43845
    },
    {
      "entropy": 5.848644018173218,
      "epoch": 4.383265856950067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003222268715861569,
      "loss": 5.2352,
      "mean_token_accuracy": 0.21751800775527955,
      "num_tokens": 85763453.0,
      "step": 43850
    },
    {
      "entropy": 5.888930702209473,
      "epoch": 4.383765682011296,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003221919769202889,
      "loss": 5.3104,
      "mean_token_accuracy": 0.21080219149589538,
      "num_tokens": 85773749.0,
      "step": 43855
    },
    {
      "entropy": 6.017475748062134,
      "epoch": 4.3842655070725245,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032215708106708284,
      "loss": 5.4586,
      "mean_token_accuracy": 0.19310705214738846,
      "num_tokens": 85784044.0,
      "step": 43860
    },
    {
      "entropy": 5.883640003204346,
      "epoch": 4.384765332133753,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003221221840274168,
      "loss": 5.2789,
      "mean_token_accuracy": 0.20836907029151916,
      "num_tokens": 85793362.0,
      "step": 43865
    },
    {
      "entropy": 5.957608795166015,
      "epoch": 4.385265157194982,
      "grad_norm": 1.109375,
      "learning_rate": 0.00032208728580216864,
      "loss": 5.4574,
      "mean_token_accuracy": 0.20299697071313857,
      "num_tokens": 85803305.0,
      "step": 43870
    },
    {
      "entropy": 5.945132446289063,
      "epoch": 4.385764982256211,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00032205238639221653,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2026263430714607,
      "num_tokens": 85814438.0,
      "step": 43875
    },
    {
      "entropy": 5.923778343200683,
      "epoch": 4.386264807317439,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003220174857984385,
      "loss": 5.3356,
      "mean_token_accuracy": 0.20698966830968857,
      "num_tokens": 85824472.0,
      "step": 43880
    },
    {
      "entropy": 5.973520088195801,
      "epoch": 4.386764632378667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032198258402171267,
      "loss": 5.3346,
      "mean_token_accuracy": 0.20750296264886856,
      "num_tokens": 85834735.0,
      "step": 43885
    },
    {
      "entropy": 5.920005941390992,
      "epoch": 4.387264457439896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032194768106291705,
      "loss": 5.3546,
      "mean_token_accuracy": 0.2090218260884285,
      "num_tokens": 85844575.0,
      "step": 43890
    },
    {
      "entropy": 5.880358266830444,
      "epoch": 4.3877642825011245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00032191277692292984,
      "loss": 5.2404,
      "mean_token_accuracy": 0.22090959399938584,
      "num_tokens": 85854142.0,
      "step": 43895
    },
    {
      "entropy": 5.891748237609863,
      "epoch": 4.388264107562353,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032187787160262924,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2114601120352745,
      "num_tokens": 85862182.0,
      "step": 43900
    },
    {
      "entropy": 5.935897016525269,
      "epoch": 4.388763932623582,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003218429651028935,
      "loss": 5.4372,
      "mean_token_accuracy": 0.20636371225118638,
      "num_tokens": 85872497.0,
      "step": 43905
    },
    {
      "entropy": 5.963131713867187,
      "epoch": 4.389263757684811,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032180805742460086,
      "loss": 5.3939,
      "mean_token_accuracy": 0.20607681572437286,
      "num_tokens": 85882744.0,
      "step": 43910
    },
    {
      "entropy": 5.907522678375244,
      "epoch": 4.389763582746039,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00032177314856862943,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2217534750699997,
      "num_tokens": 85892774.0,
      "step": 43915
    },
    {
      "entropy": 5.906961822509766,
      "epoch": 4.390263407807267,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003217382385358577,
      "loss": 5.2903,
      "mean_token_accuracy": 0.2085677742958069,
      "num_tokens": 85902470.0,
      "step": 43920
    },
    {
      "entropy": 5.866881418228149,
      "epoch": 4.390763232868496,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003217033273271639,
      "loss": 5.2945,
      "mean_token_accuracy": 0.22090629041194915,
      "num_tokens": 85912325.0,
      "step": 43925
    },
    {
      "entropy": 5.943573331832885,
      "epoch": 4.3912630579297245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003216684149434263,
      "loss": 5.3769,
      "mean_token_accuracy": 0.20676788091659545,
      "num_tokens": 85922145.0,
      "step": 43930
    },
    {
      "entropy": 5.929688739776611,
      "epoch": 4.391762882990953,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003216335013855235,
      "loss": 5.4146,
      "mean_token_accuracy": 0.20270097851753235,
      "num_tokens": 85933471.0,
      "step": 43935
    },
    {
      "entropy": 5.814212369918823,
      "epoch": 4.392262708052182,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00032159858665433376,
      "loss": 5.2665,
      "mean_token_accuracy": 0.22896598726511003,
      "num_tokens": 85943071.0,
      "step": 43940
    },
    {
      "entropy": 5.88787350654602,
      "epoch": 4.392762533113411,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032156367075073563,
      "loss": 5.2526,
      "mean_token_accuracy": 0.2114947646856308,
      "num_tokens": 85953096.0,
      "step": 43945
    },
    {
      "entropy": 5.9244688034057615,
      "epoch": 4.393262358174639,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003215287536756074,
      "loss": 5.3218,
      "mean_token_accuracy": 0.20974087417125703,
      "num_tokens": 85962547.0,
      "step": 43950
    },
    {
      "entropy": 5.885758686065674,
      "epoch": 4.393762183235867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00032149383542982786,
      "loss": 5.2576,
      "mean_token_accuracy": 0.21478597670793534,
      "num_tokens": 85972057.0,
      "step": 43955
    },
    {
      "entropy": 5.8583684921264645,
      "epoch": 4.394262008297096,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003214589160142753,
      "loss": 5.3052,
      "mean_token_accuracy": 0.20713797211647034,
      "num_tokens": 85981592.0,
      "step": 43960
    },
    {
      "entropy": 5.934215927124024,
      "epoch": 4.3947618333583245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032142399542982834,
      "loss": 5.4363,
      "mean_token_accuracy": 0.20243093073368074,
      "num_tokens": 85992193.0,
      "step": 43965
    },
    {
      "entropy": 5.914000129699707,
      "epoch": 4.395261658419553,
      "grad_norm": 1.328125,
      "learning_rate": 0.00032138907367736564,
      "loss": 5.3083,
      "mean_token_accuracy": 0.21435363590717316,
      "num_tokens": 86001557.0,
      "step": 43970
    },
    {
      "entropy": 5.972091865539551,
      "epoch": 4.395761483480782,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003213541507577658,
      "loss": 5.3808,
      "mean_token_accuracy": 0.2093152552843094,
      "num_tokens": 86010935.0,
      "step": 43975
    },
    {
      "entropy": 6.0181496143341064,
      "epoch": 4.396261308542011,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003213192266719074,
      "loss": 5.401,
      "mean_token_accuracy": 0.20207332223653793,
      "num_tokens": 86019735.0,
      "step": 43980
    },
    {
      "entropy": 5.8160628318786625,
      "epoch": 4.396761133603239,
      "grad_norm": 0.99609375,
      "learning_rate": 0.00032128430142066926,
      "loss": 5.3163,
      "mean_token_accuracy": 0.21222808063030243,
      "num_tokens": 86030419.0,
      "step": 43985
    },
    {
      "entropy": 5.946775293350219,
      "epoch": 4.397260958664467,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032124937500492995,
      "loss": 5.4031,
      "mean_token_accuracy": 0.20477544665336608,
      "num_tokens": 86041769.0,
      "step": 43990
    },
    {
      "entropy": 5.866556978225708,
      "epoch": 4.397760783725696,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003212144474255682,
      "loss": 5.3074,
      "mean_token_accuracy": 0.21144883781671525,
      "num_tokens": 86053517.0,
      "step": 43995
    },
    {
      "entropy": 5.938699340820312,
      "epoch": 4.3982606087869245,
      "grad_norm": 1.1875,
      "learning_rate": 0.000321179518683463,
      "loss": 5.4215,
      "mean_token_accuracy": 0.20367067009210588,
      "num_tokens": 86063605.0,
      "step": 44000
    },
    {
      "entropy": 5.964485549926758,
      "epoch": 4.398760433848153,
      "grad_norm": 1.25,
      "learning_rate": 0.00032114458877949287,
      "loss": 5.3081,
      "mean_token_accuracy": 0.21849497854709626,
      "num_tokens": 86073753.0,
      "step": 44005
    },
    {
      "entropy": 5.933976411819458,
      "epoch": 4.399260258909382,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003211096577145368,
      "loss": 5.258,
      "mean_token_accuracy": 0.21825719028711318,
      "num_tokens": 86083434.0,
      "step": 44010
    },
    {
      "entropy": 5.913566207885742,
      "epoch": 4.399760083970611,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003210747254894736,
      "loss": 5.3785,
      "mean_token_accuracy": 0.20427274107933044,
      "num_tokens": 86093254.0,
      "step": 44015
    },
    {
      "entropy": 5.890655517578125,
      "epoch": 4.400259909031838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003210397921051822,
      "loss": 5.2656,
      "mean_token_accuracy": 0.2100321128964424,
      "num_tokens": 86103257.0,
      "step": 44020
    },
    {
      "entropy": 5.883397054672241,
      "epoch": 4.400759734093067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003210048575625414,
      "loss": 5.3827,
      "mean_token_accuracy": 0.20446165800094604,
      "num_tokens": 86113335.0,
      "step": 44025
    },
    {
      "entropy": 5.944206285476684,
      "epoch": 4.401259559154296,
      "grad_norm": 1.296875,
      "learning_rate": 0.00032096992186243025,
      "loss": 5.4162,
      "mean_token_accuracy": 0.196287339925766,
      "num_tokens": 86123643.0,
      "step": 44030
    },
    {
      "entropy": 5.919605493545532,
      "epoch": 4.4017593842155245,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003209349850057277,
      "loss": 5.3098,
      "mean_token_accuracy": 0.21883776038885117,
      "num_tokens": 86133421.0,
      "step": 44035
    },
    {
      "entropy": 5.926914978027344,
      "epoch": 4.402259209276753,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00032090004699331277,
      "loss": 5.3167,
      "mean_token_accuracy": 0.20827930569648742,
      "num_tokens": 86143518.0,
      "step": 44040
    },
    {
      "entropy": 5.919296503067017,
      "epoch": 4.402759034337982,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00032086510782606447,
      "loss": 5.324,
      "mean_token_accuracy": 0.20843631029129028,
      "num_tokens": 86152472.0,
      "step": 44045
    },
    {
      "entropy": 6.016710090637207,
      "epoch": 4.403258859399211,
      "grad_norm": 1.25,
      "learning_rate": 0.0003208301675048619,
      "loss": 5.3268,
      "mean_token_accuracy": 0.20965345203876495,
      "num_tokens": 86163306.0,
      "step": 44050
    },
    {
      "entropy": 5.867187070846557,
      "epoch": 4.403758684460438,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032079522603058403,
      "loss": 5.3097,
      "mean_token_accuracy": 0.22032616883516312,
      "num_tokens": 86173690.0,
      "step": 44055
    },
    {
      "entropy": 5.853123664855957,
      "epoch": 4.404258509521667,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003207602834041101,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2251799985766411,
      "num_tokens": 86183326.0,
      "step": 44060
    },
    {
      "entropy": 5.922001457214355,
      "epoch": 4.404758334582896,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00032072533962631915,
      "loss": 5.3343,
      "mean_token_accuracy": 0.20375039130449296,
      "num_tokens": 86193000.0,
      "step": 44065
    },
    {
      "entropy": 5.897479248046875,
      "epoch": 4.4052581596441245,
      "grad_norm": 1.0,
      "learning_rate": 0.00032069039469809045,
      "loss": 5.2766,
      "mean_token_accuracy": 0.20799532979726792,
      "num_tokens": 86203315.0,
      "step": 44070
    },
    {
      "entropy": 5.9168675422668455,
      "epoch": 4.405757984705353,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003206554486203032,
      "loss": 5.3667,
      "mean_token_accuracy": 0.20889964401721955,
      "num_tokens": 86213242.0,
      "step": 44075
    },
    {
      "entropy": 5.878184127807617,
      "epoch": 4.406257809766582,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003206205013938366,
      "loss": 5.3518,
      "mean_token_accuracy": 0.2118893727660179,
      "num_tokens": 86222651.0,
      "step": 44080
    },
    {
      "entropy": 5.9815834045410154,
      "epoch": 4.406757634827811,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032058555301957,
      "loss": 5.4049,
      "mean_token_accuracy": 0.19813822209835052,
      "num_tokens": 86232056.0,
      "step": 44085
    },
    {
      "entropy": 5.998028755187988,
      "epoch": 4.407257459889038,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003205506034983825,
      "loss": 5.3907,
      "mean_token_accuracy": 0.20374191999435426,
      "num_tokens": 86241818.0,
      "step": 44090
    },
    {
      "entropy": 5.931292152404785,
      "epoch": 4.407757284950267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00032051565283115356,
      "loss": 5.3282,
      "mean_token_accuracy": 0.2101542443037033,
      "num_tokens": 86251661.0,
      "step": 44095
    },
    {
      "entropy": 5.866398096084595,
      "epoch": 4.408257110011496,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003204807010187625,
      "loss": 5.2812,
      "mean_token_accuracy": 0.20497573614120485,
      "num_tokens": 86261064.0,
      "step": 44100
    },
    {
      "entropy": 5.954205274581909,
      "epoch": 4.4087569350727245,
      "grad_norm": 1.21875,
      "learning_rate": 0.00032044574806208877,
      "loss": 5.3967,
      "mean_token_accuracy": 0.2085517331957817,
      "num_tokens": 86270621.0,
      "step": 44105
    },
    {
      "entropy": 5.945240879058838,
      "epoch": 4.409256760133953,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032041079396201167,
      "loss": 5.3475,
      "mean_token_accuracy": 0.20460176318883896,
      "num_tokens": 86280553.0,
      "step": 44110
    },
    {
      "entropy": 6.086656713485718,
      "epoch": 4.409756585195182,
      "grad_norm": 1.234375,
      "learning_rate": 0.00032037583871941073,
      "loss": 5.477,
      "mean_token_accuracy": 0.19654002487659455,
      "num_tokens": 86290376.0,
      "step": 44115
    },
    {
      "entropy": 5.950437879562378,
      "epoch": 4.410256410256411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003203408823351653,
      "loss": 5.2993,
      "mean_token_accuracy": 0.20813087970018387,
      "num_tokens": 86300087.0,
      "step": 44120
    },
    {
      "entropy": 5.946750640869141,
      "epoch": 4.410756235317638,
      "grad_norm": 1.203125,
      "learning_rate": 0.00032030592481015495,
      "loss": 5.374,
      "mean_token_accuracy": 0.19885123372077942,
      "num_tokens": 86309624.0,
      "step": 44125
    },
    {
      "entropy": 5.906940269470215,
      "epoch": 4.411256060378867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00032027096614525917,
      "loss": 5.2637,
      "mean_token_accuracy": 0.21333847045898438,
      "num_tokens": 86318791.0,
      "step": 44130
    },
    {
      "entropy": 5.909314584732056,
      "epoch": 4.411755885440096,
      "grad_norm": 1.9140625,
      "learning_rate": 0.00032023600634135756,
      "loss": 5.2982,
      "mean_token_accuracy": 0.2159011721611023,
      "num_tokens": 86328679.0,
      "step": 44135
    },
    {
      "entropy": 5.945114850997925,
      "epoch": 4.4122557105013245,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003202010453993297,
      "loss": 5.4499,
      "mean_token_accuracy": 0.20080929398536682,
      "num_tokens": 86337587.0,
      "step": 44140
    },
    {
      "entropy": 5.952417516708374,
      "epoch": 4.412755535562553,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00032016608332005516,
      "loss": 5.3914,
      "mean_token_accuracy": 0.20085454136133193,
      "num_tokens": 86347225.0,
      "step": 44145
    },
    {
      "entropy": 5.957817220687867,
      "epoch": 4.413255360623782,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003201311201044136,
      "loss": 5.3506,
      "mean_token_accuracy": 0.2013239711523056,
      "num_tokens": 86357137.0,
      "step": 44150
    },
    {
      "entropy": 5.98213906288147,
      "epoch": 4.413755185685011,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00032009615575328453,
      "loss": 5.4269,
      "mean_token_accuracy": 0.19723293781280518,
      "num_tokens": 86366642.0,
      "step": 44155
    },
    {
      "entropy": 5.898977327346802,
      "epoch": 4.414255010746238,
      "grad_norm": 1.078125,
      "learning_rate": 0.000320061190267548,
      "loss": 5.3338,
      "mean_token_accuracy": 0.20736535936594008,
      "num_tokens": 86378163.0,
      "step": 44160
    },
    {
      "entropy": 5.903712797164917,
      "epoch": 4.414754835807467,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00032002622364808334,
      "loss": 5.2699,
      "mean_token_accuracy": 0.21505753248929976,
      "num_tokens": 86388163.0,
      "step": 44165
    },
    {
      "entropy": 5.8961474895477295,
      "epoch": 4.415254660868696,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031999125589577055,
      "loss": 5.3214,
      "mean_token_accuracy": 0.2079936906695366,
      "num_tokens": 86397389.0,
      "step": 44170
    },
    {
      "entropy": 5.93249135017395,
      "epoch": 4.4157544859299245,
      "grad_norm": 1.515625,
      "learning_rate": 0.0003199562870114894,
      "loss": 5.4065,
      "mean_token_accuracy": 0.1955591171979904,
      "num_tokens": 86407223.0,
      "step": 44175
    },
    {
      "entropy": 5.945970058441162,
      "epoch": 4.416254310991153,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031992131699611954,
      "loss": 5.3486,
      "mean_token_accuracy": 0.20319111049175262,
      "num_tokens": 86416778.0,
      "step": 44180
    },
    {
      "entropy": 5.930273389816284,
      "epoch": 4.416754136052382,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003198863458505409,
      "loss": 5.2682,
      "mean_token_accuracy": 0.2211553066968918,
      "num_tokens": 86426720.0,
      "step": 44185
    },
    {
      "entropy": 5.964242839813233,
      "epoch": 4.417253961113611,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003198513735756334,
      "loss": 5.3867,
      "mean_token_accuracy": 0.21580284982919692,
      "num_tokens": 86437002.0,
      "step": 44190
    },
    {
      "entropy": 5.754559898376465,
      "epoch": 4.417753786174838,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031981640017227694,
      "loss": 5.2142,
      "mean_token_accuracy": 0.2245133712887764,
      "num_tokens": 86447425.0,
      "step": 44195
    },
    {
      "entropy": 5.868289232254028,
      "epoch": 4.418253611236067,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031978142564135126,
      "loss": 5.2085,
      "mean_token_accuracy": 0.2274120882153511,
      "num_tokens": 86456863.0,
      "step": 44200
    },
    {
      "entropy": 5.92948546409607,
      "epoch": 4.418753436297296,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031974644998373645,
      "loss": 5.3098,
      "mean_token_accuracy": 0.21756741106510163,
      "num_tokens": 86466097.0,
      "step": 44205
    },
    {
      "entropy": 5.933594036102295,
      "epoch": 4.4192532613585245,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031971147320031257,
      "loss": 5.3077,
      "mean_token_accuracy": 0.20613616704940796,
      "num_tokens": 86474646.0,
      "step": 44210
    },
    {
      "entropy": 5.918063688278198,
      "epoch": 4.419753086419753,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031967649529195943,
      "loss": 5.2784,
      "mean_token_accuracy": 0.2195335417985916,
      "num_tokens": 86483902.0,
      "step": 44215
    },
    {
      "entropy": 5.914741849899292,
      "epoch": 4.420252911480982,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003196415162595572,
      "loss": 5.4208,
      "mean_token_accuracy": 0.19640214145183563,
      "num_tokens": 86494273.0,
      "step": 44220
    },
    {
      "entropy": 5.881097793579102,
      "epoch": 4.420752736542211,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003196065361039859,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2134188175201416,
      "num_tokens": 86503912.0,
      "step": 44225
    },
    {
      "entropy": 5.97427921295166,
      "epoch": 4.421252561603438,
      "grad_norm": 1.0625,
      "learning_rate": 0.00031957155482612563,
      "loss": 5.381,
      "mean_token_accuracy": 0.20835675597190856,
      "num_tokens": 86513241.0,
      "step": 44230
    },
    {
      "entropy": 5.956184911727905,
      "epoch": 4.421752386664667,
      "grad_norm": 1.125,
      "learning_rate": 0.00031953657242685655,
      "loss": 5.3641,
      "mean_token_accuracy": 0.2017914205789566,
      "num_tokens": 86523367.0,
      "step": 44235
    },
    {
      "entropy": 5.976665735244751,
      "epoch": 4.422252211725896,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003195015889070588,
      "loss": 5.4397,
      "mean_token_accuracy": 0.20403016805648805,
      "num_tokens": 86534533.0,
      "step": 44240
    },
    {
      "entropy": 5.970625638961792,
      "epoch": 4.4227520367871245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031946660426761237,
      "loss": 5.3185,
      "mean_token_accuracy": 0.2045479252934456,
      "num_tokens": 86543150.0,
      "step": 44245
    },
    {
      "entropy": 5.901555967330933,
      "epoch": 4.423251861848353,
      "grad_norm": 1.25,
      "learning_rate": 0.00031943161850939774,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2093432679772377,
      "num_tokens": 86552991.0,
      "step": 44250
    },
    {
      "entropy": 5.811426639556885,
      "epoch": 4.423751686909582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00031939663163329504,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21819376498460769,
      "num_tokens": 86561976.0,
      "step": 44255
    },
    {
      "entropy": 5.884646654129028,
      "epoch": 4.424251511970811,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003193616436401844,
      "loss": 5.3953,
      "mean_token_accuracy": 0.20559686720371245,
      "num_tokens": 86570854.0,
      "step": 44260
    },
    {
      "entropy": 5.822417974472046,
      "epoch": 4.424751337032038,
      "grad_norm": 1.125,
      "learning_rate": 0.0003193266545309464,
      "loss": 5.2835,
      "mean_token_accuracy": 0.21898176819086074,
      "num_tokens": 86579920.0,
      "step": 44265
    },
    {
      "entropy": 5.922923469543457,
      "epoch": 4.425251162093267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031929166430646105,
      "loss": 5.329,
      "mean_token_accuracy": 0.20818594396114348,
      "num_tokens": 86589697.0,
      "step": 44270
    },
    {
      "entropy": 5.968634366989136,
      "epoch": 4.425750987154496,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003192566729676088,
      "loss": 5.3988,
      "mean_token_accuracy": 0.2022494077682495,
      "num_tokens": 86599033.0,
      "step": 44275
    },
    {
      "entropy": 6.012994050979614,
      "epoch": 4.4262508122157245,
      "grad_norm": 1.125,
      "learning_rate": 0.00031922168051527,
      "loss": 5.4487,
      "mean_token_accuracy": 0.20013859868049622,
      "num_tokens": 86609219.0,
      "step": 44280
    },
    {
      "entropy": 6.106365060806274,
      "epoch": 4.426750637276953,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003191866869503252,
      "loss": 5.5384,
      "mean_token_accuracy": 0.191427281498909,
      "num_tokens": 86618919.0,
      "step": 44285
    },
    {
      "entropy": 5.9769984722137455,
      "epoch": 4.427250462338182,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003191516922736547,
      "loss": 5.3399,
      "mean_token_accuracy": 0.21096887588500976,
      "num_tokens": 86627752.0,
      "step": 44290
    },
    {
      "entropy": 5.885395240783692,
      "epoch": 4.427750287399411,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000319116696486139,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21693717390298844,
      "num_tokens": 86638926.0,
      "step": 44295
    },
    {
      "entropy": 5.904186391830445,
      "epoch": 4.428250112460638,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003190816995886586,
      "loss": 5.384,
      "mean_token_accuracy": 0.2063776358962059,
      "num_tokens": 86649142.0,
      "step": 44300
    },
    {
      "entropy": 5.901856565475464,
      "epoch": 4.428749937521867,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031904670158209404,
      "loss": 5.2864,
      "mean_token_accuracy": 0.21383069604635238,
      "num_tokens": 86658946.0,
      "step": 44305
    },
    {
      "entropy": 5.874195241928101,
      "epoch": 4.429249762583096,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031901170246732564,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2150489181280136,
      "num_tokens": 86668304.0,
      "step": 44310
    },
    {
      "entropy": 5.767045640945435,
      "epoch": 4.4297495876443245,
      "grad_norm": 1.125,
      "learning_rate": 0.00031897670224523413,
      "loss": 5.178,
      "mean_token_accuracy": 0.22473168075084687,
      "num_tokens": 86678116.0,
      "step": 44315
    },
    {
      "entropy": 5.991534423828125,
      "epoch": 4.430249412705553,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031894170091670016,
      "loss": 5.4618,
      "mean_token_accuracy": 0.1965702459216118,
      "num_tokens": 86689213.0,
      "step": 44320
    },
    {
      "entropy": 5.954588317871094,
      "epoch": 4.430749237766782,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031890669848260437,
      "loss": 5.4441,
      "mean_token_accuracy": 0.21143665462732314,
      "num_tokens": 86698111.0,
      "step": 44325
    },
    {
      "entropy": 6.019769287109375,
      "epoch": 4.431249062828011,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003188716949438273,
      "loss": 5.4554,
      "mean_token_accuracy": 0.1970852181315422,
      "num_tokens": 86707412.0,
      "step": 44330
    },
    {
      "entropy": 5.93936128616333,
      "epoch": 4.431748887889238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031883669030124967,
      "loss": 5.3237,
      "mean_token_accuracy": 0.21490567773580552,
      "num_tokens": 86716223.0,
      "step": 44335
    },
    {
      "entropy": 5.958867120742798,
      "epoch": 4.432248712950467,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031880168455575216,
      "loss": 5.426,
      "mean_token_accuracy": 0.20571840703487396,
      "num_tokens": 86725879.0,
      "step": 44340
    },
    {
      "entropy": 5.983691644668579,
      "epoch": 4.432748538011696,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031876667770821556,
      "loss": 5.4028,
      "mean_token_accuracy": 0.20399300903081893,
      "num_tokens": 86735681.0,
      "step": 44345
    },
    {
      "entropy": 5.886743640899658,
      "epoch": 4.4332483630729245,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003187316697595206,
      "loss": 5.3184,
      "mean_token_accuracy": 0.2217646524310112,
      "num_tokens": 86744647.0,
      "step": 44350
    },
    {
      "entropy": 5.999347591400147,
      "epoch": 4.433748188134153,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003186966607105481,
      "loss": 5.3918,
      "mean_token_accuracy": 0.19736944437026976,
      "num_tokens": 86755020.0,
      "step": 44355
    },
    {
      "entropy": 6.058748483657837,
      "epoch": 4.434248013195382,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031866165056217887,
      "loss": 5.509,
      "mean_token_accuracy": 0.19731165021657943,
      "num_tokens": 86766179.0,
      "step": 44360
    },
    {
      "entropy": 6.021730947494507,
      "epoch": 4.434747838256611,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003186266393152937,
      "loss": 5.3944,
      "mean_token_accuracy": 0.20258630067110062,
      "num_tokens": 86776823.0,
      "step": 44365
    },
    {
      "entropy": 5.905215835571289,
      "epoch": 4.435247663317838,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003185916269707736,
      "loss": 5.3953,
      "mean_token_accuracy": 0.2042653188109398,
      "num_tokens": 86787256.0,
      "step": 44370
    },
    {
      "entropy": 5.9494242668151855,
      "epoch": 4.435747488379067,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031855661352949934,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20525456368923187,
      "num_tokens": 86796378.0,
      "step": 44375
    },
    {
      "entropy": 5.991200351715088,
      "epoch": 4.436247313440296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003185215989923519,
      "loss": 5.3854,
      "mean_token_accuracy": 0.2057402178645134,
      "num_tokens": 86806410.0,
      "step": 44380
    },
    {
      "entropy": 5.931319904327393,
      "epoch": 4.4367471385015245,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003184865833602121,
      "loss": 5.3778,
      "mean_token_accuracy": 0.20738694220781326,
      "num_tokens": 86815244.0,
      "step": 44385
    },
    {
      "entropy": 5.9082786560058596,
      "epoch": 4.437246963562753,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031845156663396125,
      "loss": 5.306,
      "mean_token_accuracy": 0.21004513800144195,
      "num_tokens": 86825911.0,
      "step": 44390
    },
    {
      "entropy": 5.874542284011841,
      "epoch": 4.437746788623982,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031841654881448005,
      "loss": 5.2798,
      "mean_token_accuracy": 0.22023552507162095,
      "num_tokens": 86836488.0,
      "step": 44395
    },
    {
      "entropy": 5.851139259338379,
      "epoch": 4.438246613685211,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031838152990264966,
      "loss": 5.2025,
      "mean_token_accuracy": 0.22319728434085845,
      "num_tokens": 86846061.0,
      "step": 44400
    },
    {
      "entropy": 5.939628744125367,
      "epoch": 4.438746438746438,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003183465098993512,
      "loss": 5.3619,
      "mean_token_accuracy": 0.21175449192523957,
      "num_tokens": 86854640.0,
      "step": 44405
    },
    {
      "entropy": 5.912424802780151,
      "epoch": 4.439246263807667,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031831148880546566,
      "loss": 5.3316,
      "mean_token_accuracy": 0.21084719598293306,
      "num_tokens": 86864256.0,
      "step": 44410
    },
    {
      "entropy": 5.945914554595947,
      "epoch": 4.439746088868896,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003182764666218742,
      "loss": 5.3417,
      "mean_token_accuracy": 0.20794398039579393,
      "num_tokens": 86873812.0,
      "step": 44415
    },
    {
      "entropy": 5.973021221160889,
      "epoch": 4.4402459139301245,
      "grad_norm": 1.1875,
      "learning_rate": 0.000318241443349458,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20415844172239303,
      "num_tokens": 86883103.0,
      "step": 44420
    },
    {
      "entropy": 6.003882789611817,
      "epoch": 4.440745738991353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003182064189890982,
      "loss": 5.4179,
      "mean_token_accuracy": 0.2049158662557602,
      "num_tokens": 86892263.0,
      "step": 44425
    },
    {
      "entropy": 5.924920892715454,
      "epoch": 4.441245564052582,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000318171393541676,
      "loss": 5.3178,
      "mean_token_accuracy": 0.2077888086438179,
      "num_tokens": 86901749.0,
      "step": 44430
    },
    {
      "entropy": 5.967228603363037,
      "epoch": 4.441745389113811,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031813636700807265,
      "loss": 5.3318,
      "mean_token_accuracy": 0.2109321191906929,
      "num_tokens": 86911838.0,
      "step": 44435
    },
    {
      "entropy": 5.973479843139648,
      "epoch": 4.442245214175038,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003181013393891695,
      "loss": 5.4182,
      "mean_token_accuracy": 0.2067021533846855,
      "num_tokens": 86921199.0,
      "step": 44440
    },
    {
      "entropy": 5.886992359161377,
      "epoch": 4.442745039236267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031806631068584764,
      "loss": 5.3757,
      "mean_token_accuracy": 0.207111194729805,
      "num_tokens": 86931940.0,
      "step": 44445
    },
    {
      "entropy": 5.931340169906616,
      "epoch": 4.443244864297496,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031803128089898845,
      "loss": 5.4371,
      "mean_token_accuracy": 0.20194060504436492,
      "num_tokens": 86942049.0,
      "step": 44450
    },
    {
      "entropy": 5.961050271987915,
      "epoch": 4.4437446893587245,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031799625002947333,
      "loss": 5.3881,
      "mean_token_accuracy": 0.20635571479797363,
      "num_tokens": 86952654.0,
      "step": 44455
    },
    {
      "entropy": 5.959321165084839,
      "epoch": 4.444244514419953,
      "grad_norm": 1.125,
      "learning_rate": 0.00031796121807818367,
      "loss": 5.3226,
      "mean_token_accuracy": 0.21194271892309188,
      "num_tokens": 86963549.0,
      "step": 44460
    },
    {
      "entropy": 5.9979472160339355,
      "epoch": 4.444744339481182,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003179261850460008,
      "loss": 5.4406,
      "mean_token_accuracy": 0.1959592193365097,
      "num_tokens": 86973464.0,
      "step": 44465
    },
    {
      "entropy": 5.904558944702148,
      "epoch": 4.445244164542411,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031789115093380614,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20779130458831788,
      "num_tokens": 86982983.0,
      "step": 44470
    },
    {
      "entropy": 5.895366144180298,
      "epoch": 4.445743989603638,
      "grad_norm": 1.109375,
      "learning_rate": 0.00031785611574248123,
      "loss": 5.3423,
      "mean_token_accuracy": 0.2092152163386345,
      "num_tokens": 86993688.0,
      "step": 44475
    },
    {
      "entropy": 6.010613536834716,
      "epoch": 4.446243814664867,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031782107947290745,
      "loss": 5.3716,
      "mean_token_accuracy": 0.2160276621580124,
      "num_tokens": 87003157.0,
      "step": 44480
    },
    {
      "entropy": 5.909497880935669,
      "epoch": 4.446743639726096,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0003177860421259663,
      "loss": 5.3262,
      "mean_token_accuracy": 0.20700348168611526,
      "num_tokens": 87012826.0,
      "step": 44485
    },
    {
      "entropy": 5.950806951522827,
      "epoch": 4.4472434647873245,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003177510037025394,
      "loss": 5.3833,
      "mean_token_accuracy": 0.19863465428352356,
      "num_tokens": 87022682.0,
      "step": 44490
    },
    {
      "entropy": 5.949761199951172,
      "epoch": 4.447743289848553,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031771596420350815,
      "loss": 5.3622,
      "mean_token_accuracy": 0.20934540927410125,
      "num_tokens": 87031180.0,
      "step": 44495
    },
    {
      "entropy": 6.006914472579956,
      "epoch": 4.448243114909782,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003176809236297542,
      "loss": 5.4218,
      "mean_token_accuracy": 0.20047540217638016,
      "num_tokens": 87041791.0,
      "step": 44500
    },
    {
      "entropy": 5.924683046340943,
      "epoch": 4.448742939971011,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003176458819821593,
      "loss": 5.2894,
      "mean_token_accuracy": 0.21223543137311934,
      "num_tokens": 87051390.0,
      "step": 44505
    },
    {
      "entropy": 5.8588768482208256,
      "epoch": 4.449242765032238,
      "grad_norm": 1.140625,
      "learning_rate": 0.000317610839261605,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21637419760227203,
      "num_tokens": 87060581.0,
      "step": 44510
    },
    {
      "entropy": 5.948165464401245,
      "epoch": 4.449742590093467,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00031757579546897287,
      "loss": 5.4014,
      "mean_token_accuracy": 0.21209798604249955,
      "num_tokens": 87069551.0,
      "step": 44515
    },
    {
      "entropy": 5.99910535812378,
      "epoch": 4.450242415154696,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031754075060514477,
      "loss": 5.372,
      "mean_token_accuracy": 0.20878180861473083,
      "num_tokens": 87079294.0,
      "step": 44520
    },
    {
      "entropy": 5.858710289001465,
      "epoch": 4.4507422402159245,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031750570467100226,
      "loss": 5.2898,
      "mean_token_accuracy": 0.20961983948946,
      "num_tokens": 87088415.0,
      "step": 44525
    },
    {
      "entropy": 5.978798866271973,
      "epoch": 4.451242065277153,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003174706576674271,
      "loss": 5.4856,
      "mean_token_accuracy": 0.1983731970191002,
      "num_tokens": 87098816.0,
      "step": 44530
    },
    {
      "entropy": 5.980418014526367,
      "epoch": 4.451741890338382,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003174356095953012,
      "loss": 5.3676,
      "mean_token_accuracy": 0.20132853984832763,
      "num_tokens": 87108540.0,
      "step": 44535
    },
    {
      "entropy": 5.960381746292114,
      "epoch": 4.4522417153996106,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003174005604555063,
      "loss": 5.3665,
      "mean_token_accuracy": 0.20961882919073105,
      "num_tokens": 87118823.0,
      "step": 44540
    },
    {
      "entropy": 5.919907331466675,
      "epoch": 4.452741540460838,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031736551024892423,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20953748971223832,
      "num_tokens": 87127608.0,
      "step": 44545
    },
    {
      "entropy": 5.941288232803345,
      "epoch": 4.453241365522067,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003173304589764368,
      "loss": 5.3199,
      "mean_token_accuracy": 0.21512936651706696,
      "num_tokens": 87137841.0,
      "step": 44550
    },
    {
      "entropy": 5.999456453323364,
      "epoch": 4.453741190583296,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031729540663892586,
      "loss": 5.4513,
      "mean_token_accuracy": 0.19803453534841536,
      "num_tokens": 87147133.0,
      "step": 44555
    },
    {
      "entropy": 5.858573627471924,
      "epoch": 4.4542410156445245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031726035323727334,
      "loss": 5.3132,
      "mean_token_accuracy": 0.21413826793432236,
      "num_tokens": 87157947.0,
      "step": 44560
    },
    {
      "entropy": 5.919586849212647,
      "epoch": 4.454740840705753,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031722529877236126,
      "loss": 5.359,
      "mean_token_accuracy": 0.2155407413840294,
      "num_tokens": 87168202.0,
      "step": 44565
    },
    {
      "entropy": 5.9661472797393795,
      "epoch": 4.455240665766982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003171902432450716,
      "loss": 5.3347,
      "mean_token_accuracy": 0.21202169358730316,
      "num_tokens": 87177674.0,
      "step": 44570
    },
    {
      "entropy": 5.987973546981811,
      "epoch": 4.4557404908282106,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031715518665628625,
      "loss": 5.3814,
      "mean_token_accuracy": 0.20163249522447585,
      "num_tokens": 87187491.0,
      "step": 44575
    },
    {
      "entropy": 6.001182937622071,
      "epoch": 4.456240315889438,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031712012900688726,
      "loss": 5.4176,
      "mean_token_accuracy": 0.19801740050315858,
      "num_tokens": 87197711.0,
      "step": 44580
    },
    {
      "entropy": 5.927639961242676,
      "epoch": 4.456740140950667,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003170850702977566,
      "loss": 5.333,
      "mean_token_accuracy": 0.20098597109317778,
      "num_tokens": 87207103.0,
      "step": 44585
    },
    {
      "entropy": 5.979972648620605,
      "epoch": 4.457239966011896,
      "grad_norm": 1.296875,
      "learning_rate": 0.00031705001052977647,
      "loss": 5.3823,
      "mean_token_accuracy": 0.2031203478574753,
      "num_tokens": 87215749.0,
      "step": 44590
    },
    {
      "entropy": 5.915714311599731,
      "epoch": 4.4577397910731245,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003170149497038288,
      "loss": 5.3185,
      "mean_token_accuracy": 0.2154226243495941,
      "num_tokens": 87225569.0,
      "step": 44595
    },
    {
      "entropy": 5.962472724914551,
      "epoch": 4.458239616134353,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031697988782079583,
      "loss": 5.2859,
      "mean_token_accuracy": 0.21341866254806519,
      "num_tokens": 87235353.0,
      "step": 44600
    },
    {
      "entropy": 5.953993034362793,
      "epoch": 4.458739441195582,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031694482488155975,
      "loss": 5.3566,
      "mean_token_accuracy": 0.21357100903987886,
      "num_tokens": 87246115.0,
      "step": 44605
    },
    {
      "entropy": 5.928322792053223,
      "epoch": 4.4592392662568106,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003169097608870026,
      "loss": 5.3164,
      "mean_token_accuracy": 0.21005581319332123,
      "num_tokens": 87256289.0,
      "step": 44610
    },
    {
      "entropy": 5.979756307601929,
      "epoch": 4.459739091318038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003168746958380067,
      "loss": 5.4664,
      "mean_token_accuracy": 0.1885116621851921,
      "num_tokens": 87265113.0,
      "step": 44615
    },
    {
      "entropy": 5.922792196273804,
      "epoch": 4.460238916379267,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031683962973545423,
      "loss": 5.3796,
      "mean_token_accuracy": 0.2066232055425644,
      "num_tokens": 87274682.0,
      "step": 44620
    },
    {
      "entropy": 6.041534376144409,
      "epoch": 4.460738741440496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003168045625802273,
      "loss": 5.4264,
      "mean_token_accuracy": 0.2033565804362297,
      "num_tokens": 87284750.0,
      "step": 44625
    },
    {
      "entropy": 5.702398061752319,
      "epoch": 4.4612385665017245,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003167694943732085,
      "loss": 5.0787,
      "mean_token_accuracy": 0.24264516234397887,
      "num_tokens": 87293815.0,
      "step": 44630
    },
    {
      "entropy": 5.899704790115356,
      "epoch": 4.461738391562953,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031673442511527994,
      "loss": 5.3012,
      "mean_token_accuracy": 0.2164505675435066,
      "num_tokens": 87303552.0,
      "step": 44635
    },
    {
      "entropy": 5.959008979797363,
      "epoch": 4.462238216624182,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031669935480732397,
      "loss": 5.3642,
      "mean_token_accuracy": 0.20287504494190217,
      "num_tokens": 87312818.0,
      "step": 44640
    },
    {
      "entropy": 5.891730880737304,
      "epoch": 4.4627380416854106,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031666428345022296,
      "loss": 5.398,
      "mean_token_accuracy": 0.2086274430155754,
      "num_tokens": 87323401.0,
      "step": 44645
    },
    {
      "entropy": 5.910454559326172,
      "epoch": 4.463237866746638,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003166292110448593,
      "loss": 5.347,
      "mean_token_accuracy": 0.2077852725982666,
      "num_tokens": 87332614.0,
      "step": 44650
    },
    {
      "entropy": 5.916568088531494,
      "epoch": 4.463737691807867,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003165941375921154,
      "loss": 5.3003,
      "mean_token_accuracy": 0.22076791673898696,
      "num_tokens": 87342113.0,
      "step": 44655
    },
    {
      "entropy": 6.024185609817505,
      "epoch": 4.464237516869096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003165590630928737,
      "loss": 5.4338,
      "mean_token_accuracy": 0.20323160886764527,
      "num_tokens": 87352955.0,
      "step": 44660
    },
    {
      "entropy": 5.952051401138306,
      "epoch": 4.4647373419303245,
      "grad_norm": 1.25,
      "learning_rate": 0.00031652398754801665,
      "loss": 5.3923,
      "mean_token_accuracy": 0.20540271401405336,
      "num_tokens": 87362669.0,
      "step": 44665
    },
    {
      "entropy": 5.843736839294434,
      "epoch": 4.465237166991553,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003164889109584268,
      "loss": 5.2482,
      "mean_token_accuracy": 0.2169782191514969,
      "num_tokens": 87371709.0,
      "step": 44670
    },
    {
      "entropy": 5.942991828918457,
      "epoch": 4.465736992052782,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003164538333249866,
      "loss": 5.3777,
      "mean_token_accuracy": 0.20540163069963455,
      "num_tokens": 87380870.0,
      "step": 44675
    },
    {
      "entropy": 5.9224145889282225,
      "epoch": 4.4662368171140105,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003164187546485786,
      "loss": 5.3551,
      "mean_token_accuracy": 0.2115221217274666,
      "num_tokens": 87390524.0,
      "step": 44680
    },
    {
      "entropy": 5.950509023666382,
      "epoch": 4.466736642175238,
      "grad_norm": 1.234375,
      "learning_rate": 0.00031638367493008545,
      "loss": 5.4027,
      "mean_token_accuracy": 0.20494784712791442,
      "num_tokens": 87401681.0,
      "step": 44685
    },
    {
      "entropy": 5.948384189605713,
      "epoch": 4.467236467236467,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031634859417038963,
      "loss": 5.3779,
      "mean_token_accuracy": 0.2069287747144699,
      "num_tokens": 87411577.0,
      "step": 44690
    },
    {
      "entropy": 5.958170747756958,
      "epoch": 4.467736292297696,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031631351237037386,
      "loss": 5.3457,
      "mean_token_accuracy": 0.21147508323192596,
      "num_tokens": 87422798.0,
      "step": 44695
    },
    {
      "entropy": 6.036318302154541,
      "epoch": 4.4682361173589245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031627842953092073,
      "loss": 5.4674,
      "mean_token_accuracy": 0.19661998003721237,
      "num_tokens": 87432610.0,
      "step": 44700
    },
    {
      "entropy": 5.9121533870697025,
      "epoch": 4.468735942420153,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031624334565291294,
      "loss": 5.2973,
      "mean_token_accuracy": 0.21210936456918716,
      "num_tokens": 87442569.0,
      "step": 44705
    },
    {
      "entropy": 5.948671579360962,
      "epoch": 4.469235767481382,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003162082607372332,
      "loss": 5.3492,
      "mean_token_accuracy": 0.21602843850851058,
      "num_tokens": 87451524.0,
      "step": 44710
    },
    {
      "entropy": 5.931614589691162,
      "epoch": 4.4697355925426105,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003161731747847642,
      "loss": 5.3372,
      "mean_token_accuracy": 0.2041474163532257,
      "num_tokens": 87461244.0,
      "step": 44715
    },
    {
      "entropy": 5.863542985916138,
      "epoch": 4.470235417603838,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031613808779638874,
      "loss": 5.2042,
      "mean_token_accuracy": 0.2225293532013893,
      "num_tokens": 87471327.0,
      "step": 44720
    },
    {
      "entropy": 5.957743215560913,
      "epoch": 4.470735242665067,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003161029997729895,
      "loss": 5.4051,
      "mean_token_accuracy": 0.20712730437517166,
      "num_tokens": 87481038.0,
      "step": 44725
    },
    {
      "entropy": 5.913999891281128,
      "epoch": 4.471235067726296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031606791071544946,
      "loss": 5.312,
      "mean_token_accuracy": 0.20761414915323256,
      "num_tokens": 87491252.0,
      "step": 44730
    },
    {
      "entropy": 5.993042659759522,
      "epoch": 4.4717348927875245,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003160328206246514,
      "loss": 5.452,
      "mean_token_accuracy": 0.19431764334440232,
      "num_tokens": 87500073.0,
      "step": 44735
    },
    {
      "entropy": 5.9409260749816895,
      "epoch": 4.472234717848753,
      "grad_norm": 1.171875,
      "learning_rate": 0.000315997729501478,
      "loss": 5.2933,
      "mean_token_accuracy": 0.2051396057009697,
      "num_tokens": 87508939.0,
      "step": 44740
    },
    {
      "entropy": 6.018612813949585,
      "epoch": 4.472734542909982,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003159626373468123,
      "loss": 5.4702,
      "mean_token_accuracy": 0.1986849680542946,
      "num_tokens": 87518484.0,
      "step": 44745
    },
    {
      "entropy": 5.931123113632202,
      "epoch": 4.4732343679712105,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031592754416153725,
      "loss": 5.3048,
      "mean_token_accuracy": 0.21685673147439957,
      "num_tokens": 87530027.0,
      "step": 44750
    },
    {
      "entropy": 6.015460872650147,
      "epoch": 4.473734193032438,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031589244994653567,
      "loss": 5.447,
      "mean_token_accuracy": 0.20102680027484893,
      "num_tokens": 87540805.0,
      "step": 44755
    },
    {
      "entropy": 5.918259477615356,
      "epoch": 4.474234018093667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031585735470269057,
      "loss": 5.3828,
      "mean_token_accuracy": 0.21546898484230043,
      "num_tokens": 87551421.0,
      "step": 44760
    },
    {
      "entropy": 5.92141695022583,
      "epoch": 4.474733843154896,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000315822258430885,
      "loss": 5.3591,
      "mean_token_accuracy": 0.20326176285743713,
      "num_tokens": 87561517.0,
      "step": 44765
    },
    {
      "entropy": 5.9532054424285885,
      "epoch": 4.4752336682161244,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003157871611320018,
      "loss": 5.3527,
      "mean_token_accuracy": 0.21042413562536239,
      "num_tokens": 87570772.0,
      "step": 44770
    },
    {
      "entropy": 5.977375268936157,
      "epoch": 4.475733493277353,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003157520628069242,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20413402915000917,
      "num_tokens": 87580345.0,
      "step": 44775
    },
    {
      "entropy": 5.9258713722229,
      "epoch": 4.476233318338582,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031571696345653516,
      "loss": 5.3505,
      "mean_token_accuracy": 0.20675129145383836,
      "num_tokens": 87589336.0,
      "step": 44780
    },
    {
      "entropy": 5.939808940887451,
      "epoch": 4.47673314339981,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031568186308171775,
      "loss": 5.3268,
      "mean_token_accuracy": 0.2099213942885399,
      "num_tokens": 87599160.0,
      "step": 44785
    },
    {
      "entropy": 5.982759141921997,
      "epoch": 4.477232968461038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031564676168335527,
      "loss": 5.3703,
      "mean_token_accuracy": 0.20767330080270768,
      "num_tokens": 87608530.0,
      "step": 44790
    },
    {
      "entropy": 5.924206304550171,
      "epoch": 4.477732793522267,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031561165926233064,
      "loss": 5.3186,
      "mean_token_accuracy": 0.21440740525722504,
      "num_tokens": 87618362.0,
      "step": 44795
    },
    {
      "entropy": 5.995800971984863,
      "epoch": 4.478232618583496,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031557655581952715,
      "loss": 5.5025,
      "mean_token_accuracy": 0.19340969771146774,
      "num_tokens": 87629043.0,
      "step": 44800
    },
    {
      "entropy": 5.925760316848755,
      "epoch": 4.4787324436447244,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00031554145135582784,
      "loss": 5.3665,
      "mean_token_accuracy": 0.20687450617551803,
      "num_tokens": 87638877.0,
      "step": 44805
    },
    {
      "entropy": 5.881593704223633,
      "epoch": 4.479232268705953,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031550634587211617,
      "loss": 5.283,
      "mean_token_accuracy": 0.2139514684677124,
      "num_tokens": 87648187.0,
      "step": 44810
    },
    {
      "entropy": 5.8795520782470705,
      "epoch": 4.479732093767182,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031547123936927516,
      "loss": 5.21,
      "mean_token_accuracy": 0.21894301772117614,
      "num_tokens": 87658209.0,
      "step": 44815
    },
    {
      "entropy": 5.978931856155396,
      "epoch": 4.48023191882841,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031543613184818834,
      "loss": 5.41,
      "mean_token_accuracy": 0.20846029967069626,
      "num_tokens": 87668328.0,
      "step": 44820
    },
    {
      "entropy": 5.960484313964844,
      "epoch": 4.480731743889638,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003154010233097387,
      "loss": 5.3713,
      "mean_token_accuracy": 0.2002704843878746,
      "num_tokens": 87677758.0,
      "step": 44825
    },
    {
      "entropy": 5.866744899749756,
      "epoch": 4.481231568950867,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031536591375480984,
      "loss": 5.3389,
      "mean_token_accuracy": 0.2099955350160599,
      "num_tokens": 87688130.0,
      "step": 44830
    },
    {
      "entropy": 6.005120038986206,
      "epoch": 4.481731394012096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003153308031842849,
      "loss": 5.4873,
      "mean_token_accuracy": 0.20011866092681885,
      "num_tokens": 87697491.0,
      "step": 44835
    },
    {
      "entropy": 5.922216844558716,
      "epoch": 4.4822312190733244,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031529569159904723,
      "loss": 5.3316,
      "mean_token_accuracy": 0.21543993055820465,
      "num_tokens": 87706412.0,
      "step": 44840
    },
    {
      "entropy": 6.064517259597778,
      "epoch": 4.482731044134553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031526057899998043,
      "loss": 5.5005,
      "mean_token_accuracy": 0.19435991644859313,
      "num_tokens": 87717469.0,
      "step": 44845
    },
    {
      "entropy": 5.986492395401001,
      "epoch": 4.483230869195782,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003152254653879679,
      "loss": 5.354,
      "mean_token_accuracy": 0.20503717064857482,
      "num_tokens": 87726876.0,
      "step": 44850
    },
    {
      "entropy": 5.830844068527222,
      "epoch": 4.48373069425701,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031519035076389284,
      "loss": 5.2925,
      "mean_token_accuracy": 0.21803344339132308,
      "num_tokens": 87736475.0,
      "step": 44855
    },
    {
      "entropy": 5.863243103027344,
      "epoch": 4.484230519318238,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000315155235128639,
      "loss": 5.2861,
      "mean_token_accuracy": 0.20545416623353957,
      "num_tokens": 87745332.0,
      "step": 44860
    },
    {
      "entropy": 5.955864953994751,
      "epoch": 4.484730344379467,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003151201184830899,
      "loss": 5.3802,
      "mean_token_accuracy": 0.2025051772594452,
      "num_tokens": 87755793.0,
      "step": 44865
    },
    {
      "entropy": 5.897295951843262,
      "epoch": 4.485230169440696,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00031508500082812874,
      "loss": 5.2581,
      "mean_token_accuracy": 0.22282962501049042,
      "num_tokens": 87764847.0,
      "step": 44870
    },
    {
      "entropy": 5.9624107837677,
      "epoch": 4.4857299945019244,
      "grad_norm": 1.125,
      "learning_rate": 0.00031504988216463927,
      "loss": 5.3724,
      "mean_token_accuracy": 0.20980723947286606,
      "num_tokens": 87775000.0,
      "step": 44875
    },
    {
      "entropy": 5.953543424606323,
      "epoch": 4.486229819563153,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003150147624935052,
      "loss": 5.3495,
      "mean_token_accuracy": 0.21813207268714904,
      "num_tokens": 87784580.0,
      "step": 44880
    },
    {
      "entropy": 5.955189990997314,
      "epoch": 4.486729644624382,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003149796418156099,
      "loss": 5.3629,
      "mean_token_accuracy": 0.2114956185221672,
      "num_tokens": 87793830.0,
      "step": 44885
    },
    {
      "entropy": 5.903805780410766,
      "epoch": 4.48722946968561,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003149445201318372,
      "loss": 5.3651,
      "mean_token_accuracy": 0.2111235335469246,
      "num_tokens": 87803824.0,
      "step": 44890
    },
    {
      "entropy": 5.921395587921142,
      "epoch": 4.487729294746838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031490939744307063,
      "loss": 5.3399,
      "mean_token_accuracy": 0.21135840713977813,
      "num_tokens": 87812833.0,
      "step": 44895
    },
    {
      "entropy": 5.836358594894409,
      "epoch": 4.488229119808067,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003148742737501938,
      "loss": 5.1771,
      "mean_token_accuracy": 0.22638890743255616,
      "num_tokens": 87822638.0,
      "step": 44900
    },
    {
      "entropy": 5.808003234863281,
      "epoch": 4.488728944869296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031483914905409055,
      "loss": 5.2255,
      "mean_token_accuracy": 0.22276883721351623,
      "num_tokens": 87832508.0,
      "step": 44905
    },
    {
      "entropy": 5.928801822662353,
      "epoch": 4.489228769930524,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003148040233556446,
      "loss": 5.3662,
      "mean_token_accuracy": 0.20793063044548035,
      "num_tokens": 87841953.0,
      "step": 44910
    },
    {
      "entropy": 5.9834167003631595,
      "epoch": 4.489728594991753,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003147688966557396,
      "loss": 5.4406,
      "mean_token_accuracy": 0.20163181573152542,
      "num_tokens": 87852287.0,
      "step": 44915
    },
    {
      "entropy": 6.00482816696167,
      "epoch": 4.490228420052982,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003147337689552594,
      "loss": 5.4815,
      "mean_token_accuracy": 0.19825955331325532,
      "num_tokens": 87862060.0,
      "step": 44920
    },
    {
      "entropy": 5.846614360809326,
      "epoch": 4.49072824511421,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003146986402550878,
      "loss": 5.2257,
      "mean_token_accuracy": 0.22544414401054383,
      "num_tokens": 87871324.0,
      "step": 44925
    },
    {
      "entropy": 5.915621948242188,
      "epoch": 4.491228070175438,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031466351055610875,
      "loss": 5.354,
      "mean_token_accuracy": 0.20880573838949204,
      "num_tokens": 87880691.0,
      "step": 44930
    },
    {
      "entropy": 5.9486898422241214,
      "epoch": 4.491727895236667,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031462837985920584,
      "loss": 5.3368,
      "mean_token_accuracy": 0.20795900076627732,
      "num_tokens": 87891572.0,
      "step": 44935
    },
    {
      "entropy": 5.980926036834717,
      "epoch": 4.492227720297896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031459324816526313,
      "loss": 5.4169,
      "mean_token_accuracy": 0.20479271411895753,
      "num_tokens": 87902320.0,
      "step": 44940
    },
    {
      "entropy": 5.986262655258178,
      "epoch": 4.492727545359124,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031455811547516447,
      "loss": 5.4553,
      "mean_token_accuracy": 0.20300431549549103,
      "num_tokens": 87912740.0,
      "step": 44945
    },
    {
      "entropy": 6.005709648132324,
      "epoch": 4.493227370420353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031452298178979385,
      "loss": 5.2895,
      "mean_token_accuracy": 0.21439919471740723,
      "num_tokens": 87922212.0,
      "step": 44950
    },
    {
      "entropy": 5.856326580047607,
      "epoch": 4.493727195481582,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003144878471100351,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21447834819555284,
      "num_tokens": 87931103.0,
      "step": 44955
    },
    {
      "entropy": 5.859505367279053,
      "epoch": 4.49422702054281,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031445271143677235,
      "loss": 5.3359,
      "mean_token_accuracy": 0.20787180960178375,
      "num_tokens": 87941345.0,
      "step": 44960
    },
    {
      "entropy": 5.991488265991211,
      "epoch": 4.494726845604038,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031441757477088957,
      "loss": 5.3815,
      "mean_token_accuracy": 0.20426299721002578,
      "num_tokens": 87951892.0,
      "step": 44965
    },
    {
      "entropy": 6.040091037750244,
      "epoch": 4.495226670665267,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031438243711327064,
      "loss": 5.3886,
      "mean_token_accuracy": 0.1959208741784096,
      "num_tokens": 87961742.0,
      "step": 44970
    },
    {
      "entropy": 5.887376499176026,
      "epoch": 4.495726495726496,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003143472984647998,
      "loss": 5.3502,
      "mean_token_accuracy": 0.21276991218328475,
      "num_tokens": 87971559.0,
      "step": 44975
    },
    {
      "entropy": 6.013719749450684,
      "epoch": 4.496226320787724,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000314312158826361,
      "loss": 5.4289,
      "mean_token_accuracy": 0.2018509179353714,
      "num_tokens": 87982442.0,
      "step": 44980
    },
    {
      "entropy": 5.927216958999634,
      "epoch": 4.496726145848953,
      "grad_norm": 1.28125,
      "learning_rate": 0.00031427701819883845,
      "loss": 5.2814,
      "mean_token_accuracy": 0.2108465775847435,
      "num_tokens": 87992416.0,
      "step": 44985
    },
    {
      "entropy": 5.957387208938599,
      "epoch": 4.497225970910182,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003142418765831162,
      "loss": 5.361,
      "mean_token_accuracy": 0.2047596275806427,
      "num_tokens": 88000676.0,
      "step": 44990
    },
    {
      "entropy": 5.939512252807617,
      "epoch": 4.49772579597141,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031420673398007846,
      "loss": 5.3916,
      "mean_token_accuracy": 0.20352351665496826,
      "num_tokens": 88010693.0,
      "step": 44995
    },
    {
      "entropy": 5.951118040084839,
      "epoch": 4.498225621032638,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031417159039060936,
      "loss": 5.3436,
      "mean_token_accuracy": 0.2105160892009735,
      "num_tokens": 88020191.0,
      "step": 45000
    },
    {
      "epoch": 4.498225621032638,
      "eval_entropy": 5.714778974725241,
      "eval_loss": 5.488633155822754,
      "eval_mean_token_accuracy": 0.21013609411401774,
      "eval_num_tokens": 88020191.0,
      "eval_runtime": 24.1384,
      "eval_samples_per_second": 1286.126,
      "eval_steps_per_second": 160.781,
      "step": 45000
    },
    {
      "entropy": 5.930775308609009,
      "epoch": 4.498725446093867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003141364458155931,
      "loss": 5.27,
      "mean_token_accuracy": 0.21321622133255005,
      "num_tokens": 88030460.0,
      "step": 45005
    },
    {
      "entropy": 5.902327251434326,
      "epoch": 4.499225271155096,
      "grad_norm": 1.203125,
      "learning_rate": 0.000314101300255914,
      "loss": 5.3556,
      "mean_token_accuracy": 0.2167203262448311,
      "num_tokens": 88042128.0,
      "step": 45010
    },
    {
      "entropy": 5.962719535827636,
      "epoch": 4.499725096216324,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003140661537124562,
      "loss": 5.3189,
      "mean_token_accuracy": 0.20621041357517242,
      "num_tokens": 88050720.0,
      "step": 45015
    },
    {
      "entropy": 5.955811977386475,
      "epoch": 4.500224921277553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000314031006186104,
      "loss": 5.3891,
      "mean_token_accuracy": 0.20710988491773605,
      "num_tokens": 88060347.0,
      "step": 45020
    },
    {
      "entropy": 5.804159259796142,
      "epoch": 4.500724746338781,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003139958576777418,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21428649574518205,
      "num_tokens": 88069942.0,
      "step": 45025
    },
    {
      "entropy": 5.878178453445434,
      "epoch": 4.50122457140001,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003139607081882538,
      "loss": 5.2855,
      "mean_token_accuracy": 0.21335354298353196,
      "num_tokens": 88080309.0,
      "step": 45030
    },
    {
      "entropy": 6.025648498535157,
      "epoch": 4.501724396461238,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031392555771852447,
      "loss": 5.397,
      "mean_token_accuracy": 0.20537475049495696,
      "num_tokens": 88089519.0,
      "step": 45035
    },
    {
      "entropy": 5.961237907409668,
      "epoch": 4.502224221522467,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003138904062694381,
      "loss": 5.3525,
      "mean_token_accuracy": 0.2084752306342125,
      "num_tokens": 88099552.0,
      "step": 45040
    },
    {
      "entropy": 6.010258293151855,
      "epoch": 4.502724046583696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003138552538418791,
      "loss": 5.4401,
      "mean_token_accuracy": 0.1994522914290428,
      "num_tokens": 88108897.0,
      "step": 45045
    },
    {
      "entropy": 5.875880908966065,
      "epoch": 4.503223871644924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031382010043673193,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20844531953334808,
      "num_tokens": 88118384.0,
      "step": 45050
    },
    {
      "entropy": 5.963056850433349,
      "epoch": 4.503723696706153,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003137849460548809,
      "loss": 5.4033,
      "mean_token_accuracy": 0.20876942425966263,
      "num_tokens": 88129301.0,
      "step": 45055
    },
    {
      "entropy": 5.994611215591431,
      "epoch": 4.504223521767381,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031374979069721077,
      "loss": 5.3649,
      "mean_token_accuracy": 0.20874007493257524,
      "num_tokens": 88138394.0,
      "step": 45060
    },
    {
      "entropy": 6.034371328353882,
      "epoch": 4.50472334682861,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031371463436460586,
      "loss": 5.4203,
      "mean_token_accuracy": 0.20455629229545594,
      "num_tokens": 88148474.0,
      "step": 45065
    },
    {
      "entropy": 5.812115859985352,
      "epoch": 4.505223171889838,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031367947705795063,
      "loss": 5.2494,
      "mean_token_accuracy": 0.21764795929193498,
      "num_tokens": 88158284.0,
      "step": 45070
    },
    {
      "entropy": 6.0131782531738285,
      "epoch": 4.505722996951067,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003136443187781298,
      "loss": 5.4281,
      "mean_token_accuracy": 0.20624678432941437,
      "num_tokens": 88167846.0,
      "step": 45075
    },
    {
      "entropy": 5.996267557144165,
      "epoch": 4.506222822012296,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031360915952602776,
      "loss": 5.416,
      "mean_token_accuracy": 0.20237179100513458,
      "num_tokens": 88177804.0,
      "step": 45080
    },
    {
      "entropy": 6.003507852554321,
      "epoch": 4.506722647073524,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003135739993025292,
      "loss": 5.4184,
      "mean_token_accuracy": 0.19779224395751954,
      "num_tokens": 88187588.0,
      "step": 45085
    },
    {
      "entropy": 5.95502462387085,
      "epoch": 4.507222472134753,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003135388381085188,
      "loss": 5.3554,
      "mean_token_accuracy": 0.21214455515146255,
      "num_tokens": 88196589.0,
      "step": 45090
    },
    {
      "entropy": 5.926339721679687,
      "epoch": 4.507722297195981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031350367594488115,
      "loss": 5.2764,
      "mean_token_accuracy": 0.20662112087011336,
      "num_tokens": 88206437.0,
      "step": 45095
    },
    {
      "entropy": 5.988366031646729,
      "epoch": 4.50822212225721,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031346851281250087,
      "loss": 5.2834,
      "mean_token_accuracy": 0.20994166135787964,
      "num_tokens": 88216217.0,
      "step": 45100
    },
    {
      "entropy": 5.908532428741455,
      "epoch": 4.508721947318438,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003134333487122628,
      "loss": 5.2747,
      "mean_token_accuracy": 0.21716489791870117,
      "num_tokens": 88226570.0,
      "step": 45105
    },
    {
      "entropy": 5.851913213729858,
      "epoch": 4.509221772379667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031339818364505153,
      "loss": 5.2415,
      "mean_token_accuracy": 0.21872662305831908,
      "num_tokens": 88236179.0,
      "step": 45110
    },
    {
      "entropy": 5.92716121673584,
      "epoch": 4.509721597440896,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003133630176117518,
      "loss": 5.4194,
      "mean_token_accuracy": 0.20458938777446747,
      "num_tokens": 88245341.0,
      "step": 45115
    },
    {
      "entropy": 6.026682996749878,
      "epoch": 4.510221422502124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003133278506132484,
      "loss": 5.4728,
      "mean_token_accuracy": 0.1958287015557289,
      "num_tokens": 88254516.0,
      "step": 45120
    },
    {
      "entropy": 5.973591756820679,
      "epoch": 4.510721247563353,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003132926826504262,
      "loss": 5.377,
      "mean_token_accuracy": 0.20390546470880508,
      "num_tokens": 88264084.0,
      "step": 45125
    },
    {
      "entropy": 5.945250368118286,
      "epoch": 4.511221072624581,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00031325751372416987,
      "loss": 5.2674,
      "mean_token_accuracy": 0.21995058208703994,
      "num_tokens": 88273797.0,
      "step": 45130
    },
    {
      "entropy": 5.824966621398926,
      "epoch": 4.51172089768581,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003132223438353644,
      "loss": 5.1981,
      "mean_token_accuracy": 0.22939865291118622,
      "num_tokens": 88284165.0,
      "step": 45135
    },
    {
      "entropy": 5.86846022605896,
      "epoch": 4.512220722747038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003131871729848947,
      "loss": 5.3235,
      "mean_token_accuracy": 0.2077735111117363,
      "num_tokens": 88294004.0,
      "step": 45140
    },
    {
      "entropy": 5.959335088729858,
      "epoch": 4.512720547808267,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031315200117364546,
      "loss": 5.3398,
      "mean_token_accuracy": 0.21006741672754287,
      "num_tokens": 88303444.0,
      "step": 45145
    },
    {
      "entropy": 5.906006765365601,
      "epoch": 4.513220372869496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003131168284025017,
      "loss": 5.2903,
      "mean_token_accuracy": 0.21416866183280944,
      "num_tokens": 88312339.0,
      "step": 45150
    },
    {
      "entropy": 5.9582914352417,
      "epoch": 4.513720197930724,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031308165467234834,
      "loss": 5.3175,
      "mean_token_accuracy": 0.21197519749403,
      "num_tokens": 88322399.0,
      "step": 45155
    },
    {
      "entropy": 5.911545896530152,
      "epoch": 4.514220022991953,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031304647998407027,
      "loss": 5.3121,
      "mean_token_accuracy": 0.20947432965040208,
      "num_tokens": 88331467.0,
      "step": 45160
    },
    {
      "entropy": 5.920316410064697,
      "epoch": 4.514719848053181,
      "grad_norm": 1.140625,
      "learning_rate": 0.00031301130433855265,
      "loss": 5.3541,
      "mean_token_accuracy": 0.2052411839365959,
      "num_tokens": 88340857.0,
      "step": 45165
    },
    {
      "entropy": 5.903994512557984,
      "epoch": 4.51521967311441,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003129761277366803,
      "loss": 5.3204,
      "mean_token_accuracy": 0.2147686004638672,
      "num_tokens": 88349536.0,
      "step": 45170
    },
    {
      "entropy": 5.8737226009368895,
      "epoch": 4.515719498175638,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00031294095017933847,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2164631024003029,
      "num_tokens": 88358108.0,
      "step": 45175
    },
    {
      "entropy": 5.942861938476563,
      "epoch": 4.516219323236867,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031290577166741203,
      "loss": 5.3379,
      "mean_token_accuracy": 0.21157876253128052,
      "num_tokens": 88367034.0,
      "step": 45180
    },
    {
      "entropy": 5.930906867980957,
      "epoch": 4.516719148298096,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003128705922017862,
      "loss": 5.3338,
      "mean_token_accuracy": 0.2092461034655571,
      "num_tokens": 88376226.0,
      "step": 45185
    },
    {
      "entropy": 5.927394008636474,
      "epoch": 4.517218973359324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031283541178334594,
      "loss": 5.2954,
      "mean_token_accuracy": 0.20878197252750397,
      "num_tokens": 88385963.0,
      "step": 45190
    },
    {
      "entropy": 5.879012107849121,
      "epoch": 4.517718798420553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003128002304129764,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2122438743710518,
      "num_tokens": 88396651.0,
      "step": 45195
    },
    {
      "entropy": 5.922211933135986,
      "epoch": 4.518218623481781,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031276504809156285,
      "loss": 5.3229,
      "mean_token_accuracy": 0.21259128600358962,
      "num_tokens": 88407810.0,
      "step": 45200
    },
    {
      "entropy": 5.957756376266479,
      "epoch": 4.51871844854301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031272986481999045,
      "loss": 5.3945,
      "mean_token_accuracy": 0.20684174001216887,
      "num_tokens": 88417797.0,
      "step": 45205
    },
    {
      "entropy": 5.9088681697845455,
      "epoch": 4.519218273604238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031269468059914434,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20592878758907318,
      "num_tokens": 88428214.0,
      "step": 45210
    },
    {
      "entropy": 5.891322803497315,
      "epoch": 4.519718098665467,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00031265949542990973,
      "loss": 5.284,
      "mean_token_accuracy": 0.20988856703042985,
      "num_tokens": 88438018.0,
      "step": 45215
    },
    {
      "entropy": 5.991840505599976,
      "epoch": 4.520217923726696,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000312624309313172,
      "loss": 5.4037,
      "mean_token_accuracy": 0.20480893105268477,
      "num_tokens": 88447462.0,
      "step": 45220
    },
    {
      "entropy": 5.899428272247315,
      "epoch": 4.520717748787924,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031258912224981617,
      "loss": 5.298,
      "mean_token_accuracy": 0.21192774921655655,
      "num_tokens": 88458071.0,
      "step": 45225
    },
    {
      "entropy": 5.946587705612183,
      "epoch": 4.521217573849153,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031255393424072783,
      "loss": 5.346,
      "mean_token_accuracy": 0.20409026443958284,
      "num_tokens": 88467383.0,
      "step": 45230
    },
    {
      "entropy": 5.888313055038452,
      "epoch": 4.521717398910381,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003125187452867921,
      "loss": 5.2546,
      "mean_token_accuracy": 0.22466745376586914,
      "num_tokens": 88476884.0,
      "step": 45235
    },
    {
      "entropy": 5.90536322593689,
      "epoch": 4.52221722397161,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003124835553888944,
      "loss": 5.3085,
      "mean_token_accuracy": 0.20962051600217818,
      "num_tokens": 88486627.0,
      "step": 45240
    },
    {
      "entropy": 5.943721914291382,
      "epoch": 4.522717049032838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003124483645479201,
      "loss": 5.418,
      "mean_token_accuracy": 0.20563091188669205,
      "num_tokens": 88498173.0,
      "step": 45245
    },
    {
      "entropy": 5.971826124191284,
      "epoch": 4.523216874094067,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003124131727647547,
      "loss": 5.3506,
      "mean_token_accuracy": 0.2099450185894966,
      "num_tokens": 88508473.0,
      "step": 45250
    },
    {
      "entropy": 6.040633630752564,
      "epoch": 4.523716699155296,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031237798004028327,
      "loss": 5.4284,
      "mean_token_accuracy": 0.20711313039064408,
      "num_tokens": 88518235.0,
      "step": 45255
    },
    {
      "entropy": 5.9658959865570065,
      "epoch": 4.524216524216524,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031234278637539163,
      "loss": 5.3754,
      "mean_token_accuracy": 0.20967860668897628,
      "num_tokens": 88527988.0,
      "step": 45260
    },
    {
      "entropy": 5.9752233028411865,
      "epoch": 4.524716349277753,
      "grad_norm": 1.21875,
      "learning_rate": 0.000312307591770965,
      "loss": 5.4129,
      "mean_token_accuracy": 0.20100825130939484,
      "num_tokens": 88539103.0,
      "step": 45265
    },
    {
      "entropy": 5.986906099319458,
      "epoch": 4.525216174338981,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00031227239622788904,
      "loss": 5.4308,
      "mean_token_accuracy": 0.19503696858882905,
      "num_tokens": 88548642.0,
      "step": 45270
    },
    {
      "entropy": 5.910177230834961,
      "epoch": 4.52571599940021,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003122371997470492,
      "loss": 5.3064,
      "mean_token_accuracy": 0.21064850986003875,
      "num_tokens": 88558327.0,
      "step": 45275
    },
    {
      "entropy": 5.88244514465332,
      "epoch": 4.526215824461438,
      "grad_norm": 1.171875,
      "learning_rate": 0.000312202002329331,
      "loss": 5.2854,
      "mean_token_accuracy": 0.21142912060022354,
      "num_tokens": 88568103.0,
      "step": 45280
    },
    {
      "entropy": 5.822341251373291,
      "epoch": 4.526715649522667,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003121668039756199,
      "loss": 5.2276,
      "mean_token_accuracy": 0.21520173251628877,
      "num_tokens": 88577794.0,
      "step": 45285
    },
    {
      "entropy": 5.8893077850341795,
      "epoch": 4.527215474583896,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031213160468680166,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21118118315935136,
      "num_tokens": 88587948.0,
      "step": 45290
    },
    {
      "entropy": 5.927646493911743,
      "epoch": 4.527715299645124,
      "grad_norm": 1.078125,
      "learning_rate": 0.00031209640446376173,
      "loss": 5.3596,
      "mean_token_accuracy": 0.20807333588600158,
      "num_tokens": 88597619.0,
      "step": 45295
    },
    {
      "entropy": 5.902676105499268,
      "epoch": 4.528215124706353,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00031206120330738593,
      "loss": 5.305,
      "mean_token_accuracy": 0.21425082236528398,
      "num_tokens": 88608404.0,
      "step": 45300
    },
    {
      "entropy": 5.948941659927368,
      "epoch": 4.528714949767581,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003120260012185597,
      "loss": 5.3415,
      "mean_token_accuracy": 0.21029331535100937,
      "num_tokens": 88618442.0,
      "step": 45305
    },
    {
      "entropy": 5.882995510101319,
      "epoch": 4.52921477482881,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003119907981981689,
      "loss": 5.2089,
      "mean_token_accuracy": 0.2177291616797447,
      "num_tokens": 88628040.0,
      "step": 45310
    },
    {
      "entropy": 5.913762426376342,
      "epoch": 4.529714599890038,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00031195559424709895,
      "loss": 5.357,
      "mean_token_accuracy": 0.20657015591859818,
      "num_tokens": 88638653.0,
      "step": 45315
    },
    {
      "entropy": 5.876624202728271,
      "epoch": 4.530214424951267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031192038936623585,
      "loss": 5.3177,
      "mean_token_accuracy": 0.20919030755758286,
      "num_tokens": 88649053.0,
      "step": 45320
    },
    {
      "entropy": 5.934948444366455,
      "epoch": 4.530714250012496,
      "grad_norm": 1.21875,
      "learning_rate": 0.00031188518355646536,
      "loss": 5.3749,
      "mean_token_accuracy": 0.20579117238521577,
      "num_tokens": 88658833.0,
      "step": 45325
    },
    {
      "entropy": 5.8862498760223385,
      "epoch": 4.531214075073724,
      "grad_norm": 1.203125,
      "learning_rate": 0.000311849976818673,
      "loss": 5.3014,
      "mean_token_accuracy": 0.21585540920495988,
      "num_tokens": 88668457.0,
      "step": 45330
    },
    {
      "entropy": 5.8589280128479,
      "epoch": 4.531713900134953,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003118147691537448,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2173326075077057,
      "num_tokens": 88677459.0,
      "step": 45335
    },
    {
      "entropy": 5.981380939483643,
      "epoch": 4.532213725196181,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003117795605625664,
      "loss": 5.4429,
      "mean_token_accuracy": 0.19966061413288116,
      "num_tokens": 88687662.0,
      "step": 45340
    },
    {
      "entropy": 5.990866184234619,
      "epoch": 4.53271355025741,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031174435104602377,
      "loss": 5.3528,
      "mean_token_accuracy": 0.20251227170228958,
      "num_tokens": 88697798.0,
      "step": 45345
    },
    {
      "entropy": 5.97138991355896,
      "epoch": 4.533213375318638,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003117091406050027,
      "loss": 5.3388,
      "mean_token_accuracy": 0.2081328436732292,
      "num_tokens": 88707247.0,
      "step": 45350
    },
    {
      "entropy": 5.929505491256714,
      "epoch": 4.533713200379867,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003116739292403891,
      "loss": 5.3451,
      "mean_token_accuracy": 0.21151744127273558,
      "num_tokens": 88716734.0,
      "step": 45355
    },
    {
      "entropy": 5.940722036361694,
      "epoch": 4.534213025441096,
      "grad_norm": 1.203125,
      "learning_rate": 0.00031163871695306895,
      "loss": 5.3085,
      "mean_token_accuracy": 0.2133166268467903,
      "num_tokens": 88725499.0,
      "step": 45360
    },
    {
      "entropy": 5.937921333312988,
      "epoch": 4.534712850502324,
      "grad_norm": 1.109375,
      "learning_rate": 0.000311603503743928,
      "loss": 5.2876,
      "mean_token_accuracy": 0.2097460925579071,
      "num_tokens": 88735901.0,
      "step": 45365
    },
    {
      "entropy": 5.951512002944947,
      "epoch": 4.535212675563553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00031156828961385235,
      "loss": 5.3354,
      "mean_token_accuracy": 0.21305018663406372,
      "num_tokens": 88745081.0,
      "step": 45370
    },
    {
      "entropy": 5.995963478088379,
      "epoch": 4.535712500624781,
      "grad_norm": 1.328125,
      "learning_rate": 0.000311533074563728,
      "loss": 5.3732,
      "mean_token_accuracy": 0.2026958167552948,
      "num_tokens": 88754124.0,
      "step": 45375
    },
    {
      "entropy": 6.009591007232666,
      "epoch": 4.53621232568601,
      "grad_norm": 0.98046875,
      "learning_rate": 0.00031149785859444087,
      "loss": 5.4487,
      "mean_token_accuracy": 0.20321014225482942,
      "num_tokens": 88765302.0,
      "step": 45380
    },
    {
      "entropy": 5.8611626625061035,
      "epoch": 4.536712150747238,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00031146264170687696,
      "loss": 5.28,
      "mean_token_accuracy": 0.20837893337011337,
      "num_tokens": 88776512.0,
      "step": 45385
    },
    {
      "entropy": 5.895390748977661,
      "epoch": 4.537211975808467,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00031142742390192243,
      "loss": 5.3086,
      "mean_token_accuracy": 0.2124835103750229,
      "num_tokens": 88786225.0,
      "step": 45390
    },
    {
      "entropy": 5.924438381195069,
      "epoch": 4.537711800869696,
      "grad_norm": 1.125,
      "learning_rate": 0.00031139220518046324,
      "loss": 5.2914,
      "mean_token_accuracy": 0.21157679408788682,
      "num_tokens": 88795363.0,
      "step": 45395
    },
    {
      "entropy": 5.852251148223877,
      "epoch": 4.538211625930924,
      "grad_norm": 1.125,
      "learning_rate": 0.00031135698554338555,
      "loss": 5.2648,
      "mean_token_accuracy": 0.21601064056158065,
      "num_tokens": 88805340.0,
      "step": 45400
    },
    {
      "entropy": 5.908453416824341,
      "epoch": 4.538711450992153,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00031132176499157544,
      "loss": 5.3106,
      "mean_token_accuracy": 0.22012666016817092,
      "num_tokens": 88815242.0,
      "step": 45405
    },
    {
      "entropy": 5.936482381820679,
      "epoch": 4.539211276053381,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031128654352591907,
      "loss": 5.3327,
      "mean_token_accuracy": 0.21241173148155212,
      "num_tokens": 88824716.0,
      "step": 45410
    },
    {
      "entropy": 5.829859399795533,
      "epoch": 4.53971110111461,
      "grad_norm": 1.1875,
      "learning_rate": 0.00031125132114730267,
      "loss": 5.2394,
      "mean_token_accuracy": 0.2184965655207634,
      "num_tokens": 88833616.0,
      "step": 45415
    },
    {
      "entropy": 5.8787843704223635,
      "epoch": 4.540210926175838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00031121609785661225,
      "loss": 5.31,
      "mean_token_accuracy": 0.21347470879554747,
      "num_tokens": 88842224.0,
      "step": 45420
    },
    {
      "entropy": 5.97101902961731,
      "epoch": 4.540710751237067,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00031118087365473426,
      "loss": 5.393,
      "mean_token_accuracy": 0.20750393867492675,
      "num_tokens": 88852135.0,
      "step": 45425
    },
    {
      "entropy": 5.972012901306153,
      "epoch": 4.541210576298296,
      "grad_norm": 1.3125,
      "learning_rate": 0.00031114564854255465,
      "loss": 5.418,
      "mean_token_accuracy": 0.19875564873218537,
      "num_tokens": 88861769.0,
      "step": 45430
    },
    {
      "entropy": 5.960900545120239,
      "epoch": 4.541710401359524,
      "grad_norm": 1.171875,
      "learning_rate": 0.00031111042252095997,
      "loss": 5.4107,
      "mean_token_accuracy": 0.2064364358782768,
      "num_tokens": 88871279.0,
      "step": 45435
    },
    {
      "entropy": 5.965633773803711,
      "epoch": 4.542210226420753,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00031107519559083624,
      "loss": 5.313,
      "mean_token_accuracy": 0.21143611371517182,
      "num_tokens": 88880697.0,
      "step": 45440
    },
    {
      "entropy": 6.015780687332153,
      "epoch": 4.542710051481981,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003110399677530699,
      "loss": 5.4089,
      "mean_token_accuracy": 0.20104511380195617,
      "num_tokens": 88892220.0,
      "step": 45445
    },
    {
      "entropy": 6.037422561645508,
      "epoch": 4.54320987654321,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031100473900854726,
      "loss": 5.4634,
      "mean_token_accuracy": 0.198288132250309,
      "num_tokens": 88902281.0,
      "step": 45450
    },
    {
      "entropy": 5.9334033012390135,
      "epoch": 4.543709701604438,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031096950935815464,
      "loss": 5.2986,
      "mean_token_accuracy": 0.21165602058172225,
      "num_tokens": 88912726.0,
      "step": 45455
    },
    {
      "entropy": 5.88907527923584,
      "epoch": 4.544209526665667,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003109342788027784,
      "loss": 5.2498,
      "mean_token_accuracy": 0.2150368332862854,
      "num_tokens": 88922455.0,
      "step": 45460
    },
    {
      "entropy": 5.887277698516845,
      "epoch": 4.544709351726896,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000310899047343305,
      "loss": 5.2366,
      "mean_token_accuracy": 0.2187169075012207,
      "num_tokens": 88932829.0,
      "step": 45465
    },
    {
      "entropy": 5.823785591125488,
      "epoch": 4.545209176788124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003108638149806208,
      "loss": 5.2533,
      "mean_token_accuracy": 0.21824074983596803,
      "num_tokens": 88942330.0,
      "step": 45470
    },
    {
      "entropy": 5.843288993835449,
      "epoch": 4.545709001849353,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003108285817156122,
      "loss": 5.2844,
      "mean_token_accuracy": 0.21638815402984618,
      "num_tokens": 88952060.0,
      "step": 45475
    },
    {
      "entropy": 6.045968723297119,
      "epoch": 4.546208826910581,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00031079334754916573,
      "loss": 5.4689,
      "mean_token_accuracy": 0.20161453932523726,
      "num_tokens": 88962359.0,
      "step": 45480
    },
    {
      "entropy": 5.888232660293579,
      "epoch": 4.54670865197181,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003107581124821679,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21930980831384658,
      "num_tokens": 88971889.0,
      "step": 45485
    },
    {
      "entropy": 5.964999961853027,
      "epoch": 4.547208477033038,
      "grad_norm": 1.15625,
      "learning_rate": 0.00031072287651550507,
      "loss": 5.3366,
      "mean_token_accuracy": 0.19877809882164002,
      "num_tokens": 88980636.0,
      "step": 45490
    },
    {
      "entropy": 5.914403200149536,
      "epoch": 4.547708302094267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003106876396500639,
      "loss": 5.2597,
      "mean_token_accuracy": 0.22694265395402907,
      "num_tokens": 88990965.0,
      "step": 45495
    },
    {
      "entropy": 5.9840686321258545,
      "epoch": 4.548208127155496,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003106524018867309,
      "loss": 5.4638,
      "mean_token_accuracy": 0.19986363500356674,
      "num_tokens": 89001637.0,
      "step": 45500
    },
    {
      "entropy": 6.025525045394898,
      "epoch": 4.548707952216724,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00031061716322639265,
      "loss": 5.4865,
      "mean_token_accuracy": 0.19661663919687272,
      "num_tokens": 89011636.0,
      "step": 45505
    },
    {
      "entropy": 5.941561603546143,
      "epoch": 4.549207777277953,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00031058192366993566,
      "loss": 5.2428,
      "mean_token_accuracy": 0.21991736739873885,
      "num_tokens": 89021826.0,
      "step": 45510
    },
    {
      "entropy": 5.960551452636719,
      "epoch": 4.549707602339181,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00031054668321824677,
      "loss": 5.2774,
      "mean_token_accuracy": 0.21560332477092742,
      "num_tokens": 89031330.0,
      "step": 45515
    },
    {
      "entropy": 5.890732669830323,
      "epoch": 4.55020742740041,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003105114418722124,
      "loss": 5.2957,
      "mean_token_accuracy": 0.21935313194990158,
      "num_tokens": 89040732.0,
      "step": 45520
    },
    {
      "entropy": 5.921638345718383,
      "epoch": 4.550707252461638,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003104761996327193,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2231445237994194,
      "num_tokens": 89051721.0,
      "step": 45525
    },
    {
      "entropy": 5.915173625946045,
      "epoch": 4.551207077522867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003104409565006541,
      "loss": 5.3357,
      "mean_token_accuracy": 0.20119151175022126,
      "num_tokens": 89062098.0,
      "step": 45530
    },
    {
      "entropy": 5.926179075241089,
      "epoch": 4.551706902584096,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003104057124769036,
      "loss": 5.3786,
      "mean_token_accuracy": 0.2038012906908989,
      "num_tokens": 89071173.0,
      "step": 45535
    },
    {
      "entropy": 5.886969995498657,
      "epoch": 4.552206727645324,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003103704675623544,
      "loss": 5.2756,
      "mean_token_accuracy": 0.21848673075437547,
      "num_tokens": 89080996.0,
      "step": 45540
    },
    {
      "entropy": 5.977472972869873,
      "epoch": 4.552706552706553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003103352217578935,
      "loss": 5.3585,
      "mean_token_accuracy": 0.21241610050201415,
      "num_tokens": 89089799.0,
      "step": 45545
    },
    {
      "entropy": 5.881862354278565,
      "epoch": 4.553206377767781,
      "grad_norm": 1.03125,
      "learning_rate": 0.00031029997506440735,
      "loss": 5.3433,
      "mean_token_accuracy": 0.21146129816770554,
      "num_tokens": 89099872.0,
      "step": 45550
    },
    {
      "entropy": 5.919081544876098,
      "epoch": 4.55370620282901,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003102647274827831,
      "loss": 5.3245,
      "mean_token_accuracy": 0.21117850095033647,
      "num_tokens": 89109418.0,
      "step": 45555
    },
    {
      "entropy": 5.8441077709198,
      "epoch": 4.554206027890238,
      "grad_norm": 1.09375,
      "learning_rate": 0.00031022947901390724,
      "loss": 5.1902,
      "mean_token_accuracy": 0.2231324091553688,
      "num_tokens": 89119222.0,
      "step": 45560
    },
    {
      "entropy": 5.911742687225342,
      "epoch": 4.554705852951467,
      "grad_norm": 1.265625,
      "learning_rate": 0.00031019422965866683,
      "loss": 5.3485,
      "mean_token_accuracy": 0.20459483712911605,
      "num_tokens": 89128559.0,
      "step": 45565
    },
    {
      "entropy": 5.978163480758667,
      "epoch": 4.555205678012696,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003101589794179486,
      "loss": 5.3905,
      "mean_token_accuracy": 0.20250810831785201,
      "num_tokens": 89138750.0,
      "step": 45570
    },
    {
      "entropy": 5.894885301589966,
      "epoch": 4.555705503073924,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003101237282926394,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21885441690683366,
      "num_tokens": 89148125.0,
      "step": 45575
    },
    {
      "entropy": 5.819274806976319,
      "epoch": 4.556205328135153,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0003100884762836264,
      "loss": 5.154,
      "mean_token_accuracy": 0.23581029176712037,
      "num_tokens": 89157407.0,
      "step": 45580
    },
    {
      "entropy": 5.815608501434326,
      "epoch": 4.556705153196381,
      "grad_norm": 1.28125,
      "learning_rate": 0.0003100532233917963,
      "loss": 5.1865,
      "mean_token_accuracy": 0.22771327793598176,
      "num_tokens": 89168162.0,
      "step": 45585
    },
    {
      "entropy": 5.930879831314087,
      "epoch": 4.55720497825761,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00031001796961803614,
      "loss": 5.3728,
      "mean_token_accuracy": 0.2114837184548378,
      "num_tokens": 89177551.0,
      "step": 45590
    },
    {
      "entropy": 6.000883722305298,
      "epoch": 4.557704803318838,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030998271496323286,
      "loss": 5.3741,
      "mean_token_accuracy": 0.2066744565963745,
      "num_tokens": 89186461.0,
      "step": 45595
    },
    {
      "entropy": 5.946384525299072,
      "epoch": 4.558204628380067,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030994745942827345,
      "loss": 5.4135,
      "mean_token_accuracy": 0.20759313702583312,
      "num_tokens": 89196959.0,
      "step": 45600
    },
    {
      "entropy": 5.887782335281372,
      "epoch": 4.558704453441296,
      "grad_norm": 1.25,
      "learning_rate": 0.000309912203014045,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21565690785646438,
      "num_tokens": 89206727.0,
      "step": 45605
    },
    {
      "entropy": 5.736345958709717,
      "epoch": 4.559204278502524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003098769457214345,
      "loss": 5.1509,
      "mean_token_accuracy": 0.22816329151391984,
      "num_tokens": 89216502.0,
      "step": 45610
    },
    {
      "entropy": 5.969074296951294,
      "epoch": 4.559704103563753,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000309841687551329,
      "loss": 5.4022,
      "mean_token_accuracy": 0.2031787022948265,
      "num_tokens": 89226616.0,
      "step": 45615
    },
    {
      "entropy": 5.985136938095093,
      "epoch": 4.560203928624981,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003098064285046156,
      "loss": 5.3327,
      "mean_token_accuracy": 0.20949507504701614,
      "num_tokens": 89237346.0,
      "step": 45620
    },
    {
      "entropy": 5.815692472457886,
      "epoch": 4.56070375368621,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003097711685821815,
      "loss": 5.187,
      "mean_token_accuracy": 0.2198987439274788,
      "num_tokens": 89246691.0,
      "step": 45625
    },
    {
      "entropy": 5.732631254196167,
      "epoch": 4.561203578747438,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030973590778491354,
      "loss": 5.2278,
      "mean_token_accuracy": 0.22913377583026887,
      "num_tokens": 89255781.0,
      "step": 45630
    },
    {
      "entropy": 5.893719339370728,
      "epoch": 4.561703403808667,
      "grad_norm": 1.125,
      "learning_rate": 0.00030970064611369927,
      "loss": 5.3171,
      "mean_token_accuracy": 0.21388984024524688,
      "num_tokens": 89265254.0,
      "step": 45635
    },
    {
      "entropy": 5.911330604553223,
      "epoch": 4.562203228869896,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003096653835694256,
      "loss": 5.3551,
      "mean_token_accuracy": 0.21269649863243104,
      "num_tokens": 89274581.0,
      "step": 45640
    },
    {
      "entropy": 5.972276639938355,
      "epoch": 4.562703053931124,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003096301201529798,
      "loss": 5.229,
      "mean_token_accuracy": 0.2177905485033989,
      "num_tokens": 89283926.0,
      "step": 45645
    },
    {
      "entropy": 5.851644802093506,
      "epoch": 4.563202878992353,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003095948558652492,
      "loss": 5.2346,
      "mean_token_accuracy": 0.21804803907871245,
      "num_tokens": 89294894.0,
      "step": 45650
    },
    {
      "entropy": 5.8458000183105465,
      "epoch": 4.563702704053581,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030955959070712075,
      "loss": 5.3246,
      "mean_token_accuracy": 0.21037036180496216,
      "num_tokens": 89305080.0,
      "step": 45655
    },
    {
      "entropy": 5.927121877670288,
      "epoch": 4.56420252911481,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030952432467948196,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20714798271656037,
      "num_tokens": 89314604.0,
      "step": 45660
    },
    {
      "entropy": 6.006836843490601,
      "epoch": 4.564702354176038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003094890577832201,
      "loss": 5.3686,
      "mean_token_accuracy": 0.2121601328253746,
      "num_tokens": 89324115.0,
      "step": 45665
    },
    {
      "entropy": 5.908967351913452,
      "epoch": 4.565202179237267,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030945379001922236,
      "loss": 5.3512,
      "mean_token_accuracy": 0.2142158716917038,
      "num_tokens": 89334241.0,
      "step": 45670
    },
    {
      "entropy": 5.97304892539978,
      "epoch": 4.565702004298496,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003094185213883761,
      "loss": 5.4224,
      "mean_token_accuracy": 0.19957073032855988,
      "num_tokens": 89345327.0,
      "step": 45675
    },
    {
      "entropy": 5.921043014526367,
      "epoch": 4.566201829359724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003093832518915688,
      "loss": 5.3774,
      "mean_token_accuracy": 0.20795324742794036,
      "num_tokens": 89354641.0,
      "step": 45680
    },
    {
      "entropy": 5.896199703216553,
      "epoch": 4.566701654420953,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003093479815296877,
      "loss": 5.3727,
      "mean_token_accuracy": 0.20548345297574996,
      "num_tokens": 89364407.0,
      "step": 45685
    },
    {
      "entropy": 6.0053098678588865,
      "epoch": 4.567201479482181,
      "grad_norm": 1.25,
      "learning_rate": 0.00030931271030362013,
      "loss": 5.344,
      "mean_token_accuracy": 0.2082348346710205,
      "num_tokens": 89373559.0,
      "step": 45690
    },
    {
      "entropy": 5.888160562515258,
      "epoch": 4.56770130454341,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003092774382142537,
      "loss": 5.2431,
      "mean_token_accuracy": 0.22422612458467484,
      "num_tokens": 89383659.0,
      "step": 45695
    },
    {
      "entropy": 5.863918304443359,
      "epoch": 4.568201129604638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030924216526247573,
      "loss": 5.2664,
      "mean_token_accuracy": 0.2220575138926506,
      "num_tokens": 89392260.0,
      "step": 45700
    },
    {
      "entropy": 5.870566415786743,
      "epoch": 4.568700954665867,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003092068914491736,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2072972908616066,
      "num_tokens": 89401704.0,
      "step": 45705
    },
    {
      "entropy": 5.8657479763031,
      "epoch": 4.569200779727096,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030917161677523496,
      "loss": 5.2676,
      "mean_token_accuracy": 0.2227622464299202,
      "num_tokens": 89411237.0,
      "step": 45710
    },
    {
      "entropy": 5.966302585601807,
      "epoch": 4.569700604788324,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003091363412415473,
      "loss": 5.3522,
      "mean_token_accuracy": 0.2016542911529541,
      "num_tokens": 89420109.0,
      "step": 45715
    },
    {
      "entropy": 5.953137302398682,
      "epoch": 4.570200429849553,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030910106484899796,
      "loss": 5.4342,
      "mean_token_accuracy": 0.19790462404489517,
      "num_tokens": 89429607.0,
      "step": 45720
    },
    {
      "entropy": 5.897829008102417,
      "epoch": 4.570700254910781,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003090657875984747,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21854755282402039,
      "num_tokens": 89440602.0,
      "step": 45725
    },
    {
      "entropy": 5.903642559051514,
      "epoch": 4.57120007997201,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030903050949086493,
      "loss": 5.3074,
      "mean_token_accuracy": 0.21080104857683182,
      "num_tokens": 89450172.0,
      "step": 45730
    },
    {
      "entropy": 5.8629372119903564,
      "epoch": 4.571699905033238,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003089952305270563,
      "loss": 5.2971,
      "mean_token_accuracy": 0.21279758512973784,
      "num_tokens": 89460102.0,
      "step": 45735
    },
    {
      "entropy": 5.8712951183319095,
      "epoch": 4.572199730094467,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003089599507079364,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21794395595788957,
      "num_tokens": 89469034.0,
      "step": 45740
    },
    {
      "entropy": 5.898723840713501,
      "epoch": 4.572699555155696,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030892467003439296,
      "loss": 5.3265,
      "mean_token_accuracy": 0.21452469378709793,
      "num_tokens": 89479170.0,
      "step": 45745
    },
    {
      "entropy": 6.000688266754151,
      "epoch": 4.573199380216924,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030888938850731347,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19239316880702972,
      "num_tokens": 89490077.0,
      "step": 45750
    },
    {
      "entropy": 5.939696550369263,
      "epoch": 4.573699205278153,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030885410612758573,
      "loss": 5.3324,
      "mean_token_accuracy": 0.20996961891651153,
      "num_tokens": 89498784.0,
      "step": 45755
    },
    {
      "entropy": 6.013583946228027,
      "epoch": 4.574199030339381,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030881882289609733,
      "loss": 5.4016,
      "mean_token_accuracy": 0.20402883291244506,
      "num_tokens": 89509354.0,
      "step": 45760
    },
    {
      "entropy": 5.981541395187378,
      "epoch": 4.57469885540061,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003087835388137361,
      "loss": 5.4427,
      "mean_token_accuracy": 0.2022375300526619,
      "num_tokens": 89518023.0,
      "step": 45765
    },
    {
      "entropy": 5.999418783187866,
      "epoch": 4.575198680461838,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003087482538813897,
      "loss": 5.4692,
      "mean_token_accuracy": 0.19546595066785813,
      "num_tokens": 89526781.0,
      "step": 45770
    },
    {
      "entropy": 5.911930322647095,
      "epoch": 4.575698505523067,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030871296809994596,
      "loss": 5.3625,
      "mean_token_accuracy": 0.20735154002904893,
      "num_tokens": 89536793.0,
      "step": 45775
    },
    {
      "entropy": 5.991241312026977,
      "epoch": 4.576198330584296,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003086776814702926,
      "loss": 5.4161,
      "mean_token_accuracy": 0.2014296382665634,
      "num_tokens": 89546753.0,
      "step": 45780
    },
    {
      "entropy": 5.895236921310425,
      "epoch": 4.576698155645524,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030864239399331744,
      "loss": 5.2926,
      "mean_token_accuracy": 0.21592307090759277,
      "num_tokens": 89558152.0,
      "step": 45785
    },
    {
      "entropy": 5.929335641860962,
      "epoch": 4.577197980706753,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003086071056699083,
      "loss": 5.3168,
      "mean_token_accuracy": 0.2019724056124687,
      "num_tokens": 89567917.0,
      "step": 45790
    },
    {
      "entropy": 5.741006660461426,
      "epoch": 4.577697805767981,
      "grad_norm": 1.1875,
      "learning_rate": 0.000308571816500953,
      "loss": 5.1816,
      "mean_token_accuracy": 0.22226639986038207,
      "num_tokens": 89577658.0,
      "step": 45795
    },
    {
      "entropy": 5.9677001476287845,
      "epoch": 4.57819763082921,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003085365264873395,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20331211090087892,
      "num_tokens": 89586850.0,
      "step": 45800
    },
    {
      "entropy": 6.0321259021759035,
      "epoch": 4.578697455890438,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003085012356299556,
      "loss": 5.4148,
      "mean_token_accuracy": 0.20257012397050858,
      "num_tokens": 89595875.0,
      "step": 45805
    },
    {
      "entropy": 5.883709955215454,
      "epoch": 4.579197280951667,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030846594392968925,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2066540911793709,
      "num_tokens": 89605708.0,
      "step": 45810
    },
    {
      "entropy": 5.909187650680542,
      "epoch": 4.579697106012896,
      "grad_norm": 1.125,
      "learning_rate": 0.00030843065138742826,
      "loss": 5.2864,
      "mean_token_accuracy": 0.2180247873067856,
      "num_tokens": 89615199.0,
      "step": 45815
    },
    {
      "entropy": 5.99703459739685,
      "epoch": 4.580196931074124,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00030839535800406085,
      "loss": 5.362,
      "mean_token_accuracy": 0.20503070801496506,
      "num_tokens": 89623936.0,
      "step": 45820
    },
    {
      "entropy": 5.972574615478516,
      "epoch": 4.580696756135353,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030836006378047464,
      "loss": 5.3659,
      "mean_token_accuracy": 0.21180226951837539,
      "num_tokens": 89634167.0,
      "step": 45825
    },
    {
      "entropy": 5.9842010021209715,
      "epoch": 4.581196581196581,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003083247687175579,
      "loss": 5.3594,
      "mean_token_accuracy": 0.20563629120588303,
      "num_tokens": 89643802.0,
      "step": 45830
    },
    {
      "entropy": 5.9848002910614015,
      "epoch": 4.58169640625781,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030828947281619863,
      "loss": 5.4294,
      "mean_token_accuracy": 0.20088664293289185,
      "num_tokens": 89653583.0,
      "step": 45835
    },
    {
      "entropy": 5.797790384292602,
      "epoch": 4.582196231319038,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003082541760772847,
      "loss": 5.19,
      "mean_token_accuracy": 0.22312134951353074,
      "num_tokens": 89663542.0,
      "step": 45840
    },
    {
      "entropy": 5.83638596534729,
      "epoch": 4.582696056380267,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030821887850170424,
      "loss": 5.3366,
      "mean_token_accuracy": 0.2105124443769455,
      "num_tokens": 89673881.0,
      "step": 45845
    },
    {
      "entropy": 5.901673126220703,
      "epoch": 4.583195881441496,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030818358009034535,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21484507769346237,
      "num_tokens": 89684255.0,
      "step": 45850
    },
    {
      "entropy": 5.888594770431519,
      "epoch": 4.583695706502724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030814828084409614,
      "loss": 5.2932,
      "mean_token_accuracy": 0.21890255957841873,
      "num_tokens": 89692710.0,
      "step": 45855
    },
    {
      "entropy": 5.914256525039673,
      "epoch": 4.584195531563953,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030811298076384474,
      "loss": 5.3422,
      "mean_token_accuracy": 0.20967935770750046,
      "num_tokens": 89702357.0,
      "step": 45860
    },
    {
      "entropy": 5.944733905792236,
      "epoch": 4.584695356625181,
      "grad_norm": 1.03125,
      "learning_rate": 0.00030807767985047927,
      "loss": 5.3021,
      "mean_token_accuracy": 0.20960677415132523,
      "num_tokens": 89712669.0,
      "step": 45865
    },
    {
      "entropy": 5.995772838592529,
      "epoch": 4.58519518168641,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003080423781048879,
      "loss": 5.3439,
      "mean_token_accuracy": 0.21141692250967026,
      "num_tokens": 89721578.0,
      "step": 45870
    },
    {
      "entropy": 5.900690984725952,
      "epoch": 4.585695006747638,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030800707552795873,
      "loss": 5.2755,
      "mean_token_accuracy": 0.21599923372268676,
      "num_tokens": 89731270.0,
      "step": 45875
    },
    {
      "entropy": 5.946036958694458,
      "epoch": 4.586194831808867,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030797177212058,
      "loss": 5.3333,
      "mean_token_accuracy": 0.20312947034835815,
      "num_tokens": 89740786.0,
      "step": 45880
    },
    {
      "entropy": 5.827971601486206,
      "epoch": 4.586694656870096,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003079364678836401,
      "loss": 5.2432,
      "mean_token_accuracy": 0.22585353553295134,
      "num_tokens": 89749616.0,
      "step": 45885
    },
    {
      "entropy": 5.833753824234009,
      "epoch": 4.587194481931324,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030790116281802704,
      "loss": 5.2662,
      "mean_token_accuracy": 0.20874898582696916,
      "num_tokens": 89758461.0,
      "step": 45890
    },
    {
      "entropy": 5.692119598388672,
      "epoch": 4.587694306992553,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003078658569246293,
      "loss": 5.0581,
      "mean_token_accuracy": 0.23723353147506715,
      "num_tokens": 89768707.0,
      "step": 45895
    },
    {
      "entropy": 5.94467601776123,
      "epoch": 4.588194132053781,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003078305502043349,
      "loss": 5.31,
      "mean_token_accuracy": 0.20637002885341643,
      "num_tokens": 89778428.0,
      "step": 45900
    },
    {
      "entropy": 5.913442945480346,
      "epoch": 4.58869395711501,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003077952426580324,
      "loss": 5.2958,
      "mean_token_accuracy": 0.21461859047412873,
      "num_tokens": 89789113.0,
      "step": 45905
    },
    {
      "entropy": 6.052037382125855,
      "epoch": 4.589193782176238,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003077599342866101,
      "loss": 5.4666,
      "mean_token_accuracy": 0.20799022763967515,
      "num_tokens": 89799663.0,
      "step": 45910
    },
    {
      "entropy": 5.939077711105346,
      "epoch": 4.589693607237467,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030772462509095615,
      "loss": 5.4173,
      "mean_token_accuracy": 0.20504551380872726,
      "num_tokens": 89810916.0,
      "step": 45915
    },
    {
      "entropy": 5.790314674377441,
      "epoch": 4.590193432298696,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030768931507195907,
      "loss": 5.1483,
      "mean_token_accuracy": 0.23388005048036575,
      "num_tokens": 89821050.0,
      "step": 45920
    },
    {
      "entropy": 5.976540279388428,
      "epoch": 4.590693257359924,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030765400423050725,
      "loss": 5.4317,
      "mean_token_accuracy": 0.20377224385738374,
      "num_tokens": 89831456.0,
      "step": 45925
    },
    {
      "entropy": 5.950364112854004,
      "epoch": 4.591193082421153,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030761869256748915,
      "loss": 5.3645,
      "mean_token_accuracy": 0.2132189005613327,
      "num_tokens": 89842564.0,
      "step": 45930
    },
    {
      "entropy": 5.90964035987854,
      "epoch": 4.591692907482381,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003075833800837931,
      "loss": 5.2302,
      "mean_token_accuracy": 0.21462469846010207,
      "num_tokens": 89852138.0,
      "step": 45935
    },
    {
      "entropy": 5.906953048706055,
      "epoch": 4.59219273254361,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003075480667803076,
      "loss": 5.4241,
      "mean_token_accuracy": 0.20221638828516006,
      "num_tokens": 89863497.0,
      "step": 45940
    },
    {
      "entropy": 5.862451410293579,
      "epoch": 4.592692557604838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003075127526579211,
      "loss": 5.2117,
      "mean_token_accuracy": 0.2205385074019432,
      "num_tokens": 89873024.0,
      "step": 45945
    },
    {
      "entropy": 5.953746175765991,
      "epoch": 4.593192382666067,
      "grad_norm": 1.125,
      "learning_rate": 0.0003074774377175221,
      "loss": 5.3453,
      "mean_token_accuracy": 0.208685265481472,
      "num_tokens": 89883633.0,
      "step": 45950
    },
    {
      "entropy": 5.922483968734741,
      "epoch": 4.593692207727296,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003074421219599992,
      "loss": 5.3988,
      "mean_token_accuracy": 0.20290178060531616,
      "num_tokens": 89893715.0,
      "step": 45955
    },
    {
      "entropy": 5.890619468688965,
      "epoch": 4.594192032788524,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030740680538624075,
      "loss": 5.319,
      "mean_token_accuracy": 0.20665154457092286,
      "num_tokens": 89903551.0,
      "step": 45960
    },
    {
      "entropy": 5.950121545791626,
      "epoch": 4.594691857849752,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003073714879971355,
      "loss": 5.311,
      "mean_token_accuracy": 0.20917905271053314,
      "num_tokens": 89914089.0,
      "step": 45965
    },
    {
      "entropy": 5.935127878189087,
      "epoch": 4.595191682910981,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030733616979357185,
      "loss": 5.3456,
      "mean_token_accuracy": 0.2175429046154022,
      "num_tokens": 89924806.0,
      "step": 45970
    },
    {
      "entropy": 5.89199914932251,
      "epoch": 4.59569150797221,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003073008507764386,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21709855198860167,
      "num_tokens": 89934819.0,
      "step": 45975
    },
    {
      "entropy": 5.7649740219116214,
      "epoch": 4.596191333033438,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030726553094662416,
      "loss": 5.0898,
      "mean_token_accuracy": 0.230587737262249,
      "num_tokens": 89944408.0,
      "step": 45980
    },
    {
      "entropy": 5.977861595153809,
      "epoch": 4.596691158094667,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003072302103050173,
      "loss": 5.4092,
      "mean_token_accuracy": 0.2016238287091255,
      "num_tokens": 89953686.0,
      "step": 45985
    },
    {
      "entropy": 5.920377206802368,
      "epoch": 4.597190983155896,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003071948888525067,
      "loss": 5.358,
      "mean_token_accuracy": 0.20300329923629762,
      "num_tokens": 89963002.0,
      "step": 45990
    },
    {
      "entropy": 5.885420560836792,
      "epoch": 4.597690808217124,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003071595665899809,
      "loss": 5.261,
      "mean_token_accuracy": 0.21882621496915816,
      "num_tokens": 89973155.0,
      "step": 45995
    },
    {
      "entropy": 5.941894388198852,
      "epoch": 4.598190633278352,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030712424351832875,
      "loss": 5.3213,
      "mean_token_accuracy": 0.21129991859197617,
      "num_tokens": 89982268.0,
      "step": 46000
    },
    {
      "entropy": 5.938103151321411,
      "epoch": 4.598690458339581,
      "grad_norm": 1.125,
      "learning_rate": 0.0003070889196384388,
      "loss": 5.3012,
      "mean_token_accuracy": 0.2178995668888092,
      "num_tokens": 89993160.0,
      "step": 46005
    },
    {
      "entropy": 5.950436401367187,
      "epoch": 4.59919028340081,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030705359495119996,
      "loss": 5.3761,
      "mean_token_accuracy": 0.2086759999394417,
      "num_tokens": 90002367.0,
      "step": 46010
    },
    {
      "entropy": 5.883538055419922,
      "epoch": 4.599690108462038,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003070182694575009,
      "loss": 5.2978,
      "mean_token_accuracy": 0.21010163128376008,
      "num_tokens": 90011550.0,
      "step": 46015
    },
    {
      "entropy": 6.064204359054566,
      "epoch": 4.600189933523267,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003069829431582304,
      "loss": 5.4858,
      "mean_token_accuracy": 0.19846067428588868,
      "num_tokens": 90020857.0,
      "step": 46020
    },
    {
      "entropy": 5.89028263092041,
      "epoch": 4.600689758584496,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003069476160542773,
      "loss": 5.3037,
      "mean_token_accuracy": 0.21111615151166915,
      "num_tokens": 90030551.0,
      "step": 46025
    },
    {
      "entropy": 5.909535455703735,
      "epoch": 4.6011895836457235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003069122881465305,
      "loss": 5.2901,
      "mean_token_accuracy": 0.22067215889692307,
      "num_tokens": 90041024.0,
      "step": 46030
    },
    {
      "entropy": 5.935566854476929,
      "epoch": 4.601689408706952,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030687695943587854,
      "loss": 5.3491,
      "mean_token_accuracy": 0.20822479724884033,
      "num_tokens": 90050673.0,
      "step": 46035
    },
    {
      "entropy": 5.957933044433593,
      "epoch": 4.602189233768181,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003068416299232107,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20911093503236772,
      "num_tokens": 90060134.0,
      "step": 46040
    },
    {
      "entropy": 5.935232734680175,
      "epoch": 4.60268905882941,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003068062996094155,
      "loss": 5.2858,
      "mean_token_accuracy": 0.215269273519516,
      "num_tokens": 90068330.0,
      "step": 46045
    },
    {
      "entropy": 5.911519622802734,
      "epoch": 4.603188883890638,
      "grad_norm": 1.203125,
      "learning_rate": 0.000306770968495382,
      "loss": 5.3635,
      "mean_token_accuracy": 0.20130140632390975,
      "num_tokens": 90078504.0,
      "step": 46050
    },
    {
      "entropy": 5.951037883758545,
      "epoch": 4.603688708951867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003067356365819992,
      "loss": 5.3879,
      "mean_token_accuracy": 0.20471150130033494,
      "num_tokens": 90088091.0,
      "step": 46055
    },
    {
      "entropy": 5.931247425079346,
      "epoch": 4.604188534013096,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030670030387015585,
      "loss": 5.2805,
      "mean_token_accuracy": 0.21695020347833632,
      "num_tokens": 90098078.0,
      "step": 46060
    },
    {
      "entropy": 5.960692548751831,
      "epoch": 4.6046883590743235,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030666497036074104,
      "loss": 5.413,
      "mean_token_accuracy": 0.20218108892440795,
      "num_tokens": 90107626.0,
      "step": 46065
    },
    {
      "entropy": 5.895064020156861,
      "epoch": 4.605188184135552,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003066296360546437,
      "loss": 5.2436,
      "mean_token_accuracy": 0.2206966444849968,
      "num_tokens": 90116504.0,
      "step": 46070
    },
    {
      "entropy": 5.898382091522217,
      "epoch": 4.605688009196781,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030659430095275283,
      "loss": 5.2481,
      "mean_token_accuracy": 0.2181527778506279,
      "num_tokens": 90125972.0,
      "step": 46075
    },
    {
      "entropy": 5.850804996490479,
      "epoch": 4.60618783425801,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003065589650559575,
      "loss": 5.2161,
      "mean_token_accuracy": 0.21970392912626266,
      "num_tokens": 90135715.0,
      "step": 46080
    },
    {
      "entropy": 5.89878396987915,
      "epoch": 4.606687659319238,
      "grad_norm": 1.125,
      "learning_rate": 0.00030652362836514687,
      "loss": 5.2764,
      "mean_token_accuracy": 0.21993841081857682,
      "num_tokens": 90144921.0,
      "step": 46085
    },
    {
      "entropy": 5.936382150650024,
      "epoch": 4.607187484380467,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030648829088120974,
      "loss": 5.3901,
      "mean_token_accuracy": 0.20003775954246522,
      "num_tokens": 90154518.0,
      "step": 46090
    },
    {
      "entropy": 5.984733390808105,
      "epoch": 4.607687309441696,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003064529526050353,
      "loss": 5.4836,
      "mean_token_accuracy": 0.19425608366727828,
      "num_tokens": 90165841.0,
      "step": 46095
    },
    {
      "entropy": 5.945958614349365,
      "epoch": 4.6081871345029235,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003064176135375127,
      "loss": 5.343,
      "mean_token_accuracy": 0.21073115319013597,
      "num_tokens": 90175708.0,
      "step": 46100
    },
    {
      "entropy": 5.936034679412842,
      "epoch": 4.608686959564152,
      "grad_norm": 1.171875,
      "learning_rate": 0.000306382273679531,
      "loss": 5.3716,
      "mean_token_accuracy": 0.2035694569349289,
      "num_tokens": 90185908.0,
      "step": 46105
    },
    {
      "entropy": 5.891686296463012,
      "epoch": 4.609186784625381,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030634693303197935,
      "loss": 5.2455,
      "mean_token_accuracy": 0.21973129957914353,
      "num_tokens": 90195200.0,
      "step": 46110
    },
    {
      "entropy": 5.864194583892822,
      "epoch": 4.60968660968661,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030631159159574703,
      "loss": 5.2329,
      "mean_token_accuracy": 0.21564197391271592,
      "num_tokens": 90205532.0,
      "step": 46115
    },
    {
      "entropy": 5.899693012237549,
      "epoch": 4.610186434747838,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030627624937172305,
      "loss": 5.3804,
      "mean_token_accuracy": 0.2107738345861435,
      "num_tokens": 90216288.0,
      "step": 46120
    },
    {
      "entropy": 5.965274286270142,
      "epoch": 4.610686259809067,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003062409063607967,
      "loss": 5.2943,
      "mean_token_accuracy": 0.21562719196081162,
      "num_tokens": 90225717.0,
      "step": 46125
    },
    {
      "entropy": 5.945536136627197,
      "epoch": 4.611186084870296,
      "grad_norm": 1.125,
      "learning_rate": 0.0003062055625638572,
      "loss": 5.3607,
      "mean_token_accuracy": 0.21045325994491576,
      "num_tokens": 90235624.0,
      "step": 46130
    },
    {
      "entropy": 5.943359994888306,
      "epoch": 4.6116859099315235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003061702179817937,
      "loss": 5.3432,
      "mean_token_accuracy": 0.2113630548119545,
      "num_tokens": 90246465.0,
      "step": 46135
    },
    {
      "entropy": 5.882848405838013,
      "epoch": 4.612185734992752,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003061348726154956,
      "loss": 5.2454,
      "mean_token_accuracy": 0.22230808287858964,
      "num_tokens": 90256249.0,
      "step": 46140
    },
    {
      "entropy": 5.950532341003418,
      "epoch": 4.612685560053981,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003060995264658521,
      "loss": 5.4485,
      "mean_token_accuracy": 0.20085462033748627,
      "num_tokens": 90267110.0,
      "step": 46145
    },
    {
      "entropy": 5.949943685531617,
      "epoch": 4.61318538511521,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003060641795337526,
      "loss": 5.3758,
      "mean_token_accuracy": 0.20794351845979692,
      "num_tokens": 90276881.0,
      "step": 46150
    },
    {
      "entropy": 5.916820478439331,
      "epoch": 4.613685210176438,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030602883182008625,
      "loss": 5.289,
      "mean_token_accuracy": 0.20942394733428954,
      "num_tokens": 90285941.0,
      "step": 46155
    },
    {
      "entropy": 5.849282073974609,
      "epoch": 4.614185035237667,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003059934833257425,
      "loss": 5.2574,
      "mean_token_accuracy": 0.21837549209594725,
      "num_tokens": 90295739.0,
      "step": 46160
    },
    {
      "entropy": 5.910872459411621,
      "epoch": 4.614684860298896,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030595813405161063,
      "loss": 5.3601,
      "mean_token_accuracy": 0.2058721140027046,
      "num_tokens": 90306243.0,
      "step": 46165
    },
    {
      "entropy": 5.865127992630005,
      "epoch": 4.6151846853601235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003059227839985801,
      "loss": 5.2284,
      "mean_token_accuracy": 0.21716260462999343,
      "num_tokens": 90316610.0,
      "step": 46170
    },
    {
      "entropy": 5.941087865829468,
      "epoch": 4.615684510421352,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003058874331675403,
      "loss": 5.3382,
      "mean_token_accuracy": 0.2060970261693001,
      "num_tokens": 90326632.0,
      "step": 46175
    },
    {
      "entropy": 5.9450116634368895,
      "epoch": 4.616184335482581,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003058520815593807,
      "loss": 5.3073,
      "mean_token_accuracy": 0.2121728241443634,
      "num_tokens": 90335923.0,
      "step": 46180
    },
    {
      "entropy": 5.876146411895752,
      "epoch": 4.61668416054381,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030581672917499056,
      "loss": 5.2582,
      "mean_token_accuracy": 0.21861644238233566,
      "num_tokens": 90346020.0,
      "step": 46185
    },
    {
      "entropy": 5.967067241668701,
      "epoch": 4.617183985605038,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030578137601525954,
      "loss": 5.3272,
      "mean_token_accuracy": 0.20610061138868332,
      "num_tokens": 90355808.0,
      "step": 46190
    },
    {
      "entropy": 5.990263414382935,
      "epoch": 4.617683810666267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003057460220810769,
      "loss": 5.4179,
      "mean_token_accuracy": 0.20991218388080596,
      "num_tokens": 90367202.0,
      "step": 46195
    },
    {
      "entropy": 5.994425582885742,
      "epoch": 4.618183635727496,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030571066737333227,
      "loss": 5.3038,
      "mean_token_accuracy": 0.21508779674768447,
      "num_tokens": 90376392.0,
      "step": 46200
    },
    {
      "entropy": 5.891765594482422,
      "epoch": 4.6186834607887235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003056753118929152,
      "loss": 5.3665,
      "mean_token_accuracy": 0.2102070778608322,
      "num_tokens": 90386262.0,
      "step": 46205
    },
    {
      "entropy": 5.958794403076172,
      "epoch": 4.619183285849952,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003056399556407152,
      "loss": 5.3765,
      "mean_token_accuracy": 0.2095910668373108,
      "num_tokens": 90395374.0,
      "step": 46210
    },
    {
      "entropy": 5.931639385223389,
      "epoch": 4.619683110911181,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030560459861762174,
      "loss": 5.2826,
      "mean_token_accuracy": 0.21225636452436447,
      "num_tokens": 90405469.0,
      "step": 46215
    },
    {
      "entropy": 5.947641468048095,
      "epoch": 4.62018293597241,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030556924082452455,
      "loss": 5.4155,
      "mean_token_accuracy": 0.19871770292520524,
      "num_tokens": 90416022.0,
      "step": 46220
    },
    {
      "entropy": 5.987388658523559,
      "epoch": 4.620682761033638,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030553388226231304,
      "loss": 5.3178,
      "mean_token_accuracy": 0.20677438527345657,
      "num_tokens": 90426346.0,
      "step": 46225
    },
    {
      "entropy": 5.940309381484985,
      "epoch": 4.621182586094867,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030549852293187686,
      "loss": 5.3014,
      "mean_token_accuracy": 0.2111608773469925,
      "num_tokens": 90436285.0,
      "step": 46230
    },
    {
      "entropy": 5.901546859741211,
      "epoch": 4.621682411156096,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003054631628341058,
      "loss": 5.2381,
      "mean_token_accuracy": 0.2233700543642044,
      "num_tokens": 90447314.0,
      "step": 46235
    },
    {
      "entropy": 5.925978755950927,
      "epoch": 4.6221822362173235,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030542780196988936,
      "loss": 5.3608,
      "mean_token_accuracy": 0.202962289750576,
      "num_tokens": 90456786.0,
      "step": 46240
    },
    {
      "entropy": 5.884938478469849,
      "epoch": 4.622682061278552,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030539244034011727,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2164239138364792,
      "num_tokens": 90465805.0,
      "step": 46245
    },
    {
      "entropy": 6.042507553100586,
      "epoch": 4.623181886339781,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003053570779456792,
      "loss": 5.5084,
      "mean_token_accuracy": 0.1993957355618477,
      "num_tokens": 90475512.0,
      "step": 46250
    },
    {
      "entropy": 5.987387371063233,
      "epoch": 4.62368171140101,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030532171478746484,
      "loss": 5.3827,
      "mean_token_accuracy": 0.20790136456489564,
      "num_tokens": 90484911.0,
      "step": 46255
    },
    {
      "entropy": 5.916531991958618,
      "epoch": 4.624181536462238,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003052863508663639,
      "loss": 5.411,
      "mean_token_accuracy": 0.19598933905363083,
      "num_tokens": 90495752.0,
      "step": 46260
    },
    {
      "entropy": 5.866849184036255,
      "epoch": 4.624681361523467,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00030525098618326625,
      "loss": 5.3,
      "mean_token_accuracy": 0.21618552207946778,
      "num_tokens": 90505332.0,
      "step": 46265
    },
    {
      "entropy": 5.958974170684814,
      "epoch": 4.625181186584696,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030521562073906156,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20644464939832688,
      "num_tokens": 90515321.0,
      "step": 46270
    },
    {
      "entropy": 5.924850130081177,
      "epoch": 4.6256810116459235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003051802545346396,
      "loss": 5.3279,
      "mean_token_accuracy": 0.21080789864063262,
      "num_tokens": 90525193.0,
      "step": 46275
    },
    {
      "entropy": 5.973572015762329,
      "epoch": 4.626180836707152,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030514488757089013,
      "loss": 5.313,
      "mean_token_accuracy": 0.2134695142507553,
      "num_tokens": 90534088.0,
      "step": 46280
    },
    {
      "entropy": 5.913407468795777,
      "epoch": 4.626680661768381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003051095198487032,
      "loss": 5.3124,
      "mean_token_accuracy": 0.205995175242424,
      "num_tokens": 90543508.0,
      "step": 46285
    },
    {
      "entropy": 5.828043985366821,
      "epoch": 4.62718048682961,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030507415136896837,
      "loss": 5.2352,
      "mean_token_accuracy": 0.22319044023752213,
      "num_tokens": 90553657.0,
      "step": 46290
    },
    {
      "entropy": 5.868379545211792,
      "epoch": 4.627680311890838,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003050387821325757,
      "loss": 5.2635,
      "mean_token_accuracy": 0.2155291348695755,
      "num_tokens": 90563250.0,
      "step": 46295
    },
    {
      "entropy": 6.043000364303589,
      "epoch": 4.628180136952067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000305003412140415,
      "loss": 5.3568,
      "mean_token_accuracy": 0.20757049024105073,
      "num_tokens": 90572453.0,
      "step": 46300
    },
    {
      "entropy": 5.983371019363403,
      "epoch": 4.628679962013296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030496804139337617,
      "loss": 5.3662,
      "mean_token_accuracy": 0.2074785053730011,
      "num_tokens": 90583156.0,
      "step": 46305
    },
    {
      "entropy": 5.930089902877808,
      "epoch": 4.6291797870745235,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030493266989234906,
      "loss": 5.3023,
      "mean_token_accuracy": 0.21458421051502227,
      "num_tokens": 90592007.0,
      "step": 46310
    },
    {
      "entropy": 5.944245004653931,
      "epoch": 4.629679612135752,
      "grad_norm": 1.125,
      "learning_rate": 0.00030489729763822374,
      "loss": 5.3627,
      "mean_token_accuracy": 0.20866645872592926,
      "num_tokens": 90601180.0,
      "step": 46315
    },
    {
      "entropy": 5.905314588546753,
      "epoch": 4.630179437196981,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030486192463189017,
      "loss": 5.3077,
      "mean_token_accuracy": 0.2127835139632225,
      "num_tokens": 90611227.0,
      "step": 46320
    },
    {
      "entropy": 5.920443439483643,
      "epoch": 4.63067926225821,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003048265508742381,
      "loss": 5.3762,
      "mean_token_accuracy": 0.21094487756490707,
      "num_tokens": 90621177.0,
      "step": 46325
    },
    {
      "entropy": 5.948083400726318,
      "epoch": 4.631179087319438,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030479117636615785,
      "loss": 5.3567,
      "mean_token_accuracy": 0.21198831647634506,
      "num_tokens": 90632256.0,
      "step": 46330
    },
    {
      "entropy": 5.952708292007446,
      "epoch": 4.631678912380667,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00030475580110853913,
      "loss": 5.2828,
      "mean_token_accuracy": 0.20924118310213088,
      "num_tokens": 90642096.0,
      "step": 46335
    },
    {
      "entropy": 5.878924798965454,
      "epoch": 4.632178737441896,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003047204251022722,
      "loss": 5.1893,
      "mean_token_accuracy": 0.22247953414916993,
      "num_tokens": 90651619.0,
      "step": 46340
    },
    {
      "entropy": 5.880587768554688,
      "epoch": 4.6326785625031235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000304685048348247,
      "loss": 5.3197,
      "mean_token_accuracy": 0.21138970404863358,
      "num_tokens": 90662126.0,
      "step": 46345
    },
    {
      "entropy": 5.996316289901733,
      "epoch": 4.633178387564352,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030464967084735357,
      "loss": 5.3307,
      "mean_token_accuracy": 0.209879569709301,
      "num_tokens": 90672732.0,
      "step": 46350
    },
    {
      "entropy": 5.926824760437012,
      "epoch": 4.633678212625581,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030461429260048213,
      "loss": 5.2759,
      "mean_token_accuracy": 0.21286593973636628,
      "num_tokens": 90683078.0,
      "step": 46355
    },
    {
      "entropy": 5.84903130531311,
      "epoch": 4.63417803768681,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030457891360852267,
      "loss": 5.2706,
      "mean_token_accuracy": 0.2247814416885376,
      "num_tokens": 90693741.0,
      "step": 46360
    },
    {
      "entropy": 5.910353565216065,
      "epoch": 4.634677862748038,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003045435338723653,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21462051719427108,
      "num_tokens": 90705122.0,
      "step": 46365
    },
    {
      "entropy": 5.895315742492675,
      "epoch": 4.635177687809267,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030450815339290026,
      "loss": 5.2988,
      "mean_token_accuracy": 0.21681384593248368,
      "num_tokens": 90715482.0,
      "step": 46370
    },
    {
      "entropy": 5.98322172164917,
      "epoch": 4.635677512870496,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030447277217101774,
      "loss": 5.3976,
      "mean_token_accuracy": 0.20509357750415802,
      "num_tokens": 90724506.0,
      "step": 46375
    },
    {
      "entropy": 5.980562162399292,
      "epoch": 4.6361773379317235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003044373902076079,
      "loss": 5.3684,
      "mean_token_accuracy": 0.2039617657661438,
      "num_tokens": 90734067.0,
      "step": 46380
    },
    {
      "entropy": 6.016969299316406,
      "epoch": 4.636677162992952,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030440200750356066,
      "loss": 5.3941,
      "mean_token_accuracy": 0.20832414627075196,
      "num_tokens": 90743911.0,
      "step": 46385
    },
    {
      "entropy": 5.843105983734131,
      "epoch": 4.637176988054181,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003043666240597667,
      "loss": 5.2319,
      "mean_token_accuracy": 0.21801613122224808,
      "num_tokens": 90753171.0,
      "step": 46390
    },
    {
      "entropy": 5.845906925201416,
      "epoch": 4.63767681311541,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003043312398771159,
      "loss": 5.2132,
      "mean_token_accuracy": 0.22148719280958176,
      "num_tokens": 90762684.0,
      "step": 46395
    },
    {
      "entropy": 5.8631758213043215,
      "epoch": 4.638176638176638,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030429585495649877,
      "loss": 5.2799,
      "mean_token_accuracy": 0.21239570379257203,
      "num_tokens": 90772390.0,
      "step": 46400
    },
    {
      "entropy": 5.874543809890747,
      "epoch": 4.638676463237867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003042604692988054,
      "loss": 5.3052,
      "mean_token_accuracy": 0.21005569249391556,
      "num_tokens": 90781244.0,
      "step": 46405
    },
    {
      "entropy": 5.983946514129639,
      "epoch": 4.639176288299096,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003042250829049261,
      "loss": 5.3692,
      "mean_token_accuracy": 0.2052116423845291,
      "num_tokens": 90791029.0,
      "step": 46410
    },
    {
      "entropy": 5.94920392036438,
      "epoch": 4.6396761133603235,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003041896957757513,
      "loss": 5.4053,
      "mean_token_accuracy": 0.19926839768886567,
      "num_tokens": 90801861.0,
      "step": 46415
    },
    {
      "entropy": 5.909753370285034,
      "epoch": 4.640175938421552,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003041543079121713,
      "loss": 5.2595,
      "mean_token_accuracy": 0.2093081921339035,
      "num_tokens": 90810847.0,
      "step": 46420
    },
    {
      "entropy": 5.940272521972656,
      "epoch": 4.640675763482781,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030411891931507635,
      "loss": 5.2973,
      "mean_token_accuracy": 0.20765940099954605,
      "num_tokens": 90820624.0,
      "step": 46425
    },
    {
      "entropy": 6.008355951309204,
      "epoch": 4.64117558854401,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030408352998535695,
      "loss": 5.3835,
      "mean_token_accuracy": 0.20225855559110642,
      "num_tokens": 90829975.0,
      "step": 46430
    },
    {
      "entropy": 5.897468090057373,
      "epoch": 4.641675413605238,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030404813992390337,
      "loss": 5.2857,
      "mean_token_accuracy": 0.2164342701435089,
      "num_tokens": 90838966.0,
      "step": 46435
    },
    {
      "entropy": 5.868356323242187,
      "epoch": 4.642175238666467,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003040127491316061,
      "loss": 5.29,
      "mean_token_accuracy": 0.21992080211639403,
      "num_tokens": 90847782.0,
      "step": 46440
    },
    {
      "entropy": 5.83433256149292,
      "epoch": 4.642675063727696,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030397735760935547,
      "loss": 5.2356,
      "mean_token_accuracy": 0.21487153321504593,
      "num_tokens": 90859640.0,
      "step": 46445
    },
    {
      "entropy": 5.961032056808472,
      "epoch": 4.6431748887889235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030394196535804205,
      "loss": 5.3965,
      "mean_token_accuracy": 0.20481764227151872,
      "num_tokens": 90868206.0,
      "step": 46450
    },
    {
      "entropy": 5.8770348072052006,
      "epoch": 4.643674713850152,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030390657237855606,
      "loss": 5.2564,
      "mean_token_accuracy": 0.2162618637084961,
      "num_tokens": 90877116.0,
      "step": 46455
    },
    {
      "entropy": 5.94220380783081,
      "epoch": 4.644174538911381,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003038711786717883,
      "loss": 5.3764,
      "mean_token_accuracy": 0.21393353193998338,
      "num_tokens": 90886346.0,
      "step": 46460
    },
    {
      "entropy": 5.937142372131348,
      "epoch": 4.64467436397261,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030383578423862914,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20432263016700744,
      "num_tokens": 90895794.0,
      "step": 46465
    },
    {
      "entropy": 5.873591613769531,
      "epoch": 4.645174189033838,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030380038907996905,
      "loss": 5.3065,
      "mean_token_accuracy": 0.21051767468452454,
      "num_tokens": 90906574.0,
      "step": 46470
    },
    {
      "entropy": 5.93496789932251,
      "epoch": 4.645674014095067,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003037649931966985,
      "loss": 5.3701,
      "mean_token_accuracy": 0.2099166363477707,
      "num_tokens": 90916228.0,
      "step": 46475
    },
    {
      "entropy": 5.923582935333252,
      "epoch": 4.646173839156296,
      "grad_norm": 1.125,
      "learning_rate": 0.0003037295965897082,
      "loss": 5.3041,
      "mean_token_accuracy": 0.21785597652196884,
      "num_tokens": 90926574.0,
      "step": 46480
    },
    {
      "entropy": 6.000755882263183,
      "epoch": 4.6466736642175235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003036941992598885,
      "loss": 5.362,
      "mean_token_accuracy": 0.21021779477596284,
      "num_tokens": 90937091.0,
      "step": 46485
    },
    {
      "entropy": 5.932620000839234,
      "epoch": 4.647173489278752,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003036588012081303,
      "loss": 5.2546,
      "mean_token_accuracy": 0.21709613800048827,
      "num_tokens": 90946037.0,
      "step": 46490
    },
    {
      "entropy": 5.903549575805664,
      "epoch": 4.647673314339981,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030362340243532393,
      "loss": 5.332,
      "mean_token_accuracy": 0.21279948204755783,
      "num_tokens": 90956323.0,
      "step": 46495
    },
    {
      "entropy": 6.028162431716919,
      "epoch": 4.64817313940121,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003035880029423602,
      "loss": 5.4468,
      "mean_token_accuracy": 0.2026719257235527,
      "num_tokens": 90965049.0,
      "step": 46500
    },
    {
      "entropy": 5.975199270248413,
      "epoch": 4.648672964462438,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003035526027301297,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20572523921728134,
      "num_tokens": 90974930.0,
      "step": 46505
    },
    {
      "entropy": 5.942128467559814,
      "epoch": 4.649172789523667,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030351720179952295,
      "loss": 5.3921,
      "mean_token_accuracy": 0.21138125360012056,
      "num_tokens": 90983969.0,
      "step": 46510
    },
    {
      "entropy": 5.827342414855957,
      "epoch": 4.649672614584896,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030348180015143077,
      "loss": 5.2392,
      "mean_token_accuracy": 0.22034142166376114,
      "num_tokens": 90993186.0,
      "step": 46515
    },
    {
      "entropy": 5.9320512294769285,
      "epoch": 4.6501724396461235,
      "grad_norm": 1.125,
      "learning_rate": 0.0003034463977867439,
      "loss": 5.3078,
      "mean_token_accuracy": 0.2141284942626953,
      "num_tokens": 91003136.0,
      "step": 46520
    },
    {
      "entropy": 5.952093839645386,
      "epoch": 4.650672264707352,
      "grad_norm": 1.0625,
      "learning_rate": 0.00030341099470635293,
      "loss": 5.3425,
      "mean_token_accuracy": 0.20696912258863448,
      "num_tokens": 91013118.0,
      "step": 46525
    },
    {
      "entropy": 5.756537199020386,
      "epoch": 4.651172089768581,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030337559091114856,
      "loss": 5.1237,
      "mean_token_accuracy": 0.23161277323961257,
      "num_tokens": 91023916.0,
      "step": 46530
    },
    {
      "entropy": 5.807536840438843,
      "epoch": 4.65167191482981,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030334018640202184,
      "loss": 5.1851,
      "mean_token_accuracy": 0.22157154977321625,
      "num_tokens": 91033217.0,
      "step": 46535
    },
    {
      "entropy": 5.987301921844482,
      "epoch": 4.652171739891038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030330478117986316,
      "loss": 5.4222,
      "mean_token_accuracy": 0.1990013226866722,
      "num_tokens": 91042858.0,
      "step": 46540
    },
    {
      "entropy": 5.927576160430908,
      "epoch": 4.652671564952267,
      "grad_norm": 1.046875,
      "learning_rate": 0.00030326937524556346,
      "loss": 5.3787,
      "mean_token_accuracy": 0.2130279242992401,
      "num_tokens": 91053218.0,
      "step": 46545
    },
    {
      "entropy": 5.9033458709716795,
      "epoch": 4.653171390013496,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030323396860001365,
      "loss": 5.3321,
      "mean_token_accuracy": 0.20173907279968262,
      "num_tokens": 91064046.0,
      "step": 46550
    },
    {
      "entropy": 5.936505508422852,
      "epoch": 4.6536712150747235,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003031985612441044,
      "loss": 5.3479,
      "mean_token_accuracy": 0.2152385249733925,
      "num_tokens": 91074163.0,
      "step": 46555
    },
    {
      "entropy": 5.970142412185669,
      "epoch": 4.654171040135952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030316315317872655,
      "loss": 5.3663,
      "mean_token_accuracy": 0.21235226839780807,
      "num_tokens": 91083477.0,
      "step": 46560
    },
    {
      "entropy": 5.80818452835083,
      "epoch": 4.654670865197181,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003031277444047711,
      "loss": 5.1825,
      "mean_token_accuracy": 0.2334347426891327,
      "num_tokens": 91092160.0,
      "step": 46565
    },
    {
      "entropy": 5.955813550949097,
      "epoch": 4.65517069025841,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003030923349231288,
      "loss": 5.4237,
      "mean_token_accuracy": 0.19985852092504502,
      "num_tokens": 91102400.0,
      "step": 46570
    },
    {
      "entropy": 5.964024448394776,
      "epoch": 4.655670515319638,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003030569247346906,
      "loss": 5.329,
      "mean_token_accuracy": 0.20591771155595778,
      "num_tokens": 91111749.0,
      "step": 46575
    },
    {
      "entropy": 5.926247596740723,
      "epoch": 4.656170340380867,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003030215138403474,
      "loss": 5.3261,
      "mean_token_accuracy": 0.21226766407489778,
      "num_tokens": 91121284.0,
      "step": 46580
    },
    {
      "entropy": 5.940508270263672,
      "epoch": 4.656670165442096,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00030298610224099,
      "loss": 5.328,
      "mean_token_accuracy": 0.20451659709215164,
      "num_tokens": 91131710.0,
      "step": 46585
    },
    {
      "entropy": 5.9238028049469,
      "epoch": 4.6571699905033235,
      "grad_norm": 1.125,
      "learning_rate": 0.00030295068993750965,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21125089675188063,
      "num_tokens": 91142382.0,
      "step": 46590
    },
    {
      "entropy": 5.865528917312622,
      "epoch": 4.657669815564552,
      "grad_norm": 1.125,
      "learning_rate": 0.00030291527693079704,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21780197322368622,
      "num_tokens": 91153127.0,
      "step": 46595
    },
    {
      "entropy": 5.866145229339599,
      "epoch": 4.658169640625781,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00030287986322174326,
      "loss": 5.3034,
      "mean_token_accuracy": 0.217154061794281,
      "num_tokens": 91162025.0,
      "step": 46600
    },
    {
      "entropy": 5.970807933807373,
      "epoch": 4.65866946568701,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00030284444881123934,
      "loss": 5.4139,
      "mean_token_accuracy": 0.20833584666252136,
      "num_tokens": 91171222.0,
      "step": 46605
    },
    {
      "entropy": 5.921177530288697,
      "epoch": 4.659169290748238,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003028090337001762,
      "loss": 5.3158,
      "mean_token_accuracy": 0.21581947803497314,
      "num_tokens": 91180075.0,
      "step": 46610
    },
    {
      "entropy": 5.923544263839721,
      "epoch": 4.659669115809467,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030277361788944495,
      "loss": 5.3076,
      "mean_token_accuracy": 0.21600010395050048,
      "num_tokens": 91188980.0,
      "step": 46615
    },
    {
      "entropy": 5.869872236251831,
      "epoch": 4.660168940870696,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003027382013799366,
      "loss": 5.2448,
      "mean_token_accuracy": 0.22168002277612686,
      "num_tokens": 91199032.0,
      "step": 46620
    },
    {
      "entropy": 5.894418239593506,
      "epoch": 4.6606687659319235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030270278417254224,
      "loss": 5.327,
      "mean_token_accuracy": 0.21092455834150314,
      "num_tokens": 91208406.0,
      "step": 46625
    },
    {
      "entropy": 5.934592151641846,
      "epoch": 4.661168590993152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000302667366268153,
      "loss": 5.3964,
      "mean_token_accuracy": 0.20954694151878356,
      "num_tokens": 91218565.0,
      "step": 46630
    },
    {
      "entropy": 5.903501796722412,
      "epoch": 4.661668416054381,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003026319476676599,
      "loss": 5.2856,
      "mean_token_accuracy": 0.21882229298353195,
      "num_tokens": 91227682.0,
      "step": 46635
    },
    {
      "entropy": 5.877967119216919,
      "epoch": 4.66216824111561,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030259652837195413,
      "loss": 5.2664,
      "mean_token_accuracy": 0.22215624153614044,
      "num_tokens": 91237539.0,
      "step": 46640
    },
    {
      "entropy": 5.978039741516113,
      "epoch": 4.662668066176838,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030256110838192677,
      "loss": 5.4329,
      "mean_token_accuracy": 0.2047516331076622,
      "num_tokens": 91247980.0,
      "step": 46645
    },
    {
      "entropy": 5.979646635055542,
      "epoch": 4.663167891238067,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003025256876984691,
      "loss": 5.342,
      "mean_token_accuracy": 0.21649424731731415,
      "num_tokens": 91257230.0,
      "step": 46650
    },
    {
      "entropy": 5.926923084259033,
      "epoch": 4.663667716299296,
      "grad_norm": 1.1875,
      "learning_rate": 0.00030249026632247214,
      "loss": 5.3,
      "mean_token_accuracy": 0.21351563781499863,
      "num_tokens": 91266019.0,
      "step": 46655
    },
    {
      "entropy": 5.819618988037109,
      "epoch": 4.6641675413605235,
      "grad_norm": 1.25,
      "learning_rate": 0.00030245484425482714,
      "loss": 5.1944,
      "mean_token_accuracy": 0.2235887363553047,
      "num_tokens": 91275337.0,
      "step": 46660
    },
    {
      "entropy": 5.904912853240967,
      "epoch": 4.664667366421752,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030241942149642533,
      "loss": 5.3253,
      "mean_token_accuracy": 0.20543072521686553,
      "num_tokens": 91286131.0,
      "step": 46665
    },
    {
      "entropy": 5.960052967071533,
      "epoch": 4.665167191482981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000302383998048158,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21231261938810347,
      "num_tokens": 91295781.0,
      "step": 46670
    },
    {
      "entropy": 5.911096858978271,
      "epoch": 4.66566701654421,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030234857391091627,
      "loss": 5.3128,
      "mean_token_accuracy": 0.21126004159450532,
      "num_tokens": 91305936.0,
      "step": 46675
    },
    {
      "entropy": 5.932642126083374,
      "epoch": 4.666166841605438,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030231314908559144,
      "loss": 5.352,
      "mean_token_accuracy": 0.20777006596326827,
      "num_tokens": 91316436.0,
      "step": 46680
    },
    {
      "entropy": 5.861638689041138,
      "epoch": 4.666666666666667,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003022777235730749,
      "loss": 5.2594,
      "mean_token_accuracy": 0.2215786650776863,
      "num_tokens": 91326969.0,
      "step": 46685
    },
    {
      "entropy": 5.961474943161011,
      "epoch": 4.667166491727896,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030224229737425774,
      "loss": 5.3614,
      "mean_token_accuracy": 0.20691573321819307,
      "num_tokens": 91335624.0,
      "step": 46690
    },
    {
      "entropy": 5.936259841918945,
      "epoch": 4.6676663167891235,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003022068704900314,
      "loss": 5.3124,
      "mean_token_accuracy": 0.20566235035657882,
      "num_tokens": 91345148.0,
      "step": 46695
    },
    {
      "entropy": 5.861268806457519,
      "epoch": 4.668166141850352,
      "grad_norm": 1.125,
      "learning_rate": 0.0003021714429212872,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21317061483860017,
      "num_tokens": 91354385.0,
      "step": 46700
    },
    {
      "entropy": 5.828505802154541,
      "epoch": 4.668665966911581,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030213601466891663,
      "loss": 5.2034,
      "mean_token_accuracy": 0.2212006315588951,
      "num_tokens": 91364610.0,
      "step": 46705
    },
    {
      "entropy": 5.9547388553619385,
      "epoch": 4.66916579197281,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003021005857338107,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20856361389160155,
      "num_tokens": 91374589.0,
      "step": 46710
    },
    {
      "entropy": 5.9521856784820555,
      "epoch": 4.669665617034038,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003020651561168612,
      "loss": 5.3775,
      "mean_token_accuracy": 0.2065439522266388,
      "num_tokens": 91383809.0,
      "step": 46715
    },
    {
      "entropy": 5.926552152633667,
      "epoch": 4.670165442095267,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030202972581895923,
      "loss": 5.3474,
      "mean_token_accuracy": 0.21693425327539445,
      "num_tokens": 91393171.0,
      "step": 46720
    },
    {
      "entropy": 5.939682102203369,
      "epoch": 4.670665267156496,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003019942948409963,
      "loss": 5.3563,
      "mean_token_accuracy": 0.21249259263277054,
      "num_tokens": 91403319.0,
      "step": 46725
    },
    {
      "entropy": 5.903029251098633,
      "epoch": 4.6711650922177235,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00030195886318386397,
      "loss": 5.3214,
      "mean_token_accuracy": 0.20865988731384277,
      "num_tokens": 91412316.0,
      "step": 46730
    },
    {
      "entropy": 5.9592029571533205,
      "epoch": 4.671664917278952,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00030192343084845343,
      "loss": 5.3558,
      "mean_token_accuracy": 0.2179841473698616,
      "num_tokens": 91421924.0,
      "step": 46735
    },
    {
      "entropy": 5.991092348098755,
      "epoch": 4.672164742340181,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003018879978356564,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21887896955013275,
      "num_tokens": 91431277.0,
      "step": 46740
    },
    {
      "entropy": 5.85684905052185,
      "epoch": 4.67266456740141,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00030185256414636425,
      "loss": 5.2658,
      "mean_token_accuracy": 0.22039713263511657,
      "num_tokens": 91441155.0,
      "step": 46745
    },
    {
      "entropy": 5.953279876708985,
      "epoch": 4.673164392462638,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00030181712978146846,
      "loss": 5.4004,
      "mean_token_accuracy": 0.1978810504078865,
      "num_tokens": 91450360.0,
      "step": 46750
    },
    {
      "entropy": 5.903624868392944,
      "epoch": 4.673664217523867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030178169474186055,
      "loss": 5.328,
      "mean_token_accuracy": 0.20932896584272384,
      "num_tokens": 91460125.0,
      "step": 46755
    },
    {
      "entropy": 5.879952478408813,
      "epoch": 4.674164042585096,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00030174625902843215,
      "loss": 5.3387,
      "mean_token_accuracy": 0.21099245250225068,
      "num_tokens": 91470646.0,
      "step": 46760
    },
    {
      "entropy": 5.895189523696899,
      "epoch": 4.6746638676463235,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003017108226420748,
      "loss": 5.2959,
      "mean_token_accuracy": 0.2141447901725769,
      "num_tokens": 91480390.0,
      "step": 46765
    },
    {
      "entropy": 5.8936480522155765,
      "epoch": 4.675163692707552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003016753855836799,
      "loss": 5.351,
      "mean_token_accuracy": 0.20268674790859223,
      "num_tokens": 91491433.0,
      "step": 46770
    },
    {
      "entropy": 5.968962097167969,
      "epoch": 4.675663517768781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003016399478541393,
      "loss": 5.344,
      "mean_token_accuracy": 0.20730580538511276,
      "num_tokens": 91500841.0,
      "step": 46775
    },
    {
      "entropy": 5.930877733230591,
      "epoch": 4.67616334283001,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030160450945434435,
      "loss": 5.2952,
      "mean_token_accuracy": 0.21118044555187226,
      "num_tokens": 91510964.0,
      "step": 46780
    },
    {
      "entropy": 5.968635892868042,
      "epoch": 4.676663167891238,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003015690703851869,
      "loss": 5.3108,
      "mean_token_accuracy": 0.20964843928813934,
      "num_tokens": 91520589.0,
      "step": 46785
    },
    {
      "entropy": 5.837492942810059,
      "epoch": 4.677162992952467,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003015336306475584,
      "loss": 5.2598,
      "mean_token_accuracy": 0.21575121730566024,
      "num_tokens": 91529280.0,
      "step": 46790
    },
    {
      "entropy": 5.928328800201416,
      "epoch": 4.677662818013696,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003014981902423506,
      "loss": 5.3494,
      "mean_token_accuracy": 0.20660273134708404,
      "num_tokens": 91537972.0,
      "step": 46795
    },
    {
      "entropy": 5.908085060119629,
      "epoch": 4.6781626430749235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003014627491704551,
      "loss": 5.3732,
      "mean_token_accuracy": 0.20855677276849746,
      "num_tokens": 91548020.0,
      "step": 46800
    },
    {
      "entropy": 5.890750360488892,
      "epoch": 4.678662468136152,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0003014273074327637,
      "loss": 5.1874,
      "mean_token_accuracy": 0.2168978676199913,
      "num_tokens": 91556731.0,
      "step": 46805
    },
    {
      "entropy": 5.9645607471466064,
      "epoch": 4.679162293197381,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003013918650301681,
      "loss": 5.3874,
      "mean_token_accuracy": 0.2121100202202797,
      "num_tokens": 91566323.0,
      "step": 46810
    },
    {
      "entropy": 5.8641441822052,
      "epoch": 4.67966211825861,
      "grad_norm": 1.109375,
      "learning_rate": 0.00030135642196355987,
      "loss": 5.2215,
      "mean_token_accuracy": 0.21691035479307175,
      "num_tokens": 91576221.0,
      "step": 46815
    },
    {
      "entropy": 5.805199241638183,
      "epoch": 4.680161943319838,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003013209782338309,
      "loss": 5.1203,
      "mean_token_accuracy": 0.22834620475769044,
      "num_tokens": 91586158.0,
      "step": 46820
    },
    {
      "entropy": 5.921179389953613,
      "epoch": 4.680661768381067,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003012855338418729,
      "loss": 5.3883,
      "mean_token_accuracy": 0.21009311079978943,
      "num_tokens": 91596278.0,
      "step": 46825
    },
    {
      "entropy": 5.87925968170166,
      "epoch": 4.681161593442296,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003012500887885775,
      "loss": 5.2486,
      "mean_token_accuracy": 0.21862830072641373,
      "num_tokens": 91606272.0,
      "step": 46830
    },
    {
      "entropy": 5.786139106750488,
      "epoch": 4.6816614185035235,
      "grad_norm": 1.15625,
      "learning_rate": 0.00030121464307483677,
      "loss": 5.1265,
      "mean_token_accuracy": 0.2268224000930786,
      "num_tokens": 91615468.0,
      "step": 46835
    },
    {
      "entropy": 5.852934741973877,
      "epoch": 4.682161243564752,
      "grad_norm": 1.125,
      "learning_rate": 0.0003011791967015423,
      "loss": 5.288,
      "mean_token_accuracy": 0.2174074500799179,
      "num_tokens": 91625395.0,
      "step": 46840
    },
    {
      "entropy": 5.858391284942627,
      "epoch": 4.682661068625981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00030114374966958604,
      "loss": 5.2258,
      "mean_token_accuracy": 0.2199304074048996,
      "num_tokens": 91634385.0,
      "step": 46845
    },
    {
      "entropy": 5.909961366653443,
      "epoch": 4.68316089368721,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003011083019798597,
      "loss": 5.2084,
      "mean_token_accuracy": 0.21805829256772996,
      "num_tokens": 91643870.0,
      "step": 46850
    },
    {
      "entropy": 5.900843906402588,
      "epoch": 4.683660718748438,
      "grad_norm": 1.125,
      "learning_rate": 0.0003010728536332552,
      "loss": 5.3467,
      "mean_token_accuracy": 0.20997948795557023,
      "num_tokens": 91653215.0,
      "step": 46855
    },
    {
      "entropy": 5.896719026565552,
      "epoch": 4.684160543809667,
      "grad_norm": 1.125,
      "learning_rate": 0.0003010374046306644,
      "loss": 5.2808,
      "mean_token_accuracy": 0.2178802952170372,
      "num_tokens": 91663276.0,
      "step": 46860
    },
    {
      "entropy": 5.994978523254394,
      "epoch": 4.684660368870896,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003010019549729793,
      "loss": 5.3926,
      "mean_token_accuracy": 0.1958774209022522,
      "num_tokens": 91671682.0,
      "step": 46865
    },
    {
      "entropy": 5.895542240142822,
      "epoch": 4.6851601939321235,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003009665046610916,
      "loss": 5.2749,
      "mean_token_accuracy": 0.2196480855345726,
      "num_tokens": 91680926.0,
      "step": 46870
    },
    {
      "entropy": 5.926209545135498,
      "epoch": 4.685660018993352,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003009310536958934,
      "loss": 5.316,
      "mean_token_accuracy": 0.2122906744480133,
      "num_tokens": 91690544.0,
      "step": 46875
    },
    {
      "entropy": 5.910192060470581,
      "epoch": 4.686159844054581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00030089560207827653,
      "loss": 5.331,
      "mean_token_accuracy": 0.2142694979906082,
      "num_tokens": 91700871.0,
      "step": 46880
    },
    {
      "entropy": 5.946951818466187,
      "epoch": 4.68665966911581,
      "grad_norm": 1.265625,
      "learning_rate": 0.00030086014980913294,
      "loss": 5.2948,
      "mean_token_accuracy": 0.21613363027572632,
      "num_tokens": 91710268.0,
      "step": 46885
    },
    {
      "entropy": 5.935460186004638,
      "epoch": 4.687159494177038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00030082469688935465,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21288712322711945,
      "num_tokens": 91720565.0,
      "step": 46890
    },
    {
      "entropy": 5.94823260307312,
      "epoch": 4.687659319238267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003007892433198336,
      "loss": 5.4243,
      "mean_token_accuracy": 0.20798580944538117,
      "num_tokens": 91730718.0,
      "step": 46895
    },
    {
      "entropy": 5.905328702926636,
      "epoch": 4.688159144299496,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000300753789101462,
      "loss": 5.2732,
      "mean_token_accuracy": 0.21766662150621413,
      "num_tokens": 91739890.0,
      "step": 46900
    },
    {
      "entropy": 5.91396656036377,
      "epoch": 4.6886589693607235,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003007183342351315,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21955904513597488,
      "num_tokens": 91749631.0,
      "step": 46905
    },
    {
      "entropy": 5.998407077789307,
      "epoch": 4.689158794421952,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003006828787217344,
      "loss": 5.4189,
      "mean_token_accuracy": 0.2007814183831215,
      "num_tokens": 91758930.0,
      "step": 46910
    },
    {
      "entropy": 5.793988227844238,
      "epoch": 4.689658619483181,
      "grad_norm": 1.09375,
      "learning_rate": 0.00030064742256216264,
      "loss": 5.207,
      "mean_token_accuracy": 0.2208620488643646,
      "num_tokens": 91768963.0,
      "step": 46915
    },
    {
      "entropy": 5.912706422805786,
      "epoch": 4.6901584445444096,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003006119657573084,
      "loss": 5.3465,
      "mean_token_accuracy": 0.20015559792518617,
      "num_tokens": 91778202.0,
      "step": 46920
    },
    {
      "entropy": 5.921932125091553,
      "epoch": 4.690658269605638,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00030057650830806365,
      "loss": 5.3481,
      "mean_token_accuracy": 0.21483950465917587,
      "num_tokens": 91788797.0,
      "step": 46925
    },
    {
      "entropy": 5.948334836959839,
      "epoch": 4.691158094666867,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030054105021532054,
      "loss": 5.3746,
      "mean_token_accuracy": 0.20722620338201522,
      "num_tokens": 91799135.0,
      "step": 46930
    },
    {
      "entropy": 5.907985067367553,
      "epoch": 4.691657919728096,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00030050559147997113,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22441240698099135,
      "num_tokens": 91808344.0,
      "step": 46935
    },
    {
      "entropy": 5.945448684692383,
      "epoch": 4.6921577447893235,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003004701321029076,
      "loss": 5.3395,
      "mean_token_accuracy": 0.20611186772584916,
      "num_tokens": 91817911.0,
      "step": 46940
    },
    {
      "entropy": 5.962925291061401,
      "epoch": 4.692657569850552,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00030043467208502216,
      "loss": 5.4648,
      "mean_token_accuracy": 0.19732718765735627,
      "num_tokens": 91828511.0,
      "step": 46945
    },
    {
      "entropy": 5.979320812225342,
      "epoch": 4.693157394911781,
      "grad_norm": 1.078125,
      "learning_rate": 0.000300399211427207,
      "loss": 5.3945,
      "mean_token_accuracy": 0.2104260042309761,
      "num_tokens": 91838493.0,
      "step": 46950
    },
    {
      "entropy": 5.957865810394287,
      "epoch": 4.6936572199730096,
      "grad_norm": 1.140625,
      "learning_rate": 0.00030036375013035404,
      "loss": 5.3248,
      "mean_token_accuracy": 0.2127443328499794,
      "num_tokens": 91849226.0,
      "step": 46955
    },
    {
      "entropy": 5.947462940216065,
      "epoch": 4.694157045034238,
      "grad_norm": 1.234375,
      "learning_rate": 0.00030032828819535575,
      "loss": 5.3517,
      "mean_token_accuracy": 0.2101467490196228,
      "num_tokens": 91858586.0,
      "step": 46960
    },
    {
      "entropy": 5.985937023162842,
      "epoch": 4.694656870095467,
      "grad_norm": 1.28125,
      "learning_rate": 0.00030029282562310424,
      "loss": 5.4114,
      "mean_token_accuracy": 0.20337785184383392,
      "num_tokens": 91867756.0,
      "step": 46965
    },
    {
      "entropy": 5.9941801071167,
      "epoch": 4.695156695156696,
      "grad_norm": 1.296875,
      "learning_rate": 0.0003002573624144917,
      "loss": 5.4546,
      "mean_token_accuracy": 0.1973084956407547,
      "num_tokens": 91878767.0,
      "step": 46970
    },
    {
      "entropy": 5.853880167007446,
      "epoch": 4.6956565202179235,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00030022189857041045,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21876170635223388,
      "num_tokens": 91888698.0,
      "step": 46975
    },
    {
      "entropy": 5.863966846466065,
      "epoch": 4.696156345279152,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003001864340917527,
      "loss": 5.2849,
      "mean_token_accuracy": 0.2124399170279503,
      "num_tokens": 91897465.0,
      "step": 46980
    },
    {
      "entropy": 5.959905481338501,
      "epoch": 4.696656170340381,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0003001509689794108,
      "loss": 5.3852,
      "mean_token_accuracy": 0.20951230376958846,
      "num_tokens": 91907594.0,
      "step": 46985
    },
    {
      "entropy": 5.94402666091919,
      "epoch": 4.6971559954016096,
      "grad_norm": 1.078125,
      "learning_rate": 0.00030011550323427694,
      "loss": 5.3771,
      "mean_token_accuracy": 0.20557720512151717,
      "num_tokens": 91918733.0,
      "step": 46990
    },
    {
      "entropy": 5.91992335319519,
      "epoch": 4.697655820462838,
      "grad_norm": 1.203125,
      "learning_rate": 0.00030008003685724353,
      "loss": 5.2557,
      "mean_token_accuracy": 0.2167016938328743,
      "num_tokens": 91928521.0,
      "step": 46995
    },
    {
      "entropy": 5.912286853790283,
      "epoch": 4.698155645524067,
      "grad_norm": 1.21875,
      "learning_rate": 0.00030004456984920283,
      "loss": 5.3227,
      "mean_token_accuracy": 0.21445380598306657,
      "num_tokens": 91938711.0,
      "step": 47000
    },
    {
      "entropy": 5.872951459884644,
      "epoch": 4.698655470585296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003000091022110472,
      "loss": 5.28,
      "mean_token_accuracy": 0.2167276993393898,
      "num_tokens": 91948103.0,
      "step": 47005
    },
    {
      "entropy": 5.968273591995239,
      "epoch": 4.6991552956465235,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002999736339436689,
      "loss": 5.3906,
      "mean_token_accuracy": 0.20311685800552368,
      "num_tokens": 91957204.0,
      "step": 47010
    },
    {
      "entropy": 5.9405347347259525,
      "epoch": 4.699655120707752,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029993816504796057,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20975107997655867,
      "num_tokens": 91966828.0,
      "step": 47015
    },
    {
      "entropy": 5.902144432067871,
      "epoch": 4.700154945768981,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0002999026955248143,
      "loss": 5.3333,
      "mean_token_accuracy": 0.211388997733593,
      "num_tokens": 91975379.0,
      "step": 47020
    },
    {
      "entropy": 5.934229516983033,
      "epoch": 4.7006547708302096,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002998672253751227,
      "loss": 5.333,
      "mean_token_accuracy": 0.20377584397792817,
      "num_tokens": 91984923.0,
      "step": 47025
    },
    {
      "entropy": 5.941374731063843,
      "epoch": 4.701154595891438,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029983175459977803,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2077582761645317,
      "num_tokens": 91996245.0,
      "step": 47030
    },
    {
      "entropy": 5.910394859313965,
      "epoch": 4.701654420952667,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002997962831996728,
      "loss": 5.2373,
      "mean_token_accuracy": 0.21890913546085358,
      "num_tokens": 92005741.0,
      "step": 47035
    },
    {
      "entropy": 5.975667953491211,
      "epoch": 4.702154246013895,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029976081117569957,
      "loss": 5.3478,
      "mean_token_accuracy": 0.21431318819522857,
      "num_tokens": 92015528.0,
      "step": 47040
    },
    {
      "entropy": 6.025884675979614,
      "epoch": 4.7026540710751235,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029972533852875064,
      "loss": 5.4909,
      "mean_token_accuracy": 0.19810211062431335,
      "num_tokens": 92026302.0,
      "step": 47045
    },
    {
      "entropy": 5.779359292984009,
      "epoch": 4.703153896136352,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00029968986525971854,
      "loss": 5.1146,
      "mean_token_accuracy": 0.23712856322526932,
      "num_tokens": 92035693.0,
      "step": 47050
    },
    {
      "entropy": 5.992165851593017,
      "epoch": 4.703653721197581,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002996543913694958,
      "loss": 5.4169,
      "mean_token_accuracy": 0.20131695419549941,
      "num_tokens": 92044633.0,
      "step": 47055
    },
    {
      "entropy": 5.87149395942688,
      "epoch": 4.7041535462588095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029961891685897484,
      "loss": 5.272,
      "mean_token_accuracy": 0.21372276097536086,
      "num_tokens": 92053869.0,
      "step": 47060
    },
    {
      "entropy": 5.870800447463989,
      "epoch": 4.704653371320038,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002995834417290484,
      "loss": 5.2947,
      "mean_token_accuracy": 0.20690253376960754,
      "num_tokens": 92062713.0,
      "step": 47065
    },
    {
      "entropy": 5.905218172073364,
      "epoch": 4.705153196381267,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002995479659806088,
      "loss": 5.303,
      "mean_token_accuracy": 0.21388487964868547,
      "num_tokens": 92072846.0,
      "step": 47070
    },
    {
      "entropy": 5.906236410140991,
      "epoch": 4.705653021442495,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029951248961454876,
      "loss": 5.3212,
      "mean_token_accuracy": 0.21768362671136857,
      "num_tokens": 92081523.0,
      "step": 47075
    },
    {
      "entropy": 5.975221824645996,
      "epoch": 4.7061528465037235,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029947701263176074,
      "loss": 5.3253,
      "mean_token_accuracy": 0.21218984723091125,
      "num_tokens": 92091287.0,
      "step": 47080
    },
    {
      "entropy": 5.978882360458374,
      "epoch": 4.706652671564952,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002994415350331373,
      "loss": 5.3318,
      "mean_token_accuracy": 0.20686050355434418,
      "num_tokens": 92101459.0,
      "step": 47085
    },
    {
      "entropy": 5.927949857711792,
      "epoch": 4.707152496626181,
      "grad_norm": 1.125,
      "learning_rate": 0.0002994060568195713,
      "loss": 5.2926,
      "mean_token_accuracy": 0.21220371574163438,
      "num_tokens": 92110232.0,
      "step": 47090
    },
    {
      "entropy": 5.8315582275390625,
      "epoch": 4.7076523216874095,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002993705779919551,
      "loss": 5.2193,
      "mean_token_accuracy": 0.22605021446943283,
      "num_tokens": 92120175.0,
      "step": 47095
    },
    {
      "entropy": 5.902080011367798,
      "epoch": 4.708152146748638,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002993350985511814,
      "loss": 5.3101,
      "mean_token_accuracy": 0.21692576259374619,
      "num_tokens": 92129085.0,
      "step": 47100
    },
    {
      "entropy": 5.935134363174439,
      "epoch": 4.708651971809866,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002992996184981429,
      "loss": 5.3486,
      "mean_token_accuracy": 0.21347821205854417,
      "num_tokens": 92139762.0,
      "step": 47105
    },
    {
      "entropy": 5.933944225311279,
      "epoch": 4.709151796871095,
      "grad_norm": 1.40625,
      "learning_rate": 0.00029926413783373237,
      "loss": 5.3263,
      "mean_token_accuracy": 0.21867117583751677,
      "num_tokens": 92149888.0,
      "step": 47110
    },
    {
      "entropy": 5.895630502700806,
      "epoch": 4.7096516219323235,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002992286565588422,
      "loss": 5.3261,
      "mean_token_accuracy": 0.2070007622241974,
      "num_tokens": 92160617.0,
      "step": 47115
    },
    {
      "entropy": 5.9826469898223875,
      "epoch": 4.710151446993552,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002991931746743653,
      "loss": 5.4314,
      "mean_token_accuracy": 0.2010769724845886,
      "num_tokens": 92171439.0,
      "step": 47120
    },
    {
      "entropy": 5.914934778213501,
      "epoch": 4.710651272054781,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002991576921811945,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21287021338939666,
      "num_tokens": 92180285.0,
      "step": 47125
    },
    {
      "entropy": 5.923205232620239,
      "epoch": 4.7111510971160095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029912220908022224,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20463513880968093,
      "num_tokens": 92189848.0,
      "step": 47130
    },
    {
      "entropy": 5.842607307434082,
      "epoch": 4.711650922177238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029908672537234146,
      "loss": 5.1863,
      "mean_token_accuracy": 0.23305098563432694,
      "num_tokens": 92200856.0,
      "step": 47135
    },
    {
      "entropy": 5.9507440567016605,
      "epoch": 4.712150747238466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029905124105844485,
      "loss": 5.314,
      "mean_token_accuracy": 0.21498293429613113,
      "num_tokens": 92210548.0,
      "step": 47140
    },
    {
      "entropy": 5.883554458618164,
      "epoch": 4.712650572299695,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002990157561394252,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21092554181814194,
      "num_tokens": 92219742.0,
      "step": 47145
    },
    {
      "entropy": 5.853910636901856,
      "epoch": 4.7131503973609234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029898027061617526,
      "loss": 5.3218,
      "mean_token_accuracy": 0.2126156911253929,
      "num_tokens": 92229127.0,
      "step": 47150
    },
    {
      "entropy": 5.8982024669647215,
      "epoch": 4.713650222422152,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002989447844895879,
      "loss": 5.3237,
      "mean_token_accuracy": 0.21869784593582153,
      "num_tokens": 92239730.0,
      "step": 47155
    },
    {
      "entropy": 5.956342506408691,
      "epoch": 4.714150047483381,
      "grad_norm": 1.078125,
      "learning_rate": 0.000298909297760556,
      "loss": 5.2918,
      "mean_token_accuracy": 0.21013015061616896,
      "num_tokens": 92249879.0,
      "step": 47160
    },
    {
      "entropy": 5.876937437057495,
      "epoch": 4.7146498725446095,
      "grad_norm": 1.25,
      "learning_rate": 0.00029887381042997227,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21871360391378403,
      "num_tokens": 92259487.0,
      "step": 47165
    },
    {
      "entropy": 5.894783687591553,
      "epoch": 4.715149697605838,
      "grad_norm": 1.125,
      "learning_rate": 0.0002988383224987296,
      "loss": 5.2804,
      "mean_token_accuracy": 0.21304874122142792,
      "num_tokens": 92268865.0,
      "step": 47170
    },
    {
      "entropy": 5.94210205078125,
      "epoch": 4.715649522667066,
      "grad_norm": 1.25,
      "learning_rate": 0.00029880283396772087,
      "loss": 5.3274,
      "mean_token_accuracy": 0.21135850697755815,
      "num_tokens": 92277572.0,
      "step": 47175
    },
    {
      "entropy": 5.852270221710205,
      "epoch": 4.716149347728295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029876734483783897,
      "loss": 5.2725,
      "mean_token_accuracy": 0.21793339699506759,
      "num_tokens": 92287847.0,
      "step": 47180
    },
    {
      "entropy": 5.919137096405029,
      "epoch": 4.7166491727895234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029873185510997686,
      "loss": 5.368,
      "mean_token_accuracy": 0.20989506244659423,
      "num_tokens": 92297530.0,
      "step": 47185
    },
    {
      "entropy": 5.906767559051514,
      "epoch": 4.717148997850752,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029869636478502726,
      "loss": 5.3497,
      "mean_token_accuracy": 0.20613878220319748,
      "num_tokens": 92306796.0,
      "step": 47190
    },
    {
      "entropy": 5.893945837020874,
      "epoch": 4.717648822911981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029866087386388334,
      "loss": 5.2333,
      "mean_token_accuracy": 0.22317758649587632,
      "num_tokens": 92316666.0,
      "step": 47195
    },
    {
      "entropy": 5.9544977188110355,
      "epoch": 4.7181486479732095,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00029862538234743784,
      "loss": 5.3376,
      "mean_token_accuracy": 0.20998909771442414,
      "num_tokens": 92324898.0,
      "step": 47200
    },
    {
      "entropy": 5.895790863037109,
      "epoch": 4.718648473034438,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029858989023658376,
      "loss": 5.3439,
      "mean_token_accuracy": 0.21230154782533645,
      "num_tokens": 92334006.0,
      "step": 47205
    },
    {
      "entropy": 5.93406457901001,
      "epoch": 4.719148298095666,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002985543975322142,
      "loss": 5.2999,
      "mean_token_accuracy": 0.21861447691917418,
      "num_tokens": 92343768.0,
      "step": 47210
    },
    {
      "entropy": 5.947649002075195,
      "epoch": 4.719648123156895,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000298518904235222,
      "loss": 5.3548,
      "mean_token_accuracy": 0.20567528307437896,
      "num_tokens": 92352981.0,
      "step": 47215
    },
    {
      "entropy": 5.956731653213501,
      "epoch": 4.7201479482181234,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029848341034650026,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21002081483602525,
      "num_tokens": 92362957.0,
      "step": 47220
    },
    {
      "entropy": 5.971572303771973,
      "epoch": 4.720647773279352,
      "grad_norm": 1.0625,
      "learning_rate": 0.000298447915866942,
      "loss": 5.3958,
      "mean_token_accuracy": 0.2074788823723793,
      "num_tokens": 92373560.0,
      "step": 47225
    },
    {
      "entropy": 5.975148105621338,
      "epoch": 4.721147598340581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002984124207974402,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20831928849220277,
      "num_tokens": 92384227.0,
      "step": 47230
    },
    {
      "entropy": 5.935011529922486,
      "epoch": 4.7216474234018095,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029837692513888785,
      "loss": 5.3114,
      "mean_token_accuracy": 0.21015432924032212,
      "num_tokens": 92394827.0,
      "step": 47235
    },
    {
      "entropy": 5.967491722106933,
      "epoch": 4.722147248463038,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002983414288921781,
      "loss": 5.4292,
      "mean_token_accuracy": 0.2042316421866417,
      "num_tokens": 92404830.0,
      "step": 47240
    },
    {
      "entropy": 5.9634106159210205,
      "epoch": 4.722647073524266,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029830593205820407,
      "loss": 5.3789,
      "mean_token_accuracy": 0.21096863299608232,
      "num_tokens": 92413900.0,
      "step": 47245
    },
    {
      "entropy": 6.002840185165406,
      "epoch": 4.723146898585495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002982704346378587,
      "loss": 5.4183,
      "mean_token_accuracy": 0.19725316017866135,
      "num_tokens": 92423654.0,
      "step": 47250
    },
    {
      "entropy": 5.91124267578125,
      "epoch": 4.7236467236467234,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00029823493663203515,
      "loss": 5.2873,
      "mean_token_accuracy": 0.2166161611676216,
      "num_tokens": 92433637.0,
      "step": 47255
    },
    {
      "entropy": 5.89517993927002,
      "epoch": 4.724146548707952,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002981994380416267,
      "loss": 5.2803,
      "mean_token_accuracy": 0.21892447173595428,
      "num_tokens": 92444232.0,
      "step": 47260
    },
    {
      "entropy": 5.974546003341675,
      "epoch": 4.724646373769181,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002981639388675263,
      "loss": 5.3972,
      "mean_token_accuracy": 0.20745883733034134,
      "num_tokens": 92453992.0,
      "step": 47265
    },
    {
      "entropy": 5.963426017761231,
      "epoch": 4.7251461988304095,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002981284391106271,
      "loss": 5.3287,
      "mean_token_accuracy": 0.20848190039396286,
      "num_tokens": 92463225.0,
      "step": 47270
    },
    {
      "entropy": 5.921622371673584,
      "epoch": 4.725646023891638,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002980929387718224,
      "loss": 5.3539,
      "mean_token_accuracy": 0.20607620477676392,
      "num_tokens": 92471632.0,
      "step": 47275
    },
    {
      "entropy": 5.856128168106079,
      "epoch": 4.726145848952866,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002980574378520053,
      "loss": 5.2696,
      "mean_token_accuracy": 0.21259932965040207,
      "num_tokens": 92480532.0,
      "step": 47280
    },
    {
      "entropy": 5.923752355575561,
      "epoch": 4.726645674014095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029802193635206895,
      "loss": 5.3425,
      "mean_token_accuracy": 0.20835408419370652,
      "num_tokens": 92490851.0,
      "step": 47285
    },
    {
      "entropy": 5.85295615196228,
      "epoch": 4.727145499075323,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002979864342729067,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22607093900442124,
      "num_tokens": 92501001.0,
      "step": 47290
    },
    {
      "entropy": 5.946095657348633,
      "epoch": 4.727645324136552,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029795093161541156,
      "loss": 5.3144,
      "mean_token_accuracy": 0.21834306567907333,
      "num_tokens": 92511075.0,
      "step": 47295
    },
    {
      "entropy": 5.953096675872803,
      "epoch": 4.728145149197781,
      "grad_norm": 1.1875,
      "learning_rate": 0.000297915428380477,
      "loss": 5.3296,
      "mean_token_accuracy": 0.2061417818069458,
      "num_tokens": 92520589.0,
      "step": 47300
    },
    {
      "entropy": 5.966013431549072,
      "epoch": 4.7286449742590095,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029787992456899604,
      "loss": 5.3609,
      "mean_token_accuracy": 0.21025815159082412,
      "num_tokens": 92529693.0,
      "step": 47305
    },
    {
      "entropy": 5.858836936950683,
      "epoch": 4.729144799320238,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002978444201818621,
      "loss": 5.2486,
      "mean_token_accuracy": 0.2202352315187454,
      "num_tokens": 92538716.0,
      "step": 47310
    },
    {
      "entropy": 5.9364090919494625,
      "epoch": 4.729644624381466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029780891521996843,
      "loss": 5.3927,
      "mean_token_accuracy": 0.2054824560880661,
      "num_tokens": 92548282.0,
      "step": 47315
    },
    {
      "entropy": 5.9087189674377445,
      "epoch": 4.730144449442695,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002977734096842084,
      "loss": 5.2631,
      "mean_token_accuracy": 0.2161826401948929,
      "num_tokens": 92558286.0,
      "step": 47320
    },
    {
      "entropy": 5.959177112579345,
      "epoch": 4.730644274503923,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029773790357547514,
      "loss": 5.3596,
      "mean_token_accuracy": 0.20834974199533463,
      "num_tokens": 92568537.0,
      "step": 47325
    },
    {
      "entropy": 5.934364175796508,
      "epoch": 4.731144099565152,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029770239689466216,
      "loss": 5.4116,
      "mean_token_accuracy": 0.20052585899829864,
      "num_tokens": 92579376.0,
      "step": 47330
    },
    {
      "entropy": 5.860497283935547,
      "epoch": 4.731643924626381,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029766688964266263,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21285671144723892,
      "num_tokens": 92589686.0,
      "step": 47335
    },
    {
      "entropy": 5.895515489578247,
      "epoch": 4.7321437496876095,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002976313818203701,
      "loss": 5.2945,
      "mean_token_accuracy": 0.21451304256916046,
      "num_tokens": 92599146.0,
      "step": 47340
    },
    {
      "entropy": 5.856056833267212,
      "epoch": 4.732643574748838,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002975958734286777,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21223331093788148,
      "num_tokens": 92608822.0,
      "step": 47345
    },
    {
      "entropy": 5.846861696243286,
      "epoch": 4.733143399810066,
      "grad_norm": 1.03125,
      "learning_rate": 0.000297560364468479,
      "loss": 5.2877,
      "mean_token_accuracy": 0.22073679864406587,
      "num_tokens": 92618863.0,
      "step": 47350
    },
    {
      "entropy": 5.944500875473023,
      "epoch": 4.733643224871295,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002975248549406673,
      "loss": 5.3004,
      "mean_token_accuracy": 0.21806042045354843,
      "num_tokens": 92628904.0,
      "step": 47355
    },
    {
      "entropy": 5.925597333908081,
      "epoch": 4.734143049932523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002974893448461361,
      "loss": 5.2918,
      "mean_token_accuracy": 0.216277414560318,
      "num_tokens": 92638330.0,
      "step": 47360
    },
    {
      "entropy": 5.983364057540894,
      "epoch": 4.734642874993752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029745383418577876,
      "loss": 5.3978,
      "mean_token_accuracy": 0.2057942643761635,
      "num_tokens": 92649621.0,
      "step": 47365
    },
    {
      "entropy": 5.897803258895874,
      "epoch": 4.735142700054981,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002974183229604888,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21740195900201797,
      "num_tokens": 92658934.0,
      "step": 47370
    },
    {
      "entropy": 5.969980239868164,
      "epoch": 4.7356425251162095,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029738281117115944,
      "loss": 5.3442,
      "mean_token_accuracy": 0.20431675761938095,
      "num_tokens": 92668077.0,
      "step": 47375
    },
    {
      "entropy": 5.919949722290039,
      "epoch": 4.736142350177438,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002973472988186844,
      "loss": 5.2897,
      "mean_token_accuracy": 0.21641198992729188,
      "num_tokens": 92679170.0,
      "step": 47380
    },
    {
      "entropy": 5.974450016021729,
      "epoch": 4.736642175238666,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002973117859039571,
      "loss": 5.3946,
      "mean_token_accuracy": 0.2060984954237938,
      "num_tokens": 92688411.0,
      "step": 47385
    },
    {
      "entropy": 5.90905442237854,
      "epoch": 4.737142000299895,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029727627242787097,
      "loss": 5.315,
      "mean_token_accuracy": 0.21261193603277206,
      "num_tokens": 92698152.0,
      "step": 47390
    },
    {
      "entropy": 5.882533025741577,
      "epoch": 4.737641825361123,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029724075839131957,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21858099848031998,
      "num_tokens": 92707746.0,
      "step": 47395
    },
    {
      "entropy": 5.935127925872803,
      "epoch": 4.738141650422352,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002972052437951963,
      "loss": 5.3317,
      "mean_token_accuracy": 0.20559089183807372,
      "num_tokens": 92717543.0,
      "step": 47400
    },
    {
      "entropy": 5.923886394500732,
      "epoch": 4.738641475483581,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029716972864039505,
      "loss": 5.271,
      "mean_token_accuracy": 0.2099712535738945,
      "num_tokens": 92727708.0,
      "step": 47405
    },
    {
      "entropy": 5.920446348190308,
      "epoch": 4.7391413005448095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029713421292780896,
      "loss": 5.3151,
      "mean_token_accuracy": 0.2090280681848526,
      "num_tokens": 92737952.0,
      "step": 47410
    },
    {
      "entropy": 5.9237961769104,
      "epoch": 4.739641125606038,
      "grad_norm": 1.09375,
      "learning_rate": 0.00029709869665833186,
      "loss": 5.3054,
      "mean_token_accuracy": 0.21157119870185853,
      "num_tokens": 92748760.0,
      "step": 47415
    },
    {
      "entropy": 5.909577178955078,
      "epoch": 4.740140950667266,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029706317983285713,
      "loss": 5.3131,
      "mean_token_accuracy": 0.2132699966430664,
      "num_tokens": 92758461.0,
      "step": 47420
    },
    {
      "entropy": 5.993689060211182,
      "epoch": 4.740640775728495,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002970276624522785,
      "loss": 5.3277,
      "mean_token_accuracy": 0.2093439668416977,
      "num_tokens": 92768027.0,
      "step": 47425
    },
    {
      "entropy": 5.865153217315674,
      "epoch": 4.741140600789723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00029699214451748955,
      "loss": 5.2646,
      "mean_token_accuracy": 0.21355636268854142,
      "num_tokens": 92778461.0,
      "step": 47430
    },
    {
      "entropy": 5.864753389358521,
      "epoch": 4.741640425850952,
      "grad_norm": 1.296875,
      "learning_rate": 0.00029695662602938395,
      "loss": 5.216,
      "mean_token_accuracy": 0.22314897030591965,
      "num_tokens": 92787897.0,
      "step": 47435
    },
    {
      "entropy": 5.798651266098022,
      "epoch": 4.742140250912181,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029692110698885534,
      "loss": 5.207,
      "mean_token_accuracy": 0.2195616438984871,
      "num_tokens": 92797061.0,
      "step": 47440
    },
    {
      "entropy": 5.898748970031738,
      "epoch": 4.7426400759734095,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002968855873967972,
      "loss": 5.285,
      "mean_token_accuracy": 0.21894439607858657,
      "num_tokens": 92806107.0,
      "step": 47445
    },
    {
      "entropy": 5.928333425521851,
      "epoch": 4.743139901034638,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002968500672541034,
      "loss": 5.2743,
      "mean_token_accuracy": 0.2190675601363182,
      "num_tokens": 92815331.0,
      "step": 47450
    },
    {
      "entropy": 5.966222333908081,
      "epoch": 4.743639726095866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029681454656166746,
      "loss": 5.3312,
      "mean_token_accuracy": 0.2082212299108505,
      "num_tokens": 92824742.0,
      "step": 47455
    },
    {
      "entropy": 6.004625749588013,
      "epoch": 4.744139551157095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029677902532038316,
      "loss": 5.4367,
      "mean_token_accuracy": 0.2027524843811989,
      "num_tokens": 92834807.0,
      "step": 47460
    },
    {
      "entropy": 5.958695936203003,
      "epoch": 4.744639376218323,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002967435035311442,
      "loss": 5.3256,
      "mean_token_accuracy": 0.2174132153391838,
      "num_tokens": 92844778.0,
      "step": 47465
    },
    {
      "entropy": 5.875651741027832,
      "epoch": 4.745139201279552,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002967079811948443,
      "loss": 5.32,
      "mean_token_accuracy": 0.21053240001201629,
      "num_tokens": 92853992.0,
      "step": 47470
    },
    {
      "entropy": 5.952959871292114,
      "epoch": 4.745639026340781,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002966724583123772,
      "loss": 5.4611,
      "mean_token_accuracy": 0.1959928587079048,
      "num_tokens": 92863872.0,
      "step": 47475
    },
    {
      "entropy": 5.915010023117065,
      "epoch": 4.7461388514020095,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002966369348846366,
      "loss": 5.3186,
      "mean_token_accuracy": 0.21090490818023683,
      "num_tokens": 92872921.0,
      "step": 47480
    },
    {
      "entropy": 5.95894079208374,
      "epoch": 4.746638676463238,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029660141091251627,
      "loss": 5.3424,
      "mean_token_accuracy": 0.21616947501897812,
      "num_tokens": 92882171.0,
      "step": 47485
    },
    {
      "entropy": 5.804140424728393,
      "epoch": 4.747138501524466,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002965658863969099,
      "loss": 5.149,
      "mean_token_accuracy": 0.22155124545097352,
      "num_tokens": 92894341.0,
      "step": 47490
    },
    {
      "entropy": 5.771945381164551,
      "epoch": 4.747638326585695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029653036133871154,
      "loss": 5.1173,
      "mean_token_accuracy": 0.23936618119478226,
      "num_tokens": 92904786.0,
      "step": 47495
    },
    {
      "entropy": 5.873742198944091,
      "epoch": 4.748138151646923,
      "grad_norm": 1.125,
      "learning_rate": 0.0002964948357388148,
      "loss": 5.3654,
      "mean_token_accuracy": 0.20876393467187881,
      "num_tokens": 92914212.0,
      "step": 47500
    },
    {
      "entropy": 5.988464593887329,
      "epoch": 4.748637976708152,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002964593095981135,
      "loss": 5.4997,
      "mean_token_accuracy": 0.19967336505651473,
      "num_tokens": 92923144.0,
      "step": 47505
    },
    {
      "entropy": 5.925479078292847,
      "epoch": 4.749137801769381,
      "grad_norm": 1.125,
      "learning_rate": 0.0002964237829175015,
      "loss": 5.3304,
      "mean_token_accuracy": 0.21179707497358322,
      "num_tokens": 92934252.0,
      "step": 47510
    },
    {
      "entropy": 5.919706964492798,
      "epoch": 4.7496376268306095,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002963882556978726,
      "loss": 5.324,
      "mean_token_accuracy": 0.21269208043813706,
      "num_tokens": 92943955.0,
      "step": 47515
    },
    {
      "entropy": 5.923021554946899,
      "epoch": 4.750137451891838,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029635272794012067,
      "loss": 5.3226,
      "mean_token_accuracy": 0.20852981507778168,
      "num_tokens": 92953663.0,
      "step": 47520
    },
    {
      "entropy": 5.881335878372193,
      "epoch": 4.750637276953066,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002963171996451397,
      "loss": 5.2787,
      "mean_token_accuracy": 0.2178661361336708,
      "num_tokens": 92963840.0,
      "step": 47525
    },
    {
      "entropy": 5.905475330352783,
      "epoch": 4.751137102014295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002962816708138234,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21324069052934647,
      "num_tokens": 92973744.0,
      "step": 47530
    },
    {
      "entropy": 5.989779853820801,
      "epoch": 4.751636927075523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002962461414470658,
      "loss": 5.3926,
      "mean_token_accuracy": 0.20678350031375886,
      "num_tokens": 92983727.0,
      "step": 47535
    },
    {
      "entropy": 5.872087860107422,
      "epoch": 4.752136752136752,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002962106115457607,
      "loss": 5.2633,
      "mean_token_accuracy": 0.2085179790854454,
      "num_tokens": 92993833.0,
      "step": 47540
    },
    {
      "entropy": 5.924894666671753,
      "epoch": 4.752636577197981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000296175081110802,
      "loss": 5.3461,
      "mean_token_accuracy": 0.20308481603860856,
      "num_tokens": 93003715.0,
      "step": 47545
    },
    {
      "entropy": 5.802413368225098,
      "epoch": 4.7531364022592095,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029613955014308387,
      "loss": 5.2283,
      "mean_token_accuracy": 0.21855124980211257,
      "num_tokens": 93013761.0,
      "step": 47550
    },
    {
      "entropy": 5.906579637527466,
      "epoch": 4.753636227320438,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029610401864349993,
      "loss": 5.264,
      "mean_token_accuracy": 0.22263456732034684,
      "num_tokens": 93022609.0,
      "step": 47555
    },
    {
      "entropy": 5.943468236923218,
      "epoch": 4.754136052381666,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029606848661294446,
      "loss": 5.2722,
      "mean_token_accuracy": 0.21768531650304795,
      "num_tokens": 93031164.0,
      "step": 47560
    },
    {
      "entropy": 5.895484733581543,
      "epoch": 4.754635877442895,
      "grad_norm": 1.25,
      "learning_rate": 0.00029603295405231117,
      "loss": 5.3446,
      "mean_token_accuracy": 0.21205396354198455,
      "num_tokens": 93041102.0,
      "step": 47565
    },
    {
      "entropy": 5.959145784378052,
      "epoch": 4.755135702504123,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002959974209624943,
      "loss": 5.317,
      "mean_token_accuracy": 0.20845882892608641,
      "num_tokens": 93050940.0,
      "step": 47570
    },
    {
      "entropy": 5.98532338142395,
      "epoch": 4.755635527565352,
      "grad_norm": 1.125,
      "learning_rate": 0.00029596188734438755,
      "loss": 5.3491,
      "mean_token_accuracy": 0.21108904778957366,
      "num_tokens": 93060150.0,
      "step": 47575
    },
    {
      "entropy": 5.885281801223755,
      "epoch": 4.756135352626581,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002959263531988853,
      "loss": 5.2838,
      "mean_token_accuracy": 0.21816509664058686,
      "num_tokens": 93069072.0,
      "step": 47580
    },
    {
      "entropy": 5.8898073673248295,
      "epoch": 4.7566351776878095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002958908185268812,
      "loss": 5.375,
      "mean_token_accuracy": 0.20126451700925826,
      "num_tokens": 93080224.0,
      "step": 47585
    },
    {
      "entropy": 5.938671827316284,
      "epoch": 4.757135002749038,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002958552833292696,
      "loss": 5.276,
      "mean_token_accuracy": 0.2139037236571312,
      "num_tokens": 93089779.0,
      "step": 47590
    },
    {
      "entropy": 5.816318225860596,
      "epoch": 4.757634827810266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002958197476069444,
      "loss": 5.151,
      "mean_token_accuracy": 0.22875814139842987,
      "num_tokens": 93099038.0,
      "step": 47595
    },
    {
      "entropy": 5.784551382064819,
      "epoch": 4.758134652871495,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002957842113607997,
      "loss": 5.2365,
      "mean_token_accuracy": 0.2260700136423111,
      "num_tokens": 93109618.0,
      "step": 47600
    },
    {
      "entropy": 5.907494068145752,
      "epoch": 4.758634477932723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002957486745917296,
      "loss": 5.3397,
      "mean_token_accuracy": 0.22036864012479782,
      "num_tokens": 93118955.0,
      "step": 47605
    },
    {
      "entropy": 5.954043817520142,
      "epoch": 4.759134302993952,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002957131373006282,
      "loss": 5.3138,
      "mean_token_accuracy": 0.217591592669487,
      "num_tokens": 93128809.0,
      "step": 47610
    },
    {
      "entropy": 6.022547674179077,
      "epoch": 4.759634128055181,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029567759948838964,
      "loss": 5.3883,
      "mean_token_accuracy": 0.20199509412050248,
      "num_tokens": 93138616.0,
      "step": 47615
    },
    {
      "entropy": 5.921207332611084,
      "epoch": 4.7601339531164095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002956420611559079,
      "loss": 5.2722,
      "mean_token_accuracy": 0.21986999809741975,
      "num_tokens": 93148112.0,
      "step": 47620
    },
    {
      "entropy": 5.8730635166168215,
      "epoch": 4.760633778177638,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029560652230407723,
      "loss": 5.2989,
      "mean_token_accuracy": 0.21627179980278016,
      "num_tokens": 93159053.0,
      "step": 47625
    },
    {
      "entropy": 5.9278980731964115,
      "epoch": 4.761133603238866,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029557098293379186,
      "loss": 5.2509,
      "mean_token_accuracy": 0.2239125117659569,
      "num_tokens": 93169204.0,
      "step": 47630
    },
    {
      "entropy": 5.94557294845581,
      "epoch": 4.761633428300095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002955354430459457,
      "loss": 5.2981,
      "mean_token_accuracy": 0.21738902032375335,
      "num_tokens": 93178034.0,
      "step": 47635
    },
    {
      "entropy": 5.878702592849732,
      "epoch": 4.762133253361323,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029549990264143323,
      "loss": 5.2772,
      "mean_token_accuracy": 0.21961890608072282,
      "num_tokens": 93187525.0,
      "step": 47640
    },
    {
      "entropy": 5.984529733657837,
      "epoch": 4.762633078422552,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00029546436172114837,
      "loss": 5.4405,
      "mean_token_accuracy": 0.20441157072782518,
      "num_tokens": 93198824.0,
      "step": 47645
    },
    {
      "entropy": 6.028190755844117,
      "epoch": 4.763132903483781,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002954288202859855,
      "loss": 5.4045,
      "mean_token_accuracy": 0.20219219326972962,
      "num_tokens": 93208344.0,
      "step": 47650
    },
    {
      "entropy": 5.897140979766846,
      "epoch": 4.7636327285450095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002953932783368388,
      "loss": 5.3585,
      "mean_token_accuracy": 0.20380486100912093,
      "num_tokens": 93217793.0,
      "step": 47655
    },
    {
      "entropy": 5.915743017196656,
      "epoch": 4.764132553606238,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029535773587460235,
      "loss": 5.3281,
      "mean_token_accuracy": 0.2162521004676819,
      "num_tokens": 93226543.0,
      "step": 47660
    },
    {
      "entropy": 5.8722731590271,
      "epoch": 4.764632378667466,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002953221929001706,
      "loss": 5.2812,
      "mean_token_accuracy": 0.209122496843338,
      "num_tokens": 93236766.0,
      "step": 47665
    },
    {
      "entropy": 5.964857339859009,
      "epoch": 4.765132203728695,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00029528664941443766,
      "loss": 5.3948,
      "mean_token_accuracy": 0.2087309867143631,
      "num_tokens": 93246654.0,
      "step": 47670
    },
    {
      "entropy": 5.930879068374634,
      "epoch": 4.765632028789923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002952511054182978,
      "loss": 5.2862,
      "mean_token_accuracy": 0.21507685482501984,
      "num_tokens": 93256157.0,
      "step": 47675
    },
    {
      "entropy": 5.962446498870849,
      "epoch": 4.766131853851152,
      "grad_norm": 1.28125,
      "learning_rate": 0.00029521556091264547,
      "loss": 5.403,
      "mean_token_accuracy": 0.20894753336906433,
      "num_tokens": 93265935.0,
      "step": 47680
    },
    {
      "entropy": 5.891399288177491,
      "epoch": 4.766631678912381,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029518001589837466,
      "loss": 5.3376,
      "mean_token_accuracy": 0.21830516159534455,
      "num_tokens": 93275849.0,
      "step": 47685
    },
    {
      "entropy": 5.995737457275391,
      "epoch": 4.7671315039736095,
      "grad_norm": 1.25,
      "learning_rate": 0.00029514447037638,
      "loss": 5.4086,
      "mean_token_accuracy": 0.2074425309896469,
      "num_tokens": 93284915.0,
      "step": 47690
    },
    {
      "entropy": 5.977061128616333,
      "epoch": 4.767631329034838,
      "grad_norm": 1.203125,
      "learning_rate": 0.00029510892434755556,
      "loss": 5.4112,
      "mean_token_accuracy": 0.2015232875943184,
      "num_tokens": 93293835.0,
      "step": 47695
    },
    {
      "entropy": 5.990554618835449,
      "epoch": 4.768131154096066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002950733778127958,
      "loss": 5.4111,
      "mean_token_accuracy": 0.2075910434126854,
      "num_tokens": 93303867.0,
      "step": 47700
    },
    {
      "entropy": 5.896244192123413,
      "epoch": 4.768630979157295,
      "grad_norm": 1.15625,
      "learning_rate": 0.000295037830772995,
      "loss": 5.2775,
      "mean_token_accuracy": 0.2170364513993263,
      "num_tokens": 93314217.0,
      "step": 47705
    },
    {
      "entropy": 5.951617670059204,
      "epoch": 4.769130804218523,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002950022832290475,
      "loss": 5.3522,
      "mean_token_accuracy": 0.21662811785936356,
      "num_tokens": 93323888.0,
      "step": 47710
    },
    {
      "entropy": 5.943874788284302,
      "epoch": 4.769630629279752,
      "grad_norm": 1.109375,
      "learning_rate": 0.00029496673518184777,
      "loss": 5.3758,
      "mean_token_accuracy": 0.21221452504396437,
      "num_tokens": 93334901.0,
      "step": 47715
    },
    {
      "entropy": 5.979932403564453,
      "epoch": 4.770130454340981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002949311866322901,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20959332585334778,
      "num_tokens": 93345757.0,
      "step": 47720
    },
    {
      "entropy": 5.981586074829101,
      "epoch": 4.7706302794022095,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029489563758126886,
      "loss": 5.3675,
      "mean_token_accuracy": 0.21074625998735427,
      "num_tokens": 93355750.0,
      "step": 47725
    },
    {
      "entropy": 5.912063455581665,
      "epoch": 4.771130104463438,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002948600880296785,
      "loss": 5.2818,
      "mean_token_accuracy": 0.2101954162120819,
      "num_tokens": 93365121.0,
      "step": 47730
    },
    {
      "entropy": 5.95076847076416,
      "epoch": 4.771629929524666,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002948245379784135,
      "loss": 5.3536,
      "mean_token_accuracy": 0.21229879409074784,
      "num_tokens": 93374991.0,
      "step": 47735
    },
    {
      "entropy": 5.981712436676025,
      "epoch": 4.772129754585895,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029478898742836817,
      "loss": 5.3828,
      "mean_token_accuracy": 0.2061815470457077,
      "num_tokens": 93384824.0,
      "step": 47740
    },
    {
      "entropy": 5.944541931152344,
      "epoch": 4.772629579647123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029475343638043697,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2195473626255989,
      "num_tokens": 93393593.0,
      "step": 47745
    },
    {
      "entropy": 5.962139844894409,
      "epoch": 4.773129404708352,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029471788483551443,
      "loss": 5.4109,
      "mean_token_accuracy": 0.20876242369413375,
      "num_tokens": 93403684.0,
      "step": 47750
    },
    {
      "entropy": 5.896848535537719,
      "epoch": 4.773629229769581,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002946823327944951,
      "loss": 5.2602,
      "mean_token_accuracy": 0.21718572825193405,
      "num_tokens": 93413419.0,
      "step": 47755
    },
    {
      "entropy": 5.924592351913452,
      "epoch": 4.7741290548308095,
      "grad_norm": 1.25,
      "learning_rate": 0.0002946467802582732,
      "loss": 5.3737,
      "mean_token_accuracy": 0.2059113472700119,
      "num_tokens": 93423006.0,
      "step": 47760
    },
    {
      "entropy": 5.999466180801392,
      "epoch": 4.774628879892038,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002946112272277434,
      "loss": 5.3213,
      "mean_token_accuracy": 0.2111528366804123,
      "num_tokens": 93431936.0,
      "step": 47765
    },
    {
      "entropy": 5.964332818984985,
      "epoch": 4.775128704953266,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002945756737038002,
      "loss": 5.3693,
      "mean_token_accuracy": 0.21362667083740233,
      "num_tokens": 93441768.0,
      "step": 47770
    },
    {
      "entropy": 5.961275243759156,
      "epoch": 4.775628530014495,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029454011968733804,
      "loss": 5.4084,
      "mean_token_accuracy": 0.2038000077009201,
      "num_tokens": 93451532.0,
      "step": 47775
    },
    {
      "entropy": 5.962594652175904,
      "epoch": 4.776128355075723,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002945045651792515,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21071975976228713,
      "num_tokens": 93461266.0,
      "step": 47780
    },
    {
      "entropy": 6.017410755157471,
      "epoch": 4.776628180136952,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00029446901018043516,
      "loss": 5.4252,
      "mean_token_accuracy": 0.2045282781124115,
      "num_tokens": 93471248.0,
      "step": 47785
    },
    {
      "entropy": 5.871835947036743,
      "epoch": 4.777128005198181,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029443345469178355,
      "loss": 5.2486,
      "mean_token_accuracy": 0.22605710923671724,
      "num_tokens": 93481778.0,
      "step": 47790
    },
    {
      "entropy": 5.8793376922607425,
      "epoch": 4.7776278302594095,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029439789871419113,
      "loss": 5.2747,
      "mean_token_accuracy": 0.22033848166465758,
      "num_tokens": 93491425.0,
      "step": 47795
    },
    {
      "entropy": 5.922244834899902,
      "epoch": 4.778127655320638,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002943623422485527,
      "loss": 5.3541,
      "mean_token_accuracy": 0.20902158915996552,
      "num_tokens": 93500903.0,
      "step": 47800
    },
    {
      "entropy": 5.866174983978271,
      "epoch": 4.778627480381866,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002943267852957626,
      "loss": 5.2571,
      "mean_token_accuracy": 0.21641040742397308,
      "num_tokens": 93509423.0,
      "step": 47805
    },
    {
      "entropy": 5.9953040599823,
      "epoch": 4.779127305443095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00029429122785671556,
      "loss": 5.3888,
      "mean_token_accuracy": 0.20709904581308364,
      "num_tokens": 93521295.0,
      "step": 47810
    },
    {
      "entropy": 5.94424352645874,
      "epoch": 4.779627130504323,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029425566993230624,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2205972582101822,
      "num_tokens": 93531783.0,
      "step": 47815
    },
    {
      "entropy": 6.048399591445923,
      "epoch": 4.780126955565552,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002942201115234292,
      "loss": 5.4286,
      "mean_token_accuracy": 0.20316366404294967,
      "num_tokens": 93542079.0,
      "step": 47820
    },
    {
      "entropy": 6.040271186828614,
      "epoch": 4.780626780626781,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029418455263097913,
      "loss": 5.4966,
      "mean_token_accuracy": 0.2037704825401306,
      "num_tokens": 93552491.0,
      "step": 47825
    },
    {
      "entropy": 5.972675895690918,
      "epoch": 4.7811266056880095,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029414899325585066,
      "loss": 5.3412,
      "mean_token_accuracy": 0.21406623870134353,
      "num_tokens": 93561332.0,
      "step": 47830
    },
    {
      "entropy": 5.951337242126465,
      "epoch": 4.781626430749238,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029411343339893835,
      "loss": 5.2744,
      "mean_token_accuracy": 0.2103965014219284,
      "num_tokens": 93570242.0,
      "step": 47835
    },
    {
      "entropy": 5.929914855957032,
      "epoch": 4.782126255810466,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029407787306113707,
      "loss": 5.3303,
      "mean_token_accuracy": 0.20585291534662248,
      "num_tokens": 93579291.0,
      "step": 47840
    },
    {
      "entropy": 5.8759359359741214,
      "epoch": 4.782626080871695,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029404231224334137,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21515064984560012,
      "num_tokens": 93588914.0,
      "step": 47845
    },
    {
      "entropy": 5.995321750640869,
      "epoch": 4.783125905932923,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029400675094644606,
      "loss": 5.3896,
      "mean_token_accuracy": 0.2001393511891365,
      "num_tokens": 93599350.0,
      "step": 47850
    },
    {
      "entropy": 5.901359748840332,
      "epoch": 4.783625730994152,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029397118917134576,
      "loss": 5.28,
      "mean_token_accuracy": 0.21101246774196625,
      "num_tokens": 93609127.0,
      "step": 47855
    },
    {
      "entropy": 5.929903507232666,
      "epoch": 4.784125556055381,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002939356269189352,
      "loss": 5.267,
      "mean_token_accuracy": 0.21507840007543563,
      "num_tokens": 93619218.0,
      "step": 47860
    },
    {
      "entropy": 5.79547381401062,
      "epoch": 4.7846253811166095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029390006419010916,
      "loss": 5.1809,
      "mean_token_accuracy": 0.2312793478369713,
      "num_tokens": 93629512.0,
      "step": 47865
    },
    {
      "entropy": 5.7099042415618895,
      "epoch": 4.785125206177838,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0002938645009857623,
      "loss": 5.0627,
      "mean_token_accuracy": 0.23342495411634445,
      "num_tokens": 93639016.0,
      "step": 47870
    },
    {
      "entropy": 5.930753469467163,
      "epoch": 4.785625031239066,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002938289373067895,
      "loss": 5.3851,
      "mean_token_accuracy": 0.2080097034573555,
      "num_tokens": 93649270.0,
      "step": 47875
    },
    {
      "entropy": 5.85277624130249,
      "epoch": 4.786124856300295,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002937933731540854,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21518107652664184,
      "num_tokens": 93659015.0,
      "step": 47880
    },
    {
      "entropy": 5.926639461517334,
      "epoch": 4.786624681361523,
      "grad_norm": 1.28125,
      "learning_rate": 0.000293757808528545,
      "loss": 5.2413,
      "mean_token_accuracy": 0.21301834434270858,
      "num_tokens": 93667846.0,
      "step": 47885
    },
    {
      "entropy": 5.991694450378418,
      "epoch": 4.787124506422752,
      "grad_norm": 1.125,
      "learning_rate": 0.0002937222434310629,
      "loss": 5.3905,
      "mean_token_accuracy": 0.2030473381280899,
      "num_tokens": 93678076.0,
      "step": 47890
    },
    {
      "entropy": 5.914833688735962,
      "epoch": 4.787624331483981,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029368667786253404,
      "loss": 5.3146,
      "mean_token_accuracy": 0.21399558186531067,
      "num_tokens": 93688566.0,
      "step": 47895
    },
    {
      "entropy": 5.8209803104400635,
      "epoch": 4.7881241565452095,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029365111182385305,
      "loss": 5.2813,
      "mean_token_accuracy": 0.21555336713790893,
      "num_tokens": 93698009.0,
      "step": 47900
    },
    {
      "entropy": 5.943331241607666,
      "epoch": 4.788623981606438,
      "grad_norm": 1.203125,
      "learning_rate": 0.000293615545315915,
      "loss": 5.3359,
      "mean_token_accuracy": 0.21691856384277344,
      "num_tokens": 93707720.0,
      "step": 47905
    },
    {
      "entropy": 5.909830045700073,
      "epoch": 4.789123806667666,
      "grad_norm": 1.140625,
      "learning_rate": 0.00029357997833961454,
      "loss": 5.3,
      "mean_token_accuracy": 0.21885122060775758,
      "num_tokens": 93716832.0,
      "step": 47910
    },
    {
      "entropy": 6.008914756774902,
      "epoch": 4.789623631728895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029354441089584667,
      "loss": 5.4173,
      "mean_token_accuracy": 0.20268016755580903,
      "num_tokens": 93727423.0,
      "step": 47915
    },
    {
      "entropy": 5.910327386856079,
      "epoch": 4.790123456790123,
      "grad_norm": 1.21875,
      "learning_rate": 0.00029350884298550616,
      "loss": 5.3198,
      "mean_token_accuracy": 0.213981232047081,
      "num_tokens": 93736664.0,
      "step": 47920
    },
    {
      "entropy": 5.919117975234985,
      "epoch": 4.790623281851352,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029347327460948786,
      "loss": 5.3908,
      "mean_token_accuracy": 0.20307970941066741,
      "num_tokens": 93747059.0,
      "step": 47925
    },
    {
      "entropy": 5.814618396759033,
      "epoch": 4.791123106912581,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029343770576868684,
      "loss": 5.1918,
      "mean_token_accuracy": 0.23104537576436995,
      "num_tokens": 93756460.0,
      "step": 47930
    },
    {
      "entropy": 5.939321899414063,
      "epoch": 4.7916229319738095,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002934021364639978,
      "loss": 5.3767,
      "mean_token_accuracy": 0.1957899272441864,
      "num_tokens": 93765496.0,
      "step": 47935
    },
    {
      "entropy": 5.863092756271362,
      "epoch": 4.792122757035038,
      "grad_norm": 1.125,
      "learning_rate": 0.0002933665666963158,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21808324456214906,
      "num_tokens": 93775189.0,
      "step": 47940
    },
    {
      "entropy": 5.936755037307739,
      "epoch": 4.792622582096266,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029333099646653565,
      "loss": 5.3297,
      "mean_token_accuracy": 0.21035303324460983,
      "num_tokens": 93784283.0,
      "step": 47945
    },
    {
      "entropy": 5.981506824493408,
      "epoch": 4.793122407157495,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002932954257755524,
      "loss": 5.3288,
      "mean_token_accuracy": 0.21670925468206406,
      "num_tokens": 93793115.0,
      "step": 47950
    },
    {
      "entropy": 5.886587858200073,
      "epoch": 4.793622232218723,
      "grad_norm": 1.109375,
      "learning_rate": 0.000293259854624261,
      "loss": 5.258,
      "mean_token_accuracy": 0.22325324267148972,
      "num_tokens": 93804106.0,
      "step": 47955
    },
    {
      "entropy": 5.9012532234191895,
      "epoch": 4.794122057279952,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00029322428301355624,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21996518224477768,
      "num_tokens": 93813991.0,
      "step": 47960
    },
    {
      "entropy": 6.009780597686768,
      "epoch": 4.794621882341181,
      "grad_norm": 1.234375,
      "learning_rate": 0.00029318871094433324,
      "loss": 5.3794,
      "mean_token_accuracy": 0.21381361782550812,
      "num_tokens": 93825024.0,
      "step": 47965
    },
    {
      "entropy": 6.0271618366241455,
      "epoch": 4.7951217074024095,
      "grad_norm": 1.078125,
      "learning_rate": 0.00029315313841748704,
      "loss": 5.3697,
      "mean_token_accuracy": 0.2018410548567772,
      "num_tokens": 93836173.0,
      "step": 47970
    },
    {
      "entropy": 5.953880500793457,
      "epoch": 4.795621532463638,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00029311756543391244,
      "loss": 5.3396,
      "mean_token_accuracy": 0.20630372017621995,
      "num_tokens": 93846519.0,
      "step": 47975
    },
    {
      "entropy": 5.912417984008789,
      "epoch": 4.796121357524866,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002930819919945046,
      "loss": 5.3339,
      "mean_token_accuracy": 0.20984984189271927,
      "num_tokens": 93856324.0,
      "step": 47980
    },
    {
      "entropy": 5.977431201934815,
      "epoch": 4.796621182586095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002930464181001585,
      "loss": 5.4249,
      "mean_token_accuracy": 0.2033650279045105,
      "num_tokens": 93865181.0,
      "step": 47985
    },
    {
      "entropy": 5.978052997589112,
      "epoch": 4.797121007647323,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002930108437517691,
      "loss": 5.3756,
      "mean_token_accuracy": 0.2038713052868843,
      "num_tokens": 93874577.0,
      "step": 47990
    },
    {
      "entropy": 5.914386463165283,
      "epoch": 4.797620832708552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029297526895023156,
      "loss": 5.2369,
      "mean_token_accuracy": 0.21590153276920318,
      "num_tokens": 93883129.0,
      "step": 47995
    },
    {
      "entropy": 5.86671199798584,
      "epoch": 4.798120657769781,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002929396936964409,
      "loss": 5.3238,
      "mean_token_accuracy": 0.22297086268663407,
      "num_tokens": 93893971.0,
      "step": 48000
    },
    {
      "epoch": 4.798120657769781,
      "eval_entropy": 5.702854934840803,
      "eval_loss": 5.463287353515625,
      "eval_mean_token_accuracy": 0.2135035098261346,
      "eval_num_tokens": 93893971.0,
      "eval_runtime": 24.1702,
      "eval_samples_per_second": 1284.434,
      "eval_steps_per_second": 160.57,
      "step": 48000
    },
    {
      "entropy": 5.940311479568481,
      "epoch": 4.7986204828310095,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002929041179912921,
      "loss": 5.3212,
      "mean_token_accuracy": 0.21445972621440887,
      "num_tokens": 93903899.0,
      "step": 48005
    },
    {
      "entropy": 5.985433053970337,
      "epoch": 4.799120307892238,
      "grad_norm": 1.265625,
      "learning_rate": 0.00029286854183568033,
      "loss": 5.3517,
      "mean_token_accuracy": 0.20529019981622695,
      "num_tokens": 93914031.0,
      "step": 48010
    },
    {
      "entropy": 5.987464332580567,
      "epoch": 4.799620132953466,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002928329652305007,
      "loss": 5.3804,
      "mean_token_accuracy": 0.207891283929348,
      "num_tokens": 93923560.0,
      "step": 48015
    },
    {
      "entropy": 5.951763010025024,
      "epoch": 4.800119958014695,
      "grad_norm": 1.25,
      "learning_rate": 0.0002927973881766482,
      "loss": 5.3356,
      "mean_token_accuracy": 0.21715320348739625,
      "num_tokens": 93932907.0,
      "step": 48020
    },
    {
      "entropy": 5.937940454483032,
      "epoch": 4.800619783075923,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000292761810675018,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21468447744846345,
      "num_tokens": 93942595.0,
      "step": 48025
    },
    {
      "entropy": 5.974768447875976,
      "epoch": 4.801119608137152,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002927262327265052,
      "loss": 5.4038,
      "mean_token_accuracy": 0.1996644213795662,
      "num_tokens": 93952522.0,
      "step": 48030
    },
    {
      "entropy": 5.967318344116211,
      "epoch": 4.801619433198381,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029269065433200494,
      "loss": 5.3846,
      "mean_token_accuracy": 0.20612021237611772,
      "num_tokens": 93961862.0,
      "step": 48035
    },
    {
      "entropy": 5.9650125980377195,
      "epoch": 4.8021192582596095,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002926550754924124,
      "loss": 5.2468,
      "mean_token_accuracy": 0.22495240867137908,
      "num_tokens": 93971119.0,
      "step": 48040
    },
    {
      "entropy": 5.940281200408935,
      "epoch": 4.802619083320838,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002926194962086227,
      "loss": 5.3608,
      "mean_token_accuracy": 0.21165546774864197,
      "num_tokens": 93981259.0,
      "step": 48045
    },
    {
      "entropy": 5.947931098937988,
      "epoch": 4.803118908382066,
      "grad_norm": 1.078125,
      "learning_rate": 0.000292583916481531,
      "loss": 5.3236,
      "mean_token_accuracy": 0.21577334702014922,
      "num_tokens": 93991652.0,
      "step": 48050
    },
    {
      "entropy": 5.926641845703125,
      "epoch": 4.803618733443295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002925483363120325,
      "loss": 5.3668,
      "mean_token_accuracy": 0.20879281610250472,
      "num_tokens": 94002590.0,
      "step": 48055
    },
    {
      "entropy": 5.902906322479248,
      "epoch": 4.804118558504523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002925127557010223,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21125276982784272,
      "num_tokens": 94011687.0,
      "step": 48060
    },
    {
      "entropy": 5.922893953323364,
      "epoch": 4.804618383565752,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002924771746493958,
      "loss": 5.263,
      "mean_token_accuracy": 0.21444906592369078,
      "num_tokens": 94020860.0,
      "step": 48065
    },
    {
      "entropy": 5.923592662811279,
      "epoch": 4.805118208626981,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029244159315804796,
      "loss": 5.3099,
      "mean_token_accuracy": 0.2148596704006195,
      "num_tokens": 94030718.0,
      "step": 48070
    },
    {
      "entropy": 5.886378335952759,
      "epoch": 4.8056180336882095,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002924060112278742,
      "loss": 5.2595,
      "mean_token_accuracy": 0.21622909903526305,
      "num_tokens": 94040815.0,
      "step": 48075
    },
    {
      "entropy": 5.915976428985596,
      "epoch": 4.806117858749437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029237042885976966,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2147481068968773,
      "num_tokens": 94050182.0,
      "step": 48080
    },
    {
      "entropy": 5.896638536453247,
      "epoch": 4.806617683810666,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00029233484605462956,
      "loss": 5.3646,
      "mean_token_accuracy": 0.20696485191583633,
      "num_tokens": 94059565.0,
      "step": 48085
    },
    {
      "entropy": 5.919974708557129,
      "epoch": 4.807117508871895,
      "grad_norm": 1.15625,
      "learning_rate": 0.00029229926281334925,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2104208156466484,
      "num_tokens": 94068514.0,
      "step": 48090
    },
    {
      "entropy": 6.001124095916748,
      "epoch": 4.807617333933123,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002922636791368239,
      "loss": 5.3887,
      "mean_token_accuracy": 0.21592275500297547,
      "num_tokens": 94078216.0,
      "step": 48095
    },
    {
      "entropy": 6.020588159561157,
      "epoch": 4.808117158994352,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029222809502594887,
      "loss": 5.3899,
      "mean_token_accuracy": 0.20778195261955262,
      "num_tokens": 94086918.0,
      "step": 48100
    },
    {
      "entropy": 6.018507385253907,
      "epoch": 4.808616984055581,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029219251048161936,
      "loss": 5.4931,
      "mean_token_accuracy": 0.2027353674173355,
      "num_tokens": 94096733.0,
      "step": 48105
    },
    {
      "entropy": 5.996832370758057,
      "epoch": 4.8091168091168095,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029215692550473075,
      "loss": 5.3901,
      "mean_token_accuracy": 0.20348696410655975,
      "num_tokens": 94106179.0,
      "step": 48110
    },
    {
      "entropy": 5.969534158706665,
      "epoch": 4.809616634178037,
      "grad_norm": 1.125,
      "learning_rate": 0.0002921213400961783,
      "loss": 5.2951,
      "mean_token_accuracy": 0.21273401379585266,
      "num_tokens": 94116649.0,
      "step": 48115
    },
    {
      "entropy": 5.956820106506347,
      "epoch": 4.810116459239266,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00029208575425685725,
      "loss": 5.3044,
      "mean_token_accuracy": 0.20946028530597688,
      "num_tokens": 94126219.0,
      "step": 48120
    },
    {
      "entropy": 5.9540300369262695,
      "epoch": 4.810616284300495,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002920501679876631,
      "loss": 5.3969,
      "mean_token_accuracy": 0.20569603592157365,
      "num_tokens": 94135924.0,
      "step": 48125
    },
    {
      "entropy": 5.964822912216187,
      "epoch": 4.811116109361723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002920145812894911,
      "loss": 5.3661,
      "mean_token_accuracy": 0.20797918289899825,
      "num_tokens": 94147748.0,
      "step": 48130
    },
    {
      "entropy": 6.027128124237061,
      "epoch": 4.811615934422952,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002919789941632366,
      "loss": 5.4698,
      "mean_token_accuracy": 0.1983620122075081,
      "num_tokens": 94157393.0,
      "step": 48135
    },
    {
      "entropy": 6.00886869430542,
      "epoch": 4.812115759484181,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00029194340660979506,
      "loss": 5.3374,
      "mean_token_accuracy": 0.21011281311511992,
      "num_tokens": 94167073.0,
      "step": 48140
    },
    {
      "entropy": 5.9098673343658445,
      "epoch": 4.812615584545409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029190781863006174,
      "loss": 5.3033,
      "mean_token_accuracy": 0.21546362489461898,
      "num_tokens": 94177277.0,
      "step": 48145
    },
    {
      "entropy": 5.976425075531006,
      "epoch": 4.813115409606637,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000291872230224932,
      "loss": 5.4015,
      "mean_token_accuracy": 0.19728058725595474,
      "num_tokens": 94187813.0,
      "step": 48150
    },
    {
      "entropy": 6.020329427719116,
      "epoch": 4.813615234667866,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029183664139530125,
      "loss": 5.3814,
      "mean_token_accuracy": 0.1993684411048889,
      "num_tokens": 94197215.0,
      "step": 48155
    },
    {
      "entropy": 5.934050893783569,
      "epoch": 4.814115059729095,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000291801052142065,
      "loss": 5.323,
      "mean_token_accuracy": 0.21637239754199983,
      "num_tokens": 94208127.0,
      "step": 48160
    },
    {
      "entropy": 5.9300439834594725,
      "epoch": 4.814614884790323,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029176546246611855,
      "loss": 5.2892,
      "mean_token_accuracy": 0.22320551723241805,
      "num_tokens": 94218504.0,
      "step": 48165
    },
    {
      "entropy": 6.048795986175537,
      "epoch": 4.815114709851552,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002917298723683575,
      "loss": 5.3998,
      "mean_token_accuracy": 0.2001920834183693,
      "num_tokens": 94227238.0,
      "step": 48170
    },
    {
      "entropy": 6.019071054458618,
      "epoch": 4.815614534912781,
      "grad_norm": 1.25,
      "learning_rate": 0.00029169428184967706,
      "loss": 5.4546,
      "mean_token_accuracy": 0.20164516419172288,
      "num_tokens": 94236865.0,
      "step": 48175
    },
    {
      "entropy": 5.866669511795044,
      "epoch": 4.816114359974009,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00029165869091097283,
      "loss": 5.2354,
      "mean_token_accuracy": 0.2200516030192375,
      "num_tokens": 94247397.0,
      "step": 48180
    },
    {
      "entropy": 5.902822542190552,
      "epoch": 4.816614185035237,
      "grad_norm": 1.125,
      "learning_rate": 0.0002916230995531402,
      "loss": 5.3629,
      "mean_token_accuracy": 0.20728123635053636,
      "num_tokens": 94258091.0,
      "step": 48185
    },
    {
      "entropy": 5.8746508121490475,
      "epoch": 4.817114010096466,
      "grad_norm": 1.171875,
      "learning_rate": 0.00029158750777707467,
      "loss": 5.2437,
      "mean_token_accuracy": 0.22529429644346238,
      "num_tokens": 94267563.0,
      "step": 48190
    },
    {
      "entropy": 5.9097230434417725,
      "epoch": 4.817613835157695,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002915519155836717,
      "loss": 5.2338,
      "mean_token_accuracy": 0.2220060020685196,
      "num_tokens": 94277312.0,
      "step": 48195
    },
    {
      "entropy": 5.894767951965332,
      "epoch": 4.818113660218923,
      "grad_norm": 1.125,
      "learning_rate": 0.0002915163229738268,
      "loss": 5.2633,
      "mean_token_accuracy": 0.2194003254175186,
      "num_tokens": 94286688.0,
      "step": 48200
    },
    {
      "entropy": 5.82127799987793,
      "epoch": 4.818613485280152,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002914807299484355,
      "loss": 5.2829,
      "mean_token_accuracy": 0.2183946490287781,
      "num_tokens": 94296103.0,
      "step": 48205
    },
    {
      "entropy": 5.976625442504883,
      "epoch": 4.819113310341381,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002914451365083932,
      "loss": 5.4025,
      "mean_token_accuracy": 0.20047547817230224,
      "num_tokens": 94306676.0,
      "step": 48210
    },
    {
      "entropy": 5.903481483459473,
      "epoch": 4.819613135402609,
      "grad_norm": 1.125,
      "learning_rate": 0.00029140954265459554,
      "loss": 5.2707,
      "mean_token_accuracy": 0.21405522227287294,
      "num_tokens": 94316672.0,
      "step": 48215
    },
    {
      "entropy": 5.933163404464722,
      "epoch": 4.820112960463837,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000291373948387938,
      "loss": 5.385,
      "mean_token_accuracy": 0.20580518543720244,
      "num_tokens": 94326212.0,
      "step": 48220
    },
    {
      "entropy": 5.953744602203369,
      "epoch": 4.820612785525066,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002913383537093161,
      "loss": 5.3743,
      "mean_token_accuracy": 0.2152107536792755,
      "num_tokens": 94335582.0,
      "step": 48225
    },
    {
      "entropy": 6.005784416198731,
      "epoch": 4.821112610586295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002913027586196255,
      "loss": 5.3941,
      "mean_token_accuracy": 0.19948982894420625,
      "num_tokens": 94346657.0,
      "step": 48230
    },
    {
      "entropy": 5.910033130645752,
      "epoch": 4.821612435647523,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002912671631197616,
      "loss": 5.2949,
      "mean_token_accuracy": 0.22012152075767516,
      "num_tokens": 94355908.0,
      "step": 48235
    },
    {
      "entropy": 5.98690972328186,
      "epoch": 4.822112260708752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002912315672106202,
      "loss": 5.4127,
      "mean_token_accuracy": 0.20704457014799119,
      "num_tokens": 94365926.0,
      "step": 48240
    },
    {
      "entropy": 5.925216722488403,
      "epoch": 4.822612085769981,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002911959708930966,
      "loss": 5.3685,
      "mean_token_accuracy": 0.20978707671165467,
      "num_tokens": 94375291.0,
      "step": 48245
    },
    {
      "entropy": 5.860266637802124,
      "epoch": 4.823111910831209,
      "grad_norm": 1.25,
      "learning_rate": 0.0002911603741680866,
      "loss": 5.26,
      "mean_token_accuracy": 0.22351263910532,
      "num_tokens": 94384522.0,
      "step": 48250
    },
    {
      "entropy": 5.931686735153198,
      "epoch": 4.823611735892437,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029112477703648575,
      "loss": 5.2703,
      "mean_token_accuracy": 0.2196951389312744,
      "num_tokens": 94394255.0,
      "step": 48255
    },
    {
      "entropy": 5.916392087936401,
      "epoch": 4.824111560953666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002910891794991897,
      "loss": 5.3224,
      "mean_token_accuracy": 0.20847125351428986,
      "num_tokens": 94405125.0,
      "step": 48260
    },
    {
      "entropy": 5.92047643661499,
      "epoch": 4.824611386014895,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000291053581557094,
      "loss": 5.4046,
      "mean_token_accuracy": 0.21093534380197526,
      "num_tokens": 94414893.0,
      "step": 48265
    },
    {
      "entropy": 5.981108856201172,
      "epoch": 4.825111211076123,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029101798321109427,
      "loss": 5.3716,
      "mean_token_accuracy": 0.20783700942993164,
      "num_tokens": 94424141.0,
      "step": 48270
    },
    {
      "entropy": 5.969895267486573,
      "epoch": 4.825611036137352,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00029098238446208626,
      "loss": 5.3591,
      "mean_token_accuracy": 0.2098898246884346,
      "num_tokens": 94434342.0,
      "step": 48275
    },
    {
      "entropy": 5.98207540512085,
      "epoch": 4.826110861198581,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002909467853109656,
      "loss": 5.4411,
      "mean_token_accuracy": 0.20033809542655945,
      "num_tokens": 94445208.0,
      "step": 48280
    },
    {
      "entropy": 5.9545563697814945,
      "epoch": 4.826610686259809,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002909111857586278,
      "loss": 5.3855,
      "mean_token_accuracy": 0.20309212803840637,
      "num_tokens": 94454444.0,
      "step": 48285
    },
    {
      "entropy": 5.9677777767181395,
      "epoch": 4.827110511321037,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002908755858059688,
      "loss": 5.3051,
      "mean_token_accuracy": 0.20663011521101,
      "num_tokens": 94464655.0,
      "step": 48290
    },
    {
      "entropy": 5.866626834869384,
      "epoch": 4.827610336382266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00029083998545388406,
      "loss": 5.26,
      "mean_token_accuracy": 0.2237279936671257,
      "num_tokens": 94473999.0,
      "step": 48295
    },
    {
      "entropy": 5.952513933181763,
      "epoch": 4.828110161443495,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002908043847032694,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21219706833362578,
      "num_tokens": 94483839.0,
      "step": 48300
    },
    {
      "entropy": 5.963705682754517,
      "epoch": 4.828609986504723,
      "grad_norm": 1.3125,
      "learning_rate": 0.00029076878355502046,
      "loss": 5.339,
      "mean_token_accuracy": 0.21251289248466493,
      "num_tokens": 94494713.0,
      "step": 48305
    },
    {
      "entropy": 5.88286657333374,
      "epoch": 4.829109811565952,
      "grad_norm": 1.265625,
      "learning_rate": 0.000290733182010033,
      "loss": 5.2613,
      "mean_token_accuracy": 0.22142644375562667,
      "num_tokens": 94504658.0,
      "step": 48310
    },
    {
      "entropy": 5.901621246337891,
      "epoch": 4.829609636627181,
      "grad_norm": 1.25,
      "learning_rate": 0.0002906975800692027,
      "loss": 5.2745,
      "mean_token_accuracy": 0.22396587431430817,
      "num_tokens": 94514485.0,
      "step": 48315
    },
    {
      "entropy": 5.891609573364258,
      "epoch": 4.830109461688409,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00029066197773342536,
      "loss": 5.3061,
      "mean_token_accuracy": 0.21530241668224334,
      "num_tokens": 94522735.0,
      "step": 48320
    },
    {
      "entropy": 5.927429103851319,
      "epoch": 4.830609286749637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002906263750035967,
      "loss": 5.3837,
      "mean_token_accuracy": 0.2146727904677391,
      "num_tokens": 94531952.0,
      "step": 48325
    },
    {
      "entropy": 5.837623500823975,
      "epoch": 4.831109111810866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00029059077188061245,
      "loss": 5.1277,
      "mean_token_accuracy": 0.23376839309930803,
      "num_tokens": 94542161.0,
      "step": 48330
    },
    {
      "entropy": 5.974328470230103,
      "epoch": 4.831608936872095,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002905551683653683,
      "loss": 5.3593,
      "mean_token_accuracy": 0.2124430790543556,
      "num_tokens": 94551787.0,
      "step": 48335
    },
    {
      "entropy": 5.861504030227661,
      "epoch": 4.832108761933323,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002905195644587603,
      "loss": 5.2686,
      "mean_token_accuracy": 0.21257109493017196,
      "num_tokens": 94562290.0,
      "step": 48340
    },
    {
      "entropy": 5.97735595703125,
      "epoch": 4.832608586994552,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00029048396016168396,
      "loss": 5.3759,
      "mean_token_accuracy": 0.20822445005178453,
      "num_tokens": 94571878.0,
      "step": 48345
    },
    {
      "entropy": 5.960498523712158,
      "epoch": 4.833108412055781,
      "grad_norm": 1.125,
      "learning_rate": 0.00029044835547503516,
      "loss": 5.3282,
      "mean_token_accuracy": 0.2093417078256607,
      "num_tokens": 94582592.0,
      "step": 48350
    },
    {
      "entropy": 5.94668927192688,
      "epoch": 4.833608237117009,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029041275039970964,
      "loss": 5.3219,
      "mean_token_accuracy": 0.2091667667031288,
      "num_tokens": 94591715.0,
      "step": 48355
    },
    {
      "entropy": 5.83934588432312,
      "epoch": 4.834108062178237,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0002903771449366034,
      "loss": 5.1234,
      "mean_token_accuracy": 0.23787977248430253,
      "num_tokens": 94601673.0,
      "step": 48360
    },
    {
      "entropy": 5.867923450469971,
      "epoch": 4.834607887239466,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002903415390866121,
      "loss": 5.3138,
      "mean_token_accuracy": 0.21688663065433503,
      "num_tokens": 94611150.0,
      "step": 48365
    },
    {
      "entropy": 5.945834016799926,
      "epoch": 4.835107712300695,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002903059328506317,
      "loss": 5.3285,
      "mean_token_accuracy": 0.21957822144031525,
      "num_tokens": 94621020.0,
      "step": 48370
    },
    {
      "entropy": 5.899651622772216,
      "epoch": 4.835607537361923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029027032622955777,
      "loss": 5.2171,
      "mean_token_accuracy": 0.22049637287855148,
      "num_tokens": 94630537.0,
      "step": 48375
    },
    {
      "entropy": 6.000145053863525,
      "epoch": 4.836107362423152,
      "grad_norm": 1.1875,
      "learning_rate": 0.00029023471922428657,
      "loss": 5.4338,
      "mean_token_accuracy": 0.20690441876649857,
      "num_tokens": 94640735.0,
      "step": 48380
    },
    {
      "entropy": 5.944220972061157,
      "epoch": 4.836607187484381,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002901991118357137,
      "loss": 5.2978,
      "mean_token_accuracy": 0.20772503912448884,
      "num_tokens": 94649849.0,
      "step": 48385
    },
    {
      "entropy": 5.980346822738648,
      "epoch": 4.837107012545609,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00029016350406473507,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20723129510879518,
      "num_tokens": 94659890.0,
      "step": 48390
    },
    {
      "entropy": 5.96099886894226,
      "epoch": 4.837606837606837,
      "grad_norm": 1.0625,
      "learning_rate": 0.00029012789591224657,
      "loss": 5.3408,
      "mean_token_accuracy": 0.20815657675266266,
      "num_tokens": 94668676.0,
      "step": 48395
    },
    {
      "entropy": 5.941809034347534,
      "epoch": 4.838106662668066,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002900922873791441,
      "loss": 5.2456,
      "mean_token_accuracy": 0.21270942986011504,
      "num_tokens": 94678364.0,
      "step": 48400
    },
    {
      "entropy": 5.908549356460571,
      "epoch": 4.838606487729295,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002900566784663235,
      "loss": 5.3748,
      "mean_token_accuracy": 0.20206715017557145,
      "num_tokens": 94688495.0,
      "step": 48405
    },
    {
      "entropy": 5.88356032371521,
      "epoch": 4.839106312790523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00029002106917468086,
      "loss": 5.1355,
      "mean_token_accuracy": 0.22298160195350647,
      "num_tokens": 94697258.0,
      "step": 48410
    },
    {
      "entropy": 6.0481630802154545,
      "epoch": 4.839606137851752,
      "grad_norm": 1.25,
      "learning_rate": 0.00028998545950511195,
      "loss": 5.4297,
      "mean_token_accuracy": 0.21037548035383224,
      "num_tokens": 94706696.0,
      "step": 48415
    },
    {
      "entropy": 5.8659521579742435,
      "epoch": 4.840105962912981,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028994984945851277,
      "loss": 5.3491,
      "mean_token_accuracy": 0.21303326934576033,
      "num_tokens": 94716409.0,
      "step": 48420
    },
    {
      "entropy": 5.866982889175415,
      "epoch": 4.840605787974209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002899142390357792,
      "loss": 5.2626,
      "mean_token_accuracy": 0.214222951233387,
      "num_tokens": 94726086.0,
      "step": 48425
    },
    {
      "entropy": 5.981846857070923,
      "epoch": 4.841105613035437,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028987862823780713,
      "loss": 5.2888,
      "mean_token_accuracy": 0.2150810495018959,
      "num_tokens": 94735231.0,
      "step": 48430
    },
    {
      "entropy": 5.919806241989136,
      "epoch": 4.841605438096666,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028984301706549265,
      "loss": 5.2303,
      "mean_token_accuracy": 0.22269193679094315,
      "num_tokens": 94745162.0,
      "step": 48435
    },
    {
      "entropy": 5.930755519866944,
      "epoch": 4.842105263157895,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028980740551973167,
      "loss": 5.3191,
      "mean_token_accuracy": 0.20674119442701339,
      "num_tokens": 94755029.0,
      "step": 48440
    },
    {
      "entropy": 5.877530813217163,
      "epoch": 4.842605088219123,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002897717936014202,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2150132343173027,
      "num_tokens": 94766049.0,
      "step": 48445
    },
    {
      "entropy": 5.982528305053711,
      "epoch": 4.843104913280352,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002897361813114543,
      "loss": 5.4084,
      "mean_token_accuracy": 0.20688542127609252,
      "num_tokens": 94775445.0,
      "step": 48450
    },
    {
      "entropy": 5.8704596042633055,
      "epoch": 4.843604738341581,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028970056865072974,
      "loss": 5.2959,
      "mean_token_accuracy": 0.21993755400180817,
      "num_tokens": 94785871.0,
      "step": 48455
    },
    {
      "entropy": 5.900885105133057,
      "epoch": 4.844104563402809,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028966495562014267,
      "loss": 5.266,
      "mean_token_accuracy": 0.21740818172693252,
      "num_tokens": 94795677.0,
      "step": 48460
    },
    {
      "entropy": 5.910172414779663,
      "epoch": 4.844604388464037,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002896293422205891,
      "loss": 5.3475,
      "mean_token_accuracy": 0.21374174803495408,
      "num_tokens": 94806003.0,
      "step": 48465
    },
    {
      "entropy": 5.950716114044189,
      "epoch": 4.845104213525266,
      "grad_norm": 1.25,
      "learning_rate": 0.00028959372845296504,
      "loss": 5.3033,
      "mean_token_accuracy": 0.21328891813755035,
      "num_tokens": 94816811.0,
      "step": 48470
    },
    {
      "entropy": 5.960167789459229,
      "epoch": 4.845604038586495,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002895581143181665,
      "loss": 5.3792,
      "mean_token_accuracy": 0.20089991092681886,
      "num_tokens": 94826225.0,
      "step": 48475
    },
    {
      "entropy": 5.905238437652588,
      "epoch": 4.846103863647723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002895224998170897,
      "loss": 5.2908,
      "mean_token_accuracy": 0.21386557072401047,
      "num_tokens": 94835991.0,
      "step": 48480
    },
    {
      "entropy": 5.870640563964844,
      "epoch": 4.846603688708952,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002894868849506304,
      "loss": 5.2321,
      "mean_token_accuracy": 0.22494089901447295,
      "num_tokens": 94846163.0,
      "step": 48485
    },
    {
      "entropy": 5.957788372039795,
      "epoch": 4.847103513770181,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002894512697196848,
      "loss": 5.3482,
      "mean_token_accuracy": 0.2110759884119034,
      "num_tokens": 94856080.0,
      "step": 48490
    },
    {
      "entropy": 5.981029510498047,
      "epoch": 4.847603338831409,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028941565412514907,
      "loss": 5.3597,
      "mean_token_accuracy": 0.21010201871395112,
      "num_tokens": 94866076.0,
      "step": 48495
    },
    {
      "entropy": 5.966193199157715,
      "epoch": 4.848103163892637,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028938003816791907,
      "loss": 5.3351,
      "mean_token_accuracy": 0.20393348336219788,
      "num_tokens": 94877212.0,
      "step": 48500
    },
    {
      "entropy": 5.9440892219543455,
      "epoch": 4.848602988953866,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002893444218488911,
      "loss": 5.3225,
      "mean_token_accuracy": 0.2154437154531479,
      "num_tokens": 94887340.0,
      "step": 48505
    },
    {
      "entropy": 6.012714958190918,
      "epoch": 4.849102814015095,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028930880516896113,
      "loss": 5.3638,
      "mean_token_accuracy": 0.20687283277511598,
      "num_tokens": 94898870.0,
      "step": 48510
    },
    {
      "entropy": 5.900240516662597,
      "epoch": 4.849602639076323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002892731881290253,
      "loss": 5.3076,
      "mean_token_accuracy": 0.2084478035569191,
      "num_tokens": 94909775.0,
      "step": 48515
    },
    {
      "entropy": 5.87098069190979,
      "epoch": 4.850102464137552,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002892375707299798,
      "loss": 5.3652,
      "mean_token_accuracy": 0.2095095083117485,
      "num_tokens": 94920656.0,
      "step": 48520
    },
    {
      "entropy": 5.977149724960327,
      "epoch": 4.850602289198781,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002892019529727206,
      "loss": 5.4331,
      "mean_token_accuracy": 0.20503420382738113,
      "num_tokens": 94930108.0,
      "step": 48525
    },
    {
      "entropy": 6.0048887729644775,
      "epoch": 4.851102114260009,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002891663348581439,
      "loss": 5.4322,
      "mean_token_accuracy": 0.20303452014923096,
      "num_tokens": 94940021.0,
      "step": 48530
    },
    {
      "entropy": 5.966620540618896,
      "epoch": 4.851601939321237,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002891307163871459,
      "loss": 5.3235,
      "mean_token_accuracy": 0.213341648876667,
      "num_tokens": 94950381.0,
      "step": 48535
    },
    {
      "entropy": 5.89547700881958,
      "epoch": 4.852101764382466,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028909509756062273,
      "loss": 5.3267,
      "mean_token_accuracy": 0.21459081023931503,
      "num_tokens": 94962218.0,
      "step": 48540
    },
    {
      "entropy": 5.9916675090789795,
      "epoch": 4.852601589443695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028905947837947057,
      "loss": 5.3522,
      "mean_token_accuracy": 0.20146140456199646,
      "num_tokens": 94972168.0,
      "step": 48545
    },
    {
      "entropy": 5.946703100204468,
      "epoch": 4.853101414504923,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002890238588445854,
      "loss": 5.3357,
      "mean_token_accuracy": 0.2037556692957878,
      "num_tokens": 94981213.0,
      "step": 48550
    },
    {
      "entropy": 5.970210790634155,
      "epoch": 4.853601239566152,
      "grad_norm": 1.46875,
      "learning_rate": 0.0002889882389568638,
      "loss": 5.3153,
      "mean_token_accuracy": 0.2154798462986946,
      "num_tokens": 94990252.0,
      "step": 48555
    },
    {
      "entropy": 5.991180801391602,
      "epoch": 4.854101064627381,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002889526187172015,
      "loss": 5.4492,
      "mean_token_accuracy": 0.19868503361940384,
      "num_tokens": 94999741.0,
      "step": 48560
    },
    {
      "entropy": 5.95477557182312,
      "epoch": 4.854600889688609,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028891699812649493,
      "loss": 5.4408,
      "mean_token_accuracy": 0.20846294164657592,
      "num_tokens": 95009455.0,
      "step": 48565
    },
    {
      "entropy": 5.855493545532227,
      "epoch": 4.855100714749837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002888813771856403,
      "loss": 5.1825,
      "mean_token_accuracy": 0.22541654109954834,
      "num_tokens": 95019621.0,
      "step": 48570
    },
    {
      "entropy": 5.97260274887085,
      "epoch": 4.855600539811066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002888457558955338,
      "loss": 5.313,
      "mean_token_accuracy": 0.21181371957063674,
      "num_tokens": 95029787.0,
      "step": 48575
    },
    {
      "entropy": 5.951861476898193,
      "epoch": 4.856100364872295,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028881013425707163,
      "loss": 5.3579,
      "mean_token_accuracy": 0.21402157992124557,
      "num_tokens": 95039682.0,
      "step": 48580
    },
    {
      "entropy": 5.904410171508789,
      "epoch": 4.856600189933523,
      "grad_norm": 1.1875,
      "learning_rate": 0.00028877451227115003,
      "loss": 5.3179,
      "mean_token_accuracy": 0.2125084489583969,
      "num_tokens": 95048891.0,
      "step": 48585
    },
    {
      "entropy": 5.935131168365478,
      "epoch": 4.857100014994752,
      "grad_norm": 1.078125,
      "learning_rate": 0.00028873888993866527,
      "loss": 5.3116,
      "mean_token_accuracy": 0.2072027549147606,
      "num_tokens": 95058900.0,
      "step": 48590
    },
    {
      "entropy": 5.908323049545288,
      "epoch": 4.857599840055981,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002887032672605136,
      "loss": 5.2567,
      "mean_token_accuracy": 0.22058908343315126,
      "num_tokens": 95067990.0,
      "step": 48595
    },
    {
      "entropy": 5.907969903945923,
      "epoch": 4.858099665117209,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028866764423759116,
      "loss": 5.314,
      "mean_token_accuracy": 0.2071455731987953,
      "num_tokens": 95077811.0,
      "step": 48600
    },
    {
      "entropy": 5.939372205734253,
      "epoch": 4.858599490178437,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002886320208707944,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21316867917776108,
      "num_tokens": 95087755.0,
      "step": 48605
    },
    {
      "entropy": 5.885903263092041,
      "epoch": 4.859099315239666,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002885963971610194,
      "loss": 5.2679,
      "mean_token_accuracy": 0.21769707947969436,
      "num_tokens": 95098235.0,
      "step": 48610
    },
    {
      "entropy": 5.871370410919189,
      "epoch": 4.859599140300895,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028856077310916267,
      "loss": 5.2132,
      "mean_token_accuracy": 0.21799449920654296,
      "num_tokens": 95108965.0,
      "step": 48615
    },
    {
      "entropy": 5.941296529769898,
      "epoch": 4.860098965362123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002885251487161203,
      "loss": 5.3583,
      "mean_token_accuracy": 0.20692614912986756,
      "num_tokens": 95118716.0,
      "step": 48620
    },
    {
      "entropy": 5.846583461761474,
      "epoch": 4.860598790423352,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028848952398278865,
      "loss": 5.1541,
      "mean_token_accuracy": 0.23046329915523528,
      "num_tokens": 95128301.0,
      "step": 48625
    },
    {
      "entropy": 5.998320579528809,
      "epoch": 4.861098615484581,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028845389891006415,
      "loss": 5.4326,
      "mean_token_accuracy": 0.1991291880607605,
      "num_tokens": 95137681.0,
      "step": 48630
    },
    {
      "entropy": 5.904818725585938,
      "epoch": 4.861598440545809,
      "grad_norm": 1.015625,
      "learning_rate": 0.0002884182734988429,
      "loss": 5.2378,
      "mean_token_accuracy": 0.22081475406885148,
      "num_tokens": 95148430.0,
      "step": 48635
    },
    {
      "entropy": 5.8697967529296875,
      "epoch": 4.862098265607037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028838264775002143,
      "loss": 5.2214,
      "mean_token_accuracy": 0.2180078148841858,
      "num_tokens": 95158378.0,
      "step": 48640
    },
    {
      "entropy": 5.943483591079712,
      "epoch": 4.862598090668266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028834702166449593,
      "loss": 5.3516,
      "mean_token_accuracy": 0.21633879691362382,
      "num_tokens": 95168116.0,
      "step": 48645
    },
    {
      "entropy": 5.959607458114624,
      "epoch": 4.863097915729495,
      "grad_norm": 1.125,
      "learning_rate": 0.0002883113952431629,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1997626468539238,
      "num_tokens": 95177744.0,
      "step": 48650
    },
    {
      "entropy": 5.870483112335205,
      "epoch": 4.863597740790723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028827576848691836,
      "loss": 5.2508,
      "mean_token_accuracy": 0.21664623469114302,
      "num_tokens": 95186732.0,
      "step": 48655
    },
    {
      "entropy": 5.877376890182495,
      "epoch": 4.864097565851952,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028824014139665913,
      "loss": 5.2984,
      "mean_token_accuracy": 0.20791819989681243,
      "num_tokens": 95196026.0,
      "step": 48660
    },
    {
      "entropy": 5.905556058883667,
      "epoch": 4.864597390913181,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002882045139732813,
      "loss": 5.3715,
      "mean_token_accuracy": 0.21107954829931258,
      "num_tokens": 95206099.0,
      "step": 48665
    },
    {
      "entropy": 5.961386728286743,
      "epoch": 4.865097215974409,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002881688862176812,
      "loss": 5.4166,
      "mean_token_accuracy": 0.21065672039985656,
      "num_tokens": 95214718.0,
      "step": 48670
    },
    {
      "entropy": 5.925380277633667,
      "epoch": 4.865597041035637,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002881332581307554,
      "loss": 5.3303,
      "mean_token_accuracy": 0.2086944818496704,
      "num_tokens": 95224064.0,
      "step": 48675
    },
    {
      "entropy": 5.949001264572144,
      "epoch": 4.866096866096866,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028809762971340015,
      "loss": 5.1954,
      "mean_token_accuracy": 0.22303885519504546,
      "num_tokens": 95231999.0,
      "step": 48680
    },
    {
      "entropy": 5.813930559158325,
      "epoch": 4.866596691158095,
      "grad_norm": 1.1875,
      "learning_rate": 0.000288062000966512,
      "loss": 5.2713,
      "mean_token_accuracy": 0.22075762897729873,
      "num_tokens": 95241877.0,
      "step": 48685
    },
    {
      "entropy": 5.914291858673096,
      "epoch": 4.867096516219323,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002880263718909871,
      "loss": 5.3015,
      "mean_token_accuracy": 0.21412178725004197,
      "num_tokens": 95250755.0,
      "step": 48690
    },
    {
      "entropy": 5.976723575592041,
      "epoch": 4.867596341280552,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028799074248772214,
      "loss": 5.371,
      "mean_token_accuracy": 0.20924188047647477,
      "num_tokens": 95261430.0,
      "step": 48695
    },
    {
      "entropy": 5.793975019454956,
      "epoch": 4.868096166341781,
      "grad_norm": 1.0625,
      "learning_rate": 0.00028795511275761347,
      "loss": 5.1373,
      "mean_token_accuracy": 0.23608812540769578,
      "num_tokens": 95271521.0,
      "step": 48700
    },
    {
      "entropy": 5.938408374786377,
      "epoch": 4.868595991403009,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028791948270155747,
      "loss": 5.3179,
      "mean_token_accuracy": 0.21199265867471695,
      "num_tokens": 95280416.0,
      "step": 48705
    },
    {
      "entropy": 5.870057010650635,
      "epoch": 4.869095816464237,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028788385232045066,
      "loss": 5.2996,
      "mean_token_accuracy": 0.21405091583728791,
      "num_tokens": 95291361.0,
      "step": 48710
    },
    {
      "entropy": 6.002820873260498,
      "epoch": 4.869595641525466,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002878482216151894,
      "loss": 5.4971,
      "mean_token_accuracy": 0.19805468171834945,
      "num_tokens": 95301146.0,
      "step": 48715
    },
    {
      "entropy": 5.941692590713501,
      "epoch": 4.870095466586695,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002878125905866701,
      "loss": 5.2466,
      "mean_token_accuracy": 0.2175922378897667,
      "num_tokens": 95309655.0,
      "step": 48720
    },
    {
      "entropy": 5.926134824752808,
      "epoch": 4.870595291647923,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028777695923578944,
      "loss": 5.3136,
      "mean_token_accuracy": 0.21648795455694197,
      "num_tokens": 95319444.0,
      "step": 48725
    },
    {
      "entropy": 5.938371562957764,
      "epoch": 4.871095116709152,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002877413275634438,
      "loss": 5.3692,
      "mean_token_accuracy": 0.21020067632198333,
      "num_tokens": 95329460.0,
      "step": 48730
    },
    {
      "entropy": 5.950079154968262,
      "epoch": 4.871594941770381,
      "grad_norm": 1.25,
      "learning_rate": 0.0002877056955705296,
      "loss": 5.3166,
      "mean_token_accuracy": 0.2054997891187668,
      "num_tokens": 95339610.0,
      "step": 48735
    },
    {
      "entropy": 5.950096702575683,
      "epoch": 4.872094766831609,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002876700632579433,
      "loss": 5.3203,
      "mean_token_accuracy": 0.2169183149933815,
      "num_tokens": 95350166.0,
      "step": 48740
    },
    {
      "entropy": 5.93412070274353,
      "epoch": 4.872594591892837,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002876344306265815,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21578503251075745,
      "num_tokens": 95359406.0,
      "step": 48745
    },
    {
      "entropy": 5.796223545074463,
      "epoch": 4.873094416954066,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028759879767734065,
      "loss": 5.208,
      "mean_token_accuracy": 0.22434355765581132,
      "num_tokens": 95368226.0,
      "step": 48750
    },
    {
      "entropy": 5.911952877044678,
      "epoch": 4.873594242015295,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002875631644111172,
      "loss": 5.3524,
      "mean_token_accuracy": 0.2027316004037857,
      "num_tokens": 95378397.0,
      "step": 48755
    },
    {
      "entropy": 5.9450602531433105,
      "epoch": 4.874094067076523,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002875275308288079,
      "loss": 5.3496,
      "mean_token_accuracy": 0.2193852037191391,
      "num_tokens": 95390047.0,
      "step": 48760
    },
    {
      "entropy": 5.958121967315674,
      "epoch": 4.874593892137752,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028749189693130924,
      "loss": 5.3628,
      "mean_token_accuracy": 0.20849617719650268,
      "num_tokens": 95398824.0,
      "step": 48765
    },
    {
      "entropy": 5.986031198501587,
      "epoch": 4.875093717198981,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002874562627195176,
      "loss": 5.3239,
      "mean_token_accuracy": 0.20951552391052247,
      "num_tokens": 95408292.0,
      "step": 48770
    },
    {
      "entropy": 6.024379682540894,
      "epoch": 4.875593542260209,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028742062819432955,
      "loss": 5.3719,
      "mean_token_accuracy": 0.19547915011644362,
      "num_tokens": 95417422.0,
      "step": 48775
    },
    {
      "entropy": 5.924511337280274,
      "epoch": 4.876093367321437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002873849933566417,
      "loss": 5.3596,
      "mean_token_accuracy": 0.20758183300495148,
      "num_tokens": 95428889.0,
      "step": 48780
    },
    {
      "entropy": 5.913654899597168,
      "epoch": 4.876593192382666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002873493582073506,
      "loss": 5.2931,
      "mean_token_accuracy": 0.2112714096903801,
      "num_tokens": 95437836.0,
      "step": 48785
    },
    {
      "entropy": 5.905035877227784,
      "epoch": 4.877093017443895,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002873137227473527,
      "loss": 5.2589,
      "mean_token_accuracy": 0.21717590391635894,
      "num_tokens": 95447448.0,
      "step": 48790
    },
    {
      "entropy": 5.956683731079101,
      "epoch": 4.877592842505123,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002872780869775449,
      "loss": 5.3653,
      "mean_token_accuracy": 0.207154344022274,
      "num_tokens": 95457781.0,
      "step": 48795
    },
    {
      "entropy": 5.999612188339233,
      "epoch": 4.878092667566352,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002872424508988236,
      "loss": 5.3741,
      "mean_token_accuracy": 0.19942611902952195,
      "num_tokens": 95467890.0,
      "step": 48800
    },
    {
      "entropy": 5.90711612701416,
      "epoch": 4.878592492627581,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002872068145120853,
      "loss": 5.2677,
      "mean_token_accuracy": 0.2153613656759262,
      "num_tokens": 95477344.0,
      "step": 48805
    },
    {
      "entropy": 6.044378232955933,
      "epoch": 4.879092317688809,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002871711778182267,
      "loss": 5.4769,
      "mean_token_accuracy": 0.19427685290575028,
      "num_tokens": 95487753.0,
      "step": 48810
    },
    {
      "entropy": 5.9324438095092775,
      "epoch": 4.879592142750037,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002871355408181444,
      "loss": 5.2741,
      "mean_token_accuracy": 0.21886733770370484,
      "num_tokens": 95496389.0,
      "step": 48815
    },
    {
      "entropy": 5.962320137023926,
      "epoch": 4.880091967811266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000287099903512735,
      "loss": 5.3487,
      "mean_token_accuracy": 0.21419897228479384,
      "num_tokens": 95507270.0,
      "step": 48820
    },
    {
      "entropy": 5.924404430389404,
      "epoch": 4.880591792872495,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028706426590289515,
      "loss": 5.3246,
      "mean_token_accuracy": 0.20878807306289673,
      "num_tokens": 95516611.0,
      "step": 48825
    },
    {
      "entropy": 5.94369592666626,
      "epoch": 4.881091617933723,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028702862798952143,
      "loss": 5.3239,
      "mean_token_accuracy": 0.21034125834703446,
      "num_tokens": 95525835.0,
      "step": 48830
    },
    {
      "entropy": 5.8875938892364506,
      "epoch": 4.881591442994952,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002869929897735106,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22773523926734923,
      "num_tokens": 95535365.0,
      "step": 48835
    },
    {
      "entropy": 5.928745412826538,
      "epoch": 4.882091268056181,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028695735125575924,
      "loss": 5.2877,
      "mean_token_accuracy": 0.2146351382136345,
      "num_tokens": 95544429.0,
      "step": 48840
    },
    {
      "entropy": 5.8098241806030275,
      "epoch": 4.882591093117409,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028692171243716394,
      "loss": 5.1242,
      "mean_token_accuracy": 0.23357396870851516,
      "num_tokens": 95554130.0,
      "step": 48845
    },
    {
      "entropy": 5.893874406814575,
      "epoch": 4.883090918178637,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028688607331862144,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22136051803827286,
      "num_tokens": 95564477.0,
      "step": 48850
    },
    {
      "entropy": 5.948016119003296,
      "epoch": 4.883590743239866,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028685043390102845,
      "loss": 5.3288,
      "mean_token_accuracy": 0.20668427646160126,
      "num_tokens": 95574067.0,
      "step": 48855
    },
    {
      "entropy": 5.868666410446167,
      "epoch": 4.884090568301095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002868147941852815,
      "loss": 5.2306,
      "mean_token_accuracy": 0.22031570225954056,
      "num_tokens": 95583412.0,
      "step": 48860
    },
    {
      "entropy": 5.854015588760376,
      "epoch": 4.884590393362323,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002867791541722774,
      "loss": 5.2355,
      "mean_token_accuracy": 0.2174099162220955,
      "num_tokens": 95593136.0,
      "step": 48865
    },
    {
      "entropy": 5.885258913040161,
      "epoch": 4.885090218423552,
      "grad_norm": 1.25,
      "learning_rate": 0.0002867435138629129,
      "loss": 5.3315,
      "mean_token_accuracy": 0.21276093572378157,
      "num_tokens": 95602645.0,
      "step": 48870
    },
    {
      "entropy": 5.95139479637146,
      "epoch": 4.885590043484781,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002867078732580845,
      "loss": 5.2453,
      "mean_token_accuracy": 0.220784929394722,
      "num_tokens": 95611224.0,
      "step": 48875
    },
    {
      "entropy": 5.8737386703491214,
      "epoch": 4.886089868546009,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00028667223235868906,
      "loss": 5.2857,
      "mean_token_accuracy": 0.22394140511751176,
      "num_tokens": 95621070.0,
      "step": 48880
    },
    {
      "entropy": 5.888996315002442,
      "epoch": 4.886589693607237,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028663659116562323,
      "loss": 5.2556,
      "mean_token_accuracy": 0.21877443194389343,
      "num_tokens": 95630306.0,
      "step": 48885
    },
    {
      "entropy": 5.9175361633300785,
      "epoch": 4.887089518668466,
      "grad_norm": 1.25,
      "learning_rate": 0.0002866009496797838,
      "loss": 5.3316,
      "mean_token_accuracy": 0.20653265416622163,
      "num_tokens": 95639890.0,
      "step": 48890
    },
    {
      "entropy": 5.909066390991211,
      "epoch": 4.887589343729695,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002865653079020674,
      "loss": 5.3657,
      "mean_token_accuracy": 0.2138603389263153,
      "num_tokens": 95650922.0,
      "step": 48895
    },
    {
      "entropy": 5.7217426776885985,
      "epoch": 4.888089168790923,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002865296658333709,
      "loss": 5.0733,
      "mean_token_accuracy": 0.2330014303326607,
      "num_tokens": 95659934.0,
      "step": 48900
    },
    {
      "entropy": 6.026555728912354,
      "epoch": 4.888588993852152,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028649402347459096,
      "loss": 5.4629,
      "mean_token_accuracy": 0.19955036044120789,
      "num_tokens": 95669303.0,
      "step": 48905
    },
    {
      "entropy": 6.0351417541503904,
      "epoch": 4.889088818913381,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002864583808266244,
      "loss": 5.4613,
      "mean_token_accuracy": 0.20397460013628005,
      "num_tokens": 95679349.0,
      "step": 48910
    },
    {
      "entropy": 5.987450647354126,
      "epoch": 4.889588643974609,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028642273789036783,
      "loss": 5.3483,
      "mean_token_accuracy": 0.20836348831653595,
      "num_tokens": 95689016.0,
      "step": 48915
    },
    {
      "entropy": 5.788338088989258,
      "epoch": 4.890088469035837,
      "grad_norm": 1.125,
      "learning_rate": 0.00028638709466671813,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2246704712510109,
      "num_tokens": 95699312.0,
      "step": 48920
    },
    {
      "entropy": 5.926684904098511,
      "epoch": 4.890588294097066,
      "grad_norm": 1.125,
      "learning_rate": 0.00028635145115657207,
      "loss": 5.3063,
      "mean_token_accuracy": 0.20443211644887924,
      "num_tokens": 95710566.0,
      "step": 48925
    },
    {
      "entropy": 5.9843799591064455,
      "epoch": 4.891088119158295,
      "grad_norm": 1.125,
      "learning_rate": 0.00028631580736082645,
      "loss": 5.3969,
      "mean_token_accuracy": 0.2047044515609741,
      "num_tokens": 95722165.0,
      "step": 48930
    },
    {
      "entropy": 5.930546283721924,
      "epoch": 4.891587944219523,
      "grad_norm": 1.125,
      "learning_rate": 0.000286280163280378,
      "loss": 5.2317,
      "mean_token_accuracy": 0.21925064027309418,
      "num_tokens": 95731466.0,
      "step": 48935
    },
    {
      "entropy": 5.896452474594116,
      "epoch": 4.892087769280752,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028624451891612356,
      "loss": 5.2845,
      "mean_token_accuracy": 0.21915989220142365,
      "num_tokens": 95742503.0,
      "step": 48940
    },
    {
      "entropy": 5.955569934844971,
      "epoch": 4.892587594341981,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028620887426895994,
      "loss": 5.3344,
      "mean_token_accuracy": 0.21680602431297302,
      "num_tokens": 95751106.0,
      "step": 48945
    },
    {
      "entropy": 5.906521034240723,
      "epoch": 4.893087419403209,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002861732293397839,
      "loss": 5.2237,
      "mean_token_accuracy": 0.2195459470152855,
      "num_tokens": 95760004.0,
      "step": 48950
    },
    {
      "entropy": 5.7522156715393065,
      "epoch": 4.893587244464437,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002861375841294923,
      "loss": 5.1438,
      "mean_token_accuracy": 0.22911949157714845,
      "num_tokens": 95770381.0,
      "step": 48955
    },
    {
      "entropy": 5.8090479373931885,
      "epoch": 4.894087069525666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002861019386389819,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22539485096931458,
      "num_tokens": 95780737.0,
      "step": 48960
    },
    {
      "entropy": 5.897252893447876,
      "epoch": 4.894586894586895,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002860662928691496,
      "loss": 5.3102,
      "mean_token_accuracy": 0.2150367185473442,
      "num_tokens": 95791224.0,
      "step": 48965
    },
    {
      "entropy": 5.853621768951416,
      "epoch": 4.895086719648123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028603064682089225,
      "loss": 5.1831,
      "mean_token_accuracy": 0.22405924052000045,
      "num_tokens": 95800857.0,
      "step": 48970
    },
    {
      "entropy": 5.925692510604859,
      "epoch": 4.895586544709352,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028599500049510666,
      "loss": 5.3238,
      "mean_token_accuracy": 0.22425908744335174,
      "num_tokens": 95810993.0,
      "step": 48975
    },
    {
      "entropy": 5.923717498779297,
      "epoch": 4.896086369770581,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002859593538926897,
      "loss": 5.3619,
      "mean_token_accuracy": 0.20541215240955352,
      "num_tokens": 95820799.0,
      "step": 48980
    },
    {
      "entropy": 6.004048156738281,
      "epoch": 4.896586194831809,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002859237070145381,
      "loss": 5.4099,
      "mean_token_accuracy": 0.19816433191299437,
      "num_tokens": 95830452.0,
      "step": 48985
    },
    {
      "entropy": 5.976240873336792,
      "epoch": 4.897086019893037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00028588805986154897,
      "loss": 5.4322,
      "mean_token_accuracy": 0.20652838945388793,
      "num_tokens": 95841060.0,
      "step": 48990
    },
    {
      "entropy": 5.951827526092529,
      "epoch": 4.897585844954266,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000285852412434619,
      "loss": 5.3033,
      "mean_token_accuracy": 0.20448716580867768,
      "num_tokens": 95851083.0,
      "step": 48995
    },
    {
      "entropy": 5.990576457977295,
      "epoch": 4.898085670015495,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028581676473464517,
      "loss": 5.3958,
      "mean_token_accuracy": 0.21498326510190963,
      "num_tokens": 95861094.0,
      "step": 49000
    },
    {
      "entropy": 5.90414457321167,
      "epoch": 4.898585495076723,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00028578111676252427,
      "loss": 5.2888,
      "mean_token_accuracy": 0.2123285710811615,
      "num_tokens": 95870519.0,
      "step": 49005
    },
    {
      "entropy": 5.850120782852173,
      "epoch": 4.899085320137952,
      "grad_norm": 1.125,
      "learning_rate": 0.0002857454685191533,
      "loss": 5.178,
      "mean_token_accuracy": 0.21925076693296433,
      "num_tokens": 95880006.0,
      "step": 49010
    },
    {
      "entropy": 5.919583797454834,
      "epoch": 4.899585145199181,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002857098200054291,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21334993988275527,
      "num_tokens": 95888442.0,
      "step": 49015
    },
    {
      "entropy": 5.969856262207031,
      "epoch": 4.900084970260409,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028567417122224854,
      "loss": 5.3477,
      "mean_token_accuracy": 0.20854030847549437,
      "num_tokens": 95897959.0,
      "step": 49020
    },
    {
      "entropy": 5.8965527534484865,
      "epoch": 4.900584795321637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002856385221705086,
      "loss": 5.3541,
      "mean_token_accuracy": 0.21468397080898285,
      "num_tokens": 95907976.0,
      "step": 49025
    },
    {
      "entropy": 5.8720485210418705,
      "epoch": 4.901084620382866,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028560287285110615,
      "loss": 5.2861,
      "mean_token_accuracy": 0.21892774552106858,
      "num_tokens": 95917908.0,
      "step": 49030
    },
    {
      "entropy": 5.947993803024292,
      "epoch": 4.901584445444095,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002855672232649382,
      "loss": 5.336,
      "mean_token_accuracy": 0.2085599735379219,
      "num_tokens": 95929311.0,
      "step": 49035
    },
    {
      "entropy": 5.987367725372314,
      "epoch": 4.902084270505323,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002855315734129016,
      "loss": 5.3889,
      "mean_token_accuracy": 0.2068818688392639,
      "num_tokens": 95938546.0,
      "step": 49040
    },
    {
      "entropy": 5.9505308151245115,
      "epoch": 4.902584095566552,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028549592329589333,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2138642266392708,
      "num_tokens": 95948057.0,
      "step": 49045
    },
    {
      "entropy": 5.9046660423278805,
      "epoch": 4.903083920627781,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028546027291481035,
      "loss": 5.3636,
      "mean_token_accuracy": 0.20583869367837906,
      "num_tokens": 95957760.0,
      "step": 49050
    },
    {
      "entropy": 5.97857985496521,
      "epoch": 4.903583745689009,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002854246222705496,
      "loss": 5.3172,
      "mean_token_accuracy": 0.2127836599946022,
      "num_tokens": 95968042.0,
      "step": 49055
    },
    {
      "entropy": 5.869814157485962,
      "epoch": 4.904083570750237,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028538897136400797,
      "loss": 5.2183,
      "mean_token_accuracy": 0.22264675945043563,
      "num_tokens": 95977663.0,
      "step": 49060
    },
    {
      "entropy": 5.963737964630127,
      "epoch": 4.904583395811466,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028535332019608256,
      "loss": 5.3113,
      "mean_token_accuracy": 0.20724422484636307,
      "num_tokens": 95987518.0,
      "step": 49065
    },
    {
      "entropy": 5.918208074569702,
      "epoch": 4.905083220872695,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028531766876767025,
      "loss": 5.2401,
      "mean_token_accuracy": 0.21125750094652176,
      "num_tokens": 95995847.0,
      "step": 49070
    },
    {
      "entropy": 5.8903509140014645,
      "epoch": 4.905583045933923,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028528201707966807,
      "loss": 5.236,
      "mean_token_accuracy": 0.21659648269414902,
      "num_tokens": 96004880.0,
      "step": 49075
    },
    {
      "entropy": 5.861894941329956,
      "epoch": 4.906082870995152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028524636513297297,
      "loss": 5.2908,
      "mean_token_accuracy": 0.22344912737607955,
      "num_tokens": 96014780.0,
      "step": 49080
    },
    {
      "entropy": 5.986183929443359,
      "epoch": 4.906582696056381,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028521071292848197,
      "loss": 5.4029,
      "mean_token_accuracy": 0.20217309445142745,
      "num_tokens": 96025012.0,
      "step": 49085
    },
    {
      "entropy": 5.892695140838623,
      "epoch": 4.907082521117609,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002851750604670921,
      "loss": 5.2511,
      "mean_token_accuracy": 0.2109474778175354,
      "num_tokens": 96035020.0,
      "step": 49090
    },
    {
      "entropy": 5.98517107963562,
      "epoch": 4.907582346178837,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002851394077497003,
      "loss": 5.2998,
      "mean_token_accuracy": 0.20824822634458542,
      "num_tokens": 96044719.0,
      "step": 49095
    },
    {
      "entropy": 5.862242937088013,
      "epoch": 4.908082171240066,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002851037547772035,
      "loss": 5.2021,
      "mean_token_accuracy": 0.2216073215007782,
      "num_tokens": 96054868.0,
      "step": 49100
    },
    {
      "entropy": 5.940825891494751,
      "epoch": 4.908581996301295,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028506810155049894,
      "loss": 5.3394,
      "mean_token_accuracy": 0.21341679543256759,
      "num_tokens": 96065663.0,
      "step": 49105
    },
    {
      "entropy": 5.920403337478637,
      "epoch": 4.909081821362523,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002850324480704835,
      "loss": 5.3684,
      "mean_token_accuracy": 0.20848768055438996,
      "num_tokens": 96075559.0,
      "step": 49110
    },
    {
      "entropy": 5.983905839920044,
      "epoch": 4.909581646423752,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002849967943380543,
      "loss": 5.3938,
      "mean_token_accuracy": 0.2050835058093071,
      "num_tokens": 96086047.0,
      "step": 49115
    },
    {
      "entropy": 5.992247915267944,
      "epoch": 4.910081471484981,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002849611403541083,
      "loss": 5.3298,
      "mean_token_accuracy": 0.2091886967420578,
      "num_tokens": 96096377.0,
      "step": 49120
    },
    {
      "entropy": 5.958574199676514,
      "epoch": 4.910581296546209,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028492548611954257,
      "loss": 5.3674,
      "mean_token_accuracy": 0.20747523456811906,
      "num_tokens": 96106088.0,
      "step": 49125
    },
    {
      "entropy": 5.837443733215332,
      "epoch": 4.911081121607437,
      "grad_norm": 1.125,
      "learning_rate": 0.0002848898316352541,
      "loss": 5.2002,
      "mean_token_accuracy": 0.21927003115415572,
      "num_tokens": 96115387.0,
      "step": 49130
    },
    {
      "entropy": 5.872136640548706,
      "epoch": 4.911580946668666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028485417690214004,
      "loss": 5.3398,
      "mean_token_accuracy": 0.20763092190027238,
      "num_tokens": 96124827.0,
      "step": 49135
    },
    {
      "entropy": 5.906379652023316,
      "epoch": 4.912080771729895,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002848185219210975,
      "loss": 5.3603,
      "mean_token_accuracy": 0.21621348112821578,
      "num_tokens": 96134302.0,
      "step": 49140
    },
    {
      "entropy": 6.009146499633789,
      "epoch": 4.912580596791123,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028478286669302336,
      "loss": 5.2945,
      "mean_token_accuracy": 0.20870391577482222,
      "num_tokens": 96144215.0,
      "step": 49145
    },
    {
      "entropy": 5.901526308059692,
      "epoch": 4.913080421852352,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028474721121881497,
      "loss": 5.2481,
      "mean_token_accuracy": 0.21762584745883942,
      "num_tokens": 96154045.0,
      "step": 49150
    },
    {
      "entropy": 5.902645683288574,
      "epoch": 4.91358024691358,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002847115554993691,
      "loss": 5.3398,
      "mean_token_accuracy": 0.2111796036362648,
      "num_tokens": 96164620.0,
      "step": 49155
    },
    {
      "entropy": 5.897151708602905,
      "epoch": 4.914080071974809,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002846758995355831,
      "loss": 5.3039,
      "mean_token_accuracy": 0.21017366796731948,
      "num_tokens": 96174380.0,
      "step": 49160
    },
    {
      "entropy": 5.843435192108155,
      "epoch": 4.914579897036037,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028464024332835384,
      "loss": 5.2048,
      "mean_token_accuracy": 0.22830493301153182,
      "num_tokens": 96183951.0,
      "step": 49165
    },
    {
      "entropy": 5.999520635604858,
      "epoch": 4.915079722097266,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028460458687857865,
      "loss": 5.4576,
      "mean_token_accuracy": 0.20382989048957825,
      "num_tokens": 96193913.0,
      "step": 49170
    },
    {
      "entropy": 5.877347135543824,
      "epoch": 4.915579547158495,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028456893018715453,
      "loss": 5.2366,
      "mean_token_accuracy": 0.2186734274029732,
      "num_tokens": 96202932.0,
      "step": 49175
    },
    {
      "entropy": 5.85935640335083,
      "epoch": 4.916079372219723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002845332732549785,
      "loss": 5.2304,
      "mean_token_accuracy": 0.2178074300289154,
      "num_tokens": 96211865.0,
      "step": 49180
    },
    {
      "entropy": 5.911427974700928,
      "epoch": 4.916579197280952,
      "grad_norm": 1.125,
      "learning_rate": 0.0002844976160829478,
      "loss": 5.3597,
      "mean_token_accuracy": 0.20987193286418915,
      "num_tokens": 96222217.0,
      "step": 49185
    },
    {
      "entropy": 5.929151487350464,
      "epoch": 4.91707902234218,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028446195867195954,
      "loss": 5.2887,
      "mean_token_accuracy": 0.21105606406927108,
      "num_tokens": 96232016.0,
      "step": 49190
    },
    {
      "entropy": 5.924783611297608,
      "epoch": 4.917578847403409,
      "grad_norm": 1.375,
      "learning_rate": 0.0002844263010229108,
      "loss": 5.3416,
      "mean_token_accuracy": 0.21112684905529022,
      "num_tokens": 96242464.0,
      "step": 49195
    },
    {
      "entropy": 5.98315463066101,
      "epoch": 4.918078672464637,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002843906431366988,
      "loss": 5.3645,
      "mean_token_accuracy": 0.2020418167114258,
      "num_tokens": 96252868.0,
      "step": 49200
    },
    {
      "entropy": 5.977819299697876,
      "epoch": 4.918578497525866,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028435498501422064,
      "loss": 5.3226,
      "mean_token_accuracy": 0.2201602876186371,
      "num_tokens": 96262417.0,
      "step": 49205
    },
    {
      "entropy": 5.900298833847046,
      "epoch": 4.919078322587095,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002843193266563734,
      "loss": 5.2854,
      "mean_token_accuracy": 0.2115044429898262,
      "num_tokens": 96272255.0,
      "step": 49210
    },
    {
      "entropy": 5.842066860198974,
      "epoch": 4.919578147648323,
      "grad_norm": 1.203125,
      "learning_rate": 0.00028428366806405436,
      "loss": 5.1453,
      "mean_token_accuracy": 0.2291274607181549,
      "num_tokens": 96280810.0,
      "step": 49215
    },
    {
      "entropy": 5.8889494895935055,
      "epoch": 4.920077972709551,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028424800923816066,
      "loss": 5.3279,
      "mean_token_accuracy": 0.21356389224529265,
      "num_tokens": 96291004.0,
      "step": 49220
    },
    {
      "entropy": 5.864847803115845,
      "epoch": 4.92057779777078,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028421235017958937,
      "loss": 5.2836,
      "mean_token_accuracy": 0.22207360863685607,
      "num_tokens": 96300747.0,
      "step": 49225
    },
    {
      "entropy": 5.940236282348633,
      "epoch": 4.921077622832009,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002841766908892377,
      "loss": 5.3637,
      "mean_token_accuracy": 0.20337270051240922,
      "num_tokens": 96310621.0,
      "step": 49230
    },
    {
      "entropy": 5.961501169204712,
      "epoch": 4.921577447893237,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028414103136800283,
      "loss": 5.26,
      "mean_token_accuracy": 0.22042774260044098,
      "num_tokens": 96320968.0,
      "step": 49235
    },
    {
      "entropy": 5.970427322387695,
      "epoch": 4.922077272954466,
      "grad_norm": 1.171875,
      "learning_rate": 0.00028410537161678204,
      "loss": 5.3522,
      "mean_token_accuracy": 0.21319353133440017,
      "num_tokens": 96329810.0,
      "step": 49240
    },
    {
      "entropy": 5.880372667312622,
      "epoch": 4.922577098015695,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002840697116364724,
      "loss": 5.2712,
      "mean_token_accuracy": 0.21719871163368226,
      "num_tokens": 96340009.0,
      "step": 49245
    },
    {
      "entropy": 5.898825454711914,
      "epoch": 4.923076923076923,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002840340514279711,
      "loss": 5.2952,
      "mean_token_accuracy": 0.21411075592041015,
      "num_tokens": 96349668.0,
      "step": 49250
    },
    {
      "entropy": 5.887667036056518,
      "epoch": 4.923576748138151,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002839983909921755,
      "loss": 5.2297,
      "mean_token_accuracy": 0.2203725039958954,
      "num_tokens": 96359496.0,
      "step": 49255
    },
    {
      "entropy": 5.9340314865112305,
      "epoch": 4.92407657319938,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002839627303299826,
      "loss": 5.3301,
      "mean_token_accuracy": 0.21007902026176453,
      "num_tokens": 96369280.0,
      "step": 49260
    },
    {
      "entropy": 5.81971435546875,
      "epoch": 4.924576398260609,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028392706944228976,
      "loss": 5.1257,
      "mean_token_accuracy": 0.23530785739421844,
      "num_tokens": 96380645.0,
      "step": 49265
    },
    {
      "entropy": 5.9661938667297365,
      "epoch": 4.925076223321837,
      "grad_norm": 1.265625,
      "learning_rate": 0.00028389140832999415,
      "loss": 5.3446,
      "mean_token_accuracy": 0.20928442776203154,
      "num_tokens": 96390275.0,
      "step": 49270
    },
    {
      "entropy": 6.022770929336548,
      "epoch": 4.925576048383066,
      "grad_norm": 1.140625,
      "learning_rate": 0.000283855746993993,
      "loss": 5.4591,
      "mean_token_accuracy": 0.2038000375032425,
      "num_tokens": 96400847.0,
      "step": 49275
    },
    {
      "entropy": 5.9828084945678714,
      "epoch": 4.926075873444295,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002838200854351835,
      "loss": 5.4094,
      "mean_token_accuracy": 0.20244702845811843,
      "num_tokens": 96410716.0,
      "step": 49280
    },
    {
      "entropy": 5.8867440700531,
      "epoch": 4.926575698505523,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00028378442365446294,
      "loss": 5.2746,
      "mean_token_accuracy": 0.21456361263990403,
      "num_tokens": 96419882.0,
      "step": 49285
    },
    {
      "entropy": 5.947265720367431,
      "epoch": 4.927075523566751,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002837487616527286,
      "loss": 5.3703,
      "mean_token_accuracy": 0.21110650449991225,
      "num_tokens": 96429189.0,
      "step": 49290
    },
    {
      "entropy": 5.836133241653442,
      "epoch": 4.92757534862798,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002837130994308776,
      "loss": 5.1661,
      "mean_token_accuracy": 0.23048955649137498,
      "num_tokens": 96439582.0,
      "step": 49295
    },
    {
      "entropy": 5.933042716979981,
      "epoch": 4.9280751736892086,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028367743698980727,
      "loss": 5.3389,
      "mean_token_accuracy": 0.215123251080513,
      "num_tokens": 96448472.0,
      "step": 49300
    },
    {
      "entropy": 5.8984418392181395,
      "epoch": 4.928574998750437,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002836417743304149,
      "loss": 5.3652,
      "mean_token_accuracy": 0.20790830701589585,
      "num_tokens": 96459113.0,
      "step": 49305
    },
    {
      "entropy": 5.851281595230103,
      "epoch": 4.929074823811666,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002836061114535976,
      "loss": 5.2478,
      "mean_token_accuracy": 0.2206502676010132,
      "num_tokens": 96468470.0,
      "step": 49310
    },
    {
      "entropy": 5.963623714447022,
      "epoch": 4.929574648872895,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002835704483602528,
      "loss": 5.2718,
      "mean_token_accuracy": 0.2199672728776932,
      "num_tokens": 96477571.0,
      "step": 49315
    },
    {
      "entropy": 5.9306800842285154,
      "epoch": 4.930074473934123,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002835347850512778,
      "loss": 5.3278,
      "mean_token_accuracy": 0.2129398226737976,
      "num_tokens": 96488131.0,
      "step": 49320
    },
    {
      "entropy": 5.895917749404907,
      "epoch": 4.930574298995351,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028349912152756974,
      "loss": 5.2761,
      "mean_token_accuracy": 0.22098690867424012,
      "num_tokens": 96498535.0,
      "step": 49325
    },
    {
      "entropy": 5.920440196990967,
      "epoch": 4.93107412405658,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002834634577900259,
      "loss": 5.3959,
      "mean_token_accuracy": 0.20181311517953873,
      "num_tokens": 96508943.0,
      "step": 49330
    },
    {
      "entropy": 5.890058755874634,
      "epoch": 4.9315739491178086,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028342779383954374,
      "loss": 5.2379,
      "mean_token_accuracy": 0.2247679501771927,
      "num_tokens": 96519141.0,
      "step": 49335
    },
    {
      "entropy": 5.9157829761505125,
      "epoch": 4.932073774179037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028339212967702046,
      "loss": 5.3351,
      "mean_token_accuracy": 0.20336789041757583,
      "num_tokens": 96528470.0,
      "step": 49340
    },
    {
      "entropy": 5.759436178207397,
      "epoch": 4.932573599240266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028335646530335337,
      "loss": 5.1241,
      "mean_token_accuracy": 0.2329138621687889,
      "num_tokens": 96538447.0,
      "step": 49345
    },
    {
      "entropy": 5.905398654937744,
      "epoch": 4.933073424301495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00028332080071943964,
      "loss": 5.2339,
      "mean_token_accuracy": 0.2181040331721306,
      "num_tokens": 96547377.0,
      "step": 49350
    },
    {
      "entropy": 5.912131071090698,
      "epoch": 4.933573249362723,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002832851359261768,
      "loss": 5.2788,
      "mean_token_accuracy": 0.21853470355272292,
      "num_tokens": 96557572.0,
      "step": 49355
    },
    {
      "entropy": 5.921447467803955,
      "epoch": 4.934073074423951,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028324947092446214,
      "loss": 5.34,
      "mean_token_accuracy": 0.20506581813097,
      "num_tokens": 96567439.0,
      "step": 49360
    },
    {
      "entropy": 5.964074230194091,
      "epoch": 4.93457289948518,
      "grad_norm": 1.234375,
      "learning_rate": 0.00028321380571519285,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21766191571950913,
      "num_tokens": 96577294.0,
      "step": 49365
    },
    {
      "entropy": 5.891913175582886,
      "epoch": 4.9350727245464086,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002831781402992663,
      "loss": 5.2795,
      "mean_token_accuracy": 0.20836576223373413,
      "num_tokens": 96586608.0,
      "step": 49370
    },
    {
      "entropy": 5.838748025894165,
      "epoch": 4.935572549607637,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028314247467757986,
      "loss": 5.2002,
      "mean_token_accuracy": 0.22455774396657943,
      "num_tokens": 96596681.0,
      "step": 49375
    },
    {
      "entropy": 5.96092643737793,
      "epoch": 4.936072374668866,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002831068088510308,
      "loss": 5.3898,
      "mean_token_accuracy": 0.2008427932858467,
      "num_tokens": 96606546.0,
      "step": 49380
    },
    {
      "entropy": 5.871597623825073,
      "epoch": 4.936572199730095,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002830711428205165,
      "loss": 5.244,
      "mean_token_accuracy": 0.22806657254695892,
      "num_tokens": 96616249.0,
      "step": 49385
    },
    {
      "entropy": 6.030178165435791,
      "epoch": 4.937072024791323,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00028303547658693447,
      "loss": 5.3576,
      "mean_token_accuracy": 0.2007418617606163,
      "num_tokens": 96626435.0,
      "step": 49390
    },
    {
      "entropy": 5.975962114334107,
      "epoch": 4.937571849852551,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002829998101511818,
      "loss": 5.4253,
      "mean_token_accuracy": 0.20218484252691268,
      "num_tokens": 96636583.0,
      "step": 49395
    },
    {
      "entropy": 5.927289199829102,
      "epoch": 4.93807167491378,
      "grad_norm": 1.265625,
      "learning_rate": 0.000282964143514156,
      "loss": 5.3283,
      "mean_token_accuracy": 0.21081500500440598,
      "num_tokens": 96645970.0,
      "step": 49400
    },
    {
      "entropy": 5.9371391296386715,
      "epoch": 4.9385714999750085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00028292847667675444,
      "loss": 5.2855,
      "mean_token_accuracy": 0.2188485488295555,
      "num_tokens": 96655798.0,
      "step": 49405
    },
    {
      "entropy": 5.914081764221192,
      "epoch": 4.939071325036237,
      "grad_norm": 1.125,
      "learning_rate": 0.0002828928096398743,
      "loss": 5.2789,
      "mean_token_accuracy": 0.21245219558477402,
      "num_tokens": 96665890.0,
      "step": 49410
    },
    {
      "entropy": 5.962475299835205,
      "epoch": 4.939571150097466,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028285714240441326,
      "loss": 5.3602,
      "mean_token_accuracy": 0.20782303363084792,
      "num_tokens": 96676598.0,
      "step": 49415
    },
    {
      "entropy": 5.901537609100342,
      "epoch": 4.940070975158695,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002828214749712685,
      "loss": 5.3112,
      "mean_token_accuracy": 0.2119879260659218,
      "num_tokens": 96686121.0,
      "step": 49420
    },
    {
      "entropy": 5.931945848464966,
      "epoch": 4.940570800219923,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002827858073413374,
      "loss": 5.3875,
      "mean_token_accuracy": 0.20800455659627914,
      "num_tokens": 96696666.0,
      "step": 49425
    },
    {
      "entropy": 5.921640491485595,
      "epoch": 4.941070625281151,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002827501395155174,
      "loss": 5.2823,
      "mean_token_accuracy": 0.20893157869577408,
      "num_tokens": 96705571.0,
      "step": 49430
    },
    {
      "entropy": 5.92579665184021,
      "epoch": 4.94157045034238,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002827144714947059,
      "loss": 5.3636,
      "mean_token_accuracy": 0.2143123120069504,
      "num_tokens": 96714595.0,
      "step": 49435
    },
    {
      "entropy": 5.898187494277954,
      "epoch": 4.9420702754036085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002826788032798003,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21068617552518845,
      "num_tokens": 96724404.0,
      "step": 49440
    },
    {
      "entropy": 5.9944745063781735,
      "epoch": 4.942570100464837,
      "grad_norm": 1.125,
      "learning_rate": 0.00028264313487169797,
      "loss": 5.3925,
      "mean_token_accuracy": 0.20440857857465744,
      "num_tokens": 96733713.0,
      "step": 49445
    },
    {
      "entropy": 5.952731609344482,
      "epoch": 4.943069925526066,
      "grad_norm": 1.140625,
      "learning_rate": 0.00028260746627129624,
      "loss": 5.3192,
      "mean_token_accuracy": 0.21900793612003328,
      "num_tokens": 96743160.0,
      "step": 49450
    },
    {
      "entropy": 5.89889965057373,
      "epoch": 4.943569750587295,
      "grad_norm": 1.125,
      "learning_rate": 0.0002825717974794927,
      "loss": 5.2181,
      "mean_token_accuracy": 0.22497717440128326,
      "num_tokens": 96753562.0,
      "step": 49455
    },
    {
      "entropy": 5.868499326705932,
      "epoch": 4.944069575648523,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028253612849718474,
      "loss": 5.2869,
      "mean_token_accuracy": 0.21511799842119217,
      "num_tokens": 96762499.0,
      "step": 49460
    },
    {
      "entropy": 5.907213973999023,
      "epoch": 4.944569400709751,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002825004593252697,
      "loss": 5.3366,
      "mean_token_accuracy": 0.20805287957191468,
      "num_tokens": 96772151.0,
      "step": 49465
    },
    {
      "entropy": 5.942550373077393,
      "epoch": 4.94506922577098,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028246478996464503,
      "loss": 5.3118,
      "mean_token_accuracy": 0.21079307645559311,
      "num_tokens": 96782485.0,
      "step": 49470
    },
    {
      "entropy": 5.963858652114868,
      "epoch": 4.9455690508322085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028242912041620804,
      "loss": 5.325,
      "mean_token_accuracy": 0.20863286405801773,
      "num_tokens": 96793605.0,
      "step": 49475
    },
    {
      "entropy": 5.924398946762085,
      "epoch": 4.946068875893437,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002823934506808565,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21042328476905822,
      "num_tokens": 96802960.0,
      "step": 49480
    },
    {
      "entropy": 5.950355005264282,
      "epoch": 4.946568700954666,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028235778075948756,
      "loss": 5.4263,
      "mean_token_accuracy": 0.1987549215555191,
      "num_tokens": 96811816.0,
      "step": 49485
    },
    {
      "entropy": 5.883904027938843,
      "epoch": 4.947068526015895,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028232211065299864,
      "loss": 5.3473,
      "mean_token_accuracy": 0.20850912481546402,
      "num_tokens": 96822198.0,
      "step": 49490
    },
    {
      "entropy": 5.965698480606079,
      "epoch": 4.947568351077123,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028228644036228746,
      "loss": 5.3507,
      "mean_token_accuracy": 0.20786824077367783,
      "num_tokens": 96831030.0,
      "step": 49495
    },
    {
      "entropy": 5.946888732910156,
      "epoch": 4.948068176138351,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00028225076988825124,
      "loss": 5.3298,
      "mean_token_accuracy": 0.2086975172162056,
      "num_tokens": 96839982.0,
      "step": 49500
    },
    {
      "entropy": 6.0157379627227785,
      "epoch": 4.94856800119958,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028221509923178744,
      "loss": 5.4011,
      "mean_token_accuracy": 0.20272998064756392,
      "num_tokens": 96849073.0,
      "step": 49505
    },
    {
      "entropy": 5.864186429977417,
      "epoch": 4.9490678262608085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002821794283937937,
      "loss": 5.2311,
      "mean_token_accuracy": 0.21925155371427535,
      "num_tokens": 96859937.0,
      "step": 49510
    },
    {
      "entropy": 5.886672258377075,
      "epoch": 4.949567651322037,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002821437573751674,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21646506935358048,
      "num_tokens": 96869103.0,
      "step": 49515
    },
    {
      "entropy": 5.924878120422363,
      "epoch": 4.950067476383266,
      "grad_norm": 1.125,
      "learning_rate": 0.00028210808617680597,
      "loss": 5.3039,
      "mean_token_accuracy": 0.21871919184923172,
      "num_tokens": 96879601.0,
      "step": 49520
    },
    {
      "entropy": 5.902702856063843,
      "epoch": 4.950567301444495,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002820724147996069,
      "loss": 5.3085,
      "mean_token_accuracy": 0.221651192009449,
      "num_tokens": 96889414.0,
      "step": 49525
    },
    {
      "entropy": 6.023726081848144,
      "epoch": 4.951067126505723,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028203674324446776,
      "loss": 5.4117,
      "mean_token_accuracy": 0.20728981792926787,
      "num_tokens": 96898538.0,
      "step": 49530
    },
    {
      "entropy": 5.942532682418824,
      "epoch": 4.951566951566951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028200107151228593,
      "loss": 5.3742,
      "mean_token_accuracy": 0.20765227526426316,
      "num_tokens": 96908351.0,
      "step": 49535
    },
    {
      "entropy": 5.959344339370728,
      "epoch": 4.95206677662818,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002819653996039589,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21842387914657593,
      "num_tokens": 96917789.0,
      "step": 49540
    },
    {
      "entropy": 5.9274375438690186,
      "epoch": 4.9525666016894085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002819297275203842,
      "loss": 5.2544,
      "mean_token_accuracy": 0.2103637307882309,
      "num_tokens": 96927339.0,
      "step": 49545
    },
    {
      "entropy": 5.932569360733032,
      "epoch": 4.953066426750637,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002818940552624594,
      "loss": 5.2704,
      "mean_token_accuracy": 0.22470609843730927,
      "num_tokens": 96937049.0,
      "step": 49550
    },
    {
      "entropy": 5.872968959808349,
      "epoch": 4.953566251811866,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028185838283108195,
      "loss": 5.23,
      "mean_token_accuracy": 0.22202303409576415,
      "num_tokens": 96946754.0,
      "step": 49555
    },
    {
      "entropy": 5.78153076171875,
      "epoch": 4.954066076873095,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002818227102271493,
      "loss": 5.1983,
      "mean_token_accuracy": 0.22296004146337509,
      "num_tokens": 96956526.0,
      "step": 49560
    },
    {
      "entropy": 5.9347621440887455,
      "epoch": 4.954565901934323,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002817870374515591,
      "loss": 5.307,
      "mean_token_accuracy": 0.21047571301460266,
      "num_tokens": 96965670.0,
      "step": 49565
    },
    {
      "entropy": 5.942339086532593,
      "epoch": 4.955065726995551,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028175136450520874,
      "loss": 5.376,
      "mean_token_accuracy": 0.21290843635797502,
      "num_tokens": 96976952.0,
      "step": 49570
    },
    {
      "entropy": 5.958284759521485,
      "epoch": 4.95556555205678,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028171569138899573,
      "loss": 5.3675,
      "mean_token_accuracy": 0.20975681841373445,
      "num_tokens": 96986866.0,
      "step": 49575
    },
    {
      "entropy": 5.865186023712158,
      "epoch": 4.9560653771180085,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028168001810381773,
      "loss": 5.2867,
      "mean_token_accuracy": 0.2163845956325531,
      "num_tokens": 96996732.0,
      "step": 49580
    },
    {
      "entropy": 5.979402685165406,
      "epoch": 4.956565202179237,
      "grad_norm": 1.109375,
      "learning_rate": 0.00028164434465057214,
      "loss": 5.4045,
      "mean_token_accuracy": 0.2031862646341324,
      "num_tokens": 97006820.0,
      "step": 49585
    },
    {
      "entropy": 5.904630327224732,
      "epoch": 4.957065027240466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002816086710301566,
      "loss": 5.2575,
      "mean_token_accuracy": 0.22197743505239487,
      "num_tokens": 97015065.0,
      "step": 49590
    },
    {
      "entropy": 5.976731586456299,
      "epoch": 4.957564852301695,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002815729972434686,
      "loss": 5.302,
      "mean_token_accuracy": 0.21550137996673585,
      "num_tokens": 97024299.0,
      "step": 49595
    },
    {
      "entropy": 5.939073371887207,
      "epoch": 4.958064677362923,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028153732329140563,
      "loss": 5.3457,
      "mean_token_accuracy": 0.21295849978923798,
      "num_tokens": 97035240.0,
      "step": 49600
    },
    {
      "entropy": 5.902547788619995,
      "epoch": 4.958564502424151,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002815016491748653,
      "loss": 5.279,
      "mean_token_accuracy": 0.2129595771431923,
      "num_tokens": 97044872.0,
      "step": 49605
    },
    {
      "entropy": 5.8796204090118405,
      "epoch": 4.95906432748538,
      "grad_norm": 1.375,
      "learning_rate": 0.0002814659748947452,
      "loss": 5.2823,
      "mean_token_accuracy": 0.21943867057561875,
      "num_tokens": 97053417.0,
      "step": 49610
    },
    {
      "entropy": 5.844894027709961,
      "epoch": 4.9595641525466085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002814303004519427,
      "loss": 5.2377,
      "mean_token_accuracy": 0.21198951005935668,
      "num_tokens": 97062927.0,
      "step": 49615
    },
    {
      "entropy": 5.943486452102661,
      "epoch": 4.960063977607837,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002813946258473556,
      "loss": 5.3179,
      "mean_token_accuracy": 0.21705848574638367,
      "num_tokens": 97072311.0,
      "step": 49620
    },
    {
      "entropy": 5.851626825332642,
      "epoch": 4.960563802669066,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028135895108188144,
      "loss": 5.2226,
      "mean_token_accuracy": 0.21732388883829118,
      "num_tokens": 97081827.0,
      "step": 49625
    },
    {
      "entropy": 5.889771318435669,
      "epoch": 4.961063627730295,
      "grad_norm": 1.125,
      "learning_rate": 0.00028132327615641755,
      "loss": 5.2731,
      "mean_token_accuracy": 0.2146648108959198,
      "num_tokens": 97092246.0,
      "step": 49630
    },
    {
      "entropy": 5.873631191253662,
      "epoch": 4.961563452791523,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00028128760107186174,
      "loss": 5.2512,
      "mean_token_accuracy": 0.21900286972522737,
      "num_tokens": 97101922.0,
      "step": 49635
    },
    {
      "entropy": 5.887064743041992,
      "epoch": 4.962063277852751,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028125192582911145,
      "loss": 5.2488,
      "mean_token_accuracy": 0.22286534905433655,
      "num_tokens": 97110941.0,
      "step": 49640
    },
    {
      "entropy": 5.839566087722778,
      "epoch": 4.96256310291398,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002812162504290644,
      "loss": 5.1527,
      "mean_token_accuracy": 0.22692630141973497,
      "num_tokens": 97121128.0,
      "step": 49645
    },
    {
      "entropy": 5.832159042358398,
      "epoch": 4.9630629279752085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00028118057487261806,
      "loss": 5.1725,
      "mean_token_accuracy": 0.23217913657426834,
      "num_tokens": 97131152.0,
      "step": 49650
    },
    {
      "entropy": 5.844907951354981,
      "epoch": 4.963562753036437,
      "grad_norm": 1.09375,
      "learning_rate": 0.00028114489916067006,
      "loss": 5.2225,
      "mean_token_accuracy": 0.21637094765901566,
      "num_tokens": 97140966.0,
      "step": 49655
    },
    {
      "entropy": 5.89504771232605,
      "epoch": 4.964062578097666,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00028110922329411793,
      "loss": 5.3084,
      "mean_token_accuracy": 0.20656504929065705,
      "num_tokens": 97150599.0,
      "step": 49660
    },
    {
      "entropy": 5.907152843475342,
      "epoch": 4.964562403158895,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002810735472738593,
      "loss": 5.335,
      "mean_token_accuracy": 0.20869721919298173,
      "num_tokens": 97160348.0,
      "step": 49665
    },
    {
      "entropy": 5.856291580200195,
      "epoch": 4.965062228220123,
      "grad_norm": 1.125,
      "learning_rate": 0.0002810378711007918,
      "loss": 5.2467,
      "mean_token_accuracy": 0.2264549672603607,
      "num_tokens": 97169637.0,
      "step": 49670
    },
    {
      "entropy": 5.893533992767334,
      "epoch": 4.965562053281351,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002810021947758131,
      "loss": 5.1999,
      "mean_token_accuracy": 0.22880998849868775,
      "num_tokens": 97179168.0,
      "step": 49675
    },
    {
      "entropy": 5.872187519073487,
      "epoch": 4.96606187834258,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002809665182998206,
      "loss": 5.1646,
      "mean_token_accuracy": 0.2300706312060356,
      "num_tokens": 97188890.0,
      "step": 49680
    },
    {
      "entropy": 5.909009122848511,
      "epoch": 4.9665617034038085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028093084167371215,
      "loss": 5.3562,
      "mean_token_accuracy": 0.209211964905262,
      "num_tokens": 97198487.0,
      "step": 49685
    },
    {
      "entropy": 5.845963430404663,
      "epoch": 4.967061528465037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002808951648983852,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22430433183908463,
      "num_tokens": 97207613.0,
      "step": 49690
    },
    {
      "entropy": 5.840523767471313,
      "epoch": 4.967561353526266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002808594879747374,
      "loss": 5.2542,
      "mean_token_accuracy": 0.2200156643986702,
      "num_tokens": 97218533.0,
      "step": 49695
    },
    {
      "entropy": 5.921529769897461,
      "epoch": 4.968061178587495,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002808238109036663,
      "loss": 5.3532,
      "mean_token_accuracy": 0.21665353924036027,
      "num_tokens": 97227884.0,
      "step": 49700
    },
    {
      "entropy": 5.992894506454467,
      "epoch": 4.968561003648723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002807881336860698,
      "loss": 5.3829,
      "mean_token_accuracy": 0.2065457209944725,
      "num_tokens": 97239169.0,
      "step": 49705
    },
    {
      "entropy": 5.951733064651489,
      "epoch": 4.969060828709951,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00028075245632284523,
      "loss": 5.3077,
      "mean_token_accuracy": 0.21649726927280427,
      "num_tokens": 97248681.0,
      "step": 49710
    },
    {
      "entropy": 5.982657718658447,
      "epoch": 4.96956065377118,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002807167788148904,
      "loss": 5.4577,
      "mean_token_accuracy": 0.20228002071380616,
      "num_tokens": 97258159.0,
      "step": 49715
    },
    {
      "entropy": 5.941579675674438,
      "epoch": 4.9700604788324085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00028068110116310287,
      "loss": 5.3003,
      "mean_token_accuracy": 0.2117758274078369,
      "num_tokens": 97267779.0,
      "step": 49720
    },
    {
      "entropy": 5.899548864364624,
      "epoch": 4.970560303893637,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002806454233683803,
      "loss": 5.2675,
      "mean_token_accuracy": 0.22068409621715546,
      "num_tokens": 97276282.0,
      "step": 49725
    },
    {
      "entropy": 5.998537206649781,
      "epoch": 4.971060128954866,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002806097454316202,
      "loss": 5.3657,
      "mean_token_accuracy": 0.20848201215267181,
      "num_tokens": 97285860.0,
      "step": 49730
    },
    {
      "entropy": 5.850255966186523,
      "epoch": 4.971559954016095,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002805740673537205,
      "loss": 5.21,
      "mean_token_accuracy": 0.21517930775880814,
      "num_tokens": 97295795.0,
      "step": 49735
    },
    {
      "entropy": 5.840193223953247,
      "epoch": 4.972059779077323,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002805383891355787,
      "loss": 5.222,
      "mean_token_accuracy": 0.22579921633005143,
      "num_tokens": 97305892.0,
      "step": 49740
    },
    {
      "entropy": 5.847502613067627,
      "epoch": 4.972559604138551,
      "grad_norm": 1.28125,
      "learning_rate": 0.00028050271077809244,
      "loss": 5.2658,
      "mean_token_accuracy": 0.2072983369231224,
      "num_tokens": 97314832.0,
      "step": 49745
    },
    {
      "entropy": 5.927683401107788,
      "epoch": 4.97305942919978,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002804670322821594,
      "loss": 5.349,
      "mean_token_accuracy": 0.20719780921936035,
      "num_tokens": 97323878.0,
      "step": 49750
    },
    {
      "entropy": 5.929150533676148,
      "epoch": 4.9735592542610085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00028043135364867717,
      "loss": 5.2205,
      "mean_token_accuracy": 0.2151888743042946,
      "num_tokens": 97332538.0,
      "step": 49755
    },
    {
      "entropy": 6.069140625,
      "epoch": 4.974059079322237,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00028039567487854347,
      "loss": 5.3925,
      "mean_token_accuracy": 0.21530162245035173,
      "num_tokens": 97341772.0,
      "step": 49760
    },
    {
      "entropy": 5.913581228256225,
      "epoch": 4.974558904383466,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000280359995972656,
      "loss": 5.2778,
      "mean_token_accuracy": 0.21279263943433763,
      "num_tokens": 97350425.0,
      "step": 49765
    },
    {
      "entropy": 5.904622793197632,
      "epoch": 4.975058729444695,
      "grad_norm": 1.25,
      "learning_rate": 0.00028032431693191244,
      "loss": 5.2478,
      "mean_token_accuracy": 0.21384229511022568,
      "num_tokens": 97360843.0,
      "step": 49770
    },
    {
      "entropy": 5.9193987369537355,
      "epoch": 4.975558554505923,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002802886377572105,
      "loss": 5.3051,
      "mean_token_accuracy": 0.21855303645133972,
      "num_tokens": 97371390.0,
      "step": 49775
    },
    {
      "entropy": 5.9770894050598145,
      "epoch": 4.976058379567151,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002802529584494477,
      "loss": 5.385,
      "mean_token_accuracy": 0.20872333645820618,
      "num_tokens": 97381010.0,
      "step": 49780
    },
    {
      "entropy": 5.86341290473938,
      "epoch": 4.97655820462838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002802172790095219,
      "loss": 5.2331,
      "mean_token_accuracy": 0.2177701398730278,
      "num_tokens": 97391503.0,
      "step": 49785
    },
    {
      "entropy": 5.9149669170379635,
      "epoch": 4.9770580296896085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00028018159943833047,
      "loss": 5.29,
      "mean_token_accuracy": 0.2098624125123024,
      "num_tokens": 97401205.0,
      "step": 49790
    },
    {
      "entropy": 6.0309898853302,
      "epoch": 4.977557854750837,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002801459197367716,
      "loss": 5.5618,
      "mean_token_accuracy": 0.1926150441169739,
      "num_tokens": 97412256.0,
      "step": 49795
    },
    {
      "entropy": 5.9280904769897464,
      "epoch": 4.978057679812066,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002801102399057426,
      "loss": 5.3102,
      "mean_token_accuracy": 0.21578384935855865,
      "num_tokens": 97422391.0,
      "step": 49800
    },
    {
      "entropy": 5.965473556518555,
      "epoch": 4.978557504873295,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00028007455994614127,
      "loss": 5.3436,
      "mean_token_accuracy": 0.20973751693964005,
      "num_tokens": 97432213.0,
      "step": 49805
    },
    {
      "entropy": 5.9765716075897215,
      "epoch": 4.979057329934523,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002800388798588654,
      "loss": 5.3466,
      "mean_token_accuracy": 0.20773872435092927,
      "num_tokens": 97441243.0,
      "step": 49810
    },
    {
      "entropy": 5.905559492111206,
      "epoch": 4.979557154995751,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00028000319964481253,
      "loss": 5.341,
      "mean_token_accuracy": 0.21168835312128068,
      "num_tokens": 97450235.0,
      "step": 49815
    },
    {
      "entropy": 5.945976972579956,
      "epoch": 4.98005698005698,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027996751930488045,
      "loss": 5.3855,
      "mean_token_accuracy": 0.2122814640402794,
      "num_tokens": 97460382.0,
      "step": 49820
    },
    {
      "entropy": 5.9499822616577145,
      "epoch": 4.9805568051182085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027993183883996686,
      "loss": 5.3849,
      "mean_token_accuracy": 0.20809912085533142,
      "num_tokens": 97469880.0,
      "step": 49825
    },
    {
      "entropy": 5.858642482757569,
      "epoch": 4.981056630179437,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00027989615825096953,
      "loss": 5.2815,
      "mean_token_accuracy": 0.21795784085988998,
      "num_tokens": 97480060.0,
      "step": 49830
    },
    {
      "entropy": 5.897536087036133,
      "epoch": 4.981556455240666,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027986047753878607,
      "loss": 5.3033,
      "mean_token_accuracy": 0.2107116848230362,
      "num_tokens": 97490042.0,
      "step": 49835
    },
    {
      "entropy": 5.982069635391236,
      "epoch": 4.982056280301895,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027982479670431425,
      "loss": 5.3649,
      "mean_token_accuracy": 0.21125311106443406,
      "num_tokens": 97501023.0,
      "step": 49840
    },
    {
      "entropy": 5.961733865737915,
      "epoch": 4.982556105363123,
      "grad_norm": 1.125,
      "learning_rate": 0.0002797891157484519,
      "loss": 5.284,
      "mean_token_accuracy": 0.214468814432621,
      "num_tokens": 97509718.0,
      "step": 49845
    },
    {
      "entropy": 5.9300134658813475,
      "epoch": 4.983055930424351,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002797534346720965,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2156299740076065,
      "num_tokens": 97518986.0,
      "step": 49850
    },
    {
      "entropy": 5.813758897781372,
      "epoch": 4.98355575548558,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00027971775347614586,
      "loss": 5.2304,
      "mean_token_accuracy": 0.21803992241621017,
      "num_tokens": 97529708.0,
      "step": 49855
    },
    {
      "entropy": 5.9659655570983885,
      "epoch": 4.9840555805468085,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002796820721614978,
      "loss": 5.3517,
      "mean_token_accuracy": 0.21494084149599074,
      "num_tokens": 97539530.0,
      "step": 49860
    },
    {
      "entropy": 6.036644792556762,
      "epoch": 4.984555405608037,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027964639072905007,
      "loss": 5.52,
      "mean_token_accuracy": 0.1940552607178688,
      "num_tokens": 97550632.0,
      "step": 49865
    },
    {
      "entropy": 5.925092697143555,
      "epoch": 4.985055230669266,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027961070917970033,
      "loss": 5.276,
      "mean_token_accuracy": 0.20920155197381973,
      "num_tokens": 97559760.0,
      "step": 49870
    },
    {
      "entropy": 5.890880298614502,
      "epoch": 4.985555055730495,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027957502751434633,
      "loss": 5.2773,
      "mean_token_accuracy": 0.21539511233568193,
      "num_tokens": 97569879.0,
      "step": 49875
    },
    {
      "entropy": 5.832244443893432,
      "epoch": 4.986054880791723,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002795393457338858,
      "loss": 5.1357,
      "mean_token_accuracy": 0.2238672584295273,
      "num_tokens": 97579754.0,
      "step": 49880
    },
    {
      "entropy": 5.897765636444092,
      "epoch": 4.986554705852951,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002795036638392165,
      "loss": 5.2517,
      "mean_token_accuracy": 0.21961734741926192,
      "num_tokens": 97589271.0,
      "step": 49885
    },
    {
      "entropy": 5.971505355834961,
      "epoch": 4.98705453091418,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002794679818312362,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20496093332767487,
      "num_tokens": 97599157.0,
      "step": 49890
    },
    {
      "entropy": 5.864169263839722,
      "epoch": 4.9875543559754085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027943229971084255,
      "loss": 5.235,
      "mean_token_accuracy": 0.21910274028778076,
      "num_tokens": 97608898.0,
      "step": 49895
    },
    {
      "entropy": 5.878494167327881,
      "epoch": 4.988054181036637,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027939661747893344,
      "loss": 5.1916,
      "mean_token_accuracy": 0.2288840964436531,
      "num_tokens": 97618212.0,
      "step": 49900
    },
    {
      "entropy": 5.914476490020752,
      "epoch": 4.988554006097866,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027936093513640655,
      "loss": 5.3094,
      "mean_token_accuracy": 0.22049743682146072,
      "num_tokens": 97627102.0,
      "step": 49905
    },
    {
      "entropy": 5.978977298736572,
      "epoch": 4.989053831159095,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002793252526841597,
      "loss": 5.391,
      "mean_token_accuracy": 0.20985322296619416,
      "num_tokens": 97637206.0,
      "step": 49910
    },
    {
      "entropy": 5.914826154708862,
      "epoch": 4.989553656220323,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002792895701230906,
      "loss": 5.3699,
      "mean_token_accuracy": 0.205813367664814,
      "num_tokens": 97646517.0,
      "step": 49915
    },
    {
      "entropy": 5.842992687225342,
      "epoch": 4.990053481281551,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027925388745409695,
      "loss": 5.2461,
      "mean_token_accuracy": 0.22203669399023057,
      "num_tokens": 97655222.0,
      "step": 49920
    },
    {
      "entropy": 5.905836868286133,
      "epoch": 4.99055330634278,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002792182046780767,
      "loss": 5.251,
      "mean_token_accuracy": 0.21710333377122878,
      "num_tokens": 97664725.0,
      "step": 49925
    },
    {
      "entropy": 6.040573453903198,
      "epoch": 4.9910531314040085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002791825217959274,
      "loss": 5.4278,
      "mean_token_accuracy": 0.20556624829769135,
      "num_tokens": 97675027.0,
      "step": 49930
    },
    {
      "entropy": 5.871016263961792,
      "epoch": 4.991552956465237,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002791468388085469,
      "loss": 5.1295,
      "mean_token_accuracy": 0.22885008603334428,
      "num_tokens": 97684397.0,
      "step": 49935
    },
    {
      "entropy": 5.949992704391479,
      "epoch": 4.992052781526466,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002791111557168331,
      "loss": 5.3841,
      "mean_token_accuracy": 0.20513938814401628,
      "num_tokens": 97694628.0,
      "step": 49940
    },
    {
      "entropy": 6.0330911636352536,
      "epoch": 4.992552606587695,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002790754725216836,
      "loss": 5.4553,
      "mean_token_accuracy": 0.2055195078253746,
      "num_tokens": 97703842.0,
      "step": 49945
    },
    {
      "entropy": 5.899306774139404,
      "epoch": 4.993052431648923,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027903978922399635,
      "loss": 5.2573,
      "mean_token_accuracy": 0.2115131288766861,
      "num_tokens": 97713726.0,
      "step": 49950
    },
    {
      "entropy": 5.969115352630615,
      "epoch": 4.993552256710151,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027900410582466903,
      "loss": 5.3199,
      "mean_token_accuracy": 0.21347678303718567,
      "num_tokens": 97723686.0,
      "step": 49955
    },
    {
      "entropy": 5.953483629226684,
      "epoch": 4.99405208177138,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002789684223245993,
      "loss": 5.3471,
      "mean_token_accuracy": 0.20798405408859252,
      "num_tokens": 97731979.0,
      "step": 49960
    },
    {
      "entropy": 5.960399532318116,
      "epoch": 4.9945519068326085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027893273872468523,
      "loss": 5.3599,
      "mean_token_accuracy": 0.20511974692344664,
      "num_tokens": 97740718.0,
      "step": 49965
    },
    {
      "entropy": 5.9221110343933105,
      "epoch": 4.995051731893837,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002788970550258244,
      "loss": 5.266,
      "mean_token_accuracy": 0.21436766535043716,
      "num_tokens": 97749650.0,
      "step": 49970
    },
    {
      "entropy": 5.924710941314697,
      "epoch": 4.995551556955066,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002788613712289147,
      "loss": 5.297,
      "mean_token_accuracy": 0.2115079790353775,
      "num_tokens": 97758813.0,
      "step": 49975
    },
    {
      "entropy": 5.825883483886718,
      "epoch": 4.996051382016295,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002788256873348539,
      "loss": 5.1277,
      "mean_token_accuracy": 0.22759630233049394,
      "num_tokens": 97768146.0,
      "step": 49980
    },
    {
      "entropy": 5.859021425247192,
      "epoch": 4.996551207077523,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002787900033445398,
      "loss": 5.2795,
      "mean_token_accuracy": 0.21695657074451447,
      "num_tokens": 97777216.0,
      "step": 49985
    },
    {
      "entropy": 5.9390233039855955,
      "epoch": 4.997051032138751,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002787543192588702,
      "loss": 5.3848,
      "mean_token_accuracy": 0.20363133102655412,
      "num_tokens": 97788324.0,
      "step": 49990
    },
    {
      "entropy": 5.87857494354248,
      "epoch": 4.99755085719998,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027871863507874286,
      "loss": 5.2719,
      "mean_token_accuracy": 0.22730314433574678,
      "num_tokens": 97798411.0,
      "step": 49995
    },
    {
      "entropy": 5.921813535690307,
      "epoch": 4.9980506822612085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027868295080505555,
      "loss": 5.2983,
      "mean_token_accuracy": 0.21688667982816695,
      "num_tokens": 97808351.0,
      "step": 50000
    },
    {
      "entropy": 5.933462285995484,
      "epoch": 4.998550507322437,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027864726643870625,
      "loss": 5.3245,
      "mean_token_accuracy": 0.21234092563390733,
      "num_tokens": 97818372.0,
      "step": 50005
    },
    {
      "entropy": 5.98574914932251,
      "epoch": 4.999050332383666,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027861158198059263,
      "loss": 5.4593,
      "mean_token_accuracy": 0.20199592411518097,
      "num_tokens": 97828866.0,
      "step": 50010
    },
    {
      "entropy": 5.859716081619263,
      "epoch": 4.999550157444895,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002785758974316125,
      "loss": 5.231,
      "mean_token_accuracy": 0.21877532303333283,
      "num_tokens": 97839816.0,
      "step": 50015
    },
    {
      "entropy": 5.8568620681762695,
      "epoch": 5.0,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0002785402127926638,
      "loss": 5.2763,
      "mean_token_accuracy": 0.21932174265384674,
      "num_tokens": 97848300.0,
      "step": 50020
    },
    {
      "entropy": 5.8696449279785154,
      "epoch": 5.000499825061229,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002785045280646442,
      "loss": 5.2047,
      "mean_token_accuracy": 0.22153721749782562,
      "num_tokens": 97857399.0,
      "step": 50025
    },
    {
      "entropy": 5.9218391418457035,
      "epoch": 5.000999650122457,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027846884324845157,
      "loss": 5.2503,
      "mean_token_accuracy": 0.21961478441953658,
      "num_tokens": 97866637.0,
      "step": 50030
    },
    {
      "entropy": 5.956479215621949,
      "epoch": 5.001499475183686,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027843315834498374,
      "loss": 5.3006,
      "mean_token_accuracy": 0.2177873209118843,
      "num_tokens": 97876861.0,
      "step": 50035
    },
    {
      "entropy": 5.938960266113281,
      "epoch": 5.001999300244914,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027839747335513856,
      "loss": 5.2183,
      "mean_token_accuracy": 0.21283847093582153,
      "num_tokens": 97886249.0,
      "step": 50040
    },
    {
      "entropy": 5.87806658744812,
      "epoch": 5.002499125306143,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027836178827981374,
      "loss": 5.2173,
      "mean_token_accuracy": 0.2313746154308319,
      "num_tokens": 97896307.0,
      "step": 50045
    },
    {
      "entropy": 5.88699631690979,
      "epoch": 5.002998950367371,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027832610311990727,
      "loss": 5.2144,
      "mean_token_accuracy": 0.217950701713562,
      "num_tokens": 97905393.0,
      "step": 50050
    },
    {
      "entropy": 5.925769233703614,
      "epoch": 5.0034987754286,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002782904178763169,
      "loss": 5.3031,
      "mean_token_accuracy": 0.21272107511758803,
      "num_tokens": 97914443.0,
      "step": 50055
    },
    {
      "entropy": 5.792198371887207,
      "epoch": 5.003998600489829,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027825473254994044,
      "loss": 5.1405,
      "mean_token_accuracy": 0.2341163232922554,
      "num_tokens": 97924894.0,
      "step": 50060
    },
    {
      "entropy": 5.904084634780884,
      "epoch": 5.004498425551057,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002782190471416758,
      "loss": 5.2339,
      "mean_token_accuracy": 0.21352019011974335,
      "num_tokens": 97934162.0,
      "step": 50065
    },
    {
      "entropy": 5.952024745941162,
      "epoch": 5.004998250612286,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002781833616524207,
      "loss": 5.2876,
      "mean_token_accuracy": 0.21864989250898362,
      "num_tokens": 97943949.0,
      "step": 50070
    },
    {
      "entropy": 5.978005218505859,
      "epoch": 5.005498075673514,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027814767608307295,
      "loss": 5.3034,
      "mean_token_accuracy": 0.21677447706460953,
      "num_tokens": 97953778.0,
      "step": 50075
    },
    {
      "entropy": 5.902772998809814,
      "epoch": 5.005997900734743,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027811199043453066,
      "loss": 5.3053,
      "mean_token_accuracy": 0.2158697172999382,
      "num_tokens": 97963076.0,
      "step": 50080
    },
    {
      "entropy": 5.932824230194091,
      "epoch": 5.006497725795971,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002780763047076914,
      "loss": 5.2614,
      "mean_token_accuracy": 0.21355756372213364,
      "num_tokens": 97973005.0,
      "step": 50085
    },
    {
      "entropy": 5.971277189254761,
      "epoch": 5.0069975508572,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002780406189034532,
      "loss": 5.2229,
      "mean_token_accuracy": 0.21922763139009477,
      "num_tokens": 97983309.0,
      "step": 50090
    },
    {
      "entropy": 5.862642097473144,
      "epoch": 5.007497375918429,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027800493302271366,
      "loss": 5.1911,
      "mean_token_accuracy": 0.22372956275939943,
      "num_tokens": 97992559.0,
      "step": 50095
    },
    {
      "entropy": 5.824261617660523,
      "epoch": 5.007997200979657,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027796924706637087,
      "loss": 5.215,
      "mean_token_accuracy": 0.22276999652385712,
      "num_tokens": 98001980.0,
      "step": 50100
    },
    {
      "entropy": 5.955246782302856,
      "epoch": 5.008497026040886,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027793356103532254,
      "loss": 5.3061,
      "mean_token_accuracy": 0.21867656260728835,
      "num_tokens": 98012364.0,
      "step": 50105
    },
    {
      "entropy": 6.0810792446136475,
      "epoch": 5.008996851102114,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002778978749304666,
      "loss": 5.3917,
      "mean_token_accuracy": 0.20557007938623428,
      "num_tokens": 98022385.0,
      "step": 50110
    },
    {
      "entropy": 5.932157039642334,
      "epoch": 5.009496676163343,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027786218875270084,
      "loss": 5.2626,
      "mean_token_accuracy": 0.2169555053114891,
      "num_tokens": 98033401.0,
      "step": 50115
    },
    {
      "entropy": 5.7305749416351315,
      "epoch": 5.009996501224571,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027782650250292323,
      "loss": 5.0593,
      "mean_token_accuracy": 0.24080633223056794,
      "num_tokens": 98043132.0,
      "step": 50120
    },
    {
      "entropy": 5.930476903915405,
      "epoch": 5.0104963262858,
      "grad_norm": 1.25,
      "learning_rate": 0.0002777908161820315,
      "loss": 5.3055,
      "mean_token_accuracy": 0.20654358416795732,
      "num_tokens": 98053578.0,
      "step": 50125
    },
    {
      "entropy": 5.885733604431152,
      "epoch": 5.010996151347029,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002777551297909236,
      "loss": 5.2327,
      "mean_token_accuracy": 0.21637217253446578,
      "num_tokens": 98063345.0,
      "step": 50130
    },
    {
      "entropy": 5.9872386932373045,
      "epoch": 5.011495976408257,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002777194433304972,
      "loss": 5.266,
      "mean_token_accuracy": 0.22333761155605317,
      "num_tokens": 98072279.0,
      "step": 50135
    },
    {
      "entropy": 5.852004957199097,
      "epoch": 5.011995801469486,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002776837568016504,
      "loss": 5.2025,
      "mean_token_accuracy": 0.22735895067453385,
      "num_tokens": 98082146.0,
      "step": 50140
    },
    {
      "entropy": 5.804745197296143,
      "epoch": 5.012495626530714,
      "grad_norm": 1.171875,
      "learning_rate": 0.000277648070205281,
      "loss": 5.1148,
      "mean_token_accuracy": 0.2196881115436554,
      "num_tokens": 98092042.0,
      "step": 50145
    },
    {
      "entropy": 5.956239652633667,
      "epoch": 5.012995451591943,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002776123835422868,
      "loss": 5.3223,
      "mean_token_accuracy": 0.2106342598795891,
      "num_tokens": 98102918.0,
      "step": 50150
    },
    {
      "entropy": 5.899165916442871,
      "epoch": 5.013495276653171,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002775766968135657,
      "loss": 5.2096,
      "mean_token_accuracy": 0.2260236993432045,
      "num_tokens": 98112790.0,
      "step": 50155
    },
    {
      "entropy": 5.974233675003052,
      "epoch": 5.0139951017144,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002775410100200156,
      "loss": 5.2413,
      "mean_token_accuracy": 0.22118262350559234,
      "num_tokens": 98122079.0,
      "step": 50160
    },
    {
      "entropy": 5.940438604354858,
      "epoch": 5.014494926775629,
      "grad_norm": 1.21875,
      "learning_rate": 0.00027750532316253427,
      "loss": 5.2411,
      "mean_token_accuracy": 0.2156943216919899,
      "num_tokens": 98130639.0,
      "step": 50165
    },
    {
      "entropy": 5.952657175064087,
      "epoch": 5.014994751836857,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027746963624201965,
      "loss": 5.3343,
      "mean_token_accuracy": 0.20910554528236389,
      "num_tokens": 98141050.0,
      "step": 50170
    },
    {
      "entropy": 5.9664027214050295,
      "epoch": 5.015494576898086,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002774339492593697,
      "loss": 5.3571,
      "mean_token_accuracy": 0.20508363097906113,
      "num_tokens": 98150563.0,
      "step": 50175
    },
    {
      "entropy": 5.89503493309021,
      "epoch": 5.015994401959314,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002773982622154821,
      "loss": 5.2043,
      "mean_token_accuracy": 0.21860343366861343,
      "num_tokens": 98160368.0,
      "step": 50180
    },
    {
      "entropy": 5.896290063858032,
      "epoch": 5.016494227020543,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027736257511125497,
      "loss": 5.2475,
      "mean_token_accuracy": 0.2164519190788269,
      "num_tokens": 98168921.0,
      "step": 50185
    },
    {
      "entropy": 5.849658870697022,
      "epoch": 5.016994052081771,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027732688794758606,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22139434218406678,
      "num_tokens": 98178711.0,
      "step": 50190
    },
    {
      "entropy": 5.974678659439087,
      "epoch": 5.017493877143,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002772912007253732,
      "loss": 5.3172,
      "mean_token_accuracy": 0.2161855846643448,
      "num_tokens": 98188219.0,
      "step": 50195
    },
    {
      "entropy": 5.951593112945557,
      "epoch": 5.017993702204229,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002772555134455143,
      "loss": 5.2604,
      "mean_token_accuracy": 0.21202590316534042,
      "num_tokens": 98197717.0,
      "step": 50200
    },
    {
      "entropy": 5.775051212310791,
      "epoch": 5.018493527265457,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027721982610890724,
      "loss": 5.0817,
      "mean_token_accuracy": 0.24187415838241577,
      "num_tokens": 98208541.0,
      "step": 50205
    },
    {
      "entropy": 5.815793657302857,
      "epoch": 5.018993352326686,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027718413871645,
      "loss": 5.1761,
      "mean_token_accuracy": 0.23028359264135362,
      "num_tokens": 98218457.0,
      "step": 50210
    },
    {
      "entropy": 5.928699064254761,
      "epoch": 5.019493177387914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002771484512690403,
      "loss": 5.2251,
      "mean_token_accuracy": 0.22186840176582337,
      "num_tokens": 98228062.0,
      "step": 50215
    },
    {
      "entropy": 5.910341548919678,
      "epoch": 5.019993002449143,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002771127637675762,
      "loss": 5.1894,
      "mean_token_accuracy": 0.2224041759967804,
      "num_tokens": 98238383.0,
      "step": 50220
    },
    {
      "entropy": 5.931057071685791,
      "epoch": 5.020492827510371,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027707707621295547,
      "loss": 5.1991,
      "mean_token_accuracy": 0.21816978007555007,
      "num_tokens": 98247259.0,
      "step": 50225
    },
    {
      "entropy": 5.914842653274536,
      "epoch": 5.0209926525716,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002770413886060761,
      "loss": 5.2745,
      "mean_token_accuracy": 0.22302350103855134,
      "num_tokens": 98257213.0,
      "step": 50230
    },
    {
      "entropy": 5.814332675933838,
      "epoch": 5.021492477632829,
      "grad_norm": 1.25,
      "learning_rate": 0.00027700570094783583,
      "loss": 5.194,
      "mean_token_accuracy": 0.22727978527545928,
      "num_tokens": 98266928.0,
      "step": 50235
    },
    {
      "entropy": 5.8258904933929445,
      "epoch": 5.021992302694057,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002769700132391327,
      "loss": 5.1591,
      "mean_token_accuracy": 0.2286942258477211,
      "num_tokens": 98276919.0,
      "step": 50240
    },
    {
      "entropy": 5.982861661911011,
      "epoch": 5.022492127755286,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002769343254808645,
      "loss": 5.265,
      "mean_token_accuracy": 0.21874528080224992,
      "num_tokens": 98286785.0,
      "step": 50245
    },
    {
      "entropy": 5.957802486419678,
      "epoch": 5.022991952816514,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027689863767392915,
      "loss": 5.3108,
      "mean_token_accuracy": 0.21406244188547136,
      "num_tokens": 98296151.0,
      "step": 50250
    },
    {
      "entropy": 5.834206581115723,
      "epoch": 5.023491777877743,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002768629498192247,
      "loss": 5.0796,
      "mean_token_accuracy": 0.23001111596822738,
      "num_tokens": 98305781.0,
      "step": 50255
    },
    {
      "entropy": 5.970646095275879,
      "epoch": 5.023991602938971,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002768272619176488,
      "loss": 5.3493,
      "mean_token_accuracy": 0.20673584938049316,
      "num_tokens": 98315984.0,
      "step": 50260
    },
    {
      "entropy": 5.8588934421539305,
      "epoch": 5.0244914280002,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027679157397009945,
      "loss": 5.1978,
      "mean_token_accuracy": 0.22482058852910997,
      "num_tokens": 98325113.0,
      "step": 50265
    },
    {
      "entropy": 5.9988383769989015,
      "epoch": 5.024991253061429,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027675588597747455,
      "loss": 5.3169,
      "mean_token_accuracy": 0.21755292862653733,
      "num_tokens": 98336071.0,
      "step": 50270
    },
    {
      "entropy": 5.965944147109985,
      "epoch": 5.025491078122657,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027672019794067204,
      "loss": 5.2771,
      "mean_token_accuracy": 0.2121326059103012,
      "num_tokens": 98346285.0,
      "step": 50275
    },
    {
      "entropy": 5.817959213256836,
      "epoch": 5.025990903183885,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002766845098605898,
      "loss": 5.1282,
      "mean_token_accuracy": 0.23037065267562867,
      "num_tokens": 98356776.0,
      "step": 50280
    },
    {
      "entropy": 5.909163475036621,
      "epoch": 5.026490728245114,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002766488217381258,
      "loss": 5.2765,
      "mean_token_accuracy": 0.2122844785451889,
      "num_tokens": 98366991.0,
      "step": 50285
    },
    {
      "entropy": 5.9306182861328125,
      "epoch": 5.026990553306343,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002766131335741778,
      "loss": 5.2938,
      "mean_token_accuracy": 0.2165020599961281,
      "num_tokens": 98377923.0,
      "step": 50290
    },
    {
      "entropy": 5.945695066452027,
      "epoch": 5.027490378367571,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027657744536964374,
      "loss": 5.2837,
      "mean_token_accuracy": 0.21473449915647508,
      "num_tokens": 98388824.0,
      "step": 50295
    },
    {
      "entropy": 5.944133281707764,
      "epoch": 5.0279902034288,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00027654175712542154,
      "loss": 5.3544,
      "mean_token_accuracy": 0.20987141579389573,
      "num_tokens": 98398635.0,
      "step": 50300
    },
    {
      "entropy": 5.988809204101562,
      "epoch": 5.028490028490029,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002765060688424092,
      "loss": 5.3515,
      "mean_token_accuracy": 0.2044375419616699,
      "num_tokens": 98409204.0,
      "step": 50305
    },
    {
      "entropy": 5.976169681549072,
      "epoch": 5.028989853551257,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002764703805215045,
      "loss": 5.1962,
      "mean_token_accuracy": 0.22047998160123825,
      "num_tokens": 98418472.0,
      "step": 50310
    },
    {
      "entropy": 5.841271162033081,
      "epoch": 5.029489678612485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027643469216360545,
      "loss": 5.1901,
      "mean_token_accuracy": 0.23476382344961166,
      "num_tokens": 98427324.0,
      "step": 50315
    },
    {
      "entropy": 5.906825923919678,
      "epoch": 5.029989503673714,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027639900376960993,
      "loss": 5.2654,
      "mean_token_accuracy": 0.21430581361055373,
      "num_tokens": 98438559.0,
      "step": 50320
    },
    {
      "entropy": 5.871568202972412,
      "epoch": 5.030489328734943,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027636331534041574,
      "loss": 5.1498,
      "mean_token_accuracy": 0.22055660635232927,
      "num_tokens": 98448204.0,
      "step": 50325
    },
    {
      "entropy": 5.988841438293457,
      "epoch": 5.030989153796171,
      "grad_norm": 1.25,
      "learning_rate": 0.0002763276268769209,
      "loss": 5.3022,
      "mean_token_accuracy": 0.21467722356319427,
      "num_tokens": 98458237.0,
      "step": 50330
    },
    {
      "entropy": 5.852079057693482,
      "epoch": 5.0314889788574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002762919383800235,
      "loss": 5.139,
      "mean_token_accuracy": 0.22786931544542313,
      "num_tokens": 98468286.0,
      "step": 50335
    },
    {
      "entropy": 5.94019045829773,
      "epoch": 5.031988803918629,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002762562498506211,
      "loss": 5.3223,
      "mean_token_accuracy": 0.21089109480381013,
      "num_tokens": 98478808.0,
      "step": 50340
    },
    {
      "entropy": 5.793464469909668,
      "epoch": 5.032488628979857,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027622056128961183,
      "loss": 5.148,
      "mean_token_accuracy": 0.23603783249855043,
      "num_tokens": 98489390.0,
      "step": 50345
    },
    {
      "entropy": 5.965744352340698,
      "epoch": 5.032988454041085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002761848726978935,
      "loss": 5.2968,
      "mean_token_accuracy": 0.20870790630578995,
      "num_tokens": 98499415.0,
      "step": 50350
    },
    {
      "entropy": 5.817391872406006,
      "epoch": 5.033488279102314,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027614918407636407,
      "loss": 5.2108,
      "mean_token_accuracy": 0.22222494035959245,
      "num_tokens": 98511343.0,
      "step": 50355
    },
    {
      "entropy": 5.937727069854736,
      "epoch": 5.033988104163543,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00027611349542592146,
      "loss": 5.2498,
      "mean_token_accuracy": 0.2222360387444496,
      "num_tokens": 98521162.0,
      "step": 50360
    },
    {
      "entropy": 5.912037754058838,
      "epoch": 5.034487929224771,
      "grad_norm": 1.34375,
      "learning_rate": 0.00027607780674746365,
      "loss": 5.264,
      "mean_token_accuracy": 0.21715232580900193,
      "num_tokens": 98531890.0,
      "step": 50365
    },
    {
      "entropy": 5.886953401565552,
      "epoch": 5.034987754286,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002760421180418885,
      "loss": 5.2671,
      "mean_token_accuracy": 0.21719308197498322,
      "num_tokens": 98541752.0,
      "step": 50370
    },
    {
      "entropy": 5.828549385070801,
      "epoch": 5.035487579347229,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002760064293100939,
      "loss": 5.181,
      "mean_token_accuracy": 0.23132795095443726,
      "num_tokens": 98551253.0,
      "step": 50375
    },
    {
      "entropy": 5.877287817001343,
      "epoch": 5.035987404408457,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027597074055297783,
      "loss": 5.2256,
      "mean_token_accuracy": 0.2218511149287224,
      "num_tokens": 98561256.0,
      "step": 50380
    },
    {
      "entropy": 5.890732049942017,
      "epoch": 5.036487229469685,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027593505177143815,
      "loss": 5.1834,
      "mean_token_accuracy": 0.21950939148664475,
      "num_tokens": 98570627.0,
      "step": 50385
    },
    {
      "entropy": 5.873935556411743,
      "epoch": 5.036987054530914,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002758993629663727,
      "loss": 5.247,
      "mean_token_accuracy": 0.21493353843688964,
      "num_tokens": 98580462.0,
      "step": 50390
    },
    {
      "entropy": 5.849591541290283,
      "epoch": 5.037486879592143,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002758636741386796,
      "loss": 5.22,
      "mean_token_accuracy": 0.22455086559057236,
      "num_tokens": 98591538.0,
      "step": 50395
    },
    {
      "entropy": 5.997117900848389,
      "epoch": 5.037986704653371,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002758279852892568,
      "loss": 5.3074,
      "mean_token_accuracy": 0.2106959655880928,
      "num_tokens": 98600963.0,
      "step": 50400
    },
    {
      "entropy": 5.945454025268555,
      "epoch": 5.0384865297146,
      "grad_norm": 1.28125,
      "learning_rate": 0.000275792296419002,
      "loss": 5.208,
      "mean_token_accuracy": 0.2270010456442833,
      "num_tokens": 98612361.0,
      "step": 50405
    },
    {
      "entropy": 5.831157636642456,
      "epoch": 5.038986354775829,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002757566075288132,
      "loss": 5.177,
      "mean_token_accuracy": 0.22282124906778336,
      "num_tokens": 98620429.0,
      "step": 50410
    },
    {
      "entropy": 5.897075462341308,
      "epoch": 5.039486179837057,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027572091861958835,
      "loss": 5.2628,
      "mean_token_accuracy": 0.2184514507651329,
      "num_tokens": 98631586.0,
      "step": 50415
    },
    {
      "entropy": 5.914002227783203,
      "epoch": 5.039986004898285,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002756852296922254,
      "loss": 5.2182,
      "mean_token_accuracy": 0.22136931717395783,
      "num_tokens": 98641610.0,
      "step": 50420
    },
    {
      "entropy": 5.950397968292236,
      "epoch": 5.040485829959514,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002756495407476223,
      "loss": 5.3493,
      "mean_token_accuracy": 0.20565101504325867,
      "num_tokens": 98652545.0,
      "step": 50425
    },
    {
      "entropy": 5.901377964019775,
      "epoch": 5.040985655020743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002756138517866768,
      "loss": 5.255,
      "mean_token_accuracy": 0.21148046106100082,
      "num_tokens": 98661786.0,
      "step": 50430
    },
    {
      "entropy": 5.871461009979248,
      "epoch": 5.041485480081971,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027557816281028707,
      "loss": 5.1804,
      "mean_token_accuracy": 0.22233442664146424,
      "num_tokens": 98670748.0,
      "step": 50435
    },
    {
      "entropy": 5.9244646549224855,
      "epoch": 5.0419853051432,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002755424738193509,
      "loss": 5.2856,
      "mean_token_accuracy": 0.2145271271467209,
      "num_tokens": 98680624.0,
      "step": 50440
    },
    {
      "entropy": 5.853133344650269,
      "epoch": 5.042485130204429,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027550678481476614,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2153811976313591,
      "num_tokens": 98690489.0,
      "step": 50445
    },
    {
      "entropy": 5.748083209991455,
      "epoch": 5.042984955265657,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027547109579743087,
      "loss": 5.0463,
      "mean_token_accuracy": 0.23832281082868575,
      "num_tokens": 98701908.0,
      "step": 50450
    },
    {
      "entropy": 5.923515176773071,
      "epoch": 5.043484780326885,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002754354067682429,
      "loss": 5.3005,
      "mean_token_accuracy": 0.2170780345797539,
      "num_tokens": 98712299.0,
      "step": 50455
    },
    {
      "entropy": 5.927550506591797,
      "epoch": 5.043984605388114,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002753997177281002,
      "loss": 5.266,
      "mean_token_accuracy": 0.22272022664546967,
      "num_tokens": 98722668.0,
      "step": 50460
    },
    {
      "entropy": 5.948711490631103,
      "epoch": 5.044484430449343,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027536402867790077,
      "loss": 5.3623,
      "mean_token_accuracy": 0.2098515123128891,
      "num_tokens": 98732306.0,
      "step": 50465
    },
    {
      "entropy": 5.888740205764771,
      "epoch": 5.044984255510571,
      "grad_norm": 1.140625,
      "learning_rate": 0.00027532833961854247,
      "loss": 5.2187,
      "mean_token_accuracy": 0.21878920644521713,
      "num_tokens": 98741724.0,
      "step": 50470
    },
    {
      "entropy": 5.992630910873413,
      "epoch": 5.0454840805718,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027529265055092325,
      "loss": 5.404,
      "mean_token_accuracy": 0.20896031111478805,
      "num_tokens": 98752178.0,
      "step": 50475
    },
    {
      "entropy": 5.966972780227661,
      "epoch": 5.045983905633029,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027525696147594095,
      "loss": 5.2827,
      "mean_token_accuracy": 0.2099640339612961,
      "num_tokens": 98763476.0,
      "step": 50480
    },
    {
      "entropy": 5.9262604236602785,
      "epoch": 5.046483730694257,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002752212723944935,
      "loss": 5.2825,
      "mean_token_accuracy": 0.22024406045675277,
      "num_tokens": 98773019.0,
      "step": 50485
    },
    {
      "entropy": 5.935505723953247,
      "epoch": 5.046983555755485,
      "grad_norm": 1.15625,
      "learning_rate": 0.000275185583307479,
      "loss": 5.3009,
      "mean_token_accuracy": 0.2216259151697159,
      "num_tokens": 98782157.0,
      "step": 50490
    },
    {
      "entropy": 5.976366281509399,
      "epoch": 5.047483380816714,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027514989421579525,
      "loss": 5.278,
      "mean_token_accuracy": 0.20681261420249938,
      "num_tokens": 98792971.0,
      "step": 50495
    },
    {
      "entropy": 5.921484422683716,
      "epoch": 5.047983205877943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002751142051203402,
      "loss": 5.3002,
      "mean_token_accuracy": 0.21951870173215865,
      "num_tokens": 98802956.0,
      "step": 50500
    },
    {
      "entropy": 5.916595554351806,
      "epoch": 5.048483030939171,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002750785160220118,
      "loss": 5.2771,
      "mean_token_accuracy": 0.21550239324569703,
      "num_tokens": 98812366.0,
      "step": 50505
    },
    {
      "entropy": 5.864920043945313,
      "epoch": 5.0489828560004,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000275042826921708,
      "loss": 5.2203,
      "mean_token_accuracy": 0.22385040074586868,
      "num_tokens": 98822170.0,
      "step": 50510
    },
    {
      "entropy": 5.860459709167481,
      "epoch": 5.049482681061629,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027500713782032657,
      "loss": 5.212,
      "mean_token_accuracy": 0.22772597372531891,
      "num_tokens": 98833116.0,
      "step": 50515
    },
    {
      "entropy": 5.889001798629761,
      "epoch": 5.049982506122857,
      "grad_norm": 1.125,
      "learning_rate": 0.00027497144871876553,
      "loss": 5.2657,
      "mean_token_accuracy": 0.21439141929149627,
      "num_tokens": 98842795.0,
      "step": 50520
    },
    {
      "entropy": 5.896967267990112,
      "epoch": 5.050482331184085,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027493575961792306,
      "loss": 5.229,
      "mean_token_accuracy": 0.22086413204669952,
      "num_tokens": 98853549.0,
      "step": 50525
    },
    {
      "entropy": 5.930978775024414,
      "epoch": 5.050982156245314,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002749000705186967,
      "loss": 5.2884,
      "mean_token_accuracy": 0.21978836059570311,
      "num_tokens": 98862864.0,
      "step": 50530
    },
    {
      "entropy": 5.9686180591583256,
      "epoch": 5.051481981306543,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027486438142198454,
      "loss": 5.2734,
      "mean_token_accuracy": 0.22087911367416382,
      "num_tokens": 98874083.0,
      "step": 50535
    },
    {
      "entropy": 5.949223852157592,
      "epoch": 5.051981806367771,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002748286923286845,
      "loss": 5.2374,
      "mean_token_accuracy": 0.2169602856040001,
      "num_tokens": 98883433.0,
      "step": 50540
    },
    {
      "entropy": 5.935040044784546,
      "epoch": 5.052481631429,
      "grad_norm": 1.296875,
      "learning_rate": 0.00027479300323969455,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21932041347026826,
      "num_tokens": 98893347.0,
      "step": 50545
    },
    {
      "entropy": 5.929821395874024,
      "epoch": 5.052981456490229,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027475731415591253,
      "loss": 5.3244,
      "mean_token_accuracy": 0.2146676704287529,
      "num_tokens": 98902726.0,
      "step": 50550
    },
    {
      "entropy": 5.8559814453125,
      "epoch": 5.053481281551457,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00027472162507823653,
      "loss": 5.2022,
      "mean_token_accuracy": 0.2203521728515625,
      "num_tokens": 98912425.0,
      "step": 50555
    },
    {
      "entropy": 5.9104899883270265,
      "epoch": 5.053981106612685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00027468593600756425,
      "loss": 5.2001,
      "mean_token_accuracy": 0.2231924831867218,
      "num_tokens": 98923336.0,
      "step": 50560
    },
    {
      "entropy": 5.980555868148803,
      "epoch": 5.054480931673914,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027465024694479386,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21731678396463394,
      "num_tokens": 98932018.0,
      "step": 50565
    },
    {
      "entropy": 5.955374050140381,
      "epoch": 5.054980756735143,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027461455789082307,
      "loss": 5.3248,
      "mean_token_accuracy": 0.21109301447868348,
      "num_tokens": 98941750.0,
      "step": 50570
    },
    {
      "entropy": 5.939123201370239,
      "epoch": 5.055480581796371,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002745788688465499,
      "loss": 5.224,
      "mean_token_accuracy": 0.2205628052353859,
      "num_tokens": 98951012.0,
      "step": 50575
    },
    {
      "entropy": 5.841425085067749,
      "epoch": 5.0559804068576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027454317981287236,
      "loss": 5.1572,
      "mean_token_accuracy": 0.22980161905288696,
      "num_tokens": 98960118.0,
      "step": 50580
    },
    {
      "entropy": 5.875634336471558,
      "epoch": 5.056480231918829,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002745074907906882,
      "loss": 5.274,
      "mean_token_accuracy": 0.21281767189502715,
      "num_tokens": 98968916.0,
      "step": 50585
    },
    {
      "entropy": 5.888440036773682,
      "epoch": 5.056980056980057,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002744718017808955,
      "loss": 5.2242,
      "mean_token_accuracy": 0.2158466935157776,
      "num_tokens": 98978412.0,
      "step": 50590
    },
    {
      "entropy": 5.978471136093139,
      "epoch": 5.057479882041285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027443611278439226,
      "loss": 5.295,
      "mean_token_accuracy": 0.21594469100236893,
      "num_tokens": 98988250.0,
      "step": 50595
    },
    {
      "entropy": 5.8331469058990475,
      "epoch": 5.057979707102514,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002744004238020761,
      "loss": 5.1216,
      "mean_token_accuracy": 0.2322192057967186,
      "num_tokens": 98997577.0,
      "step": 50600
    },
    {
      "entropy": 5.978709745407104,
      "epoch": 5.058479532163743,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027436473483484513,
      "loss": 5.3403,
      "mean_token_accuracy": 0.20642503648996352,
      "num_tokens": 99008065.0,
      "step": 50605
    },
    {
      "entropy": 5.840170526504517,
      "epoch": 5.058979357224971,
      "grad_norm": 1.265625,
      "learning_rate": 0.00027432904588359735,
      "loss": 5.1942,
      "mean_token_accuracy": 0.2291780725121498,
      "num_tokens": 99019571.0,
      "step": 50610
    },
    {
      "entropy": 5.992439270019531,
      "epoch": 5.0594791822862,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002742933569492306,
      "loss": 5.3194,
      "mean_token_accuracy": 0.21435533612966537,
      "num_tokens": 99030255.0,
      "step": 50615
    },
    {
      "entropy": 5.903344345092774,
      "epoch": 5.059979007347429,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027425766803264274,
      "loss": 5.2405,
      "mean_token_accuracy": 0.2197907656431198,
      "num_tokens": 99039511.0,
      "step": 50620
    },
    {
      "entropy": 5.939495992660523,
      "epoch": 5.060478832408657,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002742219791347318,
      "loss": 5.341,
      "mean_token_accuracy": 0.20524048060178757,
      "num_tokens": 99049193.0,
      "step": 50625
    },
    {
      "entropy": 5.974797868728638,
      "epoch": 5.060978657469885,
      "grad_norm": 1.125,
      "learning_rate": 0.0002741862902563957,
      "loss": 5.3305,
      "mean_token_accuracy": 0.20822399556636811,
      "num_tokens": 99060285.0,
      "step": 50630
    },
    {
      "entropy": 5.982345151901245,
      "epoch": 5.061478482531114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002741506013985323,
      "loss": 5.2186,
      "mean_token_accuracy": 0.2221916362643242,
      "num_tokens": 99069324.0,
      "step": 50635
    },
    {
      "entropy": 5.824138975143432,
      "epoch": 5.061978307592343,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027411491256203944,
      "loss": 5.1716,
      "mean_token_accuracy": 0.231273290514946,
      "num_tokens": 99078877.0,
      "step": 50640
    },
    {
      "entropy": 5.88238639831543,
      "epoch": 5.062478132653571,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027407922374781524,
      "loss": 5.2853,
      "mean_token_accuracy": 0.21903891116380692,
      "num_tokens": 99088706.0,
      "step": 50645
    },
    {
      "entropy": 5.870233154296875,
      "epoch": 5.0629779577148,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00027404353495675757,
      "loss": 5.2386,
      "mean_token_accuracy": 0.22335668206214904,
      "num_tokens": 99098834.0,
      "step": 50650
    },
    {
      "entropy": 5.876333713531494,
      "epoch": 5.063477782776029,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00027400784618976425,
      "loss": 5.2302,
      "mean_token_accuracy": 0.22156603783369064,
      "num_tokens": 99108680.0,
      "step": 50655
    },
    {
      "entropy": 5.715574073791504,
      "epoch": 5.063977607837257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027397215744773333,
      "loss": 5.1025,
      "mean_token_accuracy": 0.23448552340269088,
      "num_tokens": 99119540.0,
      "step": 50660
    },
    {
      "entropy": 5.991948461532592,
      "epoch": 5.064477432898485,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002739364687315627,
      "loss": 5.3151,
      "mean_token_accuracy": 0.2123686894774437,
      "num_tokens": 99129247.0,
      "step": 50665
    },
    {
      "entropy": 5.910509014129639,
      "epoch": 5.064977257959714,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027390078004215017,
      "loss": 5.2441,
      "mean_token_accuracy": 0.22071988582611085,
      "num_tokens": 99138538.0,
      "step": 50670
    },
    {
      "entropy": 5.998839569091797,
      "epoch": 5.065477083020943,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00027386509138039374,
      "loss": 5.369,
      "mean_token_accuracy": 0.2083880662918091,
      "num_tokens": 99149057.0,
      "step": 50675
    },
    {
      "entropy": 5.99597749710083,
      "epoch": 5.065976908082171,
      "grad_norm": 1.421875,
      "learning_rate": 0.0002738294027471913,
      "loss": 5.286,
      "mean_token_accuracy": 0.2091246247291565,
      "num_tokens": 99158449.0,
      "step": 50680
    },
    {
      "entropy": 6.064051961898803,
      "epoch": 5.0664767331434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002737937141434408,
      "loss": 5.4014,
      "mean_token_accuracy": 0.20859834998846055,
      "num_tokens": 99168971.0,
      "step": 50685
    },
    {
      "entropy": 5.892239284515381,
      "epoch": 5.066976558204629,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002737580255700402,
      "loss": 5.2139,
      "mean_token_accuracy": 0.21803447008132934,
      "num_tokens": 99179214.0,
      "step": 50690
    },
    {
      "entropy": 5.817209768295288,
      "epoch": 5.067476383265857,
      "grad_norm": 1.078125,
      "learning_rate": 0.00027372233702788727,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22577409595251083,
      "num_tokens": 99189327.0,
      "step": 50695
    },
    {
      "entropy": 5.83325605392456,
      "epoch": 5.067976208327085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027368664851788006,
      "loss": 5.2444,
      "mean_token_accuracy": 0.2242395594716072,
      "num_tokens": 99199551.0,
      "step": 50700
    },
    {
      "entropy": 5.902382755279541,
      "epoch": 5.068476033388314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002736509600409164,
      "loss": 5.1538,
      "mean_token_accuracy": 0.22694237679243087,
      "num_tokens": 99209608.0,
      "step": 50705
    },
    {
      "entropy": 5.827608871459961,
      "epoch": 5.068975858449543,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027361527159789424,
      "loss": 5.1152,
      "mean_token_accuracy": 0.22271163612604142,
      "num_tokens": 99218336.0,
      "step": 50710
    },
    {
      "entropy": 5.932565975189209,
      "epoch": 5.069475683510771,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002735795831897115,
      "loss": 5.2654,
      "mean_token_accuracy": 0.2129209354519844,
      "num_tokens": 99227394.0,
      "step": 50715
    },
    {
      "entropy": 5.97167181968689,
      "epoch": 5.069975508572,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002735438948172661,
      "loss": 5.3239,
      "mean_token_accuracy": 0.20333881676197052,
      "num_tokens": 99236406.0,
      "step": 50720
    },
    {
      "entropy": 5.839057159423828,
      "epoch": 5.070475333633229,
      "grad_norm": 1.125,
      "learning_rate": 0.0002735082064814559,
      "loss": 5.2068,
      "mean_token_accuracy": 0.21973580121994019,
      "num_tokens": 99246830.0,
      "step": 50725
    },
    {
      "entropy": 5.92297773361206,
      "epoch": 5.070975158694457,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027347251818317885,
      "loss": 5.3165,
      "mean_token_accuracy": 0.2149897888302803,
      "num_tokens": 99256298.0,
      "step": 50730
    },
    {
      "entropy": 5.902908515930176,
      "epoch": 5.071474983755685,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027343682992333297,
      "loss": 5.1333,
      "mean_token_accuracy": 0.2305156633257866,
      "num_tokens": 99265094.0,
      "step": 50735
    },
    {
      "entropy": 5.961174821853637,
      "epoch": 5.071974808816914,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002734011417028159,
      "loss": 5.3016,
      "mean_token_accuracy": 0.21269350349903107,
      "num_tokens": 99276717.0,
      "step": 50740
    },
    {
      "entropy": 5.815595245361328,
      "epoch": 5.072474633878143,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002733654535225257,
      "loss": 5.1294,
      "mean_token_accuracy": 0.22332161217927932,
      "num_tokens": 99286092.0,
      "step": 50745
    },
    {
      "entropy": 5.848996496200561,
      "epoch": 5.072974458939371,
      "grad_norm": 1.484375,
      "learning_rate": 0.00027332976538336037,
      "loss": 5.2039,
      "mean_token_accuracy": 0.2162235140800476,
      "num_tokens": 99294951.0,
      "step": 50750
    },
    {
      "entropy": 5.87067928314209,
      "epoch": 5.0734742840006,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002732940772862176,
      "loss": 5.2602,
      "mean_token_accuracy": 0.22075649350881577,
      "num_tokens": 99304667.0,
      "step": 50755
    },
    {
      "entropy": 5.967836380004883,
      "epoch": 5.073974109061829,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027325838923199545,
      "loss": 5.2567,
      "mean_token_accuracy": 0.2215104654431343,
      "num_tokens": 99313999.0,
      "step": 50760
    },
    {
      "entropy": 5.903146600723266,
      "epoch": 5.074473934123057,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00027322270122159175,
      "loss": 5.2068,
      "mean_token_accuracy": 0.2210762158036232,
      "num_tokens": 99324030.0,
      "step": 50765
    },
    {
      "entropy": 5.935994815826416,
      "epoch": 5.074973759184285,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027318701325590444,
      "loss": 5.2351,
      "mean_token_accuracy": 0.21808376163244247,
      "num_tokens": 99334212.0,
      "step": 50770
    },
    {
      "entropy": 5.827916145324707,
      "epoch": 5.075473584245514,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002731513253358315,
      "loss": 5.1368,
      "mean_token_accuracy": 0.22920405566692353,
      "num_tokens": 99344560.0,
      "step": 50775
    },
    {
      "entropy": 6.045470333099365,
      "epoch": 5.075973409306743,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002731156374622706,
      "loss": 5.4579,
      "mean_token_accuracy": 0.21014512777328492,
      "num_tokens": 99354440.0,
      "step": 50780
    },
    {
      "entropy": 5.905497026443482,
      "epoch": 5.076473234367971,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00027307994963611987,
      "loss": 5.156,
      "mean_token_accuracy": 0.22476569265127183,
      "num_tokens": 99362976.0,
      "step": 50785
    },
    {
      "entropy": 5.905390882492066,
      "epoch": 5.0769730594292,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002730442618582771,
      "loss": 5.2108,
      "mean_token_accuracy": 0.22257261574268342,
      "num_tokens": 99372245.0,
      "step": 50790
    },
    {
      "entropy": 5.990296792984009,
      "epoch": 5.077472884490429,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002730085741296402,
      "loss": 5.3477,
      "mean_token_accuracy": 0.21176024824380874,
      "num_tokens": 99382575.0,
      "step": 50795
    },
    {
      "entropy": 5.923951148986816,
      "epoch": 5.077972709551657,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027297288645110703,
      "loss": 5.245,
      "mean_token_accuracy": 0.21870600134134294,
      "num_tokens": 99392995.0,
      "step": 50800
    },
    {
      "entropy": 5.911412954330444,
      "epoch": 5.078472534612885,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002729371988235756,
      "loss": 5.2821,
      "mean_token_accuracy": 0.214989972114563,
      "num_tokens": 99402523.0,
      "step": 50805
    },
    {
      "entropy": 5.921037864685059,
      "epoch": 5.078972359674114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027290151124794365,
      "loss": 5.2016,
      "mean_token_accuracy": 0.22136307507753372,
      "num_tokens": 99412083.0,
      "step": 50810
    },
    {
      "entropy": 5.993298864364624,
      "epoch": 5.079472184735343,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027286582372510913,
      "loss": 5.3282,
      "mean_token_accuracy": 0.21254232078790664,
      "num_tokens": 99422009.0,
      "step": 50815
    },
    {
      "entropy": 5.97157130241394,
      "epoch": 5.079972009796571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027283013625597,
      "loss": 5.2925,
      "mean_token_accuracy": 0.2233183741569519,
      "num_tokens": 99432401.0,
      "step": 50820
    },
    {
      "entropy": 5.786825180053711,
      "epoch": 5.0804718348578,
      "grad_norm": 1.15625,
      "learning_rate": 0.00027279444884142397,
      "loss": 5.0703,
      "mean_token_accuracy": 0.23374861627817153,
      "num_tokens": 99443127.0,
      "step": 50825
    },
    {
      "entropy": 5.933169555664063,
      "epoch": 5.080971659919029,
      "grad_norm": 1.1875,
      "learning_rate": 0.00027275876148236903,
      "loss": 5.3175,
      "mean_token_accuracy": 0.212155582010746,
      "num_tokens": 99453617.0,
      "step": 50830
    },
    {
      "entropy": 5.749474096298218,
      "epoch": 5.081471484980257,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027272307417970323,
      "loss": 5.087,
      "mean_token_accuracy": 0.23900283575057985,
      "num_tokens": 99463808.0,
      "step": 50835
    },
    {
      "entropy": 5.896447992324829,
      "epoch": 5.081971310041485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002726873869343242,
      "loss": 5.2396,
      "mean_token_accuracy": 0.21615921556949616,
      "num_tokens": 99472621.0,
      "step": 50840
    },
    {
      "entropy": 5.893190240859985,
      "epoch": 5.082471135102714,
      "grad_norm": 1.234375,
      "learning_rate": 0.00027265169974712987,
      "loss": 5.2166,
      "mean_token_accuracy": 0.22481896430253984,
      "num_tokens": 99481687.0,
      "step": 50845
    },
    {
      "entropy": 5.964850950241089,
      "epoch": 5.082970960163943,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00027261601261901824,
      "loss": 5.3572,
      "mean_token_accuracy": 0.20959886610507966,
      "num_tokens": 99492786.0,
      "step": 50850
    },
    {
      "entropy": 5.9709978103637695,
      "epoch": 5.083470785225171,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002725803255508871,
      "loss": 5.2184,
      "mean_token_accuracy": 0.22026341110467912,
      "num_tokens": 99502127.0,
      "step": 50855
    },
    {
      "entropy": 5.8711829662323,
      "epoch": 5.0839706102864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00027254463854363436,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22132952362298966,
      "num_tokens": 99512572.0,
      "step": 50860
    },
    {
      "entropy": 5.88122067451477,
      "epoch": 5.084470435347629,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002725089515981579,
      "loss": 5.1859,
      "mean_token_accuracy": 0.23089118152856827,
      "num_tokens": 99522099.0,
      "step": 50865
    },
    {
      "entropy": 5.909152030944824,
      "epoch": 5.0849702604088565,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002724732647153555,
      "loss": 5.2304,
      "mean_token_accuracy": 0.22323129773139955,
      "num_tokens": 99532822.0,
      "step": 50870
    },
    {
      "entropy": 5.923115587234497,
      "epoch": 5.085470085470085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027243757789612526,
      "loss": 5.2646,
      "mean_token_accuracy": 0.22068621963262558,
      "num_tokens": 99541689.0,
      "step": 50875
    },
    {
      "entropy": 5.911649990081787,
      "epoch": 5.085969910531314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002724018911413648,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2199433133006096,
      "num_tokens": 99553030.0,
      "step": 50880
    },
    {
      "entropy": 5.925674104690552,
      "epoch": 5.086469735592543,
      "grad_norm": 1.25,
      "learning_rate": 0.00027236620445197213,
      "loss": 5.279,
      "mean_token_accuracy": 0.2086809903383255,
      "num_tokens": 99562673.0,
      "step": 50885
    },
    {
      "entropy": 5.919861602783203,
      "epoch": 5.086969560653771,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000272330517828845,
      "loss": 5.3603,
      "mean_token_accuracy": 0.2083367094397545,
      "num_tokens": 99572738.0,
      "step": 50890
    },
    {
      "entropy": 5.910038042068481,
      "epoch": 5.087469385715,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027229483127288133,
      "loss": 5.2645,
      "mean_token_accuracy": 0.22058916687965394,
      "num_tokens": 99583064.0,
      "step": 50895
    },
    {
      "entropy": 5.921819877624512,
      "epoch": 5.087969210776229,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002722591447849791,
      "loss": 5.1895,
      "mean_token_accuracy": 0.22140755951404573,
      "num_tokens": 99591854.0,
      "step": 50900
    },
    {
      "entropy": 5.8805643081665036,
      "epoch": 5.0884690358374565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002722234583660361,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21746349334716797,
      "num_tokens": 99601343.0,
      "step": 50905
    },
    {
      "entropy": 5.901079559326172,
      "epoch": 5.088968860898685,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002721877720169501,
      "loss": 5.2577,
      "mean_token_accuracy": 0.22051242589950562,
      "num_tokens": 99611533.0,
      "step": 50910
    },
    {
      "entropy": 5.907919979095459,
      "epoch": 5.089468685959914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002721520857386191,
      "loss": 5.1834,
      "mean_token_accuracy": 0.22388456016778946,
      "num_tokens": 99620969.0,
      "step": 50915
    },
    {
      "entropy": 5.88465986251831,
      "epoch": 5.089968511021143,
      "grad_norm": 1.25,
      "learning_rate": 0.0002721163995319408,
      "loss": 5.2039,
      "mean_token_accuracy": 0.21859706044197083,
      "num_tokens": 99630525.0,
      "step": 50920
    },
    {
      "entropy": 5.870405483245849,
      "epoch": 5.090468336082371,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002720807133978133,
      "loss": 5.2628,
      "mean_token_accuracy": 0.21985906660556792,
      "num_tokens": 99641141.0,
      "step": 50925
    },
    {
      "entropy": 5.95664873123169,
      "epoch": 5.0909681611436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002720450273371341,
      "loss": 5.2814,
      "mean_token_accuracy": 0.21689508259296417,
      "num_tokens": 99650849.0,
      "step": 50930
    },
    {
      "entropy": 5.95890998840332,
      "epoch": 5.091467986204829,
      "grad_norm": 1.0625,
      "learning_rate": 0.00027200934135080133,
      "loss": 5.2783,
      "mean_token_accuracy": 0.21537819802761077,
      "num_tokens": 99661228.0,
      "step": 50935
    },
    {
      "entropy": 5.826691341400147,
      "epoch": 5.0919678112660565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002719736554397128,
      "loss": 5.1568,
      "mean_token_accuracy": 0.22610730677843094,
      "num_tokens": 99671675.0,
      "step": 50940
    },
    {
      "entropy": 5.8901125431060795,
      "epoch": 5.092467636327285,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002719379696047664,
      "loss": 5.2022,
      "mean_token_accuracy": 0.21681557446718216,
      "num_tokens": 99681326.0,
      "step": 50945
    },
    {
      "entropy": 5.88326210975647,
      "epoch": 5.092967461388514,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002719022838468598,
      "loss": 5.2865,
      "mean_token_accuracy": 0.2118035152554512,
      "num_tokens": 99691142.0,
      "step": 50950
    },
    {
      "entropy": 5.985912466049195,
      "epoch": 5.093467286449743,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002718665981668909,
      "loss": 5.3267,
      "mean_token_accuracy": 0.2113363489508629,
      "num_tokens": 99700952.0,
      "step": 50955
    },
    {
      "entropy": 5.956589841842652,
      "epoch": 5.093967111510971,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00027183091256575764,
      "loss": 5.2918,
      "mean_token_accuracy": 0.2160968527197838,
      "num_tokens": 99709830.0,
      "step": 50960
    },
    {
      "entropy": 5.951415109634399,
      "epoch": 5.0944669365722,
      "grad_norm": 1.25,
      "learning_rate": 0.0002717952270443578,
      "loss": 5.3116,
      "mean_token_accuracy": 0.21052974015474318,
      "num_tokens": 99719146.0,
      "step": 50965
    },
    {
      "entropy": 5.9529016494750975,
      "epoch": 5.094966761633429,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002717595416035892,
      "loss": 5.2913,
      "mean_token_accuracy": 0.21495803594589233,
      "num_tokens": 99729655.0,
      "step": 50970
    },
    {
      "entropy": 5.962782192230224,
      "epoch": 5.0954665866946565,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002717238562443497,
      "loss": 5.2672,
      "mean_token_accuracy": 0.2236764669418335,
      "num_tokens": 99739305.0,
      "step": 50975
    },
    {
      "entropy": 5.901079368591309,
      "epoch": 5.095966411755885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002716881709675372,
      "loss": 5.2585,
      "mean_token_accuracy": 0.22301173508167266,
      "num_tokens": 99749527.0,
      "step": 50980
    },
    {
      "entropy": 5.839287948608399,
      "epoch": 5.096466236817114,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002716524857740494,
      "loss": 5.201,
      "mean_token_accuracy": 0.22034262716770173,
      "num_tokens": 99759359.0,
      "step": 50985
    },
    {
      "entropy": 5.975713157653809,
      "epoch": 5.096966061878343,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002716168006647841,
      "loss": 5.3252,
      "mean_token_accuracy": 0.21451921612024308,
      "num_tokens": 99770174.0,
      "step": 50990
    },
    {
      "entropy": 5.956063175201416,
      "epoch": 5.097465886939571,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002715811156406394,
      "loss": 5.2647,
      "mean_token_accuracy": 0.22384758442640304,
      "num_tokens": 99780794.0,
      "step": 50995
    },
    {
      "entropy": 5.885151004791259,
      "epoch": 5.0979657120008,
      "grad_norm": 1.25,
      "learning_rate": 0.00027154543070251287,
      "loss": 5.2213,
      "mean_token_accuracy": 0.22356868535280228,
      "num_tokens": 99789674.0,
      "step": 51000
    },
    {
      "epoch": 5.0979657120008,
      "eval_entropy": 5.670754340652706,
      "eval_loss": 5.443912029266357,
      "eval_mean_token_accuracy": 0.21604956212604484,
      "eval_num_tokens": 99789674.0,
      "eval_runtime": 23.8653,
      "eval_samples_per_second": 1300.843,
      "eval_steps_per_second": 162.621,
      "step": 51000
    },
    {
      "entropy": 5.888837146759033,
      "epoch": 5.098465537062029,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002715097458513024,
      "loss": 5.2759,
      "mean_token_accuracy": 0.2158835604786873,
      "num_tokens": 99800125.0,
      "step": 51005
    },
    {
      "entropy": 5.923949956893921,
      "epoch": 5.0989653621232565,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002714740610879059,
      "loss": 5.3328,
      "mean_token_accuracy": 0.20653459578752517,
      "num_tokens": 99810044.0,
      "step": 51010
    },
    {
      "entropy": 5.918823337554931,
      "epoch": 5.099465187184485,
      "grad_norm": 1.25,
      "learning_rate": 0.0002714383764132211,
      "loss": 5.184,
      "mean_token_accuracy": 0.22480694353580474,
      "num_tokens": 99819853.0,
      "step": 51015
    },
    {
      "entropy": 5.900511693954468,
      "epoch": 5.099965012245714,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002714026918281458,
      "loss": 5.2128,
      "mean_token_accuracy": 0.21708579659461974,
      "num_tokens": 99828978.0,
      "step": 51020
    },
    {
      "entropy": 5.764459800720215,
      "epoch": 5.100464837306943,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002713670073335778,
      "loss": 5.0832,
      "mean_token_accuracy": 0.2350397825241089,
      "num_tokens": 99838780.0,
      "step": 51025
    },
    {
      "entropy": 5.9029707431793215,
      "epoch": 5.100964662368171,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002713313229304151,
      "loss": 5.2938,
      "mean_token_accuracy": 0.20866354703903198,
      "num_tokens": 99849449.0,
      "step": 51030
    },
    {
      "entropy": 5.918328428268433,
      "epoch": 5.1014644874294,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00027129563861955525,
      "loss": 5.2102,
      "mean_token_accuracy": 0.2213315948843956,
      "num_tokens": 99859807.0,
      "step": 51035
    },
    {
      "entropy": 5.866599893569946,
      "epoch": 5.101964312490629,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002712599544018962,
      "loss": 5.196,
      "mean_token_accuracy": 0.2261943906545639,
      "num_tokens": 99869397.0,
      "step": 51040
    },
    {
      "entropy": 5.839798641204834,
      "epoch": 5.1024641375518565,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027122427027833577,
      "loss": 5.1401,
      "mean_token_accuracy": 0.2287616729736328,
      "num_tokens": 99878600.0,
      "step": 51045
    },
    {
      "entropy": 5.872896146774292,
      "epoch": 5.102963962613085,
      "grad_norm": 1.28125,
      "learning_rate": 0.00027118858624977166,
      "loss": 5.2065,
      "mean_token_accuracy": 0.22218798249959945,
      "num_tokens": 99887574.0,
      "step": 51050
    },
    {
      "entropy": 5.775452899932861,
      "epoch": 5.103463787674314,
      "grad_norm": 1.09375,
      "learning_rate": 0.00027115290231710177,
      "loss": 5.0995,
      "mean_token_accuracy": 0.23125012516975402,
      "num_tokens": 99897994.0,
      "step": 51055
    },
    {
      "entropy": 5.872834825515747,
      "epoch": 5.1039636127355426,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002711172184812239,
      "loss": 5.1967,
      "mean_token_accuracy": 0.22547030448913574,
      "num_tokens": 99907649.0,
      "step": 51060
    },
    {
      "entropy": 5.854964828491211,
      "epoch": 5.104463437796771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002710815347430358,
      "loss": 5.172,
      "mean_token_accuracy": 0.22140808254480362,
      "num_tokens": 99917226.0,
      "step": 51065
    },
    {
      "entropy": 5.888394498825074,
      "epoch": 5.104963262858,
      "grad_norm": 1.25,
      "learning_rate": 0.00027104585110343526,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22218859046697617,
      "num_tokens": 99925928.0,
      "step": 51070
    },
    {
      "entropy": 5.875984144210816,
      "epoch": 5.105463087919229,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002710101675633201,
      "loss": 5.2187,
      "mean_token_accuracy": 0.2219436526298523,
      "num_tokens": 99936034.0,
      "step": 51075
    },
    {
      "entropy": 5.915892505645752,
      "epoch": 5.1059629129804565,
      "grad_norm": 1.25,
      "learning_rate": 0.0002709744841235881,
      "loss": 5.2753,
      "mean_token_accuracy": 0.2192482829093933,
      "num_tokens": 99945647.0,
      "step": 51080
    },
    {
      "entropy": 5.761041212081909,
      "epoch": 5.106462738041685,
      "grad_norm": 1.25,
      "learning_rate": 0.0002709388007851371,
      "loss": 5.0421,
      "mean_token_accuracy": 0.23645109534263611,
      "num_tokens": 99955717.0,
      "step": 51085
    },
    {
      "entropy": 5.919731330871582,
      "epoch": 5.106962563102914,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002709031175488648,
      "loss": 5.31,
      "mean_token_accuracy": 0.21019418984651567,
      "num_tokens": 99965745.0,
      "step": 51090
    },
    {
      "entropy": 5.897796249389648,
      "epoch": 5.1074623881641426,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027086743441566896,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2219824805855751,
      "num_tokens": 99975640.0,
      "step": 51095
    },
    {
      "entropy": 5.865556144714356,
      "epoch": 5.107962213225371,
      "grad_norm": 1.3125,
      "learning_rate": 0.00027083175138644744,
      "loss": 5.2179,
      "mean_token_accuracy": 0.22035635709762574,
      "num_tokens": 99984128.0,
      "step": 51100
    },
    {
      "entropy": 5.989470720291138,
      "epoch": 5.1084620382866,
      "grad_norm": 1.234375,
      "learning_rate": 0.000270796068462098,
      "loss": 5.3709,
      "mean_token_accuracy": 0.20475796461105347,
      "num_tokens": 99994247.0,
      "step": 51105
    },
    {
      "entropy": 5.939775657653809,
      "epoch": 5.108961863347829,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00027076038564351836,
      "loss": 5.2513,
      "mean_token_accuracy": 0.21734825521707535,
      "num_tokens": 100003922.0,
      "step": 51110
    },
    {
      "entropy": 5.939203262329102,
      "epoch": 5.1094616884090565,
      "grad_norm": 1.25,
      "learning_rate": 0.0002707247029316062,
      "loss": 5.2702,
      "mean_token_accuracy": 0.21237855702638625,
      "num_tokens": 100013327.0,
      "step": 51115
    },
    {
      "entropy": 5.922021293640137,
      "epoch": 5.109961513470285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00027068902032725963,
      "loss": 5.2807,
      "mean_token_accuracy": 0.21353448629379274,
      "num_tokens": 100023880.0,
      "step": 51120
    },
    {
      "entropy": 5.8058568954467775,
      "epoch": 5.110461338531514,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002706533378313761,
      "loss": 5.0678,
      "mean_token_accuracy": 0.23922114670276642,
      "num_tokens": 100033432.0,
      "step": 51125
    },
    {
      "entropy": 5.864303064346314,
      "epoch": 5.1109611635927426,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027061765544485346,
      "loss": 5.2545,
      "mean_token_accuracy": 0.21782848685979844,
      "num_tokens": 100043197.0,
      "step": 51130
    },
    {
      "entropy": 5.934717273712158,
      "epoch": 5.111460988653971,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002705819731685895,
      "loss": 5.2945,
      "mean_token_accuracy": 0.21932778656482696,
      "num_tokens": 100053625.0,
      "step": 51135
    },
    {
      "entropy": 5.946099138259887,
      "epoch": 5.1119608137152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000270546291003482,
      "loss": 5.2658,
      "mean_token_accuracy": 0.22646425664424896,
      "num_tokens": 100063347.0,
      "step": 51140
    },
    {
      "entropy": 5.880619382858276,
      "epoch": 5.112460638776429,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002705106089504286,
      "loss": 5.3255,
      "mean_token_accuracy": 0.2156536504626274,
      "num_tokens": 100073916.0,
      "step": 51145
    },
    {
      "entropy": 5.818142080307007,
      "epoch": 5.1129604638376565,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027047492701032716,
      "loss": 5.1602,
      "mean_token_accuracy": 0.23293067812919616,
      "num_tokens": 100083812.0,
      "step": 51150
    },
    {
      "entropy": 5.975958251953125,
      "epoch": 5.113460288898885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002704392451840755,
      "loss": 5.3484,
      "mean_token_accuracy": 0.21099046617746353,
      "num_tokens": 100094235.0,
      "step": 51155
    },
    {
      "entropy": 5.906800317764282,
      "epoch": 5.113960113960114,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002704035634725711,
      "loss": 5.2921,
      "mean_token_accuracy": 0.21156727373600007,
      "num_tokens": 100104014.0,
      "step": 51160
    },
    {
      "entropy": 5.876590347290039,
      "epoch": 5.1144599390213425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00027036788187671194,
      "loss": 5.2419,
      "mean_token_accuracy": 0.22433536648750305,
      "num_tokens": 100113989.0,
      "step": 51165
    },
    {
      "entropy": 5.9260694026947025,
      "epoch": 5.114959764082571,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00027033220039739564,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22836766839027406,
      "num_tokens": 100123965.0,
      "step": 51170
    },
    {
      "entropy": 5.948879289627075,
      "epoch": 5.1154595891438,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002702965190355199,
      "loss": 5.2842,
      "mean_token_accuracy": 0.21964612156152724,
      "num_tokens": 100132739.0,
      "step": 51175
    },
    {
      "entropy": 5.879740381240845,
      "epoch": 5.115959414205029,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002702608377919827,
      "loss": 5.256,
      "mean_token_accuracy": 0.21684223264455796,
      "num_tokens": 100141494.0,
      "step": 51180
    },
    {
      "entropy": 5.887273120880127,
      "epoch": 5.1164592392662565,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0002702251566676815,
      "loss": 5.2669,
      "mean_token_accuracy": 0.21083150058984756,
      "num_tokens": 100150815.0,
      "step": 51185
    },
    {
      "entropy": 5.872071313858032,
      "epoch": 5.116959064327485,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027018947566351416,
      "loss": 5.221,
      "mean_token_accuracy": 0.2242719054222107,
      "num_tokens": 100159514.0,
      "step": 51190
    },
    {
      "entropy": 5.919856834411621,
      "epoch": 5.117458889388714,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002701537947803783,
      "loss": 5.1625,
      "mean_token_accuracy": 0.22237952649593354,
      "num_tokens": 100169221.0,
      "step": 51195
    },
    {
      "entropy": 5.886425399780274,
      "epoch": 5.1179587144499425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00027011811401917184,
      "loss": 5.2127,
      "mean_token_accuracy": 0.2179187536239624,
      "num_tokens": 100179220.0,
      "step": 51200
    },
    {
      "entropy": 5.974803638458252,
      "epoch": 5.118458539511171,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002700824333807923,
      "loss": 5.2707,
      "mean_token_accuracy": 0.21573852151632308,
      "num_tokens": 100188364.0,
      "step": 51205
    },
    {
      "entropy": 5.865871572494507,
      "epoch": 5.1189583645724,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002700467528661375,
      "loss": 5.1657,
      "mean_token_accuracy": 0.23022190630435943,
      "num_tokens": 100197574.0,
      "step": 51210
    },
    {
      "entropy": 5.854446601867676,
      "epoch": 5.119458189633629,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00027001107247610505,
      "loss": 5.2236,
      "mean_token_accuracy": 0.21949059069156646,
      "num_tokens": 100206991.0,
      "step": 51215
    },
    {
      "entropy": 5.92347559928894,
      "epoch": 5.1199580146948565,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002699753922115928,
      "loss": 5.2888,
      "mean_token_accuracy": 0.21204583197832108,
      "num_tokens": 100216186.0,
      "step": 51220
    },
    {
      "entropy": 5.9637120246887205,
      "epoch": 5.120457839756085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002699397120734985,
      "loss": 5.32,
      "mean_token_accuracy": 0.21489985436201095,
      "num_tokens": 100226992.0,
      "step": 51225
    },
    {
      "entropy": 5.981601715087891,
      "epoch": 5.120957664817314,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002699040320627196,
      "loss": 5.3173,
      "mean_token_accuracy": 0.2173980787396431,
      "num_tokens": 100238575.0,
      "step": 51230
    },
    {
      "entropy": 5.954367065429688,
      "epoch": 5.1214574898785425,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026986835218015395,
      "loss": 5.2816,
      "mean_token_accuracy": 0.21267151236534118,
      "num_tokens": 100248609.0,
      "step": 51235
    },
    {
      "entropy": 5.9304862976074215,
      "epoch": 5.121957314939771,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026983267242669924,
      "loss": 5.2478,
      "mean_token_accuracy": 0.2168465808033943,
      "num_tokens": 100257373.0,
      "step": 51240
    },
    {
      "entropy": 5.910837888717651,
      "epoch": 5.122457140001,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026979699280325317,
      "loss": 5.3385,
      "mean_token_accuracy": 0.21085745394229888,
      "num_tokens": 100267834.0,
      "step": 51245
    },
    {
      "entropy": 5.953196477890015,
      "epoch": 5.122956965062229,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026976131331071345,
      "loss": 5.2739,
      "mean_token_accuracy": 0.21718096286058425,
      "num_tokens": 100277102.0,
      "step": 51250
    },
    {
      "entropy": 5.9224255084991455,
      "epoch": 5.1234567901234565,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026972563394997776,
      "loss": 5.2035,
      "mean_token_accuracy": 0.21421603858470917,
      "num_tokens": 100286868.0,
      "step": 51255
    },
    {
      "entropy": 5.837898445129395,
      "epoch": 5.123956615184685,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002696899547219437,
      "loss": 5.1269,
      "mean_token_accuracy": 0.23420356214046478,
      "num_tokens": 100297098.0,
      "step": 51260
    },
    {
      "entropy": 5.893913984298706,
      "epoch": 5.124456440245914,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000269654275627509,
      "loss": 5.2725,
      "mean_token_accuracy": 0.2128689169883728,
      "num_tokens": 100306536.0,
      "step": 51265
    },
    {
      "entropy": 5.937500858306885,
      "epoch": 5.1249562653071425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026961859666757137,
      "loss": 5.2586,
      "mean_token_accuracy": 0.21970753967761994,
      "num_tokens": 100316679.0,
      "step": 51270
    },
    {
      "entropy": 5.970708799362183,
      "epoch": 5.125456090368371,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026958291784302845,
      "loss": 5.3461,
      "mean_token_accuracy": 0.21495922952890395,
      "num_tokens": 100326319.0,
      "step": 51275
    },
    {
      "entropy": 5.889137649536133,
      "epoch": 5.1259559154296,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000269547239154778,
      "loss": 5.1886,
      "mean_token_accuracy": 0.2208217576146126,
      "num_tokens": 100336056.0,
      "step": 51280
    },
    {
      "entropy": 5.878980493545532,
      "epoch": 5.126455740490829,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026951156060371753,
      "loss": 5.21,
      "mean_token_accuracy": 0.21928517371416092,
      "num_tokens": 100346076.0,
      "step": 51285
    },
    {
      "entropy": 5.892460775375366,
      "epoch": 5.1269555655520564,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002694758821907448,
      "loss": 5.2205,
      "mean_token_accuracy": 0.2181606724858284,
      "num_tokens": 100354822.0,
      "step": 51290
    },
    {
      "entropy": 6.0019735336303714,
      "epoch": 5.127455390613285,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026944020391675747,
      "loss": 5.3365,
      "mean_token_accuracy": 0.21210345923900603,
      "num_tokens": 100363685.0,
      "step": 51295
    },
    {
      "entropy": 5.968027496337891,
      "epoch": 5.127955215674514,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026940452578265325,
      "loss": 5.2452,
      "mean_token_accuracy": 0.2167486682534218,
      "num_tokens": 100372814.0,
      "step": 51300
    },
    {
      "entropy": 5.900681447982788,
      "epoch": 5.1284550407357425,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002693688477893296,
      "loss": 5.3085,
      "mean_token_accuracy": 0.21613068878650665,
      "num_tokens": 100382696.0,
      "step": 51305
    },
    {
      "entropy": 5.933055400848389,
      "epoch": 5.128954865796971,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026933316993768433,
      "loss": 5.2733,
      "mean_token_accuracy": 0.2194740042090416,
      "num_tokens": 100391688.0,
      "step": 51310
    },
    {
      "entropy": 5.920795822143555,
      "epoch": 5.1294546908582,
      "grad_norm": 1.25,
      "learning_rate": 0.00026929749222861507,
      "loss": 5.2546,
      "mean_token_accuracy": 0.2154442250728607,
      "num_tokens": 100401930.0,
      "step": 51315
    },
    {
      "entropy": 5.898633909225464,
      "epoch": 5.129954515919428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002692618146630194,
      "loss": 5.1835,
      "mean_token_accuracy": 0.22600822448730468,
      "num_tokens": 100411442.0,
      "step": 51320
    },
    {
      "entropy": 6.014916563034058,
      "epoch": 5.1304543409806564,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026922613724179504,
      "loss": 5.3167,
      "mean_token_accuracy": 0.2109946072101593,
      "num_tokens": 100422114.0,
      "step": 51325
    },
    {
      "entropy": 5.889159631729126,
      "epoch": 5.130954166041885,
      "grad_norm": 1.25,
      "learning_rate": 0.00026919045996583955,
      "loss": 5.192,
      "mean_token_accuracy": 0.22469041794538497,
      "num_tokens": 100431446.0,
      "step": 51330
    },
    {
      "entropy": 5.941508769989014,
      "epoch": 5.131453991103114,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026915478283605056,
      "loss": 5.3542,
      "mean_token_accuracy": 0.20677613765001296,
      "num_tokens": 100441516.0,
      "step": 51335
    },
    {
      "entropy": 5.8269792079925535,
      "epoch": 5.1319538161643425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002691191058533258,
      "loss": 5.0563,
      "mean_token_accuracy": 0.23768753409385682,
      "num_tokens": 100451054.0,
      "step": 51340
    },
    {
      "entropy": 5.91642198562622,
      "epoch": 5.132453641225571,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026908342901856274,
      "loss": 5.2411,
      "mean_token_accuracy": 0.22282002568244935,
      "num_tokens": 100461219.0,
      "step": 51345
    },
    {
      "entropy": 5.936973953247071,
      "epoch": 5.1329534662868,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026904775233265913,
      "loss": 5.2693,
      "mean_token_accuracy": 0.21611248254776,
      "num_tokens": 100470906.0,
      "step": 51350
    },
    {
      "entropy": 5.864470720291138,
      "epoch": 5.133453291348028,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026901207579651257,
      "loss": 5.1558,
      "mean_token_accuracy": 0.22533252239227294,
      "num_tokens": 100479098.0,
      "step": 51355
    },
    {
      "entropy": 5.922656440734864,
      "epoch": 5.1339531164092564,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002689763994110207,
      "loss": 5.2565,
      "mean_token_accuracy": 0.21761879771947862,
      "num_tokens": 100488762.0,
      "step": 51360
    },
    {
      "entropy": 5.819287061691284,
      "epoch": 5.134452941470485,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026894072317708095,
      "loss": 5.1458,
      "mean_token_accuracy": 0.23360592424869536,
      "num_tokens": 100498979.0,
      "step": 51365
    },
    {
      "entropy": 5.861579132080078,
      "epoch": 5.134952766531714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002689050470955911,
      "loss": 5.2164,
      "mean_token_accuracy": 0.21920612305402756,
      "num_tokens": 100508697.0,
      "step": 51370
    },
    {
      "entropy": 5.8271231174469,
      "epoch": 5.1354525915929425,
      "grad_norm": 1.125,
      "learning_rate": 0.0002688693711674487,
      "loss": 5.1822,
      "mean_token_accuracy": 0.23052888363599777,
      "num_tokens": 100518054.0,
      "step": 51375
    },
    {
      "entropy": 5.928694343566894,
      "epoch": 5.135952416654171,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026883369539355134,
      "loss": 5.2224,
      "mean_token_accuracy": 0.21814118921756745,
      "num_tokens": 100526612.0,
      "step": 51380
    },
    {
      "entropy": 5.83723635673523,
      "epoch": 5.1364522417154,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026879801977479654,
      "loss": 5.1508,
      "mean_token_accuracy": 0.22593817263841628,
      "num_tokens": 100536956.0,
      "step": 51385
    },
    {
      "entropy": 5.883675909042358,
      "epoch": 5.136952066776628,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002687623443120821,
      "loss": 5.2258,
      "mean_token_accuracy": 0.2236466959118843,
      "num_tokens": 100547721.0,
      "step": 51390
    },
    {
      "entropy": 5.854878950119018,
      "epoch": 5.137451891837856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002687266690063053,
      "loss": 5.2636,
      "mean_token_accuracy": 0.21909440010786058,
      "num_tokens": 100556909.0,
      "step": 51395
    },
    {
      "entropy": 5.95061674118042,
      "epoch": 5.137951716899085,
      "grad_norm": 1.140625,
      "learning_rate": 0.000268690993858364,
      "loss": 5.3246,
      "mean_token_accuracy": 0.20930406153202058,
      "num_tokens": 100566565.0,
      "step": 51400
    },
    {
      "entropy": 5.960211372375488,
      "epoch": 5.138451541960314,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026865531886915566,
      "loss": 5.2858,
      "mean_token_accuracy": 0.21464523524045945,
      "num_tokens": 100577497.0,
      "step": 51405
    },
    {
      "entropy": 5.972915506362915,
      "epoch": 5.1389513670215425,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002686196440395779,
      "loss": 5.274,
      "mean_token_accuracy": 0.2185841515660286,
      "num_tokens": 100586431.0,
      "step": 51410
    },
    {
      "entropy": 5.875321292877198,
      "epoch": 5.139451192082771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002685839693705282,
      "loss": 5.2157,
      "mean_token_accuracy": 0.21964974850416183,
      "num_tokens": 100595289.0,
      "step": 51415
    },
    {
      "entropy": 5.922457313537597,
      "epoch": 5.139951017144,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026854829486290424,
      "loss": 5.3177,
      "mean_token_accuracy": 0.21770597100257874,
      "num_tokens": 100605742.0,
      "step": 51420
    },
    {
      "entropy": 5.963001108169555,
      "epoch": 5.140450842205228,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002685126205176035,
      "loss": 5.214,
      "mean_token_accuracy": 0.22986705601215363,
      "num_tokens": 100615120.0,
      "step": 51425
    },
    {
      "entropy": 5.93152084350586,
      "epoch": 5.140950667266456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002684769463355236,
      "loss": 5.2511,
      "mean_token_accuracy": 0.21523120552301406,
      "num_tokens": 100624670.0,
      "step": 51430
    },
    {
      "entropy": 5.861659812927246,
      "epoch": 5.141450492327685,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026844127231756196,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22734990566968918,
      "num_tokens": 100633422.0,
      "step": 51435
    },
    {
      "entropy": 5.890003252029419,
      "epoch": 5.141950317388914,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002684055984646163,
      "loss": 5.233,
      "mean_token_accuracy": 0.2138666555285454,
      "num_tokens": 100642364.0,
      "step": 51440
    },
    {
      "entropy": 5.8511762619018555,
      "epoch": 5.1424501424501425,
      "grad_norm": 1.25,
      "learning_rate": 0.000268369924777584,
      "loss": 5.1846,
      "mean_token_accuracy": 0.2188478097319603,
      "num_tokens": 100651323.0,
      "step": 51445
    },
    {
      "entropy": 5.92910327911377,
      "epoch": 5.142949967511371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002683342512573628,
      "loss": 5.3004,
      "mean_token_accuracy": 0.21409388929605483,
      "num_tokens": 100661423.0,
      "step": 51450
    },
    {
      "entropy": 5.866994714736938,
      "epoch": 5.1434497925726,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00026829857790485,
      "loss": 5.1431,
      "mean_token_accuracy": 0.23162361085414887,
      "num_tokens": 100671209.0,
      "step": 51455
    },
    {
      "entropy": 5.820308017730713,
      "epoch": 5.143949617633828,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026826290472094325,
      "loss": 5.172,
      "mean_token_accuracy": 0.22535018771886825,
      "num_tokens": 100679999.0,
      "step": 51460
    },
    {
      "entropy": 5.8792119979858395,
      "epoch": 5.144449442695056,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026822723170654014,
      "loss": 5.2722,
      "mean_token_accuracy": 0.22020171582698822,
      "num_tokens": 100689899.0,
      "step": 51465
    },
    {
      "entropy": 5.899321174621582,
      "epoch": 5.144949267756285,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026819155886253815,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21910445392131805,
      "num_tokens": 100698825.0,
      "step": 51470
    },
    {
      "entropy": 5.880952453613281,
      "epoch": 5.145449092817514,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002681558861898348,
      "loss": 5.141,
      "mean_token_accuracy": 0.2256525456905365,
      "num_tokens": 100707689.0,
      "step": 51475
    },
    {
      "entropy": 5.871819972991943,
      "epoch": 5.1459489178787425,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002681202136893275,
      "loss": 5.2579,
      "mean_token_accuracy": 0.22345397025346755,
      "num_tokens": 100718716.0,
      "step": 51480
    },
    {
      "entropy": 5.898675346374512,
      "epoch": 5.146448742939971,
      "grad_norm": 1.25,
      "learning_rate": 0.00026808454136191393,
      "loss": 5.2636,
      "mean_token_accuracy": 0.2120214119553566,
      "num_tokens": 100728053.0,
      "step": 51485
    },
    {
      "entropy": 5.96498441696167,
      "epoch": 5.1469485680012,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026804886920849154,
      "loss": 5.2741,
      "mean_token_accuracy": 0.2119307518005371,
      "num_tokens": 100738225.0,
      "step": 51490
    },
    {
      "entropy": 5.904953193664551,
      "epoch": 5.147448393062428,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026801319722995784,
      "loss": 5.2211,
      "mean_token_accuracy": 0.223033045232296,
      "num_tokens": 100748889.0,
      "step": 51495
    },
    {
      "entropy": 5.953137302398682,
      "epoch": 5.147948218123656,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026797752542721023,
      "loss": 5.3082,
      "mean_token_accuracy": 0.21330418437719345,
      "num_tokens": 100758907.0,
      "step": 51500
    },
    {
      "entropy": 5.975183296203613,
      "epoch": 5.148448043184885,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026794185380114627,
      "loss": 5.3332,
      "mean_token_accuracy": 0.21115291565656663,
      "num_tokens": 100768803.0,
      "step": 51505
    },
    {
      "entropy": 5.958587169647217,
      "epoch": 5.148947868246114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002679061823526635,
      "loss": 5.2871,
      "mean_token_accuracy": 0.21811345219612122,
      "num_tokens": 100777776.0,
      "step": 51510
    },
    {
      "entropy": 6.030852365493774,
      "epoch": 5.1494476933073425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002678705110826593,
      "loss": 5.3927,
      "mean_token_accuracy": 0.20382563322782515,
      "num_tokens": 100787686.0,
      "step": 51515
    },
    {
      "entropy": 5.941105079650879,
      "epoch": 5.149947518368571,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002678348399920312,
      "loss": 5.2636,
      "mean_token_accuracy": 0.220365509390831,
      "num_tokens": 100797053.0,
      "step": 51520
    },
    {
      "entropy": 5.9347803592681885,
      "epoch": 5.1504473434298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002677991690816767,
      "loss": 5.2348,
      "mean_token_accuracy": 0.22202992290258408,
      "num_tokens": 100806717.0,
      "step": 51525
    },
    {
      "entropy": 5.97438416481018,
      "epoch": 5.150947168491028,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002677634983524933,
      "loss": 5.3492,
      "mean_token_accuracy": 0.2157475858926773,
      "num_tokens": 100816551.0,
      "step": 51530
    },
    {
      "entropy": 5.985807943344116,
      "epoch": 5.151446993552256,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026772782780537833,
      "loss": 5.3343,
      "mean_token_accuracy": 0.21145545542240143,
      "num_tokens": 100826610.0,
      "step": 51535
    },
    {
      "entropy": 5.924520063400268,
      "epoch": 5.151946818613485,
      "grad_norm": 1.125,
      "learning_rate": 0.0002676921574412294,
      "loss": 5.2593,
      "mean_token_accuracy": 0.21958347409963608,
      "num_tokens": 100836835.0,
      "step": 51540
    },
    {
      "entropy": 5.858581829071045,
      "epoch": 5.152446643674714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026765648726094396,
      "loss": 5.1647,
      "mean_token_accuracy": 0.2277456745505333,
      "num_tokens": 100846873.0,
      "step": 51545
    },
    {
      "entropy": 5.858871412277222,
      "epoch": 5.1529464687359425,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026762081726541934,
      "loss": 5.2199,
      "mean_token_accuracy": 0.21194122284650802,
      "num_tokens": 100857106.0,
      "step": 51550
    },
    {
      "entropy": 5.982828664779663,
      "epoch": 5.153446293797171,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026758514745555306,
      "loss": 5.3383,
      "mean_token_accuracy": 0.2144048884510994,
      "num_tokens": 100867875.0,
      "step": 51555
    },
    {
      "entropy": 5.921804809570313,
      "epoch": 5.1539461188584,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026754947783224253,
      "loss": 5.2261,
      "mean_token_accuracy": 0.22687255442142487,
      "num_tokens": 100877598.0,
      "step": 51560
    },
    {
      "entropy": 5.992115640640259,
      "epoch": 5.154445943919628,
      "grad_norm": 1.25,
      "learning_rate": 0.0002675138083963853,
      "loss": 5.2843,
      "mean_token_accuracy": 0.21716414391994476,
      "num_tokens": 100887033.0,
      "step": 51565
    },
    {
      "entropy": 5.928901433944702,
      "epoch": 5.154945768980856,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002674781391488788,
      "loss": 5.2994,
      "mean_token_accuracy": 0.2192514255642891,
      "num_tokens": 100897260.0,
      "step": 51570
    },
    {
      "entropy": 5.917486095428467,
      "epoch": 5.155445594042085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002674424700906203,
      "loss": 5.3162,
      "mean_token_accuracy": 0.21062458157539368,
      "num_tokens": 100907542.0,
      "step": 51575
    },
    {
      "entropy": 5.877924203872681,
      "epoch": 5.155945419103314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002674068012225072,
      "loss": 5.2041,
      "mean_token_accuracy": 0.2306123897433281,
      "num_tokens": 100917345.0,
      "step": 51580
    },
    {
      "entropy": 5.833837938308716,
      "epoch": 5.1564452441645425,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002673711325454371,
      "loss": 5.1675,
      "mean_token_accuracy": 0.22915214896202088,
      "num_tokens": 100927337.0,
      "step": 51585
    },
    {
      "entropy": 5.833637952804565,
      "epoch": 5.156945069225771,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002673354640603073,
      "loss": 5.156,
      "mean_token_accuracy": 0.2273454800248146,
      "num_tokens": 100937811.0,
      "step": 51590
    },
    {
      "entropy": 5.937223768234253,
      "epoch": 5.157444894287,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026729979576801537,
      "loss": 5.2411,
      "mean_token_accuracy": 0.2213631346821785,
      "num_tokens": 100948663.0,
      "step": 51595
    },
    {
      "entropy": 5.8662937641143795,
      "epoch": 5.157944719348228,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002672641276694585,
      "loss": 5.2062,
      "mean_token_accuracy": 0.21516793966293335,
      "num_tokens": 100958316.0,
      "step": 51600
    },
    {
      "entropy": 5.955812215805054,
      "epoch": 5.158444544409456,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026722845976553423,
      "loss": 5.3189,
      "mean_token_accuracy": 0.21167022436857225,
      "num_tokens": 100966897.0,
      "step": 51605
    },
    {
      "entropy": 5.96432523727417,
      "epoch": 5.158944369470685,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002671927920571399,
      "loss": 5.2637,
      "mean_token_accuracy": 0.2177809566259384,
      "num_tokens": 100976444.0,
      "step": 51610
    },
    {
      "entropy": 5.913388013839722,
      "epoch": 5.159444194531914,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002671571245451729,
      "loss": 5.2148,
      "mean_token_accuracy": 0.22405993193387985,
      "num_tokens": 100988016.0,
      "step": 51615
    },
    {
      "entropy": 5.9036284446716305,
      "epoch": 5.1599440195931425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026712145723053066,
      "loss": 5.2933,
      "mean_token_accuracy": 0.21116309463977814,
      "num_tokens": 100997336.0,
      "step": 51620
    },
    {
      "entropy": 5.8486686706542965,
      "epoch": 5.160443844654371,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002670857901141105,
      "loss": 5.1731,
      "mean_token_accuracy": 0.21961697190999985,
      "num_tokens": 101006662.0,
      "step": 51625
    },
    {
      "entropy": 5.886180448532104,
      "epoch": 5.1609436697156,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002670501231968098,
      "loss": 5.1928,
      "mean_token_accuracy": 0.22207177430391312,
      "num_tokens": 101016396.0,
      "step": 51630
    },
    {
      "entropy": 5.922270345687866,
      "epoch": 5.161443494776828,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000267014456479526,
      "loss": 5.2929,
      "mean_token_accuracy": 0.2179170146584511,
      "num_tokens": 101026434.0,
      "step": 51635
    },
    {
      "entropy": 5.937516880035401,
      "epoch": 5.161943319838056,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026697878996315654,
      "loss": 5.2794,
      "mean_token_accuracy": 0.212788724899292,
      "num_tokens": 101035672.0,
      "step": 51640
    },
    {
      "entropy": 5.89917402267456,
      "epoch": 5.162443144899285,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002669431236485985,
      "loss": 5.1863,
      "mean_token_accuracy": 0.22597110122442246,
      "num_tokens": 101044994.0,
      "step": 51645
    },
    {
      "entropy": 5.880944061279297,
      "epoch": 5.162942969960514,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002669074575367494,
      "loss": 5.2193,
      "mean_token_accuracy": 0.2185230150818825,
      "num_tokens": 101054520.0,
      "step": 51650
    },
    {
      "entropy": 5.911609840393067,
      "epoch": 5.1634427950217425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026687179162850663,
      "loss": 5.2058,
      "mean_token_accuracy": 0.2209336817264557,
      "num_tokens": 101063517.0,
      "step": 51655
    },
    {
      "entropy": 5.701434469223022,
      "epoch": 5.163942620082971,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026683612592476747,
      "loss": 5.0583,
      "mean_token_accuracy": 0.24950083047151567,
      "num_tokens": 101073124.0,
      "step": 51660
    },
    {
      "entropy": 5.907196283340454,
      "epoch": 5.1644424451442,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026680046042642933,
      "loss": 5.3056,
      "mean_token_accuracy": 0.2140115901827812,
      "num_tokens": 101082924.0,
      "step": 51665
    },
    {
      "entropy": 5.966027450561524,
      "epoch": 5.164942270205428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002667647951343894,
      "loss": 5.3359,
      "mean_token_accuracy": 0.21359696835279465,
      "num_tokens": 101092060.0,
      "step": 51670
    },
    {
      "entropy": 5.935229539871216,
      "epoch": 5.165442095266656,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026672913004954513,
      "loss": 5.3319,
      "mean_token_accuracy": 0.2149966448545456,
      "num_tokens": 101103052.0,
      "step": 51675
    },
    {
      "entropy": 5.935626554489136,
      "epoch": 5.165941920327885,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002666934651727938,
      "loss": 5.2885,
      "mean_token_accuracy": 0.21664652228355408,
      "num_tokens": 101112647.0,
      "step": 51680
    },
    {
      "entropy": 5.9210748195648195,
      "epoch": 5.166441745389114,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002666578005050328,
      "loss": 5.2199,
      "mean_token_accuracy": 0.2182817205786705,
      "num_tokens": 101121401.0,
      "step": 51685
    },
    {
      "entropy": 5.9468292713165285,
      "epoch": 5.1669415704503425,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026662213604715943,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21589524298906326,
      "num_tokens": 101131813.0,
      "step": 51690
    },
    {
      "entropy": 5.872857713699341,
      "epoch": 5.167441395511571,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002665864718000709,
      "loss": 5.2632,
      "mean_token_accuracy": 0.21690128445625306,
      "num_tokens": 101141494.0,
      "step": 51695
    },
    {
      "entropy": 5.930013036727905,
      "epoch": 5.1679412205728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002665508077646646,
      "loss": 5.2517,
      "mean_token_accuracy": 0.215396948158741,
      "num_tokens": 101151095.0,
      "step": 51700
    },
    {
      "entropy": 6.000193023681641,
      "epoch": 5.168441045634028,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026651514394183777,
      "loss": 5.3299,
      "mean_token_accuracy": 0.2111525356769562,
      "num_tokens": 101160190.0,
      "step": 51705
    },
    {
      "entropy": 6.015290260314941,
      "epoch": 5.168940870695256,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026647948033248784,
      "loss": 5.3527,
      "mean_token_accuracy": 0.21098435372114183,
      "num_tokens": 101169643.0,
      "step": 51710
    },
    {
      "entropy": 5.924065160751343,
      "epoch": 5.169440695756485,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002664438169375119,
      "loss": 5.3258,
      "mean_token_accuracy": 0.2061459392309189,
      "num_tokens": 101180218.0,
      "step": 51715
    },
    {
      "entropy": 5.866309547424317,
      "epoch": 5.169940520817714,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026640815375780736,
      "loss": 5.1756,
      "mean_token_accuracy": 0.22677373886108398,
      "num_tokens": 101189690.0,
      "step": 51720
    },
    {
      "entropy": 5.9078504085540775,
      "epoch": 5.1704403458789425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002663724907942714,
      "loss": 5.2767,
      "mean_token_accuracy": 0.21518756151199342,
      "num_tokens": 101199942.0,
      "step": 51725
    },
    {
      "entropy": 5.851655912399292,
      "epoch": 5.170940170940171,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002663368280478014,
      "loss": 5.1646,
      "mean_token_accuracy": 0.22970513999462128,
      "num_tokens": 101208164.0,
      "step": 51730
    },
    {
      "entropy": 5.822091102600098,
      "epoch": 5.1714399960014,
      "grad_norm": 1.25,
      "learning_rate": 0.00026630116551929454,
      "loss": 5.1595,
      "mean_token_accuracy": 0.22549229711294175,
      "num_tokens": 101217781.0,
      "step": 51735
    },
    {
      "entropy": 5.887960004806518,
      "epoch": 5.171939821062628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026626550320964817,
      "loss": 5.2847,
      "mean_token_accuracy": 0.21813465803861617,
      "num_tokens": 101226642.0,
      "step": 51740
    },
    {
      "entropy": 5.819229650497436,
      "epoch": 5.172439646123856,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002662298411197594,
      "loss": 5.125,
      "mean_token_accuracy": 0.22895860970020293,
      "num_tokens": 101236858.0,
      "step": 51745
    },
    {
      "entropy": 5.890370559692383,
      "epoch": 5.172939471185085,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00026619417925052564,
      "loss": 5.1674,
      "mean_token_accuracy": 0.22762753069400787,
      "num_tokens": 101246684.0,
      "step": 51750
    },
    {
      "entropy": 5.875774908065796,
      "epoch": 5.173439296246314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002661585176028441,
      "loss": 5.1938,
      "mean_token_accuracy": 0.2254611536860466,
      "num_tokens": 101255696.0,
      "step": 51755
    },
    {
      "entropy": 5.886921834945679,
      "epoch": 5.1739391213075425,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002661228561776119,
      "loss": 5.2786,
      "mean_token_accuracy": 0.22145086973905564,
      "num_tokens": 101265056.0,
      "step": 51760
    },
    {
      "entropy": 5.843166494369507,
      "epoch": 5.174438946368771,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002660871949757264,
      "loss": 5.1583,
      "mean_token_accuracy": 0.23333198130130767,
      "num_tokens": 101274598.0,
      "step": 51765
    },
    {
      "entropy": 5.927523374557495,
      "epoch": 5.17493877143,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002660515339980848,
      "loss": 5.2521,
      "mean_token_accuracy": 0.21269466131925582,
      "num_tokens": 101283964.0,
      "step": 51770
    },
    {
      "entropy": 5.916775226593018,
      "epoch": 5.175438596491228,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002660158732455843,
      "loss": 5.2068,
      "mean_token_accuracy": 0.22359678894281387,
      "num_tokens": 101293092.0,
      "step": 51775
    },
    {
      "entropy": 5.875547790527344,
      "epoch": 5.175938421552456,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026598021271912207,
      "loss": 5.2986,
      "mean_token_accuracy": 0.20826773345470428,
      "num_tokens": 101304252.0,
      "step": 51780
    },
    {
      "entropy": 5.908199119567871,
      "epoch": 5.176438246613685,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002659445524195955,
      "loss": 5.3151,
      "mean_token_accuracy": 0.20798137933015823,
      "num_tokens": 101314023.0,
      "step": 51785
    },
    {
      "entropy": 5.945268011093139,
      "epoch": 5.176938071674914,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002659088923479015,
      "loss": 5.2378,
      "mean_token_accuracy": 0.21698783487081527,
      "num_tokens": 101322836.0,
      "step": 51790
    },
    {
      "entropy": 5.891138744354248,
      "epoch": 5.1774378967361425,
      "grad_norm": 1.109375,
      "learning_rate": 0.0002658732325049375,
      "loss": 5.0874,
      "mean_token_accuracy": 0.23374895602464676,
      "num_tokens": 101333089.0,
      "step": 51795
    },
    {
      "entropy": 5.924261856079101,
      "epoch": 5.177937721797371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002658375728916006,
      "loss": 5.2577,
      "mean_token_accuracy": 0.21619167029857636,
      "num_tokens": 101343118.0,
      "step": 51800
    },
    {
      "entropy": 5.912112140655518,
      "epoch": 5.1784375468586,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026580191350878795,
      "loss": 5.2034,
      "mean_token_accuracy": 0.22499053925275803,
      "num_tokens": 101353582.0,
      "step": 51805
    },
    {
      "entropy": 5.924660301208496,
      "epoch": 5.178937371919828,
      "grad_norm": 1.328125,
      "learning_rate": 0.00026576625435739687,
      "loss": 5.3065,
      "mean_token_accuracy": 0.21563172936439515,
      "num_tokens": 101363688.0,
      "step": 51810
    },
    {
      "entropy": 5.9564680576324465,
      "epoch": 5.179437196981056,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00026573059543832445,
      "loss": 5.3489,
      "mean_token_accuracy": 0.21359171271324157,
      "num_tokens": 101372887.0,
      "step": 51815
    },
    {
      "entropy": 5.9570554256439205,
      "epoch": 5.179937022042285,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002656949367524679,
      "loss": 5.2319,
      "mean_token_accuracy": 0.2199731707572937,
      "num_tokens": 101382657.0,
      "step": 51820
    },
    {
      "entropy": 5.843328809738159,
      "epoch": 5.180436847103514,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002656592783007242,
      "loss": 5.1462,
      "mean_token_accuracy": 0.2288498044013977,
      "num_tokens": 101392110.0,
      "step": 51825
    },
    {
      "entropy": 5.9602789878845215,
      "epoch": 5.1809366721647425,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026562362008399076,
      "loss": 5.3773,
      "mean_token_accuracy": 0.21270921677350998,
      "num_tokens": 101401061.0,
      "step": 51830
    },
    {
      "entropy": 5.893081712722778,
      "epoch": 5.181436497225971,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002655879621031647,
      "loss": 5.2649,
      "mean_token_accuracy": 0.21916741579771043,
      "num_tokens": 101410612.0,
      "step": 51835
    },
    {
      "entropy": 5.892965698242188,
      "epoch": 5.1819363222872,
      "grad_norm": 1.109375,
      "learning_rate": 0.000265552304359143,
      "loss": 5.2595,
      "mean_token_accuracy": 0.22142406553030014,
      "num_tokens": 101420805.0,
      "step": 51840
    },
    {
      "entropy": 5.989022874832154,
      "epoch": 5.182436147348428,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002655166468528228,
      "loss": 5.3005,
      "mean_token_accuracy": 0.21664773672819138,
      "num_tokens": 101430383.0,
      "step": 51845
    },
    {
      "entropy": 5.878669595718383,
      "epoch": 5.182935972409656,
      "grad_norm": 1.09375,
      "learning_rate": 0.00026548098958510154,
      "loss": 5.1827,
      "mean_token_accuracy": 0.22435664683580397,
      "num_tokens": 101439561.0,
      "step": 51850
    },
    {
      "entropy": 5.904480934143066,
      "epoch": 5.183435797470885,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026544533255687603,
      "loss": 5.2441,
      "mean_token_accuracy": 0.222216334939003,
      "num_tokens": 101448817.0,
      "step": 51855
    },
    {
      "entropy": 6.000495719909668,
      "epoch": 5.183935622532114,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026540967576904346,
      "loss": 5.3564,
      "mean_token_accuracy": 0.2060113713145256,
      "num_tokens": 101459827.0,
      "step": 51860
    },
    {
      "entropy": 6.000712966918945,
      "epoch": 5.1844354475933425,
      "grad_norm": 1.15625,
      "learning_rate": 0.000265374019222501,
      "loss": 5.3932,
      "mean_token_accuracy": 0.2087208941578865,
      "num_tokens": 101470300.0,
      "step": 51865
    },
    {
      "entropy": 5.950753259658813,
      "epoch": 5.184935272654571,
      "grad_norm": 1.0625,
      "learning_rate": 0.00026533836291814573,
      "loss": 5.2994,
      "mean_token_accuracy": 0.2199219360947609,
      "num_tokens": 101481050.0,
      "step": 51870
    },
    {
      "entropy": 5.907164287567139,
      "epoch": 5.1854350977158,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026530270685687477,
      "loss": 5.2098,
      "mean_token_accuracy": 0.22650499194860457,
      "num_tokens": 101490980.0,
      "step": 51875
    },
    {
      "entropy": 5.933209419250488,
      "epoch": 5.185934922777028,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026526705103958523,
      "loss": 5.3314,
      "mean_token_accuracy": 0.2044832080602646,
      "num_tokens": 101501512.0,
      "step": 51880
    },
    {
      "entropy": 5.909036016464233,
      "epoch": 5.186434747838256,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026523139546717414,
      "loss": 5.1815,
      "mean_token_accuracy": 0.21659165173768996,
      "num_tokens": 101511118.0,
      "step": 51885
    },
    {
      "entropy": 5.964182710647583,
      "epoch": 5.186934572899485,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026519574014053867,
      "loss": 5.2558,
      "mean_token_accuracy": 0.2185792937874794,
      "num_tokens": 101520722.0,
      "step": 51890
    },
    {
      "entropy": 5.889034652709961,
      "epoch": 5.187434397960714,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026516008506057584,
      "loss": 5.2585,
      "mean_token_accuracy": 0.22125582098960878,
      "num_tokens": 101530286.0,
      "step": 51895
    },
    {
      "entropy": 5.910959482192993,
      "epoch": 5.1879342230219425,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002651244302281827,
      "loss": 5.2662,
      "mean_token_accuracy": 0.2217221662402153,
      "num_tokens": 101540582.0,
      "step": 51900
    },
    {
      "entropy": 5.90667757987976,
      "epoch": 5.188434048083171,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026508877564425636,
      "loss": 5.1912,
      "mean_token_accuracy": 0.22398703843355178,
      "num_tokens": 101549886.0,
      "step": 51905
    },
    {
      "entropy": 5.875924062728882,
      "epoch": 5.1889338731444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002650531213096939,
      "loss": 5.2465,
      "mean_token_accuracy": 0.2108827918767929,
      "num_tokens": 101560473.0,
      "step": 51910
    },
    {
      "entropy": 5.845443201065064,
      "epoch": 5.189433698205628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026501746722539237,
      "loss": 5.1619,
      "mean_token_accuracy": 0.22405891567468644,
      "num_tokens": 101570459.0,
      "step": 51915
    },
    {
      "entropy": 6.008775758743286,
      "epoch": 5.189933523266856,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026498181339224886,
      "loss": 5.4489,
      "mean_token_accuracy": 0.20437659323215485,
      "num_tokens": 101581358.0,
      "step": 51920
    },
    {
      "entropy": 5.959244346618652,
      "epoch": 5.190433348328085,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002649461598111602,
      "loss": 5.2786,
      "mean_token_accuracy": 0.2172714188694954,
      "num_tokens": 101590877.0,
      "step": 51925
    },
    {
      "entropy": 5.765812253952026,
      "epoch": 5.190933173389314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002649105064830237,
      "loss": 5.1337,
      "mean_token_accuracy": 0.23057982623577117,
      "num_tokens": 101600741.0,
      "step": 51930
    },
    {
      "entropy": 5.921507215499878,
      "epoch": 5.1914329984505425,
      "grad_norm": 1.078125,
      "learning_rate": 0.0002648748534087361,
      "loss": 5.2602,
      "mean_token_accuracy": 0.221511210501194,
      "num_tokens": 101611236.0,
      "step": 51935
    },
    {
      "entropy": 5.78811674118042,
      "epoch": 5.191932823511771,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002648392005891947,
      "loss": 5.0808,
      "mean_token_accuracy": 0.23735015243291854,
      "num_tokens": 101620812.0,
      "step": 51940
    },
    {
      "entropy": 5.928217840194702,
      "epoch": 5.192432648572999,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00026480354802529645,
      "loss": 5.2242,
      "mean_token_accuracy": 0.2131195619702339,
      "num_tokens": 101631059.0,
      "step": 51945
    },
    {
      "entropy": 5.875123596191406,
      "epoch": 5.192932473634228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002647678957179382,
      "loss": 5.2656,
      "mean_token_accuracy": 0.223517107963562,
      "num_tokens": 101640334.0,
      "step": 51950
    },
    {
      "entropy": 5.620136356353759,
      "epoch": 5.193432298695456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026473224366801713,
      "loss": 4.9759,
      "mean_token_accuracy": 0.2515754118561745,
      "num_tokens": 101649710.0,
      "step": 51955
    },
    {
      "entropy": 6.026166105270386,
      "epoch": 5.193932123756685,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002646965918764301,
      "loss": 5.4594,
      "mean_token_accuracy": 0.20216923654079438,
      "num_tokens": 101659226.0,
      "step": 51960
    },
    {
      "entropy": 5.886401224136352,
      "epoch": 5.194431948817914,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026466094034407415,
      "loss": 5.2347,
      "mean_token_accuracy": 0.2202809527516365,
      "num_tokens": 101668929.0,
      "step": 51965
    },
    {
      "entropy": 5.945434665679931,
      "epoch": 5.1949317738791425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026462528907184637,
      "loss": 5.2444,
      "mean_token_accuracy": 0.21846967339515685,
      "num_tokens": 101679837.0,
      "step": 51970
    },
    {
      "entropy": 5.922263336181641,
      "epoch": 5.195431598940371,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002645896380606436,
      "loss": 5.2844,
      "mean_token_accuracy": 0.2264489486813545,
      "num_tokens": 101689021.0,
      "step": 51975
    },
    {
      "entropy": 5.895827007293701,
      "epoch": 5.195931424001599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002645539873113629,
      "loss": 5.2567,
      "mean_token_accuracy": 0.21943375766277312,
      "num_tokens": 101698408.0,
      "step": 51980
    },
    {
      "entropy": 5.853606033325195,
      "epoch": 5.196431249062828,
      "grad_norm": 1.59375,
      "learning_rate": 0.00026451833682490113,
      "loss": 5.1818,
      "mean_token_accuracy": 0.22028999030590057,
      "num_tokens": 101708751.0,
      "step": 51985
    },
    {
      "entropy": 5.878904390335083,
      "epoch": 5.196931074124056,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002644826866021554,
      "loss": 5.2366,
      "mean_token_accuracy": 0.22026500552892686,
      "num_tokens": 101718934.0,
      "step": 51990
    },
    {
      "entropy": 5.884392547607422,
      "epoch": 5.197430899185285,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026444703664402246,
      "loss": 5.2109,
      "mean_token_accuracy": 0.21645719856023787,
      "num_tokens": 101729128.0,
      "step": 51995
    },
    {
      "entropy": 5.828579235076904,
      "epoch": 5.197930724246514,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026441138695139946,
      "loss": 5.1791,
      "mean_token_accuracy": 0.22702284008264542,
      "num_tokens": 101738516.0,
      "step": 52000
    },
    {
      "entropy": 5.8604669094085695,
      "epoch": 5.1984305493077425,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002643757375251832,
      "loss": 5.1712,
      "mean_token_accuracy": 0.22727174013853074,
      "num_tokens": 101747879.0,
      "step": 52005
    },
    {
      "entropy": 5.9288829326629635,
      "epoch": 5.198930374368971,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002643400883662706,
      "loss": 5.2519,
      "mean_token_accuracy": 0.2176240414381027,
      "num_tokens": 101756946.0,
      "step": 52010
    },
    {
      "entropy": 5.877235841751099,
      "epoch": 5.199430199430199,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002643044394755587,
      "loss": 5.2752,
      "mean_token_accuracy": 0.2176487147808075,
      "num_tokens": 101766805.0,
      "step": 52015
    },
    {
      "entropy": 5.854606246948242,
      "epoch": 5.199930024491428,
      "grad_norm": 1.125,
      "learning_rate": 0.0002642687908539443,
      "loss": 5.1873,
      "mean_token_accuracy": 0.22185317128896714,
      "num_tokens": 101776442.0,
      "step": 52020
    },
    {
      "entropy": 5.904419040679931,
      "epoch": 5.200429849552656,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002642331425023244,
      "loss": 5.1996,
      "mean_token_accuracy": 0.21582892537117004,
      "num_tokens": 101785992.0,
      "step": 52025
    },
    {
      "entropy": 5.872819948196411,
      "epoch": 5.200929674613885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002641974944215958,
      "loss": 5.2438,
      "mean_token_accuracy": 0.2237142950296402,
      "num_tokens": 101796011.0,
      "step": 52030
    },
    {
      "entropy": 5.843243598937988,
      "epoch": 5.201429499675114,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026416184661265553,
      "loss": 5.1832,
      "mean_token_accuracy": 0.22681754678487778,
      "num_tokens": 101806260.0,
      "step": 52035
    },
    {
      "entropy": 5.930645418167114,
      "epoch": 5.2019293247363425,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002641261990764004,
      "loss": 5.2546,
      "mean_token_accuracy": 0.22118206620216369,
      "num_tokens": 101815847.0,
      "step": 52040
    },
    {
      "entropy": 5.994344711303711,
      "epoch": 5.202429149797571,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002640905518137273,
      "loss": 5.3722,
      "mean_token_accuracy": 0.20803817063570024,
      "num_tokens": 101825677.0,
      "step": 52045
    },
    {
      "entropy": 5.938155555725098,
      "epoch": 5.202928974858799,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002640549048255331,
      "loss": 5.2882,
      "mean_token_accuracy": 0.21327054351568223,
      "num_tokens": 101836051.0,
      "step": 52050
    },
    {
      "entropy": 5.849883985519409,
      "epoch": 5.203428799920028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002640192581127147,
      "loss": 5.1692,
      "mean_token_accuracy": 0.22566402703523636,
      "num_tokens": 101846173.0,
      "step": 52055
    },
    {
      "entropy": 5.860017871856689,
      "epoch": 5.203928624981256,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002639836116761689,
      "loss": 5.1842,
      "mean_token_accuracy": 0.2308667555451393,
      "num_tokens": 101856339.0,
      "step": 52060
    },
    {
      "entropy": 5.848341417312622,
      "epoch": 5.204428450042485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002639479655167926,
      "loss": 5.191,
      "mean_token_accuracy": 0.219666288793087,
      "num_tokens": 101865905.0,
      "step": 52065
    },
    {
      "entropy": 5.919669675827026,
      "epoch": 5.204928275103714,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002639123196354827,
      "loss": 5.2224,
      "mean_token_accuracy": 0.2237250551581383,
      "num_tokens": 101876143.0,
      "step": 52070
    },
    {
      "entropy": 5.966641664505005,
      "epoch": 5.2054281001649425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002638766740331359,
      "loss": 5.3181,
      "mean_token_accuracy": 0.2087393209338188,
      "num_tokens": 101885844.0,
      "step": 52075
    },
    {
      "entropy": 5.942576503753662,
      "epoch": 5.205927925226171,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002638410287106492,
      "loss": 5.228,
      "mean_token_accuracy": 0.21719643026590346,
      "num_tokens": 101896442.0,
      "step": 52080
    },
    {
      "entropy": 5.9659104347229,
      "epoch": 5.206427750287399,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002638053836689193,
      "loss": 5.3472,
      "mean_token_accuracy": 0.21682388037443162,
      "num_tokens": 101906422.0,
      "step": 52085
    },
    {
      "entropy": 5.836633968353271,
      "epoch": 5.206927575348628,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026376973890884304,
      "loss": 5.1851,
      "mean_token_accuracy": 0.23179447948932647,
      "num_tokens": 101915843.0,
      "step": 52090
    },
    {
      "entropy": 5.879825687408447,
      "epoch": 5.207427400409856,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002637340944313173,
      "loss": 5.172,
      "mean_token_accuracy": 0.21562418788671495,
      "num_tokens": 101924393.0,
      "step": 52095
    },
    {
      "entropy": 5.856560182571411,
      "epoch": 5.207927225471085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002636984502372388,
      "loss": 5.1776,
      "mean_token_accuracy": 0.23073481768369675,
      "num_tokens": 101934557.0,
      "step": 52100
    },
    {
      "entropy": 5.932419061660767,
      "epoch": 5.208427050532314,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00026366280632750445,
      "loss": 5.2517,
      "mean_token_accuracy": 0.212410831451416,
      "num_tokens": 101943652.0,
      "step": 52105
    },
    {
      "entropy": 5.929090213775635,
      "epoch": 5.2089268755935425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026362716270301095,
      "loss": 5.2325,
      "mean_token_accuracy": 0.22257447838783265,
      "num_tokens": 101953071.0,
      "step": 52110
    },
    {
      "entropy": 5.935085201263428,
      "epoch": 5.209426700654771,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026359151936465516,
      "loss": 5.2967,
      "mean_token_accuracy": 0.21450239270925522,
      "num_tokens": 101962054.0,
      "step": 52115
    },
    {
      "entropy": 5.8790911674499515,
      "epoch": 5.209926525715999,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002635558763133337,
      "loss": 5.2046,
      "mean_token_accuracy": 0.21888316720724105,
      "num_tokens": 101970650.0,
      "step": 52120
    },
    {
      "entropy": 5.929623651504516,
      "epoch": 5.210426350777228,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026352023354994354,
      "loss": 5.2926,
      "mean_token_accuracy": 0.21173711866140366,
      "num_tokens": 101980879.0,
      "step": 52125
    },
    {
      "entropy": 5.805775594711304,
      "epoch": 5.210926175838456,
      "grad_norm": 1.25,
      "learning_rate": 0.0002634845910753814,
      "loss": 5.1073,
      "mean_token_accuracy": 0.23347501456737518,
      "num_tokens": 101990358.0,
      "step": 52130
    },
    {
      "entropy": 5.963003635406494,
      "epoch": 5.211426000899685,
      "grad_norm": 1.140625,
      "learning_rate": 0.00026344894889054383,
      "loss": 5.332,
      "mean_token_accuracy": 0.2134471833705902,
      "num_tokens": 102001687.0,
      "step": 52135
    },
    {
      "entropy": 5.806221437454224,
      "epoch": 5.211925825960914,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002634133069963278,
      "loss": 5.0847,
      "mean_token_accuracy": 0.24240682125091553,
      "num_tokens": 102011083.0,
      "step": 52140
    },
    {
      "entropy": 5.831836414337158,
      "epoch": 5.2124256510221425,
      "grad_norm": 1.03125,
      "learning_rate": 0.00026337766539363003,
      "loss": 5.1407,
      "mean_token_accuracy": 0.22832648009061812,
      "num_tokens": 102022617.0,
      "step": 52145
    },
    {
      "entropy": 5.899537849426269,
      "epoch": 5.212925476083371,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026334202408334715,
      "loss": 5.2825,
      "mean_token_accuracy": 0.22427311092615126,
      "num_tokens": 102031863.0,
      "step": 52150
    },
    {
      "entropy": 5.807322454452515,
      "epoch": 5.213425301144599,
      "grad_norm": 1.25,
      "learning_rate": 0.00026330638306637586,
      "loss": 5.1378,
      "mean_token_accuracy": 0.23750994056463243,
      "num_tokens": 102042850.0,
      "step": 52155
    },
    {
      "entropy": 5.978599262237549,
      "epoch": 5.213925126205828,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002632707423436131,
      "loss": 5.3206,
      "mean_token_accuracy": 0.20922909080982208,
      "num_tokens": 102053499.0,
      "step": 52160
    },
    {
      "entropy": 5.814971303939819,
      "epoch": 5.214424951267056,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002632351019159554,
      "loss": 5.1629,
      "mean_token_accuracy": 0.23144769221544265,
      "num_tokens": 102064575.0,
      "step": 52165
    },
    {
      "entropy": 5.901663970947266,
      "epoch": 5.214924776328285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00026319946178429953,
      "loss": 5.2272,
      "mean_token_accuracy": 0.2263159081339836,
      "num_tokens": 102074207.0,
      "step": 52170
    },
    {
      "entropy": 5.914719009399414,
      "epoch": 5.215424601389514,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026316382194954214,
      "loss": 5.2917,
      "mean_token_accuracy": 0.21887362450361253,
      "num_tokens": 102083159.0,
      "step": 52175
    },
    {
      "entropy": 5.967415761947632,
      "epoch": 5.2159244264507425,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026312818241257995,
      "loss": 5.2734,
      "mean_token_accuracy": 0.21735763549804688,
      "num_tokens": 102091952.0,
      "step": 52180
    },
    {
      "entropy": 5.887585067749024,
      "epoch": 5.216424251511971,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026309254317430966,
      "loss": 5.2434,
      "mean_token_accuracy": 0.2213097855448723,
      "num_tokens": 102101684.0,
      "step": 52185
    },
    {
      "entropy": 5.70667724609375,
      "epoch": 5.216924076573199,
      "grad_norm": 1.296875,
      "learning_rate": 0.00026305690423562785,
      "loss": 4.9401,
      "mean_token_accuracy": 0.24976910799741744,
      "num_tokens": 102110823.0,
      "step": 52190
    },
    {
      "entropy": 5.866295671463012,
      "epoch": 5.217423901634428,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026302126559743126,
      "loss": 5.1858,
      "mean_token_accuracy": 0.22679820507764817,
      "num_tokens": 102119222.0,
      "step": 52195
    },
    {
      "entropy": 5.951625728607178,
      "epoch": 5.217923726695656,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026298562726061674,
      "loss": 5.3145,
      "mean_token_accuracy": 0.2087318032979965,
      "num_tokens": 102129529.0,
      "step": 52200
    },
    {
      "entropy": 5.734704971313477,
      "epoch": 5.218423551756885,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002629499892260806,
      "loss": 5.0994,
      "mean_token_accuracy": 0.24037254899740218,
      "num_tokens": 102140487.0,
      "step": 52205
    },
    {
      "entropy": 5.969981479644775,
      "epoch": 5.218923376818114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002629143514947196,
      "loss": 5.3516,
      "mean_token_accuracy": 0.20868657976388932,
      "num_tokens": 102150268.0,
      "step": 52210
    },
    {
      "entropy": 5.980891513824463,
      "epoch": 5.2194232018793425,
      "grad_norm": 1.25,
      "learning_rate": 0.00026287871406743043,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21473568379879,
      "num_tokens": 102160117.0,
      "step": 52215
    },
    {
      "entropy": 5.908208894729614,
      "epoch": 5.219923026940571,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002628430769451097,
      "loss": 5.2922,
      "mean_token_accuracy": 0.21680417358875276,
      "num_tokens": 102170045.0,
      "step": 52220
    },
    {
      "entropy": 5.8955738067626955,
      "epoch": 5.220422852001799,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000262807440128654,
      "loss": 5.2195,
      "mean_token_accuracy": 0.22178870141506196,
      "num_tokens": 102179159.0,
      "step": 52225
    },
    {
      "entropy": 5.880153846740723,
      "epoch": 5.220922677063028,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026277180361896,
      "loss": 5.2231,
      "mean_token_accuracy": 0.2283179998397827,
      "num_tokens": 102188350.0,
      "step": 52230
    },
    {
      "entropy": 5.938544368743896,
      "epoch": 5.221422502124256,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002627361674169243,
      "loss": 5.2429,
      "mean_token_accuracy": 0.21637650430202485,
      "num_tokens": 102197907.0,
      "step": 52235
    },
    {
      "entropy": 5.921374797821045,
      "epoch": 5.221922327185485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002627005315234434,
      "loss": 5.3412,
      "mean_token_accuracy": 0.20727015882730485,
      "num_tokens": 102206806.0,
      "step": 52240
    },
    {
      "entropy": 5.899582099914551,
      "epoch": 5.222422152246714,
      "grad_norm": 1.140625,
      "learning_rate": 0.000262664895939414,
      "loss": 5.231,
      "mean_token_accuracy": 0.21925055533647536,
      "num_tokens": 102217425.0,
      "step": 52245
    },
    {
      "entropy": 5.917146873474121,
      "epoch": 5.2229219773079425,
      "grad_norm": 1.25,
      "learning_rate": 0.0002626292606657326,
      "loss": 5.2361,
      "mean_token_accuracy": 0.22197837084531785,
      "num_tokens": 102226145.0,
      "step": 52250
    },
    {
      "entropy": 5.944002771377564,
      "epoch": 5.223421802369171,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026259362570329585,
      "loss": 5.2436,
      "mean_token_accuracy": 0.218004447221756,
      "num_tokens": 102235653.0,
      "step": 52255
    },
    {
      "entropy": 5.907772541046143,
      "epoch": 5.223921627430399,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002625579910530003,
      "loss": 5.2744,
      "mean_token_accuracy": 0.21901050060987473,
      "num_tokens": 102246192.0,
      "step": 52260
    },
    {
      "entropy": 5.842292785644531,
      "epoch": 5.224421452491628,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026252235671574256,
      "loss": 5.245,
      "mean_token_accuracy": 0.22315199971199035,
      "num_tokens": 102255261.0,
      "step": 52265
    },
    {
      "entropy": 5.85981297492981,
      "epoch": 5.224921277552856,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00026248672269241907,
      "loss": 5.1518,
      "mean_token_accuracy": 0.2211461842060089,
      "num_tokens": 102264735.0,
      "step": 52270
    },
    {
      "entropy": 5.884776926040649,
      "epoch": 5.225421102614085,
      "grad_norm": 1.234375,
      "learning_rate": 0.00026245108898392637,
      "loss": 5.2624,
      "mean_token_accuracy": 0.22617256492376328,
      "num_tokens": 102274389.0,
      "step": 52275
    },
    {
      "entropy": 5.994459581375122,
      "epoch": 5.225920927675314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002624154555911611,
      "loss": 5.3599,
      "mean_token_accuracy": 0.20626871138811112,
      "num_tokens": 102284928.0,
      "step": 52280
    },
    {
      "entropy": 5.938519620895386,
      "epoch": 5.2264207527365425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026237982251501967,
      "loss": 5.2442,
      "mean_token_accuracy": 0.22398667186498641,
      "num_tokens": 102295657.0,
      "step": 52285
    },
    {
      "entropy": 5.844380521774292,
      "epoch": 5.226920577797771,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026234418975639866,
      "loss": 5.1444,
      "mean_token_accuracy": 0.22821570932865143,
      "num_tokens": 102305213.0,
      "step": 52290
    },
    {
      "entropy": 5.8499773979187015,
      "epoch": 5.227420402858999,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026230855731619466,
      "loss": 5.1211,
      "mean_token_accuracy": 0.22418375462293624,
      "num_tokens": 102314533.0,
      "step": 52295
    },
    {
      "entropy": 5.8746734142303465,
      "epoch": 5.227920227920228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00026227292519530397,
      "loss": 5.2393,
      "mean_token_accuracy": 0.21803298443555832,
      "num_tokens": 102324256.0,
      "step": 52300
    },
    {
      "entropy": 5.8111591815948485,
      "epoch": 5.228420052981456,
      "grad_norm": 1.25,
      "learning_rate": 0.00026223729339462334,
      "loss": 5.1298,
      "mean_token_accuracy": 0.237807996571064,
      "num_tokens": 102333538.0,
      "step": 52305
    },
    {
      "entropy": 5.943480110168457,
      "epoch": 5.228919878042685,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026220166191504897,
      "loss": 5.2223,
      "mean_token_accuracy": 0.22106225788593292,
      "num_tokens": 102342800.0,
      "step": 52310
    },
    {
      "entropy": 5.936939334869384,
      "epoch": 5.229419703103914,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002621660307574776,
      "loss": 5.2767,
      "mean_token_accuracy": 0.2244391605257988,
      "num_tokens": 102352343.0,
      "step": 52315
    },
    {
      "entropy": 5.8904218673706055,
      "epoch": 5.2299195281651425,
      "grad_norm": 1.1875,
      "learning_rate": 0.00026213039992280555,
      "loss": 5.2107,
      "mean_token_accuracy": 0.22595517635345458,
      "num_tokens": 102361267.0,
      "step": 52320
    },
    {
      "entropy": 5.817598962783814,
      "epoch": 5.230419353226371,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002620947694119294,
      "loss": 5.2461,
      "mean_token_accuracy": 0.2327801212668419,
      "num_tokens": 102370957.0,
      "step": 52325
    },
    {
      "entropy": 5.871937847137451,
      "epoch": 5.230919178287599,
      "grad_norm": 1.046875,
      "learning_rate": 0.00026205913922574544,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22076735496520997,
      "num_tokens": 102381022.0,
      "step": 52330
    },
    {
      "entropy": 5.933165216445923,
      "epoch": 5.231419003348828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002620235093651502,
      "loss": 5.2491,
      "mean_token_accuracy": 0.21679594963788987,
      "num_tokens": 102389809.0,
      "step": 52335
    },
    {
      "entropy": 5.866895627975464,
      "epoch": 5.231918828410056,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00026198787983104026,
      "loss": 5.1458,
      "mean_token_accuracy": 0.22928087860345842,
      "num_tokens": 102399685.0,
      "step": 52340
    },
    {
      "entropy": 5.997967100143432,
      "epoch": 5.232418653471285,
      "grad_norm": 1.25,
      "learning_rate": 0.0002619522506243118,
      "loss": 5.3572,
      "mean_token_accuracy": 0.21055438220500947,
      "num_tokens": 102409306.0,
      "step": 52345
    },
    {
      "entropy": 6.00295057296753,
      "epoch": 5.232918478532514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00026191662174586145,
      "loss": 5.3028,
      "mean_token_accuracy": 0.20871490687131883,
      "num_tokens": 102418664.0,
      "step": 52350
    },
    {
      "entropy": 5.7920379638671875,
      "epoch": 5.2334183035937425,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026188099319658545,
      "loss": 5.1405,
      "mean_token_accuracy": 0.23789812475442887,
      "num_tokens": 102428881.0,
      "step": 52355
    },
    {
      "entropy": 5.95226526260376,
      "epoch": 5.23391812865497,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002618453649773803,
      "loss": 5.3075,
      "mean_token_accuracy": 0.21091231107711791,
      "num_tokens": 102440176.0,
      "step": 52360
    },
    {
      "entropy": 5.93756365776062,
      "epoch": 5.234417953716199,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002618097370891423,
      "loss": 5.3069,
      "mean_token_accuracy": 0.21450473368167877,
      "num_tokens": 102450284.0,
      "step": 52365
    },
    {
      "entropy": 5.895134973526001,
      "epoch": 5.234917778777428,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002617741095327681,
      "loss": 5.2791,
      "mean_token_accuracy": 0.21878644227981567,
      "num_tokens": 102462078.0,
      "step": 52370
    },
    {
      "entropy": 5.988227128982544,
      "epoch": 5.235417603838656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002617384823091537,
      "loss": 5.267,
      "mean_token_accuracy": 0.21442488729953765,
      "num_tokens": 102473241.0,
      "step": 52375
    },
    {
      "entropy": 5.907278680801392,
      "epoch": 5.235917428899885,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002617028554191958,
      "loss": 5.2311,
      "mean_token_accuracy": 0.22426724731922149,
      "num_tokens": 102483687.0,
      "step": 52380
    },
    {
      "entropy": 5.94072847366333,
      "epoch": 5.236417253961114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00026166722886379064,
      "loss": 5.3496,
      "mean_token_accuracy": 0.2095867097377777,
      "num_tokens": 102493568.0,
      "step": 52385
    },
    {
      "entropy": 5.911797189712525,
      "epoch": 5.2369170790223425,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002616316026438346,
      "loss": 5.1771,
      "mean_token_accuracy": 0.22263796776533126,
      "num_tokens": 102503193.0,
      "step": 52390
    },
    {
      "entropy": 5.873530483245849,
      "epoch": 5.23741690408357,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002615959767602239,
      "loss": 5.1824,
      "mean_token_accuracy": 0.23340669125318528,
      "num_tokens": 102512317.0,
      "step": 52395
    },
    {
      "entropy": 5.740651702880859,
      "epoch": 5.237916729144799,
      "grad_norm": 1.234375,
      "learning_rate": 0.000261560351213855,
      "loss": 5.2096,
      "mean_token_accuracy": 0.22005325108766555,
      "num_tokens": 102523879.0,
      "step": 52400
    },
    {
      "entropy": 5.849501085281372,
      "epoch": 5.238416554206028,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00026152472600562423,
      "loss": 5.2628,
      "mean_token_accuracy": 0.20871769040822982,
      "num_tokens": 102534037.0,
      "step": 52405
    },
    {
      "entropy": 5.8707245826721195,
      "epoch": 5.238916379267256,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026148910113642796,
      "loss": 5.1876,
      "mean_token_accuracy": 0.2145090490579605,
      "num_tokens": 102544557.0,
      "step": 52410
    },
    {
      "entropy": 5.868761920928955,
      "epoch": 5.239416204328485,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002614534766071623,
      "loss": 5.1486,
      "mean_token_accuracy": 0.2268352434039116,
      "num_tokens": 102553975.0,
      "step": 52415
    },
    {
      "entropy": 5.927842855453491,
      "epoch": 5.239916029389714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002614178524187237,
      "loss": 5.2391,
      "mean_token_accuracy": 0.2258972480893135,
      "num_tokens": 102563055.0,
      "step": 52420
    },
    {
      "entropy": 5.907087707519532,
      "epoch": 5.2404158544509425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002613822285720085,
      "loss": 5.2548,
      "mean_token_accuracy": 0.22086631059646605,
      "num_tokens": 102572582.0,
      "step": 52425
    },
    {
      "entropy": 5.904215097427368,
      "epoch": 5.24091567951217,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002613466050679129,
      "loss": 5.1786,
      "mean_token_accuracy": 0.22557021528482438,
      "num_tokens": 102581941.0,
      "step": 52430
    },
    {
      "entropy": 5.905588293075562,
      "epoch": 5.241415504573399,
      "grad_norm": 1.25,
      "learning_rate": 0.00026131098190733317,
      "loss": 5.2384,
      "mean_token_accuracy": 0.22150621116161345,
      "num_tokens": 102591057.0,
      "step": 52435
    },
    {
      "entropy": 5.932579278945923,
      "epoch": 5.241915329634628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002612753590911656,
      "loss": 5.3687,
      "mean_token_accuracy": 0.21417678594589235,
      "num_tokens": 102600593.0,
      "step": 52440
    },
    {
      "entropy": 5.907410144805908,
      "epoch": 5.242415154695856,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026123973662030655,
      "loss": 5.2684,
      "mean_token_accuracy": 0.22021348625421525,
      "num_tokens": 102609860.0,
      "step": 52445
    },
    {
      "entropy": 5.943293905258178,
      "epoch": 5.242914979757085,
      "grad_norm": 1.15625,
      "learning_rate": 0.00026120411449565207,
      "loss": 5.2713,
      "mean_token_accuracy": 0.21130329221487046,
      "num_tokens": 102619015.0,
      "step": 52450
    },
    {
      "entropy": 6.024734878540039,
      "epoch": 5.243414804818314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002611684927180986,
      "loss": 5.3371,
      "mean_token_accuracy": 0.21309872418642045,
      "num_tokens": 102629110.0,
      "step": 52455
    },
    {
      "entropy": 5.9613594055175785,
      "epoch": 5.2439146298795425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002611328712885423,
      "loss": 5.3437,
      "mean_token_accuracy": 0.21309296637773514,
      "num_tokens": 102638867.0,
      "step": 52460
    },
    {
      "entropy": 5.926687908172608,
      "epoch": 5.24441445494077,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002610972502078794,
      "loss": 5.2507,
      "mean_token_accuracy": 0.21283649951219558,
      "num_tokens": 102648198.0,
      "step": 52465
    },
    {
      "entropy": 5.88712511062622,
      "epoch": 5.244914280001999,
      "grad_norm": 1.3125,
      "learning_rate": 0.00026106162947700604,
      "loss": 5.2088,
      "mean_token_accuracy": 0.21731491535902023,
      "num_tokens": 102657759.0,
      "step": 52470
    },
    {
      "entropy": 5.9325737953186035,
      "epoch": 5.245414105063228,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0002610260090968185,
      "loss": 5.3542,
      "mean_token_accuracy": 0.2127269595861435,
      "num_tokens": 102667921.0,
      "step": 52475
    },
    {
      "entropy": 5.815478086471558,
      "epoch": 5.245913930124456,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002609903890682131,
      "loss": 5.0807,
      "mean_token_accuracy": 0.23545426577329637,
      "num_tokens": 102677899.0,
      "step": 52480
    },
    {
      "entropy": 5.813585233688355,
      "epoch": 5.246413755185685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00026095476939208583,
      "loss": 5.166,
      "mean_token_accuracy": 0.2266755446791649,
      "num_tokens": 102687993.0,
      "step": 52485
    },
    {
      "entropy": 5.968959856033325,
      "epoch": 5.246913580246914,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002609191500693329,
      "loss": 5.3406,
      "mean_token_accuracy": 0.21791375875473024,
      "num_tokens": 102697260.0,
      "step": 52490
    },
    {
      "entropy": 5.873729658126831,
      "epoch": 5.2474134053081425,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002608835311008506,
      "loss": 5.2393,
      "mean_token_accuracy": 0.2224482461810112,
      "num_tokens": 102707571.0,
      "step": 52495
    },
    {
      "entropy": 5.841374444961548,
      "epoch": 5.24791323036937,
      "grad_norm": 1.203125,
      "learning_rate": 0.000260847912487535,
      "loss": 5.2077,
      "mean_token_accuracy": 0.2266658827662468,
      "num_tokens": 102717465.0,
      "step": 52500
    },
    {
      "entropy": 5.960282897949218,
      "epoch": 5.248413055430599,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002608122942302822,
      "loss": 5.3196,
      "mean_token_accuracy": 0.2161643996834755,
      "num_tokens": 102727368.0,
      "step": 52505
    },
    {
      "entropy": 5.934225368499756,
      "epoch": 5.248912880491828,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00026077667632998854,
      "loss": 5.2823,
      "mean_token_accuracy": 0.21635385900735854,
      "num_tokens": 102736081.0,
      "step": 52510
    },
    {
      "entropy": 5.891939973831176,
      "epoch": 5.249412705553056,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00026074105878754996,
      "loss": 5.1652,
      "mean_token_accuracy": 0.2188998907804489,
      "num_tokens": 102745563.0,
      "step": 52515
    },
    {
      "entropy": 5.9267548561096195,
      "epoch": 5.249912530614285,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002607054416038627,
      "loss": 5.2602,
      "mean_token_accuracy": 0.21688430160284042,
      "num_tokens": 102755101.0,
      "step": 52520
    },
    {
      "entropy": 5.855008268356324,
      "epoch": 5.250412355675514,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002606698247798229,
      "loss": 5.2752,
      "mean_token_accuracy": 0.21987975537776946,
      "num_tokens": 102764382.0,
      "step": 52525
    },
    {
      "entropy": 5.812178373336792,
      "epoch": 5.2509121807367425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002606342083163266,
      "loss": 5.0936,
      "mean_token_accuracy": 0.23364078849554062,
      "num_tokens": 102773709.0,
      "step": 52530
    },
    {
      "entropy": 5.844165277481079,
      "epoch": 5.25141200579797,
      "grad_norm": 1.28125,
      "learning_rate": 0.00026059859221426977,
      "loss": 5.2171,
      "mean_token_accuracy": 0.2232628807425499,
      "num_tokens": 102782345.0,
      "step": 52535
    },
    {
      "entropy": 5.90045485496521,
      "epoch": 5.251911830859199,
      "grad_norm": 1.078125,
      "learning_rate": 0.00026056297647454883,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21322107464075088,
      "num_tokens": 102792618.0,
      "step": 52540
    },
    {
      "entropy": 5.888785457611084,
      "epoch": 5.252411655920428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002605273610980597,
      "loss": 5.2576,
      "mean_token_accuracy": 0.22343017160892487,
      "num_tokens": 102802012.0,
      "step": 52545
    },
    {
      "entropy": 5.943533611297608,
      "epoch": 5.252911480981656,
      "grad_norm": 1.25,
      "learning_rate": 0.0002604917460856985,
      "loss": 5.2866,
      "mean_token_accuracy": 0.2211022213101387,
      "num_tokens": 102812504.0,
      "step": 52550
    },
    {
      "entropy": 5.9466156482696535,
      "epoch": 5.253411306042885,
      "grad_norm": 1.21875,
      "learning_rate": 0.00026045613143836106,
      "loss": 5.2837,
      "mean_token_accuracy": 0.20982359498739242,
      "num_tokens": 102822416.0,
      "step": 52555
    },
    {
      "entropy": 5.905839252471924,
      "epoch": 5.253911131104114,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00026042051715694366,
      "loss": 5.2796,
      "mean_token_accuracy": 0.21996346712112427,
      "num_tokens": 102832837.0,
      "step": 52560
    },
    {
      "entropy": 5.7996796607971195,
      "epoch": 5.2544109561653425,
      "grad_norm": 1.265625,
      "learning_rate": 0.00026038490324234225,
      "loss": 5.1137,
      "mean_token_accuracy": 0.2327963188290596,
      "num_tokens": 102843412.0,
      "step": 52565
    },
    {
      "entropy": 5.850241184234619,
      "epoch": 5.25491078122657,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000260349289695453,
      "loss": 5.1772,
      "mean_token_accuracy": 0.22182457894086838,
      "num_tokens": 102853396.0,
      "step": 52570
    },
    {
      "entropy": 5.901471185684204,
      "epoch": 5.255410606287799,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00026031367651717184,
      "loss": 5.2531,
      "mean_token_accuracy": 0.2177146315574646,
      "num_tokens": 102863818.0,
      "step": 52575
    },
    {
      "entropy": 5.911546897888184,
      "epoch": 5.255910431349028,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002602780637083948,
      "loss": 5.232,
      "mean_token_accuracy": 0.21485411524772643,
      "num_tokens": 102873451.0,
      "step": 52580
    },
    {
      "entropy": 5.9285179615020756,
      "epoch": 5.256410256410256,
      "grad_norm": 1.125,
      "learning_rate": 0.0002602424512700178,
      "loss": 5.2692,
      "mean_token_accuracy": 0.21413127332925797,
      "num_tokens": 102883163.0,
      "step": 52585
    },
    {
      "entropy": 5.919151353836059,
      "epoch": 5.256910081471485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002602068392029371,
      "loss": 5.264,
      "mean_token_accuracy": 0.21430748850107192,
      "num_tokens": 102892298.0,
      "step": 52590
    },
    {
      "entropy": 5.907843065261841,
      "epoch": 5.257409906532714,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00026017122750804833,
      "loss": 5.3173,
      "mean_token_accuracy": 0.21514222919940948,
      "num_tokens": 102901517.0,
      "step": 52595
    },
    {
      "entropy": 5.813120937347412,
      "epoch": 5.2579097315939425,
      "grad_norm": 1.171875,
      "learning_rate": 0.00026013561618624774,
      "loss": 5.2058,
      "mean_token_accuracy": 0.22497093081474304,
      "num_tokens": 102911082.0,
      "step": 52600
    },
    {
      "entropy": 5.997603082656861,
      "epoch": 5.25840955665517,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00026010000523843125,
      "loss": 5.3336,
      "mean_token_accuracy": 0.2057389348745346,
      "num_tokens": 102920547.0,
      "step": 52605
    },
    {
      "entropy": 5.9546185493469235,
      "epoch": 5.258909381716399,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026006439466549474,
      "loss": 5.3107,
      "mean_token_accuracy": 0.2137971669435501,
      "num_tokens": 102930634.0,
      "step": 52610
    },
    {
      "entropy": 5.847706413269043,
      "epoch": 5.259409206777628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002600287844683343,
      "loss": 5.2004,
      "mean_token_accuracy": 0.23020504862070085,
      "num_tokens": 102939889.0,
      "step": 52615
    },
    {
      "entropy": 5.926544857025147,
      "epoch": 5.259909031838856,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002599931746478458,
      "loss": 5.2688,
      "mean_token_accuracy": 0.21575772613286973,
      "num_tokens": 102949645.0,
      "step": 52620
    },
    {
      "entropy": 5.876717662811279,
      "epoch": 5.260408856900085,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002599575652049251,
      "loss": 5.2104,
      "mean_token_accuracy": 0.2276263266801834,
      "num_tokens": 102959379.0,
      "step": 52625
    },
    {
      "entropy": 5.897995853424073,
      "epoch": 5.260908681961314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002599219561404682,
      "loss": 5.2423,
      "mean_token_accuracy": 0.22259788066148758,
      "num_tokens": 102969118.0,
      "step": 52630
    },
    {
      "entropy": 5.939806842803955,
      "epoch": 5.2614085070225425,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000259886347455371,
      "loss": 5.2902,
      "mean_token_accuracy": 0.21794696748256684,
      "num_tokens": 102980513.0,
      "step": 52635
    },
    {
      "entropy": 5.95763053894043,
      "epoch": 5.26190833208377,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002598507391505294,
      "loss": 5.3043,
      "mean_token_accuracy": 0.21323752254247666,
      "num_tokens": 102991255.0,
      "step": 52640
    },
    {
      "entropy": 5.84154577255249,
      "epoch": 5.262408157144999,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002598151312268393,
      "loss": 5.1415,
      "mean_token_accuracy": 0.22564081251621246,
      "num_tokens": 102999851.0,
      "step": 52645
    },
    {
      "entropy": 5.829542398452759,
      "epoch": 5.262907982206228,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025977952368519663,
      "loss": 5.2467,
      "mean_token_accuracy": 0.22088207751512529,
      "num_tokens": 103009418.0,
      "step": 52650
    },
    {
      "entropy": 5.834468364715576,
      "epoch": 5.263407807267456,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002597439165264972,
      "loss": 5.2004,
      "mean_token_accuracy": 0.22461569160223008,
      "num_tokens": 103019366.0,
      "step": 52655
    },
    {
      "entropy": 5.90981879234314,
      "epoch": 5.263907632328685,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002597083097516368,
      "loss": 5.2232,
      "mean_token_accuracy": 0.22483326196670533,
      "num_tokens": 103028567.0,
      "step": 52660
    },
    {
      "entropy": 5.87438817024231,
      "epoch": 5.264407457389914,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002596727033615116,
      "loss": 5.2767,
      "mean_token_accuracy": 0.22108595967292785,
      "num_tokens": 103038296.0,
      "step": 52665
    },
    {
      "entropy": 5.871750020980835,
      "epoch": 5.2649072824511425,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025963709735701715,
      "loss": 5.1943,
      "mean_token_accuracy": 0.22774988561868667,
      "num_tokens": 103048221.0,
      "step": 52670
    },
    {
      "entropy": 5.975310277938843,
      "epoch": 5.26540710751237,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025960149173904937,
      "loss": 5.3483,
      "mean_token_accuracy": 0.2121160849928856,
      "num_tokens": 103056986.0,
      "step": 52675
    },
    {
      "entropy": 5.897495889663697,
      "epoch": 5.265906932573599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025956588650850404,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2176060363650322,
      "num_tokens": 103067073.0,
      "step": 52680
    },
    {
      "entropy": 5.915001916885376,
      "epoch": 5.266406757634828,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002595302816662771,
      "loss": 5.2547,
      "mean_token_accuracy": 0.2182882845401764,
      "num_tokens": 103076798.0,
      "step": 52685
    },
    {
      "entropy": 5.868221950531006,
      "epoch": 5.266906582696056,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0002594946772132643,
      "loss": 5.1792,
      "mean_token_accuracy": 0.22169763296842576,
      "num_tokens": 103085847.0,
      "step": 52690
    },
    {
      "entropy": 5.832156419754028,
      "epoch": 5.267406407757285,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025945907315036144,
      "loss": 5.1447,
      "mean_token_accuracy": 0.2341935008764267,
      "num_tokens": 103095758.0,
      "step": 52695
    },
    {
      "entropy": 5.8463905334472654,
      "epoch": 5.267906232818514,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025942346947846424,
      "loss": 5.2154,
      "mean_token_accuracy": 0.2200687900185585,
      "num_tokens": 103105287.0,
      "step": 52700
    },
    {
      "entropy": 5.840678024291992,
      "epoch": 5.2684060578797425,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002593878661984686,
      "loss": 5.2115,
      "mean_token_accuracy": 0.22647070735692978,
      "num_tokens": 103115018.0,
      "step": 52705
    },
    {
      "entropy": 5.894847631454468,
      "epoch": 5.26890588294097,
      "grad_norm": 1.125,
      "learning_rate": 0.0002593522633112702,
      "loss": 5.2391,
      "mean_token_accuracy": 0.22511493861675264,
      "num_tokens": 103124540.0,
      "step": 52710
    },
    {
      "entropy": 5.947214412689209,
      "epoch": 5.269405708002199,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002593166608177648,
      "loss": 5.2776,
      "mean_token_accuracy": 0.2202374145388603,
      "num_tokens": 103133595.0,
      "step": 52715
    },
    {
      "entropy": 5.829920625686645,
      "epoch": 5.269905533063428,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002592810587188482,
      "loss": 5.235,
      "mean_token_accuracy": 0.22204990088939666,
      "num_tokens": 103143892.0,
      "step": 52720
    },
    {
      "entropy": 5.890667963027954,
      "epoch": 5.270405358124656,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0002592454570154161,
      "loss": 5.1674,
      "mean_token_accuracy": 0.22948518842458726,
      "num_tokens": 103155758.0,
      "step": 52725
    },
    {
      "entropy": 5.9602522373199465,
      "epoch": 5.270905183185885,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025920985570836424,
      "loss": 5.2919,
      "mean_token_accuracy": 0.2136231005191803,
      "num_tokens": 103165966.0,
      "step": 52730
    },
    {
      "entropy": 5.966122055053711,
      "epoch": 5.271405008247114,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002591742547985884,
      "loss": 5.3097,
      "mean_token_accuracy": 0.21770006269216538,
      "num_tokens": 103175958.0,
      "step": 52735
    },
    {
      "entropy": 5.937549829483032,
      "epoch": 5.2719048333083425,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025913865428698425,
      "loss": 5.2737,
      "mean_token_accuracy": 0.21715812385082245,
      "num_tokens": 103184779.0,
      "step": 52740
    },
    {
      "entropy": 5.935439491271973,
      "epoch": 5.27240465836957,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002591030541744475,
      "loss": 5.2891,
      "mean_token_accuracy": 0.2164314031600952,
      "num_tokens": 103194425.0,
      "step": 52745
    },
    {
      "entropy": 5.930576467514038,
      "epoch": 5.272904483430799,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0002590674544618737,
      "loss": 5.3171,
      "mean_token_accuracy": 0.20882369726896285,
      "num_tokens": 103204040.0,
      "step": 52750
    },
    {
      "entropy": 5.866606950759888,
      "epoch": 5.273404308492028,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025903185515015875,
      "loss": 5.1997,
      "mean_token_accuracy": 0.22305389493703842,
      "num_tokens": 103213435.0,
      "step": 52755
    },
    {
      "entropy": 5.899068117141724,
      "epoch": 5.273904133553256,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025899625624019817,
      "loss": 5.1866,
      "mean_token_accuracy": 0.21949539035558702,
      "num_tokens": 103223724.0,
      "step": 52760
    },
    {
      "entropy": 5.989722013473511,
      "epoch": 5.274403958614485,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0002589606577328876,
      "loss": 5.3057,
      "mean_token_accuracy": 0.21043492257595062,
      "num_tokens": 103233326.0,
      "step": 52765
    },
    {
      "entropy": 5.922237586975098,
      "epoch": 5.274903783675714,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002589250596291229,
      "loss": 5.1817,
      "mean_token_accuracy": 0.22432285845279692,
      "num_tokens": 103243805.0,
      "step": 52770
    },
    {
      "entropy": 5.83468108177185,
      "epoch": 5.2754036087369425,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025888946192979947,
      "loss": 5.1856,
      "mean_token_accuracy": 0.2265207752585411,
      "num_tokens": 103253519.0,
      "step": 52775
    },
    {
      "entropy": 5.880078411102295,
      "epoch": 5.27590343379817,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000258853864635813,
      "loss": 5.2083,
      "mean_token_accuracy": 0.2150769367814064,
      "num_tokens": 103262231.0,
      "step": 52780
    },
    {
      "entropy": 5.85491909980774,
      "epoch": 5.276403258859399,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025881826774805914,
      "loss": 5.2169,
      "mean_token_accuracy": 0.22241325825452804,
      "num_tokens": 103272671.0,
      "step": 52785
    },
    {
      "entropy": 5.922617721557617,
      "epoch": 5.276903083920628,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025878267126743356,
      "loss": 5.2424,
      "mean_token_accuracy": 0.21957534551620483,
      "num_tokens": 103282220.0,
      "step": 52790
    },
    {
      "entropy": 5.9177085876464846,
      "epoch": 5.277402908981856,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025874707519483184,
      "loss": 5.2079,
      "mean_token_accuracy": 0.22811536937952043,
      "num_tokens": 103291973.0,
      "step": 52795
    },
    {
      "entropy": 5.910307264328003,
      "epoch": 5.277902734043085,
      "grad_norm": 1.25,
      "learning_rate": 0.00025871147953114945,
      "loss": 5.2388,
      "mean_token_accuracy": 0.2204159528017044,
      "num_tokens": 103300978.0,
      "step": 52800
    },
    {
      "entropy": 5.930911540985107,
      "epoch": 5.278402559104314,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002586758842772821,
      "loss": 5.2381,
      "mean_token_accuracy": 0.2101680874824524,
      "num_tokens": 103310132.0,
      "step": 52805
    },
    {
      "entropy": 5.990840148925781,
      "epoch": 5.2789023841655425,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002586402894341252,
      "loss": 5.2723,
      "mean_token_accuracy": 0.20949028581380844,
      "num_tokens": 103318798.0,
      "step": 52810
    },
    {
      "entropy": 5.895647478103638,
      "epoch": 5.27940220922677,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002586046950025745,
      "loss": 5.2512,
      "mean_token_accuracy": 0.22118860781192778,
      "num_tokens": 103328461.0,
      "step": 52815
    },
    {
      "entropy": 5.912070178985596,
      "epoch": 5.279902034287999,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025856910098352536,
      "loss": 5.2909,
      "mean_token_accuracy": 0.21866931170225143,
      "num_tokens": 103340004.0,
      "step": 52820
    },
    {
      "entropy": 5.967623376846314,
      "epoch": 5.280401859349228,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002585335073778736,
      "loss": 5.3143,
      "mean_token_accuracy": 0.20814612954854966,
      "num_tokens": 103349467.0,
      "step": 52825
    },
    {
      "entropy": 6.003772974014282,
      "epoch": 5.280901684410456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002584979141865144,
      "loss": 5.343,
      "mean_token_accuracy": 0.20950988531112671,
      "num_tokens": 103359776.0,
      "step": 52830
    },
    {
      "entropy": 5.875549936294556,
      "epoch": 5.281401509471685,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002584623214103435,
      "loss": 5.2018,
      "mean_token_accuracy": 0.22071751803159714,
      "num_tokens": 103369581.0,
      "step": 52835
    },
    {
      "entropy": 5.8402125358581545,
      "epoch": 5.281901334532914,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002584267290502562,
      "loss": 5.1793,
      "mean_token_accuracy": 0.22700927257537842,
      "num_tokens": 103379386.0,
      "step": 52840
    },
    {
      "entropy": 5.91510238647461,
      "epoch": 5.2824011595941425,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002583911371071483,
      "loss": 5.2906,
      "mean_token_accuracy": 0.21345308125019075,
      "num_tokens": 103388706.0,
      "step": 52845
    },
    {
      "entropy": 5.9720264911651615,
      "epoch": 5.28290098465537,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025835554558191503,
      "loss": 5.2626,
      "mean_token_accuracy": 0.21591711938381195,
      "num_tokens": 103398732.0,
      "step": 52850
    },
    {
      "entropy": 5.892266464233399,
      "epoch": 5.283400809716599,
      "grad_norm": 1.125,
      "learning_rate": 0.000258319954475452,
      "loss": 5.2334,
      "mean_token_accuracy": 0.21978906989097596,
      "num_tokens": 103408797.0,
      "step": 52855
    },
    {
      "entropy": 6.004144334793091,
      "epoch": 5.283900634777828,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002582843637886546,
      "loss": 5.3265,
      "mean_token_accuracy": 0.20800641775131226,
      "num_tokens": 103418685.0,
      "step": 52860
    },
    {
      "entropy": 5.874409770965576,
      "epoch": 5.284400459839056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002582487735224183,
      "loss": 5.1922,
      "mean_token_accuracy": 0.22263199537992479,
      "num_tokens": 103428060.0,
      "step": 52865
    },
    {
      "entropy": 5.985567378997803,
      "epoch": 5.284900284900285,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00025821318367763856,
      "loss": 5.2679,
      "mean_token_accuracy": 0.219546315073967,
      "num_tokens": 103438255.0,
      "step": 52870
    },
    {
      "entropy": 5.9073645114898685,
      "epoch": 5.285400109961514,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025817759425521075,
      "loss": 5.267,
      "mean_token_accuracy": 0.21902344077825547,
      "num_tokens": 103447900.0,
      "step": 52875
    },
    {
      "entropy": 5.891483783721924,
      "epoch": 5.2858999350227425,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002581420052560303,
      "loss": 5.2297,
      "mean_token_accuracy": 0.21286522150039672,
      "num_tokens": 103456841.0,
      "step": 52880
    },
    {
      "entropy": 5.912940883636475,
      "epoch": 5.28639976008397,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0002581064166809928,
      "loss": 5.292,
      "mean_token_accuracy": 0.2179124355316162,
      "num_tokens": 103465869.0,
      "step": 52885
    },
    {
      "entropy": 5.897954940795898,
      "epoch": 5.286899585145199,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025807082853099343,
      "loss": 5.3095,
      "mean_token_accuracy": 0.2100964903831482,
      "num_tokens": 103476562.0,
      "step": 52890
    },
    {
      "entropy": 5.685016775131226,
      "epoch": 5.287399410206428,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025803524080692783,
      "loss": 5.0396,
      "mean_token_accuracy": 0.24176277965307236,
      "num_tokens": 103486935.0,
      "step": 52895
    },
    {
      "entropy": 5.915978813171387,
      "epoch": 5.287899235267656,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000257999653509691,
      "loss": 5.2313,
      "mean_token_accuracy": 0.21685445606708526,
      "num_tokens": 103496599.0,
      "step": 52900
    },
    {
      "entropy": 5.905636501312256,
      "epoch": 5.288399060328885,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002579640666401785,
      "loss": 5.241,
      "mean_token_accuracy": 0.22563211619853973,
      "num_tokens": 103506858.0,
      "step": 52905
    },
    {
      "entropy": 5.924311590194702,
      "epoch": 5.288898885390114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025792848019928576,
      "loss": 5.2624,
      "mean_token_accuracy": 0.21646825075149537,
      "num_tokens": 103516023.0,
      "step": 52910
    },
    {
      "entropy": 5.903849935531616,
      "epoch": 5.2893987104513425,
      "grad_norm": 1.328125,
      "learning_rate": 0.00025789289418790814,
      "loss": 5.216,
      "mean_token_accuracy": 0.22447367161512374,
      "num_tokens": 103526506.0,
      "step": 52915
    },
    {
      "entropy": 5.902511930465698,
      "epoch": 5.28989853551257,
      "grad_norm": 1.40625,
      "learning_rate": 0.0002578573086069408,
      "loss": 5.2155,
      "mean_token_accuracy": 0.22032740265130996,
      "num_tokens": 103535881.0,
      "step": 52920
    },
    {
      "entropy": 5.990655755996704,
      "epoch": 5.290398360573799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002578217234572792,
      "loss": 5.3157,
      "mean_token_accuracy": 0.21893662959337234,
      "num_tokens": 103546115.0,
      "step": 52925
    },
    {
      "entropy": 5.8700464248657225,
      "epoch": 5.290898185635028,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0002577861387398186,
      "loss": 5.1546,
      "mean_token_accuracy": 0.2268998146057129,
      "num_tokens": 103555044.0,
      "step": 52930
    },
    {
      "entropy": 5.795075511932373,
      "epoch": 5.291398010696256,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002577505544554543,
      "loss": 5.1278,
      "mean_token_accuracy": 0.22707681655883788,
      "num_tokens": 103564816.0,
      "step": 52935
    },
    {
      "entropy": 5.900357389450074,
      "epoch": 5.291897835757485,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002577149706050817,
      "loss": 5.2974,
      "mean_token_accuracy": 0.21986227631568908,
      "num_tokens": 103575645.0,
      "step": 52940
    },
    {
      "entropy": 5.915937614440918,
      "epoch": 5.292397660818714,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002576793871895959,
      "loss": 5.2779,
      "mean_token_accuracy": 0.21913521289825438,
      "num_tokens": 103586161.0,
      "step": 52945
    },
    {
      "entropy": 5.908196258544922,
      "epoch": 5.2928974858799425,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025764380420989226,
      "loss": 5.251,
      "mean_token_accuracy": 0.2247714176774025,
      "num_tokens": 103595689.0,
      "step": 52950
    },
    {
      "entropy": 5.962975549697876,
      "epoch": 5.29339731094117,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002576082216668661,
      "loss": 5.3106,
      "mean_token_accuracy": 0.2111186295747757,
      "num_tokens": 103605763.0,
      "step": 52955
    },
    {
      "entropy": 5.952384567260742,
      "epoch": 5.293897136002399,
      "grad_norm": 1.390625,
      "learning_rate": 0.0002575726395614126,
      "loss": 5.329,
      "mean_token_accuracy": 0.21730370223522186,
      "num_tokens": 103615167.0,
      "step": 52960
    },
    {
      "entropy": 5.903452444076538,
      "epoch": 5.294396961063628,
      "grad_norm": 1.734375,
      "learning_rate": 0.000257537057894427,
      "loss": 5.2817,
      "mean_token_accuracy": 0.21933430880308152,
      "num_tokens": 103626621.0,
      "step": 52965
    },
    {
      "entropy": 5.870958948135376,
      "epoch": 5.294896786124856,
      "grad_norm": 1.125,
      "learning_rate": 0.00025750147666680454,
      "loss": 5.2355,
      "mean_token_accuracy": 0.22583871185779572,
      "num_tokens": 103637050.0,
      "step": 52970
    },
    {
      "entropy": 5.899656534194946,
      "epoch": 5.295396611186085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002574658958794404,
      "loss": 5.2287,
      "mean_token_accuracy": 0.22392577975988387,
      "num_tokens": 103647104.0,
      "step": 52975
    },
    {
      "entropy": 5.938037157058716,
      "epoch": 5.295896436247314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002574303155332298,
      "loss": 5.2735,
      "mean_token_accuracy": 0.2189952939748764,
      "num_tokens": 103656503.0,
      "step": 52980
    },
    {
      "entropy": 5.95068039894104,
      "epoch": 5.2963962613085425,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00025739473562906795,
      "loss": 5.3079,
      "mean_token_accuracy": 0.21506889313459396,
      "num_tokens": 103666661.0,
      "step": 52985
    },
    {
      "entropy": 5.9607014656066895,
      "epoch": 5.29689608636977,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00025735915616785005,
      "loss": 5.3331,
      "mean_token_accuracy": 0.2188686653971672,
      "num_tokens": 103676542.0,
      "step": 52990
    },
    {
      "entropy": 5.965414667129517,
      "epoch": 5.297395911430999,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025732357715047124,
      "loss": 5.3172,
      "mean_token_accuracy": 0.21146595925092698,
      "num_tokens": 103686313.0,
      "step": 52995
    },
    {
      "entropy": 5.990096759796143,
      "epoch": 5.297895736492228,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025728799857782663,
      "loss": 5.3227,
      "mean_token_accuracy": 0.2116812840104103,
      "num_tokens": 103695855.0,
      "step": 53000
    },
    {
      "entropy": 5.9500891208648685,
      "epoch": 5.298395561553456,
      "grad_norm": 1.25,
      "learning_rate": 0.0002572524204508114,
      "loss": 5.2827,
      "mean_token_accuracy": 0.21994890570640563,
      "num_tokens": 103705861.0,
      "step": 53005
    },
    {
      "entropy": 5.96044397354126,
      "epoch": 5.298895386614685,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00025721684277032076,
      "loss": 5.382,
      "mean_token_accuracy": 0.20922535061836242,
      "num_tokens": 103715285.0,
      "step": 53010
    },
    {
      "entropy": 5.895293664932251,
      "epoch": 5.299395211675914,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025718126553724973,
      "loss": 5.2143,
      "mean_token_accuracy": 0.2266175001859665,
      "num_tokens": 103725339.0,
      "step": 53015
    },
    {
      "entropy": 5.857324647903442,
      "epoch": 5.2998950367371425,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025714568875249353,
      "loss": 5.253,
      "mean_token_accuracy": 0.21811531931161882,
      "num_tokens": 103735110.0,
      "step": 53020
    },
    {
      "entropy": 5.7883148193359375,
      "epoch": 5.30039486179837,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002571101124169472,
      "loss": 5.1002,
      "mean_token_accuracy": 0.23205496966838837,
      "num_tokens": 103744342.0,
      "step": 53025
    },
    {
      "entropy": 5.904771375656128,
      "epoch": 5.300894686859599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002570745365315058,
      "loss": 5.27,
      "mean_token_accuracy": 0.2227024808526039,
      "num_tokens": 103753765.0,
      "step": 53030
    },
    {
      "entropy": 5.898449182510376,
      "epoch": 5.301394511920828,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002570389610970645,
      "loss": 5.2371,
      "mean_token_accuracy": 0.22569110244512558,
      "num_tokens": 103762870.0,
      "step": 53035
    },
    {
      "entropy": 6.042620468139648,
      "epoch": 5.301894336982056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025700338611451835,
      "loss": 5.3557,
      "mean_token_accuracy": 0.2112701416015625,
      "num_tokens": 103773203.0,
      "step": 53040
    },
    {
      "entropy": 5.834501552581787,
      "epoch": 5.302394162043285,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00025696781158476227,
      "loss": 5.1472,
      "mean_token_accuracy": 0.23570453971624375,
      "num_tokens": 103784033.0,
      "step": 53045
    },
    {
      "entropy": 5.908428525924682,
      "epoch": 5.302893987104514,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002569322375086915,
      "loss": 5.2722,
      "mean_token_accuracy": 0.21537960320711136,
      "num_tokens": 103792981.0,
      "step": 53050
    },
    {
      "entropy": 5.835320329666137,
      "epoch": 5.3033938121657425,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025689666388720097,
      "loss": 5.1389,
      "mean_token_accuracy": 0.22885358482599258,
      "num_tokens": 103801876.0,
      "step": 53055
    },
    {
      "entropy": 5.943123435974121,
      "epoch": 5.30389363722697,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025686109072118566,
      "loss": 5.2547,
      "mean_token_accuracy": 0.21271005570888518,
      "num_tokens": 103811650.0,
      "step": 53060
    },
    {
      "entropy": 5.911903524398804,
      "epoch": 5.304393462288199,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025682551801154064,
      "loss": 5.2466,
      "mean_token_accuracy": 0.21652437150478362,
      "num_tokens": 103821538.0,
      "step": 53065
    },
    {
      "entropy": 5.867128229141235,
      "epoch": 5.304893287349428,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002567899457591609,
      "loss": 5.1203,
      "mean_token_accuracy": 0.22794503569602967,
      "num_tokens": 103830928.0,
      "step": 53070
    },
    {
      "entropy": 5.907455587387085,
      "epoch": 5.305393112410656,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002567543739649415,
      "loss": 5.2898,
      "mean_token_accuracy": 0.21272925436496734,
      "num_tokens": 103840431.0,
      "step": 53075
    },
    {
      "entropy": 5.860916948318481,
      "epoch": 5.305892937471885,
      "grad_norm": 1.25,
      "learning_rate": 0.00025671880262977733,
      "loss": 5.1636,
      "mean_token_accuracy": 0.2194179117679596,
      "num_tokens": 103850011.0,
      "step": 53080
    },
    {
      "entropy": 5.995817947387695,
      "epoch": 5.306392762533114,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025668323175456336,
      "loss": 5.2662,
      "mean_token_accuracy": 0.21446956247091292,
      "num_tokens": 103859205.0,
      "step": 53085
    },
    {
      "entropy": 5.893934345245361,
      "epoch": 5.306892587594342,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002566476613401946,
      "loss": 5.0896,
      "mean_token_accuracy": 0.2339630529284477,
      "num_tokens": 103867941.0,
      "step": 53090
    },
    {
      "entropy": 5.922832918167114,
      "epoch": 5.30739241265557,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025661209138756595,
      "loss": 5.2589,
      "mean_token_accuracy": 0.21352357566356658,
      "num_tokens": 103877789.0,
      "step": 53095
    },
    {
      "entropy": 5.9862974166870115,
      "epoch": 5.307892237716799,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025657652189757236,
      "loss": 5.3906,
      "mean_token_accuracy": 0.2045430451631546,
      "num_tokens": 103888000.0,
      "step": 53100
    },
    {
      "entropy": 5.939726400375366,
      "epoch": 5.308392062778028,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025654095287110877,
      "loss": 5.2591,
      "mean_token_accuracy": 0.2159729689359665,
      "num_tokens": 103897741.0,
      "step": 53105
    },
    {
      "entropy": 5.896569442749024,
      "epoch": 5.308891887839256,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025650538430907,
      "loss": 5.1747,
      "mean_token_accuracy": 0.21543753147125244,
      "num_tokens": 103906147.0,
      "step": 53110
    },
    {
      "entropy": 5.838596296310425,
      "epoch": 5.309391712900485,
      "grad_norm": 1.125,
      "learning_rate": 0.00025646981621235107,
      "loss": 5.2208,
      "mean_token_accuracy": 0.21588409543037415,
      "num_tokens": 103916700.0,
      "step": 53115
    },
    {
      "entropy": 5.735418701171875,
      "epoch": 5.309891537961714,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002564342485818467,
      "loss": 5.1699,
      "mean_token_accuracy": 0.2302851215004921,
      "num_tokens": 103927027.0,
      "step": 53120
    },
    {
      "entropy": 5.952391910552978,
      "epoch": 5.310391363022942,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025639868141845185,
      "loss": 5.261,
      "mean_token_accuracy": 0.22622265070676803,
      "num_tokens": 103935962.0,
      "step": 53125
    },
    {
      "entropy": 5.993727159500122,
      "epoch": 5.31089118808417,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025636311472306147,
      "loss": 5.3061,
      "mean_token_accuracy": 0.2169194146990776,
      "num_tokens": 103947198.0,
      "step": 53130
    },
    {
      "entropy": 5.829628372192383,
      "epoch": 5.311391013145399,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025632754849657025,
      "loss": 5.1654,
      "mean_token_accuracy": 0.22719730883836747,
      "num_tokens": 103957425.0,
      "step": 53135
    },
    {
      "entropy": 5.901275157928467,
      "epoch": 5.311890838206628,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0002562919827398732,
      "loss": 5.1978,
      "mean_token_accuracy": 0.22807809859514236,
      "num_tokens": 103967197.0,
      "step": 53140
    },
    {
      "entropy": 5.879093074798584,
      "epoch": 5.312390663267856,
      "grad_norm": 1.265625,
      "learning_rate": 0.000256256417453865,
      "loss": 5.2543,
      "mean_token_accuracy": 0.2213377133011818,
      "num_tokens": 103976742.0,
      "step": 53145
    },
    {
      "entropy": 5.944814157485962,
      "epoch": 5.312890488329085,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002562208526394405,
      "loss": 5.3276,
      "mean_token_accuracy": 0.2102476477622986,
      "num_tokens": 103986834.0,
      "step": 53150
    },
    {
      "entropy": 5.890039110183716,
      "epoch": 5.313390313390314,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002561852882974945,
      "loss": 5.2193,
      "mean_token_accuracy": 0.227518929541111,
      "num_tokens": 103997082.0,
      "step": 53155
    },
    {
      "entropy": 5.919327116012573,
      "epoch": 5.313890138451542,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025614972442892186,
      "loss": 5.2212,
      "mean_token_accuracy": 0.21464178711175919,
      "num_tokens": 104005844.0,
      "step": 53160
    },
    {
      "entropy": 5.8391303539276125,
      "epoch": 5.31438996351277,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002561141610346172,
      "loss": 5.1422,
      "mean_token_accuracy": 0.2291133686900139,
      "num_tokens": 104016540.0,
      "step": 53165
    },
    {
      "entropy": 5.8837910175323485,
      "epoch": 5.314889788573999,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002560785981154755,
      "loss": 5.2569,
      "mean_token_accuracy": 0.22213864475488662,
      "num_tokens": 104025815.0,
      "step": 53170
    },
    {
      "entropy": 5.92651686668396,
      "epoch": 5.315389613635228,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025604303567239143,
      "loss": 5.2637,
      "mean_token_accuracy": 0.2132747381925583,
      "num_tokens": 104035019.0,
      "step": 53175
    },
    {
      "entropy": 5.8557363986969,
      "epoch": 5.315889438696456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025600747370625965,
      "loss": 5.2127,
      "mean_token_accuracy": 0.224453404545784,
      "num_tokens": 104044604.0,
      "step": 53180
    },
    {
      "entropy": 5.8656445980072025,
      "epoch": 5.316389263757685,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000255971912217975,
      "loss": 5.2417,
      "mean_token_accuracy": 0.22644531726837158,
      "num_tokens": 104054763.0,
      "step": 53185
    },
    {
      "entropy": 5.983596277236939,
      "epoch": 5.316889088818914,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025593635120843203,
      "loss": 5.36,
      "mean_token_accuracy": 0.20350251346826553,
      "num_tokens": 104064020.0,
      "step": 53190
    },
    {
      "entropy": 5.869630193710327,
      "epoch": 5.317388913880142,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025590079067852557,
      "loss": 5.1776,
      "mean_token_accuracy": 0.22055005729198457,
      "num_tokens": 104073702.0,
      "step": 53195
    },
    {
      "entropy": 5.899699258804321,
      "epoch": 5.31788873894137,
      "grad_norm": 1.390625,
      "learning_rate": 0.00025586523062915036,
      "loss": 5.251,
      "mean_token_accuracy": 0.22351984679698944,
      "num_tokens": 104083548.0,
      "step": 53200
    },
    {
      "entropy": 5.8341185569763185,
      "epoch": 5.318388564002599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025582967106120097,
      "loss": 5.1671,
      "mean_token_accuracy": 0.22447037249803542,
      "num_tokens": 104093534.0,
      "step": 53205
    },
    {
      "entropy": 5.886226558685303,
      "epoch": 5.318888389063828,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002557941119755722,
      "loss": 5.2714,
      "mean_token_accuracy": 0.22151773124933244,
      "num_tokens": 104103013.0,
      "step": 53210
    },
    {
      "entropy": 5.87653169631958,
      "epoch": 5.319388214125056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002557585533731586,
      "loss": 5.2251,
      "mean_token_accuracy": 0.22683295905590056,
      "num_tokens": 104112198.0,
      "step": 53215
    },
    {
      "entropy": 5.879458236694336,
      "epoch": 5.319888039186285,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002557229952548548,
      "loss": 5.1691,
      "mean_token_accuracy": 0.2225756898522377,
      "num_tokens": 104121267.0,
      "step": 53220
    },
    {
      "entropy": 5.919272994995117,
      "epoch": 5.320387864247514,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002556874376215555,
      "loss": 5.259,
      "mean_token_accuracy": 0.22492325454950332,
      "num_tokens": 104131210.0,
      "step": 53225
    },
    {
      "entropy": 5.884804725646973,
      "epoch": 5.320887689308742,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025565188047415525,
      "loss": 5.2358,
      "mean_token_accuracy": 0.2252672240138054,
      "num_tokens": 104140762.0,
      "step": 53230
    },
    {
      "entropy": 5.9083717346191404,
      "epoch": 5.32138751436997,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002556163238135488,
      "loss": 5.3265,
      "mean_token_accuracy": 0.22128693461418153,
      "num_tokens": 104150690.0,
      "step": 53235
    },
    {
      "entropy": 5.865398645401001,
      "epoch": 5.321887339431199,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025558076764063063,
      "loss": 5.1951,
      "mean_token_accuracy": 0.22104474008083344,
      "num_tokens": 104161418.0,
      "step": 53240
    },
    {
      "entropy": 5.922825193405151,
      "epoch": 5.322387164492428,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002555452119562955,
      "loss": 5.2551,
      "mean_token_accuracy": 0.2144236147403717,
      "num_tokens": 104171130.0,
      "step": 53245
    },
    {
      "entropy": 5.9370361328125,
      "epoch": 5.322886989553656,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025550965676143756,
      "loss": 5.2991,
      "mean_token_accuracy": 0.21801159679889678,
      "num_tokens": 104181450.0,
      "step": 53250
    },
    {
      "entropy": 5.85220193862915,
      "epoch": 5.323386814614885,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025547410205695176,
      "loss": 5.2012,
      "mean_token_accuracy": 0.22399470657110215,
      "num_tokens": 104192804.0,
      "step": 53255
    },
    {
      "entropy": 5.846236515045166,
      "epoch": 5.323886639676114,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025543854784373254,
      "loss": 5.1767,
      "mean_token_accuracy": 0.22494471073150635,
      "num_tokens": 104202544.0,
      "step": 53260
    },
    {
      "entropy": 5.868071126937866,
      "epoch": 5.324386464737342,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025540299412267443,
      "loss": 5.2472,
      "mean_token_accuracy": 0.22571391463279725,
      "num_tokens": 104211872.0,
      "step": 53265
    },
    {
      "entropy": 5.910371875762939,
      "epoch": 5.32488628979857,
      "grad_norm": 1.25,
      "learning_rate": 0.0002553674408946719,
      "loss": 5.2262,
      "mean_token_accuracy": 0.21038477271795272,
      "num_tokens": 104221126.0,
      "step": 53270
    },
    {
      "entropy": 5.876013851165771,
      "epoch": 5.325386114859799,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002553318881606195,
      "loss": 5.2209,
      "mean_token_accuracy": 0.21952515095472336,
      "num_tokens": 104231249.0,
      "step": 53275
    },
    {
      "entropy": 5.916838884353638,
      "epoch": 5.325885939921028,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025529633592141185,
      "loss": 5.2841,
      "mean_token_accuracy": 0.21810927391052246,
      "num_tokens": 104240893.0,
      "step": 53280
    },
    {
      "entropy": 5.870070219039917,
      "epoch": 5.326385764982256,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00025526078417794317,
      "loss": 5.2716,
      "mean_token_accuracy": 0.22799807488918306,
      "num_tokens": 104251377.0,
      "step": 53285
    },
    {
      "entropy": 5.9330747604370115,
      "epoch": 5.326885590043485,
      "grad_norm": 1.359375,
      "learning_rate": 0.0002552252329311081,
      "loss": 5.233,
      "mean_token_accuracy": 0.21853076219558715,
      "num_tokens": 104260207.0,
      "step": 53290
    },
    {
      "entropy": 5.895871782302857,
      "epoch": 5.327385415104714,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025518968218180113,
      "loss": 5.2005,
      "mean_token_accuracy": 0.22343168407678604,
      "num_tokens": 104270805.0,
      "step": 53295
    },
    {
      "entropy": 5.9417437553405765,
      "epoch": 5.327885240165942,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002551541319309166,
      "loss": 5.4018,
      "mean_token_accuracy": 0.20950444638729096,
      "num_tokens": 104280995.0,
      "step": 53300
    },
    {
      "entropy": 6.017018365859985,
      "epoch": 5.32838506522717,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025511858217934914,
      "loss": 5.3349,
      "mean_token_accuracy": 0.21202118694782257,
      "num_tokens": 104290097.0,
      "step": 53305
    },
    {
      "entropy": 5.943179130554199,
      "epoch": 5.328884890288399,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002550830329279929,
      "loss": 5.2231,
      "mean_token_accuracy": 0.22612735778093337,
      "num_tokens": 104299168.0,
      "step": 53310
    },
    {
      "entropy": 5.858639097213745,
      "epoch": 5.329384715349628,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025504748417774255,
      "loss": 5.1796,
      "mean_token_accuracy": 0.23201383501291276,
      "num_tokens": 104308965.0,
      "step": 53315
    },
    {
      "entropy": 5.877280282974243,
      "epoch": 5.329884540410856,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002550119359294923,
      "loss": 5.2364,
      "mean_token_accuracy": 0.226655812561512,
      "num_tokens": 104318737.0,
      "step": 53320
    },
    {
      "entropy": 5.829895973205566,
      "epoch": 5.330384365472085,
      "grad_norm": 1.125,
      "learning_rate": 0.00025497638818413655,
      "loss": 5.1173,
      "mean_token_accuracy": 0.22951022386550904,
      "num_tokens": 104329198.0,
      "step": 53325
    },
    {
      "entropy": 5.916851615905761,
      "epoch": 5.330884190533314,
      "grad_norm": 1.265625,
      "learning_rate": 0.00025494084094256975,
      "loss": 5.3266,
      "mean_token_accuracy": 0.20987309515476227,
      "num_tokens": 104339258.0,
      "step": 53330
    },
    {
      "entropy": 5.8447473526000975,
      "epoch": 5.331384015594542,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002549052942056863,
      "loss": 5.1038,
      "mean_token_accuracy": 0.22488575875759126,
      "num_tokens": 104350326.0,
      "step": 53335
    },
    {
      "entropy": 5.897191905975342,
      "epoch": 5.33188384065577,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002548697479743804,
      "loss": 5.2675,
      "mean_token_accuracy": 0.21544315069913864,
      "num_tokens": 104360233.0,
      "step": 53340
    },
    {
      "entropy": 5.844606685638428,
      "epoch": 5.332383665716999,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025483420224954644,
      "loss": 5.1373,
      "mean_token_accuracy": 0.22427905201911927,
      "num_tokens": 104369561.0,
      "step": 53345
    },
    {
      "entropy": 5.892821788787842,
      "epoch": 5.332883490778228,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002547986570320788,
      "loss": 5.2748,
      "mean_token_accuracy": 0.21800672858953477,
      "num_tokens": 104379093.0,
      "step": 53350
    },
    {
      "entropy": 5.895610427856445,
      "epoch": 5.333383315839456,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025476311232287173,
      "loss": 5.1874,
      "mean_token_accuracy": 0.22258628755807877,
      "num_tokens": 104387420.0,
      "step": 53355
    },
    {
      "entropy": 5.840835857391357,
      "epoch": 5.333883140900685,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002547275681228196,
      "loss": 5.1886,
      "mean_token_accuracy": 0.22718882262706758,
      "num_tokens": 104397529.0,
      "step": 53360
    },
    {
      "entropy": 5.862765789031982,
      "epoch": 5.334382965961913,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002546920244328166,
      "loss": 5.1749,
      "mean_token_accuracy": 0.21923595666885376,
      "num_tokens": 104407749.0,
      "step": 53365
    },
    {
      "entropy": 5.813943004608154,
      "epoch": 5.334882791023142,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025465648125375703,
      "loss": 5.1101,
      "mean_token_accuracy": 0.23478541374206544,
      "num_tokens": 104416802.0,
      "step": 53370
    },
    {
      "entropy": 5.932362985610962,
      "epoch": 5.33538261608437,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025462093858653515,
      "loss": 5.3398,
      "mean_token_accuracy": 0.2060275286436081,
      "num_tokens": 104426290.0,
      "step": 53375
    },
    {
      "entropy": 5.848778533935547,
      "epoch": 5.335882441145599,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002545853964320453,
      "loss": 5.2004,
      "mean_token_accuracy": 0.21826415061950682,
      "num_tokens": 104436192.0,
      "step": 53380
    },
    {
      "entropy": 5.871652746200562,
      "epoch": 5.336382266206828,
      "grad_norm": 1.28125,
      "learning_rate": 0.0002545498547911816,
      "loss": 5.2619,
      "mean_token_accuracy": 0.21418429762125016,
      "num_tokens": 104446479.0,
      "step": 53385
    },
    {
      "entropy": 5.869728660583496,
      "epoch": 5.336882091268056,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002545143136648382,
      "loss": 5.1846,
      "mean_token_accuracy": 0.22679048627614976,
      "num_tokens": 104456089.0,
      "step": 53390
    },
    {
      "entropy": 5.825775003433227,
      "epoch": 5.337381916329285,
      "grad_norm": 1.28125,
      "learning_rate": 0.00025447877305390945,
      "loss": 5.1466,
      "mean_token_accuracy": 0.23078955560922623,
      "num_tokens": 104466702.0,
      "step": 53395
    },
    {
      "entropy": 5.939886856079101,
      "epoch": 5.337881741390513,
      "grad_norm": 1.25,
      "learning_rate": 0.00025444323295928953,
      "loss": 5.2677,
      "mean_token_accuracy": 0.21466211080551148,
      "num_tokens": 104476311.0,
      "step": 53400
    },
    {
      "entropy": 5.894931793212891,
      "epoch": 5.3383815664517416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002544076933818725,
      "loss": 5.2365,
      "mean_token_accuracy": 0.22577834129333496,
      "num_tokens": 104486774.0,
      "step": 53405
    },
    {
      "entropy": 5.895291471481324,
      "epoch": 5.33888139151297,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002543721543225526,
      "loss": 5.2185,
      "mean_token_accuracy": 0.21345227658748628,
      "num_tokens": 104496356.0,
      "step": 53410
    },
    {
      "entropy": 5.888087797164917,
      "epoch": 5.339381216574199,
      "grad_norm": 1.25,
      "learning_rate": 0.000254336615782224,
      "loss": 5.2559,
      "mean_token_accuracy": 0.21432948261499404,
      "num_tokens": 104504984.0,
      "step": 53415
    },
    {
      "entropy": 5.905791616439819,
      "epoch": 5.339881041635428,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025430107776178086,
      "loss": 5.211,
      "mean_token_accuracy": 0.2172063410282135,
      "num_tokens": 104513491.0,
      "step": 53420
    },
    {
      "entropy": 5.895603942871094,
      "epoch": 5.340380866696656,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002542655402621173,
      "loss": 5.2853,
      "mean_token_accuracy": 0.21797226667404174,
      "num_tokens": 104523630.0,
      "step": 53425
    },
    {
      "entropy": 5.898230123519897,
      "epoch": 5.340880691757885,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002542300032841274,
      "loss": 5.2924,
      "mean_token_accuracy": 0.21464165896177292,
      "num_tokens": 104534112.0,
      "step": 53430
    },
    {
      "entropy": 5.947571325302124,
      "epoch": 5.341380516819113,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002541944668287052,
      "loss": 5.294,
      "mean_token_accuracy": 0.2093284696340561,
      "num_tokens": 104544062.0,
      "step": 53435
    },
    {
      "entropy": 5.870019483566284,
      "epoch": 5.3418803418803416,
      "grad_norm": 1.140625,
      "learning_rate": 0.000254158930896745,
      "loss": 5.2081,
      "mean_token_accuracy": 0.22216475754976273,
      "num_tokens": 104553951.0,
      "step": 53440
    },
    {
      "entropy": 6.013809108734131,
      "epoch": 5.34238016694157,
      "grad_norm": 1.09375,
      "learning_rate": 0.0002541233954891406,
      "loss": 5.4256,
      "mean_token_accuracy": 0.19941211342811585,
      "num_tokens": 104564778.0,
      "step": 53445
    },
    {
      "entropy": 5.892360496520996,
      "epoch": 5.342879992002799,
      "grad_norm": 1.109375,
      "learning_rate": 0.00025408786060678626,
      "loss": 5.2328,
      "mean_token_accuracy": 0.21453596353530885,
      "num_tokens": 104574797.0,
      "step": 53450
    },
    {
      "entropy": 5.92187647819519,
      "epoch": 5.343379817064028,
      "grad_norm": 1.28125,
      "learning_rate": 0.000254052326250576,
      "loss": 5.2436,
      "mean_token_accuracy": 0.22696960866451263,
      "num_tokens": 104584350.0,
      "step": 53455
    },
    {
      "entropy": 5.918367195129394,
      "epoch": 5.343879642125256,
      "grad_norm": 1.359375,
      "learning_rate": 0.00025401679242140386,
      "loss": 5.2699,
      "mean_token_accuracy": 0.2189572721719742,
      "num_tokens": 104594010.0,
      "step": 53460
    },
    {
      "entropy": 5.932449054718018,
      "epoch": 5.344379467186485,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025398125912016366,
      "loss": 5.2533,
      "mean_token_accuracy": 0.21630078107118605,
      "num_tokens": 104605000.0,
      "step": 53465
    },
    {
      "entropy": 5.752292823791504,
      "epoch": 5.344879292247713,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002539457263477497,
      "loss": 5.065,
      "mean_token_accuracy": 0.22889928370714188,
      "num_tokens": 104616349.0,
      "step": 53470
    },
    {
      "entropy": 5.903519773483277,
      "epoch": 5.3453791173089416,
      "grad_norm": 1.125,
      "learning_rate": 0.00025391019410505575,
      "loss": 5.2796,
      "mean_token_accuracy": 0.2159622311592102,
      "num_tokens": 104626701.0,
      "step": 53475
    },
    {
      "entropy": 5.91808819770813,
      "epoch": 5.34587894237017,
      "grad_norm": 1.140625,
      "learning_rate": 0.000253874662392976,
      "loss": 5.2923,
      "mean_token_accuracy": 0.21068740487098694,
      "num_tokens": 104636551.0,
      "step": 53480
    },
    {
      "entropy": 5.794997119903565,
      "epoch": 5.346378767431399,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025383913121240423,
      "loss": 5.2179,
      "mean_token_accuracy": 0.22100985199213027,
      "num_tokens": 104645071.0,
      "step": 53485
    },
    {
      "entropy": 5.945149660110474,
      "epoch": 5.346878592492628,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00025380360056423445,
      "loss": 5.2859,
      "mean_token_accuracy": 0.21749117523431777,
      "num_tokens": 104654691.0,
      "step": 53490
    },
    {
      "entropy": 5.8212120056152346,
      "epoch": 5.347378417553856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002537680704493607,
      "loss": 5.1265,
      "mean_token_accuracy": 0.23312256038188933,
      "num_tokens": 104664930.0,
      "step": 53495
    },
    {
      "entropy": 5.960046195983887,
      "epoch": 5.347878242615085,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00025373254086867683,
      "loss": 5.3136,
      "mean_token_accuracy": 0.2169810026884079,
      "num_tokens": 104675635.0,
      "step": 53500
    },
    {
      "entropy": 5.910912704467774,
      "epoch": 5.348378067676313,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002536970118230767,
      "loss": 5.2258,
      "mean_token_accuracy": 0.21450374275445938,
      "num_tokens": 104684468.0,
      "step": 53505
    },
    {
      "entropy": 6.0887839794158936,
      "epoch": 5.3488778927375416,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002536614833134543,
      "loss": 5.4638,
      "mean_token_accuracy": 0.19929826706647874,
      "num_tokens": 104694048.0,
      "step": 53510
    },
    {
      "entropy": 5.836046266555786,
      "epoch": 5.34937771779877,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002536259553407034,
      "loss": 5.2041,
      "mean_token_accuracy": 0.23320764899253846,
      "num_tokens": 104702882.0,
      "step": 53515
    },
    {
      "entropy": 5.878085327148438,
      "epoch": 5.349877542859999,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002535904279057181,
      "loss": 5.2195,
      "mean_token_accuracy": 0.21749678254127502,
      "num_tokens": 104713237.0,
      "step": 53520
    },
    {
      "entropy": 5.744829702377319,
      "epoch": 5.350377367921228,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002535549010093922,
      "loss": 5.0466,
      "mean_token_accuracy": 0.2438777729868889,
      "num_tokens": 104723183.0,
      "step": 53525
    },
    {
      "entropy": 5.809292793273926,
      "epoch": 5.350877192982456,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002535193746526194,
      "loss": 5.0769,
      "mean_token_accuracy": 0.24152970761060716,
      "num_tokens": 104733133.0,
      "step": 53530
    },
    {
      "entropy": 5.896665477752686,
      "epoch": 5.351377018043685,
      "grad_norm": 1.3125,
      "learning_rate": 0.00025348384883629345,
      "loss": 5.2365,
      "mean_token_accuracy": 0.22153977900743485,
      "num_tokens": 104742117.0,
      "step": 53535
    },
    {
      "entropy": 5.954399633407593,
      "epoch": 5.351876843104913,
      "grad_norm": 1.328125,
      "learning_rate": 0.0002534483235613085,
      "loss": 5.3242,
      "mean_token_accuracy": 0.2172090381383896,
      "num_tokens": 104752032.0,
      "step": 53540
    },
    {
      "entropy": 5.890709257125854,
      "epoch": 5.3523766681661415,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00025341279882855805,
      "loss": 5.2983,
      "mean_token_accuracy": 0.21827629655599595,
      "num_tokens": 104761664.0,
      "step": 53545
    },
    {
      "entropy": 5.953831768035888,
      "epoch": 5.35287649322737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002533772746389361,
      "loss": 5.3055,
      "mean_token_accuracy": 0.22228527665138245,
      "num_tokens": 104771172.0,
      "step": 53550
    },
    {
      "entropy": 6.027396297454834,
      "epoch": 5.353376318288599,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002533417509933364,
      "loss": 5.3856,
      "mean_token_accuracy": 0.21078705191612243,
      "num_tokens": 104780608.0,
      "step": 53555
    },
    {
      "entropy": 5.976304149627685,
      "epoch": 5.353876143349828,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025330622789265263,
      "loss": 5.2546,
      "mean_token_accuracy": 0.2160801351070404,
      "num_tokens": 104789488.0,
      "step": 53560
    },
    {
      "entropy": 5.923370790481568,
      "epoch": 5.354375968411056,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002532707053377786,
      "loss": 5.3725,
      "mean_token_accuracy": 0.2071722060441971,
      "num_tokens": 104799745.0,
      "step": 53565
    },
    {
      "entropy": 5.813714790344238,
      "epoch": 5.354875793472285,
      "grad_norm": 2.0625,
      "learning_rate": 0.00025323518332960793,
      "loss": 5.1034,
      "mean_token_accuracy": 0.22549257129430772,
      "num_tokens": 104809980.0,
      "step": 53570
    },
    {
      "entropy": 5.945400524139404,
      "epoch": 5.355375618533513,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002531996618690345,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21718083173036576,
      "num_tokens": 104819145.0,
      "step": 53575
    },
    {
      "entropy": 5.861328649520874,
      "epoch": 5.3558754435947415,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000253164140956952,
      "loss": 5.1855,
      "mean_token_accuracy": 0.22520305812358857,
      "num_tokens": 104827923.0,
      "step": 53580
    },
    {
      "entropy": 5.887405633926392,
      "epoch": 5.35637526865597,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025312862059425413,
      "loss": 5.2447,
      "mean_token_accuracy": 0.22333357036113738,
      "num_tokens": 104837695.0,
      "step": 53585
    },
    {
      "entropy": 5.855331039428711,
      "epoch": 5.356875093717199,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025309310078183445,
      "loss": 5.1755,
      "mean_token_accuracy": 0.2276241585612297,
      "num_tokens": 104847183.0,
      "step": 53590
    },
    {
      "entropy": 5.92544379234314,
      "epoch": 5.357374918778428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025305758152058684,
      "loss": 5.3208,
      "mean_token_accuracy": 0.21307066679000855,
      "num_tokens": 104857520.0,
      "step": 53595
    },
    {
      "entropy": 5.873132753372192,
      "epoch": 5.357874743839656,
      "grad_norm": 1.25,
      "learning_rate": 0.0002530220628114048,
      "loss": 5.1296,
      "mean_token_accuracy": 0.22310500144958495,
      "num_tokens": 104865900.0,
      "step": 53600
    },
    {
      "entropy": 5.726172304153442,
      "epoch": 5.358374568900885,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002529865446551819,
      "loss": 5.0871,
      "mean_token_accuracy": 0.22974826246500016,
      "num_tokens": 104875512.0,
      "step": 53605
    },
    {
      "entropy": 5.849279546737671,
      "epoch": 5.358874393962113,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000252951027052812,
      "loss": 5.1585,
      "mean_token_accuracy": 0.22730198502540588,
      "num_tokens": 104884067.0,
      "step": 53610
    },
    {
      "entropy": 5.897010707855225,
      "epoch": 5.3593742190233415,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025291551000518846,
      "loss": 5.2573,
      "mean_token_accuracy": 0.22219193130731582,
      "num_tokens": 104893691.0,
      "step": 53615
    },
    {
      "entropy": 5.910955381393433,
      "epoch": 5.35987404408457,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025287999351320504,
      "loss": 5.287,
      "mean_token_accuracy": 0.22517820745706557,
      "num_tokens": 104903705.0,
      "step": 53620
    },
    {
      "entropy": 5.865476369857788,
      "epoch": 5.360373869145799,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025284447757775534,
      "loss": 5.1698,
      "mean_token_accuracy": 0.22907553017139434,
      "num_tokens": 104914271.0,
      "step": 53625
    },
    {
      "entropy": 5.890599489212036,
      "epoch": 5.360873694207028,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0002528089621997329,
      "loss": 5.2322,
      "mean_token_accuracy": 0.22090981155633926,
      "num_tokens": 104924059.0,
      "step": 53630
    },
    {
      "entropy": 5.929154300689698,
      "epoch": 5.361373519268256,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025277344738003116,
      "loss": 5.2427,
      "mean_token_accuracy": 0.22253280133008957,
      "num_tokens": 104934999.0,
      "step": 53635
    },
    {
      "entropy": 5.882481241226197,
      "epoch": 5.361873344329485,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002527379331195439,
      "loss": 5.1611,
      "mean_token_accuracy": 0.2313899055123329,
      "num_tokens": 104944448.0,
      "step": 53640
    },
    {
      "entropy": 5.779135656356812,
      "epoch": 5.362373169390713,
      "grad_norm": 1.1875,
      "learning_rate": 0.0002527024194191644,
      "loss": 5.1627,
      "mean_token_accuracy": 0.23290696889162063,
      "num_tokens": 104954464.0,
      "step": 53645
    },
    {
      "entropy": 5.778068208694458,
      "epoch": 5.3628729944519415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025266690627978627,
      "loss": 5.0736,
      "mean_token_accuracy": 0.23351194709539413,
      "num_tokens": 104963950.0,
      "step": 53650
    },
    {
      "entropy": 5.910799312591553,
      "epoch": 5.36337281951317,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002526313937023031,
      "loss": 5.2054,
      "mean_token_accuracy": 0.2216020405292511,
      "num_tokens": 104973583.0,
      "step": 53655
    },
    {
      "entropy": 5.9413543224334715,
      "epoch": 5.363872644574399,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025259588168760833,
      "loss": 5.2837,
      "mean_token_accuracy": 0.22024635076522828,
      "num_tokens": 104982955.0,
      "step": 53660
    },
    {
      "entropy": 5.917276239395141,
      "epoch": 5.364372469635628,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0002525603702365954,
      "loss": 5.3252,
      "mean_token_accuracy": 0.21776444613933563,
      "num_tokens": 104994206.0,
      "step": 53665
    },
    {
      "entropy": 5.8678779125213625,
      "epoch": 5.364872294696856,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002525248593501578,
      "loss": 5.1919,
      "mean_token_accuracy": 0.22551386803388596,
      "num_tokens": 105004273.0,
      "step": 53670
    },
    {
      "entropy": 5.936918115615844,
      "epoch": 5.365372119758085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002524893490291889,
      "loss": 5.335,
      "mean_token_accuracy": 0.21438994109630585,
      "num_tokens": 105014477.0,
      "step": 53675
    },
    {
      "entropy": 5.878109550476074,
      "epoch": 5.365871944819313,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002524538392745822,
      "loss": 5.1704,
      "mean_token_accuracy": 0.22797974646091462,
      "num_tokens": 105024459.0,
      "step": 53680
    },
    {
      "entropy": 5.91923680305481,
      "epoch": 5.3663717698805415,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002524183300872311,
      "loss": 5.2518,
      "mean_token_accuracy": 0.21559584885835648,
      "num_tokens": 105034147.0,
      "step": 53685
    },
    {
      "entropy": 5.936181879043579,
      "epoch": 5.36687159494177,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00025238282146802907,
      "loss": 5.3781,
      "mean_token_accuracy": 0.2102620020508766,
      "num_tokens": 105044947.0,
      "step": 53690
    },
    {
      "entropy": 5.976980543136596,
      "epoch": 5.367371420002999,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00025234731341786934,
      "loss": 5.313,
      "mean_token_accuracy": 0.21073837876319884,
      "num_tokens": 105054791.0,
      "step": 53695
    },
    {
      "entropy": 5.945271825790405,
      "epoch": 5.367871245064228,
      "grad_norm": 1.09375,
      "learning_rate": 0.00025231180593764536,
      "loss": 5.2754,
      "mean_token_accuracy": 0.21637871414422988,
      "num_tokens": 105065325.0,
      "step": 53700
    },
    {
      "entropy": 5.91864709854126,
      "epoch": 5.368371070125456,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0002522762990282506,
      "loss": 5.2051,
      "mean_token_accuracy": 0.22743114531040193,
      "num_tokens": 105075859.0,
      "step": 53705
    },
    {
      "entropy": 5.898510837554932,
      "epoch": 5.368870895186685,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002522407926905782,
      "loss": 5.2211,
      "mean_token_accuracy": 0.22017985433340073,
      "num_tokens": 105084802.0,
      "step": 53710
    },
    {
      "entropy": 5.889670991897583,
      "epoch": 5.369370720247913,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002522052869255217,
      "loss": 5.1818,
      "mean_token_accuracy": 0.22040736228227614,
      "num_tokens": 105093386.0,
      "step": 53715
    },
    {
      "entropy": 5.906203031539917,
      "epoch": 5.3698705453091415,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025216978173397427,
      "loss": 5.2949,
      "mean_token_accuracy": 0.21607047468423843,
      "num_tokens": 105102008.0,
      "step": 53720
    },
    {
      "entropy": 5.904891014099121,
      "epoch": 5.37037037037037,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025213427711682926,
      "loss": 5.3409,
      "mean_token_accuracy": 0.2132373034954071,
      "num_tokens": 105112932.0,
      "step": 53725
    },
    {
      "entropy": 5.912059545516968,
      "epoch": 5.370870195431599,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0002520987730749799,
      "loss": 5.2784,
      "mean_token_accuracy": 0.22082493752241134,
      "num_tokens": 105123018.0,
      "step": 53730
    },
    {
      "entropy": 5.884818887710571,
      "epoch": 5.371370020492828,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00025206326960931964,
      "loss": 5.1724,
      "mean_token_accuracy": 0.2304197981953621,
      "num_tokens": 105132017.0,
      "step": 53735
    },
    {
      "entropy": 5.972125625610351,
      "epoch": 5.371869845554056,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002520277667207416,
      "loss": 5.2815,
      "mean_token_accuracy": 0.2161395087838173,
      "num_tokens": 105142267.0,
      "step": 53740
    },
    {
      "entropy": 5.904192066192627,
      "epoch": 5.372369670615285,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025199226441013895,
      "loss": 5.2375,
      "mean_token_accuracy": 0.21552546918392182,
      "num_tokens": 105151389.0,
      "step": 53745
    },
    {
      "entropy": 5.873010492324829,
      "epoch": 5.372869495676513,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025195676267840503,
      "loss": 5.2573,
      "mean_token_accuracy": 0.21468819826841354,
      "num_tokens": 105162112.0,
      "step": 53750
    },
    {
      "entropy": 5.921914720535279,
      "epoch": 5.3733693207377415,
      "grad_norm": 1.125,
      "learning_rate": 0.00025192126152643305,
      "loss": 5.2522,
      "mean_token_accuracy": 0.2193876564502716,
      "num_tokens": 105172592.0,
      "step": 53755
    },
    {
      "entropy": 5.912960052490234,
      "epoch": 5.37386914579897,
      "grad_norm": 1.15625,
      "learning_rate": 0.00025188576095511617,
      "loss": 5.3062,
      "mean_token_accuracy": 0.21223266422748566,
      "num_tokens": 105181821.0,
      "step": 53760
    },
    {
      "entropy": 5.942865085601807,
      "epoch": 5.374368970860199,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025185026096534755,
      "loss": 5.2515,
      "mean_token_accuracy": 0.21766164004802704,
      "num_tokens": 105191198.0,
      "step": 53765
    },
    {
      "entropy": 5.819816398620605,
      "epoch": 5.374868795921428,
      "grad_norm": 1.125,
      "learning_rate": 0.00025181476155802053,
      "loss": 5.1934,
      "mean_token_accuracy": 0.22071620523929597,
      "num_tokens": 105201357.0,
      "step": 53770
    },
    {
      "entropy": 5.855884742736817,
      "epoch": 5.375368620982656,
      "grad_norm": 1.0625,
      "learning_rate": 0.00025177926273402805,
      "loss": 5.1586,
      "mean_token_accuracy": 0.23135473132133483,
      "num_tokens": 105211340.0,
      "step": 53775
    },
    {
      "entropy": 5.921875715255737,
      "epoch": 5.375868446043885,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0002517437644942634,
      "loss": 5.2977,
      "mean_token_accuracy": 0.21306149959564208,
      "num_tokens": 105220409.0,
      "step": 53780
    },
    {
      "entropy": 5.806761074066162,
      "epoch": 5.376368271105113,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00025170826683961964,
      "loss": 5.1309,
      "mean_token_accuracy": 0.22262706756591796,
      "num_tokens": 105231159.0,
      "step": 53785
    },
    {
      "entropy": 5.89838171005249,
      "epoch": 5.3768680961663415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025167276977098983,
      "loss": 5.2078,
      "mean_token_accuracy": 0.2231139361858368,
      "num_tokens": 105242530.0,
      "step": 53790
    },
    {
      "entropy": 5.8868245601654055,
      "epoch": 5.37736792122757,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00025163727328926717,
      "loss": 5.2609,
      "mean_token_accuracy": 0.22417206466197967,
      "num_tokens": 105252032.0,
      "step": 53795
    },
    {
      "entropy": 5.924555826187134,
      "epoch": 5.377867746288799,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002516017773953447,
      "loss": 5.2656,
      "mean_token_accuracy": 0.21314172446727753,
      "num_tokens": 105261690.0,
      "step": 53800
    },
    {
      "entropy": 5.968781566619873,
      "epoch": 5.378367571350028,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00025156628209011564,
      "loss": 5.3065,
      "mean_token_accuracy": 0.21451641470193863,
      "num_tokens": 105270533.0,
      "step": 53805
    },
    {
      "entropy": 5.942224740982056,
      "epoch": 5.378867396411256,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002515307873744727,
      "loss": 5.2459,
      "mean_token_accuracy": 0.21822102814912797,
      "num_tokens": 105280033.0,
      "step": 53810
    },
    {
      "entropy": 5.896073865890503,
      "epoch": 5.379367221472485,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002514952932493092,
      "loss": 5.199,
      "mean_token_accuracy": 0.21362369656562805,
      "num_tokens": 105289944.0,
      "step": 53815
    },
    {
      "entropy": 5.888131856918335,
      "epoch": 5.379867046533713,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025145979971551803,
      "loss": 5.2426,
      "mean_token_accuracy": 0.21305591911077498,
      "num_tokens": 105300017.0,
      "step": 53820
    },
    {
      "entropy": 5.914373922348022,
      "epoch": 5.3803668715949415,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002514243067739922,
      "loss": 5.2237,
      "mean_token_accuracy": 0.21757371425628663,
      "num_tokens": 105309915.0,
      "step": 53825
    },
    {
      "entropy": 5.895397520065307,
      "epoch": 5.38086669665617,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002513888144256248,
      "loss": 5.3292,
      "mean_token_accuracy": 0.20856766402721405,
      "num_tokens": 105320573.0,
      "step": 53830
    },
    {
      "entropy": 5.942885971069336,
      "epoch": 5.381366521717399,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002513533226713088,
      "loss": 5.249,
      "mean_token_accuracy": 0.21928179115056992,
      "num_tokens": 105329564.0,
      "step": 53835
    },
    {
      "entropy": 5.914640045166015,
      "epoch": 5.381866346778628,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025131783151193703,
      "loss": 5.219,
      "mean_token_accuracy": 0.21973295360803605,
      "num_tokens": 105339327.0,
      "step": 53840
    },
    {
      "entropy": 5.873067808151245,
      "epoch": 5.382366171839856,
      "grad_norm": 1.125,
      "learning_rate": 0.00025128234094840256,
      "loss": 5.1525,
      "mean_token_accuracy": 0.2261306270956993,
      "num_tokens": 105348881.0,
      "step": 53845
    },
    {
      "entropy": 5.899156188964843,
      "epoch": 5.382865996901085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002512468509815983,
      "loss": 5.237,
      "mean_token_accuracy": 0.21194611340761185,
      "num_tokens": 105358374.0,
      "step": 53850
    },
    {
      "entropy": 5.895746469497681,
      "epoch": 5.383365821962313,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00025121136161241714,
      "loss": 5.2737,
      "mean_token_accuracy": 0.2182368144392967,
      "num_tokens": 105368394.0,
      "step": 53855
    },
    {
      "entropy": 5.8714686870574955,
      "epoch": 5.3838656470235415,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00025117587284175197,
      "loss": 5.2654,
      "mean_token_accuracy": 0.22321318536996843,
      "num_tokens": 105377335.0,
      "step": 53860
    },
    {
      "entropy": 5.892246675491333,
      "epoch": 5.38436547208477,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0002511403846704957,
      "loss": 5.2371,
      "mean_token_accuracy": 0.2205905109643936,
      "num_tokens": 105387083.0,
      "step": 53865
    },
    {
      "entropy": 5.927938652038574,
      "epoch": 5.384865297145999,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025110489709954134,
      "loss": 5.2574,
      "mean_token_accuracy": 0.2239128828048706,
      "num_tokens": 105396266.0,
      "step": 53870
    },
    {
      "entropy": 5.953767728805542,
      "epoch": 5.385365122207228,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025106941012978157,
      "loss": 5.3005,
      "mean_token_accuracy": 0.22293566018342972,
      "num_tokens": 105406351.0,
      "step": 53875
    },
    {
      "entropy": 5.97831506729126,
      "epoch": 5.385864947268456,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002510339237621092,
      "loss": 5.335,
      "mean_token_accuracy": 0.21036955416202546,
      "num_tokens": 105416137.0,
      "step": 53880
    },
    {
      "entropy": 5.933020877838135,
      "epoch": 5.386364772329685,
      "grad_norm": 1.203125,
      "learning_rate": 0.00025099843799741725,
      "loss": 5.249,
      "mean_token_accuracy": 0.21414522975683212,
      "num_tokens": 105426654.0,
      "step": 53885
    },
    {
      "entropy": 5.90992169380188,
      "epoch": 5.386864597390913,
      "grad_norm": 1.25,
      "learning_rate": 0.0002509629528365984,
      "loss": 5.1841,
      "mean_token_accuracy": 0.22528807073831558,
      "num_tokens": 105435243.0,
      "step": 53890
    },
    {
      "entropy": 5.866309976577758,
      "epoch": 5.3873644224521415,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00025092746828054544,
      "loss": 5.1998,
      "mean_token_accuracy": 0.22392271012067794,
      "num_tokens": 105445832.0,
      "step": 53895
    },
    {
      "entropy": 5.875956916809082,
      "epoch": 5.38786424751337,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025089198433015113,
      "loss": 5.1998,
      "mean_token_accuracy": 0.22171074897050858,
      "num_tokens": 105455223.0,
      "step": 53900
    },
    {
      "entropy": 5.9063152313232425,
      "epoch": 5.388364072574599,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00025085650098630836,
      "loss": 5.2192,
      "mean_token_accuracy": 0.21901085823774338,
      "num_tokens": 105464558.0,
      "step": 53905
    },
    {
      "entropy": 5.811249494552612,
      "epoch": 5.388863897635828,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00025082101824990975,
      "loss": 5.2256,
      "mean_token_accuracy": 0.22062042057514192,
      "num_tokens": 105474203.0,
      "step": 53910
    },
    {
      "entropy": 5.944071960449219,
      "epoch": 5.389363722697056,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002507855361218482,
      "loss": 5.3456,
      "mean_token_accuracy": 0.20986157208681105,
      "num_tokens": 105484249.0,
      "step": 53915
    },
    {
      "entropy": 6.00152473449707,
      "epoch": 5.389863547758285,
      "grad_norm": 1.296875,
      "learning_rate": 0.0002507500546030162,
      "loss": 5.3694,
      "mean_token_accuracy": 0.21127646863460542,
      "num_tokens": 105493699.0,
      "step": 53920
    },
    {
      "entropy": 5.9528828144073485,
      "epoch": 5.390363372819513,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025071457369430666,
      "loss": 5.3517,
      "mean_token_accuracy": 0.21083521842956543,
      "num_tokens": 105505636.0,
      "step": 53925
    },
    {
      "entropy": 5.884893608093262,
      "epoch": 5.3908631978807415,
      "grad_norm": 1.21875,
      "learning_rate": 0.00025067909339661224,
      "loss": 5.1762,
      "mean_token_accuracy": 0.222226682305336,
      "num_tokens": 105515036.0,
      "step": 53930
    },
    {
      "entropy": 5.9247715950012205,
      "epoch": 5.39136302294197,
      "grad_norm": 1.234375,
      "learning_rate": 0.0002506436137108255,
      "loss": 5.2768,
      "mean_token_accuracy": 0.21222528517246247,
      "num_tokens": 105525324.0,
      "step": 53935
    },
    {
      "entropy": 5.86968584060669,
      "epoch": 5.391862848003199,
      "grad_norm": 1.3125,
      "learning_rate": 0.0002506081346378392,
      "loss": 5.1588,
      "mean_token_accuracy": 0.22971939593553542,
      "num_tokens": 105534378.0,
      "step": 53940
    },
    {
      "entropy": 5.872314357757569,
      "epoch": 5.392362673064428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025057265617854603,
      "loss": 5.1303,
      "mean_token_accuracy": 0.22636563032865525,
      "num_tokens": 105543255.0,
      "step": 53945
    },
    {
      "entropy": 5.904870414733887,
      "epoch": 5.392862498125656,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002505371783338385,
      "loss": 5.2339,
      "mean_token_accuracy": 0.2242514505982399,
      "num_tokens": 105551976.0,
      "step": 53950
    },
    {
      "entropy": 5.903673124313355,
      "epoch": 5.393362323186885,
      "grad_norm": 1.203125,
      "learning_rate": 0.0002505017011046092,
      "loss": 5.2791,
      "mean_token_accuracy": 0.21174365729093553,
      "num_tokens": 105561964.0,
      "step": 53955
    },
    {
      "entropy": 5.910647344589234,
      "epoch": 5.393862148248113,
      "grad_norm": 1.171875,
      "learning_rate": 0.00025046622449175084,
      "loss": 5.2108,
      "mean_token_accuracy": 0.22499549835920335,
      "num_tokens": 105572212.0,
      "step": 53960
    },
    {
      "entropy": 5.897049140930176,
      "epoch": 5.3943619733093415,
      "grad_norm": 1.296875,
      "learning_rate": 0.00025043074849615595,
      "loss": 5.192,
      "mean_token_accuracy": 0.2255934238433838,
      "num_tokens": 105581772.0,
      "step": 53965
    },
    {
      "entropy": 5.931192827224732,
      "epoch": 5.39486179837057,
      "grad_norm": 1.140625,
      "learning_rate": 0.00025039527311871714,
      "loss": 5.3232,
      "mean_token_accuracy": 0.21268365979194642,
      "num_tokens": 105591716.0,
      "step": 53970
    },
    {
      "entropy": 5.941907167434692,
      "epoch": 5.395361623431799,
      "grad_norm": 1.234375,
      "learning_rate": 0.00025035979836032687,
      "loss": 5.3089,
      "mean_token_accuracy": 0.21215150505304337,
      "num_tokens": 105601653.0,
      "step": 53975
    },
    {
      "entropy": 5.980674743652344,
      "epoch": 5.395861448493028,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002503243242218777,
      "loss": 5.3157,
      "mean_token_accuracy": 0.21651553958654404,
      "num_tokens": 105610441.0,
      "step": 53980
    },
    {
      "entropy": 5.964120388031006,
      "epoch": 5.396361273554256,
      "grad_norm": 1.25,
      "learning_rate": 0.0002502888507042623,
      "loss": 5.2681,
      "mean_token_accuracy": 0.21796846240758896,
      "num_tokens": 105619398.0,
      "step": 53985
    },
    {
      "entropy": 5.868376541137695,
      "epoch": 5.396861098615485,
      "grad_norm": 1.21875,
      "learning_rate": 0.000250253377808373,
      "loss": 5.2409,
      "mean_token_accuracy": 0.21601803302764894,
      "num_tokens": 105628355.0,
      "step": 53990
    },
    {
      "entropy": 5.822594928741455,
      "epoch": 5.397360923676713,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002502179055351023,
      "loss": 5.1243,
      "mean_token_accuracy": 0.23134371489286423,
      "num_tokens": 105638442.0,
      "step": 53995
    },
    {
      "entropy": 5.87192668914795,
      "epoch": 5.3978607487379415,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0002501824338853428,
      "loss": 5.2343,
      "mean_token_accuracy": 0.22363251745700835,
      "num_tokens": 105646742.0,
      "step": 54000
    },
    {
      "epoch": 5.3978607487379415,
      "eval_entropy": 5.700997262417548,
      "eval_loss": 5.426027297973633,
      "eval_mean_token_accuracy": 0.21806414766322088,
      "eval_num_tokens": 105646742.0,
      "eval_runtime": 24.1701,
      "eval_samples_per_second": 1284.439,
      "eval_steps_per_second": 160.57,
      "step": 54000
    }
  ],
  "logging_steps": 5,
  "max_steps": 100030,
  "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": 1.67703129704448e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}