{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0991652921477484,
  "eval_steps": 3000,
  "global_step": 21000,
  "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
    }
  ],
  "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": 6.51033593856e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}