tur-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed10/checkpoint-33000/trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.2986954565901936,
  "eval_steps": 3000,
  "global_step": 33000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 8.173395156860352,
      "epoch": 0.00049982506122857,
      "grad_norm": 1.2265625,
      "learning_rate": 2e-06,
      "loss": 7.9212,
      "mean_token_accuracy": 0.09272176176309585,
      "num_tokens": 8821.0,
      "step": 5
    },
    {
      "entropy": 8.15390853881836,
      "epoch": 0.00099965012245714,
      "grad_norm": 1.1953125,
      "learning_rate": 4.5e-06,
      "loss": 7.8813,
      "mean_token_accuracy": 0.0939578540623188,
      "num_tokens": 18288.0,
      "step": 10
    },
    {
      "entropy": 8.147719764709473,
      "epoch": 0.00149947518368571,
      "grad_norm": 0.96875,
      "learning_rate": 7e-06,
      "loss": 8.0013,
      "mean_token_accuracy": 0.08617180287837982,
      "num_tokens": 28664.0,
      "step": 15
    },
    {
      "entropy": 8.167346668243407,
      "epoch": 0.00199930024491428,
      "grad_norm": 2.0,
      "learning_rate": 9.5e-06,
      "loss": 7.9255,
      "mean_token_accuracy": 0.0916962169110775,
      "num_tokens": 37807.0,
      "step": 20
    },
    {
      "entropy": 8.143739604949952,
      "epoch": 0.00249912530614285,
      "grad_norm": 0.9765625,
      "learning_rate": 1.2e-05,
      "loss": 8.0114,
      "mean_token_accuracy": 0.08655632585287094,
      "num_tokens": 48086.0,
      "step": 25
    },
    {
      "entropy": 8.162391948699952,
      "epoch": 0.00299895036737142,
      "grad_norm": 1.125,
      "learning_rate": 1.4500000000000002e-05,
      "loss": 7.9097,
      "mean_token_accuracy": 0.09338272213935853,
      "num_tokens": 57427.0,
      "step": 30
    },
    {
      "entropy": 8.124217319488526,
      "epoch": 0.00349877542859999,
      "grad_norm": 1.078125,
      "learning_rate": 1.7000000000000003e-05,
      "loss": 7.8375,
      "mean_token_accuracy": 0.09512318149209023,
      "num_tokens": 66662.0,
      "step": 35
    },
    {
      "entropy": 8.138414859771729,
      "epoch": 0.00399860048982856,
      "grad_norm": 0.94140625,
      "learning_rate": 1.95e-05,
      "loss": 7.9693,
      "mean_token_accuracy": 0.08915233314037323,
      "num_tokens": 77388.0,
      "step": 40
    },
    {
      "entropy": 8.167016983032227,
      "epoch": 0.00449842555105713,
      "grad_norm": 1.1875,
      "learning_rate": 2.2e-05,
      "loss": 7.8205,
      "mean_token_accuracy": 0.09171208068728447,
      "num_tokens": 86093.0,
      "step": 45
    },
    {
      "entropy": 8.184170150756836,
      "epoch": 0.0049982506122857,
      "grad_norm": 1.078125,
      "learning_rate": 2.4500000000000003e-05,
      "loss": 7.9262,
      "mean_token_accuracy": 0.09001774936914445,
      "num_tokens": 96097.0,
      "step": 50
    },
    {
      "entropy": 8.148734855651856,
      "epoch": 0.00549807567351427,
      "grad_norm": 0.953125,
      "learning_rate": 2.7e-05,
      "loss": 7.9451,
      "mean_token_accuracy": 0.08841896653175355,
      "num_tokens": 106328.0,
      "step": 55
    },
    {
      "entropy": 8.164736080169678,
      "epoch": 0.00599790073474284,
      "grad_norm": 1.09375,
      "learning_rate": 2.95e-05,
      "loss": 7.8896,
      "mean_token_accuracy": 0.09127758592367172,
      "num_tokens": 114822.0,
      "step": 60
    },
    {
      "entropy": 8.159956741333009,
      "epoch": 0.00649772579597141,
      "grad_norm": 1.1640625,
      "learning_rate": 3.2e-05,
      "loss": 7.9015,
      "mean_token_accuracy": 0.0870350182056427,
      "num_tokens": 124984.0,
      "step": 65
    },
    {
      "entropy": 8.099590635299682,
      "epoch": 0.00699755085719998,
      "grad_norm": 1.0625,
      "learning_rate": 3.4500000000000005e-05,
      "loss": 7.8877,
      "mean_token_accuracy": 0.09039942994713783,
      "num_tokens": 134054.0,
      "step": 70
    },
    {
      "entropy": 8.094889259338379,
      "epoch": 0.00749737591842855,
      "grad_norm": 1.078125,
      "learning_rate": 3.7e-05,
      "loss": 7.8934,
      "mean_token_accuracy": 0.09580017626285553,
      "num_tokens": 144561.0,
      "step": 75
    },
    {
      "entropy": 8.143476486206055,
      "epoch": 0.00799720097965712,
      "grad_norm": 1.0703125,
      "learning_rate": 3.95e-05,
      "loss": 7.9329,
      "mean_token_accuracy": 0.08533147275447846,
      "num_tokens": 153701.0,
      "step": 80
    },
    {
      "entropy": 8.166958999633788,
      "epoch": 0.00849702604088569,
      "grad_norm": 1.15625,
      "learning_rate": 4.2000000000000004e-05,
      "loss": 7.8542,
      "mean_token_accuracy": 0.09379108399152755,
      "num_tokens": 164291.0,
      "step": 85
    },
    {
      "entropy": 8.122808074951172,
      "epoch": 0.00899685110211426,
      "grad_norm": 1.1328125,
      "learning_rate": 4.45e-05,
      "loss": 7.86,
      "mean_token_accuracy": 0.09373755380511284,
      "num_tokens": 174136.0,
      "step": 90
    },
    {
      "entropy": 8.091786527633667,
      "epoch": 0.00949667616334283,
      "grad_norm": 0.98046875,
      "learning_rate": 4.7000000000000004e-05,
      "loss": 7.9022,
      "mean_token_accuracy": 0.09065977856516838,
      "num_tokens": 184368.0,
      "step": 95
    },
    {
      "entropy": 8.127646446228027,
      "epoch": 0.0099965012245714,
      "grad_norm": 0.99609375,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 7.9161,
      "mean_token_accuracy": 0.08597206845879554,
      "num_tokens": 194659.0,
      "step": 100
    },
    {
      "entropy": 8.213620471954346,
      "epoch": 0.01049632628579997,
      "grad_norm": 0.96875,
      "learning_rate": 5.2e-05,
      "loss": 7.9287,
      "mean_token_accuracy": 0.09312979727983475,
      "num_tokens": 204875.0,
      "step": 105
    },
    {
      "entropy": 8.149574470520019,
      "epoch": 0.01099615134702854,
      "grad_norm": 1.46875,
      "learning_rate": 5.45e-05,
      "loss": 7.9428,
      "mean_token_accuracy": 0.09381906017661094,
      "num_tokens": 217033.0,
      "step": 110
    },
    {
      "entropy": 8.211454486846923,
      "epoch": 0.01149597640825711,
      "grad_norm": 1.1015625,
      "learning_rate": 5.7e-05,
      "loss": 7.8896,
      "mean_token_accuracy": 0.09600678682327271,
      "num_tokens": 225946.0,
      "step": 115
    },
    {
      "entropy": 8.188517951965332,
      "epoch": 0.01199580146948568,
      "grad_norm": 1.0703125,
      "learning_rate": 5.9499999999999996e-05,
      "loss": 7.871,
      "mean_token_accuracy": 0.09085513278841972,
      "num_tokens": 235413.0,
      "step": 120
    },
    {
      "entropy": 8.094170618057252,
      "epoch": 0.01249562653071425,
      "grad_norm": 0.90625,
      "learning_rate": 6.2e-05,
      "loss": 7.9383,
      "mean_token_accuracy": 0.08237705007195473,
      "num_tokens": 247913.0,
      "step": 125
    },
    {
      "entropy": 8.203950881958008,
      "epoch": 0.01299545159194282,
      "grad_norm": 1.09375,
      "learning_rate": 6.450000000000001e-05,
      "loss": 7.9114,
      "mean_token_accuracy": 0.09513814747333527,
      "num_tokens": 258486.0,
      "step": 130
    },
    {
      "entropy": 8.173905754089356,
      "epoch": 0.01349527665317139,
      "grad_norm": 1.0078125,
      "learning_rate": 6.7e-05,
      "loss": 7.8574,
      "mean_token_accuracy": 0.08711725249886512,
      "num_tokens": 268027.0,
      "step": 135
    },
    {
      "entropy": 8.097312259674073,
      "epoch": 0.01399510171439996,
      "grad_norm": 1.1015625,
      "learning_rate": 6.950000000000001e-05,
      "loss": 7.8699,
      "mean_token_accuracy": 0.09045619666576385,
      "num_tokens": 277470.0,
      "step": 140
    },
    {
      "entropy": 8.162187767028808,
      "epoch": 0.01449492677562853,
      "grad_norm": 1.03125,
      "learning_rate": 7.2e-05,
      "loss": 7.8441,
      "mean_token_accuracy": 0.08916656076908111,
      "num_tokens": 286891.0,
      "step": 145
    },
    {
      "entropy": 8.138840484619141,
      "epoch": 0.0149947518368571,
      "grad_norm": 1.1875,
      "learning_rate": 7.45e-05,
      "loss": 7.8221,
      "mean_token_accuracy": 0.09832631349563599,
      "num_tokens": 295636.0,
      "step": 150
    },
    {
      "entropy": 8.100650215148926,
      "epoch": 0.01549457689808567,
      "grad_norm": 0.98828125,
      "learning_rate": 7.7e-05,
      "loss": 7.807,
      "mean_token_accuracy": 0.09535080567002296,
      "num_tokens": 306542.0,
      "step": 155
    },
    {
      "entropy": 8.136592578887939,
      "epoch": 0.01599440195931424,
      "grad_norm": 1.0546875,
      "learning_rate": 7.950000000000001e-05,
      "loss": 7.8128,
      "mean_token_accuracy": 0.09563795924186706,
      "num_tokens": 315956.0,
      "step": 160
    },
    {
      "entropy": 8.075897836685181,
      "epoch": 0.01649422702054281,
      "grad_norm": 1.03125,
      "learning_rate": 8.2e-05,
      "loss": 7.7835,
      "mean_token_accuracy": 0.09556139633059502,
      "num_tokens": 325500.0,
      "step": 165
    },
    {
      "entropy": 8.13330774307251,
      "epoch": 0.01699405208177138,
      "grad_norm": 1.140625,
      "learning_rate": 8.450000000000001e-05,
      "loss": 7.8728,
      "mean_token_accuracy": 0.09840985015034676,
      "num_tokens": 335194.0,
      "step": 170
    },
    {
      "entropy": 8.178883743286132,
      "epoch": 0.01749387714299995,
      "grad_norm": 1.1328125,
      "learning_rate": 8.7e-05,
      "loss": 7.8081,
      "mean_token_accuracy": 0.09958252236247063,
      "num_tokens": 344717.0,
      "step": 175
    },
    {
      "entropy": 8.01774287223816,
      "epoch": 0.01799370220422852,
      "grad_norm": 1.0703125,
      "learning_rate": 8.95e-05,
      "loss": 7.7695,
      "mean_token_accuracy": 0.09593843147158623,
      "num_tokens": 354476.0,
      "step": 180
    },
    {
      "entropy": 8.1070050239563,
      "epoch": 0.01849352726545709,
      "grad_norm": 1.21875,
      "learning_rate": 9.2e-05,
      "loss": 7.837,
      "mean_token_accuracy": 0.09266547411680222,
      "num_tokens": 363939.0,
      "step": 185
    },
    {
      "entropy": 8.165003967285156,
      "epoch": 0.01899335232668566,
      "grad_norm": 1.2109375,
      "learning_rate": 9.45e-05,
      "loss": 7.8697,
      "mean_token_accuracy": 0.08987938240170479,
      "num_tokens": 373187.0,
      "step": 190
    },
    {
      "entropy": 8.064292907714844,
      "epoch": 0.01949317738791423,
      "grad_norm": 1.0078125,
      "learning_rate": 9.7e-05,
      "loss": 7.8155,
      "mean_token_accuracy": 0.09654008895158768,
      "num_tokens": 382999.0,
      "step": 195
    },
    {
      "entropy": 8.19270725250244,
      "epoch": 0.0199930024491428,
      "grad_norm": 1.0546875,
      "learning_rate": 9.95e-05,
      "loss": 7.8707,
      "mean_token_accuracy": 0.09406451731920243,
      "num_tokens": 392639.0,
      "step": 200
    },
    {
      "entropy": 8.113720321655274,
      "epoch": 0.02049282751037137,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000102,
      "loss": 7.8688,
      "mean_token_accuracy": 0.0874548889696598,
      "num_tokens": 403089.0,
      "step": 205
    },
    {
      "entropy": 8.125567817687989,
      "epoch": 0.02099265257159994,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00010449999999999999,
      "loss": 7.8475,
      "mean_token_accuracy": 0.08455977067351342,
      "num_tokens": 412321.0,
      "step": 210
    },
    {
      "entropy": 8.179554748535157,
      "epoch": 0.02149247763282851,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000107,
      "loss": 7.8661,
      "mean_token_accuracy": 0.08831506222486496,
      "num_tokens": 421599.0,
      "step": 215
    },
    {
      "entropy": 8.149456882476807,
      "epoch": 0.02199230269405708,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001095,
      "loss": 7.8491,
      "mean_token_accuracy": 0.08751258924603462,
      "num_tokens": 432012.0,
      "step": 220
    },
    {
      "entropy": 8.103495025634766,
      "epoch": 0.02249212775528565,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000112,
      "loss": 7.8785,
      "mean_token_accuracy": 0.08636855706572533,
      "num_tokens": 441989.0,
      "step": 225
    },
    {
      "entropy": 8.124602317810059,
      "epoch": 0.02299195281651422,
      "grad_norm": 1.078125,
      "learning_rate": 0.0001145,
      "loss": 7.8288,
      "mean_token_accuracy": 0.09801522344350815,
      "num_tokens": 451532.0,
      "step": 230
    },
    {
      "entropy": 8.118602657318116,
      "epoch": 0.02349177787774279,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00011700000000000001,
      "loss": 7.8538,
      "mean_token_accuracy": 0.09045086875557899,
      "num_tokens": 461815.0,
      "step": 235
    },
    {
      "entropy": 8.224643325805664,
      "epoch": 0.02399160293897136,
      "grad_norm": 1.328125,
      "learning_rate": 0.00011949999999999999,
      "loss": 7.9141,
      "mean_token_accuracy": 0.08622012361884117,
      "num_tokens": 470860.0,
      "step": 240
    },
    {
      "entropy": 8.085096645355225,
      "epoch": 0.02449142800019993,
      "grad_norm": 1.109375,
      "learning_rate": 0.000122,
      "loss": 7.8658,
      "mean_token_accuracy": 0.08970579206943512,
      "num_tokens": 480649.0,
      "step": 245
    },
    {
      "entropy": 8.196398639678955,
      "epoch": 0.0249912530614285,
      "grad_norm": 1.328125,
      "learning_rate": 0.0001245,
      "loss": 7.791,
      "mean_token_accuracy": 0.09182884991168976,
      "num_tokens": 489969.0,
      "step": 250
    },
    {
      "entropy": 7.999597597122192,
      "epoch": 0.02549107812265707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000127,
      "loss": 7.8234,
      "mean_token_accuracy": 0.09344998225569726,
      "num_tokens": 500000.0,
      "step": 255
    },
    {
      "entropy": 8.161861276626587,
      "epoch": 0.02599090318388564,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001295,
      "loss": 7.7993,
      "mean_token_accuracy": 0.09596751630306244,
      "num_tokens": 510013.0,
      "step": 260
    },
    {
      "entropy": 8.045274543762208,
      "epoch": 0.026490728245114212,
      "grad_norm": 1.15625,
      "learning_rate": 0.000132,
      "loss": 7.8612,
      "mean_token_accuracy": 0.09452076926827431,
      "num_tokens": 519698.0,
      "step": 265
    },
    {
      "entropy": 8.16001009941101,
      "epoch": 0.02699055330634278,
      "grad_norm": 1.09375,
      "learning_rate": 0.00013450000000000002,
      "loss": 7.8053,
      "mean_token_accuracy": 0.09544829428195953,
      "num_tokens": 531631.0,
      "step": 270
    },
    {
      "entropy": 8.094933128356933,
      "epoch": 0.02749037836757135,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00013700000000000002,
      "loss": 7.8554,
      "mean_token_accuracy": 0.09051231518387795,
      "num_tokens": 542204.0,
      "step": 275
    },
    {
      "entropy": 8.072505760192872,
      "epoch": 0.02799020342879992,
      "grad_norm": 1.125,
      "learning_rate": 0.0001395,
      "loss": 7.7632,
      "mean_token_accuracy": 0.09450774788856506,
      "num_tokens": 552269.0,
      "step": 280
    },
    {
      "entropy": 8.006140613555909,
      "epoch": 0.02849002849002849,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00014199999999999998,
      "loss": 7.7554,
      "mean_token_accuracy": 0.09378736913204193,
      "num_tokens": 561907.0,
      "step": 285
    },
    {
      "entropy": 8.243315029144288,
      "epoch": 0.02898985355125706,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0001445,
      "loss": 7.8846,
      "mean_token_accuracy": 0.08584100976586342,
      "num_tokens": 571283.0,
      "step": 290
    },
    {
      "entropy": 7.974696350097656,
      "epoch": 0.02948967861248563,
      "grad_norm": 1.203125,
      "learning_rate": 0.000147,
      "loss": 7.7631,
      "mean_token_accuracy": 0.09530414417386054,
      "num_tokens": 579896.0,
      "step": 295
    },
    {
      "entropy": 8.0429096698761,
      "epoch": 0.0299895036737142,
      "grad_norm": 1.21875,
      "learning_rate": 0.0001495,
      "loss": 7.7369,
      "mean_token_accuracy": 0.09685113728046417,
      "num_tokens": 588856.0,
      "step": 300
    },
    {
      "entropy": 8.04590039253235,
      "epoch": 0.03048932873494277,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000152,
      "loss": 7.7846,
      "mean_token_accuracy": 0.09521858915686607,
      "num_tokens": 598910.0,
      "step": 305
    },
    {
      "entropy": 8.120123100280761,
      "epoch": 0.03098915379617134,
      "grad_norm": 1.28125,
      "learning_rate": 0.00015450000000000001,
      "loss": 7.8292,
      "mean_token_accuracy": 0.09306757375597954,
      "num_tokens": 609248.0,
      "step": 310
    },
    {
      "entropy": 8.05753664970398,
      "epoch": 0.03148897885739991,
      "grad_norm": 1.125,
      "learning_rate": 0.000157,
      "loss": 7.8521,
      "mean_token_accuracy": 0.09199029803276063,
      "num_tokens": 619352.0,
      "step": 315
    },
    {
      "entropy": 8.209227561950684,
      "epoch": 0.03198880391862848,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0001595,
      "loss": 7.824,
      "mean_token_accuracy": 0.08493875190615655,
      "num_tokens": 628300.0,
      "step": 320
    },
    {
      "entropy": 8.080143022537232,
      "epoch": 0.03248862897985705,
      "grad_norm": 1.0625,
      "learning_rate": 0.000162,
      "loss": 7.803,
      "mean_token_accuracy": 0.09538680538535119,
      "num_tokens": 637969.0,
      "step": 325
    },
    {
      "entropy": 8.04603624343872,
      "epoch": 0.03298845404108562,
      "grad_norm": 1.015625,
      "learning_rate": 0.00016450000000000001,
      "loss": 7.7472,
      "mean_token_accuracy": 0.09739784002304078,
      "num_tokens": 647518.0,
      "step": 330
    },
    {
      "entropy": 8.218221759796142,
      "epoch": 0.03348827910231419,
      "grad_norm": 2.40625,
      "learning_rate": 0.00016700000000000002,
      "loss": 7.8654,
      "mean_token_accuracy": 0.08986226171255111,
      "num_tokens": 658292.0,
      "step": 335
    },
    {
      "entropy": 8.023103809356689,
      "epoch": 0.03398810416354276,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00016950000000000003,
      "loss": 7.7836,
      "mean_token_accuracy": 0.09298105165362358,
      "num_tokens": 667872.0,
      "step": 340
    },
    {
      "entropy": 8.046844100952148,
      "epoch": 0.03448792922477133,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00017199999999999998,
      "loss": 7.6883,
      "mean_token_accuracy": 0.0971065603196621,
      "num_tokens": 676976.0,
      "step": 345
    },
    {
      "entropy": 8.024046325683594,
      "epoch": 0.0349877542859999,
      "grad_norm": 1.75,
      "learning_rate": 0.00017449999999999999,
      "loss": 7.7503,
      "mean_token_accuracy": 0.09279201328754424,
      "num_tokens": 686149.0,
      "step": 350
    },
    {
      "entropy": 8.001934909820557,
      "epoch": 0.03548757934722847,
      "grad_norm": 1.25,
      "learning_rate": 0.000177,
      "loss": 7.6529,
      "mean_token_accuracy": 0.10482841432094574,
      "num_tokens": 696439.0,
      "step": 355
    },
    {
      "entropy": 7.9410679817199705,
      "epoch": 0.03598740440845704,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0001795,
      "loss": 7.6498,
      "mean_token_accuracy": 0.10238936245441436,
      "num_tokens": 705902.0,
      "step": 360
    },
    {
      "entropy": 8.022974014282227,
      "epoch": 0.03648722946968561,
      "grad_norm": 1.1875,
      "learning_rate": 0.000182,
      "loss": 7.8524,
      "mean_token_accuracy": 0.09500225633382797,
      "num_tokens": 715973.0,
      "step": 365
    },
    {
      "entropy": 8.071463918685913,
      "epoch": 0.03698705453091418,
      "grad_norm": 1.171875,
      "learning_rate": 0.0001845,
      "loss": 7.7332,
      "mean_token_accuracy": 0.09426685646176339,
      "num_tokens": 725463.0,
      "step": 370
    },
    {
      "entropy": 8.053776121139526,
      "epoch": 0.03748687959214275,
      "grad_norm": 1.296875,
      "learning_rate": 0.000187,
      "loss": 7.7399,
      "mean_token_accuracy": 0.09462474286556244,
      "num_tokens": 734941.0,
      "step": 375
    },
    {
      "entropy": 8.029221868515014,
      "epoch": 0.03798670465337132,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0001895,
      "loss": 7.777,
      "mean_token_accuracy": 0.09015404880046844,
      "num_tokens": 744302.0,
      "step": 380
    },
    {
      "entropy": 7.99616174697876,
      "epoch": 0.03848652971459989,
      "grad_norm": 1.203125,
      "learning_rate": 0.000192,
      "loss": 7.7275,
      "mean_token_accuracy": 0.10049449354410171,
      "num_tokens": 754341.0,
      "step": 385
    },
    {
      "entropy": 8.145702171325684,
      "epoch": 0.03898635477582846,
      "grad_norm": 1.359375,
      "learning_rate": 0.0001945,
      "loss": 7.852,
      "mean_token_accuracy": 0.08332133702933789,
      "num_tokens": 764087.0,
      "step": 390
    },
    {
      "entropy": 7.944943904876709,
      "epoch": 0.03948617983705703,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00019700000000000002,
      "loss": 7.7887,
      "mean_token_accuracy": 0.08848774060606956,
      "num_tokens": 773728.0,
      "step": 395
    },
    {
      "entropy": 8.168435096740723,
      "epoch": 0.0399860048982856,
      "grad_norm": 1.03125,
      "learning_rate": 0.00019950000000000002,
      "loss": 7.822,
      "mean_token_accuracy": 0.08284367993474007,
      "num_tokens": 784839.0,
      "step": 400
    },
    {
      "entropy": 7.93064603805542,
      "epoch": 0.04048582995951417,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000202,
      "loss": 7.7929,
      "mean_token_accuracy": 0.09478651136159896,
      "num_tokens": 794514.0,
      "step": 405
    },
    {
      "entropy": 8.11144437789917,
      "epoch": 0.04098565502074274,
      "grad_norm": 1.109375,
      "learning_rate": 0.00020449999999999998,
      "loss": 7.7581,
      "mean_token_accuracy": 0.09314420819282532,
      "num_tokens": 805254.0,
      "step": 410
    },
    {
      "entropy": 7.990531015396118,
      "epoch": 0.04148548008197131,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000207,
      "loss": 7.6794,
      "mean_token_accuracy": 0.10085502117872239,
      "num_tokens": 814135.0,
      "step": 415
    },
    {
      "entropy": 7.9582349300384525,
      "epoch": 0.04198530514319988,
      "grad_norm": 1.15625,
      "learning_rate": 0.0002095,
      "loss": 7.7453,
      "mean_token_accuracy": 0.0957567848265171,
      "num_tokens": 823531.0,
      "step": 420
    },
    {
      "entropy": 7.954767322540283,
      "epoch": 0.04248513020442845,
      "grad_norm": 1.234375,
      "learning_rate": 0.000212,
      "loss": 7.6464,
      "mean_token_accuracy": 0.10113119706511497,
      "num_tokens": 832833.0,
      "step": 425
    },
    {
      "entropy": 7.997408628463745,
      "epoch": 0.04298495526565702,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0002145,
      "loss": 7.6441,
      "mean_token_accuracy": 0.09947716444730759,
      "num_tokens": 841541.0,
      "step": 430
    },
    {
      "entropy": 7.836303043365478,
      "epoch": 0.04348478032688559,
      "grad_norm": 1.25,
      "learning_rate": 0.00021700000000000002,
      "loss": 7.6644,
      "mean_token_accuracy": 0.09943394884467124,
      "num_tokens": 850689.0,
      "step": 435
    },
    {
      "entropy": 7.932573461532593,
      "epoch": 0.04398460538811416,
      "grad_norm": 1.03125,
      "learning_rate": 0.0002195,
      "loss": 7.624,
      "mean_token_accuracy": 0.10144704058766366,
      "num_tokens": 860573.0,
      "step": 440
    },
    {
      "entropy": 8.005117082595826,
      "epoch": 0.04448443044934273,
      "grad_norm": 1.25,
      "learning_rate": 0.000222,
      "loss": 7.7274,
      "mean_token_accuracy": 0.090135994926095,
      "num_tokens": 869784.0,
      "step": 445
    },
    {
      "entropy": 8.012764692306519,
      "epoch": 0.0449842555105713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0002245,
      "loss": 7.6871,
      "mean_token_accuracy": 0.09482940882444382,
      "num_tokens": 878716.0,
      "step": 450
    },
    {
      "entropy": 7.914750099182129,
      "epoch": 0.04548408057179987,
      "grad_norm": 1.171875,
      "learning_rate": 0.00022700000000000002,
      "loss": 7.7939,
      "mean_token_accuracy": 0.09040690958499908,
      "num_tokens": 888334.0,
      "step": 455
    },
    {
      "entropy": 8.02715539932251,
      "epoch": 0.04598390563302844,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00022950000000000002,
      "loss": 7.684,
      "mean_token_accuracy": 0.09270390123128891,
      "num_tokens": 897701.0,
      "step": 460
    },
    {
      "entropy": 7.970236349105835,
      "epoch": 0.04648373069425701,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00023200000000000003,
      "loss": 7.6925,
      "mean_token_accuracy": 0.09443332329392433,
      "num_tokens": 907278.0,
      "step": 465
    },
    {
      "entropy": 7.925041341781617,
      "epoch": 0.04698355575548558,
      "grad_norm": 1.1875,
      "learning_rate": 0.00023449999999999998,
      "loss": 7.7006,
      "mean_token_accuracy": 0.09271714985370635,
      "num_tokens": 916194.0,
      "step": 470
    },
    {
      "entropy": 7.967321968078613,
      "epoch": 0.04748338081671415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000237,
      "loss": 7.7231,
      "mean_token_accuracy": 0.09261031746864319,
      "num_tokens": 926743.0,
      "step": 475
    },
    {
      "entropy": 8.015508699417115,
      "epoch": 0.04798320587794272,
      "grad_norm": 1.171875,
      "learning_rate": 0.0002395,
      "loss": 7.7504,
      "mean_token_accuracy": 0.08958175107836723,
      "num_tokens": 936059.0,
      "step": 480
    },
    {
      "entropy": 7.923208951950073,
      "epoch": 0.04848303093917129,
      "grad_norm": 1.265625,
      "learning_rate": 0.000242,
      "loss": 7.6733,
      "mean_token_accuracy": 0.09590794965624809,
      "num_tokens": 945990.0,
      "step": 485
    },
    {
      "entropy": 8.079939603805542,
      "epoch": 0.04898285600039986,
      "grad_norm": 1.375,
      "learning_rate": 0.0002445,
      "loss": 7.7223,
      "mean_token_accuracy": 0.10266662836074829,
      "num_tokens": 956626.0,
      "step": 490
    },
    {
      "entropy": 7.87365756034851,
      "epoch": 0.04948268106162843,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000247,
      "loss": 7.7187,
      "mean_token_accuracy": 0.09226388484239578,
      "num_tokens": 966200.0,
      "step": 495
    },
    {
      "entropy": 7.9425849437713625,
      "epoch": 0.049982506122857,
      "grad_norm": 1.140625,
      "learning_rate": 0.0002495,
      "loss": 7.662,
      "mean_token_accuracy": 0.09558572769165039,
      "num_tokens": 975651.0,
      "step": 500
    },
    {
      "entropy": 7.913667392730713,
      "epoch": 0.05048233118408557,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000252,
      "loss": 7.6696,
      "mean_token_accuracy": 0.0961525209248066,
      "num_tokens": 985264.0,
      "step": 505
    },
    {
      "entropy": 7.940064716339111,
      "epoch": 0.05098215624531414,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002545,
      "loss": 7.65,
      "mean_token_accuracy": 0.09761169999837875,
      "num_tokens": 996723.0,
      "step": 510
    },
    {
      "entropy": 7.919295358657837,
      "epoch": 0.05148198130654271,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000257,
      "loss": 7.6735,
      "mean_token_accuracy": 0.09729609489440919,
      "num_tokens": 1006098.0,
      "step": 515
    },
    {
      "entropy": 8.019027423858642,
      "epoch": 0.05198180636777128,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002595,
      "loss": 7.719,
      "mean_token_accuracy": 0.09270480126142502,
      "num_tokens": 1016595.0,
      "step": 520
    },
    {
      "entropy": 7.7847301959991455,
      "epoch": 0.05248163142899985,
      "grad_norm": 1.265625,
      "learning_rate": 0.000262,
      "loss": 7.5765,
      "mean_token_accuracy": 0.10053420215845107,
      "num_tokens": 1026281.0,
      "step": 525
    },
    {
      "entropy": 7.8970424175262455,
      "epoch": 0.052981456490228424,
      "grad_norm": 0.98828125,
      "learning_rate": 0.00026450000000000003,
      "loss": 7.6903,
      "mean_token_accuracy": 0.10703469291329384,
      "num_tokens": 1036697.0,
      "step": 530
    },
    {
      "entropy": 7.789024686813354,
      "epoch": 0.05348128155145699,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00026700000000000004,
      "loss": 7.6547,
      "mean_token_accuracy": 0.09684248864650727,
      "num_tokens": 1047433.0,
      "step": 535
    },
    {
      "entropy": 7.969110536575317,
      "epoch": 0.05398110661268556,
      "grad_norm": 1.109375,
      "learning_rate": 0.00026950000000000005,
      "loss": 7.6915,
      "mean_token_accuracy": 0.09020176380872727,
      "num_tokens": 1056774.0,
      "step": 540
    },
    {
      "entropy": 7.943284511566162,
      "epoch": 0.05448093167391413,
      "grad_norm": 1.109375,
      "learning_rate": 0.00027200000000000005,
      "loss": 7.5707,
      "mean_token_accuracy": 0.10046748295426369,
      "num_tokens": 1068056.0,
      "step": 545
    },
    {
      "entropy": 7.795782899856567,
      "epoch": 0.0549807567351427,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0002745,
      "loss": 7.6036,
      "mean_token_accuracy": 0.10044094324111938,
      "num_tokens": 1077698.0,
      "step": 550
    },
    {
      "entropy": 7.8286316871643065,
      "epoch": 0.05548058179637127,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000277,
      "loss": 7.5719,
      "mean_token_accuracy": 0.09958620220422745,
      "num_tokens": 1086485.0,
      "step": 555
    },
    {
      "entropy": 7.919136619567871,
      "epoch": 0.05598040685759984,
      "grad_norm": 1.21875,
      "learning_rate": 0.0002795,
      "loss": 7.6477,
      "mean_token_accuracy": 0.0894574172794819,
      "num_tokens": 1097329.0,
      "step": 560
    },
    {
      "entropy": 7.742609977722168,
      "epoch": 0.05648023191882841,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00028199999999999997,
      "loss": 7.6188,
      "mean_token_accuracy": 0.09743130803108216,
      "num_tokens": 1107990.0,
      "step": 565
    },
    {
      "entropy": 7.8571289539337155,
      "epoch": 0.05698005698005698,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0002845,
      "loss": 7.5419,
      "mean_token_accuracy": 0.10700362995266914,
      "num_tokens": 1117798.0,
      "step": 570
    },
    {
      "entropy": 7.760644626617432,
      "epoch": 0.05747988204128555,
      "grad_norm": 1.109375,
      "learning_rate": 0.000287,
      "loss": 7.6946,
      "mean_token_accuracy": 0.09315006881952285,
      "num_tokens": 1126976.0,
      "step": 575
    },
    {
      "entropy": 7.9075384616851805,
      "epoch": 0.05797970710251412,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002895,
      "loss": 7.6043,
      "mean_token_accuracy": 0.10048359259963036,
      "num_tokens": 1136410.0,
      "step": 580
    },
    {
      "entropy": 7.852177858352661,
      "epoch": 0.05847953216374269,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000292,
      "loss": 7.6589,
      "mean_token_accuracy": 0.09688285812735557,
      "num_tokens": 1147162.0,
      "step": 585
    },
    {
      "entropy": 7.778061771392823,
      "epoch": 0.05897935722497126,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0002945,
      "loss": 7.5948,
      "mean_token_accuracy": 0.09637040942907334,
      "num_tokens": 1157578.0,
      "step": 590
    },
    {
      "entropy": 7.852608489990234,
      "epoch": 0.05947918228619983,
      "grad_norm": 1.265625,
      "learning_rate": 0.000297,
      "loss": 7.5039,
      "mean_token_accuracy": 0.10554338097572327,
      "num_tokens": 1166914.0,
      "step": 595
    },
    {
      "entropy": 7.852068567276001,
      "epoch": 0.0599790073474284,
      "grad_norm": 1.265625,
      "learning_rate": 0.0002995,
      "loss": 7.6895,
      "mean_token_accuracy": 0.08630285412073135,
      "num_tokens": 1176740.0,
      "step": 600
    },
    {
      "entropy": 7.816021823883057,
      "epoch": 0.06047883240865697,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000302,
      "loss": 7.5753,
      "mean_token_accuracy": 0.09594688341021537,
      "num_tokens": 1185497.0,
      "step": 605
    },
    {
      "entropy": 7.933215665817261,
      "epoch": 0.06097865746988554,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003045,
      "loss": 7.6755,
      "mean_token_accuracy": 0.09170238226652146,
      "num_tokens": 1197390.0,
      "step": 610
    },
    {
      "entropy": 7.677834701538086,
      "epoch": 0.06147848253111411,
      "grad_norm": 1.0625,
      "learning_rate": 0.000307,
      "loss": 7.5827,
      "mean_token_accuracy": 0.09688961803913117,
      "num_tokens": 1208293.0,
      "step": 615
    },
    {
      "entropy": 7.839175367355347,
      "epoch": 0.06197830759234268,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003095,
      "loss": 7.5679,
      "mean_token_accuracy": 0.09913268983364106,
      "num_tokens": 1217860.0,
      "step": 620
    },
    {
      "entropy": 7.758148860931397,
      "epoch": 0.06247813265357125,
      "grad_norm": 1.25,
      "learning_rate": 0.000312,
      "loss": 7.5767,
      "mean_token_accuracy": 0.09510255008935928,
      "num_tokens": 1227259.0,
      "step": 625
    },
    {
      "entropy": 7.715111303329468,
      "epoch": 0.06297795771479982,
      "grad_norm": 1.359375,
      "learning_rate": 0.0003145,
      "loss": 7.466,
      "mean_token_accuracy": 0.10256180614233017,
      "num_tokens": 1236775.0,
      "step": 630
    },
    {
      "entropy": 7.763536643981934,
      "epoch": 0.06347778277602839,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000317,
      "loss": 7.6092,
      "mean_token_accuracy": 0.09816056564450264,
      "num_tokens": 1246307.0,
      "step": 635
    },
    {
      "entropy": 7.900314426422119,
      "epoch": 0.06397760783725696,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0003195,
      "loss": 7.5545,
      "mean_token_accuracy": 0.09914906919002534,
      "num_tokens": 1255197.0,
      "step": 640
    },
    {
      "entropy": 7.679573202133179,
      "epoch": 0.06447743289848552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000322,
      "loss": 7.5638,
      "mean_token_accuracy": 0.10264531224966049,
      "num_tokens": 1265056.0,
      "step": 645
    },
    {
      "entropy": 7.778749275207519,
      "epoch": 0.0649772579597141,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00032450000000000003,
      "loss": 7.5018,
      "mean_token_accuracy": 0.09979739040136337,
      "num_tokens": 1274718.0,
      "step": 650
    },
    {
      "entropy": 7.823064184188842,
      "epoch": 0.06547708302094267,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00032700000000000003,
      "loss": 7.6216,
      "mean_token_accuracy": 0.09702884927392005,
      "num_tokens": 1284364.0,
      "step": 655
    },
    {
      "entropy": 7.74606237411499,
      "epoch": 0.06597690808217124,
      "grad_norm": 1.265625,
      "learning_rate": 0.00032950000000000004,
      "loss": 7.5995,
      "mean_token_accuracy": 0.09949352443218232,
      "num_tokens": 1294186.0,
      "step": 660
    },
    {
      "entropy": 7.743727207183838,
      "epoch": 0.06647673314339982,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00033200000000000005,
      "loss": 7.5094,
      "mean_token_accuracy": 0.10350405499339103,
      "num_tokens": 1303393.0,
      "step": 665
    },
    {
      "entropy": 7.670324659347534,
      "epoch": 0.06697655820462838,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00033450000000000005,
      "loss": 7.5125,
      "mean_token_accuracy": 0.09411343447864055,
      "num_tokens": 1312819.0,
      "step": 670
    },
    {
      "entropy": 7.773294258117676,
      "epoch": 0.06747638326585695,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000337,
      "loss": 7.5869,
      "mean_token_accuracy": 0.09254579842090607,
      "num_tokens": 1323001.0,
      "step": 675
    },
    {
      "entropy": 7.692234659194947,
      "epoch": 0.06797620832708552,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003395,
      "loss": 7.5356,
      "mean_token_accuracy": 0.09674872979521751,
      "num_tokens": 1333260.0,
      "step": 680
    },
    {
      "entropy": 7.714693260192871,
      "epoch": 0.0684760333883141,
      "grad_norm": 1.171875,
      "learning_rate": 0.000342,
      "loss": 7.5253,
      "mean_token_accuracy": 0.1035267636179924,
      "num_tokens": 1342696.0,
      "step": 685
    },
    {
      "entropy": 7.71041955947876,
      "epoch": 0.06897585844954265,
      "grad_norm": 1.171875,
      "learning_rate": 0.00034449999999999997,
      "loss": 7.5005,
      "mean_token_accuracy": 0.1020401656627655,
      "num_tokens": 1353104.0,
      "step": 690
    },
    {
      "entropy": 7.7471325397491455,
      "epoch": 0.06947568351077123,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000347,
      "loss": 7.5799,
      "mean_token_accuracy": 0.09462966471910476,
      "num_tokens": 1362871.0,
      "step": 695
    },
    {
      "entropy": 7.774437189102173,
      "epoch": 0.0699755085719998,
      "grad_norm": 1.046875,
      "learning_rate": 0.0003495,
      "loss": 7.5494,
      "mean_token_accuracy": 0.09824139177799225,
      "num_tokens": 1373622.0,
      "step": 700
    },
    {
      "entropy": 7.5870044231414795,
      "epoch": 0.07047533363322837,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000352,
      "loss": 7.4243,
      "mean_token_accuracy": 0.10344729945063591,
      "num_tokens": 1382707.0,
      "step": 705
    },
    {
      "entropy": 7.7422362804412845,
      "epoch": 0.07097515869445695,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003545,
      "loss": 7.5181,
      "mean_token_accuracy": 0.09872173815965653,
      "num_tokens": 1392048.0,
      "step": 710
    },
    {
      "entropy": 7.511621618270874,
      "epoch": 0.0714749837556855,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000357,
      "loss": 7.3933,
      "mean_token_accuracy": 0.10354112610220909,
      "num_tokens": 1401387.0,
      "step": 715
    },
    {
      "entropy": 7.69247579574585,
      "epoch": 0.07197480881691408,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0003595,
      "loss": 7.4989,
      "mean_token_accuracy": 0.09920497015118598,
      "num_tokens": 1411404.0,
      "step": 720
    },
    {
      "entropy": 7.644815969467163,
      "epoch": 0.07247463387814265,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000362,
      "loss": 7.5194,
      "mean_token_accuracy": 0.09978774040937424,
      "num_tokens": 1420292.0,
      "step": 725
    },
    {
      "entropy": 7.674730587005615,
      "epoch": 0.07297445893937123,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003645,
      "loss": 7.4475,
      "mean_token_accuracy": 0.10385879799723625,
      "num_tokens": 1429990.0,
      "step": 730
    },
    {
      "entropy": 7.65957236289978,
      "epoch": 0.07347428400059978,
      "grad_norm": 1.234375,
      "learning_rate": 0.000367,
      "loss": 7.4864,
      "mean_token_accuracy": 0.09656506404280663,
      "num_tokens": 1440148.0,
      "step": 735
    },
    {
      "entropy": 7.612310075759888,
      "epoch": 0.07397410906182836,
      "grad_norm": 1.375,
      "learning_rate": 0.0003695,
      "loss": 7.4143,
      "mean_token_accuracy": 0.09919942021369935,
      "num_tokens": 1450015.0,
      "step": 740
    },
    {
      "entropy": 7.595290803909302,
      "epoch": 0.07447393412305693,
      "grad_norm": 1.1875,
      "learning_rate": 0.000372,
      "loss": 7.4663,
      "mean_token_accuracy": 0.09626267701387406,
      "num_tokens": 1459582.0,
      "step": 745
    },
    {
      "entropy": 7.677187204360962,
      "epoch": 0.0749737591842855,
      "grad_norm": 1.125,
      "learning_rate": 0.0003745,
      "loss": 7.5235,
      "mean_token_accuracy": 0.09861956685781478,
      "num_tokens": 1468151.0,
      "step": 750
    },
    {
      "entropy": 7.5536364078521725,
      "epoch": 0.07547358424551406,
      "grad_norm": 1.0625,
      "learning_rate": 0.000377,
      "loss": 7.4131,
      "mean_token_accuracy": 0.10350600630044937,
      "num_tokens": 1479099.0,
      "step": 755
    },
    {
      "entropy": 7.644258260726929,
      "epoch": 0.07597340930674264,
      "grad_norm": 1.125,
      "learning_rate": 0.0003795,
      "loss": 7.5166,
      "mean_token_accuracy": 0.10102088227868081,
      "num_tokens": 1488060.0,
      "step": 760
    },
    {
      "entropy": 7.6076555252075195,
      "epoch": 0.07647323436797121,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000382,
      "loss": 7.5237,
      "mean_token_accuracy": 0.09237042292952538,
      "num_tokens": 1498230.0,
      "step": 765
    },
    {
      "entropy": 7.625078392028809,
      "epoch": 0.07697305942919978,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003845,
      "loss": 7.4769,
      "mean_token_accuracy": 0.10016548410058021,
      "num_tokens": 1507962.0,
      "step": 770
    },
    {
      "entropy": 7.708804130554199,
      "epoch": 0.07747288449042836,
      "grad_norm": 1.1875,
      "learning_rate": 0.00038700000000000003,
      "loss": 7.5617,
      "mean_token_accuracy": 0.09648449942469597,
      "num_tokens": 1518607.0,
      "step": 775
    },
    {
      "entropy": 7.697318172454834,
      "epoch": 0.07797270955165692,
      "grad_norm": 1.203125,
      "learning_rate": 0.00038950000000000003,
      "loss": 7.4201,
      "mean_token_accuracy": 0.09753957316279412,
      "num_tokens": 1527166.0,
      "step": 780
    },
    {
      "entropy": 7.599863815307617,
      "epoch": 0.07847253461288549,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039200000000000004,
      "loss": 7.5271,
      "mean_token_accuracy": 0.09614471271634102,
      "num_tokens": 1536683.0,
      "step": 785
    },
    {
      "entropy": 7.574948406219482,
      "epoch": 0.07897235967411406,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039450000000000005,
      "loss": 7.4201,
      "mean_token_accuracy": 0.09909558668732643,
      "num_tokens": 1546243.0,
      "step": 790
    },
    {
      "entropy": 7.641019535064697,
      "epoch": 0.07947218473534264,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039700000000000005,
      "loss": 7.435,
      "mean_token_accuracy": 0.09994245618581772,
      "num_tokens": 1555916.0,
      "step": 795
    },
    {
      "entropy": 7.438870811462403,
      "epoch": 0.0799720097965712,
      "grad_norm": 1.015625,
      "learning_rate": 0.0003995,
      "loss": 7.369,
      "mean_token_accuracy": 0.10524533912539483,
      "num_tokens": 1566617.0,
      "step": 800
    },
    {
      "entropy": 7.572072315216064,
      "epoch": 0.08047183485779977,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000402,
      "loss": 7.4247,
      "mean_token_accuracy": 0.09702792316675186,
      "num_tokens": 1575413.0,
      "step": 805
    },
    {
      "entropy": 7.568275737762451,
      "epoch": 0.08097165991902834,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004045,
      "loss": 7.4234,
      "mean_token_accuracy": 0.09980287402868271,
      "num_tokens": 1586863.0,
      "step": 810
    },
    {
      "entropy": 7.546803331375122,
      "epoch": 0.08147148498025691,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040699999999999997,
      "loss": 7.5189,
      "mean_token_accuracy": 0.10001372396945954,
      "num_tokens": 1597685.0,
      "step": 815
    },
    {
      "entropy": 7.595981025695801,
      "epoch": 0.08197131004148547,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095,
      "loss": 7.3956,
      "mean_token_accuracy": 0.10019611120223999,
      "num_tokens": 1607103.0,
      "step": 820
    },
    {
      "entropy": 7.573238372802734,
      "epoch": 0.08247113510271405,
      "grad_norm": 1.078125,
      "learning_rate": 0.000412,
      "loss": 7.4421,
      "mean_token_accuracy": 0.10041169673204423,
      "num_tokens": 1617919.0,
      "step": 825
    },
    {
      "entropy": 7.571632623672485,
      "epoch": 0.08297096016394262,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004145,
      "loss": 7.4844,
      "mean_token_accuracy": 0.09938613995909691,
      "num_tokens": 1627293.0,
      "step": 830
    },
    {
      "entropy": 7.595695161819458,
      "epoch": 0.08347078522517119,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000417,
      "loss": 7.5219,
      "mean_token_accuracy": 0.0978666827082634,
      "num_tokens": 1638440.0,
      "step": 835
    },
    {
      "entropy": 7.5352317810058596,
      "epoch": 0.08397061028639977,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004195,
      "loss": 7.4249,
      "mean_token_accuracy": 0.10109311267733574,
      "num_tokens": 1648148.0,
      "step": 840
    },
    {
      "entropy": 7.512708568572998,
      "epoch": 0.08447043534762833,
      "grad_norm": 1.21875,
      "learning_rate": 0.000422,
      "loss": 7.4451,
      "mean_token_accuracy": 0.10042148157954216,
      "num_tokens": 1658334.0,
      "step": 845
    },
    {
      "entropy": 7.617625761032104,
      "epoch": 0.0849702604088569,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004245,
      "loss": 7.4846,
      "mean_token_accuracy": 0.09633303582668304,
      "num_tokens": 1667963.0,
      "step": 850
    },
    {
      "entropy": 7.4715556621551515,
      "epoch": 0.08547008547008547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000427,
      "loss": 7.3979,
      "mean_token_accuracy": 0.10262207016348839,
      "num_tokens": 1677270.0,
      "step": 855
    },
    {
      "entropy": 7.523194360733032,
      "epoch": 0.08596991053131404,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004295,
      "loss": 7.3859,
      "mean_token_accuracy": 0.10481420755386353,
      "num_tokens": 1686770.0,
      "step": 860
    },
    {
      "entropy": 7.529279613494873,
      "epoch": 0.0864697355925426,
      "grad_norm": 1.03125,
      "learning_rate": 0.000432,
      "loss": 7.4609,
      "mean_token_accuracy": 0.09573153108358383,
      "num_tokens": 1697507.0,
      "step": 865
    },
    {
      "entropy": 7.387008571624756,
      "epoch": 0.08696956065377118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004345,
      "loss": 7.3945,
      "mean_token_accuracy": 0.10062696561217307,
      "num_tokens": 1708253.0,
      "step": 870
    },
    {
      "entropy": 7.612355899810791,
      "epoch": 0.08746938571499975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000437,
      "loss": 7.4601,
      "mean_token_accuracy": 0.09850295558571816,
      "num_tokens": 1717827.0,
      "step": 875
    },
    {
      "entropy": 7.336903047561646,
      "epoch": 0.08796921077622832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004395,
      "loss": 7.2649,
      "mean_token_accuracy": 0.11100565418601036,
      "num_tokens": 1726678.0,
      "step": 880
    },
    {
      "entropy": 7.556534910202027,
      "epoch": 0.0884690358374569,
      "grad_norm": 1.015625,
      "learning_rate": 0.000442,
      "loss": 7.3976,
      "mean_token_accuracy": 0.09947276636958122,
      "num_tokens": 1737518.0,
      "step": 885
    },
    {
      "entropy": 7.438637018203735,
      "epoch": 0.08896886089868546,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004445,
      "loss": 7.2915,
      "mean_token_accuracy": 0.10892865434288979,
      "num_tokens": 1746338.0,
      "step": 890
    },
    {
      "entropy": 7.425498580932617,
      "epoch": 0.08946868595991403,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000447,
      "loss": 7.3736,
      "mean_token_accuracy": 0.10575593486428261,
      "num_tokens": 1756538.0,
      "step": 895
    },
    {
      "entropy": 7.557357740402222,
      "epoch": 0.0899685110211426,
      "grad_norm": 1.03125,
      "learning_rate": 0.00044950000000000003,
      "loss": 7.4052,
      "mean_token_accuracy": 0.10165804401040077,
      "num_tokens": 1766722.0,
      "step": 900
    },
    {
      "entropy": 7.337777996063233,
      "epoch": 0.09046833608237118,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045200000000000004,
      "loss": 7.3223,
      "mean_token_accuracy": 0.10339211374521255,
      "num_tokens": 1777039.0,
      "step": 905
    },
    {
      "entropy": 7.548654174804687,
      "epoch": 0.09096816114359974,
      "grad_norm": 0.91796875,
      "learning_rate": 0.00045450000000000004,
      "loss": 7.4393,
      "mean_token_accuracy": 0.09877310320734978,
      "num_tokens": 1787621.0,
      "step": 910
    },
    {
      "entropy": 7.429612350463867,
      "epoch": 0.09146798620482831,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045700000000000005,
      "loss": 7.4313,
      "mean_token_accuracy": 0.10118876844644546,
      "num_tokens": 1798345.0,
      "step": 915
    },
    {
      "entropy": 7.510519886016846,
      "epoch": 0.09196781126605688,
      "grad_norm": 1.0625,
      "learning_rate": 0.00045950000000000006,
      "loss": 7.4015,
      "mean_token_accuracy": 0.10110852941870689,
      "num_tokens": 1807994.0,
      "step": 920
    },
    {
      "entropy": 7.367496538162231,
      "epoch": 0.09246763632728545,
      "grad_norm": 1.140625,
      "learning_rate": 0.000462,
      "loss": 7.3139,
      "mean_token_accuracy": 0.09367315769195557,
      "num_tokens": 1817965.0,
      "step": 925
    },
    {
      "entropy": 7.464560461044312,
      "epoch": 0.09296746138851401,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004645,
      "loss": 7.3635,
      "mean_token_accuracy": 0.10010309666395187,
      "num_tokens": 1827094.0,
      "step": 930
    },
    {
      "entropy": 7.467557430267334,
      "epoch": 0.09346728644974259,
      "grad_norm": 1.125,
      "learning_rate": 0.000467,
      "loss": 7.3466,
      "mean_token_accuracy": 0.10039986148476601,
      "num_tokens": 1837754.0,
      "step": 935
    },
    {
      "entropy": 7.335368013381958,
      "epoch": 0.09396711151097116,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695,
      "loss": 7.3533,
      "mean_token_accuracy": 0.10200253203511238,
      "num_tokens": 1847500.0,
      "step": 940
    },
    {
      "entropy": 7.444225072860718,
      "epoch": 0.09446693657219973,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000472,
      "loss": 7.4125,
      "mean_token_accuracy": 0.0964927189052105,
      "num_tokens": 1857663.0,
      "step": 945
    },
    {
      "entropy": 7.509044599533081,
      "epoch": 0.0949667616334283,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004745,
      "loss": 7.4629,
      "mean_token_accuracy": 0.09966767057776452,
      "num_tokens": 1867752.0,
      "step": 950
    },
    {
      "entropy": 7.410632658004761,
      "epoch": 0.09546658669465687,
      "grad_norm": 1.09375,
      "learning_rate": 0.000477,
      "loss": 7.3962,
      "mean_token_accuracy": 0.09652933329343796,
      "num_tokens": 1877479.0,
      "step": 955
    },
    {
      "entropy": 7.43900876045227,
      "epoch": 0.09596641175588544,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004795,
      "loss": 7.4226,
      "mean_token_accuracy": 0.09707022532820701,
      "num_tokens": 1887944.0,
      "step": 960
    },
    {
      "entropy": 7.46921329498291,
      "epoch": 0.09646623681711401,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000482,
      "loss": 7.3202,
      "mean_token_accuracy": 0.1051582269370556,
      "num_tokens": 1896991.0,
      "step": 965
    },
    {
      "entropy": 7.35594482421875,
      "epoch": 0.09696606187834259,
      "grad_norm": 1.921875,
      "learning_rate": 0.0004845,
      "loss": 7.3518,
      "mean_token_accuracy": 0.10959891974925995,
      "num_tokens": 1907216.0,
      "step": 970
    },
    {
      "entropy": 7.475865411758423,
      "epoch": 0.09746588693957114,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000487,
      "loss": 7.3681,
      "mean_token_accuracy": 0.09844867512583733,
      "num_tokens": 1917206.0,
      "step": 975
    },
    {
      "entropy": 7.342016315460205,
      "epoch": 0.09796571200079972,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004895,
      "loss": 7.4607,
      "mean_token_accuracy": 0.09241715744137764,
      "num_tokens": 1929669.0,
      "step": 980
    },
    {
      "entropy": 7.476624441146851,
      "epoch": 0.09846553706202829,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000492,
      "loss": 7.3029,
      "mean_token_accuracy": 0.10470175147056579,
      "num_tokens": 1939881.0,
      "step": 985
    },
    {
      "entropy": 7.328264808654785,
      "epoch": 0.09896536212325686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004945,
      "loss": 7.346,
      "mean_token_accuracy": 0.10308508202433586,
      "num_tokens": 1949423.0,
      "step": 990
    },
    {
      "entropy": 7.44049825668335,
      "epoch": 0.09946518718448542,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000497,
      "loss": 7.3721,
      "mean_token_accuracy": 0.10048017129302025,
      "num_tokens": 1959214.0,
      "step": 995
    },
    {
      "entropy": 7.437930011749268,
      "epoch": 0.099965012245714,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004995,
      "loss": 7.3243,
      "mean_token_accuracy": 0.09923652932047844,
      "num_tokens": 1968963.0,
      "step": 1000
    },
    {
      "entropy": 7.347122669219971,
      "epoch": 0.10046483730694257,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999999981884985,
      "loss": 7.3358,
      "mean_token_accuracy": 0.10006266087293625,
      "num_tokens": 1979551.0,
      "step": 1005
    },
    {
      "entropy": 7.342271709442139,
      "epoch": 0.10096466236817114,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004999999908292739,
      "loss": 7.2707,
      "mean_token_accuracy": 0.10789052844047546,
      "num_tokens": 1988432.0,
      "step": 1010
    },
    {
      "entropy": 7.405298137664795,
      "epoch": 0.10146448742939972,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999999778091072,
      "loss": 7.3429,
      "mean_token_accuracy": 0.10290407985448838,
      "num_tokens": 1997509.0,
      "step": 1015
    },
    {
      "entropy": 7.448093605041504,
      "epoch": 0.10196431249062828,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999999591279991,
      "loss": 7.5284,
      "mean_token_accuracy": 0.09559334665536881,
      "num_tokens": 2007406.0,
      "step": 1020
    },
    {
      "entropy": 7.455267906188965,
      "epoch": 0.10246413755185685,
      "grad_norm": 1.125,
      "learning_rate": 0.00049999993478595,
      "loss": 7.3217,
      "mean_token_accuracy": 0.09977051839232445,
      "num_tokens": 2016760.0,
      "step": 1025
    },
    {
      "entropy": 7.275325965881348,
      "epoch": 0.10296396261308542,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999999047829604,
      "loss": 7.3609,
      "mean_token_accuracy": 0.10224623009562492,
      "num_tokens": 2027776.0,
      "step": 1030
    },
    {
      "entropy": 7.378179454803467,
      "epoch": 0.103463787674314,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499999869119031,
      "loss": 7.3113,
      "mean_token_accuracy": 0.10338322818279266,
      "num_tokens": 2036955.0,
      "step": 1035
    },
    {
      "entropy": 7.345253419876099,
      "epoch": 0.10396361273554255,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999998277941629,
      "loss": 7.2195,
      "mean_token_accuracy": 0.10754978880286217,
      "num_tokens": 2046118.0,
      "step": 1040
    },
    {
      "entropy": 7.256439685821533,
      "epoch": 0.10446343779677113,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999997808083569,
      "loss": 7.282,
      "mean_token_accuracy": 0.10003011375665664,
      "num_tokens": 2056266.0,
      "step": 1045
    },
    {
      "entropy": 7.421114683151245,
      "epoch": 0.1049632628579997,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999997281616144,
      "loss": 7.3354,
      "mean_token_accuracy": 0.10521339923143387,
      "num_tokens": 2067416.0,
      "step": 1050
    },
    {
      "entropy": 7.324567222595215,
      "epoch": 0.10546308791922827,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999996698539366,
      "loss": 7.33,
      "mean_token_accuracy": 0.10143450647592545,
      "num_tokens": 2076926.0,
      "step": 1055
    },
    {
      "entropy": 7.306933975219726,
      "epoch": 0.10596291298045685,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499999605885325,
      "loss": 7.3169,
      "mean_token_accuracy": 0.10499019995331764,
      "num_tokens": 2088017.0,
      "step": 1060
    },
    {
      "entropy": 7.321994495391846,
      "epoch": 0.1064627380416854,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004999995362557813,
      "loss": 7.3576,
      "mean_token_accuracy": 0.10440385192632676,
      "num_tokens": 2097249.0,
      "step": 1065
    },
    {
      "entropy": 7.292541551589966,
      "epoch": 0.10696256310291398,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000499999460965307,
      "loss": 7.3116,
      "mean_token_accuracy": 0.09968645870685577,
      "num_tokens": 2105778.0,
      "step": 1070
    },
    {
      "entropy": 7.35160641670227,
      "epoch": 0.10746238816414255,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999993800139043,
      "loss": 7.3278,
      "mean_token_accuracy": 0.10377108082175254,
      "num_tokens": 2114710.0,
      "step": 1075
    },
    {
      "entropy": 7.34290714263916,
      "epoch": 0.10796221322537113,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999992934015752,
      "loss": 7.2596,
      "mean_token_accuracy": 0.10727637037634849,
      "num_tokens": 2124002.0,
      "step": 1080
    },
    {
      "entropy": 7.377143383026123,
      "epoch": 0.10846203828659969,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999992011283215,
      "loss": 7.3426,
      "mean_token_accuracy": 0.103351741284132,
      "num_tokens": 2134592.0,
      "step": 1085
    },
    {
      "entropy": 7.273125696182251,
      "epoch": 0.10896186334782826,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999991031941459,
      "loss": 7.1919,
      "mean_token_accuracy": 0.1127089373767376,
      "num_tokens": 2143641.0,
      "step": 1090
    },
    {
      "entropy": 7.344776439666748,
      "epoch": 0.10946168840905683,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004999989995990509,
      "loss": 7.2986,
      "mean_token_accuracy": 0.10175277665257454,
      "num_tokens": 2154738.0,
      "step": 1095
    },
    {
      "entropy": 7.166865539550781,
      "epoch": 0.1099615134702854,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999988903430389,
      "loss": 7.2791,
      "mean_token_accuracy": 0.10936013385653495,
      "num_tokens": 2164365.0,
      "step": 1100
    },
    {
      "entropy": 7.3008204936981205,
      "epoch": 0.11046133853151396,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999987754261126,
      "loss": 7.1491,
      "mean_token_accuracy": 0.1082618147134781,
      "num_tokens": 2173185.0,
      "step": 1105
    },
    {
      "entropy": 7.216965007781982,
      "epoch": 0.11096116359274254,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999986548482751,
      "loss": 7.2612,
      "mean_token_accuracy": 0.10522031784057617,
      "num_tokens": 2184717.0,
      "step": 1110
    },
    {
      "entropy": 7.27905764579773,
      "epoch": 0.11146098865397111,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999985286095293,
      "loss": 7.2279,
      "mean_token_accuracy": 0.10826601237058639,
      "num_tokens": 2193681.0,
      "step": 1115
    },
    {
      "entropy": 7.277110385894775,
      "epoch": 0.11196081371519968,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999983967098785,
      "loss": 7.2467,
      "mean_token_accuracy": 0.10190426334738731,
      "num_tokens": 2203366.0,
      "step": 1120
    },
    {
      "entropy": 7.271683073043823,
      "epoch": 0.11246063877642826,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999982591493258,
      "loss": 7.2459,
      "mean_token_accuracy": 0.10593178719282151,
      "num_tokens": 2213982.0,
      "step": 1125
    },
    {
      "entropy": 7.318293666839599,
      "epoch": 0.11296046383765682,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999981159278749,
      "loss": 7.3031,
      "mean_token_accuracy": 0.10162633955478668,
      "num_tokens": 2222897.0,
      "step": 1130
    },
    {
      "entropy": 7.238396883010864,
      "epoch": 0.11346028889888539,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999979670455292,
      "loss": 7.2558,
      "mean_token_accuracy": 0.11027894765138627,
      "num_tokens": 2232472.0,
      "step": 1135
    },
    {
      "entropy": 7.249378728866577,
      "epoch": 0.11396011396011396,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004999978125022926,
      "loss": 7.294,
      "mean_token_accuracy": 0.1087009809911251,
      "num_tokens": 2242666.0,
      "step": 1140
    },
    {
      "entropy": 7.319674825668335,
      "epoch": 0.11445993902134254,
      "grad_norm": 1.125,
      "learning_rate": 0.000499997652298169,
      "loss": 7.3479,
      "mean_token_accuracy": 0.09625119417905807,
      "num_tokens": 2254586.0,
      "step": 1145
    },
    {
      "entropy": 7.281134080886841,
      "epoch": 0.1149597640825711,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999974864331624,
      "loss": 7.202,
      "mean_token_accuracy": 0.11020608842372895,
      "num_tokens": 2264619.0,
      "step": 1150
    },
    {
      "entropy": 7.388929986953736,
      "epoch": 0.11545958914379967,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999973149072768,
      "loss": 7.2859,
      "mean_token_accuracy": 0.09894285909831524,
      "num_tokens": 2274114.0,
      "step": 1155
    },
    {
      "entropy": 7.06968936920166,
      "epoch": 0.11595941420502824,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999971377205167,
      "loss": 7.1888,
      "mean_token_accuracy": 0.11129210293292999,
      "num_tokens": 2283887.0,
      "step": 1160
    },
    {
      "entropy": 7.307636737823486,
      "epoch": 0.11645923926625681,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999969548728864,
      "loss": 7.2059,
      "mean_token_accuracy": 0.11073495373129845,
      "num_tokens": 2293090.0,
      "step": 1165
    },
    {
      "entropy": 7.2551915645599365,
      "epoch": 0.11695906432748537,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999967663643908,
      "loss": 7.3175,
      "mean_token_accuracy": 0.10531776547431945,
      "num_tokens": 2302828.0,
      "step": 1170
    },
    {
      "entropy": 7.2968638896942135,
      "epoch": 0.11745888938871395,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999965721950345,
      "loss": 7.1898,
      "mean_token_accuracy": 0.10970171466469765,
      "num_tokens": 2311415.0,
      "step": 1175
    },
    {
      "entropy": 7.247892379760742,
      "epoch": 0.11795871444994252,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999963723648222,
      "loss": 7.2905,
      "mean_token_accuracy": 0.1028762400150299,
      "num_tokens": 2321512.0,
      "step": 1180
    },
    {
      "entropy": 7.223724460601806,
      "epoch": 0.1184585395111711,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499996166873759,
      "loss": 7.1816,
      "mean_token_accuracy": 0.10487090200185775,
      "num_tokens": 2331398.0,
      "step": 1185
    },
    {
      "entropy": 7.114894962310791,
      "epoch": 0.11895836457239967,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999959557218502,
      "loss": 7.1949,
      "mean_token_accuracy": 0.10667292214930058,
      "num_tokens": 2341751.0,
      "step": 1190
    },
    {
      "entropy": 7.293861246109008,
      "epoch": 0.11945818963362823,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999957389091012,
      "loss": 7.1852,
      "mean_token_accuracy": 0.10581460520625115,
      "num_tokens": 2352538.0,
      "step": 1195
    },
    {
      "entropy": 7.263128328323364,
      "epoch": 0.1199580146948568,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999955164355171,
      "loss": 7.2807,
      "mean_token_accuracy": 0.101702369004488,
      "num_tokens": 2362433.0,
      "step": 1200
    },
    {
      "entropy": 7.23228964805603,
      "epoch": 0.12045783975608537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004999952883011038,
      "loss": 7.2357,
      "mean_token_accuracy": 0.10086923241615295,
      "num_tokens": 2372431.0,
      "step": 1205
    },
    {
      "entropy": 7.155064105987549,
      "epoch": 0.12095766481731395,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499995054505867,
      "loss": 7.261,
      "mean_token_accuracy": 0.11123752519488335,
      "num_tokens": 2382483.0,
      "step": 1210
    },
    {
      "entropy": 7.333055305480957,
      "epoch": 0.1214574898785425,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999948150498124,
      "loss": 7.2314,
      "mean_token_accuracy": 0.09728648811578751,
      "num_tokens": 2392835.0,
      "step": 1215
    },
    {
      "entropy": 7.157829141616821,
      "epoch": 0.12195731493977108,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004999945699329462,
      "loss": 7.2069,
      "mean_token_accuracy": 0.10083935856819153,
      "num_tokens": 2402305.0,
      "step": 1220
    },
    {
      "entropy": 7.154402446746826,
      "epoch": 0.12245714000099965,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999943191552745,
      "loss": 7.1366,
      "mean_token_accuracy": 0.1104811742901802,
      "num_tokens": 2412093.0,
      "step": 1225
    },
    {
      "entropy": 7.197684717178345,
      "epoch": 0.12295696506222822,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999940627168037,
      "loss": 7.1138,
      "mean_token_accuracy": 0.11224015802145004,
      "num_tokens": 2421349.0,
      "step": 1230
    },
    {
      "entropy": 7.113976287841797,
      "epoch": 0.12345679012345678,
      "grad_norm": 1.125,
      "learning_rate": 0.00049999380061754,
      "loss": 7.0816,
      "mean_token_accuracy": 0.12027929723262787,
      "num_tokens": 2430672.0,
      "step": 1235
    },
    {
      "entropy": 7.171084785461426,
      "epoch": 0.12395661518468536,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999935328574903,
      "loss": 7.1672,
      "mean_token_accuracy": 0.10523176565766335,
      "num_tokens": 2439355.0,
      "step": 1240
    },
    {
      "entropy": 7.167644071578979,
      "epoch": 0.12445644024591393,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999932594366613,
      "loss": 7.1629,
      "mean_token_accuracy": 0.11419163346290588,
      "num_tokens": 2448860.0,
      "step": 1245
    },
    {
      "entropy": 7.165760946273804,
      "epoch": 0.1249562653071425,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999929803550597,
      "loss": 7.2068,
      "mean_token_accuracy": 0.1069261223077774,
      "num_tokens": 2458276.0,
      "step": 1250
    },
    {
      "entropy": 7.17039475440979,
      "epoch": 0.12545609036837108,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004999926956126927,
      "loss": 7.1266,
      "mean_token_accuracy": 0.11022537723183631,
      "num_tokens": 2467791.0,
      "step": 1255
    },
    {
      "entropy": 7.146854448318481,
      "epoch": 0.12595591542959964,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999924052095672,
      "loss": 7.2725,
      "mean_token_accuracy": 0.10534469336271286,
      "num_tokens": 2477549.0,
      "step": 1260
    },
    {
      "entropy": 7.200028324127198,
      "epoch": 0.12645574049082822,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999921091456907,
      "loss": 7.1641,
      "mean_token_accuracy": 0.10597576200962067,
      "num_tokens": 2487549.0,
      "step": 1265
    },
    {
      "entropy": 7.233373498916626,
      "epoch": 0.12695556555205678,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999918074210707,
      "loss": 7.1615,
      "mean_token_accuracy": 0.11181024834513664,
      "num_tokens": 2496799.0,
      "step": 1270
    },
    {
      "entropy": 7.232607460021972,
      "epoch": 0.12745539061328534,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999915000357147,
      "loss": 7.2669,
      "mean_token_accuracy": 0.10368774533271789,
      "num_tokens": 2505795.0,
      "step": 1275
    },
    {
      "entropy": 7.204328823089599,
      "epoch": 0.12795521567451393,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004999911869896305,
      "loss": 7.0697,
      "mean_token_accuracy": 0.10817176178097725,
      "num_tokens": 2515289.0,
      "step": 1280
    },
    {
      "entropy": 7.2058745384216305,
      "epoch": 0.1284550407357425,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999908682828258,
      "loss": 7.2686,
      "mean_token_accuracy": 0.09726808145642281,
      "num_tokens": 2524563.0,
      "step": 1285
    },
    {
      "entropy": 7.179253101348877,
      "epoch": 0.12895486579697105,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004999905439153089,
      "loss": 7.1331,
      "mean_token_accuracy": 0.11065928637981415,
      "num_tokens": 2534057.0,
      "step": 1290
    },
    {
      "entropy": 7.125552940368652,
      "epoch": 0.12945469085819963,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999902138870878,
      "loss": 7.1287,
      "mean_token_accuracy": 0.11491556018590927,
      "num_tokens": 2544646.0,
      "step": 1295
    },
    {
      "entropy": 7.201935863494873,
      "epoch": 0.1299545159194282,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999898781981707,
      "loss": 7.1982,
      "mean_token_accuracy": 0.1065469078719616,
      "num_tokens": 2553843.0,
      "step": 1300
    },
    {
      "entropy": 7.158224201202392,
      "epoch": 0.13045434098065678,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999895368485662,
      "loss": 7.2568,
      "mean_token_accuracy": 0.10226461961865425,
      "num_tokens": 2564061.0,
      "step": 1305
    },
    {
      "entropy": 7.151171636581421,
      "epoch": 0.13095416604188534,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499989189838283,
      "loss": 7.1242,
      "mean_token_accuracy": 0.10985731407999992,
      "num_tokens": 2574758.0,
      "step": 1310
    },
    {
      "entropy": 7.168261194229126,
      "epoch": 0.1314539911031139,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999888371673295,
      "loss": 7.1635,
      "mean_token_accuracy": 0.10855096057057381,
      "num_tokens": 2585213.0,
      "step": 1315
    },
    {
      "entropy": 7.2063212394714355,
      "epoch": 0.13195381616434249,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999884788357147,
      "loss": 7.2354,
      "mean_token_accuracy": 0.10834863558411598,
      "num_tokens": 2595511.0,
      "step": 1320
    },
    {
      "entropy": 7.141773223876953,
      "epoch": 0.13245364122557104,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999881148434478,
      "loss": 7.1519,
      "mean_token_accuracy": 0.10698395371437072,
      "num_tokens": 2605930.0,
      "step": 1325
    },
    {
      "entropy": 7.183522272109985,
      "epoch": 0.13295346628679963,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999877451905378,
      "loss": 7.0084,
      "mean_token_accuracy": 0.12175094336271286,
      "num_tokens": 2614224.0,
      "step": 1330
    },
    {
      "entropy": 7.179550838470459,
      "epoch": 0.1334532913480282,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999873698769941,
      "loss": 7.2662,
      "mean_token_accuracy": 0.1017696812748909,
      "num_tokens": 2624879.0,
      "step": 1335
    },
    {
      "entropy": 7.120278787612915,
      "epoch": 0.13395311640925675,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499986988902826,
      "loss": 7.117,
      "mean_token_accuracy": 0.10926841422915459,
      "num_tokens": 2633836.0,
      "step": 1340
    },
    {
      "entropy": 7.138001871109009,
      "epoch": 0.13445294147048534,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999866022680431,
      "loss": 7.1771,
      "mean_token_accuracy": 0.10937484353780746,
      "num_tokens": 2643754.0,
      "step": 1345
    },
    {
      "entropy": 7.255338096618653,
      "epoch": 0.1349527665317139,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999862099726552,
      "loss": 7.2411,
      "mean_token_accuracy": 0.10106356516480446,
      "num_tokens": 2652708.0,
      "step": 1350
    },
    {
      "entropy": 7.085612630844116,
      "epoch": 0.13545259159294248,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999858120166721,
      "loss": 7.1799,
      "mean_token_accuracy": 0.10279356613755226,
      "num_tokens": 2662353.0,
      "step": 1355
    },
    {
      "entropy": 7.2352677345275875,
      "epoch": 0.13595241665417104,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999854084001039,
      "loss": 7.092,
      "mean_token_accuracy": 0.1117504045367241,
      "num_tokens": 2672376.0,
      "step": 1360
    },
    {
      "entropy": 7.125872230529785,
      "epoch": 0.1364522417153996,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004999849991229608,
      "loss": 7.1979,
      "mean_token_accuracy": 0.10238617360591888,
      "num_tokens": 2682115.0,
      "step": 1365
    },
    {
      "entropy": 7.1724754810333256,
      "epoch": 0.1369520667766282,
      "grad_norm": 1.03125,
      "learning_rate": 0.000499984584185253,
      "loss": 7.1836,
      "mean_token_accuracy": 0.10678994581103325,
      "num_tokens": 2692429.0,
      "step": 1370
    },
    {
      "entropy": 7.102124166488648,
      "epoch": 0.13745189183785675,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999841635869909,
      "loss": 7.1257,
      "mean_token_accuracy": 0.10722747519612312,
      "num_tokens": 2701373.0,
      "step": 1375
    },
    {
      "entropy": 7.279002380371094,
      "epoch": 0.1379517168990853,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004999837373281852,
      "loss": 7.3167,
      "mean_token_accuracy": 0.10132398679852486,
      "num_tokens": 2710717.0,
      "step": 1380
    },
    {
      "entropy": 7.091415548324585,
      "epoch": 0.1384515419603139,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999833054088465,
      "loss": 7.1606,
      "mean_token_accuracy": 0.10936282426118851,
      "num_tokens": 2720936.0,
      "step": 1385
    },
    {
      "entropy": 7.2388688087463375,
      "epoch": 0.13895136702154245,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999828678289858,
      "loss": 7.0993,
      "mean_token_accuracy": 0.11418369710445404,
      "num_tokens": 2730504.0,
      "step": 1390
    },
    {
      "entropy": 7.0881829261779785,
      "epoch": 0.13945119208277104,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000499982424588614,
      "loss": 7.1144,
      "mean_token_accuracy": 0.11084567457437515,
      "num_tokens": 2739069.0,
      "step": 1395
    },
    {
      "entropy": 7.147162532806396,
      "epoch": 0.1399510171439996,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004999819756877422,
      "loss": 7.1945,
      "mean_token_accuracy": 0.10329353213310241,
      "num_tokens": 2748941.0,
      "step": 1400
    },
    {
      "entropy": 7.18705563545227,
      "epoch": 0.14045084220522816,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999815211263819,
      "loss": 7.1507,
      "mean_token_accuracy": 0.10858112126588822,
      "num_tokens": 2758915.0,
      "step": 1405
    },
    {
      "entropy": 7.191486263275147,
      "epoch": 0.14095066726645675,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999810609045444,
      "loss": 7.1754,
      "mean_token_accuracy": 0.10965260490775108,
      "num_tokens": 2768972.0,
      "step": 1410
    },
    {
      "entropy": 7.156418704986573,
      "epoch": 0.1414504923276853,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999805950222412,
      "loss": 7.0278,
      "mean_token_accuracy": 0.10822951421141624,
      "num_tokens": 2777989.0,
      "step": 1415
    },
    {
      "entropy": 7.091416835784912,
      "epoch": 0.1419503173889139,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999801234794843,
      "loss": 7.1096,
      "mean_token_accuracy": 0.11228812485933304,
      "num_tokens": 2788227.0,
      "step": 1420
    },
    {
      "entropy": 7.147107839584351,
      "epoch": 0.14245014245014245,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999796462762853,
      "loss": 7.1112,
      "mean_token_accuracy": 0.11311104521155357,
      "num_tokens": 2797729.0,
      "step": 1425
    },
    {
      "entropy": 7.119664812088013,
      "epoch": 0.142949967511371,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999791634126562,
      "loss": 7.1126,
      "mean_token_accuracy": 0.11075319051742553,
      "num_tokens": 2807215.0,
      "step": 1430
    },
    {
      "entropy": 7.124571132659912,
      "epoch": 0.1434497925725996,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004999786748886094,
      "loss": 7.0246,
      "mean_token_accuracy": 0.11238908171653747,
      "num_tokens": 2816945.0,
      "step": 1435
    },
    {
      "entropy": 7.017452239990234,
      "epoch": 0.14394961763382816,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999781807041569,
      "loss": 7.1146,
      "mean_token_accuracy": 0.10776891112327576,
      "num_tokens": 2827072.0,
      "step": 1440
    },
    {
      "entropy": 7.165062618255615,
      "epoch": 0.14444944269505672,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999776808593113,
      "loss": 7.0482,
      "mean_token_accuracy": 0.11216341257095337,
      "num_tokens": 2837928.0,
      "step": 1445
    },
    {
      "entropy": 7.001790475845337,
      "epoch": 0.1449492677562853,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999771753540852,
      "loss": 7.0846,
      "mean_token_accuracy": 0.11152777597308158,
      "num_tokens": 2847861.0,
      "step": 1450
    },
    {
      "entropy": 7.149148035049438,
      "epoch": 0.14544909281751386,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999766641884912,
      "loss": 7.0848,
      "mean_token_accuracy": 0.11256931573152543,
      "num_tokens": 2857885.0,
      "step": 1455
    },
    {
      "entropy": 7.160283946990967,
      "epoch": 0.14594891787874245,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999761473625422,
      "loss": 7.0932,
      "mean_token_accuracy": 0.10660878345370292,
      "num_tokens": 2867004.0,
      "step": 1460
    },
    {
      "entropy": 7.065319442749024,
      "epoch": 0.146448742939971,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999756248762513,
      "loss": 6.9736,
      "mean_token_accuracy": 0.11344703435897827,
      "num_tokens": 2876285.0,
      "step": 1465
    },
    {
      "entropy": 7.079582357406617,
      "epoch": 0.14694856800119957,
      "grad_norm": 0.890625,
      "learning_rate": 0.0004999750967296315,
      "loss": 7.112,
      "mean_token_accuracy": 0.10987017676234245,
      "num_tokens": 2886967.0,
      "step": 1470
    },
    {
      "entropy": 7.077367973327637,
      "epoch": 0.14744839306242816,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999745629226962,
      "loss": 7.0936,
      "mean_token_accuracy": 0.10697494596242904,
      "num_tokens": 2897137.0,
      "step": 1475
    },
    {
      "entropy": 7.038984680175782,
      "epoch": 0.14794821812365672,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999740234554589,
      "loss": 7.0472,
      "mean_token_accuracy": 0.12016323506832123,
      "num_tokens": 2907658.0,
      "step": 1480
    },
    {
      "entropy": 7.136841917037964,
      "epoch": 0.1484480431848853,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999734783279327,
      "loss": 7.0572,
      "mean_token_accuracy": 0.11264566704630852,
      "num_tokens": 2917801.0,
      "step": 1485
    },
    {
      "entropy": 7.058710241317749,
      "epoch": 0.14894786824611386,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499972927540132,
      "loss": 7.1593,
      "mean_token_accuracy": 0.1104632668197155,
      "num_tokens": 2927188.0,
      "step": 1490
    },
    {
      "entropy": 7.148079061508179,
      "epoch": 0.14944769330734242,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999723710920703,
      "loss": 7.0421,
      "mean_token_accuracy": 0.10991474613547325,
      "num_tokens": 2936781.0,
      "step": 1495
    },
    {
      "entropy": 6.985206222534179,
      "epoch": 0.149947518368571,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999718089837616,
      "loss": 7.0034,
      "mean_token_accuracy": 0.11071012616157531,
      "num_tokens": 2946426.0,
      "step": 1500
    },
    {
      "entropy": 6.977966022491455,
      "epoch": 0.15044734342979957,
      "grad_norm": 1.03125,
      "learning_rate": 0.00049997124121522,
      "loss": 7.0591,
      "mean_token_accuracy": 0.11089984625577927,
      "num_tokens": 2956246.0,
      "step": 1505
    },
    {
      "entropy": 7.119696187973022,
      "epoch": 0.15094716849102813,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999706677864599,
      "loss": 7.0301,
      "mean_token_accuracy": 0.11129701286554336,
      "num_tokens": 2965930.0,
      "step": 1510
    },
    {
      "entropy": 7.027358150482177,
      "epoch": 0.15144699355225671,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004999700886974958,
      "loss": 6.9594,
      "mean_token_accuracy": 0.11398379057645798,
      "num_tokens": 2975975.0,
      "step": 1515
    },
    {
      "entropy": 7.029348134994507,
      "epoch": 0.15194681861348527,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499969503948342,
      "loss": 6.9922,
      "mean_token_accuracy": 0.11278997659683228,
      "num_tokens": 2985343.0,
      "step": 1520
    },
    {
      "entropy": 6.995435190200806,
      "epoch": 0.15244664367471386,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999689135390134,
      "loss": 7.0206,
      "mean_token_accuracy": 0.11354669854044915,
      "num_tokens": 2994420.0,
      "step": 1525
    },
    {
      "entropy": 7.102861928939819,
      "epoch": 0.15294646873594242,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999683174695249,
      "loss": 7.1314,
      "mean_token_accuracy": 0.11147044822573662,
      "num_tokens": 3004447.0,
      "step": 1530
    },
    {
      "entropy": 7.11038761138916,
      "epoch": 0.15344629379717098,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999677157398914,
      "loss": 7.0136,
      "mean_token_accuracy": 0.11158140152692794,
      "num_tokens": 3014596.0,
      "step": 1535
    },
    {
      "entropy": 7.001529932022095,
      "epoch": 0.15394611885839957,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999671083501281,
      "loss": 6.9424,
      "mean_token_accuracy": 0.11554695665836334,
      "num_tokens": 3024528.0,
      "step": 1540
    },
    {
      "entropy": 7.043489742279053,
      "epoch": 0.15444594391962813,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999664953002502,
      "loss": 7.0757,
      "mean_token_accuracy": 0.1089545451104641,
      "num_tokens": 3035233.0,
      "step": 1545
    },
    {
      "entropy": 7.057877731323242,
      "epoch": 0.1549457689808567,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499965876590273,
      "loss": 7.0639,
      "mean_token_accuracy": 0.11354411020874977,
      "num_tokens": 3045890.0,
      "step": 1550
    },
    {
      "entropy": 7.006091165542602,
      "epoch": 0.15544559404208527,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999652522202125,
      "loss": 6.9704,
      "mean_token_accuracy": 0.11927012428641319,
      "num_tokens": 3056640.0,
      "step": 1555
    },
    {
      "entropy": 6.988191604614258,
      "epoch": 0.15594541910331383,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004999646221900839,
      "loss": 6.9995,
      "mean_token_accuracy": 0.11643217280507087,
      "num_tokens": 3066938.0,
      "step": 1560
    },
    {
      "entropy": 7.100666046142578,
      "epoch": 0.15644524416454242,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999639864999034,
      "loss": 7.1224,
      "mean_token_accuracy": 0.11021771505475045,
      "num_tokens": 3076791.0,
      "step": 1565
    },
    {
      "entropy": 7.040261554718017,
      "epoch": 0.15694506922577098,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499963345149687,
      "loss": 6.9766,
      "mean_token_accuracy": 0.1124839298427105,
      "num_tokens": 3087532.0,
      "step": 1570
    },
    {
      "entropy": 7.011826610565185,
      "epoch": 0.15744489428699954,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999626981394506,
      "loss": 7.1019,
      "mean_token_accuracy": 0.11321934536099434,
      "num_tokens": 3097465.0,
      "step": 1575
    },
    {
      "entropy": 7.116326904296875,
      "epoch": 0.15794471934822812,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999620454692106,
      "loss": 7.0088,
      "mean_token_accuracy": 0.11308922916650772,
      "num_tokens": 3107162.0,
      "step": 1580
    },
    {
      "entropy": 7.002964687347412,
      "epoch": 0.15844454440945668,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004999613871389836,
      "loss": 6.9932,
      "mean_token_accuracy": 0.11376159563660622,
      "num_tokens": 3117154.0,
      "step": 1585
    },
    {
      "entropy": 7.0667582035064695,
      "epoch": 0.15894436947068527,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004999607231487858,
      "loss": 7.0118,
      "mean_token_accuracy": 0.11034693121910095,
      "num_tokens": 3127635.0,
      "step": 1590
    },
    {
      "entropy": 7.099604892730713,
      "epoch": 0.15944419453191383,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499960053498634,
      "loss": 7.0542,
      "mean_token_accuracy": 0.10722034946084022,
      "num_tokens": 3137806.0,
      "step": 1595
    },
    {
      "entropy": 6.99808521270752,
      "epoch": 0.1599440195931424,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004999593781885454,
      "loss": 7.0869,
      "mean_token_accuracy": 0.10542495176196098,
      "num_tokens": 3147704.0,
      "step": 1600
    },
    {
      "entropy": 7.039315319061279,
      "epoch": 0.16044384465437098,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004999586972185366,
      "loss": 7.0516,
      "mean_token_accuracy": 0.10607154071331024,
      "num_tokens": 3158486.0,
      "step": 1605
    },
    {
      "entropy": 7.045173120498657,
      "epoch": 0.16094366971559954,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999580105886248,
      "loss": 7.0479,
      "mean_token_accuracy": 0.11122439876198768,
      "num_tokens": 3168518.0,
      "step": 1610
    },
    {
      "entropy": 7.000042533874511,
      "epoch": 0.16144349477682812,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004999573182988275,
      "loss": 6.9298,
      "mean_token_accuracy": 0.11688626557588577,
      "num_tokens": 3177502.0,
      "step": 1615
    },
    {
      "entropy": 7.017080640792846,
      "epoch": 0.16194331983805668,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999566203491619,
      "loss": 6.9921,
      "mean_token_accuracy": 0.11503020003437996,
      "num_tokens": 3187408.0,
      "step": 1620
    },
    {
      "entropy": 7.0513831615448,
      "epoch": 0.16244314489928524,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999559167396455,
      "loss": 7.0567,
      "mean_token_accuracy": 0.11678390353918075,
      "num_tokens": 3196812.0,
      "step": 1625
    },
    {
      "entropy": 6.902290725708008,
      "epoch": 0.16294296996051383,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999552074702963,
      "loss": 6.9108,
      "mean_token_accuracy": 0.11789578422904015,
      "num_tokens": 3206159.0,
      "step": 1630
    },
    {
      "entropy": 7.007291650772094,
      "epoch": 0.1634427950217424,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004999544925411318,
      "loss": 6.8954,
      "mean_token_accuracy": 0.12061700448393822,
      "num_tokens": 3217329.0,
      "step": 1635
    },
    {
      "entropy": 6.951911449432373,
      "epoch": 0.16394262008297095,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0004999537719521703,
      "loss": 7.0219,
      "mean_token_accuracy": 0.11060998439788819,
      "num_tokens": 3227558.0,
      "step": 1640
    },
    {
      "entropy": 7.113637161254883,
      "epoch": 0.16444244514419953,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999530457034297,
      "loss": 7.1038,
      "mean_token_accuracy": 0.10778464749455452,
      "num_tokens": 3237719.0,
      "step": 1645
    },
    {
      "entropy": 7.020438766479492,
      "epoch": 0.1649422702054281,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999523137949283,
      "loss": 7.0316,
      "mean_token_accuracy": 0.11263542622327805,
      "num_tokens": 3246483.0,
      "step": 1650
    },
    {
      "entropy": 6.933847713470459,
      "epoch": 0.16544209526665668,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999515762266846,
      "loss": 6.9301,
      "mean_token_accuracy": 0.1197948582470417,
      "num_tokens": 3256955.0,
      "step": 1655
    },
    {
      "entropy": 7.0298436164855955,
      "epoch": 0.16594192032788524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999508329987173,
      "loss": 6.998,
      "mean_token_accuracy": 0.11044911220669747,
      "num_tokens": 3266799.0,
      "step": 1660
    },
    {
      "entropy": 6.953704118728638,
      "epoch": 0.1664417453891138,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999500841110447,
      "loss": 6.9928,
      "mean_token_accuracy": 0.11013735309243203,
      "num_tokens": 3276595.0,
      "step": 1665
    },
    {
      "entropy": 7.008637952804565,
      "epoch": 0.16694157045034239,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499949329563686,
      "loss": 6.9401,
      "mean_token_accuracy": 0.11440528258681297,
      "num_tokens": 3286047.0,
      "step": 1670
    },
    {
      "entropy": 7.053787803649902,
      "epoch": 0.16744139551157095,
      "grad_norm": 1.046875,
      "learning_rate": 0.00049994856935666,
      "loss": 7.0196,
      "mean_token_accuracy": 0.11221319362521172,
      "num_tokens": 3295364.0,
      "step": 1675
    },
    {
      "entropy": 6.936156606674194,
      "epoch": 0.16794122057279953,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000499947803489986,
      "loss": 6.9291,
      "mean_token_accuracy": 0.12198807075619697,
      "num_tokens": 3305319.0,
      "step": 1680
    },
    {
      "entropy": 6.941357517242432,
      "epoch": 0.1684410456340281,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999470319636829,
      "loss": 6.9625,
      "mean_token_accuracy": 0.11831458657979965,
      "num_tokens": 3315349.0,
      "step": 1685
    },
    {
      "entropy": 7.03402681350708,
      "epoch": 0.16894087069525665,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004999462547777705,
      "loss": 7.0281,
      "mean_token_accuracy": 0.10700425431132317,
      "num_tokens": 3325603.0,
      "step": 1690
    },
    {
      "entropy": 6.9488404273986815,
      "epoch": 0.16944069575648524,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999454719322682,
      "loss": 6.9345,
      "mean_token_accuracy": 0.11474086120724677,
      "num_tokens": 3334693.0,
      "step": 1695
    },
    {
      "entropy": 7.03020396232605,
      "epoch": 0.1699405208177138,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999446834271958,
      "loss": 6.9321,
      "mean_token_accuracy": 0.1120713397860527,
      "num_tokens": 3345259.0,
      "step": 1700
    },
    {
      "entropy": 6.898108196258545,
      "epoch": 0.17044034587894236,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004999438892625728,
      "loss": 6.9402,
      "mean_token_accuracy": 0.11133012548089027,
      "num_tokens": 3354072.0,
      "step": 1705
    },
    {
      "entropy": 7.105929708480835,
      "epoch": 0.17094017094017094,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004999430894384198,
      "loss": 6.9114,
      "mean_token_accuracy": 0.11355426013469697,
      "num_tokens": 3364881.0,
      "step": 1710
    },
    {
      "entropy": 6.9483030319213865,
      "epoch": 0.1714399960013995,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999422839547564,
      "loss": 6.9086,
      "mean_token_accuracy": 0.11701997593045235,
      "num_tokens": 3374030.0,
      "step": 1715
    },
    {
      "entropy": 6.9252705574035645,
      "epoch": 0.1719398210626281,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999414728116029,
      "loss": 6.8956,
      "mean_token_accuracy": 0.11267698481678963,
      "num_tokens": 3384009.0,
      "step": 1720
    },
    {
      "entropy": 7.041887044906616,
      "epoch": 0.17243964612385665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999406560089798,
      "loss": 7.0475,
      "mean_token_accuracy": 0.10910101756453514,
      "num_tokens": 3393660.0,
      "step": 1725
    },
    {
      "entropy": 7.037035226821899,
      "epoch": 0.1729394711850852,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004999398335469078,
      "loss": 7.0708,
      "mean_token_accuracy": 0.11321713551878929,
      "num_tokens": 3404364.0,
      "step": 1730
    },
    {
      "entropy": 6.9971565246582035,
      "epoch": 0.1734392962463138,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999390054254074,
      "loss": 6.9521,
      "mean_token_accuracy": 0.1106104426085949,
      "num_tokens": 3412990.0,
      "step": 1735
    },
    {
      "entropy": 6.852579402923584,
      "epoch": 0.17393912130754235,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004999381716444996,
      "loss": 6.9329,
      "mean_token_accuracy": 0.11495454907417298,
      "num_tokens": 3422339.0,
      "step": 1740
    },
    {
      "entropy": 6.959084606170654,
      "epoch": 0.17443894636877094,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004999373322042052,
      "loss": 6.9043,
      "mean_token_accuracy": 0.11547202616930008,
      "num_tokens": 3433093.0,
      "step": 1745
    },
    {
      "entropy": 7.003214359283447,
      "epoch": 0.1749387714299995,
      "grad_norm": 1.0,
      "learning_rate": 0.0004999364871045455,
      "loss": 6.9932,
      "mean_token_accuracy": 0.11007570326328278,
      "num_tokens": 3443464.0,
      "step": 1750
    },
    {
      "entropy": 6.978196811676026,
      "epoch": 0.17543859649122806,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999356363455415,
      "loss": 6.9345,
      "mean_token_accuracy": 0.11543623134493827,
      "num_tokens": 3453380.0,
      "step": 1755
    },
    {
      "entropy": 6.9858551025390625,
      "epoch": 0.17593842155245665,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000499934779927215,
      "loss": 6.9758,
      "mean_token_accuracy": 0.11607812270522118,
      "num_tokens": 3463583.0,
      "step": 1760
    },
    {
      "entropy": 6.9328991889953615,
      "epoch": 0.1764382466136852,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999339178495872,
      "loss": 6.8913,
      "mean_token_accuracy": 0.122561364620924,
      "num_tokens": 3473640.0,
      "step": 1765
    },
    {
      "entropy": 6.92822413444519,
      "epoch": 0.1769380716749138,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999330501126798,
      "loss": 6.9503,
      "mean_token_accuracy": 0.1169328160583973,
      "num_tokens": 3483827.0,
      "step": 1770
    },
    {
      "entropy": 6.91102409362793,
      "epoch": 0.17743789673614235,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999321767165148,
      "loss": 6.9004,
      "mean_token_accuracy": 0.12227763310074806,
      "num_tokens": 3493763.0,
      "step": 1775
    },
    {
      "entropy": 7.019092321395874,
      "epoch": 0.1779377217973709,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999312976611142,
      "loss": 7.0148,
      "mean_token_accuracy": 0.11686293035745621,
      "num_tokens": 3503119.0,
      "step": 1780
    },
    {
      "entropy": 7.024073219299316,
      "epoch": 0.1784375468585995,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999304129465,
      "loss": 6.9535,
      "mean_token_accuracy": 0.11344824209809304,
      "num_tokens": 3513235.0,
      "step": 1785
    },
    {
      "entropy": 7.011400938034058,
      "epoch": 0.17893737191982806,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999295225726945,
      "loss": 7.0523,
      "mean_token_accuracy": 0.10902164354920388,
      "num_tokens": 3523633.0,
      "step": 1790
    },
    {
      "entropy": 6.947716617584229,
      "epoch": 0.17943719698105662,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00049992862653972,
      "loss": 6.8941,
      "mean_token_accuracy": 0.11354788541793823,
      "num_tokens": 3532888.0,
      "step": 1795
    },
    {
      "entropy": 6.984013509750366,
      "epoch": 0.1799370220422852,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004999277248475992,
      "loss": 6.9221,
      "mean_token_accuracy": 0.11461929455399514,
      "num_tokens": 3541942.0,
      "step": 1800
    },
    {
      "entropy": 6.907126140594483,
      "epoch": 0.18043684710351376,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004999268174963547,
      "loss": 6.8983,
      "mean_token_accuracy": 0.11499835699796676,
      "num_tokens": 3551544.0,
      "step": 1805
    },
    {
      "entropy": 6.945745325088501,
      "epoch": 0.18093667216474235,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999259044860093,
      "loss": 6.8902,
      "mean_token_accuracy": 0.11644125431776046,
      "num_tokens": 3561024.0,
      "step": 1810
    },
    {
      "entropy": 6.9447986602783205,
      "epoch": 0.1814364972259709,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999249858165861,
      "loss": 6.8811,
      "mean_token_accuracy": 0.1235733337700367,
      "num_tokens": 3570962.0,
      "step": 1815
    },
    {
      "entropy": 6.819528198242187,
      "epoch": 0.18193632228719947,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004999240614881082,
      "loss": 6.8763,
      "mean_token_accuracy": 0.1207821324467659,
      "num_tokens": 3580669.0,
      "step": 1820
    },
    {
      "entropy": 7.0006632804870605,
      "epoch": 0.18243614734842806,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004999231315005987,
      "loss": 6.8879,
      "mean_token_accuracy": 0.11845619827508927,
      "num_tokens": 3590627.0,
      "step": 1825
    },
    {
      "entropy": 6.9256019115448,
      "epoch": 0.18293597240965662,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004999221958540811,
      "loss": 6.9646,
      "mean_token_accuracy": 0.10778707936406136,
      "num_tokens": 3599872.0,
      "step": 1830
    },
    {
      "entropy": 6.9726580619812015,
      "epoch": 0.1834357974708852,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004999212545485789,
      "loss": 6.9475,
      "mean_token_accuracy": 0.11592329442501068,
      "num_tokens": 3609809.0,
      "step": 1835
    },
    {
      "entropy": 7.054142475128174,
      "epoch": 0.18393562253211376,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004999203075841159,
      "loss": 6.9647,
      "mean_token_accuracy": 0.11736813709139823,
      "num_tokens": 3619385.0,
      "step": 1840
    },
    {
      "entropy": 6.828405857086182,
      "epoch": 0.18443544759334232,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004999193549607157,
      "loss": 6.8006,
      "mean_token_accuracy": 0.12424161806702613,
      "num_tokens": 3628867.0,
      "step": 1845
    },
    {
      "entropy": 6.887426710128784,
      "epoch": 0.1849352726545709,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004999183966784026,
      "loss": 6.8011,
      "mean_token_accuracy": 0.13002747148275376,
      "num_tokens": 3637528.0,
      "step": 1850
    },
    {
      "entropy": 6.9064164638519285,
      "epoch": 0.18543509771579947,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004999174327372004,
      "loss": 6.943,
      "mean_token_accuracy": 0.11500856131315232,
      "num_tokens": 3648011.0,
      "step": 1855
    },
    {
      "entropy": 6.918382453918457,
      "epoch": 0.18593492277702803,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004999164631371335,
      "loss": 6.9275,
      "mean_token_accuracy": 0.11812992170453071,
      "num_tokens": 3657648.0,
      "step": 1860
    },
    {
      "entropy": 7.01478066444397,
      "epoch": 0.18643474783825661,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999154878782262,
      "loss": 6.9333,
      "mean_token_accuracy": 0.11356185749173164,
      "num_tokens": 3666965.0,
      "step": 1865
    },
    {
      "entropy": 6.965643262863159,
      "epoch": 0.18693457289948517,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499914506960503,
      "loss": 6.9537,
      "mean_token_accuracy": 0.10866264775395393,
      "num_tokens": 3676609.0,
      "step": 1870
    },
    {
      "entropy": 6.886865043640137,
      "epoch": 0.18743439796071376,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004999135203839889,
      "loss": 6.8298,
      "mean_token_accuracy": 0.1227896973490715,
      "num_tokens": 3685884.0,
      "step": 1875
    },
    {
      "entropy": 6.844762706756592,
      "epoch": 0.18793422302194232,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004999125281487084,
      "loss": 6.9795,
      "mean_token_accuracy": 0.114560666680336,
      "num_tokens": 3695436.0,
      "step": 1880
    },
    {
      "entropy": 6.946454906463623,
      "epoch": 0.18843404808317088,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004999115302546866,
      "loss": 6.7854,
      "mean_token_accuracy": 0.12357550859451294,
      "num_tokens": 3704806.0,
      "step": 1885
    },
    {
      "entropy": 6.897518253326416,
      "epoch": 0.18893387314439947,
      "grad_norm": 1.375,
      "learning_rate": 0.0004999105267019486,
      "loss": 6.8642,
      "mean_token_accuracy": 0.1193320944905281,
      "num_tokens": 3713927.0,
      "step": 1890
    },
    {
      "entropy": 6.967895030975342,
      "epoch": 0.18943369820562803,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004999095174905195,
      "loss": 6.9635,
      "mean_token_accuracy": 0.11119818165898324,
      "num_tokens": 3723685.0,
      "step": 1895
    },
    {
      "entropy": 6.898874568939209,
      "epoch": 0.1899335232668566,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004999085026204249,
      "loss": 6.9147,
      "mean_token_accuracy": 0.11309322565793992,
      "num_tokens": 3733918.0,
      "step": 1900
    },
    {
      "entropy": 6.957060718536377,
      "epoch": 0.19043334832808517,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004999074820916903,
      "loss": 6.8693,
      "mean_token_accuracy": 0.11191848888993264,
      "num_tokens": 3744466.0,
      "step": 1905
    },
    {
      "entropy": 6.869232225418091,
      "epoch": 0.19093317338931373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004999064559043412,
      "loss": 6.827,
      "mean_token_accuracy": 0.11591498851776123,
      "num_tokens": 3754082.0,
      "step": 1910
    },
    {
      "entropy": 6.862514638900757,
      "epoch": 0.19143299845054232,
      "grad_norm": 1.125,
      "learning_rate": 0.0004999054240584037,
      "loss": 6.8541,
      "mean_token_accuracy": 0.12327471598982812,
      "num_tokens": 3763111.0,
      "step": 1915
    },
    {
      "entropy": 6.996739625930786,
      "epoch": 0.19193282351177088,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004999043865539035,
      "loss": 6.8973,
      "mean_token_accuracy": 0.11046274900436401,
      "num_tokens": 3772555.0,
      "step": 1920
    },
    {
      "entropy": 6.9612620830535885,
      "epoch": 0.19243264857299944,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999033433908667,
      "loss": 6.9552,
      "mean_token_accuracy": 0.11249120384454728,
      "num_tokens": 3783073.0,
      "step": 1925
    },
    {
      "entropy": 6.834769010543823,
      "epoch": 0.19293247363422802,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004999022945693198,
      "loss": 6.8807,
      "mean_token_accuracy": 0.11839989647269249,
      "num_tokens": 3792277.0,
      "step": 1930
    },
    {
      "entropy": 6.931283950805664,
      "epoch": 0.19343229869545658,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499901240089289,
      "loss": 6.8551,
      "mean_token_accuracy": 0.11572782248258591,
      "num_tokens": 3801098.0,
      "step": 1935
    },
    {
      "entropy": 6.933052253723145,
      "epoch": 0.19393212375668517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004999001799508009,
      "loss": 6.8941,
      "mean_token_accuracy": 0.1140223778784275,
      "num_tokens": 3810614.0,
      "step": 1940
    },
    {
      "entropy": 6.9448882102966305,
      "epoch": 0.19443194881791373,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998991141538821,
      "loss": 6.8723,
      "mean_token_accuracy": 0.11779175996780396,
      "num_tokens": 3818989.0,
      "step": 1945
    },
    {
      "entropy": 6.793414831161499,
      "epoch": 0.1949317738791423,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998980426985596,
      "loss": 6.9092,
      "mean_token_accuracy": 0.1152567982673645,
      "num_tokens": 3829319.0,
      "step": 1950
    },
    {
      "entropy": 6.958078002929687,
      "epoch": 0.19543159894037088,
      "grad_norm": 0.92578125,
      "learning_rate": 0.00049989696558486,
      "loss": 6.9356,
      "mean_token_accuracy": 0.1145101934671402,
      "num_tokens": 3839545.0,
      "step": 1955
    },
    {
      "entropy": 6.871245193481445,
      "epoch": 0.19593142400159944,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004998958828128108,
      "loss": 6.8735,
      "mean_token_accuracy": 0.1180442787706852,
      "num_tokens": 3849029.0,
      "step": 1960
    },
    {
      "entropy": 6.933473062515259,
      "epoch": 0.19643124906282802,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998947943824389,
      "loss": 6.9198,
      "mean_token_accuracy": 0.11709642335772515,
      "num_tokens": 3859463.0,
      "step": 1965
    },
    {
      "entropy": 7.0088598251342775,
      "epoch": 0.19693107412405658,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998937002937719,
      "loss": 6.9588,
      "mean_token_accuracy": 0.11571476459503174,
      "num_tokens": 3869695.0,
      "step": 1970
    },
    {
      "entropy": 6.87238917350769,
      "epoch": 0.19743089918528514,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004998926005468373,
      "loss": 6.9183,
      "mean_token_accuracy": 0.11517301499843598,
      "num_tokens": 3879608.0,
      "step": 1975
    },
    {
      "entropy": 6.92598967552185,
      "epoch": 0.19793072424651373,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998914951416627,
      "loss": 6.8308,
      "mean_token_accuracy": 0.11759625598788262,
      "num_tokens": 3888709.0,
      "step": 1980
    },
    {
      "entropy": 6.7358252048492435,
      "epoch": 0.1984305493077423,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499890384078276,
      "loss": 6.7746,
      "mean_token_accuracy": 0.12024817541241646,
      "num_tokens": 3898392.0,
      "step": 1985
    },
    {
      "entropy": 6.905943870544434,
      "epoch": 0.19893037436897085,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004998892673567051,
      "loss": 6.9924,
      "mean_token_accuracy": 0.10537268072366715,
      "num_tokens": 3909508.0,
      "step": 1990
    },
    {
      "entropy": 6.925677967071533,
      "epoch": 0.19943019943019943,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004998881449769781,
      "loss": 6.8298,
      "mean_token_accuracy": 0.11959851756691933,
      "num_tokens": 3918155.0,
      "step": 1995
    },
    {
      "entropy": 6.9149247169494625,
      "epoch": 0.199930024491428,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004998870169391232,
      "loss": 6.8808,
      "mean_token_accuracy": 0.11446636393666268,
      "num_tokens": 3928478.0,
      "step": 2000
    },
    {
      "entropy": 6.903803873062134,
      "epoch": 0.20042984955265658,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499885883243169,
      "loss": 6.9271,
      "mean_token_accuracy": 0.12026000469923019,
      "num_tokens": 3938126.0,
      "step": 2005
    },
    {
      "entropy": 6.782207250595093,
      "epoch": 0.20092967461388514,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998847438891437,
      "loss": 6.7315,
      "mean_token_accuracy": 0.11714796498417854,
      "num_tokens": 3948176.0,
      "step": 2010
    },
    {
      "entropy": 6.901465272903442,
      "epoch": 0.2014294996751137,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004998835988770761,
      "loss": 6.8789,
      "mean_token_accuracy": 0.1191024824976921,
      "num_tokens": 3958983.0,
      "step": 2015
    },
    {
      "entropy": 6.933460426330567,
      "epoch": 0.20192932473634229,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499882448206995,
      "loss": 6.8457,
      "mean_token_accuracy": 0.12024828121066093,
      "num_tokens": 3968766.0,
      "step": 2020
    },
    {
      "entropy": 6.861673736572266,
      "epoch": 0.20242914979757085,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998812918789293,
      "loss": 6.8413,
      "mean_token_accuracy": 0.11853379607200623,
      "num_tokens": 3979487.0,
      "step": 2025
    },
    {
      "entropy": 6.832311201095581,
      "epoch": 0.20292897485879943,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004998801298929084,
      "loss": 6.8225,
      "mean_token_accuracy": 0.11745660156011581,
      "num_tokens": 3988450.0,
      "step": 2030
    },
    {
      "entropy": 6.882190370559693,
      "epoch": 0.203428799920028,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004998789622489611,
      "loss": 6.825,
      "mean_token_accuracy": 0.11936729922890663,
      "num_tokens": 3997492.0,
      "step": 2035
    },
    {
      "entropy": 6.902972221374512,
      "epoch": 0.20392862498125655,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998777889471171,
      "loss": 6.8323,
      "mean_token_accuracy": 0.12065311595797538,
      "num_tokens": 4007298.0,
      "step": 2040
    },
    {
      "entropy": 6.8405660629272464,
      "epoch": 0.20442845004248514,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998766099874056,
      "loss": 6.8628,
      "mean_token_accuracy": 0.1189582072198391,
      "num_tokens": 4017727.0,
      "step": 2045
    },
    {
      "entropy": 6.887753915786743,
      "epoch": 0.2049282751037137,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004998754253698566,
      "loss": 6.8607,
      "mean_token_accuracy": 0.11793215274810791,
      "num_tokens": 4027618.0,
      "step": 2050
    },
    {
      "entropy": 6.814794063568115,
      "epoch": 0.20542810016494226,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998742350944998,
      "loss": 6.8848,
      "mean_token_accuracy": 0.11500374749302864,
      "num_tokens": 4038831.0,
      "step": 2055
    },
    {
      "entropy": 6.859647417068482,
      "epoch": 0.20592792522617084,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004998730391613651,
      "loss": 6.8482,
      "mean_token_accuracy": 0.11991979852318764,
      "num_tokens": 4048736.0,
      "step": 2060
    },
    {
      "entropy": 6.81789493560791,
      "epoch": 0.2064277502873994,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004998718375704825,
      "loss": 6.7655,
      "mean_token_accuracy": 0.11906948909163476,
      "num_tokens": 4057633.0,
      "step": 2065
    },
    {
      "entropy": 6.886240243911743,
      "epoch": 0.206927575348628,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998706303218825,
      "loss": 6.8785,
      "mean_token_accuracy": 0.1240123063325882,
      "num_tokens": 4067883.0,
      "step": 2070
    },
    {
      "entropy": 6.954527091979981,
      "epoch": 0.20742740040985655,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998694174155952,
      "loss": 6.7437,
      "mean_token_accuracy": 0.12506394237279891,
      "num_tokens": 4078147.0,
      "step": 2075
    },
    {
      "entropy": 6.741695070266724,
      "epoch": 0.2079272254710851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998681988516512,
      "loss": 6.7181,
      "mean_token_accuracy": 0.12472501173615455,
      "num_tokens": 4087130.0,
      "step": 2080
    },
    {
      "entropy": 6.781894207000732,
      "epoch": 0.2084270505323137,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998669746300811,
      "loss": 6.851,
      "mean_token_accuracy": 0.12515389323234558,
      "num_tokens": 4096840.0,
      "step": 2085
    },
    {
      "entropy": 7.006636762619019,
      "epoch": 0.20892687559354226,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998657447509159,
      "loss": 6.8735,
      "mean_token_accuracy": 0.11554861664772034,
      "num_tokens": 4105842.0,
      "step": 2090
    },
    {
      "entropy": 6.778407716751099,
      "epoch": 0.20942670065477084,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004998645092141864,
      "loss": 6.7627,
      "mean_token_accuracy": 0.12160579934716224,
      "num_tokens": 4114865.0,
      "step": 2095
    },
    {
      "entropy": 6.914173173904419,
      "epoch": 0.2099265257159994,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998632680199238,
      "loss": 6.8991,
      "mean_token_accuracy": 0.11216249763965606,
      "num_tokens": 4124954.0,
      "step": 2100
    },
    {
      "entropy": 6.829844617843628,
      "epoch": 0.21042635077722796,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998620211681591,
      "loss": 6.7727,
      "mean_token_accuracy": 0.12200651988387108,
      "num_tokens": 4134091.0,
      "step": 2105
    },
    {
      "entropy": 6.9233245849609375,
      "epoch": 0.21092617583845655,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998607686589239,
      "loss": 6.8436,
      "mean_token_accuracy": 0.11750573813915252,
      "num_tokens": 4143566.0,
      "step": 2110
    },
    {
      "entropy": 6.829675722122192,
      "epoch": 0.2114260008996851,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004998595104922496,
      "loss": 6.8147,
      "mean_token_accuracy": 0.1171520471572876,
      "num_tokens": 4153053.0,
      "step": 2115
    },
    {
      "entropy": 6.852267074584961,
      "epoch": 0.2119258259609137,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004998582466681678,
      "loss": 6.7949,
      "mean_token_accuracy": 0.12671943306922911,
      "num_tokens": 4162925.0,
      "step": 2120
    },
    {
      "entropy": 6.735289525985718,
      "epoch": 0.21242565102214225,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004998569771867105,
      "loss": 6.7167,
      "mean_token_accuracy": 0.12583983838558196,
      "num_tokens": 4172077.0,
      "step": 2125
    },
    {
      "entropy": 6.815008449554443,
      "epoch": 0.2129254760833708,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998557020479095,
      "loss": 6.807,
      "mean_token_accuracy": 0.1230967991054058,
      "num_tokens": 4182134.0,
      "step": 2130
    },
    {
      "entropy": 6.904581451416016,
      "epoch": 0.2134253011445994,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004998544212517968,
      "loss": 6.8459,
      "mean_token_accuracy": 0.11531514376401901,
      "num_tokens": 4191963.0,
      "step": 2135
    },
    {
      "entropy": 6.808885908126831,
      "epoch": 0.21392512620582796,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004998531347984047,
      "loss": 6.7897,
      "mean_token_accuracy": 0.12169457077980042,
      "num_tokens": 4202180.0,
      "step": 2140
    },
    {
      "entropy": 6.8748163223266605,
      "epoch": 0.21442495126705652,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004998518426877656,
      "loss": 6.8107,
      "mean_token_accuracy": 0.12011150643229485,
      "num_tokens": 4211874.0,
      "step": 2145
    },
    {
      "entropy": 6.78511381149292,
      "epoch": 0.2149247763282851,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998505449199122,
      "loss": 6.7715,
      "mean_token_accuracy": 0.11865760311484337,
      "num_tokens": 4221610.0,
      "step": 2150
    },
    {
      "entropy": 6.713295936584473,
      "epoch": 0.21542460138951366,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004998492414948768,
      "loss": 6.7396,
      "mean_token_accuracy": 0.12928089648485183,
      "num_tokens": 4230610.0,
      "step": 2155
    },
    {
      "entropy": 6.8231436252594,
      "epoch": 0.21592442645074225,
      "grad_norm": 0.86328125,
      "learning_rate": 0.0004998479324126924,
      "loss": 6.8422,
      "mean_token_accuracy": 0.11402832642197609,
      "num_tokens": 4241389.0,
      "step": 2160
    },
    {
      "entropy": 6.889860057830811,
      "epoch": 0.2164242515119708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998466176733918,
      "loss": 6.7815,
      "mean_token_accuracy": 0.12243804410099983,
      "num_tokens": 4250617.0,
      "step": 2165
    },
    {
      "entropy": 6.73257622718811,
      "epoch": 0.21692407657319937,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998452972770082,
      "loss": 6.7796,
      "mean_token_accuracy": 0.12635771334171295,
      "num_tokens": 4260000.0,
      "step": 2170
    },
    {
      "entropy": 6.793949365615845,
      "epoch": 0.21742390163442796,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998439712235747,
      "loss": 6.7069,
      "mean_token_accuracy": 0.12508605569601058,
      "num_tokens": 4269751.0,
      "step": 2175
    },
    {
      "entropy": 6.805910682678222,
      "epoch": 0.21792372669565652,
      "grad_norm": 1.125,
      "learning_rate": 0.0004998426395131248,
      "loss": 6.7758,
      "mean_token_accuracy": 0.11987375542521476,
      "num_tokens": 4279210.0,
      "step": 2180
    },
    {
      "entropy": 6.804451131820679,
      "epoch": 0.2184235517568851,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004998413021456919,
      "loss": 6.7386,
      "mean_token_accuracy": 0.12299165651202201,
      "num_tokens": 4288573.0,
      "step": 2185
    },
    {
      "entropy": 6.781830787658691,
      "epoch": 0.21892337681811366,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998399591213098,
      "loss": 6.7641,
      "mean_token_accuracy": 0.11898610666394234,
      "num_tokens": 4298438.0,
      "step": 2190
    },
    {
      "entropy": 6.778064155578614,
      "epoch": 0.21942320187934222,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004998386104400122,
      "loss": 6.6757,
      "mean_token_accuracy": 0.13105980008840562,
      "num_tokens": 4307174.0,
      "step": 2195
    },
    {
      "entropy": 6.902370309829712,
      "epoch": 0.2199230269405708,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998372561018329,
      "loss": 6.8756,
      "mean_token_accuracy": 0.11589000076055526,
      "num_tokens": 4317220.0,
      "step": 2200
    },
    {
      "entropy": 6.813035392761231,
      "epoch": 0.22042285200179937,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998358961068062,
      "loss": 6.7789,
      "mean_token_accuracy": 0.1260307364165783,
      "num_tokens": 4326496.0,
      "step": 2205
    },
    {
      "entropy": 6.771057558059693,
      "epoch": 0.22092267706302793,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998345304549661,
      "loss": 6.7837,
      "mean_token_accuracy": 0.12462224960327148,
      "num_tokens": 4335705.0,
      "step": 2210
    },
    {
      "entropy": 6.854783344268799,
      "epoch": 0.22142250212425652,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998331591463471,
      "loss": 6.816,
      "mean_token_accuracy": 0.12542387172579766,
      "num_tokens": 4344711.0,
      "step": 2215
    },
    {
      "entropy": 6.836872529983521,
      "epoch": 0.22192232718548507,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004998317821809837,
      "loss": 6.823,
      "mean_token_accuracy": 0.12106759324669839,
      "num_tokens": 4355448.0,
      "step": 2220
    },
    {
      "entropy": 6.691077804565429,
      "epoch": 0.22242215224671366,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004998303995589104,
      "loss": 6.6938,
      "mean_token_accuracy": 0.12454739660024643,
      "num_tokens": 4365104.0,
      "step": 2225
    },
    {
      "entropy": 6.707037878036499,
      "epoch": 0.22292197730794222,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004998290112801623,
      "loss": 6.6715,
      "mean_token_accuracy": 0.12494275271892548,
      "num_tokens": 4375219.0,
      "step": 2230
    },
    {
      "entropy": 6.891085195541382,
      "epoch": 0.22342180236917078,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499827617344774,
      "loss": 6.8096,
      "mean_token_accuracy": 0.1162020929157734,
      "num_tokens": 4385294.0,
      "step": 2235
    },
    {
      "entropy": 6.842881155014038,
      "epoch": 0.22392162743039937,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998262177527807,
      "loss": 6.7727,
      "mean_token_accuracy": 0.11394542381167412,
      "num_tokens": 4394305.0,
      "step": 2240
    },
    {
      "entropy": 6.6954185485839846,
      "epoch": 0.22442145249162793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998248125042177,
      "loss": 6.6253,
      "mean_token_accuracy": 0.1304540902376175,
      "num_tokens": 4403732.0,
      "step": 2245
    },
    {
      "entropy": 6.796254825592041,
      "epoch": 0.2249212775528565,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998234015991202,
      "loss": 6.8615,
      "mean_token_accuracy": 0.11681714579463005,
      "num_tokens": 4413730.0,
      "step": 2250
    },
    {
      "entropy": 6.899509286880493,
      "epoch": 0.22542110261408507,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998219850375238,
      "loss": 6.859,
      "mean_token_accuracy": 0.11603142693638802,
      "num_tokens": 4423051.0,
      "step": 2255
    },
    {
      "entropy": 6.866402816772461,
      "epoch": 0.22592092767531363,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004998205628194641,
      "loss": 6.8641,
      "mean_token_accuracy": 0.12003328129649163,
      "num_tokens": 4433363.0,
      "step": 2260
    },
    {
      "entropy": 6.876128149032593,
      "epoch": 0.22642075273654222,
      "grad_norm": 0.8984375,
      "learning_rate": 0.0004998191349449768,
      "loss": 6.8133,
      "mean_token_accuracy": 0.11659364998340607,
      "num_tokens": 4442984.0,
      "step": 2265
    },
    {
      "entropy": 6.778914976119995,
      "epoch": 0.22692057779777078,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998177014140981,
      "loss": 6.7622,
      "mean_token_accuracy": 0.11801078915596008,
      "num_tokens": 4452364.0,
      "step": 2270
    },
    {
      "entropy": 6.881569671630859,
      "epoch": 0.22742040285899934,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998162622268637,
      "loss": 6.7927,
      "mean_token_accuracy": 0.11506244912743568,
      "num_tokens": 4461904.0,
      "step": 2275
    },
    {
      "entropy": 6.80613260269165,
      "epoch": 0.22792022792022792,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004998148173833101,
      "loss": 6.7406,
      "mean_token_accuracy": 0.11940495148301125,
      "num_tokens": 4470633.0,
      "step": 2280
    },
    {
      "entropy": 6.768052959442139,
      "epoch": 0.22842005298145648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004998133668834735,
      "loss": 6.7526,
      "mean_token_accuracy": 0.1232437625527382,
      "num_tokens": 4480673.0,
      "step": 2285
    },
    {
      "entropy": 6.768804597854614,
      "epoch": 0.22891987804268507,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004998119107273905,
      "loss": 6.743,
      "mean_token_accuracy": 0.11722776517271996,
      "num_tokens": 4489923.0,
      "step": 2290
    },
    {
      "entropy": 6.719747304916382,
      "epoch": 0.22941970310391363,
      "grad_norm": 1.0,
      "learning_rate": 0.0004998104489150975,
      "loss": 6.7136,
      "mean_token_accuracy": 0.12137444093823432,
      "num_tokens": 4501007.0,
      "step": 2295
    },
    {
      "entropy": 6.784075307846069,
      "epoch": 0.2299195281651422,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004998089814466318,
      "loss": 6.7367,
      "mean_token_accuracy": 0.11942115277051926,
      "num_tokens": 4510460.0,
      "step": 2300
    },
    {
      "entropy": 6.869019842147827,
      "epoch": 0.23041935322637078,
      "grad_norm": 0.9453125,
      "learning_rate": 0.0004998075083220295,
      "loss": 6.8036,
      "mean_token_accuracy": 0.12088342234492302,
      "num_tokens": 4521026.0,
      "step": 2305
    },
    {
      "entropy": 6.728188514709473,
      "epoch": 0.23091917828759934,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004998060295413286,
      "loss": 6.7208,
      "mean_token_accuracy": 0.12475512400269509,
      "num_tokens": 4530818.0,
      "step": 2310
    },
    {
      "entropy": 6.693582963943482,
      "epoch": 0.23141900334882792,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004998045451045656,
      "loss": 6.6643,
      "mean_token_accuracy": 0.13362600281834602,
      "num_tokens": 4539512.0,
      "step": 2315
    },
    {
      "entropy": 6.876028394699096,
      "epoch": 0.23191882841005648,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004998030550117782,
      "loss": 6.8587,
      "mean_token_accuracy": 0.11190490797162056,
      "num_tokens": 4549030.0,
      "step": 2320
    },
    {
      "entropy": 6.743910264968872,
      "epoch": 0.23241865347128504,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004998015592630037,
      "loss": 6.698,
      "mean_token_accuracy": 0.1262820728123188,
      "num_tokens": 4559347.0,
      "step": 2325
    },
    {
      "entropy": 6.810586643218994,
      "epoch": 0.23291847853251363,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004998000578582798,
      "loss": 6.752,
      "mean_token_accuracy": 0.11423259377479553,
      "num_tokens": 4568721.0,
      "step": 2330
    },
    {
      "entropy": 6.790478038787842,
      "epoch": 0.2334183035937422,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997985507976443,
      "loss": 6.7539,
      "mean_token_accuracy": 0.1210959531366825,
      "num_tokens": 4578265.0,
      "step": 2335
    },
    {
      "entropy": 6.713445091247559,
      "epoch": 0.23391812865497075,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004997970380811351,
      "loss": 6.718,
      "mean_token_accuracy": 0.1265711396932602,
      "num_tokens": 4588694.0,
      "step": 2340
    },
    {
      "entropy": 6.8143064975738525,
      "epoch": 0.23441795371619933,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004997955197087904,
      "loss": 6.7238,
      "mean_token_accuracy": 0.1284726820886135,
      "num_tokens": 4597754.0,
      "step": 2345
    },
    {
      "entropy": 6.756000995635986,
      "epoch": 0.2349177787774279,
      "grad_norm": 1.125,
      "learning_rate": 0.0004997939956806481,
      "loss": 6.7397,
      "mean_token_accuracy": 0.12147602438926697,
      "num_tokens": 4606874.0,
      "step": 2350
    },
    {
      "entropy": 6.706974267959595,
      "epoch": 0.23541760383865648,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997924659967467,
      "loss": 6.623,
      "mean_token_accuracy": 0.12457457929849625,
      "num_tokens": 4615588.0,
      "step": 2355
    },
    {
      "entropy": 6.68896107673645,
      "epoch": 0.23591742889988504,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997909306571248,
      "loss": 6.7797,
      "mean_token_accuracy": 0.12551534697413444,
      "num_tokens": 4625919.0,
      "step": 2360
    },
    {
      "entropy": 6.834859418869018,
      "epoch": 0.2364172539611136,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997893896618209,
      "loss": 6.726,
      "mean_token_accuracy": 0.11857106536626816,
      "num_tokens": 4635539.0,
      "step": 2365
    },
    {
      "entropy": 6.732093048095703,
      "epoch": 0.2369170790223422,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004997878430108738,
      "loss": 6.7191,
      "mean_token_accuracy": 0.12062986120581627,
      "num_tokens": 4645833.0,
      "step": 2370
    },
    {
      "entropy": 6.7105419635772705,
      "epoch": 0.23741690408357075,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004997862907043224,
      "loss": 6.8265,
      "mean_token_accuracy": 0.1232969455420971,
      "num_tokens": 4655166.0,
      "step": 2375
    },
    {
      "entropy": 6.9115959167480465,
      "epoch": 0.23791672914479933,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997847327422059,
      "loss": 6.7108,
      "mean_token_accuracy": 0.12596101090312004,
      "num_tokens": 4664884.0,
      "step": 2380
    },
    {
      "entropy": 6.6541790008544925,
      "epoch": 0.2384165542060279,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997831691245632,
      "loss": 6.7907,
      "mean_token_accuracy": 0.12016936019062996,
      "num_tokens": 4675172.0,
      "step": 2385
    },
    {
      "entropy": 6.848490476608276,
      "epoch": 0.23891637926725645,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997815998514337,
      "loss": 6.7032,
      "mean_token_accuracy": 0.12810124084353447,
      "num_tokens": 4684588.0,
      "step": 2390
    },
    {
      "entropy": 6.729746675491333,
      "epoch": 0.23941620432848504,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997800249228572,
      "loss": 6.816,
      "mean_token_accuracy": 0.12293470725417137,
      "num_tokens": 4694434.0,
      "step": 2395
    },
    {
      "entropy": 6.746816682815552,
      "epoch": 0.2399160293897136,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997784443388732,
      "loss": 6.7548,
      "mean_token_accuracy": 0.12327200025320054,
      "num_tokens": 4705080.0,
      "step": 2400
    },
    {
      "entropy": 6.883273792266846,
      "epoch": 0.24041585445094216,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004997768580995212,
      "loss": 6.8304,
      "mean_token_accuracy": 0.12185543552041053,
      "num_tokens": 4714466.0,
      "step": 2405
    },
    {
      "entropy": 6.736199378967285,
      "epoch": 0.24091567951217074,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997752662048413,
      "loss": 6.7419,
      "mean_token_accuracy": 0.12363232001662254,
      "num_tokens": 4724047.0,
      "step": 2410
    },
    {
      "entropy": 6.775224685668945,
      "epoch": 0.2414155045733993,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997736686548736,
      "loss": 6.7772,
      "mean_token_accuracy": 0.11827057525515557,
      "num_tokens": 4733153.0,
      "step": 2415
    },
    {
      "entropy": 6.771958589553833,
      "epoch": 0.2419153296346279,
      "grad_norm": 0.94140625,
      "learning_rate": 0.0004997720654496582,
      "loss": 6.6351,
      "mean_token_accuracy": 0.1304004967212677,
      "num_tokens": 4743239.0,
      "step": 2420
    },
    {
      "entropy": 6.749918031692505,
      "epoch": 0.24241515469585645,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997704565892355,
      "loss": 6.6749,
      "mean_token_accuracy": 0.12188196331262588,
      "num_tokens": 4753128.0,
      "step": 2425
    },
    {
      "entropy": 6.858197546005249,
      "epoch": 0.242914979757085,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004997688420736459,
      "loss": 6.9063,
      "mean_token_accuracy": 0.10862480849027634,
      "num_tokens": 4764056.0,
      "step": 2430
    },
    {
      "entropy": 6.790316009521485,
      "epoch": 0.2434148048183136,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997672219029301,
      "loss": 6.7419,
      "mean_token_accuracy": 0.12110049352049827,
      "num_tokens": 4773354.0,
      "step": 2435
    },
    {
      "entropy": 6.717658138275146,
      "epoch": 0.24391462987954216,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997655960771288,
      "loss": 6.6883,
      "mean_token_accuracy": 0.1205076090991497,
      "num_tokens": 4783367.0,
      "step": 2440
    },
    {
      "entropy": 6.718193531036377,
      "epoch": 0.24441445494077074,
      "grad_norm": 1.6953125,
      "learning_rate": 0.000499763964596283,
      "loss": 6.5672,
      "mean_token_accuracy": 0.13306122049689292,
      "num_tokens": 4793028.0,
      "step": 2445
    },
    {
      "entropy": 6.765444898605347,
      "epoch": 0.2449142800019993,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997623274604336,
      "loss": 6.7561,
      "mean_token_accuracy": 0.11872197464108467,
      "num_tokens": 4804032.0,
      "step": 2450
    },
    {
      "entropy": 6.793762731552124,
      "epoch": 0.24541410506322786,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000499760684669622,
      "loss": 6.7854,
      "mean_token_accuracy": 0.12456026300787926,
      "num_tokens": 4813294.0,
      "step": 2455
    },
    {
      "entropy": 6.7408191680908205,
      "epoch": 0.24591393012445645,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004997590362238894,
      "loss": 6.6902,
      "mean_token_accuracy": 0.11964407935738564,
      "num_tokens": 4823461.0,
      "step": 2460
    },
    {
      "entropy": 6.703203010559082,
      "epoch": 0.246413755185685,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004997573821232772,
      "loss": 6.7469,
      "mean_token_accuracy": 0.1259793259203434,
      "num_tokens": 4832464.0,
      "step": 2465
    },
    {
      "entropy": 6.755440807342529,
      "epoch": 0.24691358024691357,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004997557223678272,
      "loss": 6.7035,
      "mean_token_accuracy": 0.11913309246301651,
      "num_tokens": 4842697.0,
      "step": 2470
    },
    {
      "entropy": 6.782284879684449,
      "epoch": 0.24741340530814215,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997540569575809,
      "loss": 6.7543,
      "mean_token_accuracy": 0.12437557801604271,
      "num_tokens": 4853129.0,
      "step": 2475
    },
    {
      "entropy": 6.83038649559021,
      "epoch": 0.2479132303693707,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997523858925805,
      "loss": 6.7974,
      "mean_token_accuracy": 0.11816194653511047,
      "num_tokens": 4864048.0,
      "step": 2480
    },
    {
      "entropy": 6.671797800064087,
      "epoch": 0.2484130554305993,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004997507091728678,
      "loss": 6.7311,
      "mean_token_accuracy": 0.12627621069550515,
      "num_tokens": 4874368.0,
      "step": 2485
    },
    {
      "entropy": 6.8561256408691404,
      "epoch": 0.24891288049182786,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997490267984853,
      "loss": 6.7148,
      "mean_token_accuracy": 0.12414158284664153,
      "num_tokens": 4884813.0,
      "step": 2490
    },
    {
      "entropy": 6.699068689346314,
      "epoch": 0.24941270555305642,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997473387694749,
      "loss": 6.7832,
      "mean_token_accuracy": 0.118557108938694,
      "num_tokens": 4895594.0,
      "step": 2495
    },
    {
      "entropy": 6.756707620620728,
      "epoch": 0.249912530614285,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997456450858793,
      "loss": 6.6781,
      "mean_token_accuracy": 0.12389897853136063,
      "num_tokens": 4904922.0,
      "step": 2500
    },
    {
      "entropy": 6.794224119186401,
      "epoch": 0.25041235567551356,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997439457477412,
      "loss": 6.6721,
      "mean_token_accuracy": 0.12629162073135375,
      "num_tokens": 4915009.0,
      "step": 2505
    },
    {
      "entropy": 6.742335557937622,
      "epoch": 0.25091218073674215,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997422407551032,
      "loss": 6.818,
      "mean_token_accuracy": 0.117496757209301,
      "num_tokens": 4924916.0,
      "step": 2510
    },
    {
      "entropy": 6.740524244308472,
      "epoch": 0.2514120057979707,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004997405301080082,
      "loss": 6.7218,
      "mean_token_accuracy": 0.12168776243925095,
      "num_tokens": 4935127.0,
      "step": 2515
    },
    {
      "entropy": 6.776787757873535,
      "epoch": 0.25191183085919927,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004997388138064994,
      "loss": 6.7088,
      "mean_token_accuracy": 0.12240934222936631,
      "num_tokens": 4944749.0,
      "step": 2520
    },
    {
      "entropy": 6.69001955986023,
      "epoch": 0.25241165592042786,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004997370918506198,
      "loss": 6.568,
      "mean_token_accuracy": 0.1336754024028778,
      "num_tokens": 4953868.0,
      "step": 2525
    },
    {
      "entropy": 6.680736446380616,
      "epoch": 0.25291148098165644,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004997353642404129,
      "loss": 6.6791,
      "mean_token_accuracy": 0.12432244718074799,
      "num_tokens": 4963282.0,
      "step": 2530
    },
    {
      "entropy": 6.6588935375213625,
      "epoch": 0.253411306042885,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997336309759219,
      "loss": 6.531,
      "mean_token_accuracy": 0.13633738234639167,
      "num_tokens": 4973099.0,
      "step": 2535
    },
    {
      "entropy": 6.730578756332397,
      "epoch": 0.25391113110411356,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004997318920571907,
      "loss": 6.6989,
      "mean_token_accuracy": 0.11887490078806877,
      "num_tokens": 4983771.0,
      "step": 2540
    },
    {
      "entropy": 6.627429342269897,
      "epoch": 0.25441095616534215,
      "grad_norm": 1.109375,
      "learning_rate": 0.000499730147484263,
      "loss": 6.5654,
      "mean_token_accuracy": 0.12911427095532418,
      "num_tokens": 4993648.0,
      "step": 2545
    },
    {
      "entropy": 6.69837760925293,
      "epoch": 0.2549107812265707,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004997283972571824,
      "loss": 6.7168,
      "mean_token_accuracy": 0.12526629120111465,
      "num_tokens": 5002821.0,
      "step": 2550
    },
    {
      "entropy": 6.832654714584351,
      "epoch": 0.25541060628779927,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997266413759934,
      "loss": 6.7327,
      "mean_token_accuracy": 0.12186189666390419,
      "num_tokens": 5012655.0,
      "step": 2555
    },
    {
      "entropy": 6.656014776229858,
      "epoch": 0.25591043134902786,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004997248798407397,
      "loss": 6.7004,
      "mean_token_accuracy": 0.12474093288183212,
      "num_tokens": 5023649.0,
      "step": 2560
    },
    {
      "entropy": 6.723897504806518,
      "epoch": 0.2564102564102564,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004997231126514659,
      "loss": 6.6708,
      "mean_token_accuracy": 0.13028084263205528,
      "num_tokens": 5031945.0,
      "step": 2565
    },
    {
      "entropy": 6.767000675201416,
      "epoch": 0.256910081471485,
      "grad_norm": 1.734375,
      "learning_rate": 0.0004997213398082165,
      "loss": 6.6893,
      "mean_token_accuracy": 0.12177317440509797,
      "num_tokens": 5042036.0,
      "step": 2570
    },
    {
      "entropy": 6.682856845855713,
      "epoch": 0.25740990653271356,
      "grad_norm": 1.0,
      "learning_rate": 0.000499719561311036,
      "loss": 6.7122,
      "mean_token_accuracy": 0.1242594949901104,
      "num_tokens": 5050986.0,
      "step": 2575
    },
    {
      "entropy": 6.787206935882568,
      "epoch": 0.2579097315939421,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004997177771599691,
      "loss": 6.7099,
      "mean_token_accuracy": 0.1260345883667469,
      "num_tokens": 5061408.0,
      "step": 2580
    },
    {
      "entropy": 6.657704401016235,
      "epoch": 0.2584095566551707,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004997159873550607,
      "loss": 6.6111,
      "mean_token_accuracy": 0.125968037545681,
      "num_tokens": 5071646.0,
      "step": 2585
    },
    {
      "entropy": 6.647315835952758,
      "epoch": 0.25890938171639927,
      "grad_norm": 0.875,
      "learning_rate": 0.0004997141918963561,
      "loss": 6.7173,
      "mean_token_accuracy": 0.12666191905736923,
      "num_tokens": 5083185.0,
      "step": 2590
    },
    {
      "entropy": 6.773238277435302,
      "epoch": 0.25940920677762785,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004997123907839001,
      "loss": 6.6628,
      "mean_token_accuracy": 0.1251784197986126,
      "num_tokens": 5092993.0,
      "step": 2595
    },
    {
      "entropy": 6.78870210647583,
      "epoch": 0.2599090318388564,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004997105840177381,
      "loss": 6.7272,
      "mean_token_accuracy": 0.12102791294455528,
      "num_tokens": 5102709.0,
      "step": 2600
    },
    {
      "entropy": 6.701229047775269,
      "epoch": 0.260408856900085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004997087715979157,
      "loss": 6.6704,
      "mean_token_accuracy": 0.11898972243070602,
      "num_tokens": 5112858.0,
      "step": 2605
    },
    {
      "entropy": 6.76274938583374,
      "epoch": 0.26090868196131356,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004997069535244785,
      "loss": 6.5925,
      "mean_token_accuracy": 0.13014568462967874,
      "num_tokens": 5123573.0,
      "step": 2610
    },
    {
      "entropy": 6.64758710861206,
      "epoch": 0.2614085070225421,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004997051297974722,
      "loss": 6.7694,
      "mean_token_accuracy": 0.12116136848926544,
      "num_tokens": 5132908.0,
      "step": 2615
    },
    {
      "entropy": 6.752505588531494,
      "epoch": 0.2619083320837707,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004997033004169425,
      "loss": 6.6314,
      "mean_token_accuracy": 0.13227704614400865,
      "num_tokens": 5142331.0,
      "step": 2620
    },
    {
      "entropy": 6.808556032180786,
      "epoch": 0.26240815714499927,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004997014653829357,
      "loss": 6.7369,
      "mean_token_accuracy": 0.12015433385968208,
      "num_tokens": 5152517.0,
      "step": 2625
    },
    {
      "entropy": 6.724541759490966,
      "epoch": 0.2629079822062278,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996996246954977,
      "loss": 6.6678,
      "mean_token_accuracy": 0.12828353345394133,
      "num_tokens": 5162418.0,
      "step": 2630
    },
    {
      "entropy": 6.698705959320068,
      "epoch": 0.2634078072674564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996977783546752,
      "loss": 6.69,
      "mean_token_accuracy": 0.1206023059785366,
      "num_tokens": 5171948.0,
      "step": 2635
    },
    {
      "entropy": 6.750772285461426,
      "epoch": 0.26390763232868497,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004996959263605143,
      "loss": 6.6106,
      "mean_token_accuracy": 0.1264311730861664,
      "num_tokens": 5180963.0,
      "step": 2640
    },
    {
      "entropy": 6.683148622512817,
      "epoch": 0.2644074573899135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996940687130617,
      "loss": 6.7276,
      "mean_token_accuracy": 0.12301484346389771,
      "num_tokens": 5192305.0,
      "step": 2645
    },
    {
      "entropy": 6.740709686279297,
      "epoch": 0.2649072824511421,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996922054123641,
      "loss": 6.6007,
      "mean_token_accuracy": 0.1260277345776558,
      "num_tokens": 5201503.0,
      "step": 2650
    },
    {
      "entropy": 6.702259254455567,
      "epoch": 0.2654071075123707,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004996903364584685,
      "loss": 6.5746,
      "mean_token_accuracy": 0.1269563689827919,
      "num_tokens": 5210206.0,
      "step": 2655
    },
    {
      "entropy": 6.542616653442383,
      "epoch": 0.26590693257359926,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004996884618514219,
      "loss": 6.596,
      "mean_token_accuracy": 0.13134641647338868,
      "num_tokens": 5220256.0,
      "step": 2660
    },
    {
      "entropy": 6.699193620681763,
      "epoch": 0.2664067576348278,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004996865815912713,
      "loss": 6.6338,
      "mean_token_accuracy": 0.13359641507267953,
      "num_tokens": 5229128.0,
      "step": 2665
    },
    {
      "entropy": 6.726186513900757,
      "epoch": 0.2669065826960564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996846956780642,
      "loss": 6.5722,
      "mean_token_accuracy": 0.12618556693196298,
      "num_tokens": 5239468.0,
      "step": 2670
    },
    {
      "entropy": 6.633094072341919,
      "epoch": 0.26740640775728497,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499682804111848,
      "loss": 6.6844,
      "mean_token_accuracy": 0.12155252695083618,
      "num_tokens": 5247957.0,
      "step": 2675
    },
    {
      "entropy": 6.73829493522644,
      "epoch": 0.2679062328185135,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996809068926702,
      "loss": 6.6877,
      "mean_token_accuracy": 0.12203190922737121,
      "num_tokens": 5257960.0,
      "step": 2680
    },
    {
      "entropy": 6.618297576904297,
      "epoch": 0.2684060578797421,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996790040205786,
      "loss": 6.538,
      "mean_token_accuracy": 0.1295544274151325,
      "num_tokens": 5266696.0,
      "step": 2685
    },
    {
      "entropy": 6.642504405975342,
      "epoch": 0.2689058829409707,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996770954956211,
      "loss": 6.52,
      "mean_token_accuracy": 0.1335084095597267,
      "num_tokens": 5275619.0,
      "step": 2690
    },
    {
      "entropy": 6.690548467636108,
      "epoch": 0.2694057080021992,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996751813178456,
      "loss": 6.6938,
      "mean_token_accuracy": 0.12373248711228371,
      "num_tokens": 5284884.0,
      "step": 2695
    },
    {
      "entropy": 6.612657642364502,
      "epoch": 0.2699055330634278,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004996732614873005,
      "loss": 6.5796,
      "mean_token_accuracy": 0.13691650182008744,
      "num_tokens": 5293909.0,
      "step": 2700
    },
    {
      "entropy": 6.598701143264771,
      "epoch": 0.2704053581246564,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996713360040339,
      "loss": 6.6494,
      "mean_token_accuracy": 0.12244352400302887,
      "num_tokens": 5303142.0,
      "step": 2705
    },
    {
      "entropy": 6.737385845184326,
      "epoch": 0.27090518318588497,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004996694048680942,
      "loss": 6.595,
      "mean_token_accuracy": 0.12717368975281715,
      "num_tokens": 5313055.0,
      "step": 2710
    },
    {
      "entropy": 6.666927337646484,
      "epoch": 0.2714050082471135,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004996674680795302,
      "loss": 6.709,
      "mean_token_accuracy": 0.11881735101342202,
      "num_tokens": 5323773.0,
      "step": 2715
    },
    {
      "entropy": 6.573076391220093,
      "epoch": 0.2719048333083421,
      "grad_norm": 0.94921875,
      "learning_rate": 0.0004996655256383905,
      "loss": 6.487,
      "mean_token_accuracy": 0.14107831716537475,
      "num_tokens": 5334613.0,
      "step": 2720
    },
    {
      "entropy": 6.6817444324493405,
      "epoch": 0.2724046583695707,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000499663577544724,
      "loss": 6.5659,
      "mean_token_accuracy": 0.13264155611395836,
      "num_tokens": 5344151.0,
      "step": 2725
    },
    {
      "entropy": 6.620177268981934,
      "epoch": 0.2729044834307992,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996616237985796,
      "loss": 6.549,
      "mean_token_accuracy": 0.1309521608054638,
      "num_tokens": 5352425.0,
      "step": 2730
    },
    {
      "entropy": 6.676284456253052,
      "epoch": 0.2734043084920278,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004996596644000066,
      "loss": 6.7542,
      "mean_token_accuracy": 0.12151754051446914,
      "num_tokens": 5361811.0,
      "step": 2735
    },
    {
      "entropy": 6.73824896812439,
      "epoch": 0.2739041335532564,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996576993490543,
      "loss": 6.6959,
      "mean_token_accuracy": 0.1195530690252781,
      "num_tokens": 5371463.0,
      "step": 2740
    },
    {
      "entropy": 6.662166357040405,
      "epoch": 0.2744039586144849,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004996557286457721,
      "loss": 6.5441,
      "mean_token_accuracy": 0.1362849183380604,
      "num_tokens": 5381464.0,
      "step": 2745
    },
    {
      "entropy": 6.587826251983643,
      "epoch": 0.2749037836757135,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004996537522902096,
      "loss": 6.634,
      "mean_token_accuracy": 0.12591350078582764,
      "num_tokens": 5391012.0,
      "step": 2750
    },
    {
      "entropy": 6.7804382801055905,
      "epoch": 0.2754036087369421,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004996517702824163,
      "loss": 6.7322,
      "mean_token_accuracy": 0.12323454171419143,
      "num_tokens": 5401198.0,
      "step": 2755
    },
    {
      "entropy": 6.769994783401489,
      "epoch": 0.2759034337981706,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004996497826224425,
      "loss": 6.6698,
      "mean_token_accuracy": 0.12419450357556343,
      "num_tokens": 5410592.0,
      "step": 2760
    },
    {
      "entropy": 6.562923526763916,
      "epoch": 0.2764032588593992,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996477893103379,
      "loss": 6.6448,
      "mean_token_accuracy": 0.12875826954841613,
      "num_tokens": 5420878.0,
      "step": 2765
    },
    {
      "entropy": 6.72771463394165,
      "epoch": 0.2769030839206278,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996457903461527,
      "loss": 6.643,
      "mean_token_accuracy": 0.12691074088215828,
      "num_tokens": 5431199.0,
      "step": 2770
    },
    {
      "entropy": 6.6997754096984865,
      "epoch": 0.2774029089818564,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004996437857299373,
      "loss": 6.5783,
      "mean_token_accuracy": 0.12888110503554345,
      "num_tokens": 5441661.0,
      "step": 2775
    },
    {
      "entropy": 6.5640599727630615,
      "epoch": 0.2779027340430849,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499641775461742,
      "loss": 6.579,
      "mean_token_accuracy": 0.12949036434292793,
      "num_tokens": 5451863.0,
      "step": 2780
    },
    {
      "entropy": 6.697029161453247,
      "epoch": 0.2784025591043135,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996397595416173,
      "loss": 6.6365,
      "mean_token_accuracy": 0.12379389107227326,
      "num_tokens": 5461600.0,
      "step": 2785
    },
    {
      "entropy": 6.774936103820801,
      "epoch": 0.2789023841655421,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004996377379696143,
      "loss": 6.6492,
      "mean_token_accuracy": 0.12262091562151908,
      "num_tokens": 5470097.0,
      "step": 2790
    },
    {
      "entropy": 6.73469181060791,
      "epoch": 0.2794022092267706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996357107457834,
      "loss": 6.814,
      "mean_token_accuracy": 0.11655468791723252,
      "num_tokens": 5480906.0,
      "step": 2795
    },
    {
      "entropy": 6.808956336975098,
      "epoch": 0.2799020342879992,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499633677870176,
      "loss": 6.6844,
      "mean_token_accuracy": 0.12597069665789604,
      "num_tokens": 5489732.0,
      "step": 2800
    },
    {
      "entropy": 6.528180980682373,
      "epoch": 0.2804018593492278,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996316393428429,
      "loss": 6.4803,
      "mean_token_accuracy": 0.13605385273694992,
      "num_tokens": 5498422.0,
      "step": 2805
    },
    {
      "entropy": 6.613443279266358,
      "epoch": 0.2809016844104563,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004996295951638357,
      "loss": 6.6409,
      "mean_token_accuracy": 0.12768530771136283,
      "num_tokens": 5507958.0,
      "step": 2810
    },
    {
      "entropy": 6.775778770446777,
      "epoch": 0.2814015094716849,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004996275453332057,
      "loss": 6.6654,
      "mean_token_accuracy": 0.1298332244157791,
      "num_tokens": 5516752.0,
      "step": 2815
    },
    {
      "entropy": 6.648133945465088,
      "epoch": 0.2819013345329135,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004996254898510043,
      "loss": 6.6731,
      "mean_token_accuracy": 0.12450689151883125,
      "num_tokens": 5527211.0,
      "step": 2820
    },
    {
      "entropy": 6.654214859008789,
      "epoch": 0.282401159594142,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996234287172835,
      "loss": 6.5656,
      "mean_token_accuracy": 0.12709278836846352,
      "num_tokens": 5536925.0,
      "step": 2825
    },
    {
      "entropy": 6.69185094833374,
      "epoch": 0.2829009846553706,
      "grad_norm": 0.984375,
      "learning_rate": 0.000499621361932095,
      "loss": 6.7517,
      "mean_token_accuracy": 0.12027538269758224,
      "num_tokens": 5546800.0,
      "step": 2830
    },
    {
      "entropy": 6.633594799041748,
      "epoch": 0.2834008097165992,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004996192894954909,
      "loss": 6.5781,
      "mean_token_accuracy": 0.12837735265493394,
      "num_tokens": 5557266.0,
      "step": 2835
    },
    {
      "entropy": 6.729125261306763,
      "epoch": 0.2839006347778278,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004996172114075232,
      "loss": 6.5663,
      "mean_token_accuracy": 0.12667735964059829,
      "num_tokens": 5566169.0,
      "step": 2840
    },
    {
      "entropy": 6.662754964828491,
      "epoch": 0.2844004598390563,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996151276682443,
      "loss": 6.6598,
      "mean_token_accuracy": 0.12701473608613015,
      "num_tokens": 5575566.0,
      "step": 2845
    },
    {
      "entropy": 6.615578985214233,
      "epoch": 0.2849002849002849,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004996130382777065,
      "loss": 6.546,
      "mean_token_accuracy": 0.13053665533661843,
      "num_tokens": 5584804.0,
      "step": 2850
    },
    {
      "entropy": 6.638299226760864,
      "epoch": 0.2854001099615135,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004996109432359625,
      "loss": 6.6009,
      "mean_token_accuracy": 0.1265592433512211,
      "num_tokens": 5594361.0,
      "step": 2855
    },
    {
      "entropy": 6.683030986785889,
      "epoch": 0.285899935022742,
      "grad_norm": 2.0625,
      "learning_rate": 0.0004996088425430651,
      "loss": 6.6219,
      "mean_token_accuracy": 0.12339444905519485,
      "num_tokens": 5605682.0,
      "step": 2860
    },
    {
      "entropy": 6.641170263290405,
      "epoch": 0.2863997600839706,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004996067361990669,
      "loss": 6.6897,
      "mean_token_accuracy": 0.12035760283470154,
      "num_tokens": 5617251.0,
      "step": 2865
    },
    {
      "entropy": 6.7068572521209715,
      "epoch": 0.2868995851451992,
      "grad_norm": 0.96484375,
      "learning_rate": 0.0004996046242040209,
      "loss": 6.5628,
      "mean_token_accuracy": 0.12945612370967866,
      "num_tokens": 5627140.0,
      "step": 2870
    },
    {
      "entropy": 6.550381660461426,
      "epoch": 0.28739941020642773,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004996025065579805,
      "loss": 6.485,
      "mean_token_accuracy": 0.13200978338718414,
      "num_tokens": 5635486.0,
      "step": 2875
    },
    {
      "entropy": 6.645271348953247,
      "epoch": 0.2878992352676563,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004996003832609988,
      "loss": 6.675,
      "mean_token_accuracy": 0.1206593632698059,
      "num_tokens": 5645621.0,
      "step": 2880
    },
    {
      "entropy": 6.811749696731567,
      "epoch": 0.2883990603288849,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004995982543131292,
      "loss": 6.7572,
      "mean_token_accuracy": 0.11400788351893425,
      "num_tokens": 5655626.0,
      "step": 2885
    },
    {
      "entropy": 6.684689092636108,
      "epoch": 0.28889888539011344,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995961197144254,
      "loss": 6.6776,
      "mean_token_accuracy": 0.12474105134606361,
      "num_tokens": 5664547.0,
      "step": 2890
    },
    {
      "entropy": 6.67263617515564,
      "epoch": 0.289398710451342,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499593979464941,
      "loss": 6.6308,
      "mean_token_accuracy": 0.12696049585938454,
      "num_tokens": 5674861.0,
      "step": 2895
    },
    {
      "entropy": 6.643967628479004,
      "epoch": 0.2898985355125706,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004995918335647298,
      "loss": 6.5706,
      "mean_token_accuracy": 0.1286863476037979,
      "num_tokens": 5684143.0,
      "step": 2900
    },
    {
      "entropy": 6.593589925765992,
      "epoch": 0.2903983605737992,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004995896820138459,
      "loss": 6.5715,
      "mean_token_accuracy": 0.12647168189287186,
      "num_tokens": 5693837.0,
      "step": 2905
    },
    {
      "entropy": 6.714412641525269,
      "epoch": 0.29089818563502773,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995875248123434,
      "loss": 6.6512,
      "mean_token_accuracy": 0.11919039636850357,
      "num_tokens": 5704827.0,
      "step": 2910
    },
    {
      "entropy": 6.721566820144654,
      "epoch": 0.2913980106962563,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995853619602766,
      "loss": 6.6848,
      "mean_token_accuracy": 0.12311125099658966,
      "num_tokens": 5714419.0,
      "step": 2915
    },
    {
      "entropy": 6.657805967330932,
      "epoch": 0.2918978357574849,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0004995831934576998,
      "loss": 6.6042,
      "mean_token_accuracy": 0.12744421064853667,
      "num_tokens": 5724983.0,
      "step": 2920
    },
    {
      "entropy": 6.611767673492432,
      "epoch": 0.29239766081871343,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004995810193046677,
      "loss": 6.5242,
      "mean_token_accuracy": 0.13742104694247245,
      "num_tokens": 5734471.0,
      "step": 2925
    },
    {
      "entropy": 6.707096099853516,
      "epoch": 0.292897485879942,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499578839501235,
      "loss": 6.6545,
      "mean_token_accuracy": 0.12848085910081863,
      "num_tokens": 5744801.0,
      "step": 2930
    },
    {
      "entropy": 6.626270914077759,
      "epoch": 0.2933973109411706,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995766540474564,
      "loss": 6.6228,
      "mean_token_accuracy": 0.12304180413484574,
      "num_tokens": 5754267.0,
      "step": 2935
    },
    {
      "entropy": 6.61904239654541,
      "epoch": 0.29389713600239914,
      "grad_norm": 0.921875,
      "learning_rate": 0.000499574462943387,
      "loss": 6.5306,
      "mean_token_accuracy": 0.13505666479468345,
      "num_tokens": 5764466.0,
      "step": 2940
    },
    {
      "entropy": 6.673902082443237,
      "epoch": 0.2943969610636277,
      "grad_norm": 0.92578125,
      "learning_rate": 0.000499572266189082,
      "loss": 6.6548,
      "mean_token_accuracy": 0.12099963128566742,
      "num_tokens": 5775158.0,
      "step": 2945
    },
    {
      "entropy": 6.636359071731567,
      "epoch": 0.2948967861248563,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004995700637845964,
      "loss": 6.4623,
      "mean_token_accuracy": 0.13724422752857207,
      "num_tokens": 5783320.0,
      "step": 2950
    },
    {
      "entropy": 6.554117107391358,
      "epoch": 0.29539661118608485,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004995678557299859,
      "loss": 6.6023,
      "mean_token_accuracy": 0.1300165981054306,
      "num_tokens": 5792980.0,
      "step": 2955
    },
    {
      "entropy": 6.71733250617981,
      "epoch": 0.29589643624731343,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995656420253058,
      "loss": 6.6511,
      "mean_token_accuracy": 0.12479741126298904,
      "num_tokens": 5802338.0,
      "step": 2960
    },
    {
      "entropy": 6.630457496643066,
      "epoch": 0.296396261308542,
      "grad_norm": 1.140625,
      "learning_rate": 0.000499563422670612,
      "loss": 6.554,
      "mean_token_accuracy": 0.12994145452976227,
      "num_tokens": 5811388.0,
      "step": 2965
    },
    {
      "entropy": 6.584511566162109,
      "epoch": 0.2968960863697706,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995611976659604,
      "loss": 6.5898,
      "mean_token_accuracy": 0.127712956815958,
      "num_tokens": 5821108.0,
      "step": 2970
    },
    {
      "entropy": 6.6266556739807125,
      "epoch": 0.29739591143099914,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995589670114067,
      "loss": 6.6609,
      "mean_token_accuracy": 0.11711250171065331,
      "num_tokens": 5831389.0,
      "step": 2975
    },
    {
      "entropy": 6.604189395904541,
      "epoch": 0.2978957364922277,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004995567307070072,
      "loss": 6.4889,
      "mean_token_accuracy": 0.13330937549471855,
      "num_tokens": 5840392.0,
      "step": 2980
    },
    {
      "entropy": 6.632587146759033,
      "epoch": 0.2983955615534563,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004995544887528182,
      "loss": 6.5082,
      "mean_token_accuracy": 0.13582244664430618,
      "num_tokens": 5850032.0,
      "step": 2985
    },
    {
      "entropy": 6.613868570327758,
      "epoch": 0.29889538661468484,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995522411488959,
      "loss": 6.574,
      "mean_token_accuracy": 0.12751223295927047,
      "num_tokens": 5859204.0,
      "step": 2990
    },
    {
      "entropy": 6.610184240341186,
      "epoch": 0.29939521167591343,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004995499878952971,
      "loss": 6.588,
      "mean_token_accuracy": 0.13020774498581886,
      "num_tokens": 5868582.0,
      "step": 2995
    },
    {
      "entropy": 6.630722141265869,
      "epoch": 0.299895036737142,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004995477289920784,
      "loss": 6.5448,
      "mean_token_accuracy": 0.12928271293640137,
      "num_tokens": 5879601.0,
      "step": 3000
    },
    {
      "epoch": 0.299895036737142,
      "eval_entropy": 6.478444874947049,
      "eval_loss": 6.631077289581299,
      "eval_mean_token_accuracy": 0.13178826568691893,
      "eval_num_tokens": 5879601.0,
      "eval_runtime": 23.8125,
      "eval_samples_per_second": 1303.725,
      "eval_steps_per_second": 162.981,
      "step": 3000
    },
    {
      "entropy": 6.6075661182403564,
      "epoch": 0.30039486179837055,
      "grad_norm": 0.953125,
      "learning_rate": 0.0004995454644392967,
      "loss": 6.587,
      "mean_token_accuracy": 0.12043059319257736,
      "num_tokens": 5890255.0,
      "step": 3005
    },
    {
      "entropy": 6.612881708145141,
      "epoch": 0.30089468685959914,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004995431942370087,
      "loss": 6.6179,
      "mean_token_accuracy": 0.13473038971424103,
      "num_tokens": 5899965.0,
      "step": 3010
    },
    {
      "entropy": 6.637569236755371,
      "epoch": 0.3013945119208277,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004995409183852718,
      "loss": 6.5262,
      "mean_token_accuracy": 0.13055912107229234,
      "num_tokens": 5910082.0,
      "step": 3015
    },
    {
      "entropy": 6.558588743209839,
      "epoch": 0.30189433698205626,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995386368841432,
      "loss": 6.6306,
      "mean_token_accuracy": 0.12224628701806069,
      "num_tokens": 5920267.0,
      "step": 3020
    },
    {
      "entropy": 6.679111623764038,
      "epoch": 0.30239416204328484,
      "grad_norm": 1.125,
      "learning_rate": 0.0004995363497336802,
      "loss": 6.609,
      "mean_token_accuracy": 0.130937509983778,
      "num_tokens": 5929275.0,
      "step": 3025
    },
    {
      "entropy": 6.681234884262085,
      "epoch": 0.30289398710451343,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995340569339405,
      "loss": 6.6373,
      "mean_token_accuracy": 0.12426983788609505,
      "num_tokens": 5939305.0,
      "step": 3030
    },
    {
      "entropy": 6.697987699508667,
      "epoch": 0.303393812165742,
      "grad_norm": 1.8515625,
      "learning_rate": 0.0004995317584849817,
      "loss": 6.7081,
      "mean_token_accuracy": 0.11630848348140717,
      "num_tokens": 5950816.0,
      "step": 3035
    },
    {
      "entropy": 6.536932706832886,
      "epoch": 0.30389363722697055,
      "grad_norm": 1.0,
      "learning_rate": 0.0004995294543868615,
      "loss": 6.4835,
      "mean_token_accuracy": 0.13960207998752594,
      "num_tokens": 5959824.0,
      "step": 3040
    },
    {
      "entropy": 6.647433090209961,
      "epoch": 0.30439346228819913,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004995271446396382,
      "loss": 6.5884,
      "mean_token_accuracy": 0.13330790847539903,
      "num_tokens": 5968752.0,
      "step": 3045
    },
    {
      "entropy": 6.600769710540772,
      "epoch": 0.3048932873494277,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995248292433697,
      "loss": 6.5454,
      "mean_token_accuracy": 0.12558241337537765,
      "num_tokens": 5979085.0,
      "step": 3050
    },
    {
      "entropy": 6.625737810134888,
      "epoch": 0.30539311241065625,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0004995225081981142,
      "loss": 6.5855,
      "mean_token_accuracy": 0.12906341180205344,
      "num_tokens": 5990378.0,
      "step": 3055
    },
    {
      "entropy": 6.598832750320435,
      "epoch": 0.30589293747188484,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004995201815039301,
      "loss": 6.4633,
      "mean_token_accuracy": 0.13708525747060776,
      "num_tokens": 5999524.0,
      "step": 3060
    },
    {
      "entropy": 6.562254190444946,
      "epoch": 0.3063927625331134,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004995178491608762,
      "loss": 6.5898,
      "mean_token_accuracy": 0.12790925949811935,
      "num_tokens": 6009928.0,
      "step": 3065
    },
    {
      "entropy": 6.536128902435303,
      "epoch": 0.30689258759434196,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004995155111690107,
      "loss": 6.4702,
      "mean_token_accuracy": 0.13199358731508254,
      "num_tokens": 6019683.0,
      "step": 3070
    },
    {
      "entropy": 6.708814334869385,
      "epoch": 0.30739241265557055,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499513167528393,
      "loss": 6.5986,
      "mean_token_accuracy": 0.12633300423622132,
      "num_tokens": 6030506.0,
      "step": 3075
    },
    {
      "entropy": 6.63917818069458,
      "epoch": 0.30789223771679913,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004995108182390815,
      "loss": 6.5794,
      "mean_token_accuracy": 0.13186729699373245,
      "num_tokens": 6039720.0,
      "step": 3080
    },
    {
      "entropy": 6.478183078765869,
      "epoch": 0.30839206277802766,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995084633011359,
      "loss": 6.4306,
      "mean_token_accuracy": 0.13840741142630578,
      "num_tokens": 6049336.0,
      "step": 3085
    },
    {
      "entropy": 6.526669502258301,
      "epoch": 0.30889188783925625,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000499506102714615,
      "loss": 6.4328,
      "mean_token_accuracy": 0.12964970394968986,
      "num_tokens": 6059420.0,
      "step": 3090
    },
    {
      "entropy": 6.593797445297241,
      "epoch": 0.30939171290048484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004995037364795782,
      "loss": 6.5344,
      "mean_token_accuracy": 0.13578092977404593,
      "num_tokens": 6069035.0,
      "step": 3095
    },
    {
      "entropy": 6.580455732345581,
      "epoch": 0.3098915379617134,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004995013645960853,
      "loss": 6.533,
      "mean_token_accuracy": 0.12749103531241418,
      "num_tokens": 6077863.0,
      "step": 3100
    },
    {
      "entropy": 6.60352201461792,
      "epoch": 0.31039136302294196,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994989870641958,
      "loss": 6.5743,
      "mean_token_accuracy": 0.1299404464662075,
      "num_tokens": 6088929.0,
      "step": 3105
    },
    {
      "entropy": 6.558046865463257,
      "epoch": 0.31089118808417054,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994966038839695,
      "loss": 6.5045,
      "mean_token_accuracy": 0.13678478747606276,
      "num_tokens": 6098673.0,
      "step": 3110
    },
    {
      "entropy": 6.573744106292724,
      "epoch": 0.31139101314539913,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004994942150554666,
      "loss": 6.4069,
      "mean_token_accuracy": 0.1336665131151676,
      "num_tokens": 6107289.0,
      "step": 3115
    },
    {
      "entropy": 6.537808895111084,
      "epoch": 0.31189083820662766,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994918205787468,
      "loss": 6.4508,
      "mean_token_accuracy": 0.13489218950271606,
      "num_tokens": 6117811.0,
      "step": 3120
    },
    {
      "entropy": 6.578155469894409,
      "epoch": 0.31239066326785625,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994894204538706,
      "loss": 6.5531,
      "mean_token_accuracy": 0.12874066233634948,
      "num_tokens": 6127740.0,
      "step": 3125
    },
    {
      "entropy": 6.6143735408782955,
      "epoch": 0.31289048832908484,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004994870146808984,
      "loss": 6.5534,
      "mean_token_accuracy": 0.12975090965628625,
      "num_tokens": 6137253.0,
      "step": 3130
    },
    {
      "entropy": 6.611574411392212,
      "epoch": 0.31339031339031337,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004994846032598906,
      "loss": 6.5853,
      "mean_token_accuracy": 0.12906740829348565,
      "num_tokens": 6147420.0,
      "step": 3135
    },
    {
      "entropy": 6.590312719345093,
      "epoch": 0.31389013845154196,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994821861909082,
      "loss": 6.5257,
      "mean_token_accuracy": 0.13443489223718644,
      "num_tokens": 6156436.0,
      "step": 3140
    },
    {
      "entropy": 6.477894592285156,
      "epoch": 0.31438996351277054,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994797634740114,
      "loss": 6.4677,
      "mean_token_accuracy": 0.12823034226894378,
      "num_tokens": 6165526.0,
      "step": 3145
    },
    {
      "entropy": 6.686942195892334,
      "epoch": 0.3148897885739991,
      "grad_norm": 1.125,
      "learning_rate": 0.0004994773351092618,
      "loss": 6.6826,
      "mean_token_accuracy": 0.12269697338342667,
      "num_tokens": 6176093.0,
      "step": 3150
    },
    {
      "entropy": 6.632466173171997,
      "epoch": 0.31538961363522766,
      "grad_norm": 1.109375,
      "learning_rate": 0.00049947490109672,
      "loss": 6.4976,
      "mean_token_accuracy": 0.13914723843336105,
      "num_tokens": 6185373.0,
      "step": 3155
    },
    {
      "entropy": 6.567219924926758,
      "epoch": 0.31588943869645625,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994724614364476,
      "loss": 6.6106,
      "mean_token_accuracy": 0.12375282943248749,
      "num_tokens": 6196252.0,
      "step": 3160
    },
    {
      "entropy": 6.682858514785766,
      "epoch": 0.31638926375768484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004994700161285058,
      "loss": 6.5178,
      "mean_token_accuracy": 0.13397348448634147,
      "num_tokens": 6206463.0,
      "step": 3165
    },
    {
      "entropy": 6.575518178939819,
      "epoch": 0.31688908881891337,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499467565172956,
      "loss": 6.5176,
      "mean_token_accuracy": 0.13587507605552673,
      "num_tokens": 6216676.0,
      "step": 3170
    },
    {
      "entropy": 6.581781101226807,
      "epoch": 0.31738891388014195,
      "grad_norm": 0.9375,
      "learning_rate": 0.0004994651085698602,
      "loss": 6.5516,
      "mean_token_accuracy": 0.12568040937185287,
      "num_tokens": 6227261.0,
      "step": 3175
    },
    {
      "entropy": 6.595517110824585,
      "epoch": 0.31788873894137054,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00049946264631928,
      "loss": 6.484,
      "mean_token_accuracy": 0.12769982665777208,
      "num_tokens": 6237468.0,
      "step": 3180
    },
    {
      "entropy": 6.529371881484986,
      "epoch": 0.3183885640025991,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004994601784212774,
      "loss": 6.4489,
      "mean_token_accuracy": 0.13162523061037062,
      "num_tokens": 6247164.0,
      "step": 3185
    },
    {
      "entropy": 6.519611978530884,
      "epoch": 0.31888838906382766,
      "grad_norm": 0.95703125,
      "learning_rate": 0.0004994577048759144,
      "loss": 6.54,
      "mean_token_accuracy": 0.13035092502832413,
      "num_tokens": 6257635.0,
      "step": 3190
    },
    {
      "entropy": 6.698423528671265,
      "epoch": 0.31938821412505625,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004994552256832533,
      "loss": 6.6056,
      "mean_token_accuracy": 0.13065118864178657,
      "num_tokens": 6266823.0,
      "step": 3195
    },
    {
      "entropy": 6.6359423160552975,
      "epoch": 0.3198880391862848,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004994527408433566,
      "loss": 6.5154,
      "mean_token_accuracy": 0.13732208013534547,
      "num_tokens": 6276527.0,
      "step": 3200
    },
    {
      "entropy": 6.59774079322815,
      "epoch": 0.32038786424751337,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004994502503562865,
      "loss": 6.5771,
      "mean_token_accuracy": 0.12412868291139603,
      "num_tokens": 6286499.0,
      "step": 3205
    },
    {
      "entropy": 6.558255195617676,
      "epoch": 0.32088768930874195,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000499447754222106,
      "loss": 6.5125,
      "mean_token_accuracy": 0.13360145688056946,
      "num_tokens": 6296550.0,
      "step": 3210
    },
    {
      "entropy": 6.580250597000122,
      "epoch": 0.3213875143699705,
      "grad_norm": 1.0,
      "learning_rate": 0.0004994452524408777,
      "loss": 6.5635,
      "mean_token_accuracy": 0.12742604464292526,
      "num_tokens": 6306951.0,
      "step": 3215
    },
    {
      "entropy": 6.687309217453003,
      "epoch": 0.32188733943119907,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004994427450126646,
      "loss": 6.5762,
      "mean_token_accuracy": 0.1241823248565197,
      "num_tokens": 6316026.0,
      "step": 3220
    },
    {
      "entropy": 6.490858936309815,
      "epoch": 0.32238716449242766,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004994402319375298,
      "loss": 6.52,
      "mean_token_accuracy": 0.13340955302119256,
      "num_tokens": 6325064.0,
      "step": 3225
    },
    {
      "entropy": 6.593712186813354,
      "epoch": 0.32288698955365625,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004994377132155366,
      "loss": 6.5463,
      "mean_token_accuracy": 0.1296512231230736,
      "num_tokens": 6334443.0,
      "step": 3230
    },
    {
      "entropy": 6.64386625289917,
      "epoch": 0.3233868146148848,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004994351888467482,
      "loss": 6.5646,
      "mean_token_accuracy": 0.12805707305669783,
      "num_tokens": 6346103.0,
      "step": 3235
    },
    {
      "entropy": 6.5068731784820555,
      "epoch": 0.32388663967611336,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004994326588312282,
      "loss": 6.5322,
      "mean_token_accuracy": 0.13017322793602942,
      "num_tokens": 6356576.0,
      "step": 3240
    },
    {
      "entropy": 6.642087459564209,
      "epoch": 0.32438646473734195,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004994301231690403,
      "loss": 6.5327,
      "mean_token_accuracy": 0.13048202469944953,
      "num_tokens": 6364796.0,
      "step": 3245
    },
    {
      "entropy": 6.612889957427979,
      "epoch": 0.3248862897985705,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004994275818602482,
      "loss": 6.6128,
      "mean_token_accuracy": 0.12709691599011422,
      "num_tokens": 6374951.0,
      "step": 3250
    },
    {
      "entropy": 6.540645408630371,
      "epoch": 0.32538611485979907,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994250349049159,
      "loss": 6.5089,
      "mean_token_accuracy": 0.1306569382548332,
      "num_tokens": 6384230.0,
      "step": 3255
    },
    {
      "entropy": 6.621853971481324,
      "epoch": 0.32588593992102766,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004994224823031075,
      "loss": 6.492,
      "mean_token_accuracy": 0.13094476982951164,
      "num_tokens": 6394070.0,
      "step": 3260
    },
    {
      "entropy": 6.502137517929077,
      "epoch": 0.3263857649822562,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994199240548873,
      "loss": 6.4865,
      "mean_token_accuracy": 0.13274573534727097,
      "num_tokens": 6403968.0,
      "step": 3265
    },
    {
      "entropy": 6.6335138320922855,
      "epoch": 0.3268855900434848,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004994173601603194,
      "loss": 6.5944,
      "mean_token_accuracy": 0.13025195598602296,
      "num_tokens": 6413254.0,
      "step": 3270
    },
    {
      "entropy": 6.542357492446899,
      "epoch": 0.32738541510471336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004994147906194684,
      "loss": 6.4343,
      "mean_token_accuracy": 0.1355332300066948,
      "num_tokens": 6421266.0,
      "step": 3275
    },
    {
      "entropy": 6.574606847763062,
      "epoch": 0.3278852401659419,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004994122154323992,
      "loss": 6.5638,
      "mean_token_accuracy": 0.12714037150144578,
      "num_tokens": 6429596.0,
      "step": 3280
    },
    {
      "entropy": 6.678783321380616,
      "epoch": 0.3283850652271705,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004994096345991763,
      "loss": 6.5592,
      "mean_token_accuracy": 0.12805289700627326,
      "num_tokens": 6440825.0,
      "step": 3285
    },
    {
      "entropy": 6.474652099609375,
      "epoch": 0.32888489028839907,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994070481198649,
      "loss": 6.5034,
      "mean_token_accuracy": 0.13392820730805396,
      "num_tokens": 6450576.0,
      "step": 3290
    },
    {
      "entropy": 6.600502347946167,
      "epoch": 0.32938471534962765,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004994044559945296,
      "loss": 6.5202,
      "mean_token_accuracy": 0.13730576187372207,
      "num_tokens": 6459673.0,
      "step": 3295
    },
    {
      "entropy": 6.560824251174926,
      "epoch": 0.3298845404108562,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004994018582232361,
      "loss": 6.5073,
      "mean_token_accuracy": 0.1271762117743492,
      "num_tokens": 6469353.0,
      "step": 3300
    },
    {
      "entropy": 6.5393877029418945,
      "epoch": 0.3303843654720848,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004993992548060497,
      "loss": 6.5087,
      "mean_token_accuracy": 0.13126223683357238,
      "num_tokens": 6479610.0,
      "step": 3305
    },
    {
      "entropy": 6.524065208435059,
      "epoch": 0.33088419053331336,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004993966457430357,
      "loss": 6.5624,
      "mean_token_accuracy": 0.13205365911126138,
      "num_tokens": 6488866.0,
      "step": 3310
    },
    {
      "entropy": 6.609571266174316,
      "epoch": 0.3313840155945419,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993940310342598,
      "loss": 6.434,
      "mean_token_accuracy": 0.1394350454211235,
      "num_tokens": 6498396.0,
      "step": 3315
    },
    {
      "entropy": 6.480389213562011,
      "epoch": 0.3318838406557705,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993914106797879,
      "loss": 6.4496,
      "mean_token_accuracy": 0.1399465799331665,
      "num_tokens": 6508347.0,
      "step": 3320
    },
    {
      "entropy": 6.6146210670471195,
      "epoch": 0.33238366571699907,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993887846796858,
      "loss": 6.5599,
      "mean_token_accuracy": 0.12741111367940902,
      "num_tokens": 6518369.0,
      "step": 3325
    },
    {
      "entropy": 6.507545328140258,
      "epoch": 0.3328834907782276,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004993861530340196,
      "loss": 6.5303,
      "mean_token_accuracy": 0.13504743203520775,
      "num_tokens": 6527477.0,
      "step": 3330
    },
    {
      "entropy": 6.551281929016113,
      "epoch": 0.3333833158394562,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004993835157428555,
      "loss": 6.4473,
      "mean_token_accuracy": 0.1366249904036522,
      "num_tokens": 6536271.0,
      "step": 3335
    },
    {
      "entropy": 6.595297527313233,
      "epoch": 0.33388314090068477,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993808728062599,
      "loss": 6.5704,
      "mean_token_accuracy": 0.13260070234537125,
      "num_tokens": 6546295.0,
      "step": 3340
    },
    {
      "entropy": 6.5805864334106445,
      "epoch": 0.3343829659619133,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004993782242242994,
      "loss": 6.4849,
      "mean_token_accuracy": 0.1375689074397087,
      "num_tokens": 6557338.0,
      "step": 3345
    },
    {
      "entropy": 6.52690167427063,
      "epoch": 0.3348827910231419,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993755699970404,
      "loss": 6.496,
      "mean_token_accuracy": 0.13343513533473014,
      "num_tokens": 6567916.0,
      "step": 3350
    },
    {
      "entropy": 6.577276420593262,
      "epoch": 0.3353826160843705,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993729101245498,
      "loss": 6.6029,
      "mean_token_accuracy": 0.1312417969107628,
      "num_tokens": 6577821.0,
      "step": 3355
    },
    {
      "entropy": 6.616403436660766,
      "epoch": 0.33588244114559906,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004993702446068945,
      "loss": 6.4814,
      "mean_token_accuracy": 0.1339896112680435,
      "num_tokens": 6587674.0,
      "step": 3360
    },
    {
      "entropy": 6.524472713470459,
      "epoch": 0.3363822662068276,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993675734441416,
      "loss": 6.5612,
      "mean_token_accuracy": 0.1359415054321289,
      "num_tokens": 6597292.0,
      "step": 3365
    },
    {
      "entropy": 6.487587928771973,
      "epoch": 0.3368820912680562,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993648966363583,
      "loss": 6.4885,
      "mean_token_accuracy": 0.13474380001425743,
      "num_tokens": 6608698.0,
      "step": 3370
    },
    {
      "entropy": 6.596515083312989,
      "epoch": 0.33738191632928477,
      "grad_norm": 1.09375,
      "learning_rate": 0.000499362214183612,
      "loss": 6.485,
      "mean_token_accuracy": 0.12874717712402345,
      "num_tokens": 6617914.0,
      "step": 3375
    },
    {
      "entropy": 6.492099285125732,
      "epoch": 0.3378817413905133,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00049935952608597,
      "loss": 6.4483,
      "mean_token_accuracy": 0.1352960519492626,
      "num_tokens": 6627457.0,
      "step": 3380
    },
    {
      "entropy": 6.613459873199463,
      "epoch": 0.3383815664517419,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993568323435002,
      "loss": 6.4821,
      "mean_token_accuracy": 0.1378979541361332,
      "num_tokens": 6636511.0,
      "step": 3385
    },
    {
      "entropy": 6.553523159027099,
      "epoch": 0.3388813915129705,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004993541329562702,
      "loss": 6.4776,
      "mean_token_accuracy": 0.1300669088959694,
      "num_tokens": 6647028.0,
      "step": 3390
    },
    {
      "entropy": 6.525398588180542,
      "epoch": 0.339381216574199,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004993514279243477,
      "loss": 6.5301,
      "mean_token_accuracy": 0.13475566282868384,
      "num_tokens": 6656407.0,
      "step": 3395
    },
    {
      "entropy": 6.531813669204712,
      "epoch": 0.3398810416354276,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004993487172478012,
      "loss": 6.4775,
      "mean_token_accuracy": 0.13646075278520584,
      "num_tokens": 6665554.0,
      "step": 3400
    },
    {
      "entropy": 6.620145225524903,
      "epoch": 0.3403808666966562,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004993460009266988,
      "loss": 6.5394,
      "mean_token_accuracy": 0.12786303758621215,
      "num_tokens": 6675730.0,
      "step": 3405
    },
    {
      "entropy": 6.488579177856446,
      "epoch": 0.3408806917578847,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004993432789611086,
      "loss": 6.4328,
      "mean_token_accuracy": 0.1401745080947876,
      "num_tokens": 6685871.0,
      "step": 3410
    },
    {
      "entropy": 6.628891897201538,
      "epoch": 0.3413805168191133,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004993405513510992,
      "loss": 6.6577,
      "mean_token_accuracy": 0.12042568475008011,
      "num_tokens": 6697503.0,
      "step": 3415
    },
    {
      "entropy": 6.622269010543823,
      "epoch": 0.3418803418803419,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004993378180967394,
      "loss": 6.46,
      "mean_token_accuracy": 0.13375439420342444,
      "num_tokens": 6708022.0,
      "step": 3420
    },
    {
      "entropy": 6.51132116317749,
      "epoch": 0.3423801669415705,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004993350791980978,
      "loss": 6.5365,
      "mean_token_accuracy": 0.13535891547799112,
      "num_tokens": 6717510.0,
      "step": 3425
    },
    {
      "entropy": 6.572848892211914,
      "epoch": 0.342879992002799,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004993323346552433,
      "loss": 6.5434,
      "mean_token_accuracy": 0.13333892300724984,
      "num_tokens": 6726538.0,
      "step": 3430
    },
    {
      "entropy": 6.582985877990723,
      "epoch": 0.3433798170640276,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004993295844682449,
      "loss": 6.4511,
      "mean_token_accuracy": 0.13790654987096787,
      "num_tokens": 6736750.0,
      "step": 3435
    },
    {
      "entropy": 6.535582828521728,
      "epoch": 0.3438796421252562,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499326828637172,
      "loss": 6.5043,
      "mean_token_accuracy": 0.12834450080990792,
      "num_tokens": 6746366.0,
      "step": 3440
    },
    {
      "entropy": 6.51208758354187,
      "epoch": 0.3443794671864847,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004993240671620938,
      "loss": 6.52,
      "mean_token_accuracy": 0.12855228558182716,
      "num_tokens": 6755887.0,
      "step": 3445
    },
    {
      "entropy": 6.592728948593139,
      "epoch": 0.3448792922477133,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004993213000430797,
      "loss": 6.5489,
      "mean_token_accuracy": 0.13966599181294442,
      "num_tokens": 6764434.0,
      "step": 3450
    },
    {
      "entropy": 6.631072187423706,
      "epoch": 0.3453791173089419,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004993185272801995,
      "loss": 6.5996,
      "mean_token_accuracy": 0.12416719496250153,
      "num_tokens": 6774237.0,
      "step": 3455
    },
    {
      "entropy": 6.655964422225952,
      "epoch": 0.3458789423701704,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993157488735228,
      "loss": 6.468,
      "mean_token_accuracy": 0.13427117168903352,
      "num_tokens": 6784537.0,
      "step": 3460
    },
    {
      "entropy": 6.486367416381836,
      "epoch": 0.346378767431399,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004993129648231197,
      "loss": 6.4571,
      "mean_token_accuracy": 0.1391492322087288,
      "num_tokens": 6794223.0,
      "step": 3465
    },
    {
      "entropy": 6.490914535522461,
      "epoch": 0.3468785924926276,
      "grad_norm": 1.078125,
      "learning_rate": 0.00049931017512906,
      "loss": 6.5119,
      "mean_token_accuracy": 0.12897660657763482,
      "num_tokens": 6803680.0,
      "step": 3470
    },
    {
      "entropy": 6.654724025726319,
      "epoch": 0.3473784175538562,
      "grad_norm": 1.125,
      "learning_rate": 0.0004993073797914141,
      "loss": 6.4942,
      "mean_token_accuracy": 0.12987780347466468,
      "num_tokens": 6812853.0,
      "step": 3475
    },
    {
      "entropy": 6.480592870712281,
      "epoch": 0.3478782426150847,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004993045788102522,
      "loss": 6.5732,
      "mean_token_accuracy": 0.12895409166812896,
      "num_tokens": 6822104.0,
      "step": 3480
    },
    {
      "entropy": 6.586609172821045,
      "epoch": 0.3483780676763133,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004993017721856449,
      "loss": 6.5376,
      "mean_token_accuracy": 0.12887432128190995,
      "num_tokens": 6832683.0,
      "step": 3485
    },
    {
      "entropy": 6.636422252655029,
      "epoch": 0.3488778927375419,
      "grad_norm": 0.93359375,
      "learning_rate": 0.0004992989599176626,
      "loss": 6.5054,
      "mean_token_accuracy": 0.12848403304815292,
      "num_tokens": 6843191.0,
      "step": 3490
    },
    {
      "entropy": 6.525516223907471,
      "epoch": 0.3493777177987704,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992961420063763,
      "loss": 6.5041,
      "mean_token_accuracy": 0.1352040521800518,
      "num_tokens": 6852205.0,
      "step": 3495
    },
    {
      "entropy": 6.4578968524932865,
      "epoch": 0.349877542859999,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992933184518567,
      "loss": 6.4386,
      "mean_token_accuracy": 0.13598101660609246,
      "num_tokens": 6863052.0,
      "step": 3500
    },
    {
      "entropy": 6.557692575454712,
      "epoch": 0.3503773679212276,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992904892541749,
      "loss": 6.5006,
      "mean_token_accuracy": 0.1390739470720291,
      "num_tokens": 6872717.0,
      "step": 3505
    },
    {
      "entropy": 6.481651687622071,
      "epoch": 0.3508771929824561,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004992876544134022,
      "loss": 6.4079,
      "mean_token_accuracy": 0.1390100099146366,
      "num_tokens": 6881351.0,
      "step": 3510
    },
    {
      "entropy": 6.561838531494141,
      "epoch": 0.3513770180436847,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992848139296098,
      "loss": 6.468,
      "mean_token_accuracy": 0.13341921865940093,
      "num_tokens": 6891353.0,
      "step": 3515
    },
    {
      "entropy": 6.578236865997314,
      "epoch": 0.3518768431049133,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992819678028692,
      "loss": 6.5466,
      "mean_token_accuracy": 0.12448048442602158,
      "num_tokens": 6902352.0,
      "step": 3520
    },
    {
      "entropy": 6.596884632110596,
      "epoch": 0.3523766681661418,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499279116033252,
      "loss": 6.5389,
      "mean_token_accuracy": 0.12417580634355545,
      "num_tokens": 6911950.0,
      "step": 3525
    },
    {
      "entropy": 6.52663893699646,
      "epoch": 0.3528764932273704,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992762586208299,
      "loss": 6.4504,
      "mean_token_accuracy": 0.13816978707909583,
      "num_tokens": 6921949.0,
      "step": 3530
    },
    {
      "entropy": 6.525535011291504,
      "epoch": 0.353376318288599,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992733955656749,
      "loss": 6.4712,
      "mean_token_accuracy": 0.13278446346521378,
      "num_tokens": 6932561.0,
      "step": 3535
    },
    {
      "entropy": 6.498945331573486,
      "epoch": 0.3538761433498276,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004992705268678589,
      "loss": 6.4724,
      "mean_token_accuracy": 0.1376880429685116,
      "num_tokens": 6941663.0,
      "step": 3540
    },
    {
      "entropy": 6.5523069381713865,
      "epoch": 0.3543759684110561,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004992676525274542,
      "loss": 6.4837,
      "mean_token_accuracy": 0.12919382229447365,
      "num_tokens": 6952373.0,
      "step": 3545
    },
    {
      "entropy": 6.507160234451294,
      "epoch": 0.3548757934722847,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992647725445331,
      "loss": 6.4961,
      "mean_token_accuracy": 0.13755922242999077,
      "num_tokens": 6961337.0,
      "step": 3550
    },
    {
      "entropy": 6.488829517364502,
      "epoch": 0.3553756185335133,
      "grad_norm": 1.046875,
      "learning_rate": 0.000499261886919168,
      "loss": 6.4398,
      "mean_token_accuracy": 0.14278729483485222,
      "num_tokens": 6971649.0,
      "step": 3555
    },
    {
      "entropy": 6.5226733684539795,
      "epoch": 0.3558754435947418,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004992589956514316,
      "loss": 6.4722,
      "mean_token_accuracy": 0.13939098119735718,
      "num_tokens": 6982261.0,
      "step": 3560
    },
    {
      "entropy": 6.550300359725952,
      "epoch": 0.3563752686559704,
      "grad_norm": 1.125,
      "learning_rate": 0.0004992560987413965,
      "loss": 6.4454,
      "mean_token_accuracy": 0.13285200521349907,
      "num_tokens": 6991324.0,
      "step": 3565
    },
    {
      "entropy": 6.43190975189209,
      "epoch": 0.356875093717199,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004992531961891356,
      "loss": 6.4296,
      "mean_token_accuracy": 0.13729097843170165,
      "num_tokens": 7001396.0,
      "step": 3570
    },
    {
      "entropy": 6.545795249938965,
      "epoch": 0.35737491877842753,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004992502879947222,
      "loss": 6.4363,
      "mean_token_accuracy": 0.1328266330063343,
      "num_tokens": 7010354.0,
      "step": 3575
    },
    {
      "entropy": 6.578396415710449,
      "epoch": 0.3578747438396561,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004992473741582291,
      "loss": 6.5387,
      "mean_token_accuracy": 0.13147649839520453,
      "num_tokens": 7019835.0,
      "step": 3580
    },
    {
      "entropy": 6.489585638046265,
      "epoch": 0.3583745689008847,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992444546797298,
      "loss": 6.4135,
      "mean_token_accuracy": 0.136285550147295,
      "num_tokens": 7030515.0,
      "step": 3585
    },
    {
      "entropy": 6.441481447219848,
      "epoch": 0.35887439396211324,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992415295592977,
      "loss": 6.4773,
      "mean_token_accuracy": 0.1365040898323059,
      "num_tokens": 7041248.0,
      "step": 3590
    },
    {
      "entropy": 6.587574148178101,
      "epoch": 0.3593742190233418,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004992385987970064,
      "loss": 6.5043,
      "mean_token_accuracy": 0.12980234399437904,
      "num_tokens": 7050578.0,
      "step": 3595
    },
    {
      "entropy": 6.600696229934693,
      "epoch": 0.3598740440845704,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004992356623929297,
      "loss": 6.4912,
      "mean_token_accuracy": 0.13227495476603507,
      "num_tokens": 7060270.0,
      "step": 3600
    },
    {
      "entropy": 6.554751539230347,
      "epoch": 0.360373869145799,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004992327203471414,
      "loss": 6.5518,
      "mean_token_accuracy": 0.12837831005454065,
      "num_tokens": 7070259.0,
      "step": 3605
    },
    {
      "entropy": 6.471230411529541,
      "epoch": 0.36087369420702753,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992297726597156,
      "loss": 6.4446,
      "mean_token_accuracy": 0.13782778307795523,
      "num_tokens": 7078784.0,
      "step": 3610
    },
    {
      "entropy": 6.57683367729187,
      "epoch": 0.3613735192682561,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004992268193307264,
      "loss": 6.5054,
      "mean_token_accuracy": 0.12678046450018882,
      "num_tokens": 7088632.0,
      "step": 3615
    },
    {
      "entropy": 6.5138507843017575,
      "epoch": 0.3618733443294847,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004992238603602481,
      "loss": 6.4903,
      "mean_token_accuracy": 0.1330701433122158,
      "num_tokens": 7097703.0,
      "step": 3620
    },
    {
      "entropy": 6.4704427242279055,
      "epoch": 0.36237316939071323,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004992208957483552,
      "loss": 6.4171,
      "mean_token_accuracy": 0.1354508228600025,
      "num_tokens": 7107183.0,
      "step": 3625
    },
    {
      "entropy": 6.619330167770386,
      "epoch": 0.3628729944519418,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004992179254951223,
      "loss": 6.5165,
      "mean_token_accuracy": 0.12851681262254716,
      "num_tokens": 7118220.0,
      "step": 3630
    },
    {
      "entropy": 6.5486711978912355,
      "epoch": 0.3633728195131704,
      "grad_norm": 1.015625,
      "learning_rate": 0.000499214949600624,
      "loss": 6.46,
      "mean_token_accuracy": 0.13207071423530578,
      "num_tokens": 7128322.0,
      "step": 3635
    },
    {
      "entropy": 6.380391597747803,
      "epoch": 0.36387264457439894,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004992119680649355,
      "loss": 6.472,
      "mean_token_accuracy": 0.13622428625822067,
      "num_tokens": 7137284.0,
      "step": 3640
    },
    {
      "entropy": 6.485237312316895,
      "epoch": 0.3643724696356275,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004992089808881313,
      "loss": 6.4123,
      "mean_token_accuracy": 0.1392784297466278,
      "num_tokens": 7146598.0,
      "step": 3645
    },
    {
      "entropy": 6.69722695350647,
      "epoch": 0.3648722946968561,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004992059880702869,
      "loss": 6.5599,
      "mean_token_accuracy": 0.13605788052082063,
      "num_tokens": 7156816.0,
      "step": 3650
    },
    {
      "entropy": 6.540158987045288,
      "epoch": 0.36537211975808465,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004992029896114775,
      "loss": 6.4996,
      "mean_token_accuracy": 0.12887731418013573,
      "num_tokens": 7166855.0,
      "step": 3655
    },
    {
      "entropy": 6.412228345870972,
      "epoch": 0.36587194481931323,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991999855117787,
      "loss": 6.3789,
      "mean_token_accuracy": 0.1410156346857548,
      "num_tokens": 7176477.0,
      "step": 3660
    },
    {
      "entropy": 6.535949945449829,
      "epoch": 0.3663717698805418,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004991969757712659,
      "loss": 6.4268,
      "mean_token_accuracy": 0.13362221270799637,
      "num_tokens": 7186173.0,
      "step": 3665
    },
    {
      "entropy": 6.410127258300781,
      "epoch": 0.3668715949417704,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004991939603900149,
      "loss": 6.3617,
      "mean_token_accuracy": 0.13872723802924156,
      "num_tokens": 7195403.0,
      "step": 3670
    },
    {
      "entropy": 6.516451835632324,
      "epoch": 0.36737142000299894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991909393681015,
      "loss": 6.4833,
      "mean_token_accuracy": 0.13445534855127333,
      "num_tokens": 7204179.0,
      "step": 3675
    },
    {
      "entropy": 6.471215152740479,
      "epoch": 0.3678712450642275,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991879127056017,
      "loss": 6.4264,
      "mean_token_accuracy": 0.13383713141083717,
      "num_tokens": 7214159.0,
      "step": 3680
    },
    {
      "entropy": 6.462945604324341,
      "epoch": 0.3683710701254561,
      "grad_norm": 0.97265625,
      "learning_rate": 0.0004991848804025918,
      "loss": 6.3955,
      "mean_token_accuracy": 0.14022568762302398,
      "num_tokens": 7224410.0,
      "step": 3685
    },
    {
      "entropy": 6.456934833526612,
      "epoch": 0.36887089518668464,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000499181842459148,
      "loss": 6.4565,
      "mean_token_accuracy": 0.14114324226975442,
      "num_tokens": 7233740.0,
      "step": 3690
    },
    {
      "entropy": 6.5373962879180905,
      "epoch": 0.36937072024791323,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004991787988753467,
      "loss": 6.4082,
      "mean_token_accuracy": 0.13846576884388923,
      "num_tokens": 7242834.0,
      "step": 3695
    },
    {
      "entropy": 6.4727240085601805,
      "epoch": 0.3698705453091418,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991757496512645,
      "loss": 6.3856,
      "mean_token_accuracy": 0.1420457072556019,
      "num_tokens": 7252098.0,
      "step": 3700
    },
    {
      "entropy": 6.597651672363281,
      "epoch": 0.37037037037037035,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991726947869782,
      "loss": 6.5514,
      "mean_token_accuracy": 0.1257934272289276,
      "num_tokens": 7261396.0,
      "step": 3705
    },
    {
      "entropy": 6.529380989074707,
      "epoch": 0.37087019543159894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004991696342825645,
      "loss": 6.4953,
      "mean_token_accuracy": 0.127850254625082,
      "num_tokens": 7271138.0,
      "step": 3710
    },
    {
      "entropy": 6.470937824249267,
      "epoch": 0.3713700204928275,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991665681381005,
      "loss": 6.4192,
      "mean_token_accuracy": 0.13879003301262854,
      "num_tokens": 7280164.0,
      "step": 3715
    },
    {
      "entropy": 6.558149003982544,
      "epoch": 0.37186984555405606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991634963536633,
      "loss": 6.4776,
      "mean_token_accuracy": 0.13312826305627823,
      "num_tokens": 7288830.0,
      "step": 3720
    },
    {
      "entropy": 6.520046949386597,
      "epoch": 0.37236967061528464,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004991604189293303,
      "loss": 6.5057,
      "mean_token_accuracy": 0.13286127969622613,
      "num_tokens": 7298955.0,
      "step": 3725
    },
    {
      "entropy": 6.541933012008667,
      "epoch": 0.37286949567651323,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991573358651787,
      "loss": 6.4104,
      "mean_token_accuracy": 0.13342548981308938,
      "num_tokens": 7309829.0,
      "step": 3730
    },
    {
      "entropy": 6.539558410644531,
      "epoch": 0.3733693207377418,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991542471612864,
      "loss": 6.5025,
      "mean_token_accuracy": 0.1323994792997837,
      "num_tokens": 7319654.0,
      "step": 3735
    },
    {
      "entropy": 6.572074699401855,
      "epoch": 0.37386914579897035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991511528177307,
      "loss": 6.4759,
      "mean_token_accuracy": 0.13173828050494193,
      "num_tokens": 7330415.0,
      "step": 3740
    },
    {
      "entropy": 6.542703151702881,
      "epoch": 0.37436897086019894,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004991480528345898,
      "loss": 6.4861,
      "mean_token_accuracy": 0.13387491255998613,
      "num_tokens": 7339511.0,
      "step": 3745
    },
    {
      "entropy": 6.603790760040283,
      "epoch": 0.3748687959214275,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004991449472119416,
      "loss": 6.4908,
      "mean_token_accuracy": 0.12859921902418137,
      "num_tokens": 7348488.0,
      "step": 3750
    },
    {
      "entropy": 6.414752101898193,
      "epoch": 0.37536862098265605,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004991418359498642,
      "loss": 6.4357,
      "mean_token_accuracy": 0.13227830901741983,
      "num_tokens": 7357804.0,
      "step": 3755
    },
    {
      "entropy": 6.497383403778076,
      "epoch": 0.37586844604388464,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004991387190484358,
      "loss": 6.4864,
      "mean_token_accuracy": 0.13240752145648002,
      "num_tokens": 7367644.0,
      "step": 3760
    },
    {
      "entropy": 6.56132402420044,
      "epoch": 0.37636827110511323,
      "grad_norm": 1.1875,
      "learning_rate": 0.000499135596507735,
      "loss": 6.3185,
      "mean_token_accuracy": 0.13788421526551248,
      "num_tokens": 7377357.0,
      "step": 3765
    },
    {
      "entropy": 6.511699247360229,
      "epoch": 0.37686809616634176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004991324683278402,
      "loss": 6.531,
      "mean_token_accuracy": 0.12990649566054344,
      "num_tokens": 7387306.0,
      "step": 3770
    },
    {
      "entropy": 6.502481603622437,
      "epoch": 0.37736792122757035,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991293345088301,
      "loss": 6.4699,
      "mean_token_accuracy": 0.1305549792945385,
      "num_tokens": 7397472.0,
      "step": 3775
    },
    {
      "entropy": 6.5302571773529055,
      "epoch": 0.37786774628879893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991261950507837,
      "loss": 6.488,
      "mean_token_accuracy": 0.135041344165802,
      "num_tokens": 7408786.0,
      "step": 3780
    },
    {
      "entropy": 6.5253747463226315,
      "epoch": 0.37836757135002747,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00049912304995378,
      "loss": 6.4688,
      "mean_token_accuracy": 0.12599104791879653,
      "num_tokens": 7417857.0,
      "step": 3785
    },
    {
      "entropy": 6.481985187530517,
      "epoch": 0.37886739641125605,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004991198992178979,
      "loss": 6.4521,
      "mean_token_accuracy": 0.1293245181441307,
      "num_tokens": 7427007.0,
      "step": 3790
    },
    {
      "entropy": 6.493896198272705,
      "epoch": 0.37936722147248464,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004991167428432168,
      "loss": 6.3763,
      "mean_token_accuracy": 0.13669300004839896,
      "num_tokens": 7437090.0,
      "step": 3795
    },
    {
      "entropy": 6.466151142120362,
      "epoch": 0.3798670465337132,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004991135808298162,
      "loss": 6.4084,
      "mean_token_accuracy": 0.1375203974545002,
      "num_tokens": 7446249.0,
      "step": 3800
    },
    {
      "entropy": 6.564642429351807,
      "epoch": 0.38036687159494176,
      "grad_norm": 1.25,
      "learning_rate": 0.0004991104131777755,
      "loss": 6.447,
      "mean_token_accuracy": 0.13011155053973197,
      "num_tokens": 7455838.0,
      "step": 3805
    },
    {
      "entropy": 6.494140672683716,
      "epoch": 0.38086669665617034,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004991072398871746,
      "loss": 6.498,
      "mean_token_accuracy": 0.12991281002759933,
      "num_tokens": 7466057.0,
      "step": 3810
    },
    {
      "entropy": 6.531226396560669,
      "epoch": 0.38136652171739893,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004991040609580932,
      "loss": 6.4806,
      "mean_token_accuracy": 0.1331527642905712,
      "num_tokens": 7475174.0,
      "step": 3815
    },
    {
      "entropy": 6.542899227142334,
      "epoch": 0.38186634677862746,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004991008763906112,
      "loss": 6.4009,
      "mean_token_accuracy": 0.1381096974015236,
      "num_tokens": 7484635.0,
      "step": 3820
    },
    {
      "entropy": 6.417492961883545,
      "epoch": 0.38236617183985605,
      "grad_norm": 0.9609375,
      "learning_rate": 0.0004990976861848087,
      "loss": 6.4668,
      "mean_token_accuracy": 0.12807572260499,
      "num_tokens": 7496428.0,
      "step": 3825
    },
    {
      "entropy": 6.520355558395385,
      "epoch": 0.38286599690108464,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004990944903407663,
      "loss": 6.4983,
      "mean_token_accuracy": 0.13231004774570465,
      "num_tokens": 7506581.0,
      "step": 3830
    },
    {
      "entropy": 6.527870559692383,
      "epoch": 0.38336582196231317,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990912888585641,
      "loss": 6.361,
      "mean_token_accuracy": 0.13924233615398407,
      "num_tokens": 7515809.0,
      "step": 3835
    },
    {
      "entropy": 6.356991004943848,
      "epoch": 0.38386564702354176,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990880817382828,
      "loss": 6.3978,
      "mean_token_accuracy": 0.13598079830408097,
      "num_tokens": 7525800.0,
      "step": 3840
    },
    {
      "entropy": 6.503269243240356,
      "epoch": 0.38436547208477034,
      "grad_norm": 1.0,
      "learning_rate": 0.000499084868980003,
      "loss": 6.5018,
      "mean_token_accuracy": 0.13475804030895233,
      "num_tokens": 7536903.0,
      "step": 3845
    },
    {
      "entropy": 6.553870105743409,
      "epoch": 0.3848652971459989,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004990816505838054,
      "loss": 6.4857,
      "mean_token_accuracy": 0.13236310184001923,
      "num_tokens": 7546004.0,
      "step": 3850
    },
    {
      "entropy": 6.550846672058105,
      "epoch": 0.38536512220722746,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004990784265497713,
      "loss": 6.59,
      "mean_token_accuracy": 0.129102224111557,
      "num_tokens": 7555971.0,
      "step": 3855
    },
    {
      "entropy": 6.625053215026855,
      "epoch": 0.38586494726845605,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004990751968779817,
      "loss": 6.4849,
      "mean_token_accuracy": 0.13418882042169572,
      "num_tokens": 7564569.0,
      "step": 3860
    },
    {
      "entropy": 6.4894555568695065,
      "epoch": 0.38636477232968464,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990719615685176,
      "loss": 6.4665,
      "mean_token_accuracy": 0.13154759109020234,
      "num_tokens": 7574134.0,
      "step": 3865
    },
    {
      "entropy": 6.52254810333252,
      "epoch": 0.38686459739091317,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990687206214608,
      "loss": 6.344,
      "mean_token_accuracy": 0.13741969615221022,
      "num_tokens": 7583631.0,
      "step": 3870
    },
    {
      "entropy": 6.379374217987061,
      "epoch": 0.38736442245214175,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990654740368925,
      "loss": 6.3054,
      "mean_token_accuracy": 0.14057383686304092,
      "num_tokens": 7593281.0,
      "step": 3875
    },
    {
      "entropy": 6.375336456298828,
      "epoch": 0.38786424751337034,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004990622218148947,
      "loss": 6.4609,
      "mean_token_accuracy": 0.13463458716869353,
      "num_tokens": 7603735.0,
      "step": 3880
    },
    {
      "entropy": 6.54819622039795,
      "epoch": 0.3883640725745989,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004990589639555489,
      "loss": 6.4262,
      "mean_token_accuracy": 0.13453493192791938,
      "num_tokens": 7613126.0,
      "step": 3885
    },
    {
      "entropy": 6.562527799606324,
      "epoch": 0.38886389763582746,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004990557004589373,
      "loss": 6.4532,
      "mean_token_accuracy": 0.13126180320978165,
      "num_tokens": 7623191.0,
      "step": 3890
    },
    {
      "entropy": 6.5247735500335695,
      "epoch": 0.38936372269705605,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004990524313251419,
      "loss": 6.4315,
      "mean_token_accuracy": 0.13880911991000175,
      "num_tokens": 7634118.0,
      "step": 3895
    },
    {
      "entropy": 6.420445775985717,
      "epoch": 0.3898635477582846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004990491565542449,
      "loss": 6.3196,
      "mean_token_accuracy": 0.14285002946853637,
      "num_tokens": 7643228.0,
      "step": 3900
    },
    {
      "entropy": 6.43741888999939,
      "epoch": 0.39036337281951317,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990458761463289,
      "loss": 6.4192,
      "mean_token_accuracy": 0.13168829903006554,
      "num_tokens": 7652578.0,
      "step": 3905
    },
    {
      "entropy": 6.491273593902588,
      "epoch": 0.39086319788074175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990425901014763,
      "loss": 6.388,
      "mean_token_accuracy": 0.13412820547819138,
      "num_tokens": 7661617.0,
      "step": 3910
    },
    {
      "entropy": 6.399303293228149,
      "epoch": 0.3913630229419703,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004990392984197697,
      "loss": 6.3904,
      "mean_token_accuracy": 0.14149613678455353,
      "num_tokens": 7670864.0,
      "step": 3915
    },
    {
      "entropy": 6.4511185646057125,
      "epoch": 0.39186284800319887,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004990360011012921,
      "loss": 6.4521,
      "mean_token_accuracy": 0.12870443016290664,
      "num_tokens": 7681528.0,
      "step": 3920
    },
    {
      "entropy": 6.505901193618774,
      "epoch": 0.39236267306442746,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004990326981461264,
      "loss": 6.3813,
      "mean_token_accuracy": 0.13232036978006362,
      "num_tokens": 7690398.0,
      "step": 3925
    },
    {
      "entropy": 6.40734224319458,
      "epoch": 0.39286249812565605,
      "grad_norm": 1.125,
      "learning_rate": 0.0004990293895543555,
      "loss": 6.4135,
      "mean_token_accuracy": 0.14027423188090324,
      "num_tokens": 7699843.0,
      "step": 3930
    },
    {
      "entropy": 6.431153154373169,
      "epoch": 0.3933623231868846,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499026075326063,
      "loss": 6.449,
      "mean_token_accuracy": 0.13698539435863494,
      "num_tokens": 7710181.0,
      "step": 3935
    },
    {
      "entropy": 6.489415884017944,
      "epoch": 0.39386214824811316,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000499022755461332,
      "loss": 6.3745,
      "mean_token_accuracy": 0.1336131788790226,
      "num_tokens": 7719148.0,
      "step": 3940
    },
    {
      "entropy": 6.326394557952881,
      "epoch": 0.39436197330934175,
      "grad_norm": 1.078125,
      "learning_rate": 0.000499019429960246,
      "loss": 6.2553,
      "mean_token_accuracy": 0.14362635761499404,
      "num_tokens": 7728551.0,
      "step": 3945
    },
    {
      "entropy": 6.501528358459472,
      "epoch": 0.3948617983705703,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004990160988228888,
      "loss": 6.4779,
      "mean_token_accuracy": 0.13136930465698243,
      "num_tokens": 7739083.0,
      "step": 3950
    },
    {
      "entropy": 6.5596575260162355,
      "epoch": 0.39536162343179887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004990127620493444,
      "loss": 6.4612,
      "mean_token_accuracy": 0.1364016979932785,
      "num_tokens": 7749684.0,
      "step": 3955
    },
    {
      "entropy": 6.395131587982178,
      "epoch": 0.39586144849302746,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004990094196396964,
      "loss": 6.3313,
      "mean_token_accuracy": 0.13728146478533745,
      "num_tokens": 7759281.0,
      "step": 3960
    },
    {
      "entropy": 6.47342586517334,
      "epoch": 0.396361273554256,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000499006071594029,
      "loss": 6.4174,
      "mean_token_accuracy": 0.1372039444744587,
      "num_tokens": 7768612.0,
      "step": 3965
    },
    {
      "entropy": 6.423118972778321,
      "epoch": 0.3968610986154846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004990027179124266,
      "loss": 6.3373,
      "mean_token_accuracy": 0.13924752101302146,
      "num_tokens": 7778333.0,
      "step": 3970
    },
    {
      "entropy": 6.506185293197632,
      "epoch": 0.39736092367671316,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989993585949733,
      "loss": 6.4849,
      "mean_token_accuracy": 0.12784129306674002,
      "num_tokens": 7787547.0,
      "step": 3975
    },
    {
      "entropy": 6.436396360397339,
      "epoch": 0.3978607487379417,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004989959936417541,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13556910753250123,
      "num_tokens": 7798407.0,
      "step": 3980
    },
    {
      "entropy": 6.452073287963867,
      "epoch": 0.3983605737991703,
      "grad_norm": 0.99609375,
      "learning_rate": 0.000498992623052853,
      "loss": 6.3615,
      "mean_token_accuracy": 0.14274941980838776,
      "num_tokens": 7807879.0,
      "step": 3985
    },
    {
      "entropy": 6.414194822311401,
      "epoch": 0.39886039886039887,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989892468283553,
      "loss": 6.3245,
      "mean_token_accuracy": 0.13929423540830613,
      "num_tokens": 7817210.0,
      "step": 3990
    },
    {
      "entropy": 6.5941883563995365,
      "epoch": 0.39936022392162746,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004989858649683457,
      "loss": 6.5686,
      "mean_token_accuracy": 0.12396848574280739,
      "num_tokens": 7827845.0,
      "step": 3995
    },
    {
      "entropy": 6.6080930709838865,
      "epoch": 0.399860048982856,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989824774729094,
      "loss": 6.4333,
      "mean_token_accuracy": 0.1379890486598015,
      "num_tokens": 7836994.0,
      "step": 4000
    },
    {
      "entropy": 6.467769432067871,
      "epoch": 0.4003598740440846,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989790843421317,
      "loss": 6.4582,
      "mean_token_accuracy": 0.13288121446967124,
      "num_tokens": 7846733.0,
      "step": 4005
    },
    {
      "entropy": 6.450557899475098,
      "epoch": 0.40085969910531316,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004989756855760979,
      "loss": 6.3774,
      "mean_token_accuracy": 0.14024601578712464,
      "num_tokens": 7856393.0,
      "step": 4010
    },
    {
      "entropy": 6.504327917098999,
      "epoch": 0.4013595241665417,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989722811748934,
      "loss": 6.4681,
      "mean_token_accuracy": 0.13493888974189758,
      "num_tokens": 7866394.0,
      "step": 4015
    },
    {
      "entropy": 6.530735111236572,
      "epoch": 0.4018593492277703,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004989688711386039,
      "loss": 6.4742,
      "mean_token_accuracy": 0.1269529640674591,
      "num_tokens": 7877084.0,
      "step": 4020
    },
    {
      "entropy": 6.575013875961304,
      "epoch": 0.40235917428899887,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004989654554673152,
      "loss": 6.4765,
      "mean_token_accuracy": 0.13267083317041398,
      "num_tokens": 7887289.0,
      "step": 4025
    },
    {
      "entropy": 6.367003679275513,
      "epoch": 0.4028589993502274,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004989620341611133,
      "loss": 6.4269,
      "mean_token_accuracy": 0.13722112998366356,
      "num_tokens": 7897315.0,
      "step": 4030
    },
    {
      "entropy": 6.603945541381836,
      "epoch": 0.403358824411456,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004989586072200843,
      "loss": 6.5344,
      "mean_token_accuracy": 0.12967253774404525,
      "num_tokens": 7905913.0,
      "step": 4035
    },
    {
      "entropy": 6.444837760925293,
      "epoch": 0.40385864947268457,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004989551746443143,
      "loss": 6.3368,
      "mean_token_accuracy": 0.14095348194241525,
      "num_tokens": 7914879.0,
      "step": 4040
    },
    {
      "entropy": 6.448266172409058,
      "epoch": 0.4043584745339131,
      "grad_norm": 1.125,
      "learning_rate": 0.0004989517364338899,
      "loss": 6.4821,
      "mean_token_accuracy": 0.13294394686818123,
      "num_tokens": 7925163.0,
      "step": 4045
    },
    {
      "entropy": 6.587580680847168,
      "epoch": 0.4048582995951417,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989482925888974,
      "loss": 6.3872,
      "mean_token_accuracy": 0.13814545199275016,
      "num_tokens": 7934380.0,
      "step": 4050
    },
    {
      "entropy": 6.46042218208313,
      "epoch": 0.4053581246563703,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004989448431094233,
      "loss": 6.4474,
      "mean_token_accuracy": 0.1337108075618744,
      "num_tokens": 7944600.0,
      "step": 4055
    },
    {
      "entropy": 6.349003171920776,
      "epoch": 0.40585794971759886,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004989413879955548,
      "loss": 6.354,
      "mean_token_accuracy": 0.14133601263165474,
      "num_tokens": 7953545.0,
      "step": 4060
    },
    {
      "entropy": 6.475251865386963,
      "epoch": 0.4063577747788274,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989379272473785,
      "loss": 6.4245,
      "mean_token_accuracy": 0.1339748039841652,
      "num_tokens": 7962931.0,
      "step": 4065
    },
    {
      "entropy": 6.523327445983886,
      "epoch": 0.406857599840056,
      "grad_norm": 1.75,
      "learning_rate": 0.0004989344608649815,
      "loss": 6.411,
      "mean_token_accuracy": 0.13698220029473304,
      "num_tokens": 7973090.0,
      "step": 4070
    },
    {
      "entropy": 6.526764011383056,
      "epoch": 0.40735742490128457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004989309888484511,
      "loss": 6.5142,
      "mean_token_accuracy": 0.12804269716143607,
      "num_tokens": 7982208.0,
      "step": 4075
    },
    {
      "entropy": 6.441964769363404,
      "epoch": 0.4078572499625131,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004989275111978748,
      "loss": 6.4852,
      "mean_token_accuracy": 0.13562206402420998,
      "num_tokens": 7993389.0,
      "step": 4080
    },
    {
      "entropy": 6.4270039081573485,
      "epoch": 0.4083570750237417,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004989240279133398,
      "loss": 6.3425,
      "mean_token_accuracy": 0.14047398939728736,
      "num_tokens": 8002749.0,
      "step": 4085
    },
    {
      "entropy": 6.494481229782105,
      "epoch": 0.4088569000849703,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004989205389949339,
      "loss": 6.3405,
      "mean_token_accuracy": 0.1379816897213459,
      "num_tokens": 8011552.0,
      "step": 4090
    },
    {
      "entropy": 6.340465354919433,
      "epoch": 0.4093567251461988,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004989170444427448,
      "loss": 6.3727,
      "mean_token_accuracy": 0.13385937213897706,
      "num_tokens": 8020622.0,
      "step": 4095
    },
    {
      "entropy": 6.392237949371338,
      "epoch": 0.4098565502074274,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004989135442568606,
      "loss": 6.3105,
      "mean_token_accuracy": 0.14349081069231034,
      "num_tokens": 8030356.0,
      "step": 4100
    },
    {
      "entropy": 6.448809289932251,
      "epoch": 0.410356375268656,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004989100384373693,
      "loss": 6.4328,
      "mean_token_accuracy": 0.13252698332071305,
      "num_tokens": 8039903.0,
      "step": 4105
    },
    {
      "entropy": 6.488882303237915,
      "epoch": 0.4108562003298845,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004989065269843588,
      "loss": 6.3604,
      "mean_token_accuracy": 0.13210076540708543,
      "num_tokens": 8050488.0,
      "step": 4110
    },
    {
      "entropy": 6.481236839294434,
      "epoch": 0.4113560253911131,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000498903009897918,
      "loss": 6.3962,
      "mean_token_accuracy": 0.1381304681301117,
      "num_tokens": 8060179.0,
      "step": 4115
    },
    {
      "entropy": 6.47611403465271,
      "epoch": 0.4118558504523417,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004988994871781349,
      "loss": 6.4489,
      "mean_token_accuracy": 0.13481701165437698,
      "num_tokens": 8070915.0,
      "step": 4120
    },
    {
      "entropy": 6.412845230102539,
      "epoch": 0.4123556755135703,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988959588250983,
      "loss": 6.3724,
      "mean_token_accuracy": 0.14198843836784364,
      "num_tokens": 8080203.0,
      "step": 4125
    },
    {
      "entropy": 6.493917083740234,
      "epoch": 0.4128555005747988,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000498892424838897,
      "loss": 6.3816,
      "mean_token_accuracy": 0.1400151565670967,
      "num_tokens": 8090552.0,
      "step": 4130
    },
    {
      "entropy": 6.468116998672485,
      "epoch": 0.4133553256360274,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004988888852196199,
      "loss": 6.3498,
      "mean_token_accuracy": 0.14506910964846612,
      "num_tokens": 8099947.0,
      "step": 4135
    },
    {
      "entropy": 6.4164337635040285,
      "epoch": 0.413855150697256,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988853399673562,
      "loss": 6.4473,
      "mean_token_accuracy": 0.1339076429605484,
      "num_tokens": 8110502.0,
      "step": 4140
    },
    {
      "entropy": 6.4882711410522464,
      "epoch": 0.4143549757584845,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004988817890821948,
      "loss": 6.4656,
      "mean_token_accuracy": 0.1324286200106144,
      "num_tokens": 8118762.0,
      "step": 4145
    },
    {
      "entropy": 6.462985801696777,
      "epoch": 0.4148548008197131,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988782325642252,
      "loss": 6.3717,
      "mean_token_accuracy": 0.13247129172086716,
      "num_tokens": 8129072.0,
      "step": 4150
    },
    {
      "entropy": 6.441734266281128,
      "epoch": 0.4153546258809417,
      "grad_norm": 1.125,
      "learning_rate": 0.000498874670413537,
      "loss": 6.4098,
      "mean_token_accuracy": 0.1296373963356018,
      "num_tokens": 8138504.0,
      "step": 4155
    },
    {
      "entropy": 6.5720216751098635,
      "epoch": 0.4158544509421702,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988711026302197,
      "loss": 6.4741,
      "mean_token_accuracy": 0.1308560349047184,
      "num_tokens": 8148534.0,
      "step": 4160
    },
    {
      "entropy": 6.507727098464966,
      "epoch": 0.4163542760033988,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004988675292143629,
      "loss": 6.5272,
      "mean_token_accuracy": 0.1274973638355732,
      "num_tokens": 8160272.0,
      "step": 4165
    },
    {
      "entropy": 6.497180700302124,
      "epoch": 0.4168541010646274,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004988639501660568,
      "loss": 6.3836,
      "mean_token_accuracy": 0.13714147731661797,
      "num_tokens": 8170053.0,
      "step": 4170
    },
    {
      "entropy": 6.519702005386352,
      "epoch": 0.4173539261258559,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004988603654853913,
      "loss": 6.4107,
      "mean_token_accuracy": 0.13379530236124992,
      "num_tokens": 8179425.0,
      "step": 4175
    },
    {
      "entropy": 6.424587154388428,
      "epoch": 0.4178537511870845,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004988567751724565,
      "loss": 6.4366,
      "mean_token_accuracy": 0.1306513100862503,
      "num_tokens": 8189846.0,
      "step": 4180
    },
    {
      "entropy": 6.542534732818604,
      "epoch": 0.4183535762483131,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498853179227343,
      "loss": 6.4442,
      "mean_token_accuracy": 0.13438463732600212,
      "num_tokens": 8198596.0,
      "step": 4185
    },
    {
      "entropy": 6.52326774597168,
      "epoch": 0.4188534013095417,
      "grad_norm": 1.171875,
      "learning_rate": 0.000498849577650141,
      "loss": 6.3668,
      "mean_token_accuracy": 0.1381375603377819,
      "num_tokens": 8206849.0,
      "step": 4190
    },
    {
      "entropy": 6.399906778335572,
      "epoch": 0.4193532263707702,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988459704409413,
      "loss": 6.3791,
      "mean_token_accuracy": 0.1348581276834011,
      "num_tokens": 8216260.0,
      "step": 4195
    },
    {
      "entropy": 6.498875951766967,
      "epoch": 0.4198530514319988,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004988423575998345,
      "loss": 6.4093,
      "mean_token_accuracy": 0.13606366738677025,
      "num_tokens": 8224650.0,
      "step": 4200
    },
    {
      "entropy": 6.329025411605835,
      "epoch": 0.4203528764932274,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004988387391269117,
      "loss": 6.2693,
      "mean_token_accuracy": 0.1433685302734375,
      "num_tokens": 8235276.0,
      "step": 4205
    },
    {
      "entropy": 6.479813098907471,
      "epoch": 0.4208527015544559,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004988351150222638,
      "loss": 6.3397,
      "mean_token_accuracy": 0.13741505444049834,
      "num_tokens": 8244687.0,
      "step": 4210
    },
    {
      "entropy": 6.44670729637146,
      "epoch": 0.4213525266156845,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004988314852859819,
      "loss": 6.4632,
      "mean_token_accuracy": 0.1346420370042324,
      "num_tokens": 8255091.0,
      "step": 4215
    },
    {
      "entropy": 6.318567132949829,
      "epoch": 0.4218523516769131,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004988278499181575,
      "loss": 6.3028,
      "mean_token_accuracy": 0.14652785509824753,
      "num_tokens": 8265172.0,
      "step": 4220
    },
    {
      "entropy": 6.4701916694641115,
      "epoch": 0.4223521767381416,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498824208918882,
      "loss": 6.3747,
      "mean_token_accuracy": 0.13661768659949303,
      "num_tokens": 8275625.0,
      "step": 4225
    },
    {
      "entropy": 6.400294065475464,
      "epoch": 0.4228520017993702,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000498820562288247,
      "loss": 6.3664,
      "mean_token_accuracy": 0.1326092280447483,
      "num_tokens": 8284343.0,
      "step": 4230
    },
    {
      "entropy": 6.50111985206604,
      "epoch": 0.4233518268605988,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004988169100263442,
      "loss": 6.4325,
      "mean_token_accuracy": 0.13281701132655144,
      "num_tokens": 8293430.0,
      "step": 4235
    },
    {
      "entropy": 6.451905584335327,
      "epoch": 0.4238516519218274,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988132521332656,
      "loss": 6.4059,
      "mean_token_accuracy": 0.13518600389361382,
      "num_tokens": 8302925.0,
      "step": 4240
    },
    {
      "entropy": 6.439618444442749,
      "epoch": 0.4243514769830559,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004988095886091031,
      "loss": 6.3684,
      "mean_token_accuracy": 0.1392774134874344,
      "num_tokens": 8312134.0,
      "step": 4245
    },
    {
      "entropy": 6.476375436782837,
      "epoch": 0.4248513020442845,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004988059194539489,
      "loss": 6.3667,
      "mean_token_accuracy": 0.13766386359930038,
      "num_tokens": 8320942.0,
      "step": 4250
    },
    {
      "entropy": 6.4668882369995115,
      "epoch": 0.4253511271055131,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004988022446678954,
      "loss": 6.387,
      "mean_token_accuracy": 0.1343892551958561,
      "num_tokens": 8331898.0,
      "step": 4255
    },
    {
      "entropy": 6.407869291305542,
      "epoch": 0.4258509521667416,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004987985642510352,
      "loss": 6.3796,
      "mean_token_accuracy": 0.14184328019618989,
      "num_tokens": 8341728.0,
      "step": 4260
    },
    {
      "entropy": 6.463212251663208,
      "epoch": 0.4263507772279702,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987948782034605,
      "loss": 6.3204,
      "mean_token_accuracy": 0.1412410818040371,
      "num_tokens": 8351611.0,
      "step": 4265
    },
    {
      "entropy": 6.433274126052856,
      "epoch": 0.4268506022891988,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987911865252644,
      "loss": 6.3506,
      "mean_token_accuracy": 0.14228583350777627,
      "num_tokens": 8360528.0,
      "step": 4270
    },
    {
      "entropy": 6.42871208190918,
      "epoch": 0.42735042735042733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987874892165395,
      "loss": 6.4159,
      "mean_token_accuracy": 0.13375010266900061,
      "num_tokens": 8369898.0,
      "step": 4275
    },
    {
      "entropy": 6.469485330581665,
      "epoch": 0.4278502524116559,
      "grad_norm": 1.046875,
      "learning_rate": 0.000498783786277379,
      "loss": 6.3535,
      "mean_token_accuracy": 0.14114384949207306,
      "num_tokens": 8379928.0,
      "step": 4280
    },
    {
      "entropy": 6.448226451873779,
      "epoch": 0.4283500774728845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987800777078759,
      "loss": 6.3864,
      "mean_token_accuracy": 0.14110791832208633,
      "num_tokens": 8389208.0,
      "step": 4285
    },
    {
      "entropy": 6.395701837539673,
      "epoch": 0.42884990253411304,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987763635081238,
      "loss": 6.3813,
      "mean_token_accuracy": 0.13884787857532502,
      "num_tokens": 8398364.0,
      "step": 4290
    },
    {
      "entropy": 6.441565322875976,
      "epoch": 0.4293497275953416,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000498772643678216,
      "loss": 6.3931,
      "mean_token_accuracy": 0.13277500122785568,
      "num_tokens": 8408756.0,
      "step": 4295
    },
    {
      "entropy": 6.462358808517456,
      "epoch": 0.4298495526565702,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987689182182459,
      "loss": 6.3572,
      "mean_token_accuracy": 0.140143720805645,
      "num_tokens": 8418318.0,
      "step": 4300
    },
    {
      "entropy": 6.39695634841919,
      "epoch": 0.4303493777177988,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004987651871283075,
      "loss": 6.3876,
      "mean_token_accuracy": 0.13188599050045013,
      "num_tokens": 8427709.0,
      "step": 4305
    },
    {
      "entropy": 6.482594633102417,
      "epoch": 0.43084920277902733,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004987614504084946,
      "loss": 6.3968,
      "mean_token_accuracy": 0.1333758682012558,
      "num_tokens": 8437848.0,
      "step": 4310
    },
    {
      "entropy": 6.467829704284668,
      "epoch": 0.4313490278402559,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987577080589012,
      "loss": 6.3444,
      "mean_token_accuracy": 0.13813810050487518,
      "num_tokens": 8447466.0,
      "step": 4315
    },
    {
      "entropy": 6.3970293521881105,
      "epoch": 0.4318488529014845,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987539600796213,
      "loss": 6.3599,
      "mean_token_accuracy": 0.14601844921708107,
      "num_tokens": 8456651.0,
      "step": 4320
    },
    {
      "entropy": 6.502226257324219,
      "epoch": 0.43234867796271303,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004987502064707494,
      "loss": 6.3991,
      "mean_token_accuracy": 0.1384994775056839,
      "num_tokens": 8466421.0,
      "step": 4325
    },
    {
      "entropy": 6.430251932144165,
      "epoch": 0.4328485030239416,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004987464472323799,
      "loss": 6.3389,
      "mean_token_accuracy": 0.13588759750127793,
      "num_tokens": 8476419.0,
      "step": 4330
    },
    {
      "entropy": 6.482425165176392,
      "epoch": 0.4333483280851702,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004987426823646074,
      "loss": 6.3669,
      "mean_token_accuracy": 0.13521523028612137,
      "num_tokens": 8486652.0,
      "step": 4335
    },
    {
      "entropy": 6.367968988418579,
      "epoch": 0.43384815314639874,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004987389118675266,
      "loss": 6.3183,
      "mean_token_accuracy": 0.13830461949110032,
      "num_tokens": 8495965.0,
      "step": 4340
    },
    {
      "entropy": 6.472572660446167,
      "epoch": 0.4343479782076273,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987351357412322,
      "loss": 6.4053,
      "mean_token_accuracy": 0.1303241431713104,
      "num_tokens": 8506018.0,
      "step": 4345
    },
    {
      "entropy": 6.418121719360352,
      "epoch": 0.4348478032688559,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004987313539858196,
      "loss": 6.3871,
      "mean_token_accuracy": 0.1389095142483711,
      "num_tokens": 8516178.0,
      "step": 4350
    },
    {
      "entropy": 6.485131502151489,
      "epoch": 0.43534762833008445,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987275666013836,
      "loss": 6.4256,
      "mean_token_accuracy": 0.1329130657017231,
      "num_tokens": 8525548.0,
      "step": 4355
    },
    {
      "entropy": 6.4853137016296385,
      "epoch": 0.43584745339131303,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004987237735880195,
      "loss": 6.4869,
      "mean_token_accuracy": 0.12986717298626899,
      "num_tokens": 8535941.0,
      "step": 4360
    },
    {
      "entropy": 6.376958274841309,
      "epoch": 0.4363472784525416,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987199749458229,
      "loss": 6.2931,
      "mean_token_accuracy": 0.14049483463168144,
      "num_tokens": 8545381.0,
      "step": 4365
    },
    {
      "entropy": 6.483648872375488,
      "epoch": 0.4368471035137702,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004987161706748893,
      "loss": 6.3855,
      "mean_token_accuracy": 0.13644311875104903,
      "num_tokens": 8555827.0,
      "step": 4370
    },
    {
      "entropy": 6.426487731933594,
      "epoch": 0.43734692857499874,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004987123607753144,
      "loss": 6.2699,
      "mean_token_accuracy": 0.14769768416881562,
      "num_tokens": 8565508.0,
      "step": 4375
    },
    {
      "entropy": 6.483022212982178,
      "epoch": 0.4378467536362273,
      "grad_norm": 1.09375,
      "learning_rate": 0.000498708545247194,
      "loss": 6.4767,
      "mean_token_accuracy": 0.12742696925997735,
      "num_tokens": 8576273.0,
      "step": 4380
    },
    {
      "entropy": 6.482770538330078,
      "epoch": 0.4383465786974559,
      "grad_norm": 1.125,
      "learning_rate": 0.0004987047240906243,
      "loss": 6.4335,
      "mean_token_accuracy": 0.13154776245355607,
      "num_tokens": 8585488.0,
      "step": 4385
    },
    {
      "entropy": 6.455898332595825,
      "epoch": 0.43884640375868444,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004987008973057012,
      "loss": 6.4639,
      "mean_token_accuracy": 0.13179367780685425,
      "num_tokens": 8595299.0,
      "step": 4390
    },
    {
      "entropy": 6.455296516418457,
      "epoch": 0.43934622881991303,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004986970648925211,
      "loss": 6.3398,
      "mean_token_accuracy": 0.13784969225525856,
      "num_tokens": 8604910.0,
      "step": 4395
    },
    {
      "entropy": 6.431913709640503,
      "epoch": 0.4398460538811416,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986932268511804,
      "loss": 6.3595,
      "mean_token_accuracy": 0.1392812706530094,
      "num_tokens": 8614966.0,
      "step": 4400
    },
    {
      "entropy": 6.377312231063843,
      "epoch": 0.44034587894237015,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004986893831817758,
      "loss": 6.395,
      "mean_token_accuracy": 0.13974217027425767,
      "num_tokens": 8624539.0,
      "step": 4405
    },
    {
      "entropy": 6.411195421218872,
      "epoch": 0.44084570400359874,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986855338844038,
      "loss": 6.2375,
      "mean_token_accuracy": 0.14474914520978927,
      "num_tokens": 8633775.0,
      "step": 4410
    },
    {
      "entropy": 6.424028539657593,
      "epoch": 0.4413455290648273,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004986816789591613,
      "loss": 6.4002,
      "mean_token_accuracy": 0.13322965949773788,
      "num_tokens": 8643798.0,
      "step": 4415
    },
    {
      "entropy": 6.314562368392944,
      "epoch": 0.44184535412605586,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986778184061453,
      "loss": 6.3162,
      "mean_token_accuracy": 0.13995987102389335,
      "num_tokens": 8653695.0,
      "step": 4420
    },
    {
      "entropy": 6.44688138961792,
      "epoch": 0.44234517918728444,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004986739522254529,
      "loss": 6.3351,
      "mean_token_accuracy": 0.13648810386657714,
      "num_tokens": 8663521.0,
      "step": 4425
    },
    {
      "entropy": 6.519506978988647,
      "epoch": 0.44284500424851303,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986700804171816,
      "loss": 6.4708,
      "mean_token_accuracy": 0.13224498629570008,
      "num_tokens": 8673108.0,
      "step": 4430
    },
    {
      "entropy": 6.444082593917846,
      "epoch": 0.4433448293097416,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004986662029814285,
      "loss": 6.3291,
      "mean_token_accuracy": 0.13684166669845582,
      "num_tokens": 8682917.0,
      "step": 4435
    },
    {
      "entropy": 6.498892307281494,
      "epoch": 0.44384465437097015,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986623199182913,
      "loss": 6.428,
      "mean_token_accuracy": 0.1265405111014843,
      "num_tokens": 8692436.0,
      "step": 4440
    },
    {
      "entropy": 6.3918274402618405,
      "epoch": 0.44434447943219874,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986584312278678,
      "loss": 6.3374,
      "mean_token_accuracy": 0.14236319363117217,
      "num_tokens": 8702627.0,
      "step": 4445
    },
    {
      "entropy": 6.422056865692139,
      "epoch": 0.4448443044934273,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004986545369102556,
      "loss": 6.3541,
      "mean_token_accuracy": 0.13138117119669915,
      "num_tokens": 8711697.0,
      "step": 4450
    },
    {
      "entropy": 6.482198238372803,
      "epoch": 0.44534412955465585,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986506369655528,
      "loss": 6.3701,
      "mean_token_accuracy": 0.13062169700860976,
      "num_tokens": 8721449.0,
      "step": 4455
    },
    {
      "entropy": 6.404720163345337,
      "epoch": 0.44584395461588444,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004986467313938575,
      "loss": 6.3618,
      "mean_token_accuracy": 0.13613314256072045,
      "num_tokens": 8731829.0,
      "step": 4460
    },
    {
      "entropy": 6.3724242687225345,
      "epoch": 0.44634377967711303,
      "grad_norm": 1.015625,
      "learning_rate": 0.000498642820195268,
      "loss": 6.1694,
      "mean_token_accuracy": 0.1527157574892044,
      "num_tokens": 8741768.0,
      "step": 4465
    },
    {
      "entropy": 6.441873168945312,
      "epoch": 0.44684360473834156,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004986389033698827,
      "loss": 6.4681,
      "mean_token_accuracy": 0.1314198724925518,
      "num_tokens": 8751473.0,
      "step": 4470
    },
    {
      "entropy": 6.505035877227783,
      "epoch": 0.44734342979957015,
      "grad_norm": 1.125,
      "learning_rate": 0.0004986349809178002,
      "loss": 6.3855,
      "mean_token_accuracy": 0.13281653970479965,
      "num_tokens": 8761229.0,
      "step": 4475
    },
    {
      "entropy": 6.354346704483032,
      "epoch": 0.44784325486079873,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004986310528391192,
      "loss": 6.337,
      "mean_token_accuracy": 0.1419730991125107,
      "num_tokens": 8770670.0,
      "step": 4480
    },
    {
      "entropy": 6.450547313690185,
      "epoch": 0.44834307992202727,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986271191339383,
      "loss": 6.3589,
      "mean_token_accuracy": 0.13514169678092003,
      "num_tokens": 8780816.0,
      "step": 4485
    },
    {
      "entropy": 6.522676181793213,
      "epoch": 0.44884290498325585,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004986231798023566,
      "loss": 6.4335,
      "mean_token_accuracy": 0.1313347890973091,
      "num_tokens": 8790304.0,
      "step": 4490
    },
    {
      "entropy": 6.396866035461426,
      "epoch": 0.44934273004448444,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004986192348444734,
      "loss": 6.3802,
      "mean_token_accuracy": 0.1358383111655712,
      "num_tokens": 8799188.0,
      "step": 4495
    },
    {
      "entropy": 6.488626098632812,
      "epoch": 0.449842555105713,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004986152842603877,
      "loss": 6.4195,
      "mean_token_accuracy": 0.13813868388533593,
      "num_tokens": 8810194.0,
      "step": 4500
    },
    {
      "entropy": 6.526076459884644,
      "epoch": 0.45034238016694156,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004986113280501989,
      "loss": 6.4292,
      "mean_token_accuracy": 0.13713254481554032,
      "num_tokens": 8820394.0,
      "step": 4505
    },
    {
      "entropy": 6.421188497543335,
      "epoch": 0.45084220522817015,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0004986073662140068,
      "loss": 6.3812,
      "mean_token_accuracy": 0.14109829887747766,
      "num_tokens": 8831308.0,
      "step": 4510
    },
    {
      "entropy": 6.408560180664063,
      "epoch": 0.45134203028939873,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004986033987519109,
      "loss": 6.316,
      "mean_token_accuracy": 0.1428537130355835,
      "num_tokens": 8841461.0,
      "step": 4515
    },
    {
      "entropy": 6.43904824256897,
      "epoch": 0.45184185535062726,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985994256640108,
      "loss": 6.3336,
      "mean_token_accuracy": 0.13536809235811234,
      "num_tokens": 8850356.0,
      "step": 4520
    },
    {
      "entropy": 6.454597043991089,
      "epoch": 0.45234168041185585,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004985954469504069,
      "loss": 6.4304,
      "mean_token_accuracy": 0.13004560098052026,
      "num_tokens": 8859863.0,
      "step": 4525
    },
    {
      "entropy": 6.4495768547058105,
      "epoch": 0.45284150547308444,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004985914626111991,
      "loss": 6.3042,
      "mean_token_accuracy": 0.1360592320561409,
      "num_tokens": 8869887.0,
      "step": 4530
    },
    {
      "entropy": 6.442905139923096,
      "epoch": 0.45334133053431297,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985874726464875,
      "loss": 6.467,
      "mean_token_accuracy": 0.13518969491124153,
      "num_tokens": 8879950.0,
      "step": 4535
    },
    {
      "entropy": 6.440595245361328,
      "epoch": 0.45384115559554156,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985834770563727,
      "loss": 6.23,
      "mean_token_accuracy": 0.14607514441013336,
      "num_tokens": 8888304.0,
      "step": 4540
    },
    {
      "entropy": 6.382362365722656,
      "epoch": 0.45434098065677014,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004985794758409552,
      "loss": 6.2371,
      "mean_token_accuracy": 0.1514820337295532,
      "num_tokens": 8898642.0,
      "step": 4545
    },
    {
      "entropy": 6.375413465499878,
      "epoch": 0.4548408057179987,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004985754690003354,
      "loss": 6.463,
      "mean_token_accuracy": 0.12876035273075104,
      "num_tokens": 8909202.0,
      "step": 4550
    },
    {
      "entropy": 6.362637329101562,
      "epoch": 0.45534063077922726,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985714565346145,
      "loss": 6.2731,
      "mean_token_accuracy": 0.14898935705423355,
      "num_tokens": 8918844.0,
      "step": 4555
    },
    {
      "entropy": 6.50612359046936,
      "epoch": 0.45584045584045585,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004985674384438933,
      "loss": 6.3859,
      "mean_token_accuracy": 0.13208258599042894,
      "num_tokens": 8928706.0,
      "step": 4560
    },
    {
      "entropy": 6.495807266235351,
      "epoch": 0.45634028090168444,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004985634147282729,
      "loss": 6.3235,
      "mean_token_accuracy": 0.1419259324669838,
      "num_tokens": 8939293.0,
      "step": 4565
    },
    {
      "entropy": 6.330229187011719,
      "epoch": 0.45684010596291297,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004985593853878544,
      "loss": 6.3021,
      "mean_token_accuracy": 0.13806715086102486,
      "num_tokens": 8947922.0,
      "step": 4570
    },
    {
      "entropy": 6.336430025100708,
      "epoch": 0.45733993102414155,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004985553504227394,
      "loss": 6.3088,
      "mean_token_accuracy": 0.14462661296129226,
      "num_tokens": 8957745.0,
      "step": 4575
    },
    {
      "entropy": 6.517952728271484,
      "epoch": 0.45783975608537014,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985513098330293,
      "loss": 6.4451,
      "mean_token_accuracy": 0.13025145679712297,
      "num_tokens": 8968331.0,
      "step": 4580
    },
    {
      "entropy": 6.442313957214355,
      "epoch": 0.4583395811465987,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004985472636188257,
      "loss": 6.2953,
      "mean_token_accuracy": 0.1470504179596901,
      "num_tokens": 8977752.0,
      "step": 4585
    },
    {
      "entropy": 6.348786783218384,
      "epoch": 0.45883940620782726,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004985432117802305,
      "loss": 6.3284,
      "mean_token_accuracy": 0.13521047681570053,
      "num_tokens": 8986701.0,
      "step": 4590
    },
    {
      "entropy": 6.356202936172485,
      "epoch": 0.45933923126905585,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985391543173456,
      "loss": 6.3793,
      "mean_token_accuracy": 0.13613076582551004,
      "num_tokens": 8996192.0,
      "step": 4595
    },
    {
      "entropy": 6.5046874523162845,
      "epoch": 0.4598390563302844,
      "grad_norm": 1.0625,
      "learning_rate": 0.000498535091230273,
      "loss": 6.3453,
      "mean_token_accuracy": 0.13811999410390854,
      "num_tokens": 9007059.0,
      "step": 4600
    },
    {
      "entropy": 6.36495680809021,
      "epoch": 0.46033888139151297,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004985310225191152,
      "loss": 6.2861,
      "mean_token_accuracy": 0.13991933912038804,
      "num_tokens": 9016923.0,
      "step": 4605
    },
    {
      "entropy": 6.417649173736573,
      "epoch": 0.46083870645274155,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004985269481839744,
      "loss": 6.4057,
      "mean_token_accuracy": 0.13580618500709535,
      "num_tokens": 9026034.0,
      "step": 4610
    },
    {
      "entropy": 6.55390625,
      "epoch": 0.4613385315139701,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004985228682249529,
      "loss": 6.414,
      "mean_token_accuracy": 0.13191715031862258,
      "num_tokens": 9036773.0,
      "step": 4615
    },
    {
      "entropy": 6.437671566009522,
      "epoch": 0.46183835657519867,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004985187826421538,
      "loss": 6.3699,
      "mean_token_accuracy": 0.13903029933571814,
      "num_tokens": 9047825.0,
      "step": 4620
    },
    {
      "entropy": 6.451393985748291,
      "epoch": 0.46233818163642726,
      "grad_norm": 1.125,
      "learning_rate": 0.0004985146914356795,
      "loss": 6.4405,
      "mean_token_accuracy": 0.13519158437848092,
      "num_tokens": 9056295.0,
      "step": 4625
    },
    {
      "entropy": 6.442917108535767,
      "epoch": 0.46283800669765585,
      "grad_norm": 1.078125,
      "learning_rate": 0.000498510594605633,
      "loss": 6.3904,
      "mean_token_accuracy": 0.13657915964722633,
      "num_tokens": 9067245.0,
      "step": 4630
    },
    {
      "entropy": 6.431668329238891,
      "epoch": 0.4633378317588844,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004985064921521177,
      "loss": 6.3562,
      "mean_token_accuracy": 0.13739126324653625,
      "num_tokens": 9077528.0,
      "step": 4635
    },
    {
      "entropy": 6.414422988891602,
      "epoch": 0.46383765682011296,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004985023840752364,
      "loss": 6.2794,
      "mean_token_accuracy": 0.14148288667201997,
      "num_tokens": 9086015.0,
      "step": 4640
    },
    {
      "entropy": 6.46856803894043,
      "epoch": 0.46433748188134155,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984982703750926,
      "loss": 6.3562,
      "mean_token_accuracy": 0.13547587618231774,
      "num_tokens": 9095963.0,
      "step": 4645
    },
    {
      "entropy": 6.324900102615357,
      "epoch": 0.4648373069425701,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004984941510517899,
      "loss": 6.1896,
      "mean_token_accuracy": 0.14242352768778802,
      "num_tokens": 9105819.0,
      "step": 4650
    },
    {
      "entropy": 6.324583625793457,
      "epoch": 0.46533713200379867,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004984900261054319,
      "loss": 6.2002,
      "mean_token_accuracy": 0.14522180408239366,
      "num_tokens": 9114574.0,
      "step": 4655
    },
    {
      "entropy": 6.430502605438233,
      "epoch": 0.46583695706502726,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004984858955361224,
      "loss": 6.4139,
      "mean_token_accuracy": 0.13110233321785927,
      "num_tokens": 9125490.0,
      "step": 4660
    },
    {
      "entropy": 6.514950084686279,
      "epoch": 0.4663367821262558,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004984817593439651,
      "loss": 6.3685,
      "mean_token_accuracy": 0.13579460829496384,
      "num_tokens": 9134886.0,
      "step": 4665
    },
    {
      "entropy": 6.368170738220215,
      "epoch": 0.4668366071874844,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984776175290643,
      "loss": 6.3113,
      "mean_token_accuracy": 0.14255902469158171,
      "num_tokens": 9143598.0,
      "step": 4670
    },
    {
      "entropy": 6.34682936668396,
      "epoch": 0.46733643224871296,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004984734700915242,
      "loss": 6.3983,
      "mean_token_accuracy": 0.12681760415434837,
      "num_tokens": 9153051.0,
      "step": 4675
    },
    {
      "entropy": 6.5240332126617435,
      "epoch": 0.4678362573099415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000498469317031449,
      "loss": 6.3257,
      "mean_token_accuracy": 0.1389312207698822,
      "num_tokens": 9162472.0,
      "step": 4680
    },
    {
      "entropy": 6.4434144496917725,
      "epoch": 0.4683360823711701,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984651583489434,
      "loss": 6.3844,
      "mean_token_accuracy": 0.13914884254336357,
      "num_tokens": 9171453.0,
      "step": 4685
    },
    {
      "entropy": 6.390880918502807,
      "epoch": 0.46883590743239867,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004984609940441119,
      "loss": 6.2935,
      "mean_token_accuracy": 0.14254987463355065,
      "num_tokens": 9181104.0,
      "step": 4690
    },
    {
      "entropy": 6.423695039749146,
      "epoch": 0.46933573249362726,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004984568241170593,
      "loss": 6.3876,
      "mean_token_accuracy": 0.14116302505135536,
      "num_tokens": 9190403.0,
      "step": 4695
    },
    {
      "entropy": 6.336381912231445,
      "epoch": 0.4698355575548558,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984526485678905,
      "loss": 6.2576,
      "mean_token_accuracy": 0.1406373657286167,
      "num_tokens": 9199132.0,
      "step": 4700
    },
    {
      "entropy": 6.456232309341431,
      "epoch": 0.4703353826160844,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004984484673967105,
      "loss": 6.4177,
      "mean_token_accuracy": 0.12666289135813713,
      "num_tokens": 9208274.0,
      "step": 4705
    },
    {
      "entropy": 6.388136005401611,
      "epoch": 0.47083520767731296,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984442806036246,
      "loss": 6.3933,
      "mean_token_accuracy": 0.14819189608097078,
      "num_tokens": 9218513.0,
      "step": 4710
    },
    {
      "entropy": 6.43345046043396,
      "epoch": 0.4713350327385415,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984400881887381,
      "loss": 6.3889,
      "mean_token_accuracy": 0.1344588428735733,
      "num_tokens": 9229197.0,
      "step": 4715
    },
    {
      "entropy": 6.464925956726074,
      "epoch": 0.4718348577997701,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004984358901521565,
      "loss": 6.2591,
      "mean_token_accuracy": 0.1347227245569229,
      "num_tokens": 9239465.0,
      "step": 4720
    },
    {
      "entropy": 6.422285795211792,
      "epoch": 0.47233468286099867,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984316864939853,
      "loss": 6.3694,
      "mean_token_accuracy": 0.14140414595603942,
      "num_tokens": 9250112.0,
      "step": 4725
    },
    {
      "entropy": 6.404574966430664,
      "epoch": 0.4728345079222272,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004984274772143304,
      "loss": 6.3721,
      "mean_token_accuracy": 0.13907759562134742,
      "num_tokens": 9259406.0,
      "step": 4730
    },
    {
      "entropy": 6.434519910812378,
      "epoch": 0.4733343329834558,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004984232623132977,
      "loss": 6.3006,
      "mean_token_accuracy": 0.13723840266466142,
      "num_tokens": 9269462.0,
      "step": 4735
    },
    {
      "entropy": 6.399371576309204,
      "epoch": 0.4738341580446844,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004984190417909932,
      "loss": 6.3621,
      "mean_token_accuracy": 0.1323429524898529,
      "num_tokens": 9280291.0,
      "step": 4740
    },
    {
      "entropy": 6.464402103424073,
      "epoch": 0.4743339831059129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004984148156475231,
      "loss": 6.3276,
      "mean_token_accuracy": 0.13442974165081978,
      "num_tokens": 9289680.0,
      "step": 4745
    },
    {
      "entropy": 6.40746283531189,
      "epoch": 0.4748338081671415,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004984105838829937,
      "loss": 6.3354,
      "mean_token_accuracy": 0.13612587079405786,
      "num_tokens": 9301043.0,
      "step": 4750
    },
    {
      "entropy": 6.197843790054321,
      "epoch": 0.4753336332283701,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004984063464975116,
      "loss": 6.3959,
      "mean_token_accuracy": 0.13443394303321837,
      "num_tokens": 9311851.0,
      "step": 4755
    },
    {
      "entropy": 6.521279525756836,
      "epoch": 0.47583345828959867,
      "grad_norm": 1.125,
      "learning_rate": 0.0004984021034911833,
      "loss": 6.3155,
      "mean_token_accuracy": 0.1389845132827759,
      "num_tokens": 9320886.0,
      "step": 4760
    },
    {
      "entropy": 6.436952018737793,
      "epoch": 0.4763332833508272,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004983978548641154,
      "loss": 6.3972,
      "mean_token_accuracy": 0.13653154224157332,
      "num_tokens": 9330518.0,
      "step": 4765
    },
    {
      "entropy": 6.341850900650025,
      "epoch": 0.4768331084120558,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983936006164151,
      "loss": 6.2379,
      "mean_token_accuracy": 0.13599972128868104,
      "num_tokens": 9339703.0,
      "step": 4770
    },
    {
      "entropy": 6.547325086593628,
      "epoch": 0.47733293347328437,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004983893407481892,
      "loss": 6.4733,
      "mean_token_accuracy": 0.127138202637434,
      "num_tokens": 9349931.0,
      "step": 4775
    },
    {
      "entropy": 6.424779891967773,
      "epoch": 0.4778327585345129,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004983850752595451,
      "loss": 6.3293,
      "mean_token_accuracy": 0.1422683008015156,
      "num_tokens": 9358984.0,
      "step": 4780
    },
    {
      "entropy": 6.4349016666412355,
      "epoch": 0.4783325835957415,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004983808041505898,
      "loss": 6.2833,
      "mean_token_accuracy": 0.13742662444710732,
      "num_tokens": 9368081.0,
      "step": 4785
    },
    {
      "entropy": 6.458392238616943,
      "epoch": 0.4788324086569701,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004983765274214311,
      "loss": 6.4138,
      "mean_token_accuracy": 0.13497077152132989,
      "num_tokens": 9377346.0,
      "step": 4790
    },
    {
      "entropy": 6.343363904953003,
      "epoch": 0.4793322337181986,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004983722450721764,
      "loss": 6.3641,
      "mean_token_accuracy": 0.13735883086919784,
      "num_tokens": 9387493.0,
      "step": 4795
    },
    {
      "entropy": 6.4110101699829105,
      "epoch": 0.4798320587794272,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004983679571029336,
      "loss": 6.3267,
      "mean_token_accuracy": 0.13297692984342574,
      "num_tokens": 9397154.0,
      "step": 4800
    },
    {
      "entropy": 6.446910238265991,
      "epoch": 0.4803318838406558,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983636635138103,
      "loss": 6.4467,
      "mean_token_accuracy": 0.13256388455629348,
      "num_tokens": 9406705.0,
      "step": 4805
    },
    {
      "entropy": 6.409035348892212,
      "epoch": 0.4808317089018843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983593643049148,
      "loss": 6.2945,
      "mean_token_accuracy": 0.14054804518818856,
      "num_tokens": 9416850.0,
      "step": 4810
    },
    {
      "entropy": 6.302218580245972,
      "epoch": 0.4813315339631129,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004983550594763552,
      "loss": 6.2822,
      "mean_token_accuracy": 0.13723476678133012,
      "num_tokens": 9425841.0,
      "step": 4815
    },
    {
      "entropy": 6.312332344055176,
      "epoch": 0.4818313590243415,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004983507490282397,
      "loss": 6.2176,
      "mean_token_accuracy": 0.14703118205070495,
      "num_tokens": 9436748.0,
      "step": 4820
    },
    {
      "entropy": 6.436319065093994,
      "epoch": 0.4823311840855701,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004983464329606768,
      "loss": 6.3686,
      "mean_token_accuracy": 0.13694579228758813,
      "num_tokens": 9446043.0,
      "step": 4825
    },
    {
      "entropy": 6.404607200622559,
      "epoch": 0.4828310091467986,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004983421112737752,
      "loss": 6.382,
      "mean_token_accuracy": 0.1413051627576351,
      "num_tokens": 9455480.0,
      "step": 4830
    },
    {
      "entropy": 6.4302928924560545,
      "epoch": 0.4833308342080272,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004983377839676435,
      "loss": 6.3611,
      "mean_token_accuracy": 0.13676040768623351,
      "num_tokens": 9464412.0,
      "step": 4835
    },
    {
      "entropy": 6.424809360504151,
      "epoch": 0.4838306592692558,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004983334510423906,
      "loss": 6.3192,
      "mean_token_accuracy": 0.13830352947115898,
      "num_tokens": 9475639.0,
      "step": 4840
    },
    {
      "entropy": 6.427857351303101,
      "epoch": 0.4843304843304843,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983291124981257,
      "loss": 6.2785,
      "mean_token_accuracy": 0.14063755720853804,
      "num_tokens": 9484721.0,
      "step": 4845
    },
    {
      "entropy": 6.299161052703857,
      "epoch": 0.4848303093917129,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004983247683349576,
      "loss": 6.2247,
      "mean_token_accuracy": 0.1456839345395565,
      "num_tokens": 9494594.0,
      "step": 4850
    },
    {
      "entropy": 6.298327255249023,
      "epoch": 0.4853301344529415,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004983204185529959,
      "loss": 6.2658,
      "mean_token_accuracy": 0.14413323402404785,
      "num_tokens": 9503707.0,
      "step": 4855
    },
    {
      "entropy": 6.3907839298248295,
      "epoch": 0.48582995951417,
      "grad_norm": 1.25,
      "learning_rate": 0.0004983160631523499,
      "loss": 6.2272,
      "mean_token_accuracy": 0.14055933132767678,
      "num_tokens": 9512473.0,
      "step": 4860
    },
    {
      "entropy": 6.3712372303009035,
      "epoch": 0.4863297845753986,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004983117021331293,
      "loss": 6.3608,
      "mean_token_accuracy": 0.1389065131545067,
      "num_tokens": 9523253.0,
      "step": 4865
    },
    {
      "entropy": 6.4325599193573,
      "epoch": 0.4868296096366272,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004983073354954436,
      "loss": 6.3142,
      "mean_token_accuracy": 0.1434569962322712,
      "num_tokens": 9531938.0,
      "step": 4870
    },
    {
      "entropy": 6.467983675003052,
      "epoch": 0.4873294346978557,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004983029632394029,
      "loss": 6.3447,
      "mean_token_accuracy": 0.13684508129954337,
      "num_tokens": 9541542.0,
      "step": 4875
    },
    {
      "entropy": 6.302601099014282,
      "epoch": 0.4878292597590843,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000498298585365117,
      "loss": 6.2624,
      "mean_token_accuracy": 0.142653389275074,
      "num_tokens": 9551521.0,
      "step": 4880
    },
    {
      "entropy": 6.383222913742065,
      "epoch": 0.4883290848203129,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004982942018726963,
      "loss": 6.389,
      "mean_token_accuracy": 0.14110652059316636,
      "num_tokens": 9561605.0,
      "step": 4885
    },
    {
      "entropy": 6.425156974792481,
      "epoch": 0.4888289098815415,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982898127622508,
      "loss": 6.3004,
      "mean_token_accuracy": 0.1347697749733925,
      "num_tokens": 9571591.0,
      "step": 4890
    },
    {
      "entropy": 6.312005710601807,
      "epoch": 0.48932873494277,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982854180338911,
      "loss": 6.32,
      "mean_token_accuracy": 0.14240961745381356,
      "num_tokens": 9580994.0,
      "step": 4895
    },
    {
      "entropy": 6.423856019973755,
      "epoch": 0.4898285600039986,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982810176877279,
      "loss": 6.3051,
      "mean_token_accuracy": 0.13638504296541215,
      "num_tokens": 9589268.0,
      "step": 4900
    },
    {
      "entropy": 6.419354677200317,
      "epoch": 0.4903283850652272,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004982766117238716,
      "loss": 6.3594,
      "mean_token_accuracy": 0.13577214404940605,
      "num_tokens": 9599461.0,
      "step": 4905
    },
    {
      "entropy": 6.396965980529785,
      "epoch": 0.4908282101264557,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004982722001424331,
      "loss": 6.3303,
      "mean_token_accuracy": 0.13797401115298272,
      "num_tokens": 9609629.0,
      "step": 4910
    },
    {
      "entropy": 6.341986274719238,
      "epoch": 0.4913280351876843,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982677829435236,
      "loss": 6.2512,
      "mean_token_accuracy": 0.14354280680418013,
      "num_tokens": 9619618.0,
      "step": 4915
    },
    {
      "entropy": 6.379724216461182,
      "epoch": 0.4918278602489129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004982633601272543,
      "loss": 6.3056,
      "mean_token_accuracy": 0.13943262919783592,
      "num_tokens": 9629042.0,
      "step": 4920
    },
    {
      "entropy": 6.377507877349854,
      "epoch": 0.4923276853101414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982589316937361,
      "loss": 6.2897,
      "mean_token_accuracy": 0.14414727091789245,
      "num_tokens": 9638596.0,
      "step": 4925
    },
    {
      "entropy": 6.442139720916748,
      "epoch": 0.49282751037137,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004982544976430808,
      "loss": 6.3673,
      "mean_token_accuracy": 0.13445642665028573,
      "num_tokens": 9648505.0,
      "step": 4930
    },
    {
      "entropy": 6.35016016960144,
      "epoch": 0.4933273354325986,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004982500579753995,
      "loss": 6.3477,
      "mean_token_accuracy": 0.13654106482863426,
      "num_tokens": 9658113.0,
      "step": 4935
    },
    {
      "entropy": 6.391938734054565,
      "epoch": 0.49382716049382713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982456126908044,
      "loss": 6.3747,
      "mean_token_accuracy": 0.13850862681865692,
      "num_tokens": 9668075.0,
      "step": 4940
    },
    {
      "entropy": 6.513855361938477,
      "epoch": 0.4943269855550557,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004982411617894071,
      "loss": 6.3345,
      "mean_token_accuracy": 0.13620950132608414,
      "num_tokens": 9678139.0,
      "step": 4945
    },
    {
      "entropy": 6.363553190231324,
      "epoch": 0.4948268106162843,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004982367052713196,
      "loss": 6.3503,
      "mean_token_accuracy": 0.13079959377646447,
      "num_tokens": 9687328.0,
      "step": 4950
    },
    {
      "entropy": 6.346356582641602,
      "epoch": 0.4953266356775129,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004982322431366541,
      "loss": 6.3152,
      "mean_token_accuracy": 0.14156793877482415,
      "num_tokens": 9697536.0,
      "step": 4955
    },
    {
      "entropy": 6.391560411453247,
      "epoch": 0.4958264607387414,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004982277753855227,
      "loss": 6.3571,
      "mean_token_accuracy": 0.14056526571512223,
      "num_tokens": 9706804.0,
      "step": 4960
    },
    {
      "entropy": 6.377194356918335,
      "epoch": 0.49632628579997,
      "grad_norm": 1.109375,
      "learning_rate": 0.000498223302018038,
      "loss": 6.2653,
      "mean_token_accuracy": 0.1481527231633663,
      "num_tokens": 9716868.0,
      "step": 4965
    },
    {
      "entropy": 6.3460808277130125,
      "epoch": 0.4968261108611986,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004982188230343124,
      "loss": 6.3554,
      "mean_token_accuracy": 0.14097662195563315,
      "num_tokens": 9728790.0,
      "step": 4970
    },
    {
      "entropy": 6.420025062561035,
      "epoch": 0.49732593592242713,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004982143384344587,
      "loss": 6.3704,
      "mean_token_accuracy": 0.1372966133058071,
      "num_tokens": 9739118.0,
      "step": 4975
    },
    {
      "entropy": 6.413408088684082,
      "epoch": 0.4978257609836557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004982098482185897,
      "loss": 6.3296,
      "mean_token_accuracy": 0.1391241133213043,
      "num_tokens": 9749347.0,
      "step": 4980
    },
    {
      "entropy": 6.454338884353637,
      "epoch": 0.4983255860448843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004982053523868182,
      "loss": 6.3706,
      "mean_token_accuracy": 0.1381652422249317,
      "num_tokens": 9759432.0,
      "step": 4985
    },
    {
      "entropy": 6.348123407363891,
      "epoch": 0.49882541110611284,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004982008509392576,
      "loss": 6.2999,
      "mean_token_accuracy": 0.1378486856818199,
      "num_tokens": 9768666.0,
      "step": 4990
    },
    {
      "entropy": 6.2872474670410154,
      "epoch": 0.4993252361673414,
      "grad_norm": 1.25,
      "learning_rate": 0.000498196343876021,
      "loss": 6.2387,
      "mean_token_accuracy": 0.14893582016229628,
      "num_tokens": 9778699.0,
      "step": 4995
    },
    {
      "entropy": 6.3792359828948975,
      "epoch": 0.49982506122857,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981918311972219,
      "loss": 6.2744,
      "mean_token_accuracy": 0.13931502103805543,
      "num_tokens": 9788156.0,
      "step": 5000
    },
    {
      "entropy": 6.346083259582519,
      "epoch": 0.5003248862897985,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004981873129029737,
      "loss": 6.2337,
      "mean_token_accuracy": 0.15014591813087463,
      "num_tokens": 9797293.0,
      "step": 5005
    },
    {
      "entropy": 6.321063566207886,
      "epoch": 0.5008247113510271,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00049818278899339,
      "loss": 6.2396,
      "mean_token_accuracy": 0.1417209066450596,
      "num_tokens": 9806962.0,
      "step": 5010
    },
    {
      "entropy": 6.397770595550537,
      "epoch": 0.5013245364122557,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000498178259468585,
      "loss": 6.3936,
      "mean_token_accuracy": 0.13865589052438737,
      "num_tokens": 9816689.0,
      "step": 5015
    },
    {
      "entropy": 6.406004858016968,
      "epoch": 0.5018243614734843,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004981737243286722,
      "loss": 6.2308,
      "mean_token_accuracy": 0.1416420593857765,
      "num_tokens": 9826321.0,
      "step": 5020
    },
    {
      "entropy": 6.444920635223388,
      "epoch": 0.5023241865347129,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004981691835737661,
      "loss": 6.3901,
      "mean_token_accuracy": 0.13233431354165076,
      "num_tokens": 9836054.0,
      "step": 5025
    },
    {
      "entropy": 6.445040464401245,
      "epoch": 0.5028240115959414,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981646372039807,
      "loss": 6.3698,
      "mean_token_accuracy": 0.137714883685112,
      "num_tokens": 9846516.0,
      "step": 5030
    },
    {
      "entropy": 6.378769016265869,
      "epoch": 0.50332383665717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981600852194305,
      "loss": 6.1809,
      "mean_token_accuracy": 0.14500134512782098,
      "num_tokens": 9856599.0,
      "step": 5035
    },
    {
      "entropy": 6.373896980285645,
      "epoch": 0.5038236617183985,
      "grad_norm": 1.09375,
      "learning_rate": 0.00049815552762023,
      "loss": 6.3034,
      "mean_token_accuracy": 0.1417168639600277,
      "num_tokens": 9866922.0,
      "step": 5040
    },
    {
      "entropy": 6.321751499176026,
      "epoch": 0.5043234867796271,
      "grad_norm": 1.25,
      "learning_rate": 0.0004981509644064937,
      "loss": 6.2818,
      "mean_token_accuracy": 0.13754203915596008,
      "num_tokens": 9876174.0,
      "step": 5045
    },
    {
      "entropy": 6.289154052734375,
      "epoch": 0.5048233118408557,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004981463955783367,
      "loss": 6.2314,
      "mean_token_accuracy": 0.14614471644163132,
      "num_tokens": 9886208.0,
      "step": 5050
    },
    {
      "entropy": 6.375028371810913,
      "epoch": 0.5053231369020843,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981418211358736,
      "loss": 6.2785,
      "mean_token_accuracy": 0.13845686763525009,
      "num_tokens": 9896522.0,
      "step": 5055
    },
    {
      "entropy": 6.387468147277832,
      "epoch": 0.5058229619633129,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004981372410792199,
      "loss": 6.2841,
      "mean_token_accuracy": 0.1429004319012165,
      "num_tokens": 9905753.0,
      "step": 5060
    },
    {
      "entropy": 6.395010375976563,
      "epoch": 0.5063227870245414,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004981326554084906,
      "loss": 6.3935,
      "mean_token_accuracy": 0.14241379648447036,
      "num_tokens": 9916566.0,
      "step": 5065
    },
    {
      "entropy": 6.370335245132447,
      "epoch": 0.50682261208577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004981280641238011,
      "loss": 6.3669,
      "mean_token_accuracy": 0.13439635932445526,
      "num_tokens": 9926126.0,
      "step": 5070
    },
    {
      "entropy": 6.426411056518555,
      "epoch": 0.5073224371469985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004981234672252669,
      "loss": 6.3579,
      "mean_token_accuracy": 0.13243716582655907,
      "num_tokens": 9937666.0,
      "step": 5075
    },
    {
      "entropy": 6.4846014976501465,
      "epoch": 0.5078222622082271,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004981188647130037,
      "loss": 6.4061,
      "mean_token_accuracy": 0.1410438321530819,
      "num_tokens": 9947107.0,
      "step": 5080
    },
    {
      "entropy": 6.404412460327149,
      "epoch": 0.5083220872694557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004981142565871272,
      "loss": 6.2555,
      "mean_token_accuracy": 0.1399041622877121,
      "num_tokens": 9956484.0,
      "step": 5085
    },
    {
      "entropy": 6.238588333129883,
      "epoch": 0.5088219123306843,
      "grad_norm": 1.125,
      "learning_rate": 0.0004981096428477535,
      "loss": 6.289,
      "mean_token_accuracy": 0.1420084461569786,
      "num_tokens": 9966731.0,
      "step": 5090
    },
    {
      "entropy": 6.405433654785156,
      "epoch": 0.5093217373919128,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004981050234949986,
      "loss": 6.2835,
      "mean_token_accuracy": 0.14363766014575957,
      "num_tokens": 9976491.0,
      "step": 5095
    },
    {
      "entropy": 6.3993744373321535,
      "epoch": 0.5098215624531414,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004981003985289786,
      "loss": 6.2361,
      "mean_token_accuracy": 0.14966873228549957,
      "num_tokens": 9985248.0,
      "step": 5100
    },
    {
      "entropy": 6.291393041610718,
      "epoch": 0.51032138751437,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980957679498101,
      "loss": 6.3129,
      "mean_token_accuracy": 0.14490401297807692,
      "num_tokens": 9995578.0,
      "step": 5105
    },
    {
      "entropy": 6.4709821224212645,
      "epoch": 0.5108212125755985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004980911317576095,
      "loss": 6.4086,
      "mean_token_accuracy": 0.13916842490434647,
      "num_tokens": 10006078.0,
      "step": 5110
    },
    {
      "entropy": 6.440082788467407,
      "epoch": 0.5113210376368271,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004980864899524933,
      "loss": 6.3618,
      "mean_token_accuracy": 0.1328499473631382,
      "num_tokens": 10015943.0,
      "step": 5115
    },
    {
      "entropy": 6.379803800582886,
      "epoch": 0.5118208626980557,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004980818425345786,
      "loss": 6.3521,
      "mean_token_accuracy": 0.13664940968155861,
      "num_tokens": 10025738.0,
      "step": 5120
    },
    {
      "entropy": 6.454934024810791,
      "epoch": 0.5123206877592843,
      "grad_norm": 1.1875,
      "learning_rate": 0.000498077189503982,
      "loss": 6.3268,
      "mean_token_accuracy": 0.13216581642627717,
      "num_tokens": 10035032.0,
      "step": 5125
    },
    {
      "entropy": 6.3814489364624025,
      "epoch": 0.5128205128205128,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004980725308608208,
      "loss": 6.2893,
      "mean_token_accuracy": 0.1422179237008095,
      "num_tokens": 10044201.0,
      "step": 5130
    },
    {
      "entropy": 6.341943168640137,
      "epoch": 0.5133203378817414,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004980678666052121,
      "loss": 6.3285,
      "mean_token_accuracy": 0.14286698028445244,
      "num_tokens": 10054163.0,
      "step": 5135
    },
    {
      "entropy": 6.394853734970093,
      "epoch": 0.51382016294297,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980631967372733,
      "loss": 6.3161,
      "mean_token_accuracy": 0.14063534066081046,
      "num_tokens": 10065225.0,
      "step": 5140
    },
    {
      "entropy": 6.406400394439697,
      "epoch": 0.5143199880041985,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004980585212571218,
      "loss": 6.3019,
      "mean_token_accuracy": 0.138229887932539,
      "num_tokens": 10074776.0,
      "step": 5145
    },
    {
      "entropy": 6.405063343048096,
      "epoch": 0.5148198130654271,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004980538401648753,
      "loss": 6.4621,
      "mean_token_accuracy": 0.1354549139738083,
      "num_tokens": 10085917.0,
      "step": 5150
    },
    {
      "entropy": 6.422722244262696,
      "epoch": 0.5153196381266557,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004980491534606517,
      "loss": 6.2556,
      "mean_token_accuracy": 0.14383360594511033,
      "num_tokens": 10094727.0,
      "step": 5155
    },
    {
      "entropy": 6.299395751953125,
      "epoch": 0.5158194631878842,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004980444611445686,
      "loss": 6.2354,
      "mean_token_accuracy": 0.14570630937814713,
      "num_tokens": 10104839.0,
      "step": 5160
    },
    {
      "entropy": 6.321631288528442,
      "epoch": 0.5163192882491128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004980397632167444,
      "loss": 6.2416,
      "mean_token_accuracy": 0.14232385382056237,
      "num_tokens": 10114501.0,
      "step": 5165
    },
    {
      "entropy": 6.415744876861572,
      "epoch": 0.5168191133103414,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980350596772971,
      "loss": 6.3369,
      "mean_token_accuracy": 0.13884129449725152,
      "num_tokens": 10124597.0,
      "step": 5170
    },
    {
      "entropy": 6.245902395248413,
      "epoch": 0.51731893837157,
      "grad_norm": 2.0,
      "learning_rate": 0.0004980303505263451,
      "loss": 6.1978,
      "mean_token_accuracy": 0.1433788277208805,
      "num_tokens": 10135695.0,
      "step": 5175
    },
    {
      "entropy": 6.245345258712769,
      "epoch": 0.5178187634327985,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004980256357640069,
      "loss": 6.2906,
      "mean_token_accuracy": 0.14047933667898177,
      "num_tokens": 10146107.0,
      "step": 5180
    },
    {
      "entropy": 6.433437156677246,
      "epoch": 0.5183185884940271,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004980209153904011,
      "loss": 6.3634,
      "mean_token_accuracy": 0.14122015684843064,
      "num_tokens": 10155983.0,
      "step": 5185
    },
    {
      "entropy": 6.432741355895996,
      "epoch": 0.5188184135552557,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004980161894056464,
      "loss": 6.3152,
      "mean_token_accuracy": 0.13615113273262977,
      "num_tokens": 10166183.0,
      "step": 5190
    },
    {
      "entropy": 6.400639295578003,
      "epoch": 0.5193182386164842,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004980114578098618,
      "loss": 6.2125,
      "mean_token_accuracy": 0.14031895250082016,
      "num_tokens": 10175524.0,
      "step": 5195
    },
    {
      "entropy": 6.353472280502319,
      "epoch": 0.5198180636777128,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004980067206031663,
      "loss": 6.3585,
      "mean_token_accuracy": 0.13098444491624833,
      "num_tokens": 10186235.0,
      "step": 5200
    },
    {
      "entropy": 6.461900520324707,
      "epoch": 0.5203178887389414,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004980019777856791,
      "loss": 6.3487,
      "mean_token_accuracy": 0.1372937597334385,
      "num_tokens": 10196166.0,
      "step": 5205
    },
    {
      "entropy": 6.439592266082764,
      "epoch": 0.52081771380017,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004979972293575195,
      "loss": 6.2917,
      "mean_token_accuracy": 0.1391717955470085,
      "num_tokens": 10205691.0,
      "step": 5210
    },
    {
      "entropy": 6.373390150070191,
      "epoch": 0.5213175388613985,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004979924753188069,
      "loss": 6.2379,
      "mean_token_accuracy": 0.1419716775417328,
      "num_tokens": 10216224.0,
      "step": 5215
    },
    {
      "entropy": 6.31008358001709,
      "epoch": 0.5218173639226271,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979877156696612,
      "loss": 6.3075,
      "mean_token_accuracy": 0.142496807128191,
      "num_tokens": 10226468.0,
      "step": 5220
    },
    {
      "entropy": 6.5411852359771725,
      "epoch": 0.5223171889838556,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979829504102018,
      "loss": 6.3187,
      "mean_token_accuracy": 0.137630794942379,
      "num_tokens": 10235932.0,
      "step": 5225
    },
    {
      "entropy": 6.442687034606934,
      "epoch": 0.5228170140450842,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004979781795405488,
      "loss": 6.4271,
      "mean_token_accuracy": 0.13533934354782104,
      "num_tokens": 10246052.0,
      "step": 5230
    },
    {
      "entropy": 6.429741287231446,
      "epoch": 0.5233168391063128,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979734030608222,
      "loss": 6.3278,
      "mean_token_accuracy": 0.13412164524197578,
      "num_tokens": 10255056.0,
      "step": 5235
    },
    {
      "entropy": 6.362507915496826,
      "epoch": 0.5238166641675414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004979686209711422,
      "loss": 6.2823,
      "mean_token_accuracy": 0.1377575747668743,
      "num_tokens": 10265403.0,
      "step": 5240
    },
    {
      "entropy": 6.301126480102539,
      "epoch": 0.5243164892287699,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000497963833271629,
      "loss": 6.2542,
      "mean_token_accuracy": 0.14617456123232841,
      "num_tokens": 10275148.0,
      "step": 5245
    },
    {
      "entropy": 6.352523279190064,
      "epoch": 0.5248163142899985,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004979590399624031,
      "loss": 6.2978,
      "mean_token_accuracy": 0.13944504261016846,
      "num_tokens": 10285083.0,
      "step": 5250
    },
    {
      "entropy": 6.3316234111785885,
      "epoch": 0.5253161393512271,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979542410435852,
      "loss": 6.2473,
      "mean_token_accuracy": 0.1363876312971115,
      "num_tokens": 10294466.0,
      "step": 5255
    },
    {
      "entropy": 6.440902662277222,
      "epoch": 0.5258159644124556,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004979494365152959,
      "loss": 6.3591,
      "mean_token_accuracy": 0.13570173531770707,
      "num_tokens": 10304333.0,
      "step": 5260
    },
    {
      "entropy": 6.436617088317871,
      "epoch": 0.5263157894736842,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004979446263776562,
      "loss": 6.2925,
      "mean_token_accuracy": 0.13742511495947837,
      "num_tokens": 10314880.0,
      "step": 5265
    },
    {
      "entropy": 6.390373945236206,
      "epoch": 0.5268156145349128,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000497939810630787,
      "loss": 6.2632,
      "mean_token_accuracy": 0.13987749367952346,
      "num_tokens": 10324060.0,
      "step": 5270
    },
    {
      "entropy": 6.323992013931274,
      "epoch": 0.5273154395961414,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004979349892748096,
      "loss": 6.3786,
      "mean_token_accuracy": 0.1369629204273224,
      "num_tokens": 10335248.0,
      "step": 5275
    },
    {
      "entropy": 6.365525913238526,
      "epoch": 0.5278152646573699,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004979301623098452,
      "loss": 6.1764,
      "mean_token_accuracy": 0.14416617378592492,
      "num_tokens": 10343974.0,
      "step": 5280
    },
    {
      "entropy": 6.409889984130859,
      "epoch": 0.5283150897185985,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004979253297360153,
      "loss": 6.2923,
      "mean_token_accuracy": 0.13761034682393075,
      "num_tokens": 10354037.0,
      "step": 5285
    },
    {
      "entropy": 6.427184581756592,
      "epoch": 0.528814914779827,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004979204915534415,
      "loss": 6.3421,
      "mean_token_accuracy": 0.13941146433353424,
      "num_tokens": 10364714.0,
      "step": 5290
    },
    {
      "entropy": 6.3684930324554445,
      "epoch": 0.5293147398410556,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004979156477622454,
      "loss": 6.321,
      "mean_token_accuracy": 0.1379815936088562,
      "num_tokens": 10373972.0,
      "step": 5295
    },
    {
      "entropy": 6.265024280548095,
      "epoch": 0.5298145649022842,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004979107983625491,
      "loss": 6.219,
      "mean_token_accuracy": 0.14653107672929763,
      "num_tokens": 10382882.0,
      "step": 5300
    },
    {
      "entropy": 6.301440572738647,
      "epoch": 0.5303143899635128,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004979059433544744,
      "loss": 6.2685,
      "mean_token_accuracy": 0.14495664685964585,
      "num_tokens": 10392257.0,
      "step": 5305
    },
    {
      "entropy": 6.395061731338501,
      "epoch": 0.5308142150247414,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004979010827381435,
      "loss": 6.3023,
      "mean_token_accuracy": 0.13960984945297242,
      "num_tokens": 10401055.0,
      "step": 5310
    },
    {
      "entropy": 6.412199068069458,
      "epoch": 0.5313140400859699,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004978962165136787,
      "loss": 6.4062,
      "mean_token_accuracy": 0.13208347037434578,
      "num_tokens": 10411443.0,
      "step": 5315
    },
    {
      "entropy": 6.459276485443115,
      "epoch": 0.5318138651471985,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004978913446812025,
      "loss": 6.3952,
      "mean_token_accuracy": 0.13661471232771874,
      "num_tokens": 10422142.0,
      "step": 5320
    },
    {
      "entropy": 6.452884864807129,
      "epoch": 0.532313690208427,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004978864672408374,
      "loss": 6.309,
      "mean_token_accuracy": 0.13937231451272963,
      "num_tokens": 10432454.0,
      "step": 5325
    },
    {
      "entropy": 6.313589382171631,
      "epoch": 0.5328135152696556,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0004978815841927061,
      "loss": 6.2813,
      "mean_token_accuracy": 0.14135064482688903,
      "num_tokens": 10443625.0,
      "step": 5330
    },
    {
      "entropy": 6.43781361579895,
      "epoch": 0.5333133403308842,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978766955369315,
      "loss": 6.3521,
      "mean_token_accuracy": 0.1418583534657955,
      "num_tokens": 10454055.0,
      "step": 5335
    },
    {
      "entropy": 6.324926519393921,
      "epoch": 0.5338131653921128,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004978718012736367,
      "loss": 6.1918,
      "mean_token_accuracy": 0.14811479896306992,
      "num_tokens": 10462683.0,
      "step": 5340
    },
    {
      "entropy": 6.282338714599609,
      "epoch": 0.5343129904533414,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004978669014029446,
      "loss": 6.2115,
      "mean_token_accuracy": 0.1487436003983021,
      "num_tokens": 10471699.0,
      "step": 5345
    },
    {
      "entropy": 6.296060609817505,
      "epoch": 0.5348128155145699,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004978619959249787,
      "loss": 6.1496,
      "mean_token_accuracy": 0.15180823877453803,
      "num_tokens": 10480790.0,
      "step": 5350
    },
    {
      "entropy": 6.321809577941894,
      "epoch": 0.5353126405757984,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004978570848398623,
      "loss": 6.2535,
      "mean_token_accuracy": 0.1351524792611599,
      "num_tokens": 10490202.0,
      "step": 5355
    },
    {
      "entropy": 6.3398229598999025,
      "epoch": 0.535812465637027,
      "grad_norm": 1.078125,
      "learning_rate": 0.000497852168147719,
      "loss": 6.3211,
      "mean_token_accuracy": 0.13673623278737068,
      "num_tokens": 10499690.0,
      "step": 5360
    },
    {
      "entropy": 6.41335916519165,
      "epoch": 0.5363122906982556,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004978472458486724,
      "loss": 6.2927,
      "mean_token_accuracy": 0.13549100011587142,
      "num_tokens": 10510120.0,
      "step": 5365
    },
    {
      "entropy": 6.417970085144043,
      "epoch": 0.5368121157594842,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004978423179428466,
      "loss": 6.4271,
      "mean_token_accuracy": 0.133928569406271,
      "num_tokens": 10520521.0,
      "step": 5370
    },
    {
      "entropy": 6.289327335357666,
      "epoch": 0.5373119408207128,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004978373844303653,
      "loss": 6.1127,
      "mean_token_accuracy": 0.14804793000221253,
      "num_tokens": 10530233.0,
      "step": 5375
    },
    {
      "entropy": 6.345554542541504,
      "epoch": 0.5378117658819414,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004978324453113528,
      "loss": 6.2398,
      "mean_token_accuracy": 0.1437473401427269,
      "num_tokens": 10539683.0,
      "step": 5380
    },
    {
      "entropy": 6.34681248664856,
      "epoch": 0.5383115909431699,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978275005859333,
      "loss": 6.2695,
      "mean_token_accuracy": 0.14396946206688882,
      "num_tokens": 10549483.0,
      "step": 5385
    },
    {
      "entropy": 6.408108568191528,
      "epoch": 0.5388114160043984,
      "grad_norm": 0.984375,
      "learning_rate": 0.0004978225502542313,
      "loss": 6.3746,
      "mean_token_accuracy": 0.1334490619599819,
      "num_tokens": 10560464.0,
      "step": 5390
    },
    {
      "entropy": 6.3947995662689205,
      "epoch": 0.539311241065627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004978175943163712,
      "loss": 6.2731,
      "mean_token_accuracy": 0.13974858000874518,
      "num_tokens": 10568660.0,
      "step": 5395
    },
    {
      "entropy": 6.407984018325806,
      "epoch": 0.5398110661268556,
      "grad_norm": 1.125,
      "learning_rate": 0.0004978126327724777,
      "loss": 6.337,
      "mean_token_accuracy": 0.1405469924211502,
      "num_tokens": 10577304.0,
      "step": 5400
    },
    {
      "entropy": 6.36352276802063,
      "epoch": 0.5403108911880842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004978076656226758,
      "loss": 6.2301,
      "mean_token_accuracy": 0.1479938320815563,
      "num_tokens": 10587591.0,
      "step": 5405
    },
    {
      "entropy": 6.3344934463500975,
      "epoch": 0.5408107162493128,
      "grad_norm": 2.140625,
      "learning_rate": 0.0004978026928670902,
      "loss": 6.1936,
      "mean_token_accuracy": 0.14546226859092712,
      "num_tokens": 10598591.0,
      "step": 5410
    },
    {
      "entropy": 6.2777117729187015,
      "epoch": 0.5413105413105413,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004977977145058463,
      "loss": 6.2212,
      "mean_token_accuracy": 0.13852583095431328,
      "num_tokens": 10608842.0,
      "step": 5415
    },
    {
      "entropy": 6.30012698173523,
      "epoch": 0.5418103663717699,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977927305390693,
      "loss": 6.3043,
      "mean_token_accuracy": 0.14666473120450974,
      "num_tokens": 10618812.0,
      "step": 5420
    },
    {
      "entropy": 6.421801662445068,
      "epoch": 0.5423101914329984,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004977877409668845,
      "loss": 6.4107,
      "mean_token_accuracy": 0.13223664835095406,
      "num_tokens": 10629244.0,
      "step": 5425
    },
    {
      "entropy": 6.456833124160767,
      "epoch": 0.542810016494227,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004977827457894174,
      "loss": 6.3447,
      "mean_token_accuracy": 0.13367147296667098,
      "num_tokens": 10638931.0,
      "step": 5430
    },
    {
      "entropy": 6.362542581558228,
      "epoch": 0.5433098415554556,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977777450067937,
      "loss": 6.2081,
      "mean_token_accuracy": 0.14372213631868364,
      "num_tokens": 10648754.0,
      "step": 5435
    },
    {
      "entropy": 6.285918760299682,
      "epoch": 0.5438096666166842,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004977727386191393,
      "loss": 6.187,
      "mean_token_accuracy": 0.14815898686647416,
      "num_tokens": 10657716.0,
      "step": 5440
    },
    {
      "entropy": 6.323311853408813,
      "epoch": 0.5443094916779128,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977677266265802,
      "loss": 6.2191,
      "mean_token_accuracy": 0.14353563338518144,
      "num_tokens": 10667621.0,
      "step": 5445
    },
    {
      "entropy": 6.348590230941772,
      "epoch": 0.5448093167391413,
      "grad_norm": 1.125,
      "learning_rate": 0.0004977627090292424,
      "loss": 6.1981,
      "mean_token_accuracy": 0.14093047380447388,
      "num_tokens": 10676640.0,
      "step": 5450
    },
    {
      "entropy": 6.300322103500366,
      "epoch": 0.5453091418003698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004977576858272521,
      "loss": 6.3171,
      "mean_token_accuracy": 0.13745146691799165,
      "num_tokens": 10687034.0,
      "step": 5455
    },
    {
      "entropy": 6.372998762130737,
      "epoch": 0.5458089668615984,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004977526570207358,
      "loss": 6.2001,
      "mean_token_accuracy": 0.14638773947954178,
      "num_tokens": 10696234.0,
      "step": 5460
    },
    {
      "entropy": 6.274461793899536,
      "epoch": 0.546308791922827,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004977476226098199,
      "loss": 6.2148,
      "mean_token_accuracy": 0.1524929493665695,
      "num_tokens": 10705733.0,
      "step": 5465
    },
    {
      "entropy": 6.352570295333862,
      "epoch": 0.5468086169840556,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004977425825946312,
      "loss": 6.249,
      "mean_token_accuracy": 0.14258571341633797,
      "num_tokens": 10714628.0,
      "step": 5470
    },
    {
      "entropy": 6.3967187881469725,
      "epoch": 0.5473084420452842,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004977375369752964,
      "loss": 6.4011,
      "mean_token_accuracy": 0.1309283420443535,
      "num_tokens": 10726120.0,
      "step": 5475
    },
    {
      "entropy": 6.370625019073486,
      "epoch": 0.5478082671065128,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004977324857519425,
      "loss": 6.2556,
      "mean_token_accuracy": 0.14103192687034607,
      "num_tokens": 10735976.0,
      "step": 5480
    },
    {
      "entropy": 6.377379703521728,
      "epoch": 0.5483080921677413,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977274289246967,
      "loss": 6.3458,
      "mean_token_accuracy": 0.13264099061489104,
      "num_tokens": 10745730.0,
      "step": 5485
    },
    {
      "entropy": 6.3523273944854735,
      "epoch": 0.5488079172289698,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004977223664936859,
      "loss": 6.2485,
      "mean_token_accuracy": 0.14141613021492958,
      "num_tokens": 10755435.0,
      "step": 5490
    },
    {
      "entropy": 6.3246574878692625,
      "epoch": 0.5493077422901984,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977172984590378,
      "loss": 6.1592,
      "mean_token_accuracy": 0.1442379593849182,
      "num_tokens": 10765483.0,
      "step": 5495
    },
    {
      "entropy": 6.300998735427856,
      "epoch": 0.549807567351427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004977122248208798,
      "loss": 6.2384,
      "mean_token_accuracy": 0.14392194896936417,
      "num_tokens": 10775034.0,
      "step": 5500
    },
    {
      "entropy": 6.292659139633178,
      "epoch": 0.5503073924126556,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004977071455793394,
      "loss": 6.2054,
      "mean_token_accuracy": 0.14822374880313874,
      "num_tokens": 10784799.0,
      "step": 5505
    },
    {
      "entropy": 6.379563999176026,
      "epoch": 0.5508072174738842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004977020607345445,
      "loss": 6.3093,
      "mean_token_accuracy": 0.1327025532722473,
      "num_tokens": 10794907.0,
      "step": 5510
    },
    {
      "entropy": 6.31625804901123,
      "epoch": 0.5513070425351128,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976969702866231,
      "loss": 6.2059,
      "mean_token_accuracy": 0.14766609147191048,
      "num_tokens": 10804374.0,
      "step": 5515
    },
    {
      "entropy": 6.332025003433228,
      "epoch": 0.5518068675963412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004976918742357033,
      "loss": 6.2386,
      "mean_token_accuracy": 0.1420293740928173,
      "num_tokens": 10814245.0,
      "step": 5520
    },
    {
      "entropy": 6.282225131988525,
      "epoch": 0.5523066926575698,
      "grad_norm": 1.0,
      "learning_rate": 0.0004976867725819133,
      "loss": 6.2497,
      "mean_token_accuracy": 0.1439930096268654,
      "num_tokens": 10824667.0,
      "step": 5525
    },
    {
      "entropy": 6.322996044158936,
      "epoch": 0.5528065177187984,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976816653253813,
      "loss": 6.196,
      "mean_token_accuracy": 0.14456692785024644,
      "num_tokens": 10834890.0,
      "step": 5530
    },
    {
      "entropy": 6.33039288520813,
      "epoch": 0.553306342780027,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976765524662359,
      "loss": 6.1911,
      "mean_token_accuracy": 0.14612701460719107,
      "num_tokens": 10844751.0,
      "step": 5535
    },
    {
      "entropy": 6.375986814498901,
      "epoch": 0.5538061678412556,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004976714340046058,
      "loss": 6.3259,
      "mean_token_accuracy": 0.14123422801494598,
      "num_tokens": 10854113.0,
      "step": 5540
    },
    {
      "entropy": 6.272052812576294,
      "epoch": 0.5543059929024842,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004976663099406196,
      "loss": 6.162,
      "mean_token_accuracy": 0.14373724684119224,
      "num_tokens": 10863468.0,
      "step": 5545
    },
    {
      "entropy": 6.368210744857788,
      "epoch": 0.5548058179637128,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004976611802744066,
      "loss": 6.2815,
      "mean_token_accuracy": 0.14646353498101233,
      "num_tokens": 10873381.0,
      "step": 5550
    },
    {
      "entropy": 6.386367702484131,
      "epoch": 0.5553056430249412,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004976560450060954,
      "loss": 6.2059,
      "mean_token_accuracy": 0.14763978868722916,
      "num_tokens": 10883457.0,
      "step": 5555
    },
    {
      "entropy": 6.244527053833008,
      "epoch": 0.5558054680861698,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004976509041358154,
      "loss": 6.1734,
      "mean_token_accuracy": 0.14715434312820436,
      "num_tokens": 10892822.0,
      "step": 5560
    },
    {
      "entropy": 6.289275741577148,
      "epoch": 0.5563052931473984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000497645757663696,
      "loss": 6.1534,
      "mean_token_accuracy": 0.15133251026272773,
      "num_tokens": 10901935.0,
      "step": 5565
    },
    {
      "entropy": 6.2485250473022464,
      "epoch": 0.556805118208627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004976406055898665,
      "loss": 6.2452,
      "mean_token_accuracy": 0.14060668796300888,
      "num_tokens": 10911863.0,
      "step": 5570
    },
    {
      "entropy": 6.329042434692383,
      "epoch": 0.5573049432698556,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976354479144568,
      "loss": 6.2083,
      "mean_token_accuracy": 0.14429977908730507,
      "num_tokens": 10921196.0,
      "step": 5575
    },
    {
      "entropy": 6.333301210403443,
      "epoch": 0.5578047683310842,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004976302846375965,
      "loss": 6.1839,
      "mean_token_accuracy": 0.14238550141453743,
      "num_tokens": 10931374.0,
      "step": 5580
    },
    {
      "entropy": 6.400436305999756,
      "epoch": 0.5583045933923126,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004976251157594155,
      "loss": 6.3403,
      "mean_token_accuracy": 0.1371694512665272,
      "num_tokens": 10941002.0,
      "step": 5585
    },
    {
      "entropy": 6.418968963623047,
      "epoch": 0.5588044184535412,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004976199412800437,
      "loss": 6.2727,
      "mean_token_accuracy": 0.14011965394020082,
      "num_tokens": 10950602.0,
      "step": 5590
    },
    {
      "entropy": 6.366796064376831,
      "epoch": 0.5593042435147698,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004976147611996118,
      "loss": 6.2315,
      "mean_token_accuracy": 0.14264914765954018,
      "num_tokens": 10960363.0,
      "step": 5595
    },
    {
      "entropy": 6.369522333145142,
      "epoch": 0.5598040685759984,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004976095755182496,
      "loss": 6.3354,
      "mean_token_accuracy": 0.13463364765048028,
      "num_tokens": 10970179.0,
      "step": 5600
    },
    {
      "entropy": 6.314689636230469,
      "epoch": 0.560303893637227,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004976043842360877,
      "loss": 6.2307,
      "mean_token_accuracy": 0.1400937870144844,
      "num_tokens": 10980446.0,
      "step": 5605
    },
    {
      "entropy": 6.312025690078736,
      "epoch": 0.5608037186984556,
      "grad_norm": 1.109375,
      "learning_rate": 0.000497599187353257,
      "loss": 6.2782,
      "mean_token_accuracy": 0.1409224160015583,
      "num_tokens": 10990883.0,
      "step": 5610
    },
    {
      "entropy": 6.3605124950408936,
      "epoch": 0.5613035437596842,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004975939848698879,
      "loss": 6.1799,
      "mean_token_accuracy": 0.1477658122777939,
      "num_tokens": 11000956.0,
      "step": 5615
    },
    {
      "entropy": 6.2973676204681395,
      "epoch": 0.5618033688209126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004975887767861113,
      "loss": 6.2397,
      "mean_token_accuracy": 0.14062583595514297,
      "num_tokens": 11010873.0,
      "step": 5620
    },
    {
      "entropy": 6.3561375617980955,
      "epoch": 0.5623031938821412,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004975835631020585,
      "loss": 6.2065,
      "mean_token_accuracy": 0.14184274300932884,
      "num_tokens": 11020192.0,
      "step": 5625
    },
    {
      "entropy": 6.32411642074585,
      "epoch": 0.5628030189433698,
      "grad_norm": 1.125,
      "learning_rate": 0.0004975783438178605,
      "loss": 6.2395,
      "mean_token_accuracy": 0.14021152928471564,
      "num_tokens": 11030466.0,
      "step": 5630
    },
    {
      "entropy": 6.265923690795899,
      "epoch": 0.5633028440045984,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004975731189336487,
      "loss": 6.2029,
      "mean_token_accuracy": 0.14996592849493026,
      "num_tokens": 11040309.0,
      "step": 5635
    },
    {
      "entropy": 6.338271522521973,
      "epoch": 0.563802669065827,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004975678884495544,
      "loss": 6.2921,
      "mean_token_accuracy": 0.14094484001398086,
      "num_tokens": 11049560.0,
      "step": 5640
    },
    {
      "entropy": 6.419568157196045,
      "epoch": 0.5643024941270556,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004975626523657093,
      "loss": 6.3675,
      "mean_token_accuracy": 0.12910672202706336,
      "num_tokens": 11059352.0,
      "step": 5645
    },
    {
      "entropy": 6.401751756668091,
      "epoch": 0.564802319188284,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004975574106822451,
      "loss": 6.2646,
      "mean_token_accuracy": 0.1445077180862427,
      "num_tokens": 11069159.0,
      "step": 5650
    },
    {
      "entropy": 6.341478443145752,
      "epoch": 0.5653021442495126,
      "grad_norm": 1.125,
      "learning_rate": 0.0004975521633992938,
      "loss": 6.3459,
      "mean_token_accuracy": 0.13364666253328322,
      "num_tokens": 11078960.0,
      "step": 5655
    },
    {
      "entropy": 6.35322265625,
      "epoch": 0.5658019693107412,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004975469105169873,
      "loss": 6.1762,
      "mean_token_accuracy": 0.15067835450172423,
      "num_tokens": 11089408.0,
      "step": 5660
    },
    {
      "entropy": 6.391804647445679,
      "epoch": 0.5663017943719698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004975416520354579,
      "loss": 6.3354,
      "mean_token_accuracy": 0.13959866464138032,
      "num_tokens": 11099485.0,
      "step": 5665
    },
    {
      "entropy": 6.334350442886352,
      "epoch": 0.5668016194331984,
      "grad_norm": 1.25,
      "learning_rate": 0.0004975363879548377,
      "loss": 6.351,
      "mean_token_accuracy": 0.1425812341272831,
      "num_tokens": 11109568.0,
      "step": 5670
    },
    {
      "entropy": 6.533071231842041,
      "epoch": 0.567301444494427,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004975311182752592,
      "loss": 6.4188,
      "mean_token_accuracy": 0.1318359337747097,
      "num_tokens": 11119433.0,
      "step": 5675
    },
    {
      "entropy": 6.496234512329101,
      "epoch": 0.5678012695556556,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004975258429968551,
      "loss": 6.3272,
      "mean_token_accuracy": 0.13728075623512268,
      "num_tokens": 11130060.0,
      "step": 5680
    },
    {
      "entropy": 6.396345376968384,
      "epoch": 0.568301094616884,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004975205621197581,
      "loss": 6.3773,
      "mean_token_accuracy": 0.13217183873057364,
      "num_tokens": 11140417.0,
      "step": 5685
    },
    {
      "entropy": 6.433355045318604,
      "epoch": 0.5688009196781126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004975152756441009,
      "loss": 6.3254,
      "mean_token_accuracy": 0.1399938814342022,
      "num_tokens": 11150830.0,
      "step": 5690
    },
    {
      "entropy": 6.3664405822753904,
      "epoch": 0.5693007447393412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004975099835700166,
      "loss": 6.1443,
      "mean_token_accuracy": 0.15007183998823165,
      "num_tokens": 11161057.0,
      "step": 5695
    },
    {
      "entropy": 6.253142976760865,
      "epoch": 0.5698005698005698,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004975046858976386,
      "loss": 6.2196,
      "mean_token_accuracy": 0.14115416407585143,
      "num_tokens": 11170182.0,
      "step": 5700
    },
    {
      "entropy": 6.260078001022339,
      "epoch": 0.5703003948617984,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004974993826270997,
      "loss": 6.1676,
      "mean_token_accuracy": 0.14473040029406548,
      "num_tokens": 11179962.0,
      "step": 5705
    },
    {
      "entropy": 6.33286337852478,
      "epoch": 0.570800219923027,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004974940737585337,
      "loss": 6.1085,
      "mean_token_accuracy": 0.15513153970241547,
      "num_tokens": 11189341.0,
      "step": 5710
    },
    {
      "entropy": 6.2625147819519045,
      "epoch": 0.5713000449842555,
      "grad_norm": 1.203125,
      "learning_rate": 0.000497488759292074,
      "loss": 6.2502,
      "mean_token_accuracy": 0.1522951252758503,
      "num_tokens": 11199264.0,
      "step": 5715
    },
    {
      "entropy": 6.2431128978729244,
      "epoch": 0.571799870045484,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004974834392278545,
      "loss": 6.2419,
      "mean_token_accuracy": 0.1471673645079136,
      "num_tokens": 11209257.0,
      "step": 5720
    },
    {
      "entropy": 6.395171689987182,
      "epoch": 0.5722996951067126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974781135660088,
      "loss": 6.2897,
      "mean_token_accuracy": 0.14230490028858184,
      "num_tokens": 11218294.0,
      "step": 5725
    },
    {
      "entropy": 6.288343524932861,
      "epoch": 0.5727995201679412,
      "grad_norm": 1.25,
      "learning_rate": 0.0004974727823066709,
      "loss": 6.1347,
      "mean_token_accuracy": 0.15037455558776855,
      "num_tokens": 11227684.0,
      "step": 5730
    },
    {
      "entropy": 6.406284475326538,
      "epoch": 0.5732993452291698,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004974674454499753,
      "loss": 6.3439,
      "mean_token_accuracy": 0.1382959380745888,
      "num_tokens": 11237195.0,
      "step": 5735
    },
    {
      "entropy": 6.413314390182495,
      "epoch": 0.5737991702903984,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004974621029960559,
      "loss": 6.311,
      "mean_token_accuracy": 0.13842954337596894,
      "num_tokens": 11246871.0,
      "step": 5740
    },
    {
      "entropy": 6.3722772121429445,
      "epoch": 0.574298995351627,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004974567549450472,
      "loss": 6.1532,
      "mean_token_accuracy": 0.15091011747717858,
      "num_tokens": 11256298.0,
      "step": 5745
    },
    {
      "entropy": 6.4279319763183596,
      "epoch": 0.5747988204128555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004974514012970839,
      "loss": 6.3627,
      "mean_token_accuracy": 0.13900440111756324,
      "num_tokens": 11265857.0,
      "step": 5750
    },
    {
      "entropy": 6.272759199142456,
      "epoch": 0.575298645474084,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004974460420523005,
      "loss": 6.1397,
      "mean_token_accuracy": 0.14941828846931457,
      "num_tokens": 11275406.0,
      "step": 5755
    },
    {
      "entropy": 6.309527015686035,
      "epoch": 0.5757984705353126,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004974406772108319,
      "loss": 6.1391,
      "mean_token_accuracy": 0.14953822940587996,
      "num_tokens": 11283910.0,
      "step": 5760
    },
    {
      "entropy": 6.307175874710083,
      "epoch": 0.5762982955965412,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004974353067728132,
      "loss": 6.2019,
      "mean_token_accuracy": 0.14291139543056489,
      "num_tokens": 11293808.0,
      "step": 5765
    },
    {
      "entropy": 6.313005542755127,
      "epoch": 0.5767981206577698,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004974299307383794,
      "loss": 6.244,
      "mean_token_accuracy": 0.14377905800938606,
      "num_tokens": 11302519.0,
      "step": 5770
    },
    {
      "entropy": 6.399163722991943,
      "epoch": 0.5772979457189984,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004974245491076656,
      "loss": 6.3541,
      "mean_token_accuracy": 0.135417915135622,
      "num_tokens": 11312609.0,
      "step": 5775
    },
    {
      "entropy": 6.3678234100341795,
      "epoch": 0.5777977707802269,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004974191618808075,
      "loss": 6.2845,
      "mean_token_accuracy": 0.1417811058461666,
      "num_tokens": 11322471.0,
      "step": 5780
    },
    {
      "entropy": 6.440444850921631,
      "epoch": 0.5782975958414555,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004974137690579405,
      "loss": 6.2783,
      "mean_token_accuracy": 0.14394563660025597,
      "num_tokens": 11333332.0,
      "step": 5785
    },
    {
      "entropy": 6.370710325241089,
      "epoch": 0.578797420902684,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004974083706392004,
      "loss": 6.2827,
      "mean_token_accuracy": 0.13744583651423453,
      "num_tokens": 11342952.0,
      "step": 5790
    },
    {
      "entropy": 6.3380354881286625,
      "epoch": 0.5792972459639126,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004974029666247228,
      "loss": 6.3595,
      "mean_token_accuracy": 0.13167960494756697,
      "num_tokens": 11353509.0,
      "step": 5795
    },
    {
      "entropy": 6.501959228515625,
      "epoch": 0.5797970710251412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004973975570146438,
      "loss": 6.3631,
      "mean_token_accuracy": 0.1348981648683548,
      "num_tokens": 11363065.0,
      "step": 5800
    },
    {
      "entropy": 6.295175361633301,
      "epoch": 0.5802968960863698,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004973921418090997,
      "loss": 6.2263,
      "mean_token_accuracy": 0.14342114701867104,
      "num_tokens": 11373720.0,
      "step": 5805
    },
    {
      "entropy": 6.239151048660278,
      "epoch": 0.5807967211475984,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004973867210082263,
      "loss": 6.2176,
      "mean_token_accuracy": 0.1432886928319931,
      "num_tokens": 11384432.0,
      "step": 5810
    },
    {
      "entropy": 6.3962414264678955,
      "epoch": 0.5812965462088269,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004973812946121603,
      "loss": 6.3836,
      "mean_token_accuracy": 0.13938908129930497,
      "num_tokens": 11394100.0,
      "step": 5815
    },
    {
      "entropy": 6.33066029548645,
      "epoch": 0.5817963712700555,
      "grad_norm": 1.25,
      "learning_rate": 0.0004973758626210382,
      "loss": 6.2488,
      "mean_token_accuracy": 0.1417160674929619,
      "num_tokens": 11403359.0,
      "step": 5820
    },
    {
      "entropy": 6.364110088348388,
      "epoch": 0.582296196331284,
      "grad_norm": 1.25,
      "learning_rate": 0.0004973704250349965,
      "loss": 6.2859,
      "mean_token_accuracy": 0.1334216959774494,
      "num_tokens": 11412921.0,
      "step": 5825
    },
    {
      "entropy": 6.37510929107666,
      "epoch": 0.5827960213925126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973649818541722,
      "loss": 6.2338,
      "mean_token_accuracy": 0.13970995396375657,
      "num_tokens": 11422459.0,
      "step": 5830
    },
    {
      "entropy": 6.358178663253784,
      "epoch": 0.5832958464537412,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004973595330787022,
      "loss": 6.1865,
      "mean_token_accuracy": 0.1506979987025261,
      "num_tokens": 11431517.0,
      "step": 5835
    },
    {
      "entropy": 6.336723709106446,
      "epoch": 0.5837956715149698,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004973540787087235,
      "loss": 6.2381,
      "mean_token_accuracy": 0.14081623926758766,
      "num_tokens": 11440662.0,
      "step": 5840
    },
    {
      "entropy": 6.235829639434814,
      "epoch": 0.5842954965761983,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004973486187443734,
      "loss": 6.0816,
      "mean_token_accuracy": 0.15424600690603257,
      "num_tokens": 11449762.0,
      "step": 5845
    },
    {
      "entropy": 6.179425573348999,
      "epoch": 0.5847953216374269,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004973431531857893,
      "loss": 6.1006,
      "mean_token_accuracy": 0.15300129354000092,
      "num_tokens": 11459963.0,
      "step": 5850
    },
    {
      "entropy": 6.3275891780853275,
      "epoch": 0.5852951466986555,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004973376820331087,
      "loss": 6.2651,
      "mean_token_accuracy": 0.14004729613661765,
      "num_tokens": 11469063.0,
      "step": 5855
    },
    {
      "entropy": 6.344074344635009,
      "epoch": 0.585794971759884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004973322052864692,
      "loss": 6.1642,
      "mean_token_accuracy": 0.15059844255447388,
      "num_tokens": 11477651.0,
      "step": 5860
    },
    {
      "entropy": 6.4009171485900875,
      "epoch": 0.5862947968211126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004973267229460087,
      "loss": 6.3628,
      "mean_token_accuracy": 0.1331980399787426,
      "num_tokens": 11487248.0,
      "step": 5865
    },
    {
      "entropy": 6.331179523468018,
      "epoch": 0.5867946218823412,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000497321235011865,
      "loss": 6.3433,
      "mean_token_accuracy": 0.1435278907418251,
      "num_tokens": 11497066.0,
      "step": 5870
    },
    {
      "entropy": 6.4113420963287355,
      "epoch": 0.5872944469435698,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004973157414841763,
      "loss": 6.3115,
      "mean_token_accuracy": 0.13302920013666153,
      "num_tokens": 11508490.0,
      "step": 5875
    },
    {
      "entropy": 6.369688367843628,
      "epoch": 0.5877942720047983,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004973102423630806,
      "loss": 6.2994,
      "mean_token_accuracy": 0.14273363649845122,
      "num_tokens": 11518851.0,
      "step": 5880
    },
    {
      "entropy": 6.323107671737671,
      "epoch": 0.5882940970660269,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004973047376487166,
      "loss": 6.2629,
      "mean_token_accuracy": 0.140333840996027,
      "num_tokens": 11529102.0,
      "step": 5885
    },
    {
      "entropy": 6.341410970687866,
      "epoch": 0.5887939221272555,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004972992273412224,
      "loss": 6.2889,
      "mean_token_accuracy": 0.13700356483459472,
      "num_tokens": 11540563.0,
      "step": 5890
    },
    {
      "entropy": 6.337323474884033,
      "epoch": 0.589293747188484,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004972937114407369,
      "loss": 6.2045,
      "mean_token_accuracy": 0.1455566741526127,
      "num_tokens": 11549955.0,
      "step": 5895
    },
    {
      "entropy": 6.304736471176147,
      "epoch": 0.5897935722497126,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000497288189947399,
      "loss": 6.1877,
      "mean_token_accuracy": 0.14470466002821922,
      "num_tokens": 11558932.0,
      "step": 5900
    },
    {
      "entropy": 6.216138076782227,
      "epoch": 0.5902933973109412,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004972826628613472,
      "loss": 6.2009,
      "mean_token_accuracy": 0.14667390063405036,
      "num_tokens": 11568901.0,
      "step": 5905
    },
    {
      "entropy": 6.339971828460693,
      "epoch": 0.5907932223721697,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004972771301827209,
      "loss": 6.2558,
      "mean_token_accuracy": 0.14320868328213693,
      "num_tokens": 11579738.0,
      "step": 5910
    },
    {
      "entropy": 6.384644412994385,
      "epoch": 0.5912930474333983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004972715919116592,
      "loss": 6.2366,
      "mean_token_accuracy": 0.14273437932133676,
      "num_tokens": 11590293.0,
      "step": 5915
    },
    {
      "entropy": 6.334532880783081,
      "epoch": 0.5917928724946269,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004972660480483015,
      "loss": 6.3015,
      "mean_token_accuracy": 0.13526504263281822,
      "num_tokens": 11599745.0,
      "step": 5920
    },
    {
      "entropy": 6.35646915435791,
      "epoch": 0.5922926975558555,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004972604985927871,
      "loss": 6.29,
      "mean_token_accuracy": 0.13545462265610694,
      "num_tokens": 11609958.0,
      "step": 5925
    },
    {
      "entropy": 6.412453746795654,
      "epoch": 0.592792522617084,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004972549435452558,
      "loss": 6.2698,
      "mean_token_accuracy": 0.1367018401622772,
      "num_tokens": 11619941.0,
      "step": 5930
    },
    {
      "entropy": 6.296717691421509,
      "epoch": 0.5932923476783126,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004972493829058474,
      "loss": 6.1477,
      "mean_token_accuracy": 0.1530252903699875,
      "num_tokens": 11630287.0,
      "step": 5935
    },
    {
      "entropy": 6.32022614479065,
      "epoch": 0.5937921727395412,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004972438166747016,
      "loss": 6.2019,
      "mean_token_accuracy": 0.14405305236577987,
      "num_tokens": 11639806.0,
      "step": 5940
    },
    {
      "entropy": 6.295377397537232,
      "epoch": 0.5942919978007697,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004972382448519586,
      "loss": 6.294,
      "mean_token_accuracy": 0.1430134057998657,
      "num_tokens": 11650132.0,
      "step": 5945
    },
    {
      "entropy": 6.320089721679688,
      "epoch": 0.5947918228619983,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004972326674377586,
      "loss": 6.21,
      "mean_token_accuracy": 0.1424887128174305,
      "num_tokens": 11659670.0,
      "step": 5950
    },
    {
      "entropy": 6.361303281784058,
      "epoch": 0.5952916479232269,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004972270844322418,
      "loss": 6.2706,
      "mean_token_accuracy": 0.14572205990552903,
      "num_tokens": 11669998.0,
      "step": 5955
    },
    {
      "entropy": 6.3370928287506105,
      "epoch": 0.5957914729844555,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004972214958355488,
      "loss": 6.2654,
      "mean_token_accuracy": 0.14094047322869302,
      "num_tokens": 11679052.0,
      "step": 5960
    },
    {
      "entropy": 6.328358221054077,
      "epoch": 0.596291298045684,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004972159016478201,
      "loss": 6.254,
      "mean_token_accuracy": 0.1448790781199932,
      "num_tokens": 11687957.0,
      "step": 5965
    },
    {
      "entropy": 6.362970972061158,
      "epoch": 0.5967911231069126,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004972103018691965,
      "loss": 6.1781,
      "mean_token_accuracy": 0.14631520956754684,
      "num_tokens": 11696922.0,
      "step": 5970
    },
    {
      "entropy": 6.295720767974854,
      "epoch": 0.5972909481681411,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004972046964998189,
      "loss": 6.1718,
      "mean_token_accuracy": 0.14752159640192986,
      "num_tokens": 11706600.0,
      "step": 5975
    },
    {
      "entropy": 6.2522577285766605,
      "epoch": 0.5977907732293697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004971990855398283,
      "loss": 6.2009,
      "mean_token_accuracy": 0.14315633177757264,
      "num_tokens": 11715933.0,
      "step": 5980
    },
    {
      "entropy": 6.289220237731934,
      "epoch": 0.5982905982905983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004971934689893658,
      "loss": 6.1558,
      "mean_token_accuracy": 0.1515709176659584,
      "num_tokens": 11724807.0,
      "step": 5985
    },
    {
      "entropy": 6.345755100250244,
      "epoch": 0.5987904233518269,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971878468485729,
      "loss": 6.2374,
      "mean_token_accuracy": 0.14083078131079674,
      "num_tokens": 11734985.0,
      "step": 5990
    },
    {
      "entropy": 6.343067836761475,
      "epoch": 0.5992902484130554,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000497182219117591,
      "loss": 6.223,
      "mean_token_accuracy": 0.14369984343647957,
      "num_tokens": 11743418.0,
      "step": 5995
    },
    {
      "entropy": 6.274335336685181,
      "epoch": 0.599790073474284,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004971765857965615,
      "loss": 6.1777,
      "mean_token_accuracy": 0.14245593249797822,
      "num_tokens": 11752953.0,
      "step": 6000
    },
    {
      "epoch": 0.599790073474284,
      "eval_entropy": 6.13612363670819,
      "eval_loss": 6.280503273010254,
      "eval_mean_token_accuracy": 0.14671862201803096,
      "eval_num_tokens": 11752953.0,
      "eval_runtime": 24.0294,
      "eval_samples_per_second": 1291.956,
      "eval_steps_per_second": 161.51,
      "step": 6000
    },
    {
      "entropy": 6.3440389156341555,
      "epoch": 0.6002898985355126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004971709468856263,
      "loss": 6.316,
      "mean_token_accuracy": 0.13903788477182388,
      "num_tokens": 11763260.0,
      "step": 6005
    },
    {
      "entropy": 6.242662239074707,
      "epoch": 0.6007897235967411,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004971653023849274,
      "loss": 6.0854,
      "mean_token_accuracy": 0.1641439639031887,
      "num_tokens": 11773255.0,
      "step": 6010
    },
    {
      "entropy": 6.377378034591675,
      "epoch": 0.6012895486579697,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004971596522946065,
      "loss": 6.2306,
      "mean_token_accuracy": 0.14113587737083436,
      "num_tokens": 11783443.0,
      "step": 6015
    },
    {
      "entropy": 6.2793535709381105,
      "epoch": 0.6017893737191983,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004971539966148059,
      "loss": 6.1885,
      "mean_token_accuracy": 0.14466064497828485,
      "num_tokens": 11791932.0,
      "step": 6020
    },
    {
      "entropy": 6.291140842437744,
      "epoch": 0.6022891987804269,
      "grad_norm": 1.125,
      "learning_rate": 0.0004971483353456681,
      "loss": 6.1631,
      "mean_token_accuracy": 0.14731817394495011,
      "num_tokens": 11801420.0,
      "step": 6025
    },
    {
      "entropy": 6.356692886352539,
      "epoch": 0.6027890238416554,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004971426684873352,
      "loss": 6.3165,
      "mean_token_accuracy": 0.13832596838474273,
      "num_tokens": 11810977.0,
      "step": 6030
    },
    {
      "entropy": 6.271351766586304,
      "epoch": 0.603288848902884,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004971369960399499,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1504375457763672,
      "num_tokens": 11821016.0,
      "step": 6035
    },
    {
      "entropy": 6.332382154464722,
      "epoch": 0.6037886739641125,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000497131318003655,
      "loss": 6.2628,
      "mean_token_accuracy": 0.13746074512600898,
      "num_tokens": 11831157.0,
      "step": 6040
    },
    {
      "entropy": 6.279897403717041,
      "epoch": 0.6042884990253411,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004971256343785932,
      "loss": 6.1728,
      "mean_token_accuracy": 0.14922841638326645,
      "num_tokens": 11840319.0,
      "step": 6045
    },
    {
      "entropy": 6.318679571151733,
      "epoch": 0.6047883240865697,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004971199451649078,
      "loss": 6.3085,
      "mean_token_accuracy": 0.13916947543621064,
      "num_tokens": 11850808.0,
      "step": 6050
    },
    {
      "entropy": 6.400442695617675,
      "epoch": 0.6052881491477983,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004971142503627415,
      "loss": 6.3542,
      "mean_token_accuracy": 0.14150375127792358,
      "num_tokens": 11861123.0,
      "step": 6055
    },
    {
      "entropy": 6.240767335891723,
      "epoch": 0.6057879742090269,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000497108549972238,
      "loss": 6.0891,
      "mean_token_accuracy": 0.14752269834280013,
      "num_tokens": 11870308.0,
      "step": 6060
    },
    {
      "entropy": 6.373895645141602,
      "epoch": 0.6062877992702554,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004971028439935404,
      "loss": 6.2318,
      "mean_token_accuracy": 0.13867005333304405,
      "num_tokens": 11880254.0,
      "step": 6065
    },
    {
      "entropy": 6.327100563049316,
      "epoch": 0.606787624331484,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004970971324267925,
      "loss": 6.2341,
      "mean_token_accuracy": 0.14898003488779069,
      "num_tokens": 11891272.0,
      "step": 6070
    },
    {
      "entropy": 6.312870216369629,
      "epoch": 0.6072874493927125,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004970914152721378,
      "loss": 6.2626,
      "mean_token_accuracy": 0.14210639670491218,
      "num_tokens": 11901176.0,
      "step": 6075
    },
    {
      "entropy": 6.399405097961425,
      "epoch": 0.6077872744539411,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004970856925297204,
      "loss": 6.2639,
      "mean_token_accuracy": 0.136920203268528,
      "num_tokens": 11910820.0,
      "step": 6080
    },
    {
      "entropy": 6.290783786773682,
      "epoch": 0.6082870995151697,
      "grad_norm": 1.296875,
      "learning_rate": 0.000497079964199684,
      "loss": 6.215,
      "mean_token_accuracy": 0.13901869729161262,
      "num_tokens": 11920458.0,
      "step": 6085
    },
    {
      "entropy": 6.176512384414673,
      "epoch": 0.6087869245763983,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970742302821729,
      "loss": 6.078,
      "mean_token_accuracy": 0.15024468004703523,
      "num_tokens": 11931150.0,
      "step": 6090
    },
    {
      "entropy": 6.303402709960937,
      "epoch": 0.6092867496376269,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970684907773312,
      "loss": 6.1903,
      "mean_token_accuracy": 0.14731646180152894,
      "num_tokens": 11939960.0,
      "step": 6095
    },
    {
      "entropy": 6.430849695205689,
      "epoch": 0.6097865746988554,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004970627456853036,
      "loss": 6.3687,
      "mean_token_accuracy": 0.13612013831734657,
      "num_tokens": 11950781.0,
      "step": 6100
    },
    {
      "entropy": 6.371213960647583,
      "epoch": 0.6102863997600839,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004970569950062343,
      "loss": 6.2074,
      "mean_token_accuracy": 0.1433712512254715,
      "num_tokens": 11960489.0,
      "step": 6105
    },
    {
      "entropy": 6.297654533386231,
      "epoch": 0.6107862248213125,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004970512387402682,
      "loss": 6.2333,
      "mean_token_accuracy": 0.1458951212465763,
      "num_tokens": 11970651.0,
      "step": 6110
    },
    {
      "entropy": 6.252086400985718,
      "epoch": 0.6112860498825411,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004970454768875501,
      "loss": 6.1962,
      "mean_token_accuracy": 0.14458660632371903,
      "num_tokens": 11980349.0,
      "step": 6115
    },
    {
      "entropy": 6.311388921737671,
      "epoch": 0.6117858749437697,
      "grad_norm": 1.125,
      "learning_rate": 0.0004970397094482249,
      "loss": 6.239,
      "mean_token_accuracy": 0.14557712599635125,
      "num_tokens": 11990076.0,
      "step": 6120
    },
    {
      "entropy": 6.281975936889649,
      "epoch": 0.6122857000049983,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004970339364224378,
      "loss": 6.1421,
      "mean_token_accuracy": 0.1484926924109459,
      "num_tokens": 11999515.0,
      "step": 6125
    },
    {
      "entropy": 6.315656471252441,
      "epoch": 0.6127855250662269,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004970281578103339,
      "loss": 6.2927,
      "mean_token_accuracy": 0.1360795460641384,
      "num_tokens": 12009224.0,
      "step": 6130
    },
    {
      "entropy": 6.254594802856445,
      "epoch": 0.6132853501274554,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004970223736120587,
      "loss": 6.158,
      "mean_token_accuracy": 0.14876777231693267,
      "num_tokens": 12019295.0,
      "step": 6135
    },
    {
      "entropy": 6.348979473114014,
      "epoch": 0.6137851751886839,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004970165838277578,
      "loss": 6.2436,
      "mean_token_accuracy": 0.13932792767882346,
      "num_tokens": 12029130.0,
      "step": 6140
    },
    {
      "entropy": 6.330397748947144,
      "epoch": 0.6142850002499125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004970107884575767,
      "loss": 6.1932,
      "mean_token_accuracy": 0.1471465826034546,
      "num_tokens": 12038676.0,
      "step": 6145
    },
    {
      "entropy": 6.27381443977356,
      "epoch": 0.6147848253111411,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004970049875016613,
      "loss": 6.1647,
      "mean_token_accuracy": 0.15583466589450837,
      "num_tokens": 12048089.0,
      "step": 6150
    },
    {
      "entropy": 6.316942548751831,
      "epoch": 0.6152846503723697,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004969991809601576,
      "loss": 6.2437,
      "mean_token_accuracy": 0.14197300672531127,
      "num_tokens": 12059531.0,
      "step": 6155
    },
    {
      "entropy": 6.361162328720093,
      "epoch": 0.6157844754335983,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004969933688332115,
      "loss": 6.2694,
      "mean_token_accuracy": 0.13537977114319802,
      "num_tokens": 12070103.0,
      "step": 6160
    },
    {
      "entropy": 6.284975576400757,
      "epoch": 0.6162843004948269,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004969875511209694,
      "loss": 6.2245,
      "mean_token_accuracy": 0.1425408072769642,
      "num_tokens": 12078481.0,
      "step": 6165
    },
    {
      "entropy": 6.3813670635223385,
      "epoch": 0.6167841255560553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004969817278235777,
      "loss": 6.3384,
      "mean_token_accuracy": 0.13671956434845925,
      "num_tokens": 12088875.0,
      "step": 6170
    },
    {
      "entropy": 6.381721639633179,
      "epoch": 0.6172839506172839,
      "grad_norm": 1.125,
      "learning_rate": 0.0004969758989411829,
      "loss": 6.2214,
      "mean_token_accuracy": 0.14257473722100258,
      "num_tokens": 12099769.0,
      "step": 6175
    },
    {
      "entropy": 6.35776538848877,
      "epoch": 0.6177837756785125,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004969700644739315,
      "loss": 6.1849,
      "mean_token_accuracy": 0.14221653789281846,
      "num_tokens": 12109334.0,
      "step": 6180
    },
    {
      "entropy": 6.349963665008545,
      "epoch": 0.6182836007397411,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969642244219705,
      "loss": 6.2111,
      "mean_token_accuracy": 0.14354735389351844,
      "num_tokens": 12119017.0,
      "step": 6185
    },
    {
      "entropy": 6.214911699295044,
      "epoch": 0.6187834258009697,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004969583787854467,
      "loss": 6.1207,
      "mean_token_accuracy": 0.15771229714155197,
      "num_tokens": 12128002.0,
      "step": 6190
    },
    {
      "entropy": 6.311770343780518,
      "epoch": 0.6192832508621983,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004969525275645071,
      "loss": 6.2245,
      "mean_token_accuracy": 0.1406985655426979,
      "num_tokens": 12137802.0,
      "step": 6195
    },
    {
      "entropy": 6.29865198135376,
      "epoch": 0.6197830759234269,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004969466707592991,
      "loss": 6.1841,
      "mean_token_accuracy": 0.14674678891897203,
      "num_tokens": 12147270.0,
      "step": 6200
    },
    {
      "entropy": 6.313567733764648,
      "epoch": 0.6202829009846553,
      "grad_norm": 1.203125,
      "learning_rate": 0.00049694080836997,
      "loss": 6.2237,
      "mean_token_accuracy": 0.14287682622671127,
      "num_tokens": 12157358.0,
      "step": 6205
    },
    {
      "entropy": 6.377605533599853,
      "epoch": 0.6207827260458839,
      "grad_norm": 1.25,
      "learning_rate": 0.0004969349403966672,
      "loss": 6.2886,
      "mean_token_accuracy": 0.14394927993416787,
      "num_tokens": 12166159.0,
      "step": 6210
    },
    {
      "entropy": 6.330509805679322,
      "epoch": 0.6212825511071125,
      "grad_norm": 1.125,
      "learning_rate": 0.0004969290668395385,
      "loss": 6.25,
      "mean_token_accuracy": 0.14399931356310844,
      "num_tokens": 12176405.0,
      "step": 6215
    },
    {
      "entropy": 6.404755306243897,
      "epoch": 0.6217823761683411,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004969231876987315,
      "loss": 6.2988,
      "mean_token_accuracy": 0.13984070494771003,
      "num_tokens": 12186021.0,
      "step": 6220
    },
    {
      "entropy": 6.312417984008789,
      "epoch": 0.6222822012295697,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004969173029743942,
      "loss": 6.1796,
      "mean_token_accuracy": 0.14615456983447075,
      "num_tokens": 12195222.0,
      "step": 6225
    },
    {
      "entropy": 6.289805746078491,
      "epoch": 0.6227820262907983,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004969114126666746,
      "loss": 6.1413,
      "mean_token_accuracy": 0.15038748309016228,
      "num_tokens": 12205443.0,
      "step": 6230
    },
    {
      "entropy": 6.365929222106933,
      "epoch": 0.6232818513520267,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004969055167757211,
      "loss": 6.3519,
      "mean_token_accuracy": 0.1405246950685978,
      "num_tokens": 12215181.0,
      "step": 6235
    },
    {
      "entropy": 6.365829992294311,
      "epoch": 0.6237816764132553,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004968996153016819,
      "loss": 6.2133,
      "mean_token_accuracy": 0.14518257826566697,
      "num_tokens": 12224327.0,
      "step": 6240
    },
    {
      "entropy": 6.293065214157105,
      "epoch": 0.6242815014744839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968937082447054,
      "loss": 6.2436,
      "mean_token_accuracy": 0.14280423074960708,
      "num_tokens": 12234962.0,
      "step": 6245
    },
    {
      "entropy": 6.341446447372436,
      "epoch": 0.6247813265357125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004968877956049403,
      "loss": 6.3128,
      "mean_token_accuracy": 0.14122455343604087,
      "num_tokens": 12245883.0,
      "step": 6250
    },
    {
      "entropy": 6.396516942977906,
      "epoch": 0.6252811515969411,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004968818773825355,
      "loss": 6.2943,
      "mean_token_accuracy": 0.13761424720287324,
      "num_tokens": 12254412.0,
      "step": 6255
    },
    {
      "entropy": 6.220240116119385,
      "epoch": 0.6257809766581697,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968759535776396,
      "loss": 6.1684,
      "mean_token_accuracy": 0.1464730307459831,
      "num_tokens": 12264894.0,
      "step": 6260
    },
    {
      "entropy": 6.33331789970398,
      "epoch": 0.6262808017193983,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004968700241904018,
      "loss": 6.2701,
      "mean_token_accuracy": 0.14257046431303025,
      "num_tokens": 12274557.0,
      "step": 6265
    },
    {
      "entropy": 6.336774826049805,
      "epoch": 0.6267806267806267,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004968640892209714,
      "loss": 6.0851,
      "mean_token_accuracy": 0.14928142875432968,
      "num_tokens": 12283696.0,
      "step": 6270
    },
    {
      "entropy": 6.241849374771118,
      "epoch": 0.6272804518418553,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968581486694975,
      "loss": 6.1659,
      "mean_token_accuracy": 0.14955419600009917,
      "num_tokens": 12292981.0,
      "step": 6275
    },
    {
      "entropy": 6.238350439071655,
      "epoch": 0.6277802769030839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968522025361297,
      "loss": 6.0721,
      "mean_token_accuracy": 0.14934796392917632,
      "num_tokens": 12302865.0,
      "step": 6280
    },
    {
      "entropy": 6.2630171298980715,
      "epoch": 0.6282801019643125,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004968462508210176,
      "loss": 6.223,
      "mean_token_accuracy": 0.1475556805729866,
      "num_tokens": 12313099.0,
      "step": 6285
    },
    {
      "entropy": 6.364171743392944,
      "epoch": 0.6287799270255411,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968402935243109,
      "loss": 6.2601,
      "mean_token_accuracy": 0.14661428183317185,
      "num_tokens": 12324155.0,
      "step": 6290
    },
    {
      "entropy": 6.240270185470581,
      "epoch": 0.6292797520867697,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004968343306461595,
      "loss": 6.0982,
      "mean_token_accuracy": 0.1493609756231308,
      "num_tokens": 12334082.0,
      "step": 6295
    },
    {
      "entropy": 6.24729905128479,
      "epoch": 0.6297795771479981,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004968283621867134,
      "loss": 6.198,
      "mean_token_accuracy": 0.14586353302001953,
      "num_tokens": 12344673.0,
      "step": 6300
    },
    {
      "entropy": 6.367838716506958,
      "epoch": 0.6302794022092267,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004968223881461227,
      "loss": 6.2032,
      "mean_token_accuracy": 0.14202526360750198,
      "num_tokens": 12355149.0,
      "step": 6305
    },
    {
      "entropy": 6.305402755737305,
      "epoch": 0.6307792272704553,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000496816408524538,
      "loss": 6.0996,
      "mean_token_accuracy": 0.15201123356819152,
      "num_tokens": 12364043.0,
      "step": 6310
    },
    {
      "entropy": 6.276588392257691,
      "epoch": 0.6312790523316839,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004968104233221094,
      "loss": 6.3046,
      "mean_token_accuracy": 0.1399997130036354,
      "num_tokens": 12373591.0,
      "step": 6315
    },
    {
      "entropy": 6.297060108184814,
      "epoch": 0.6317788773929125,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004968044325389877,
      "loss": 6.2424,
      "mean_token_accuracy": 0.14478263184428214,
      "num_tokens": 12383570.0,
      "step": 6320
    },
    {
      "entropy": 6.436004304885865,
      "epoch": 0.6322787024541411,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004967984361753235,
      "loss": 6.2261,
      "mean_token_accuracy": 0.14246728867292405,
      "num_tokens": 12393682.0,
      "step": 6325
    },
    {
      "entropy": 6.302444887161255,
      "epoch": 0.6327785275153697,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004967924342312677,
      "loss": 6.146,
      "mean_token_accuracy": 0.1438180610537529,
      "num_tokens": 12403478.0,
      "step": 6330
    },
    {
      "entropy": 6.281976413726807,
      "epoch": 0.6332783525765981,
      "grad_norm": 1.25,
      "learning_rate": 0.0004967864267069713,
      "loss": 6.2505,
      "mean_token_accuracy": 0.14182919263839722,
      "num_tokens": 12412810.0,
      "step": 6335
    },
    {
      "entropy": 6.266110277175903,
      "epoch": 0.6337781776378267,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967804136025856,
      "loss": 6.1394,
      "mean_token_accuracy": 0.15015392899513244,
      "num_tokens": 12422209.0,
      "step": 6340
    },
    {
      "entropy": 6.3207359790802,
      "epoch": 0.6342780026990553,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004967743949182617,
      "loss": 6.1998,
      "mean_token_accuracy": 0.14080120623111725,
      "num_tokens": 12431569.0,
      "step": 6345
    },
    {
      "entropy": 6.222105312347412,
      "epoch": 0.6347778277602839,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004967683706541511,
      "loss": 6.0672,
      "mean_token_accuracy": 0.16392299085855483,
      "num_tokens": 12441607.0,
      "step": 6350
    },
    {
      "entropy": 6.266947269439697,
      "epoch": 0.6352776528215125,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004967623408104054,
      "loss": 6.2071,
      "mean_token_accuracy": 0.1397886171936989,
      "num_tokens": 12451072.0,
      "step": 6355
    },
    {
      "entropy": 6.4045663356781,
      "epoch": 0.6357774778827411,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004967563053871762,
      "loss": 6.2353,
      "mean_token_accuracy": 0.14162094816565513,
      "num_tokens": 12460798.0,
      "step": 6360
    },
    {
      "entropy": 6.2242199897766115,
      "epoch": 0.6362773029439696,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967502643846155,
      "loss": 6.1594,
      "mean_token_accuracy": 0.15118182748556136,
      "num_tokens": 12471891.0,
      "step": 6365
    },
    {
      "entropy": 6.243682718276977,
      "epoch": 0.6367771280051981,
      "grad_norm": 1.25,
      "learning_rate": 0.000496744217802875,
      "loss": 6.1925,
      "mean_token_accuracy": 0.14185698032379152,
      "num_tokens": 12482212.0,
      "step": 6370
    },
    {
      "entropy": 6.332003164291382,
      "epoch": 0.6372769530664267,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004967381656421073,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14490965455770494,
      "num_tokens": 12492517.0,
      "step": 6375
    },
    {
      "entropy": 6.310923624038696,
      "epoch": 0.6377767781276553,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004967321079024643,
      "loss": 6.1312,
      "mean_token_accuracy": 0.14473602846264838,
      "num_tokens": 12502521.0,
      "step": 6380
    },
    {
      "entropy": 6.33732328414917,
      "epoch": 0.6382766031888839,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004967260445840985,
      "loss": 6.2676,
      "mean_token_accuracy": 0.14144754111766816,
      "num_tokens": 12513195.0,
      "step": 6385
    },
    {
      "entropy": 6.283482837677002,
      "epoch": 0.6387764282501125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004967199756871624,
      "loss": 6.0772,
      "mean_token_accuracy": 0.15662836655974388,
      "num_tokens": 12521640.0,
      "step": 6390
    },
    {
      "entropy": 6.282724618911743,
      "epoch": 0.6392762533113411,
      "grad_norm": 1.375,
      "learning_rate": 0.0004967139012118088,
      "loss": 6.2234,
      "mean_token_accuracy": 0.14944694191217422,
      "num_tokens": 12530807.0,
      "step": 6395
    },
    {
      "entropy": 6.32041449546814,
      "epoch": 0.6397760783725696,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004967078211581907,
      "loss": 6.2757,
      "mean_token_accuracy": 0.14207592457532883,
      "num_tokens": 12542049.0,
      "step": 6400
    },
    {
      "entropy": 6.326205015182495,
      "epoch": 0.6402759034337981,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004967017355264606,
      "loss": 6.2137,
      "mean_token_accuracy": 0.15072372779250146,
      "num_tokens": 12551699.0,
      "step": 6405
    },
    {
      "entropy": 6.312424230575561,
      "epoch": 0.6407757284950267,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496695644316772,
      "loss": 6.2969,
      "mean_token_accuracy": 0.14193230271339416,
      "num_tokens": 12561414.0,
      "step": 6410
    },
    {
      "entropy": 6.3347960948944095,
      "epoch": 0.6412755535562553,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004966895475292781,
      "loss": 6.2711,
      "mean_token_accuracy": 0.14173732250928878,
      "num_tokens": 12571950.0,
      "step": 6415
    },
    {
      "entropy": 6.4350605487823485,
      "epoch": 0.6417753786174839,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004966834451641322,
      "loss": 6.2958,
      "mean_token_accuracy": 0.14328024685382842,
      "num_tokens": 12581264.0,
      "step": 6420
    },
    {
      "entropy": 6.333584213256836,
      "epoch": 0.6422752036787125,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966773372214878,
      "loss": 6.2398,
      "mean_token_accuracy": 0.14176533818244935,
      "num_tokens": 12590950.0,
      "step": 6425
    },
    {
      "entropy": 6.285247850418091,
      "epoch": 0.642775028739941,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966712237014987,
      "loss": 6.182,
      "mean_token_accuracy": 0.1438572585582733,
      "num_tokens": 12599766.0,
      "step": 6430
    },
    {
      "entropy": 6.387751054763794,
      "epoch": 0.6432748538011696,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004966651046043186,
      "loss": 6.2175,
      "mean_token_accuracy": 0.1467558190226555,
      "num_tokens": 12610074.0,
      "step": 6435
    },
    {
      "entropy": 6.215302324295044,
      "epoch": 0.6437746788623981,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004966589799301016,
      "loss": 6.0874,
      "mean_token_accuracy": 0.1569178357720375,
      "num_tokens": 12619117.0,
      "step": 6440
    },
    {
      "entropy": 6.310656785964966,
      "epoch": 0.6442745039236267,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004966528496790015,
      "loss": 6.2281,
      "mean_token_accuracy": 0.13886816501617433,
      "num_tokens": 12627993.0,
      "step": 6445
    },
    {
      "entropy": 6.337427949905395,
      "epoch": 0.6447743289848553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966467138511729,
      "loss": 6.2144,
      "mean_token_accuracy": 0.14373488277196883,
      "num_tokens": 12637787.0,
      "step": 6450
    },
    {
      "entropy": 6.35661039352417,
      "epoch": 0.6452741540460839,
      "grad_norm": 1.59375,
      "learning_rate": 0.00049664057244677,
      "loss": 6.3397,
      "mean_token_accuracy": 0.1412152901291847,
      "num_tokens": 12648409.0,
      "step": 6455
    },
    {
      "entropy": 6.280369901657105,
      "epoch": 0.6457739791073125,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004966344254659473,
      "loss": 6.1398,
      "mean_token_accuracy": 0.1532076209783554,
      "num_tokens": 12658391.0,
      "step": 6460
    },
    {
      "entropy": 6.260202646255493,
      "epoch": 0.646273804168541,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004966282729088595,
      "loss": 6.2045,
      "mean_token_accuracy": 0.14276923686265947,
      "num_tokens": 12669926.0,
      "step": 6465
    },
    {
      "entropy": 6.261676836013794,
      "epoch": 0.6467736292297696,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004966221147756615,
      "loss": 6.2661,
      "mean_token_accuracy": 0.14284807145595552,
      "num_tokens": 12679176.0,
      "step": 6470
    },
    {
      "entropy": 6.265682744979858,
      "epoch": 0.6472734542909981,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004966159510665079,
      "loss": 6.1554,
      "mean_token_accuracy": 0.146842310577631,
      "num_tokens": 12688954.0,
      "step": 6475
    },
    {
      "entropy": 6.244898319244385,
      "epoch": 0.6477732793522267,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004966097817815542,
      "loss": 6.1533,
      "mean_token_accuracy": 0.14491694048047066,
      "num_tokens": 12697577.0,
      "step": 6480
    },
    {
      "entropy": 6.281431055068969,
      "epoch": 0.6482731044134553,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004966036069209552,
      "loss": 6.2834,
      "mean_token_accuracy": 0.13943449333310126,
      "num_tokens": 12708153.0,
      "step": 6485
    },
    {
      "entropy": 6.342728471755981,
      "epoch": 0.6487729294746839,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004965974264848666,
      "loss": 6.2392,
      "mean_token_accuracy": 0.14364442005753517,
      "num_tokens": 12717329.0,
      "step": 6490
    },
    {
      "entropy": 6.330558776855469,
      "epoch": 0.6492727545359124,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004965912404734438,
      "loss": 6.2348,
      "mean_token_accuracy": 0.14568729028105737,
      "num_tokens": 12726528.0,
      "step": 6495
    },
    {
      "entropy": 6.333059644699096,
      "epoch": 0.649772579597141,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004965850488868424,
      "loss": 6.2249,
      "mean_token_accuracy": 0.1396489255130291,
      "num_tokens": 12737341.0,
      "step": 6500
    },
    {
      "entropy": 6.32062258720398,
      "epoch": 0.6502724046583696,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004965788517252181,
      "loss": 6.1424,
      "mean_token_accuracy": 0.14806357473134996,
      "num_tokens": 12746973.0,
      "step": 6505
    },
    {
      "entropy": 6.343089914321899,
      "epoch": 0.6507722297195981,
      "grad_norm": 1.171875,
      "learning_rate": 0.000496572648988727,
      "loss": 6.1739,
      "mean_token_accuracy": 0.14911404773592948,
      "num_tokens": 12756176.0,
      "step": 6510
    },
    {
      "entropy": 6.324216270446778,
      "epoch": 0.6512720547808267,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000496566440677525,
      "loss": 6.1912,
      "mean_token_accuracy": 0.13954056277871132,
      "num_tokens": 12766349.0,
      "step": 6515
    },
    {
      "entropy": 6.212085580825805,
      "epoch": 0.6517718798420553,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004965602267917685,
      "loss": 6.1328,
      "mean_token_accuracy": 0.14660271480679513,
      "num_tokens": 12775857.0,
      "step": 6520
    },
    {
      "entropy": 6.282214832305908,
      "epoch": 0.6522717049032839,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004965540073316136,
      "loss": 6.2013,
      "mean_token_accuracy": 0.14208867996931077,
      "num_tokens": 12786376.0,
      "step": 6525
    },
    {
      "entropy": 6.286488151550293,
      "epoch": 0.6527715299645124,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004965477822972169,
      "loss": 6.2221,
      "mean_token_accuracy": 0.14130338579416274,
      "num_tokens": 12795719.0,
      "step": 6530
    },
    {
      "entropy": 6.321103763580322,
      "epoch": 0.653271355025741,
      "grad_norm": 1.125,
      "learning_rate": 0.0004965415516887349,
      "loss": 6.2102,
      "mean_token_accuracy": 0.1439807489514351,
      "num_tokens": 12806742.0,
      "step": 6535
    },
    {
      "entropy": 6.25177412033081,
      "epoch": 0.6537711800869695,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004965353155063246,
      "loss": 6.0818,
      "mean_token_accuracy": 0.14133842587471007,
      "num_tokens": 12817362.0,
      "step": 6540
    },
    {
      "entropy": 6.159471035003662,
      "epoch": 0.6542710051481981,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004965290737501427,
      "loss": 6.0441,
      "mean_token_accuracy": 0.15827706903219224,
      "num_tokens": 12826516.0,
      "step": 6545
    },
    {
      "entropy": 6.264061784744262,
      "epoch": 0.6547708302094267,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004965228264203464,
      "loss": 6.2657,
      "mean_token_accuracy": 0.13436038196086883,
      "num_tokens": 12836321.0,
      "step": 6550
    },
    {
      "entropy": 6.297969245910645,
      "epoch": 0.6552706552706553,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004965165735170928,
      "loss": 6.1134,
      "mean_token_accuracy": 0.14968342557549477,
      "num_tokens": 12845955.0,
      "step": 6555
    },
    {
      "entropy": 6.309147787094116,
      "epoch": 0.6557704803318838,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000496510315040539,
      "loss": 6.276,
      "mean_token_accuracy": 0.14372547939419747,
      "num_tokens": 12854749.0,
      "step": 6560
    },
    {
      "entropy": 6.326816082000732,
      "epoch": 0.6562703053931124,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004965040509908429,
      "loss": 6.2409,
      "mean_token_accuracy": 0.139018664509058,
      "num_tokens": 12865145.0,
      "step": 6565
    },
    {
      "entropy": 6.290833425521851,
      "epoch": 0.656770130454341,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004964977813681618,
      "loss": 6.1639,
      "mean_token_accuracy": 0.1501176491379738,
      "num_tokens": 12876109.0,
      "step": 6570
    },
    {
      "entropy": 6.270522165298462,
      "epoch": 0.6572699555155695,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004964915061726534,
      "loss": 6.2403,
      "mean_token_accuracy": 0.13995340913534166,
      "num_tokens": 12886813.0,
      "step": 6575
    },
    {
      "entropy": 6.288305282592773,
      "epoch": 0.6577697805767981,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964852254044759,
      "loss": 6.2187,
      "mean_token_accuracy": 0.14275041222572327,
      "num_tokens": 12897991.0,
      "step": 6580
    },
    {
      "entropy": 6.290961265563965,
      "epoch": 0.6582696056380267,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004964789390637871,
      "loss": 6.2867,
      "mean_token_accuracy": 0.1336333706974983,
      "num_tokens": 12907354.0,
      "step": 6585
    },
    {
      "entropy": 6.327837562561035,
      "epoch": 0.6587694306992553,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004964726471507451,
      "loss": 6.1687,
      "mean_token_accuracy": 0.15185509622097015,
      "num_tokens": 12917187.0,
      "step": 6590
    },
    {
      "entropy": 6.211197280883789,
      "epoch": 0.6592692557604838,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004964663496655083,
      "loss": 6.134,
      "mean_token_accuracy": 0.15331015586853028,
      "num_tokens": 12926470.0,
      "step": 6595
    },
    {
      "entropy": 6.251475429534912,
      "epoch": 0.6597690808217124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004964600466082351,
      "loss": 6.1932,
      "mean_token_accuracy": 0.1524736166000366,
      "num_tokens": 12936609.0,
      "step": 6600
    },
    {
      "entropy": 6.299889087677002,
      "epoch": 0.660268905882941,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004964537379790842,
      "loss": 6.1108,
      "mean_token_accuracy": 0.14824423491954802,
      "num_tokens": 12946286.0,
      "step": 6605
    },
    {
      "entropy": 6.277209854125976,
      "epoch": 0.6607687309441695,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004964474237782141,
      "loss": 6.1376,
      "mean_token_accuracy": 0.1471827894449234,
      "num_tokens": 12955881.0,
      "step": 6610
    },
    {
      "entropy": 6.2671130180358885,
      "epoch": 0.6612685560053981,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004964411040057839,
      "loss": 6.2871,
      "mean_token_accuracy": 0.14310781806707382,
      "num_tokens": 12966273.0,
      "step": 6615
    },
    {
      "entropy": 6.282306671142578,
      "epoch": 0.6617683810666267,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004964347786619524,
      "loss": 6.2574,
      "mean_token_accuracy": 0.14010658338665963,
      "num_tokens": 12975300.0,
      "step": 6620
    },
    {
      "entropy": 6.350692987442017,
      "epoch": 0.6622682061278552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004964284477468791,
      "loss": 6.1849,
      "mean_token_accuracy": 0.1533190742135048,
      "num_tokens": 12985291.0,
      "step": 6625
    },
    {
      "entropy": 6.257151174545288,
      "epoch": 0.6627680311890838,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004964221112607228,
      "loss": 6.1481,
      "mean_token_accuracy": 0.1530764639377594,
      "num_tokens": 12995935.0,
      "step": 6630
    },
    {
      "entropy": 6.296711111068726,
      "epoch": 0.6632678562503124,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004964157692036432,
      "loss": 6.1264,
      "mean_token_accuracy": 0.15155061930418015,
      "num_tokens": 13006916.0,
      "step": 6635
    },
    {
      "entropy": 6.383872938156128,
      "epoch": 0.663767681311541,
      "grad_norm": 1.125,
      "learning_rate": 0.0004964094215757998,
      "loss": 6.4002,
      "mean_token_accuracy": 0.13124821931123734,
      "num_tokens": 13017247.0,
      "step": 6640
    },
    {
      "entropy": 6.354250049591064,
      "epoch": 0.6642675063727695,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004964030683773525,
      "loss": 6.2234,
      "mean_token_accuracy": 0.1461649015545845,
      "num_tokens": 13026038.0,
      "step": 6645
    },
    {
      "entropy": 6.307520914077759,
      "epoch": 0.6647673314339981,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004963967096084608,
      "loss": 6.2715,
      "mean_token_accuracy": 0.13808465301990508,
      "num_tokens": 13036436.0,
      "step": 6650
    },
    {
      "entropy": 6.330572319030762,
      "epoch": 0.6652671564952267,
      "grad_norm": 1.25,
      "learning_rate": 0.0004963903452692849,
      "loss": 6.233,
      "mean_token_accuracy": 0.1444825440645218,
      "num_tokens": 13045594.0,
      "step": 6655
    },
    {
      "entropy": 6.359673881530762,
      "epoch": 0.6657669815564552,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004963839753599848,
      "loss": 6.2215,
      "mean_token_accuracy": 0.143858839571476,
      "num_tokens": 13056023.0,
      "step": 6660
    },
    {
      "entropy": 6.317483806610108,
      "epoch": 0.6662668066176838,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000496377599880721,
      "loss": 6.174,
      "mean_token_accuracy": 0.1491334892809391,
      "num_tokens": 13064230.0,
      "step": 6665
    },
    {
      "entropy": 6.238755607604981,
      "epoch": 0.6667666316789124,
      "grad_norm": 2.0,
      "learning_rate": 0.0004963712188316537,
      "loss": 6.0889,
      "mean_token_accuracy": 0.15473163425922393,
      "num_tokens": 13073767.0,
      "step": 6670
    },
    {
      "entropy": 6.235145568847656,
      "epoch": 0.667266456740141,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004963648322129435,
      "loss": 6.1085,
      "mean_token_accuracy": 0.14921026378870011,
      "num_tokens": 13083026.0,
      "step": 6675
    },
    {
      "entropy": 6.306944036483765,
      "epoch": 0.6677662818013695,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000496358440024751,
      "loss": 6.1501,
      "mean_token_accuracy": 0.149919793009758,
      "num_tokens": 13092029.0,
      "step": 6680
    },
    {
      "entropy": 6.313770484924317,
      "epoch": 0.6682661068625981,
      "grad_norm": 1.25,
      "learning_rate": 0.0004963520422672372,
      "loss": 6.2907,
      "mean_token_accuracy": 0.1392592266201973,
      "num_tokens": 13101859.0,
      "step": 6685
    },
    {
      "entropy": 6.310837841033935,
      "epoch": 0.6687659319238266,
      "grad_norm": 1.15625,
      "learning_rate": 0.000496345638940563,
      "loss": 6.1344,
      "mean_token_accuracy": 0.15150134414434432,
      "num_tokens": 13111007.0,
      "step": 6690
    },
    {
      "entropy": 6.215577077865601,
      "epoch": 0.6692657569850552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004963392300448894,
      "loss": 6.1836,
      "mean_token_accuracy": 0.14991192072629927,
      "num_tokens": 13120662.0,
      "step": 6695
    },
    {
      "entropy": 6.309817981719971,
      "epoch": 0.6697655820462838,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004963328155803778,
      "loss": 6.0865,
      "mean_token_accuracy": 0.14408904016017915,
      "num_tokens": 13130263.0,
      "step": 6700
    },
    {
      "entropy": 6.341976642608643,
      "epoch": 0.6702654071075124,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004963263955471895,
      "loss": 6.1448,
      "mean_token_accuracy": 0.14620930105447769,
      "num_tokens": 13139821.0,
      "step": 6705
    },
    {
      "entropy": 6.231247997283935,
      "epoch": 0.670765232168741,
      "grad_norm": 1.1875,
      "learning_rate": 0.000496319969945486,
      "loss": 6.1395,
      "mean_token_accuracy": 0.14915765598416328,
      "num_tokens": 13149731.0,
      "step": 6710
    },
    {
      "entropy": 6.282005691528321,
      "epoch": 0.6712650572299695,
      "grad_norm": 1.296875,
      "learning_rate": 0.000496313538775429,
      "loss": 6.2245,
      "mean_token_accuracy": 0.14730927273631095,
      "num_tokens": 13160689.0,
      "step": 6715
    },
    {
      "entropy": 6.274010753631591,
      "epoch": 0.6717648822911981,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004963071020371803,
      "loss": 6.3393,
      "mean_token_accuracy": 0.14334413260221482,
      "num_tokens": 13170694.0,
      "step": 6720
    },
    {
      "entropy": 6.403909540176391,
      "epoch": 0.6722647073524266,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004963006597309019,
      "loss": 6.3,
      "mean_token_accuracy": 0.1423768937587738,
      "num_tokens": 13180514.0,
      "step": 6725
    },
    {
      "entropy": 6.331519603729248,
      "epoch": 0.6727645324136552,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004962942118567559,
      "loss": 6.1034,
      "mean_token_accuracy": 0.14691800028085708,
      "num_tokens": 13190144.0,
      "step": 6730
    },
    {
      "entropy": 6.304260635375977,
      "epoch": 0.6732643574748838,
      "grad_norm": 1.25,
      "learning_rate": 0.0004962877584149043,
      "loss": 6.1565,
      "mean_token_accuracy": 0.1423011153936386,
      "num_tokens": 13199640.0,
      "step": 6735
    },
    {
      "entropy": 6.255619192123413,
      "epoch": 0.6737641825361124,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962812994055098,
      "loss": 6.1178,
      "mean_token_accuracy": 0.15270792841911315,
      "num_tokens": 13208824.0,
      "step": 6740
    },
    {
      "entropy": 6.198691463470459,
      "epoch": 0.674264007597341,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004962748348287347,
      "loss": 6.0688,
      "mean_token_accuracy": 0.1588856428861618,
      "num_tokens": 13219305.0,
      "step": 6745
    },
    {
      "entropy": 6.245419836044311,
      "epoch": 0.6747638326585695,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004962683646847416,
      "loss": 6.0901,
      "mean_token_accuracy": 0.15135168731212617,
      "num_tokens": 13227727.0,
      "step": 6750
    },
    {
      "entropy": 6.312480020523071,
      "epoch": 0.675263657719798,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004962618889736936,
      "loss": 6.151,
      "mean_token_accuracy": 0.1480355367064476,
      "num_tokens": 13237522.0,
      "step": 6755
    },
    {
      "entropy": 6.25027904510498,
      "epoch": 0.6757634827810266,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004962554076957533,
      "loss": 6.2101,
      "mean_token_accuracy": 0.1414764016866684,
      "num_tokens": 13247155.0,
      "step": 6760
    },
    {
      "entropy": 6.224506425857544,
      "epoch": 0.6762633078422552,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004962489208510838,
      "loss": 6.2534,
      "mean_token_accuracy": 0.14092431366443633,
      "num_tokens": 13257324.0,
      "step": 6765
    },
    {
      "entropy": 6.232892608642578,
      "epoch": 0.6767631329034838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004962424284398484,
      "loss": 6.2316,
      "mean_token_accuracy": 0.14328891038894653,
      "num_tokens": 13267025.0,
      "step": 6770
    },
    {
      "entropy": 6.354032373428344,
      "epoch": 0.6772629579647124,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004962359304622105,
      "loss": 6.1337,
      "mean_token_accuracy": 0.15617139488458634,
      "num_tokens": 13276709.0,
      "step": 6775
    },
    {
      "entropy": 6.288168001174927,
      "epoch": 0.677762783025941,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004962294269183335,
      "loss": 6.1372,
      "mean_token_accuracy": 0.1426538646221161,
      "num_tokens": 13286029.0,
      "step": 6780
    },
    {
      "entropy": 6.31599702835083,
      "epoch": 0.6782626080871695,
      "grad_norm": 1.296875,
      "learning_rate": 0.000496222917808381,
      "loss": 6.3013,
      "mean_token_accuracy": 0.13417728766798973,
      "num_tokens": 13296367.0,
      "step": 6785
    },
    {
      "entropy": 6.370834159851074,
      "epoch": 0.678762433148398,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004962164031325169,
      "loss": 6.2259,
      "mean_token_accuracy": 0.14463320076465608,
      "num_tokens": 13305186.0,
      "step": 6790
    },
    {
      "entropy": 6.298191785812378,
      "epoch": 0.6792622582096266,
      "grad_norm": 1.21875,
      "learning_rate": 0.000496209882890905,
      "loss": 6.1602,
      "mean_token_accuracy": 0.15014540180563926,
      "num_tokens": 13314512.0,
      "step": 6795
    },
    {
      "entropy": 6.325139904022217,
      "epoch": 0.6797620832708552,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004962033570837094,
      "loss": 6.2709,
      "mean_token_accuracy": 0.14526838138699533,
      "num_tokens": 13324423.0,
      "step": 6800
    },
    {
      "entropy": 6.303468942642212,
      "epoch": 0.6802619083320838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004961968257110941,
      "loss": 6.2591,
      "mean_token_accuracy": 0.1405862882733345,
      "num_tokens": 13334133.0,
      "step": 6805
    },
    {
      "entropy": 6.388118839263916,
      "epoch": 0.6807617333933124,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004961902887732236,
      "loss": 6.2583,
      "mean_token_accuracy": 0.1440707750618458,
      "num_tokens": 13344017.0,
      "step": 6810
    },
    {
      "entropy": 6.233449983596802,
      "epoch": 0.681261558454541,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004961837462702626,
      "loss": 6.0123,
      "mean_token_accuracy": 0.15574013590812683,
      "num_tokens": 13353922.0,
      "step": 6815
    },
    {
      "entropy": 6.261878919601441,
      "epoch": 0.6817613835157694,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004961771982023752,
      "loss": 6.1858,
      "mean_token_accuracy": 0.14553807079792022,
      "num_tokens": 13363890.0,
      "step": 6820
    },
    {
      "entropy": 6.2675285816192625,
      "epoch": 0.682261208576998,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004961706445697265,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14782624989748,
      "num_tokens": 13373726.0,
      "step": 6825
    },
    {
      "entropy": 6.198501825332642,
      "epoch": 0.6827610336382266,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004961640853724813,
      "loss": 6.0373,
      "mean_token_accuracy": 0.15897899568080903,
      "num_tokens": 13382049.0,
      "step": 6830
    },
    {
      "entropy": 6.107681846618652,
      "epoch": 0.6832608586994552,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004961575206108046,
      "loss": 5.9809,
      "mean_token_accuracy": 0.15685263127088547,
      "num_tokens": 13390977.0,
      "step": 6835
    },
    {
      "entropy": 6.29368314743042,
      "epoch": 0.6837606837606838,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004961509502848616,
      "loss": 6.192,
      "mean_token_accuracy": 0.14624952003359795,
      "num_tokens": 13400165.0,
      "step": 6840
    },
    {
      "entropy": 6.319872760772705,
      "epoch": 0.6842605088219124,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004961443743948177,
      "loss": 6.2193,
      "mean_token_accuracy": 0.13824222683906556,
      "num_tokens": 13410201.0,
      "step": 6845
    },
    {
      "entropy": 6.268837928771973,
      "epoch": 0.684760333883141,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000496137792940838,
      "loss": 6.1959,
      "mean_token_accuracy": 0.149955914914608,
      "num_tokens": 13421074.0,
      "step": 6850
    },
    {
      "entropy": 6.197783422470093,
      "epoch": 0.6852601589443694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004961312059230885,
      "loss": 6.0266,
      "mean_token_accuracy": 0.15611146837472917,
      "num_tokens": 13430387.0,
      "step": 6855
    },
    {
      "entropy": 6.211978149414063,
      "epoch": 0.685759984005598,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004961246133417348,
      "loss": 6.0954,
      "mean_token_accuracy": 0.15050722509622574,
      "num_tokens": 13439497.0,
      "step": 6860
    },
    {
      "entropy": 6.343523168563843,
      "epoch": 0.6862598090668266,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004961180151969427,
      "loss": 6.2344,
      "mean_token_accuracy": 0.14303319454193114,
      "num_tokens": 13450543.0,
      "step": 6865
    },
    {
      "entropy": 6.282239866256714,
      "epoch": 0.6867596341280552,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004961114114888782,
      "loss": 6.1662,
      "mean_token_accuracy": 0.15495624169707298,
      "num_tokens": 13459996.0,
      "step": 6870
    },
    {
      "entropy": 6.216299295425415,
      "epoch": 0.6872594591892838,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004961048022177074,
      "loss": 6.1124,
      "mean_token_accuracy": 0.15178535506129265,
      "num_tokens": 13469141.0,
      "step": 6875
    },
    {
      "entropy": 6.249474763870239,
      "epoch": 0.6877592842505124,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004960981873835968,
      "loss": 6.0934,
      "mean_token_accuracy": 0.15554443895816802,
      "num_tokens": 13478349.0,
      "step": 6880
    },
    {
      "entropy": 6.253992319107056,
      "epoch": 0.6882591093117408,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960915669867126,
      "loss": 6.0597,
      "mean_token_accuracy": 0.14935113713145257,
      "num_tokens": 13487417.0,
      "step": 6885
    },
    {
      "entropy": 6.343624305725098,
      "epoch": 0.6887589343729694,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004960849410272216,
      "loss": 6.2971,
      "mean_token_accuracy": 0.13986774384975434,
      "num_tokens": 13497607.0,
      "step": 6890
    },
    {
      "entropy": 6.311396169662475,
      "epoch": 0.689258759434198,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004960783095052903,
      "loss": 6.2558,
      "mean_token_accuracy": 0.13967615738511086,
      "num_tokens": 13508222.0,
      "step": 6895
    },
    {
      "entropy": 6.194321393966675,
      "epoch": 0.6897585844954266,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004960716724210856,
      "loss": 6.0695,
      "mean_token_accuracy": 0.15555219054222108,
      "num_tokens": 13519593.0,
      "step": 6900
    },
    {
      "entropy": 6.381457090377808,
      "epoch": 0.6902584095566552,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004960650297747746,
      "loss": 6.2735,
      "mean_token_accuracy": 0.13538890033960344,
      "num_tokens": 13531089.0,
      "step": 6905
    },
    {
      "entropy": 6.325894927978515,
      "epoch": 0.6907582346178838,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004960583815665242,
      "loss": 6.1589,
      "mean_token_accuracy": 0.147079511731863,
      "num_tokens": 13539897.0,
      "step": 6910
    },
    {
      "entropy": 6.304473161697388,
      "epoch": 0.6912580596791124,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004960517277965018,
      "loss": 6.2417,
      "mean_token_accuracy": 0.13882038742303848,
      "num_tokens": 13549986.0,
      "step": 6915
    },
    {
      "entropy": 6.32588210105896,
      "epoch": 0.6917578847403408,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004960450684648749,
      "loss": 6.2492,
      "mean_token_accuracy": 0.14710931479930878,
      "num_tokens": 13559866.0,
      "step": 6920
    },
    {
      "entropy": 6.389265441894532,
      "epoch": 0.6922577098015694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000496038403571811,
      "loss": 6.2572,
      "mean_token_accuracy": 0.1386592723429203,
      "num_tokens": 13569789.0,
      "step": 6925
    },
    {
      "entropy": 6.312770366668701,
      "epoch": 0.692757534862798,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0004960317331174777,
      "loss": 6.2462,
      "mean_token_accuracy": 0.14467992633581161,
      "num_tokens": 13579121.0,
      "step": 6930
    },
    {
      "entropy": 6.216776704788208,
      "epoch": 0.6932573599240266,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004960250571020428,
      "loss": 6.1388,
      "mean_token_accuracy": 0.15116308480501175,
      "num_tokens": 13589281.0,
      "step": 6935
    },
    {
      "entropy": 6.392689514160156,
      "epoch": 0.6937571849852552,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004960183755256744,
      "loss": 6.1444,
      "mean_token_accuracy": 0.15153247416019439,
      "num_tokens": 13598312.0,
      "step": 6940
    },
    {
      "entropy": 6.234421300888061,
      "epoch": 0.6942570100464838,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004960116883885406,
      "loss": 6.1721,
      "mean_token_accuracy": 0.1503891319036484,
      "num_tokens": 13608164.0,
      "step": 6945
    },
    {
      "entropy": 6.222692155838013,
      "epoch": 0.6947568351077124,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004960049956908095,
      "loss": 6.2069,
      "mean_token_accuracy": 0.1400441512465477,
      "num_tokens": 13618314.0,
      "step": 6950
    },
    {
      "entropy": 6.340925025939941,
      "epoch": 0.6952566601689408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004959982974326496,
      "loss": 6.1847,
      "mean_token_accuracy": 0.14271825179457664,
      "num_tokens": 13628846.0,
      "step": 6955
    },
    {
      "entropy": 6.262188482284546,
      "epoch": 0.6957564852301694,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959915936142294,
      "loss": 6.0836,
      "mean_token_accuracy": 0.15636171847581865,
      "num_tokens": 13637941.0,
      "step": 6960
    },
    {
      "entropy": 6.301395273208618,
      "epoch": 0.696256310291398,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004959848842357175,
      "loss": 6.157,
      "mean_token_accuracy": 0.14335214495658874,
      "num_tokens": 13647734.0,
      "step": 6965
    },
    {
      "entropy": 6.254585075378418,
      "epoch": 0.6967561353526266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004959781692972829,
      "loss": 6.1384,
      "mean_token_accuracy": 0.14826470315456391,
      "num_tokens": 13657929.0,
      "step": 6970
    },
    {
      "entropy": 6.263657426834106,
      "epoch": 0.6972559604138552,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004959714487990943,
      "loss": 6.2294,
      "mean_token_accuracy": 0.13924643248319626,
      "num_tokens": 13668258.0,
      "step": 6975
    },
    {
      "entropy": 6.293397760391235,
      "epoch": 0.6977557854750838,
      "grad_norm": 1.203125,
      "learning_rate": 0.000495964722741321,
      "loss": 6.1677,
      "mean_token_accuracy": 0.1528436839580536,
      "num_tokens": 13677515.0,
      "step": 6980
    },
    {
      "entropy": 6.257136583328247,
      "epoch": 0.6982556105363122,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959579911241322,
      "loss": 6.1572,
      "mean_token_accuracy": 0.1477597177028656,
      "num_tokens": 13686865.0,
      "step": 6985
    },
    {
      "entropy": 6.228747272491455,
      "epoch": 0.6987554355975408,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004959512539476971,
      "loss": 6.2665,
      "mean_token_accuracy": 0.14205617532134057,
      "num_tokens": 13696683.0,
      "step": 6990
    },
    {
      "entropy": 6.3137575626373295,
      "epoch": 0.6992552606587694,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004959445112121853,
      "loss": 6.2373,
      "mean_token_accuracy": 0.1395167514681816,
      "num_tokens": 13706452.0,
      "step": 6995
    },
    {
      "entropy": 6.3202889919281,
      "epoch": 0.699755085719998,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959377629177665,
      "loss": 6.1099,
      "mean_token_accuracy": 0.1506766676902771,
      "num_tokens": 13717474.0,
      "step": 7000
    },
    {
      "entropy": 6.222063255310059,
      "epoch": 0.7002549107812266,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959310090646104,
      "loss": 6.1191,
      "mean_token_accuracy": 0.15438238829374312,
      "num_tokens": 13726723.0,
      "step": 7005
    },
    {
      "entropy": 6.322344446182251,
      "epoch": 0.7007547358424552,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004959242496528869,
      "loss": 6.349,
      "mean_token_accuracy": 0.13866892084479332,
      "num_tokens": 13736432.0,
      "step": 7010
    },
    {
      "entropy": 6.360767126083374,
      "epoch": 0.7012545609036838,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004959174846827663,
      "loss": 6.1608,
      "mean_token_accuracy": 0.14594118297100067,
      "num_tokens": 13745767.0,
      "step": 7015
    },
    {
      "entropy": 6.349599075317383,
      "epoch": 0.7017543859649122,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004959107141544186,
      "loss": 6.1917,
      "mean_token_accuracy": 0.1485964499413967,
      "num_tokens": 13756035.0,
      "step": 7020
    },
    {
      "entropy": 6.1866546154022215,
      "epoch": 0.7022542110261408,
      "grad_norm": 1.203125,
      "learning_rate": 0.000495903938068014,
      "loss": 6.0148,
      "mean_token_accuracy": 0.15764825716614722,
      "num_tokens": 13765715.0,
      "step": 7025
    },
    {
      "entropy": 6.230158805847168,
      "epoch": 0.7027540360873694,
      "grad_norm": 1.125,
      "learning_rate": 0.0004958971564237233,
      "loss": 6.1837,
      "mean_token_accuracy": 0.1422155648469925,
      "num_tokens": 13775987.0,
      "step": 7030
    },
    {
      "entropy": 6.292929363250733,
      "epoch": 0.703253861148598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000495890369221717,
      "loss": 6.1037,
      "mean_token_accuracy": 0.15195335894823075,
      "num_tokens": 13785095.0,
      "step": 7035
    },
    {
      "entropy": 6.34586763381958,
      "epoch": 0.7037536862098266,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004958835764621657,
      "loss": 6.1719,
      "mean_token_accuracy": 0.14128438681364058,
      "num_tokens": 13794777.0,
      "step": 7040
    },
    {
      "entropy": 6.222014904022217,
      "epoch": 0.7042535112710552,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004958767781452406,
      "loss": 6.0905,
      "mean_token_accuracy": 0.1442723661661148,
      "num_tokens": 13803534.0,
      "step": 7045
    },
    {
      "entropy": 6.25515923500061,
      "epoch": 0.7047533363322837,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004958699742711126,
      "loss": 6.163,
      "mean_token_accuracy": 0.1452835828065872,
      "num_tokens": 13813941.0,
      "step": 7050
    },
    {
      "entropy": 6.320334768295288,
      "epoch": 0.7052531613935122,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004958631648399528,
      "loss": 6.1479,
      "mean_token_accuracy": 0.15006570667028427,
      "num_tokens": 13823611.0,
      "step": 7055
    },
    {
      "entropy": 6.330420589447021,
      "epoch": 0.7057529864547408,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004958563498519327,
      "loss": 6.286,
      "mean_token_accuracy": 0.13946598321199416,
      "num_tokens": 13833047.0,
      "step": 7060
    },
    {
      "entropy": 6.216289043426514,
      "epoch": 0.7062528115159694,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004958495293072235,
      "loss": 6.1961,
      "mean_token_accuracy": 0.15018607527017594,
      "num_tokens": 13843524.0,
      "step": 7065
    },
    {
      "entropy": 6.252985239028931,
      "epoch": 0.706752636577198,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004958427032059971,
      "loss": 6.0661,
      "mean_token_accuracy": 0.15229864716529845,
      "num_tokens": 13852188.0,
      "step": 7070
    },
    {
      "entropy": 6.229141330718994,
      "epoch": 0.7072524616384266,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004958358715484251,
      "loss": 6.2086,
      "mean_token_accuracy": 0.14524195045232774,
      "num_tokens": 13863540.0,
      "step": 7075
    },
    {
      "entropy": 6.332069635391235,
      "epoch": 0.7077522866996552,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004958290343346795,
      "loss": 6.1805,
      "mean_token_accuracy": 0.1465955287218094,
      "num_tokens": 13871828.0,
      "step": 7080
    },
    {
      "entropy": 6.300998258590698,
      "epoch": 0.7082521117608837,
      "grad_norm": 1.171875,
      "learning_rate": 0.000495822191564932,
      "loss": 6.2216,
      "mean_token_accuracy": 0.142748311907053,
      "num_tokens": 13882222.0,
      "step": 7085
    },
    {
      "entropy": 6.2999204158782955,
      "epoch": 0.7087519368221122,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004958153432393553,
      "loss": 6.2623,
      "mean_token_accuracy": 0.14057695791125296,
      "num_tokens": 13891689.0,
      "step": 7090
    },
    {
      "entropy": 6.293786144256591,
      "epoch": 0.7092517618833408,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000495808489358121,
      "loss": 6.1353,
      "mean_token_accuracy": 0.14565599709749222,
      "num_tokens": 13901310.0,
      "step": 7095
    },
    {
      "entropy": 6.284841108322143,
      "epoch": 0.7097515869445694,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004958016299214022,
      "loss": 6.1293,
      "mean_token_accuracy": 0.14428649023175238,
      "num_tokens": 13910350.0,
      "step": 7100
    },
    {
      "entropy": 6.253100252151489,
      "epoch": 0.710251412005798,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004957947649293712,
      "loss": 6.1809,
      "mean_token_accuracy": 0.14693159386515617,
      "num_tokens": 13920481.0,
      "step": 7105
    },
    {
      "entropy": 6.337112760543823,
      "epoch": 0.7107512370670266,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957878943822006,
      "loss": 6.1685,
      "mean_token_accuracy": 0.13915617093443872,
      "num_tokens": 13931232.0,
      "step": 7110
    },
    {
      "entropy": 6.333304405212402,
      "epoch": 0.7112510621282551,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004957810182800634,
      "loss": 6.2542,
      "mean_token_accuracy": 0.13936160653829574,
      "num_tokens": 13940717.0,
      "step": 7115
    },
    {
      "entropy": 6.286463499069214,
      "epoch": 0.7117508871894836,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004957741366231327,
      "loss": 6.0956,
      "mean_token_accuracy": 0.1488887868821621,
      "num_tokens": 13949918.0,
      "step": 7120
    },
    {
      "entropy": 6.278175210952758,
      "epoch": 0.7122507122507122,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004957672494115815,
      "loss": 6.1496,
      "mean_token_accuracy": 0.15120262354612352,
      "num_tokens": 13959914.0,
      "step": 7125
    },
    {
      "entropy": 6.259276962280273,
      "epoch": 0.7127505373119408,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495760356645583,
      "loss": 6.1534,
      "mean_token_accuracy": 0.1492697849869728,
      "num_tokens": 13970054.0,
      "step": 7130
    },
    {
      "entropy": 6.168964004516601,
      "epoch": 0.7132503623731694,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004957534583253108,
      "loss": 6.0668,
      "mean_token_accuracy": 0.15821440666913986,
      "num_tokens": 13980339.0,
      "step": 7135
    },
    {
      "entropy": 6.230792713165283,
      "epoch": 0.713750187434398,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004957465544509384,
      "loss": 6.142,
      "mean_token_accuracy": 0.14590761363506316,
      "num_tokens": 13990466.0,
      "step": 7140
    },
    {
      "entropy": 6.340567970275879,
      "epoch": 0.7142500124956266,
      "grad_norm": 1.25,
      "learning_rate": 0.0004957396450226395,
      "loss": 6.177,
      "mean_token_accuracy": 0.14110314920544625,
      "num_tokens": 13999822.0,
      "step": 7145
    },
    {
      "entropy": 6.4066338539123535,
      "epoch": 0.7147498375568551,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957327300405879,
      "loss": 6.2117,
      "mean_token_accuracy": 0.14593433663249017,
      "num_tokens": 14008939.0,
      "step": 7150
    },
    {
      "entropy": 6.255950021743774,
      "epoch": 0.7152496626180836,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004957258095049577,
      "loss": 6.1602,
      "mean_token_accuracy": 0.13879207521677017,
      "num_tokens": 14019005.0,
      "step": 7155
    },
    {
      "entropy": 6.244685506820678,
      "epoch": 0.7157494876793122,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004957188834159229,
      "loss": 6.06,
      "mean_token_accuracy": 0.14689342975616454,
      "num_tokens": 14029390.0,
      "step": 7160
    },
    {
      "entropy": 6.261938953399659,
      "epoch": 0.7162493127405408,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004957119517736577,
      "loss": 6.1392,
      "mean_token_accuracy": 0.14793641418218612,
      "num_tokens": 14038613.0,
      "step": 7165
    },
    {
      "entropy": 6.253458833694458,
      "epoch": 0.7167491378017694,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004957050145783366,
      "loss": 6.1326,
      "mean_token_accuracy": 0.1456688791513443,
      "num_tokens": 14048144.0,
      "step": 7170
    },
    {
      "entropy": 6.12710747718811,
      "epoch": 0.717248962862998,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004956980718301342,
      "loss": 6.0247,
      "mean_token_accuracy": 0.16036900132894516,
      "num_tokens": 14056894.0,
      "step": 7175
    },
    {
      "entropy": 6.227234363555908,
      "epoch": 0.7177487879242265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004956911235292251,
      "loss": 6.1495,
      "mean_token_accuracy": 0.14351562559604644,
      "num_tokens": 14066024.0,
      "step": 7180
    },
    {
      "entropy": 6.29120397567749,
      "epoch": 0.7182486129854551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956841696757841,
      "loss": 6.1491,
      "mean_token_accuracy": 0.14925071001052856,
      "num_tokens": 14075805.0,
      "step": 7185
    },
    {
      "entropy": 6.249656534194946,
      "epoch": 0.7187484380466836,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004956772102699862,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14800831228494643,
      "num_tokens": 14085229.0,
      "step": 7190
    },
    {
      "entropy": 6.27678918838501,
      "epoch": 0.7192482631079122,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004956702453120064,
      "loss": 6.1291,
      "mean_token_accuracy": 0.1453297108411789,
      "num_tokens": 14094594.0,
      "step": 7195
    },
    {
      "entropy": 6.23982081413269,
      "epoch": 0.7197480881691408,
      "grad_norm": 1.84375,
      "learning_rate": 0.0004956632748020201,
      "loss": 6.1682,
      "mean_token_accuracy": 0.1511926256120205,
      "num_tokens": 14104373.0,
      "step": 7200
    },
    {
      "entropy": 6.194260454177856,
      "epoch": 0.7202479132303694,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004956562987402026,
      "loss": 6.0632,
      "mean_token_accuracy": 0.15217120125889777,
      "num_tokens": 14113918.0,
      "step": 7205
    },
    {
      "entropy": 6.34488582611084,
      "epoch": 0.720747738291598,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004956493171267293,
      "loss": 6.2913,
      "mean_token_accuracy": 0.13412214070558548,
      "num_tokens": 14123711.0,
      "step": 7210
    },
    {
      "entropy": 6.304417562484741,
      "epoch": 0.7212475633528265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000495642329961776,
      "loss": 6.0711,
      "mean_token_accuracy": 0.15149474442005156,
      "num_tokens": 14132994.0,
      "step": 7215
    },
    {
      "entropy": 6.211438131332398,
      "epoch": 0.7217473884140551,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004956353372455185,
      "loss": 6.1141,
      "mean_token_accuracy": 0.14841899201273917,
      "num_tokens": 14143398.0,
      "step": 7220
    },
    {
      "entropy": 6.260597848892212,
      "epoch": 0.7222472134752836,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004956283389781327,
      "loss": 6.1277,
      "mean_token_accuracy": 0.15246413350105287,
      "num_tokens": 14152374.0,
      "step": 7225
    },
    {
      "entropy": 6.286064195632934,
      "epoch": 0.7227470385365122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956213351597946,
      "loss": 6.0912,
      "mean_token_accuracy": 0.1527165576815605,
      "num_tokens": 14162123.0,
      "step": 7230
    },
    {
      "entropy": 6.283297920227051,
      "epoch": 0.7232468635977408,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004956143257906805,
      "loss": 6.2731,
      "mean_token_accuracy": 0.14748100861907004,
      "num_tokens": 14172004.0,
      "step": 7235
    },
    {
      "entropy": 6.125205707550049,
      "epoch": 0.7237466886589694,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004956073108709667,
      "loss": 6.1491,
      "mean_token_accuracy": 0.1470201477408409,
      "num_tokens": 14181659.0,
      "step": 7240
    },
    {
      "entropy": 6.21792778968811,
      "epoch": 0.7242465137201979,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004956002904008298,
      "loss": 6.1211,
      "mean_token_accuracy": 0.1494100198149681,
      "num_tokens": 14191051.0,
      "step": 7245
    },
    {
      "entropy": 6.375655031204223,
      "epoch": 0.7247463387814265,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004955932643804463,
      "loss": 6.2577,
      "mean_token_accuracy": 0.13779568821191787,
      "num_tokens": 14201232.0,
      "step": 7250
    },
    {
      "entropy": 6.298529720306396,
      "epoch": 0.7252461638426551,
      "grad_norm": 1.234375,
      "learning_rate": 0.000495586232809993,
      "loss": 6.1787,
      "mean_token_accuracy": 0.1432025372982025,
      "num_tokens": 14210426.0,
      "step": 7255
    },
    {
      "entropy": 6.154688215255737,
      "epoch": 0.7257459889038836,
      "grad_norm": 1.25,
      "learning_rate": 0.0004955791956896469,
      "loss": 6.0449,
      "mean_token_accuracy": 0.15584971606731415,
      "num_tokens": 14220013.0,
      "step": 7260
    },
    {
      "entropy": 6.279833030700684,
      "epoch": 0.7262458139651122,
      "grad_norm": 1.296875,
      "learning_rate": 0.000495572153019585,
      "loss": 6.1748,
      "mean_token_accuracy": 0.1474984735250473,
      "num_tokens": 14229249.0,
      "step": 7265
    },
    {
      "entropy": 6.281010627746582,
      "epoch": 0.7267456390263408,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004955651047999844,
      "loss": 6.0642,
      "mean_token_accuracy": 0.15525839626789092,
      "num_tokens": 14238370.0,
      "step": 7270
    },
    {
      "entropy": 6.174116420745849,
      "epoch": 0.7272454640875694,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004955580510310226,
      "loss": 6.1325,
      "mean_token_accuracy": 0.1603264719247818,
      "num_tokens": 14248037.0,
      "step": 7275
    },
    {
      "entropy": 6.215904712677002,
      "epoch": 0.7277452891487979,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004955509917128771,
      "loss": 6.1732,
      "mean_token_accuracy": 0.14549778550863265,
      "num_tokens": 14257480.0,
      "step": 7280
    },
    {
      "entropy": 6.2183716773986815,
      "epoch": 0.7282451142100265,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004955439268457252,
      "loss": 6.1342,
      "mean_token_accuracy": 0.14461766853928565,
      "num_tokens": 14266114.0,
      "step": 7285
    },
    {
      "entropy": 6.1960225105285645,
      "epoch": 0.728744939271255,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004955368564297449,
      "loss": 6.1313,
      "mean_token_accuracy": 0.14872496724128723,
      "num_tokens": 14276048.0,
      "step": 7290
    },
    {
      "entropy": 6.290376043319702,
      "epoch": 0.7292447643324836,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004955297804651142,
      "loss": 6.0416,
      "mean_token_accuracy": 0.15893998593091965,
      "num_tokens": 14285225.0,
      "step": 7295
    },
    {
      "entropy": 6.325178861618042,
      "epoch": 0.7297445893937122,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004955226989520107,
      "loss": 6.1474,
      "mean_token_accuracy": 0.14429766312241554,
      "num_tokens": 14295566.0,
      "step": 7300
    },
    {
      "entropy": 6.225370311737061,
      "epoch": 0.7302444144549408,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000495515611890613,
      "loss": 6.1257,
      "mean_token_accuracy": 0.14591144621372223,
      "num_tokens": 14305072.0,
      "step": 7305
    },
    {
      "entropy": 6.162986850738525,
      "epoch": 0.7307442395161693,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004955085192810993,
      "loss": 5.978,
      "mean_token_accuracy": 0.15212754383683205,
      "num_tokens": 14315273.0,
      "step": 7310
    },
    {
      "entropy": 6.2624650478363035,
      "epoch": 0.7312440645773979,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004955014211236479,
      "loss": 6.1661,
      "mean_token_accuracy": 0.1413229674100876,
      "num_tokens": 14324818.0,
      "step": 7315
    },
    {
      "entropy": 6.3168800354003904,
      "epoch": 0.7317438896386265,
      "grad_norm": 1.375,
      "learning_rate": 0.0004954943174184375,
      "loss": 6.2342,
      "mean_token_accuracy": 0.14350414872169495,
      "num_tokens": 14335129.0,
      "step": 7320
    },
    {
      "entropy": 6.349929332733154,
      "epoch": 0.732243714699855,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004954872081656469,
      "loss": 6.2181,
      "mean_token_accuracy": 0.14031280279159547,
      "num_tokens": 14344339.0,
      "step": 7325
    },
    {
      "entropy": 6.29350528717041,
      "epoch": 0.7327435397610836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004954800933654548,
      "loss": 6.1485,
      "mean_token_accuracy": 0.14367504715919494,
      "num_tokens": 14353391.0,
      "step": 7330
    },
    {
      "entropy": 6.327314662933349,
      "epoch": 0.7332433648223122,
      "grad_norm": 1.25,
      "learning_rate": 0.0004954729730180403,
      "loss": 6.2593,
      "mean_token_accuracy": 0.1388326495885849,
      "num_tokens": 14363684.0,
      "step": 7335
    },
    {
      "entropy": 6.330690670013428,
      "epoch": 0.7337431898835408,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004954658471235825,
      "loss": 6.1515,
      "mean_token_accuracy": 0.15232093632221222,
      "num_tokens": 14372802.0,
      "step": 7340
    },
    {
      "entropy": 6.29860520362854,
      "epoch": 0.7342430149447693,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004954587156822607,
      "loss": 6.1826,
      "mean_token_accuracy": 0.13735579922795296,
      "num_tokens": 14382665.0,
      "step": 7345
    },
    {
      "entropy": 6.25233678817749,
      "epoch": 0.7347428400059979,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004954515786942544,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14868216961622238,
      "num_tokens": 14393631.0,
      "step": 7350
    },
    {
      "entropy": 6.258598518371582,
      "epoch": 0.7352426650672265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000495444436159743,
      "loss": 6.1413,
      "mean_token_accuracy": 0.1490681141614914,
      "num_tokens": 14402713.0,
      "step": 7355
    },
    {
      "entropy": 6.388405275344849,
      "epoch": 0.735742490128455,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004954372880789064,
      "loss": 6.3014,
      "mean_token_accuracy": 0.1369183234870434,
      "num_tokens": 14412777.0,
      "step": 7360
    },
    {
      "entropy": 6.374183988571167,
      "epoch": 0.7362423151896836,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004954301344519244,
      "loss": 6.2337,
      "mean_token_accuracy": 0.14146374613046647,
      "num_tokens": 14422839.0,
      "step": 7365
    },
    {
      "entropy": 6.174902868270874,
      "epoch": 0.7367421402509122,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004954229752789769,
      "loss": 6.1094,
      "mean_token_accuracy": 0.15102093666791916,
      "num_tokens": 14433239.0,
      "step": 7370
    },
    {
      "entropy": 6.286706495285034,
      "epoch": 0.7372419653121407,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004954158105602441,
      "loss": 6.2137,
      "mean_token_accuracy": 0.14214254766702653,
      "num_tokens": 14442907.0,
      "step": 7375
    },
    {
      "entropy": 6.252299070358276,
      "epoch": 0.7377417903733693,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004954086402959061,
      "loss": 6.05,
      "mean_token_accuracy": 0.1516972839832306,
      "num_tokens": 14452492.0,
      "step": 7380
    },
    {
      "entropy": 6.274751377105713,
      "epoch": 0.7382416154345979,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004954014644861435,
      "loss": 6.2011,
      "mean_token_accuracy": 0.14236974269151687,
      "num_tokens": 14461353.0,
      "step": 7385
    },
    {
      "entropy": 6.237046480178833,
      "epoch": 0.7387414404958265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004953942831311367,
      "loss": 6.0683,
      "mean_token_accuracy": 0.159549480676651,
      "num_tokens": 14470579.0,
      "step": 7390
    },
    {
      "entropy": 6.210099554061889,
      "epoch": 0.739241265557055,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004953870962310666,
      "loss": 6.039,
      "mean_token_accuracy": 0.1542980045080185,
      "num_tokens": 14481608.0,
      "step": 7395
    },
    {
      "entropy": 6.2453282356262205,
      "epoch": 0.7397410906182836,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953799037861138,
      "loss": 6.202,
      "mean_token_accuracy": 0.14560826495289803,
      "num_tokens": 14493074.0,
      "step": 7400
    },
    {
      "entropy": 6.25093183517456,
      "epoch": 0.7402409156795122,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004953727057964594,
      "loss": 6.0557,
      "mean_token_accuracy": 0.15362336188554765,
      "num_tokens": 14502469.0,
      "step": 7405
    },
    {
      "entropy": 6.263181686401367,
      "epoch": 0.7407407407407407,
      "grad_norm": 1.25,
      "learning_rate": 0.0004953655022622843,
      "loss": 6.2016,
      "mean_token_accuracy": 0.1444786325097084,
      "num_tokens": 14512939.0,
      "step": 7410
    },
    {
      "entropy": 6.291695547103882,
      "epoch": 0.7412405658019693,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004953582931837699,
      "loss": 6.1987,
      "mean_token_accuracy": 0.14422058761119844,
      "num_tokens": 14522681.0,
      "step": 7415
    },
    {
      "entropy": 6.34011025428772,
      "epoch": 0.7417403908631979,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004953510785610976,
      "loss": 6.0426,
      "mean_token_accuracy": 0.1551153466105461,
      "num_tokens": 14532115.0,
      "step": 7420
    },
    {
      "entropy": 6.249790334701538,
      "epoch": 0.7422402159244265,
      "grad_norm": 1.25,
      "learning_rate": 0.0004953438583944487,
      "loss": 6.1616,
      "mean_token_accuracy": 0.1463777095079422,
      "num_tokens": 14541147.0,
      "step": 7425
    },
    {
      "entropy": 6.221587467193603,
      "epoch": 0.742740040985655,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004953366326840052,
      "loss": 6.1851,
      "mean_token_accuracy": 0.1517923206090927,
      "num_tokens": 14550373.0,
      "step": 7430
    },
    {
      "entropy": 6.341326713562012,
      "epoch": 0.7432398660468836,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004953294014299486,
      "loss": 6.203,
      "mean_token_accuracy": 0.15120259523391724,
      "num_tokens": 14559396.0,
      "step": 7435
    },
    {
      "entropy": 6.33515477180481,
      "epoch": 0.7437396911081121,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004953221646324611,
      "loss": 6.2511,
      "mean_token_accuracy": 0.14376300275325776,
      "num_tokens": 14570292.0,
      "step": 7440
    },
    {
      "entropy": 6.224924039840698,
      "epoch": 0.7442395161693407,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004953149222917245,
      "loss": 6.1201,
      "mean_token_accuracy": 0.1470565378665924,
      "num_tokens": 14579887.0,
      "step": 7445
    },
    {
      "entropy": 6.313384675979615,
      "epoch": 0.7447393412305693,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004953076744079213,
      "loss": 6.2159,
      "mean_token_accuracy": 0.14277174770832063,
      "num_tokens": 14590260.0,
      "step": 7450
    },
    {
      "entropy": 6.295294284820557,
      "epoch": 0.7452391662917979,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004953004209812335,
      "loss": 6.1375,
      "mean_token_accuracy": 0.1440180078148842,
      "num_tokens": 14599812.0,
      "step": 7455
    },
    {
      "entropy": 6.24394645690918,
      "epoch": 0.7457389913530265,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004952931620118439,
      "loss": 6.0753,
      "mean_token_accuracy": 0.15063059329986572,
      "num_tokens": 14609147.0,
      "step": 7460
    },
    {
      "entropy": 6.296684455871582,
      "epoch": 0.746238816414255,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004952858974999351,
      "loss": 6.1698,
      "mean_token_accuracy": 0.14148258045315742,
      "num_tokens": 14618834.0,
      "step": 7465
    },
    {
      "entropy": 6.206604385375977,
      "epoch": 0.7467386414754836,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004952786274456896,
      "loss": 6.0698,
      "mean_token_accuracy": 0.15565451979637146,
      "num_tokens": 14628556.0,
      "step": 7470
    },
    {
      "entropy": 6.201365661621094,
      "epoch": 0.7472384665367121,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004952713518492907,
      "loss": 6.141,
      "mean_token_accuracy": 0.14673066288232803,
      "num_tokens": 14637486.0,
      "step": 7475
    },
    {
      "entropy": 6.199154376983643,
      "epoch": 0.7477382915979407,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004952640707109213,
      "loss": 6.1284,
      "mean_token_accuracy": 0.1510494664311409,
      "num_tokens": 14647599.0,
      "step": 7480
    },
    {
      "entropy": 6.291010332107544,
      "epoch": 0.7482381166591693,
      "grad_norm": 1.375,
      "learning_rate": 0.0004952567840307644,
      "loss": 6.0289,
      "mean_token_accuracy": 0.15243840217590332,
      "num_tokens": 14656246.0,
      "step": 7485
    },
    {
      "entropy": 6.242139911651611,
      "epoch": 0.7487379417203979,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004952494918090036,
      "loss": 6.1583,
      "mean_token_accuracy": 0.14685693830251695,
      "num_tokens": 14665550.0,
      "step": 7490
    },
    {
      "entropy": 6.3156510353088375,
      "epoch": 0.7492377667816265,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004952421940458222,
      "loss": 6.1814,
      "mean_token_accuracy": 0.1445535846054554,
      "num_tokens": 14676462.0,
      "step": 7495
    },
    {
      "entropy": 6.227349185943604,
      "epoch": 0.749737591842855,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004952348907414039,
      "loss": 6.0179,
      "mean_token_accuracy": 0.15693516954779624,
      "num_tokens": 14685963.0,
      "step": 7500
    },
    {
      "entropy": 6.2448042869567875,
      "epoch": 0.7502374169040835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004952275818959325,
      "loss": 6.1707,
      "mean_token_accuracy": 0.14609416276216508,
      "num_tokens": 14695764.0,
      "step": 7505
    },
    {
      "entropy": 6.258151483535767,
      "epoch": 0.7507372419653121,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004952202675095916,
      "loss": 6.0992,
      "mean_token_accuracy": 0.1508508086204529,
      "num_tokens": 14705839.0,
      "step": 7510
    },
    {
      "entropy": 6.238934564590454,
      "epoch": 0.7512370670265407,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004952129475825657,
      "loss": 6.0751,
      "mean_token_accuracy": 0.15591567903757095,
      "num_tokens": 14714801.0,
      "step": 7515
    },
    {
      "entropy": 6.311402273178101,
      "epoch": 0.7517368920877693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004952056221150385,
      "loss": 6.2181,
      "mean_token_accuracy": 0.15030758529901506,
      "num_tokens": 14723803.0,
      "step": 7520
    },
    {
      "entropy": 6.318091297149659,
      "epoch": 0.7522367171489979,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004951982911071945,
      "loss": 6.1139,
      "mean_token_accuracy": 0.152104152739048,
      "num_tokens": 14732598.0,
      "step": 7525
    },
    {
      "entropy": 6.27641453742981,
      "epoch": 0.7527365422102265,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004951909545592183,
      "loss": 6.164,
      "mean_token_accuracy": 0.1490496352314949,
      "num_tokens": 14742598.0,
      "step": 7530
    },
    {
      "entropy": 6.224202871322632,
      "epoch": 0.753236367271455,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004951836124712942,
      "loss": 6.1868,
      "mean_token_accuracy": 0.1542062133550644,
      "num_tokens": 14751434.0,
      "step": 7535
    },
    {
      "entropy": 6.355421304702759,
      "epoch": 0.7537361923326835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004951762648436073,
      "loss": 6.1844,
      "mean_token_accuracy": 0.1477696493268013,
      "num_tokens": 14760661.0,
      "step": 7540
    },
    {
      "entropy": 6.337145137786865,
      "epoch": 0.7542360173939121,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004951689116763419,
      "loss": 6.1808,
      "mean_token_accuracy": 0.14135282933712007,
      "num_tokens": 14770431.0,
      "step": 7545
    },
    {
      "entropy": 6.235747861862182,
      "epoch": 0.7547358424551407,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004951615529696835,
      "loss": 6.1779,
      "mean_token_accuracy": 0.14604596346616744,
      "num_tokens": 14780066.0,
      "step": 7550
    },
    {
      "entropy": 6.281822252273559,
      "epoch": 0.7552356675163693,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004951541887238172,
      "loss": 6.1834,
      "mean_token_accuracy": 0.14925995394587516,
      "num_tokens": 14789531.0,
      "step": 7555
    },
    {
      "entropy": 6.32441291809082,
      "epoch": 0.7557354925775979,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004951468189389281,
      "loss": 6.0949,
      "mean_token_accuracy": 0.14842487275600433,
      "num_tokens": 14799000.0,
      "step": 7560
    },
    {
      "entropy": 6.213815116882325,
      "epoch": 0.7562353176388265,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004951394436152016,
      "loss": 6.1315,
      "mean_token_accuracy": 0.15473720282316208,
      "num_tokens": 14808991.0,
      "step": 7565
    },
    {
      "entropy": 6.242095899581909,
      "epoch": 0.7567351427000549,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004951320627528234,
      "loss": 6.1652,
      "mean_token_accuracy": 0.15253257900476455,
      "num_tokens": 14819351.0,
      "step": 7570
    },
    {
      "entropy": 6.266114091873169,
      "epoch": 0.7572349677612835,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004951246763519791,
      "loss": 6.1829,
      "mean_token_accuracy": 0.14603662714362145,
      "num_tokens": 14828924.0,
      "step": 7575
    },
    {
      "entropy": 6.448176622390747,
      "epoch": 0.7577347928225121,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004951172844128546,
      "loss": 6.3275,
      "mean_token_accuracy": 0.1285861536860466,
      "num_tokens": 14841077.0,
      "step": 7580
    },
    {
      "entropy": 6.410078239440918,
      "epoch": 0.7582346178837407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004951098869356358,
      "loss": 6.2551,
      "mean_token_accuracy": 0.1365300476551056,
      "num_tokens": 14851223.0,
      "step": 7585
    },
    {
      "entropy": 6.320130109786987,
      "epoch": 0.7587344429449693,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000495102483920509,
      "loss": 6.088,
      "mean_token_accuracy": 0.1487969607114792,
      "num_tokens": 14861003.0,
      "step": 7590
    },
    {
      "entropy": 6.334915590286255,
      "epoch": 0.7592342680061979,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004950950753676603,
      "loss": 6.2199,
      "mean_token_accuracy": 0.14157491624355317,
      "num_tokens": 14870207.0,
      "step": 7595
    },
    {
      "entropy": 6.180481767654419,
      "epoch": 0.7597340930674265,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004950876612772763,
      "loss": 6.1095,
      "mean_token_accuracy": 0.1504347875714302,
      "num_tokens": 14879886.0,
      "step": 7600
    },
    {
      "entropy": 6.199652433395386,
      "epoch": 0.7602339181286549,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950802416495431,
      "loss": 6.0979,
      "mean_token_accuracy": 0.15191600620746612,
      "num_tokens": 14889275.0,
      "step": 7605
    },
    {
      "entropy": 6.374163436889648,
      "epoch": 0.7607337431898835,
      "grad_norm": 1.25,
      "learning_rate": 0.0004950728164846478,
      "loss": 6.1958,
      "mean_token_accuracy": 0.14734057039022447,
      "num_tokens": 14898317.0,
      "step": 7610
    },
    {
      "entropy": 6.291928720474243,
      "epoch": 0.7612335682511121,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004950653857827771,
      "loss": 6.1199,
      "mean_token_accuracy": 0.15374854654073716,
      "num_tokens": 14907183.0,
      "step": 7615
    },
    {
      "entropy": 6.262668514251709,
      "epoch": 0.7617333933123407,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004950579495441177,
      "loss": 6.2769,
      "mean_token_accuracy": 0.14063684716820718,
      "num_tokens": 14917564.0,
      "step": 7620
    },
    {
      "entropy": 6.322841644287109,
      "epoch": 0.7622332183735693,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004950505077688572,
      "loss": 6.097,
      "mean_token_accuracy": 0.15270165205001832,
      "num_tokens": 14927102.0,
      "step": 7625
    },
    {
      "entropy": 6.144235038757325,
      "epoch": 0.7627330434347979,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004950430604571824,
      "loss": 5.9942,
      "mean_token_accuracy": 0.1637290060520172,
      "num_tokens": 14937934.0,
      "step": 7630
    },
    {
      "entropy": 6.279917001724243,
      "epoch": 0.7632328684960263,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004950356076092808,
      "loss": 6.158,
      "mean_token_accuracy": 0.14633598923683167,
      "num_tokens": 14948109.0,
      "step": 7635
    },
    {
      "entropy": 6.273922777175903,
      "epoch": 0.7637326935572549,
      "grad_norm": 1.34375,
      "learning_rate": 0.00049502814922534,
      "loss": 6.1949,
      "mean_token_accuracy": 0.14215281307697297,
      "num_tokens": 14956971.0,
      "step": 7640
    },
    {
      "entropy": 6.253824615478516,
      "epoch": 0.7642325186184835,
      "grad_norm": 2.6875,
      "learning_rate": 0.0004950206853055475,
      "loss": 6.1432,
      "mean_token_accuracy": 0.15685205310583114,
      "num_tokens": 14967046.0,
      "step": 7645
    },
    {
      "entropy": 6.296667289733887,
      "epoch": 0.7647323436797121,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004950132158500913,
      "loss": 6.126,
      "mean_token_accuracy": 0.1458499863743782,
      "num_tokens": 14976234.0,
      "step": 7650
    },
    {
      "entropy": 6.1665582180023195,
      "epoch": 0.7652321687409407,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004950057408591591,
      "loss": 6.0744,
      "mean_token_accuracy": 0.15667738020420074,
      "num_tokens": 14985205.0,
      "step": 7655
    },
    {
      "entropy": 6.139488077163696,
      "epoch": 0.7657319938021693,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004949982603329391,
      "loss": 6.0707,
      "mean_token_accuracy": 0.15247628539800645,
      "num_tokens": 14994233.0,
      "step": 7660
    },
    {
      "entropy": 6.25272388458252,
      "epoch": 0.7662318188633979,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004949907742716195,
      "loss": 6.1337,
      "mean_token_accuracy": 0.15385272800922395,
      "num_tokens": 15004001.0,
      "step": 7665
    },
    {
      "entropy": 6.324283599853516,
      "epoch": 0.7667316439246263,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004949832826753886,
      "loss": 6.1577,
      "mean_token_accuracy": 0.1397901549935341,
      "num_tokens": 15015046.0,
      "step": 7670
    },
    {
      "entropy": 6.21451735496521,
      "epoch": 0.7672314689858549,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000494975785544435,
      "loss": 6.0893,
      "mean_token_accuracy": 0.1504504442214966,
      "num_tokens": 15025404.0,
      "step": 7675
    },
    {
      "entropy": 6.285001659393311,
      "epoch": 0.7677312940470835,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004949682828789472,
      "loss": 6.1872,
      "mean_token_accuracy": 0.1416396342217922,
      "num_tokens": 15036160.0,
      "step": 7680
    },
    {
      "entropy": 6.2325386047363285,
      "epoch": 0.7682311191083121,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494960774679114,
      "loss": 6.0999,
      "mean_token_accuracy": 0.15209167897701265,
      "num_tokens": 15046356.0,
      "step": 7685
    },
    {
      "entropy": 6.220134687423706,
      "epoch": 0.7687309441695407,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004949532609451242,
      "loss": 6.1047,
      "mean_token_accuracy": 0.14704525470733643,
      "num_tokens": 15056101.0,
      "step": 7690
    },
    {
      "entropy": 6.281166315078735,
      "epoch": 0.7692307692307693,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004949457416771671,
      "loss": 6.1052,
      "mean_token_accuracy": 0.15407356172800063,
      "num_tokens": 15065726.0,
      "step": 7695
    },
    {
      "entropy": 6.217519998550415,
      "epoch": 0.7697305942919977,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004949382168754316,
      "loss": 6.082,
      "mean_token_accuracy": 0.15789294242858887,
      "num_tokens": 15075211.0,
      "step": 7700
    },
    {
      "entropy": 6.176139068603516,
      "epoch": 0.7702304193532263,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004949306865401073,
      "loss": 6.0085,
      "mean_token_accuracy": 0.1518138200044632,
      "num_tokens": 15085525.0,
      "step": 7705
    },
    {
      "entropy": 6.290589666366577,
      "epoch": 0.7707302444144549,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004949231506713833,
      "loss": 6.2111,
      "mean_token_accuracy": 0.14028151333332062,
      "num_tokens": 15093851.0,
      "step": 7710
    },
    {
      "entropy": 6.390812492370605,
      "epoch": 0.7712300694756835,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004949156092694496,
      "loss": 6.2805,
      "mean_token_accuracy": 0.13829066157341002,
      "num_tokens": 15104066.0,
      "step": 7715
    },
    {
      "entropy": 6.316229248046875,
      "epoch": 0.7717298945369121,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004949080623344958,
      "loss": 6.1169,
      "mean_token_accuracy": 0.1493843212723732,
      "num_tokens": 15114047.0,
      "step": 7720
    },
    {
      "entropy": 6.217624187469482,
      "epoch": 0.7722297195981407,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004949005098667117,
      "loss": 6.1184,
      "mean_token_accuracy": 0.15427873581647872,
      "num_tokens": 15123719.0,
      "step": 7725
    },
    {
      "entropy": 6.179004573822022,
      "epoch": 0.7727295446593693,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004948929518662873,
      "loss": 6.1737,
      "mean_token_accuracy": 0.14911814630031586,
      "num_tokens": 15133978.0,
      "step": 7730
    },
    {
      "entropy": 6.300451707839966,
      "epoch": 0.7732293697205977,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004948853883334129,
      "loss": 6.199,
      "mean_token_accuracy": 0.1457667052745819,
      "num_tokens": 15145058.0,
      "step": 7735
    },
    {
      "entropy": 6.389480972290039,
      "epoch": 0.7737291947818263,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004948778192682787,
      "loss": 6.2094,
      "mean_token_accuracy": 0.1432968035340309,
      "num_tokens": 15155400.0,
      "step": 7740
    },
    {
      "entropy": 6.18777379989624,
      "epoch": 0.7742290198430549,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000494870244671075,
      "loss": 6.0416,
      "mean_token_accuracy": 0.1611133858561516,
      "num_tokens": 15164969.0,
      "step": 7745
    },
    {
      "entropy": 6.236864519119263,
      "epoch": 0.7747288449042835,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004948626645419926,
      "loss": 6.12,
      "mean_token_accuracy": 0.15284099280834199,
      "num_tokens": 15173840.0,
      "step": 7750
    },
    {
      "entropy": 6.2907976627349855,
      "epoch": 0.7752286699655121,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004948550788812222,
      "loss": 6.2204,
      "mean_token_accuracy": 0.13937411978840827,
      "num_tokens": 15184541.0,
      "step": 7755
    },
    {
      "entropy": 6.29304141998291,
      "epoch": 0.7757284950267407,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004948474876889545,
      "loss": 6.1579,
      "mean_token_accuracy": 0.15075905323028566,
      "num_tokens": 15195060.0,
      "step": 7760
    },
    {
      "entropy": 6.320642518997192,
      "epoch": 0.7762283200879692,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004948398909653806,
      "loss": 6.1885,
      "mean_token_accuracy": 0.1424543760716915,
      "num_tokens": 15206093.0,
      "step": 7765
    },
    {
      "entropy": 6.292466402053833,
      "epoch": 0.7767281451491977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004948322887106917,
      "loss": 6.1425,
      "mean_token_accuracy": 0.14284879937767983,
      "num_tokens": 15215359.0,
      "step": 7770
    },
    {
      "entropy": 6.2611757755279545,
      "epoch": 0.7772279702104263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004948246809250788,
      "loss": 6.1938,
      "mean_token_accuracy": 0.14619085863232611,
      "num_tokens": 15225283.0,
      "step": 7775
    },
    {
      "entropy": 6.220918893814087,
      "epoch": 0.7777277952716549,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004948170676087338,
      "loss": 6.0549,
      "mean_token_accuracy": 0.15439228415489198,
      "num_tokens": 15234319.0,
      "step": 7780
    },
    {
      "entropy": 6.220788431167603,
      "epoch": 0.7782276203328835,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004948094487618476,
      "loss": 6.0271,
      "mean_token_accuracy": 0.15739640295505525,
      "num_tokens": 15244410.0,
      "step": 7785
    },
    {
      "entropy": 6.22987551689148,
      "epoch": 0.7787274453941121,
      "grad_norm": 1.25,
      "learning_rate": 0.0004948018243846126,
      "loss": 6.097,
      "mean_token_accuracy": 0.15216832756996154,
      "num_tokens": 15253431.0,
      "step": 7790
    },
    {
      "entropy": 6.249270248413086,
      "epoch": 0.7792272704553407,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004947941944772199,
      "loss": 6.144,
      "mean_token_accuracy": 0.14642169028520585,
      "num_tokens": 15262566.0,
      "step": 7795
    },
    {
      "entropy": 6.298220920562744,
      "epoch": 0.7797270955165692,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000494786559039862,
      "loss": 6.2042,
      "mean_token_accuracy": 0.146491177380085,
      "num_tokens": 15272122.0,
      "step": 7800
    },
    {
      "entropy": 6.30478572845459,
      "epoch": 0.7802269205777977,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004947789180727308,
      "loss": 6.1742,
      "mean_token_accuracy": 0.14181172400712966,
      "num_tokens": 15281165.0,
      "step": 7805
    },
    {
      "entropy": 6.242979669570923,
      "epoch": 0.7807267456390263,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004947712715760186,
      "loss": 6.1421,
      "mean_token_accuracy": 0.14454442709684373,
      "num_tokens": 15290992.0,
      "step": 7810
    },
    {
      "entropy": 6.349635267257691,
      "epoch": 0.7812265707002549,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004947636195499177,
      "loss": 6.1435,
      "mean_token_accuracy": 0.1442279890179634,
      "num_tokens": 15300454.0,
      "step": 7815
    },
    {
      "entropy": 6.270070219039917,
      "epoch": 0.7817263957614835,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004947559619946207,
      "loss": 6.098,
      "mean_token_accuracy": 0.1553158387541771,
      "num_tokens": 15310471.0,
      "step": 7820
    },
    {
      "entropy": 6.281497240066528,
      "epoch": 0.7822262208227121,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004947482989103203,
      "loss": 6.1823,
      "mean_token_accuracy": 0.14365701079368592,
      "num_tokens": 15320554.0,
      "step": 7825
    },
    {
      "entropy": 6.242315721511841,
      "epoch": 0.7827260458839406,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004947406302972092,
      "loss": 6.2175,
      "mean_token_accuracy": 0.1436364881694317,
      "num_tokens": 15331623.0,
      "step": 7830
    },
    {
      "entropy": 6.28975739479065,
      "epoch": 0.7832258709451692,
      "grad_norm": 1.375,
      "learning_rate": 0.0004947329561554804,
      "loss": 6.0692,
      "mean_token_accuracy": 0.15061132088303567,
      "num_tokens": 15340954.0,
      "step": 7835
    },
    {
      "entropy": 6.231539535522461,
      "epoch": 0.7837256960063977,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004947252764853269,
      "loss": 6.1876,
      "mean_token_accuracy": 0.14777177199721336,
      "num_tokens": 15350858.0,
      "step": 7840
    },
    {
      "entropy": 6.227728748321534,
      "epoch": 0.7842255210676263,
      "grad_norm": 1.25,
      "learning_rate": 0.0004947175912869421,
      "loss": 6.0796,
      "mean_token_accuracy": 0.14888300895690917,
      "num_tokens": 15360382.0,
      "step": 7845
    },
    {
      "entropy": 6.245897674560547,
      "epoch": 0.7847253461288549,
      "grad_norm": 1.25,
      "learning_rate": 0.0004947099005605191,
      "loss": 6.1882,
      "mean_token_accuracy": 0.14180536866188048,
      "num_tokens": 15370626.0,
      "step": 7850
    },
    {
      "entropy": 6.27165904045105,
      "epoch": 0.7852251711900835,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004947022043062517,
      "loss": 6.1574,
      "mean_token_accuracy": 0.14667243137955666,
      "num_tokens": 15379582.0,
      "step": 7855
    },
    {
      "entropy": 6.261765432357788,
      "epoch": 0.7857249962513121,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004946945025243333,
      "loss": 6.0636,
      "mean_token_accuracy": 0.1543782025575638,
      "num_tokens": 15390168.0,
      "step": 7860
    },
    {
      "entropy": 6.078561210632325,
      "epoch": 0.7862248213125406,
      "grad_norm": 1.25,
      "learning_rate": 0.0004946867952149577,
      "loss": 5.9986,
      "mean_token_accuracy": 0.16364614814519882,
      "num_tokens": 15400979.0,
      "step": 7865
    },
    {
      "entropy": 6.269335794448852,
      "epoch": 0.7867246463737692,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004946790823783189,
      "loss": 6.1056,
      "mean_token_accuracy": 0.14975828379392625,
      "num_tokens": 15409954.0,
      "step": 7870
    },
    {
      "entropy": 6.258737325668335,
      "epoch": 0.7872244714349977,
      "grad_norm": 1.375,
      "learning_rate": 0.000494671364014611,
      "loss": 6.1908,
      "mean_token_accuracy": 0.14610916376113892,
      "num_tokens": 15419539.0,
      "step": 7875
    },
    {
      "entropy": 6.174856567382813,
      "epoch": 0.7877242964962263,
      "grad_norm": 1.234375,
      "learning_rate": 0.000494663640124028,
      "loss": 6.1255,
      "mean_token_accuracy": 0.14952130913734435,
      "num_tokens": 15428751.0,
      "step": 7880
    },
    {
      "entropy": 6.304116821289062,
      "epoch": 0.7882241215574549,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004946559107067643,
      "loss": 6.1376,
      "mean_token_accuracy": 0.15365694165229798,
      "num_tokens": 15438466.0,
      "step": 7885
    },
    {
      "entropy": 6.259090757369995,
      "epoch": 0.7887239466186835,
      "grad_norm": 1.25,
      "learning_rate": 0.0004946481757630146,
      "loss": 6.1296,
      "mean_token_accuracy": 0.14955679550766945,
      "num_tokens": 15448395.0,
      "step": 7890
    },
    {
      "entropy": 6.228724956512451,
      "epoch": 0.789223771679912,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004946404352929732,
      "loss": 6.1498,
      "mean_token_accuracy": 0.14119478985667228,
      "num_tokens": 15457986.0,
      "step": 7895
    },
    {
      "entropy": 6.2512843132019045,
      "epoch": 0.7897235967411406,
      "grad_norm": 1.171875,
      "learning_rate": 0.000494632689296835,
      "loss": 6.1416,
      "mean_token_accuracy": 0.1456635534763336,
      "num_tokens": 15467989.0,
      "step": 7900
    },
    {
      "entropy": 6.212045669555664,
      "epoch": 0.7902234218023692,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004946249377747947,
      "loss": 6.035,
      "mean_token_accuracy": 0.15187643766403197,
      "num_tokens": 15477537.0,
      "step": 7905
    },
    {
      "entropy": 6.246332311630249,
      "epoch": 0.7907232468635977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004946171807270476,
      "loss": 6.1058,
      "mean_token_accuracy": 0.1481271728873253,
      "num_tokens": 15488007.0,
      "step": 7910
    },
    {
      "entropy": 6.2981274127960205,
      "epoch": 0.7912230719248263,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004946094181537887,
      "loss": 6.143,
      "mean_token_accuracy": 0.1494605705142021,
      "num_tokens": 15498122.0,
      "step": 7915
    },
    {
      "entropy": 6.269918155670166,
      "epoch": 0.7917228969860549,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004946016500552134,
      "loss": 6.1629,
      "mean_token_accuracy": 0.14419534131884576,
      "num_tokens": 15507585.0,
      "step": 7920
    },
    {
      "entropy": 6.2122949123382565,
      "epoch": 0.7922227220472835,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945938764315171,
      "loss": 6.0815,
      "mean_token_accuracy": 0.15033162757754326,
      "num_tokens": 15516260.0,
      "step": 7925
    },
    {
      "entropy": 6.265510845184326,
      "epoch": 0.792722547108512,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945860972828954,
      "loss": 6.1941,
      "mean_token_accuracy": 0.14702940732240677,
      "num_tokens": 15527372.0,
      "step": 7930
    },
    {
      "entropy": 6.238861894607544,
      "epoch": 0.7932223721697406,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004945783126095439,
      "loss": 6.1658,
      "mean_token_accuracy": 0.14506676122546197,
      "num_tokens": 15537244.0,
      "step": 7935
    },
    {
      "entropy": 6.359428453445434,
      "epoch": 0.7937221972309692,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004945705224116586,
      "loss": 6.169,
      "mean_token_accuracy": 0.14562607556581497,
      "num_tokens": 15546512.0,
      "step": 7940
    },
    {
      "entropy": 6.2519354820251465,
      "epoch": 0.7942220222921977,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004945627266894354,
      "loss": 6.1266,
      "mean_token_accuracy": 0.14432251453399658,
      "num_tokens": 15557381.0,
      "step": 7945
    },
    {
      "entropy": 6.26039342880249,
      "epoch": 0.7947218473534263,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004945549254430707,
      "loss": 6.2042,
      "mean_token_accuracy": 0.14475180879235267,
      "num_tokens": 15565806.0,
      "step": 7950
    },
    {
      "entropy": 6.247186851501465,
      "epoch": 0.7952216724146549,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004945471186727603,
      "loss": 6.1235,
      "mean_token_accuracy": 0.1465785175561905,
      "num_tokens": 15577059.0,
      "step": 7955
    },
    {
      "entropy": 6.366255378723144,
      "epoch": 0.7957214974758834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004945393063787011,
      "loss": 6.3111,
      "mean_token_accuracy": 0.13643303140997887,
      "num_tokens": 15587827.0,
      "step": 7960
    },
    {
      "entropy": 6.386262559890747,
      "epoch": 0.796221322537112,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004945314885610894,
      "loss": 6.0886,
      "mean_token_accuracy": 0.14473408013582229,
      "num_tokens": 15598232.0,
      "step": 7965
    },
    {
      "entropy": 6.211160945892334,
      "epoch": 0.7967211475983406,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004945236652201219,
      "loss": 6.0492,
      "mean_token_accuracy": 0.1514046907424927,
      "num_tokens": 15606896.0,
      "step": 7970
    },
    {
      "entropy": 6.195062351226807,
      "epoch": 0.7972209726595691,
      "grad_norm": 1.25,
      "learning_rate": 0.0004945158363559955,
      "loss": 6.0659,
      "mean_token_accuracy": 0.15074210613965988,
      "num_tokens": 15616868.0,
      "step": 7975
    },
    {
      "entropy": 6.265796661376953,
      "epoch": 0.7977207977207977,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004945080019689071,
      "loss": 6.0843,
      "mean_token_accuracy": 0.1455327734351158,
      "num_tokens": 15626521.0,
      "step": 7980
    },
    {
      "entropy": 6.209790182113648,
      "epoch": 0.7982206227820263,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004945001620590538,
      "loss": 6.0783,
      "mean_token_accuracy": 0.15831656605005265,
      "num_tokens": 15635901.0,
      "step": 7985
    },
    {
      "entropy": 6.23649787902832,
      "epoch": 0.7987204478432549,
      "grad_norm": 1.359375,
      "learning_rate": 0.000494492316626633,
      "loss": 6.1337,
      "mean_token_accuracy": 0.14447178021073342,
      "num_tokens": 15645672.0,
      "step": 7990
    },
    {
      "entropy": 6.310642099380493,
      "epoch": 0.7992202729044834,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004944844656718419,
      "loss": 6.1715,
      "mean_token_accuracy": 0.14741648957133294,
      "num_tokens": 15656668.0,
      "step": 7995
    },
    {
      "entropy": 6.2679932594299315,
      "epoch": 0.799720097965712,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004944766091948783,
      "loss": 6.1111,
      "mean_token_accuracy": 0.148764169216156,
      "num_tokens": 15666176.0,
      "step": 8000
    },
    {
      "entropy": 6.252793312072754,
      "epoch": 0.8002199230269406,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004944687471959395,
      "loss": 6.1506,
      "mean_token_accuracy": 0.14224263802170753,
      "num_tokens": 15676162.0,
      "step": 8005
    },
    {
      "entropy": 6.297210264205932,
      "epoch": 0.8007197480881691,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004944608796752237,
      "loss": 6.1408,
      "mean_token_accuracy": 0.15124161988496782,
      "num_tokens": 15685393.0,
      "step": 8010
    },
    {
      "entropy": 6.234340572357178,
      "epoch": 0.8012195731493977,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004944530066329284,
      "loss": 6.0616,
      "mean_token_accuracy": 0.15364625602960585,
      "num_tokens": 15694944.0,
      "step": 8015
    },
    {
      "entropy": 6.277842998504639,
      "epoch": 0.8017193982106263,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004944451280692521,
      "loss": 6.2135,
      "mean_token_accuracy": 0.14151860475540162,
      "num_tokens": 15704410.0,
      "step": 8020
    },
    {
      "entropy": 6.284603595733643,
      "epoch": 0.8022192232718548,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004944372439843927,
      "loss": 6.1111,
      "mean_token_accuracy": 0.1499621458351612,
      "num_tokens": 15713663.0,
      "step": 8025
    },
    {
      "entropy": 6.22285418510437,
      "epoch": 0.8027190483330834,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004944293543785489,
      "loss": 6.0662,
      "mean_token_accuracy": 0.15269689112901688,
      "num_tokens": 15722662.0,
      "step": 8030
    },
    {
      "entropy": 6.26998987197876,
      "epoch": 0.803218873394312,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004944214592519188,
      "loss": 6.2397,
      "mean_token_accuracy": 0.14086568281054496,
      "num_tokens": 15732615.0,
      "step": 8035
    },
    {
      "entropy": 6.28104944229126,
      "epoch": 0.8037186984555406,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004944135586047014,
      "loss": 6.164,
      "mean_token_accuracy": 0.15626058131456375,
      "num_tokens": 15741569.0,
      "step": 8040
    },
    {
      "entropy": 6.252986097335816,
      "epoch": 0.8042185235167691,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004944056524370953,
      "loss": 6.0319,
      "mean_token_accuracy": 0.15656131207942964,
      "num_tokens": 15750734.0,
      "step": 8045
    },
    {
      "entropy": 6.241338396072388,
      "epoch": 0.8047183485779977,
      "grad_norm": 1.25,
      "learning_rate": 0.0004943977407492994,
      "loss": 6.1107,
      "mean_token_accuracy": 0.15088774263858795,
      "num_tokens": 15761664.0,
      "step": 8050
    },
    {
      "entropy": 6.215361833572388,
      "epoch": 0.8052181736392263,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004943898235415128,
      "loss": 6.054,
      "mean_token_accuracy": 0.15960732698440552,
      "num_tokens": 15771159.0,
      "step": 8055
    },
    {
      "entropy": 6.287186574935913,
      "epoch": 0.8057179987004548,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004943819008139347,
      "loss": 6.1713,
      "mean_token_accuracy": 0.14609452039003373,
      "num_tokens": 15781429.0,
      "step": 8060
    },
    {
      "entropy": 6.275288486480713,
      "epoch": 0.8062178237616834,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004943739725667644,
      "loss": 6.0811,
      "mean_token_accuracy": 0.14959293454885483,
      "num_tokens": 15791361.0,
      "step": 8065
    },
    {
      "entropy": 6.283444452285766,
      "epoch": 0.806717648822912,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004943660388002014,
      "loss": 6.2176,
      "mean_token_accuracy": 0.13855115920305253,
      "num_tokens": 15802020.0,
      "step": 8070
    },
    {
      "entropy": 6.192579841613769,
      "epoch": 0.8072174738841406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004943580995144454,
      "loss": 6.0219,
      "mean_token_accuracy": 0.1576543368399143,
      "num_tokens": 15811319.0,
      "step": 8075
    },
    {
      "entropy": 6.251135778427124,
      "epoch": 0.8077172989453691,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004943501547096961,
      "loss": 6.165,
      "mean_token_accuracy": 0.14632676020264626,
      "num_tokens": 15821337.0,
      "step": 8080
    },
    {
      "entropy": 6.276842832565308,
      "epoch": 0.8082171240065977,
      "grad_norm": 2.03125,
      "learning_rate": 0.0004943422043861531,
      "loss": 6.1612,
      "mean_token_accuracy": 0.14304060935974122,
      "num_tokens": 15832234.0,
      "step": 8085
    },
    {
      "entropy": 6.32946662902832,
      "epoch": 0.8087169490678262,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004943342485440169,
      "loss": 6.1484,
      "mean_token_accuracy": 0.14859842807054519,
      "num_tokens": 15841797.0,
      "step": 8090
    },
    {
      "entropy": 6.2384930610656735,
      "epoch": 0.8092167741290548,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004943262871834874,
      "loss": 6.1499,
      "mean_token_accuracy": 0.148804971575737,
      "num_tokens": 15851031.0,
      "step": 8095
    },
    {
      "entropy": 6.264900636672974,
      "epoch": 0.8097165991902834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004943183203047649,
      "loss": 6.1598,
      "mean_token_accuracy": 0.1434067964553833,
      "num_tokens": 15861297.0,
      "step": 8100
    },
    {
      "entropy": 6.259158325195313,
      "epoch": 0.810216424251512,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004943103479080499,
      "loss": 6.1818,
      "mean_token_accuracy": 0.14457603693008422,
      "num_tokens": 15871066.0,
      "step": 8105
    },
    {
      "entropy": 6.289132833480835,
      "epoch": 0.8107162493127406,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000494302369993543,
      "loss": 6.1429,
      "mean_token_accuracy": 0.1471883848309517,
      "num_tokens": 15880395.0,
      "step": 8110
    },
    {
      "entropy": 6.2609984397888185,
      "epoch": 0.8112160743739691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004942943865614449,
      "loss": 6.0989,
      "mean_token_accuracy": 0.15523406416177749,
      "num_tokens": 15889937.0,
      "step": 8115
    },
    {
      "entropy": 6.302601289749146,
      "epoch": 0.8117158994351977,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004942863976119565,
      "loss": 6.1147,
      "mean_token_accuracy": 0.1481298550963402,
      "num_tokens": 15899894.0,
      "step": 8120
    },
    {
      "entropy": 6.156360101699829,
      "epoch": 0.8122157244964262,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004942784031452788,
      "loss": 5.9939,
      "mean_token_accuracy": 0.16302908807992936,
      "num_tokens": 15909038.0,
      "step": 8125
    },
    {
      "entropy": 6.1888800144195555,
      "epoch": 0.8127155495576548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004942704031616127,
      "loss": 6.1083,
      "mean_token_accuracy": 0.1479555383324623,
      "num_tokens": 15917891.0,
      "step": 8130
    },
    {
      "entropy": 6.356932258605957,
      "epoch": 0.8132153746188834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004942623976611598,
      "loss": 6.2352,
      "mean_token_accuracy": 0.13697459399700165,
      "num_tokens": 15928079.0,
      "step": 8135
    },
    {
      "entropy": 6.232399320602417,
      "epoch": 0.813715199680112,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004942543866441215,
      "loss": 6.0724,
      "mean_token_accuracy": 0.1500664010643959,
      "num_tokens": 15938065.0,
      "step": 8140
    },
    {
      "entropy": 6.178165769577026,
      "epoch": 0.8142150247413406,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004942463701106992,
      "loss": 6.0959,
      "mean_token_accuracy": 0.14846319928765297,
      "num_tokens": 15947650.0,
      "step": 8145
    },
    {
      "entropy": 6.237065696716309,
      "epoch": 0.8147148498025691,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004942383480610946,
      "loss": 6.1043,
      "mean_token_accuracy": 0.15195035934448242,
      "num_tokens": 15957194.0,
      "step": 8150
    },
    {
      "entropy": 6.291400337219239,
      "epoch": 0.8152146748637976,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004942303204955096,
      "loss": 6.087,
      "mean_token_accuracy": 0.1471625730395317,
      "num_tokens": 15966068.0,
      "step": 8155
    },
    {
      "entropy": 6.212914705276489,
      "epoch": 0.8157144999250262,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004942222874141461,
      "loss": 6.0863,
      "mean_token_accuracy": 0.15084351524710654,
      "num_tokens": 15975672.0,
      "step": 8160
    },
    {
      "entropy": 6.216798114776611,
      "epoch": 0.8162143249862548,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004942142488172063,
      "loss": 6.1177,
      "mean_token_accuracy": 0.1479246884584427,
      "num_tokens": 15986216.0,
      "step": 8165
    },
    {
      "entropy": 6.22374701499939,
      "epoch": 0.8167141500474834,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004942062047048924,
      "loss": 6.0502,
      "mean_token_accuracy": 0.15558496564626695,
      "num_tokens": 15994673.0,
      "step": 8170
    },
    {
      "entropy": 6.242802095413208,
      "epoch": 0.817213975108712,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004941981550774068,
      "loss": 6.0851,
      "mean_token_accuracy": 0.15276721715927125,
      "num_tokens": 16003576.0,
      "step": 8175
    },
    {
      "entropy": 6.128625202178955,
      "epoch": 0.8177138001699406,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004941900999349521,
      "loss": 6.0744,
      "mean_token_accuracy": 0.15664535239338875,
      "num_tokens": 16013334.0,
      "step": 8180
    },
    {
      "entropy": 6.258645534515381,
      "epoch": 0.8182136252311691,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004941820392777307,
      "loss": 6.0872,
      "mean_token_accuracy": 0.15426221936941148,
      "num_tokens": 16023101.0,
      "step": 8185
    },
    {
      "entropy": 6.2186816215515135,
      "epoch": 0.8187134502923976,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004941739731059457,
      "loss": 6.1569,
      "mean_token_accuracy": 0.14787602350115775,
      "num_tokens": 16034189.0,
      "step": 8190
    },
    {
      "entropy": 6.251330137252808,
      "epoch": 0.8192132753536262,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004941659014198,
      "loss": 6.0657,
      "mean_token_accuracy": 0.15217609703540802,
      "num_tokens": 16044008.0,
      "step": 8195
    },
    {
      "entropy": 6.324179697036743,
      "epoch": 0.8197131004148548,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004941578242194965,
      "loss": 6.1555,
      "mean_token_accuracy": 0.1472798228263855,
      "num_tokens": 16053557.0,
      "step": 8200
    },
    {
      "entropy": 6.294137382507325,
      "epoch": 0.8202129254760834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004941497415052387,
      "loss": 6.0813,
      "mean_token_accuracy": 0.1436959832906723,
      "num_tokens": 16063191.0,
      "step": 8205
    },
    {
      "entropy": 6.24939923286438,
      "epoch": 0.820712750537312,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004941416532772296,
      "loss": 6.0924,
      "mean_token_accuracy": 0.14684049487113954,
      "num_tokens": 16072482.0,
      "step": 8210
    },
    {
      "entropy": 6.187750387191772,
      "epoch": 0.8212125755985406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000494133559535673,
      "loss": 6.0859,
      "mean_token_accuracy": 0.15419772565364837,
      "num_tokens": 16082848.0,
      "step": 8215
    },
    {
      "entropy": 6.2827372550964355,
      "epoch": 0.821712400659769,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004941254602807724,
      "loss": 6.2319,
      "mean_token_accuracy": 0.14404311627149582,
      "num_tokens": 16091604.0,
      "step": 8220
    },
    {
      "entropy": 6.328517723083496,
      "epoch": 0.8222122257209976,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004941173555127315,
      "loss": 6.1026,
      "mean_token_accuracy": 0.14960498064756395,
      "num_tokens": 16101427.0,
      "step": 8225
    },
    {
      "entropy": 6.179753541946411,
      "epoch": 0.8227120507822262,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004941092452317546,
      "loss": 6.0761,
      "mean_token_accuracy": 0.14994554594159126,
      "num_tokens": 16112206.0,
      "step": 8230
    },
    {
      "entropy": 6.25575442314148,
      "epoch": 0.8232118758434548,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004941011294380452,
      "loss": 6.2041,
      "mean_token_accuracy": 0.1474871329963207,
      "num_tokens": 16122652.0,
      "step": 8235
    },
    {
      "entropy": 6.22660722732544,
      "epoch": 0.8237117009046834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004940930081318079,
      "loss": 6.0749,
      "mean_token_accuracy": 0.15140161216259002,
      "num_tokens": 16132976.0,
      "step": 8240
    },
    {
      "entropy": 6.294421243667602,
      "epoch": 0.824211525965912,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004940848813132469,
      "loss": 6.1688,
      "mean_token_accuracy": 0.14806075468659402,
      "num_tokens": 16143514.0,
      "step": 8245
    },
    {
      "entropy": 6.3129771709442135,
      "epoch": 0.8247113510271405,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004940767489825666,
      "loss": 6.1556,
      "mean_token_accuracy": 0.1452951177954674,
      "num_tokens": 16154235.0,
      "step": 8250
    },
    {
      "entropy": 6.252210807800293,
      "epoch": 0.825211176088369,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004940686111399716,
      "loss": 6.0709,
      "mean_token_accuracy": 0.14720857962965966,
      "num_tokens": 16164251.0,
      "step": 8255
    },
    {
      "entropy": 6.163020896911621,
      "epoch": 0.8257110011495976,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004940604677856669,
      "loss": 6.0098,
      "mean_token_accuracy": 0.159994275867939,
      "num_tokens": 16172780.0,
      "step": 8260
    },
    {
      "entropy": 6.068484830856323,
      "epoch": 0.8262108262108262,
      "grad_norm": 1.265625,
      "learning_rate": 0.000494052318919857,
      "loss": 5.9823,
      "mean_token_accuracy": 0.1593278169631958,
      "num_tokens": 16182443.0,
      "step": 8265
    },
    {
      "entropy": 6.1880988597869875,
      "epoch": 0.8267106512720548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004940441645427473,
      "loss": 6.0799,
      "mean_token_accuracy": 0.15314688831567763,
      "num_tokens": 16192515.0,
      "step": 8270
    },
    {
      "entropy": 6.235993576049805,
      "epoch": 0.8272104763332834,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004940360046545427,
      "loss": 6.0788,
      "mean_token_accuracy": 0.15016637444496156,
      "num_tokens": 16201615.0,
      "step": 8275
    },
    {
      "entropy": 6.296337890625,
      "epoch": 0.827710301394512,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004940278392554486,
      "loss": 6.1186,
      "mean_token_accuracy": 0.14623932614922525,
      "num_tokens": 16210461.0,
      "step": 8280
    },
    {
      "entropy": 6.212397527694702,
      "epoch": 0.8282101264557404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004940196683456704,
      "loss": 6.1081,
      "mean_token_accuracy": 0.14230554327368736,
      "num_tokens": 16220293.0,
      "step": 8285
    },
    {
      "entropy": 6.192126226425171,
      "epoch": 0.828709951516969,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004940114919254137,
      "loss": 6.0461,
      "mean_token_accuracy": 0.147898331284523,
      "num_tokens": 16230020.0,
      "step": 8290
    },
    {
      "entropy": 6.2220683097839355,
      "epoch": 0.8292097765781976,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004940033099948843,
      "loss": 6.0787,
      "mean_token_accuracy": 0.14865076094865798,
      "num_tokens": 16240187.0,
      "step": 8295
    },
    {
      "entropy": 6.185304594039917,
      "epoch": 0.8297096016394262,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004939951225542879,
      "loss": 6.0881,
      "mean_token_accuracy": 0.15052397847175597,
      "num_tokens": 16250200.0,
      "step": 8300
    },
    {
      "entropy": 6.2307592868804935,
      "epoch": 0.8302094267006548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004939869296038306,
      "loss": 6.1659,
      "mean_token_accuracy": 0.14886408001184465,
      "num_tokens": 16259396.0,
      "step": 8305
    },
    {
      "entropy": 6.297973155975342,
      "epoch": 0.8307092517618834,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004939787311437185,
      "loss": 6.1629,
      "mean_token_accuracy": 0.13925077468156816,
      "num_tokens": 16270033.0,
      "step": 8310
    },
    {
      "entropy": 6.341564130783081,
      "epoch": 0.831209076823112,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004939705271741579,
      "loss": 6.1293,
      "mean_token_accuracy": 0.14668145179748535,
      "num_tokens": 16281464.0,
      "step": 8315
    },
    {
      "entropy": 6.344202089309692,
      "epoch": 0.8317089018843404,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004939623176953551,
      "loss": 6.1944,
      "mean_token_accuracy": 0.13954090774059297,
      "num_tokens": 16292125.0,
      "step": 8320
    },
    {
      "entropy": 6.198182868957519,
      "epoch": 0.832208726945569,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004939541027075168,
      "loss": 6.1069,
      "mean_token_accuracy": 0.1497854433953762,
      "num_tokens": 16302993.0,
      "step": 8325
    },
    {
      "entropy": 6.202881574630737,
      "epoch": 0.8327085520067976,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004939458822108496,
      "loss": 6.0897,
      "mean_token_accuracy": 0.1481951244175434,
      "num_tokens": 16311287.0,
      "step": 8330
    },
    {
      "entropy": 6.295244646072388,
      "epoch": 0.8332083770680262,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004939376562055602,
      "loss": 6.0758,
      "mean_token_accuracy": 0.14660119265317917,
      "num_tokens": 16320701.0,
      "step": 8335
    },
    {
      "entropy": 6.223574304580689,
      "epoch": 0.8337082021292548,
      "grad_norm": 1.375,
      "learning_rate": 0.0004939294246918558,
      "loss": 6.1108,
      "mean_token_accuracy": 0.1493000753223896,
      "num_tokens": 16329834.0,
      "step": 8340
    },
    {
      "entropy": 6.205131244659424,
      "epoch": 0.8342080271904834,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004939211876699435,
      "loss": 6.1,
      "mean_token_accuracy": 0.14822338297963142,
      "num_tokens": 16339043.0,
      "step": 8345
    },
    {
      "entropy": 6.293129634857178,
      "epoch": 0.8347078522517118,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004939129451400305,
      "loss": 6.2134,
      "mean_token_accuracy": 0.14606078416109086,
      "num_tokens": 16348774.0,
      "step": 8350
    },
    {
      "entropy": 6.264648580551148,
      "epoch": 0.8352076773129404,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000493904697102324,
      "loss": 6.077,
      "mean_token_accuracy": 0.1439671091735363,
      "num_tokens": 16357999.0,
      "step": 8355
    },
    {
      "entropy": 6.26352367401123,
      "epoch": 0.835707502374169,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004938964435570317,
      "loss": 6.1451,
      "mean_token_accuracy": 0.14471773505210878,
      "num_tokens": 16369308.0,
      "step": 8360
    },
    {
      "entropy": 6.158105087280274,
      "epoch": 0.8362073274353976,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938881845043612,
      "loss": 6.04,
      "mean_token_accuracy": 0.1513890579342842,
      "num_tokens": 16378001.0,
      "step": 8365
    },
    {
      "entropy": 6.3232310771942135,
      "epoch": 0.8367071524966262,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004938799199445202,
      "loss": 6.2509,
      "mean_token_accuracy": 0.14490632116794586,
      "num_tokens": 16387925.0,
      "step": 8370
    },
    {
      "entropy": 6.367937898635864,
      "epoch": 0.8372069775578548,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004938716498777169,
      "loss": 6.1947,
      "mean_token_accuracy": 0.140191800147295,
      "num_tokens": 16398200.0,
      "step": 8375
    },
    {
      "entropy": 6.116406965255737,
      "epoch": 0.8377068026190834,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938633743041592,
      "loss": 5.9616,
      "mean_token_accuracy": 0.15980350077152253,
      "num_tokens": 16407874.0,
      "step": 8380
    },
    {
      "entropy": 6.201684188842774,
      "epoch": 0.8382066276803118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004938550932240552,
      "loss": 6.0939,
      "mean_token_accuracy": 0.14970097839832305,
      "num_tokens": 16417825.0,
      "step": 8385
    },
    {
      "entropy": 6.239827442169189,
      "epoch": 0.8387064527415404,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004938468066376134,
      "loss": 6.0868,
      "mean_token_accuracy": 0.1496803194284439,
      "num_tokens": 16427265.0,
      "step": 8390
    },
    {
      "entropy": 6.133868980407715,
      "epoch": 0.839206277802769,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004938385145450423,
      "loss": 6.021,
      "mean_token_accuracy": 0.15879987925291061,
      "num_tokens": 16436639.0,
      "step": 8395
    },
    {
      "entropy": 6.248953580856323,
      "epoch": 0.8397061028639976,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004938302169465505,
      "loss": 6.1244,
      "mean_token_accuracy": 0.1505824476480484,
      "num_tokens": 16446411.0,
      "step": 8400
    },
    {
      "entropy": 6.289775371551514,
      "epoch": 0.8402059279252262,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004938219138423467,
      "loss": 6.1518,
      "mean_token_accuracy": 0.1493359997868538,
      "num_tokens": 16455784.0,
      "step": 8405
    },
    {
      "entropy": 6.1737555980682375,
      "epoch": 0.8407057529864548,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004938136052326399,
      "loss": 5.9757,
      "mean_token_accuracy": 0.157905213534832,
      "num_tokens": 16464411.0,
      "step": 8410
    },
    {
      "entropy": 6.200588655471802,
      "epoch": 0.8412055780476833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004938052911176392,
      "loss": 6.1569,
      "mean_token_accuracy": 0.14453719928860664,
      "num_tokens": 16473299.0,
      "step": 8415
    },
    {
      "entropy": 6.119726753234863,
      "epoch": 0.8417054031089118,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004937969714975534,
      "loss": 5.9807,
      "mean_token_accuracy": 0.15740735679864884,
      "num_tokens": 16483040.0,
      "step": 8420
    },
    {
      "entropy": 6.188728380203247,
      "epoch": 0.8422052281701404,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004937886463725923,
      "loss": 6.0587,
      "mean_token_accuracy": 0.15332949608564378,
      "num_tokens": 16492608.0,
      "step": 8425
    },
    {
      "entropy": 6.333893489837647,
      "epoch": 0.842705053231369,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004937803157429651,
      "loss": 6.2059,
      "mean_token_accuracy": 0.1431631065905094,
      "num_tokens": 16503139.0,
      "step": 8430
    },
    {
      "entropy": 6.225705146789551,
      "epoch": 0.8432048782925976,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004937719796088815,
      "loss": 6.0737,
      "mean_token_accuracy": 0.15347908735275267,
      "num_tokens": 16511584.0,
      "step": 8435
    },
    {
      "entropy": 6.15557370185852,
      "epoch": 0.8437047033538262,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004937636379705511,
      "loss": 6.0643,
      "mean_token_accuracy": 0.14917184859514238,
      "num_tokens": 16522407.0,
      "step": 8440
    },
    {
      "entropy": 6.127405500411987,
      "epoch": 0.8442045284150548,
      "grad_norm": 1.25,
      "learning_rate": 0.0004937552908281839,
      "loss": 6.0979,
      "mean_token_accuracy": 0.1465989112854004,
      "num_tokens": 16532351.0,
      "step": 8445
    },
    {
      "entropy": 6.334324598312378,
      "epoch": 0.8447043534762833,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004937469381819899,
      "loss": 6.1664,
      "mean_token_accuracy": 0.14261157140135766,
      "num_tokens": 16541673.0,
      "step": 8450
    },
    {
      "entropy": 6.2667001247406,
      "epoch": 0.8452041785375118,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004937385800321792,
      "loss": 6.1407,
      "mean_token_accuracy": 0.1521382987499237,
      "num_tokens": 16551622.0,
      "step": 8455
    },
    {
      "entropy": 6.210046052932739,
      "epoch": 0.8457040035987404,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004937302163789621,
      "loss": 6.0865,
      "mean_token_accuracy": 0.15102536603808403,
      "num_tokens": 16559074.0,
      "step": 8460
    },
    {
      "entropy": 6.162340068817139,
      "epoch": 0.846203828659969,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004937218472225491,
      "loss": 6.0602,
      "mean_token_accuracy": 0.15858377367258072,
      "num_tokens": 16568650.0,
      "step": 8465
    },
    {
      "entropy": 6.214577579498291,
      "epoch": 0.8467036537211976,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004937134725631506,
      "loss": 6.1061,
      "mean_token_accuracy": 0.15394459217786788,
      "num_tokens": 16578696.0,
      "step": 8470
    },
    {
      "entropy": 6.282371950149536,
      "epoch": 0.8472034787824262,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004937050924009775,
      "loss": 6.1594,
      "mean_token_accuracy": 0.1439043678343296,
      "num_tokens": 16589014.0,
      "step": 8475
    },
    {
      "entropy": 6.218377876281738,
      "epoch": 0.8477033038436548,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004936967067362406,
      "loss": 6.0361,
      "mean_token_accuracy": 0.15218364596366882,
      "num_tokens": 16598319.0,
      "step": 8480
    },
    {
      "entropy": 6.223678922653198,
      "epoch": 0.8482031289048833,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004936883155691507,
      "loss": 6.1755,
      "mean_token_accuracy": 0.14707616940140725,
      "num_tokens": 16607941.0,
      "step": 8485
    },
    {
      "entropy": 6.23028335571289,
      "epoch": 0.8487029539661118,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004936799188999192,
      "loss": 6.118,
      "mean_token_accuracy": 0.15177458226680757,
      "num_tokens": 16618398.0,
      "step": 8490
    },
    {
      "entropy": 6.320610618591308,
      "epoch": 0.8492027790273404,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004936715167287571,
      "loss": 6.144,
      "mean_token_accuracy": 0.14272797778248786,
      "num_tokens": 16628953.0,
      "step": 8495
    },
    {
      "entropy": 6.243732357025147,
      "epoch": 0.849702604088569,
      "grad_norm": 1.4140625,
      "learning_rate": 0.000493663109055876,
      "loss": 6.0653,
      "mean_token_accuracy": 0.15223424583673478,
      "num_tokens": 16638783.0,
      "step": 8500
    },
    {
      "entropy": 6.22808690071106,
      "epoch": 0.8502024291497976,
      "grad_norm": 1.25,
      "learning_rate": 0.0004936546958814873,
      "loss": 6.1197,
      "mean_token_accuracy": 0.15219603180885316,
      "num_tokens": 16647432.0,
      "step": 8505
    },
    {
      "entropy": 6.292452144622803,
      "epoch": 0.8507022542110262,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004936462772058026,
      "loss": 6.105,
      "mean_token_accuracy": 0.14516105651855468,
      "num_tokens": 16656839.0,
      "step": 8510
    },
    {
      "entropy": 6.264522695541382,
      "epoch": 0.8512020792722547,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004936378530290339,
      "loss": 6.1858,
      "mean_token_accuracy": 0.14216080233454703,
      "num_tokens": 16667054.0,
      "step": 8515
    },
    {
      "entropy": 6.235785388946534,
      "epoch": 0.8517019043334833,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004936294233513931,
      "loss": 6.1115,
      "mean_token_accuracy": 0.1501529909670353,
      "num_tokens": 16676851.0,
      "step": 8520
    },
    {
      "entropy": 6.284620141983032,
      "epoch": 0.8522017293947118,
      "grad_norm": 1.375,
      "learning_rate": 0.0004936209881730922,
      "loss": 6.1148,
      "mean_token_accuracy": 0.14707250893115997,
      "num_tokens": 16686433.0,
      "step": 8525
    },
    {
      "entropy": 6.2035236835479735,
      "epoch": 0.8527015544559404,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004936125474943436,
      "loss": 6.0922,
      "mean_token_accuracy": 0.15978797525167465,
      "num_tokens": 16695859.0,
      "step": 8530
    },
    {
      "entropy": 6.230979251861572,
      "epoch": 0.853201379517169,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004936041013153596,
      "loss": 6.1076,
      "mean_token_accuracy": 0.15107351392507554,
      "num_tokens": 16704391.0,
      "step": 8535
    },
    {
      "entropy": 6.24996976852417,
      "epoch": 0.8537012045783976,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004935956496363525,
      "loss": 6.1888,
      "mean_token_accuracy": 0.1405817985534668,
      "num_tokens": 16715316.0,
      "step": 8540
    },
    {
      "entropy": 6.40833158493042,
      "epoch": 0.8542010296396262,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004935871924575352,
      "loss": 6.2714,
      "mean_token_accuracy": 0.13728706762194634,
      "num_tokens": 16726215.0,
      "step": 8545
    },
    {
      "entropy": 6.266237020492554,
      "epoch": 0.8547008547008547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935787297791204,
      "loss": 6.1056,
      "mean_token_accuracy": 0.14677743539214133,
      "num_tokens": 16735768.0,
      "step": 8550
    },
    {
      "entropy": 6.179116868972779,
      "epoch": 0.8552006797620832,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004935702616013209,
      "loss": 5.9155,
      "mean_token_accuracy": 0.16017419695854188,
      "num_tokens": 16744265.0,
      "step": 8555
    },
    {
      "entropy": 6.1934630393981935,
      "epoch": 0.8557005048233118,
      "grad_norm": 1.296875,
      "learning_rate": 0.00049356178792435,
      "loss": 6.166,
      "mean_token_accuracy": 0.15210893601179123,
      "num_tokens": 16753713.0,
      "step": 8560
    },
    {
      "entropy": 6.277865028381347,
      "epoch": 0.8562003298845404,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004935533087484207,
      "loss": 6.1129,
      "mean_token_accuracy": 0.14849793463945388,
      "num_tokens": 16763670.0,
      "step": 8565
    },
    {
      "entropy": 6.223178863525391,
      "epoch": 0.856700154945769,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004935448240737464,
      "loss": 6.1255,
      "mean_token_accuracy": 0.14845992624759674,
      "num_tokens": 16773504.0,
      "step": 8570
    },
    {
      "entropy": 6.207229471206665,
      "epoch": 0.8571999800069976,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004935363339005405,
      "loss": 6.1653,
      "mean_token_accuracy": 0.14824780002236365,
      "num_tokens": 16784181.0,
      "step": 8575
    },
    {
      "entropy": 6.280117988586426,
      "epoch": 0.8576998050682261,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004935278382290168,
      "loss": 6.1476,
      "mean_token_accuracy": 0.14702174365520476,
      "num_tokens": 16793610.0,
      "step": 8580
    },
    {
      "entropy": 6.2846839904785154,
      "epoch": 0.8581996301294547,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004935193370593887,
      "loss": 6.1907,
      "mean_token_accuracy": 0.1370268478989601,
      "num_tokens": 16804068.0,
      "step": 8585
    },
    {
      "entropy": 6.231192111968994,
      "epoch": 0.8586994551906832,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004935108303918705,
      "loss": 6.1363,
      "mean_token_accuracy": 0.15340641140937805,
      "num_tokens": 16813769.0,
      "step": 8590
    },
    {
      "entropy": 6.209807109832764,
      "epoch": 0.8591992802519118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004935023182266761,
      "loss": 6.0599,
      "mean_token_accuracy": 0.15377306640148164,
      "num_tokens": 16822859.0,
      "step": 8595
    },
    {
      "entropy": 6.3580468654632565,
      "epoch": 0.8596991053131404,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004934938005640194,
      "loss": 6.2806,
      "mean_token_accuracy": 0.1398426979780197,
      "num_tokens": 16834839.0,
      "step": 8600
    },
    {
      "entropy": 6.372250032424927,
      "epoch": 0.860198930374369,
      "grad_norm": 1.34375,
      "learning_rate": 0.000493485277404115,
      "loss": 6.1219,
      "mean_token_accuracy": 0.1425008900463581,
      "num_tokens": 16845205.0,
      "step": 8605
    },
    {
      "entropy": 6.294728994369507,
      "epoch": 0.8606987554355976,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004934767487471772,
      "loss": 6.1049,
      "mean_token_accuracy": 0.14174228757619858,
      "num_tokens": 16854874.0,
      "step": 8610
    },
    {
      "entropy": 6.1609419822692875,
      "epoch": 0.8611985804968261,
      "grad_norm": 2.203125,
      "learning_rate": 0.0004934682145934207,
      "loss": 6.0113,
      "mean_token_accuracy": 0.15708543360233307,
      "num_tokens": 16864771.0,
      "step": 8615
    },
    {
      "entropy": 6.171891212463379,
      "epoch": 0.8616984055580547,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00049345967494306,
      "loss": 6.0949,
      "mean_token_accuracy": 0.14856914430856705,
      "num_tokens": 16874227.0,
      "step": 8620
    },
    {
      "entropy": 6.167325353622436,
      "epoch": 0.8621982306192832,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004934511297963102,
      "loss": 6.0322,
      "mean_token_accuracy": 0.158851907402277,
      "num_tokens": 16883895.0,
      "step": 8625
    },
    {
      "entropy": 6.261799955368042,
      "epoch": 0.8626980556805118,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004934425791533861,
      "loss": 6.0544,
      "mean_token_accuracy": 0.1530415877699852,
      "num_tokens": 16893012.0,
      "step": 8630
    },
    {
      "entropy": 6.246605920791626,
      "epoch": 0.8631978807417404,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004934340230145031,
      "loss": 6.1349,
      "mean_token_accuracy": 0.15347109436988832,
      "num_tokens": 16903831.0,
      "step": 8635
    },
    {
      "entropy": 6.277946043014526,
      "epoch": 0.863697705802969,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004934254613798761,
      "loss": 6.1795,
      "mean_token_accuracy": 0.13722404167056085,
      "num_tokens": 16913649.0,
      "step": 8640
    },
    {
      "entropy": 6.184159898757935,
      "epoch": 0.8641975308641975,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004934168942497207,
      "loss": 6.0781,
      "mean_token_accuracy": 0.15299730226397515,
      "num_tokens": 16923063.0,
      "step": 8645
    },
    {
      "entropy": 6.291852617263794,
      "epoch": 0.8646973559254261,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004934083216242526,
      "loss": 6.0788,
      "mean_token_accuracy": 0.15082671046257018,
      "num_tokens": 16932901.0,
      "step": 8650
    },
    {
      "entropy": 6.288068199157715,
      "epoch": 0.8651971809866547,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004933997435036873,
      "loss": 6.2039,
      "mean_token_accuracy": 0.14317680895328522,
      "num_tokens": 16942781.0,
      "step": 8655
    },
    {
      "entropy": 6.2505628108978275,
      "epoch": 0.8656970060478832,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004933911598882405,
      "loss": 6.0934,
      "mean_token_accuracy": 0.14813961684703827,
      "num_tokens": 16952064.0,
      "step": 8660
    },
    {
      "entropy": 6.16200885772705,
      "epoch": 0.8661968311091118,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004933825707781284,
      "loss": 6.0284,
      "mean_token_accuracy": 0.1516401693224907,
      "num_tokens": 16961528.0,
      "step": 8665
    },
    {
      "entropy": 6.233776664733886,
      "epoch": 0.8666966561703404,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004933739761735671,
      "loss": 6.1239,
      "mean_token_accuracy": 0.14984436184167862,
      "num_tokens": 16971066.0,
      "step": 8670
    },
    {
      "entropy": 6.2612748622894285,
      "epoch": 0.867196481231569,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004933653760747726,
      "loss": 6.1497,
      "mean_token_accuracy": 0.14837488532066345,
      "num_tokens": 16981438.0,
      "step": 8675
    },
    {
      "entropy": 6.221832323074341,
      "epoch": 0.8676963062927975,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004933567704819615,
      "loss": 6.0462,
      "mean_token_accuracy": 0.1588964506983757,
      "num_tokens": 16990836.0,
      "step": 8680
    },
    {
      "entropy": 6.193079948425293,
      "epoch": 0.8681961313540261,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933481593953502,
      "loss": 6.0265,
      "mean_token_accuracy": 0.15681727975606918,
      "num_tokens": 17000254.0,
      "step": 8685
    },
    {
      "entropy": 6.216556119918823,
      "epoch": 0.8686959564152547,
      "grad_norm": 1.25,
      "learning_rate": 0.0004933395428151554,
      "loss": 6.0872,
      "mean_token_accuracy": 0.15317708849906922,
      "num_tokens": 17010081.0,
      "step": 8690
    },
    {
      "entropy": 6.17724757194519,
      "epoch": 0.8691957814764832,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000493330920741594,
      "loss": 5.9406,
      "mean_token_accuracy": 0.1592693328857422,
      "num_tokens": 17018967.0,
      "step": 8695
    },
    {
      "entropy": 6.209882164001465,
      "epoch": 0.8696956065377118,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004933222931748827,
      "loss": 6.0998,
      "mean_token_accuracy": 0.14961342960596086,
      "num_tokens": 17027615.0,
      "step": 8700
    },
    {
      "entropy": 6.2211010456085205,
      "epoch": 0.8701954315989404,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004933136601152387,
      "loss": 6.0741,
      "mean_token_accuracy": 0.15184015780687332,
      "num_tokens": 17038446.0,
      "step": 8705
    },
    {
      "entropy": 6.232716035842896,
      "epoch": 0.8706952566601689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004933050215628793,
      "loss": 6.1064,
      "mean_token_accuracy": 0.14452996253967285,
      "num_tokens": 17047787.0,
      "step": 8710
    },
    {
      "entropy": 6.216891050338745,
      "epoch": 0.8711950817213975,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004932963775180216,
      "loss": 6.1253,
      "mean_token_accuracy": 0.1500982813537121,
      "num_tokens": 17056842.0,
      "step": 8715
    },
    {
      "entropy": 6.197678565979004,
      "epoch": 0.8716949067826261,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004932877279808833,
      "loss": 6.1492,
      "mean_token_accuracy": 0.14616958051919937,
      "num_tokens": 17066217.0,
      "step": 8720
    },
    {
      "entropy": 6.339457702636719,
      "epoch": 0.8721947318438547,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004932790729516818,
      "loss": 6.2127,
      "mean_token_accuracy": 0.14135319069027902,
      "num_tokens": 17077760.0,
      "step": 8725
    },
    {
      "entropy": 6.198708438873291,
      "epoch": 0.8726945569050832,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004932704124306352,
      "loss": 6.0579,
      "mean_token_accuracy": 0.15177614539861678,
      "num_tokens": 17088251.0,
      "step": 8730
    },
    {
      "entropy": 6.2632228374481205,
      "epoch": 0.8731943819663118,
      "grad_norm": 1.3125,
      "learning_rate": 0.000493261746417961,
      "loss": 6.1029,
      "mean_token_accuracy": 0.14797398522496225,
      "num_tokens": 17098192.0,
      "step": 8735
    },
    {
      "entropy": 6.260759878158569,
      "epoch": 0.8736942070275404,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932530749138775,
      "loss": 6.0776,
      "mean_token_accuracy": 0.1498183324933052,
      "num_tokens": 17107787.0,
      "step": 8740
    },
    {
      "entropy": 6.276432180404663,
      "epoch": 0.8741940320887689,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004932443979186028,
      "loss": 6.1783,
      "mean_token_accuracy": 0.15007296204566956,
      "num_tokens": 17118125.0,
      "step": 8745
    },
    {
      "entropy": 6.238847732543945,
      "epoch": 0.8746938571499975,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004932357154323552,
      "loss": 6.0485,
      "mean_token_accuracy": 0.1506869301199913,
      "num_tokens": 17127909.0,
      "step": 8750
    },
    {
      "entropy": 6.200756597518921,
      "epoch": 0.8751936822112261,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004932270274553531,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14798829555511475,
      "num_tokens": 17137672.0,
      "step": 8755
    },
    {
      "entropy": 6.210784196853638,
      "epoch": 0.8756935072724547,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004932183339878153,
      "loss": 6.0811,
      "mean_token_accuracy": 0.14794449508190155,
      "num_tokens": 17147151.0,
      "step": 8760
    },
    {
      "entropy": 6.215718841552734,
      "epoch": 0.8761933323336832,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004932096350299602,
      "loss": 5.9748,
      "mean_token_accuracy": 0.15497036427259445,
      "num_tokens": 17156615.0,
      "step": 8765
    },
    {
      "entropy": 6.220007085800171,
      "epoch": 0.8766931573949118,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004932009305820069,
      "loss": 6.1053,
      "mean_token_accuracy": 0.15130169838666915,
      "num_tokens": 17166089.0,
      "step": 8770
    },
    {
      "entropy": 6.2222010612487795,
      "epoch": 0.8771929824561403,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004931922206441743,
      "loss": 6.1141,
      "mean_token_accuracy": 0.14922061413526536,
      "num_tokens": 17176565.0,
      "step": 8775
    },
    {
      "entropy": 6.1548075675964355,
      "epoch": 0.8776928075173689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004931835052166817,
      "loss": 6.0594,
      "mean_token_accuracy": 0.15178312957286835,
      "num_tokens": 17186987.0,
      "step": 8780
    },
    {
      "entropy": 6.180013036727905,
      "epoch": 0.8781926325785975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004931747842997481,
      "loss": 6.052,
      "mean_token_accuracy": 0.1553016185760498,
      "num_tokens": 17196325.0,
      "step": 8785
    },
    {
      "entropy": 6.302029800415039,
      "epoch": 0.8786924576398261,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004931660578935931,
      "loss": 6.1119,
      "mean_token_accuracy": 0.14662524461746215,
      "num_tokens": 17206210.0,
      "step": 8790
    },
    {
      "entropy": 6.223845148086548,
      "epoch": 0.8791922827010547,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004931573259984363,
      "loss": 6.0513,
      "mean_token_accuracy": 0.15506964772939683,
      "num_tokens": 17216521.0,
      "step": 8795
    },
    {
      "entropy": 6.117723274230957,
      "epoch": 0.8796921077622832,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004931485886144972,
      "loss": 6.0691,
      "mean_token_accuracy": 0.15074310451745987,
      "num_tokens": 17225125.0,
      "step": 8800
    },
    {
      "entropy": 6.193218088150024,
      "epoch": 0.8801919328235118,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004931398457419959,
      "loss": 6.0878,
      "mean_token_accuracy": 0.14968049824237822,
      "num_tokens": 17234674.0,
      "step": 8805
    },
    {
      "entropy": 6.212806129455567,
      "epoch": 0.8806917578847403,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004931310973811521,
      "loss": 6.0378,
      "mean_token_accuracy": 0.1504512310028076,
      "num_tokens": 17244362.0,
      "step": 8810
    },
    {
      "entropy": 6.22478346824646,
      "epoch": 0.8811915829459689,
      "grad_norm": 1.296875,
      "learning_rate": 0.000493122343532186,
      "loss": 6.0024,
      "mean_token_accuracy": 0.15428674295544625,
      "num_tokens": 17253541.0,
      "step": 8815
    },
    {
      "entropy": 6.1872923374176025,
      "epoch": 0.8816914080071975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000493113584195318,
      "loss": 6.0318,
      "mean_token_accuracy": 0.15427838265895844,
      "num_tokens": 17263214.0,
      "step": 8820
    },
    {
      "entropy": 6.200919437408447,
      "epoch": 0.8821912330684261,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004931048193707681,
      "loss": 6.1164,
      "mean_token_accuracy": 0.14537843316793442,
      "num_tokens": 17273991.0,
      "step": 8825
    },
    {
      "entropy": 6.140426397323608,
      "epoch": 0.8826910581296546,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004930960490587573,
      "loss": 5.9532,
      "mean_token_accuracy": 0.15931536257266998,
      "num_tokens": 17283221.0,
      "step": 8830
    },
    {
      "entropy": 6.155590343475342,
      "epoch": 0.8831908831908832,
      "grad_norm": 1.1875,
      "learning_rate": 0.000493087273259506,
      "loss": 6.0832,
      "mean_token_accuracy": 0.15291211307048796,
      "num_tokens": 17293520.0,
      "step": 8835
    },
    {
      "entropy": 6.182627534866333,
      "epoch": 0.8836907082521117,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004930784919732349,
      "loss": 6.093,
      "mean_token_accuracy": 0.15025223344564437,
      "num_tokens": 17303881.0,
      "step": 8840
    },
    {
      "entropy": 6.1930054187774655,
      "epoch": 0.8841905333133403,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004930697052001651,
      "loss": 5.9742,
      "mean_token_accuracy": 0.16299348175525666,
      "num_tokens": 17314115.0,
      "step": 8845
    },
    {
      "entropy": 6.261048126220703,
      "epoch": 0.8846903583745689,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004930609129405177,
      "loss": 6.1412,
      "mean_token_accuracy": 0.14706263393163682,
      "num_tokens": 17324517.0,
      "step": 8850
    },
    {
      "entropy": 6.235148668289185,
      "epoch": 0.8851901834357975,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004930521151945138,
      "loss": 6.1136,
      "mean_token_accuracy": 0.1499358169734478,
      "num_tokens": 17334109.0,
      "step": 8855
    },
    {
      "entropy": 6.197334241867066,
      "epoch": 0.8856900084970261,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004930433119623749,
      "loss": 6.1104,
      "mean_token_accuracy": 0.14843170195817948,
      "num_tokens": 17344602.0,
      "step": 8860
    },
    {
      "entropy": 6.209990310668945,
      "epoch": 0.8861898335582546,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004930345032443221,
      "loss": 6.0373,
      "mean_token_accuracy": 0.16051452308893205,
      "num_tokens": 17354476.0,
      "step": 8865
    },
    {
      "entropy": 6.213006114959716,
      "epoch": 0.8866896586194832,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004930256890405777,
      "loss": 6.0252,
      "mean_token_accuracy": 0.15292552709579468,
      "num_tokens": 17364748.0,
      "step": 8870
    },
    {
      "entropy": 6.180851888656616,
      "epoch": 0.8871894836807117,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004930168693513627,
      "loss": 6.0648,
      "mean_token_accuracy": 0.15097014904022216,
      "num_tokens": 17374162.0,
      "step": 8875
    },
    {
      "entropy": 6.188289833068848,
      "epoch": 0.8876893087419403,
      "grad_norm": 1.25,
      "learning_rate": 0.0004930080441768995,
      "loss": 6.1156,
      "mean_token_accuracy": 0.1504686877131462,
      "num_tokens": 17384042.0,
      "step": 8880
    },
    {
      "entropy": 6.197012996673584,
      "epoch": 0.8881891338031689,
      "grad_norm": 1.203125,
      "learning_rate": 0.00049299921351741,
      "loss": 6.0069,
      "mean_token_accuracy": 0.1541579082608223,
      "num_tokens": 17394181.0,
      "step": 8885
    },
    {
      "entropy": 6.24964747428894,
      "epoch": 0.8886889588643975,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004929903773731164,
      "loss": 6.1932,
      "mean_token_accuracy": 0.13834179937839508,
      "num_tokens": 17402983.0,
      "step": 8890
    },
    {
      "entropy": 6.341733169555664,
      "epoch": 0.8891887839256261,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004929815357442409,
      "loss": 6.2401,
      "mean_token_accuracy": 0.1434593215584755,
      "num_tokens": 17413442.0,
      "step": 8895
    },
    {
      "entropy": 6.276982593536377,
      "epoch": 0.8896886089868546,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004929726886310062,
      "loss": 6.0578,
      "mean_token_accuracy": 0.1483489379286766,
      "num_tokens": 17423644.0,
      "step": 8900
    },
    {
      "entropy": 6.243321275711059,
      "epoch": 0.8901884340480831,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004929638360336348,
      "loss": 6.1098,
      "mean_token_accuracy": 0.1475658483803272,
      "num_tokens": 17433400.0,
      "step": 8905
    },
    {
      "entropy": 6.223323965072632,
      "epoch": 0.8906882591093117,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004929549779523492,
      "loss": 6.0763,
      "mean_token_accuracy": 0.1483346626162529,
      "num_tokens": 17443146.0,
      "step": 8910
    },
    {
      "entropy": 6.1547651290893555,
      "epoch": 0.8911880841705403,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004929461143873725,
      "loss": 5.9661,
      "mean_token_accuracy": 0.16434328854084015,
      "num_tokens": 17452880.0,
      "step": 8915
    },
    {
      "entropy": 6.137537097930908,
      "epoch": 0.8916879092317689,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004929372453389277,
      "loss": 5.9857,
      "mean_token_accuracy": 0.15564365088939666,
      "num_tokens": 17461633.0,
      "step": 8920
    },
    {
      "entropy": 6.253197717666626,
      "epoch": 0.8921877342929975,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004929283708072378,
      "loss": 6.1826,
      "mean_token_accuracy": 0.14608959779143332,
      "num_tokens": 17470892.0,
      "step": 8925
    },
    {
      "entropy": 6.243263912200928,
      "epoch": 0.8926875593542261,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004929194907925262,
      "loss": 6.0846,
      "mean_token_accuracy": 0.14708081781864166,
      "num_tokens": 17481457.0,
      "step": 8930
    },
    {
      "entropy": 6.229396486282349,
      "epoch": 0.8931873844154546,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004929106052950163,
      "loss": 6.109,
      "mean_token_accuracy": 0.14611636251211166,
      "num_tokens": 17490387.0,
      "step": 8935
    },
    {
      "entropy": 6.274730634689331,
      "epoch": 0.8936872094766831,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004929017143149317,
      "loss": 6.0762,
      "mean_token_accuracy": 0.15361128598451615,
      "num_tokens": 17499478.0,
      "step": 8940
    },
    {
      "entropy": 6.280763816833496,
      "epoch": 0.8941870345379117,
      "grad_norm": 1.46875,
      "learning_rate": 0.000492892817852496,
      "loss": 6.1843,
      "mean_token_accuracy": 0.142434324324131,
      "num_tokens": 17509396.0,
      "step": 8945
    },
    {
      "entropy": 6.163690662384033,
      "epoch": 0.8946868595991403,
      "grad_norm": 1.25,
      "learning_rate": 0.000492883915907933,
      "loss": 6.1008,
      "mean_token_accuracy": 0.15106959640979767,
      "num_tokens": 17519773.0,
      "step": 8950
    },
    {
      "entropy": 6.234408617019653,
      "epoch": 0.8951866846603689,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004928750084814668,
      "loss": 6.0931,
      "mean_token_accuracy": 0.15659594982862474,
      "num_tokens": 17529873.0,
      "step": 8955
    },
    {
      "entropy": 6.192315006256104,
      "epoch": 0.8956865097215975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004928660955733214,
      "loss": 6.0061,
      "mean_token_accuracy": 0.15732004940509797,
      "num_tokens": 17538642.0,
      "step": 8960
    },
    {
      "entropy": 6.217100954055786,
      "epoch": 0.896186334782826,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004928571771837211,
      "loss": 6.0537,
      "mean_token_accuracy": 0.14765094220638275,
      "num_tokens": 17546892.0,
      "step": 8965
    },
    {
      "entropy": 6.2676269054412845,
      "epoch": 0.8966861598440545,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004928482533128902,
      "loss": 6.1065,
      "mean_token_accuracy": 0.14703651741147042,
      "num_tokens": 17556283.0,
      "step": 8970
    },
    {
      "entropy": 6.176572465896607,
      "epoch": 0.8971859849052831,
      "grad_norm": 1.25,
      "learning_rate": 0.0004928393239610535,
      "loss": 6.0372,
      "mean_token_accuracy": 0.14964791908860206,
      "num_tokens": 17567021.0,
      "step": 8975
    },
    {
      "entropy": 6.225878524780273,
      "epoch": 0.8976858099665117,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004928303891284354,
      "loss": 6.0566,
      "mean_token_accuracy": 0.15498910546302797,
      "num_tokens": 17575927.0,
      "step": 8980
    },
    {
      "entropy": 6.2229187965393065,
      "epoch": 0.8981856350277403,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004928214488152607,
      "loss": 6.0856,
      "mean_token_accuracy": 0.1536189168691635,
      "num_tokens": 17585882.0,
      "step": 8985
    },
    {
      "entropy": 6.150031852722168,
      "epoch": 0.8986854600889689,
      "grad_norm": 1.125,
      "learning_rate": 0.0004928125030217545,
      "loss": 6.0444,
      "mean_token_accuracy": 0.15631795972585677,
      "num_tokens": 17595998.0,
      "step": 8990
    },
    {
      "entropy": 6.286819410324097,
      "epoch": 0.8991852851501975,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004928035517481416,
      "loss": 6.0581,
      "mean_token_accuracy": 0.1565806120634079,
      "num_tokens": 17605177.0,
      "step": 8995
    },
    {
      "entropy": 6.254734373092651,
      "epoch": 0.899685110211426,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004927945949946476,
      "loss": 6.1879,
      "mean_token_accuracy": 0.14481331557035446,
      "num_tokens": 17615089.0,
      "step": 9000
    },
    {
      "epoch": 0.899685110211426,
      "eval_entropy": 6.0719523968017155,
      "eval_loss": 6.127414226531982,
      "eval_mean_token_accuracy": 0.15489319380333408,
      "eval_num_tokens": 17615089.0,
      "eval_runtime": 24.011,
      "eval_samples_per_second": 1292.948,
      "eval_steps_per_second": 161.634,
      "step": 9000
    },
    {
      "entropy": 6.231038665771484,
      "epoch": 0.9001849352726545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004927856327614975,
      "loss": 6.0595,
      "mean_token_accuracy": 0.1500016190111637,
      "num_tokens": 17624708.0,
      "step": 9005
    },
    {
      "entropy": 6.23131914138794,
      "epoch": 0.9006847603338831,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000492776665048917,
      "loss": 6.0795,
      "mean_token_accuracy": 0.15019645541906357,
      "num_tokens": 17633639.0,
      "step": 9010
    },
    {
      "entropy": 6.171057796478271,
      "epoch": 0.9011845853951117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004927676918571316,
      "loss": 6.1127,
      "mean_token_accuracy": 0.14683156684041024,
      "num_tokens": 17643092.0,
      "step": 9015
    },
    {
      "entropy": 6.234998607635498,
      "epoch": 0.9016844104563403,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004927587131863671,
      "loss": 6.1382,
      "mean_token_accuracy": 0.14209688752889632,
      "num_tokens": 17652698.0,
      "step": 9020
    },
    {
      "entropy": 6.1679009914398195,
      "epoch": 0.9021842355175689,
      "grad_norm": 1.375,
      "learning_rate": 0.0004927497290368493,
      "loss": 5.9917,
      "mean_token_accuracy": 0.1590147167444229,
      "num_tokens": 17662654.0,
      "step": 9025
    },
    {
      "entropy": 6.19018931388855,
      "epoch": 0.9026840605787975,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004927407394088046,
      "loss": 6.0394,
      "mean_token_accuracy": 0.14583585262298585,
      "num_tokens": 17672476.0,
      "step": 9030
    },
    {
      "entropy": 6.191551446914673,
      "epoch": 0.9031838856400259,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004927317443024587,
      "loss": 6.1126,
      "mean_token_accuracy": 0.14732194617390632,
      "num_tokens": 17682612.0,
      "step": 9035
    },
    {
      "entropy": 6.295206022262573,
      "epoch": 0.9036837107012545,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004927227437180382,
      "loss": 6.1013,
      "mean_token_accuracy": 0.1427030920982361,
      "num_tokens": 17692020.0,
      "step": 9040
    },
    {
      "entropy": 6.220331048965454,
      "epoch": 0.9041835357624831,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004927137376557696,
      "loss": 6.078,
      "mean_token_accuracy": 0.15332088023424148,
      "num_tokens": 17702677.0,
      "step": 9045
    },
    {
      "entropy": 6.129501914978027,
      "epoch": 0.9046833608237117,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004927047261158793,
      "loss": 5.9393,
      "mean_token_accuracy": 0.16101567447185516,
      "num_tokens": 17711753.0,
      "step": 9050
    },
    {
      "entropy": 6.200072240829468,
      "epoch": 0.9051831858849403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004926957090985942,
      "loss": 6.0728,
      "mean_token_accuracy": 0.1509827569127083,
      "num_tokens": 17721293.0,
      "step": 9055
    },
    {
      "entropy": 6.317616891860962,
      "epoch": 0.9056830109461689,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004926866866041409,
      "loss": 6.1385,
      "mean_token_accuracy": 0.14871339425444602,
      "num_tokens": 17732311.0,
      "step": 9060
    },
    {
      "entropy": 6.210763263702392,
      "epoch": 0.9061828360073975,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004926776586327467,
      "loss": 6.0481,
      "mean_token_accuracy": 0.15548539906740189,
      "num_tokens": 17743040.0,
      "step": 9065
    },
    {
      "entropy": 6.165962743759155,
      "epoch": 0.9066826610686259,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004926686251846386,
      "loss": 5.9899,
      "mean_token_accuracy": 0.1647803232073784,
      "num_tokens": 17752436.0,
      "step": 9070
    },
    {
      "entropy": 6.212686586380005,
      "epoch": 0.9071824861298545,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004926595862600439,
      "loss": 6.0518,
      "mean_token_accuracy": 0.14493615478277205,
      "num_tokens": 17762043.0,
      "step": 9075
    },
    {
      "entropy": 6.19841046333313,
      "epoch": 0.9076823111910831,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004926505418591901,
      "loss": 6.0771,
      "mean_token_accuracy": 0.152317875623703,
      "num_tokens": 17772065.0,
      "step": 9080
    },
    {
      "entropy": 6.1374993324279785,
      "epoch": 0.9081821362523117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004926414919823045,
      "loss": 5.9929,
      "mean_token_accuracy": 0.15644051134586334,
      "num_tokens": 17781472.0,
      "step": 9085
    },
    {
      "entropy": 6.230915832519531,
      "epoch": 0.9086819613135403,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004926324366296151,
      "loss": 6.0721,
      "mean_token_accuracy": 0.15163581520318986,
      "num_tokens": 17789435.0,
      "step": 9090
    },
    {
      "entropy": 6.24999885559082,
      "epoch": 0.9091817863747689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004926233758013494,
      "loss": 6.0505,
      "mean_token_accuracy": 0.15046188086271287,
      "num_tokens": 17798367.0,
      "step": 9095
    },
    {
      "entropy": 6.243617916107178,
      "epoch": 0.9096816114359974,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004926143094977359,
      "loss": 6.0569,
      "mean_token_accuracy": 0.15349369496107101,
      "num_tokens": 17807660.0,
      "step": 9100
    },
    {
      "entropy": 6.136161613464355,
      "epoch": 0.9101814364972259,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004926052377190022,
      "loss": 6.0738,
      "mean_token_accuracy": 0.15352212712168695,
      "num_tokens": 17816993.0,
      "step": 9105
    },
    {
      "entropy": 6.2376378059387205,
      "epoch": 0.9106812615584545,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004925961604653765,
      "loss": 6.0444,
      "mean_token_accuracy": 0.1527748979628086,
      "num_tokens": 17826570.0,
      "step": 9110
    },
    {
      "entropy": 6.2975146770477295,
      "epoch": 0.9111810866196831,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004925870777370877,
      "loss": 6.0496,
      "mean_token_accuracy": 0.15350356101989746,
      "num_tokens": 17835921.0,
      "step": 9115
    },
    {
      "entropy": 6.19925742149353,
      "epoch": 0.9116809116809117,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004925779895343639,
      "loss": 6.0868,
      "mean_token_accuracy": 0.15331922620534896,
      "num_tokens": 17845713.0,
      "step": 9120
    },
    {
      "entropy": 6.231760787963867,
      "epoch": 0.9121807367421403,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004925688958574339,
      "loss": 6.0853,
      "mean_token_accuracy": 0.15133204609155654,
      "num_tokens": 17855808.0,
      "step": 9125
    },
    {
      "entropy": 6.255069971084595,
      "epoch": 0.9126805618033689,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004925597967065264,
      "loss": 6.0691,
      "mean_token_accuracy": 0.14928917735815048,
      "num_tokens": 17865427.0,
      "step": 9130
    },
    {
      "entropy": 6.227627325057983,
      "epoch": 0.9131803868645973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004925506920818705,
      "loss": 6.1327,
      "mean_token_accuracy": 0.14354583397507667,
      "num_tokens": 17876371.0,
      "step": 9135
    },
    {
      "entropy": 6.205776786804199,
      "epoch": 0.9136802119258259,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004925415819836951,
      "loss": 6.0232,
      "mean_token_accuracy": 0.1528637856245041,
      "num_tokens": 17886050.0,
      "step": 9140
    },
    {
      "entropy": 6.261341524124146,
      "epoch": 0.9141800369870545,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004925324664122294,
      "loss": 6.0967,
      "mean_token_accuracy": 0.15840699076652526,
      "num_tokens": 17896509.0,
      "step": 9145
    },
    {
      "entropy": 6.290964269638062,
      "epoch": 0.9146798620482831,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004925233453677029,
      "loss": 6.1281,
      "mean_token_accuracy": 0.14990468248724936,
      "num_tokens": 17906242.0,
      "step": 9150
    },
    {
      "entropy": 6.151191854476929,
      "epoch": 0.9151796871095117,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004925142188503451,
      "loss": 6.055,
      "mean_token_accuracy": 0.15152156352996826,
      "num_tokens": 17916261.0,
      "step": 9155
    },
    {
      "entropy": 6.201545238494873,
      "epoch": 0.9156795121707403,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004925050868603854,
      "loss": 6.0595,
      "mean_token_accuracy": 0.15109204202890397,
      "num_tokens": 17926486.0,
      "step": 9160
    },
    {
      "entropy": 6.218569374084472,
      "epoch": 0.9161793372319688,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004924959493980537,
      "loss": 6.0829,
      "mean_token_accuracy": 0.1467864230275154,
      "num_tokens": 17936374.0,
      "step": 9165
    },
    {
      "entropy": 6.30111255645752,
      "epoch": 0.9166791622931973,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004924868064635798,
      "loss": 6.3435,
      "mean_token_accuracy": 0.13521681651473044,
      "num_tokens": 17946930.0,
      "step": 9170
    },
    {
      "entropy": 6.36268105506897,
      "epoch": 0.9171789873544259,
      "grad_norm": 1.375,
      "learning_rate": 0.000492477658057194,
      "loss": 6.0954,
      "mean_token_accuracy": 0.1502355843782425,
      "num_tokens": 17955760.0,
      "step": 9175
    },
    {
      "entropy": 6.191945123672485,
      "epoch": 0.9176788124156545,
      "grad_norm": 1.125,
      "learning_rate": 0.0004924685041791262,
      "loss": 6.0265,
      "mean_token_accuracy": 0.15668040066957473,
      "num_tokens": 17965871.0,
      "step": 9180
    },
    {
      "entropy": 6.046727180480957,
      "epoch": 0.9181786374768831,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004924593448296068,
      "loss": 5.9758,
      "mean_token_accuracy": 0.15996017903089524,
      "num_tokens": 17974551.0,
      "step": 9185
    },
    {
      "entropy": 6.2014867782592775,
      "epoch": 0.9186784625381117,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004924501800088662,
      "loss": 6.0994,
      "mean_token_accuracy": 0.15105812400579452,
      "num_tokens": 17983675.0,
      "step": 9190
    },
    {
      "entropy": 6.242628669738769,
      "epoch": 0.9191782875993403,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004924410097171351,
      "loss": 6.0527,
      "mean_token_accuracy": 0.15568673461675644,
      "num_tokens": 17993299.0,
      "step": 9195
    },
    {
      "entropy": 6.237846422195434,
      "epoch": 0.9196781126605688,
      "grad_norm": 1.796875,
      "learning_rate": 0.000492431833954644,
      "loss": 6.0935,
      "mean_token_accuracy": 0.15041086748242377,
      "num_tokens": 18004755.0,
      "step": 9200
    },
    {
      "entropy": 6.260499095916748,
      "epoch": 0.9201779377217973,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004924226527216241,
      "loss": 6.1725,
      "mean_token_accuracy": 0.14105364829301834,
      "num_tokens": 18014908.0,
      "step": 9205
    },
    {
      "entropy": 6.222171878814697,
      "epoch": 0.9206777627830259,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004924134660183061,
      "loss": 6.0401,
      "mean_token_accuracy": 0.15047877728939058,
      "num_tokens": 18023919.0,
      "step": 9210
    },
    {
      "entropy": 6.178057622909546,
      "epoch": 0.9211775878442545,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004924042738449212,
      "loss": 5.9954,
      "mean_token_accuracy": 0.15466747879981996,
      "num_tokens": 18033706.0,
      "step": 9215
    },
    {
      "entropy": 6.255136966705322,
      "epoch": 0.9216774129054831,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004923950762017008,
      "loss": 6.1117,
      "mean_token_accuracy": 0.14287826269865037,
      "num_tokens": 18043171.0,
      "step": 9220
    },
    {
      "entropy": 6.222732400894165,
      "epoch": 0.9221772379667117,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004923858730888762,
      "loss": 6.0271,
      "mean_token_accuracy": 0.16167323291301727,
      "num_tokens": 18052567.0,
      "step": 9225
    },
    {
      "entropy": 6.214358806610107,
      "epoch": 0.9226770630279402,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000492376664506679,
      "loss": 6.0411,
      "mean_token_accuracy": 0.15918398052453994,
      "num_tokens": 18061648.0,
      "step": 9230
    },
    {
      "entropy": 6.22084674835205,
      "epoch": 0.9231768880891688,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004923674504553408,
      "loss": 6.0858,
      "mean_token_accuracy": 0.15076340436935426,
      "num_tokens": 18071079.0,
      "step": 9235
    },
    {
      "entropy": 6.187896585464477,
      "epoch": 0.9236767131503973,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004923582309350935,
      "loss": 6.1202,
      "mean_token_accuracy": 0.14868925660848617,
      "num_tokens": 18081162.0,
      "step": 9240
    },
    {
      "entropy": 6.260784482955932,
      "epoch": 0.9241765382116259,
      "grad_norm": 1.3125,
      "learning_rate": 0.000492349005946169,
      "loss": 6.0691,
      "mean_token_accuracy": 0.15349994897842406,
      "num_tokens": 18090072.0,
      "step": 9245
    },
    {
      "entropy": 6.306547069549561,
      "epoch": 0.9246763632728545,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004923397754887994,
      "loss": 6.1133,
      "mean_token_accuracy": 0.1476108178496361,
      "num_tokens": 18100565.0,
      "step": 9250
    },
    {
      "entropy": 6.233260679244995,
      "epoch": 0.9251761883340831,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004923305395632172,
      "loss": 6.065,
      "mean_token_accuracy": 0.1483036145567894,
      "num_tokens": 18109795.0,
      "step": 9255
    },
    {
      "entropy": 6.257886219024658,
      "epoch": 0.9256760133953117,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004923212981696543,
      "loss": 6.0993,
      "mean_token_accuracy": 0.1438927434384823,
      "num_tokens": 18118821.0,
      "step": 9260
    },
    {
      "entropy": 6.259016990661621,
      "epoch": 0.9261758384565402,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004923120513083436,
      "loss": 6.1098,
      "mean_token_accuracy": 0.15111489444971085,
      "num_tokens": 18128709.0,
      "step": 9265
    },
    {
      "entropy": 6.209259796142578,
      "epoch": 0.9266756635177688,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004923027989795175,
      "loss": 6.1537,
      "mean_token_accuracy": 0.14629934430122377,
      "num_tokens": 18138134.0,
      "step": 9270
    },
    {
      "entropy": 6.205793476104736,
      "epoch": 0.9271754885789973,
      "grad_norm": 1.375,
      "learning_rate": 0.000492293541183409,
      "loss": 6.0004,
      "mean_token_accuracy": 0.15127084851264955,
      "num_tokens": 18147455.0,
      "step": 9275
    },
    {
      "entropy": 6.208625650405883,
      "epoch": 0.9276753136402259,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004922842779202509,
      "loss": 6.0728,
      "mean_token_accuracy": 0.15632907748222352,
      "num_tokens": 18157787.0,
      "step": 9280
    },
    {
      "entropy": 6.129068899154663,
      "epoch": 0.9281751387014545,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004922750091902763,
      "loss": 6.0395,
      "mean_token_accuracy": 0.15793268382549286,
      "num_tokens": 18167607.0,
      "step": 9285
    },
    {
      "entropy": 6.214073991775512,
      "epoch": 0.9286749637626831,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004922657349937183,
      "loss": 6.1022,
      "mean_token_accuracy": 0.14468628019094468,
      "num_tokens": 18178927.0,
      "step": 9290
    },
    {
      "entropy": 6.1747565269470215,
      "epoch": 0.9291747888239116,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004922564553308103,
      "loss": 5.9943,
      "mean_token_accuracy": 0.15306239128112792,
      "num_tokens": 18188907.0,
      "step": 9295
    },
    {
      "entropy": 6.130193758010864,
      "epoch": 0.9296746138851402,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000492247170201786,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15720709413290024,
      "num_tokens": 18198937.0,
      "step": 9300
    },
    {
      "entropy": 6.164691591262818,
      "epoch": 0.9301744389463688,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004922378796068785,
      "loss": 6.0155,
      "mean_token_accuracy": 0.15191463530063629,
      "num_tokens": 18209306.0,
      "step": 9305
    },
    {
      "entropy": 6.318961191177368,
      "epoch": 0.9306742640075973,
      "grad_norm": 1.34375,
      "learning_rate": 0.000492228583546322,
      "loss": 6.1847,
      "mean_token_accuracy": 0.13658268228173256,
      "num_tokens": 18219665.0,
      "step": 9310
    },
    {
      "entropy": 6.270161151885986,
      "epoch": 0.9311740890688259,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004922192820203502,
      "loss": 6.0504,
      "mean_token_accuracy": 0.1484959080815315,
      "num_tokens": 18229044.0,
      "step": 9315
    },
    {
      "entropy": 6.174050140380859,
      "epoch": 0.9316739141300545,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004922099750291972,
      "loss": 6.0207,
      "mean_token_accuracy": 0.15656452178955077,
      "num_tokens": 18239373.0,
      "step": 9320
    },
    {
      "entropy": 6.198981428146363,
      "epoch": 0.9321737391912831,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004922006625730971,
      "loss": 6.0229,
      "mean_token_accuracy": 0.15291590392589569,
      "num_tokens": 18249195.0,
      "step": 9325
    },
    {
      "entropy": 6.223763084411621,
      "epoch": 0.9326735642525116,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004921913446522843,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15656565874814987,
      "num_tokens": 18258399.0,
      "step": 9330
    },
    {
      "entropy": 6.18964467048645,
      "epoch": 0.9331733893137402,
      "grad_norm": 1.9296875,
      "learning_rate": 0.000492182021266993,
      "loss": 6.0594,
      "mean_token_accuracy": 0.15399105101823807,
      "num_tokens": 18268952.0,
      "step": 9335
    },
    {
      "entropy": 6.212173175811768,
      "epoch": 0.9336732143749688,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000492172692417458,
      "loss": 6.0199,
      "mean_token_accuracy": 0.15111202225089074,
      "num_tokens": 18279482.0,
      "step": 9340
    },
    {
      "entropy": 6.187543964385986,
      "epoch": 0.9341730394361973,
      "grad_norm": 1.359375,
      "learning_rate": 0.000492163358103914,
      "loss": 6.0607,
      "mean_token_accuracy": 0.15307216495275497,
      "num_tokens": 18289069.0,
      "step": 9345
    },
    {
      "entropy": 6.184841823577881,
      "epoch": 0.9346728644974259,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004921540183265957,
      "loss": 6.01,
      "mean_token_accuracy": 0.15416917502880095,
      "num_tokens": 18298333.0,
      "step": 9350
    },
    {
      "entropy": 6.219984722137451,
      "epoch": 0.9351726895586545,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004921446730857383,
      "loss": 6.0834,
      "mean_token_accuracy": 0.14918960109353066,
      "num_tokens": 18307661.0,
      "step": 9355
    },
    {
      "entropy": 6.251865720748901,
      "epoch": 0.935672514619883,
      "grad_norm": 1.25,
      "learning_rate": 0.0004921353223815767,
      "loss": 6.1416,
      "mean_token_accuracy": 0.14653849452733994,
      "num_tokens": 18317786.0,
      "step": 9360
    },
    {
      "entropy": 6.235687494277954,
      "epoch": 0.9361723396811116,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004921259662143463,
      "loss": 6.1317,
      "mean_token_accuracy": 0.14261429905891418,
      "num_tokens": 18328430.0,
      "step": 9365
    },
    {
      "entropy": 6.254350471496582,
      "epoch": 0.9366721647423402,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004921166045842825,
      "loss": 6.1347,
      "mean_token_accuracy": 0.14653569906949998,
      "num_tokens": 18339800.0,
      "step": 9370
    },
    {
      "entropy": 6.33001937866211,
      "epoch": 0.9371719898035688,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004921072374916208,
      "loss": 6.1431,
      "mean_token_accuracy": 0.14112329632043838,
      "num_tokens": 18348819.0,
      "step": 9375
    },
    {
      "entropy": 6.186894130706787,
      "epoch": 0.9376718148647973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004920978649365968,
      "loss": 5.9902,
      "mean_token_accuracy": 0.1623859539628029,
      "num_tokens": 18358189.0,
      "step": 9380
    },
    {
      "entropy": 6.186570739746093,
      "epoch": 0.9381716399260259,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004920884869194463,
      "loss": 5.9914,
      "mean_token_accuracy": 0.15406157970428466,
      "num_tokens": 18367571.0,
      "step": 9385
    },
    {
      "entropy": 6.197534465789795,
      "epoch": 0.9386714649872545,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004920791034404054,
      "loss": 5.9892,
      "mean_token_accuracy": 0.1555450141429901,
      "num_tokens": 18376699.0,
      "step": 9390
    },
    {
      "entropy": 6.143626689910889,
      "epoch": 0.939171290048483,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004920697144997102,
      "loss": 5.8852,
      "mean_token_accuracy": 0.16139229834079744,
      "num_tokens": 18386036.0,
      "step": 9395
    },
    {
      "entropy": 6.191374158859253,
      "epoch": 0.9396711151097116,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004920603200975968,
      "loss": 6.0884,
      "mean_token_accuracy": 0.16238868981599808,
      "num_tokens": 18395363.0,
      "step": 9400
    },
    {
      "entropy": 6.24739785194397,
      "epoch": 0.9401709401709402,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004920509202343016,
      "loss": 6.0688,
      "mean_token_accuracy": 0.14848162084817887,
      "num_tokens": 18404919.0,
      "step": 9405
    },
    {
      "entropy": 6.218137788772583,
      "epoch": 0.9406707652321687,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000492041514910061,
      "loss": 6.0887,
      "mean_token_accuracy": 0.1517230212688446,
      "num_tokens": 18413520.0,
      "step": 9410
    },
    {
      "entropy": 6.22852840423584,
      "epoch": 0.9411705902933973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004920321041251118,
      "loss": 6.1081,
      "mean_token_accuracy": 0.15238470435142518,
      "num_tokens": 18422942.0,
      "step": 9415
    },
    {
      "entropy": 6.307512998580933,
      "epoch": 0.9416704153546259,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004920226878796908,
      "loss": 6.1128,
      "mean_token_accuracy": 0.14426910802721976,
      "num_tokens": 18432782.0,
      "step": 9420
    },
    {
      "entropy": 6.270034885406494,
      "epoch": 0.9421702404158544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004920132661740348,
      "loss": 6.1013,
      "mean_token_accuracy": 0.15191699415445328,
      "num_tokens": 18443058.0,
      "step": 9425
    },
    {
      "entropy": 6.2411493301391605,
      "epoch": 0.942670065477083,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004920038390083808,
      "loss": 5.9896,
      "mean_token_accuracy": 0.15453899800777435,
      "num_tokens": 18453138.0,
      "step": 9430
    },
    {
      "entropy": 6.172799110412598,
      "epoch": 0.9431698905383116,
      "grad_norm": 1.34375,
      "learning_rate": 0.000491994406382966,
      "loss": 6.1322,
      "mean_token_accuracy": 0.15028438121080398,
      "num_tokens": 18463763.0,
      "step": 9435
    },
    {
      "entropy": 6.2097571849822994,
      "epoch": 0.9436697155995402,
      "grad_norm": 1.375,
      "learning_rate": 0.0004919849682980278,
      "loss": 6.0345,
      "mean_token_accuracy": 0.15386351495981215,
      "num_tokens": 18473391.0,
      "step": 9440
    },
    {
      "entropy": 6.220196533203125,
      "epoch": 0.9441695406607687,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004919755247538037,
      "loss": 6.0639,
      "mean_token_accuracy": 0.1476370945572853,
      "num_tokens": 18483002.0,
      "step": 9445
    },
    {
      "entropy": 6.279415893554687,
      "epoch": 0.9446693657219973,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004919660757505312,
      "loss": 6.0655,
      "mean_token_accuracy": 0.15239429101347923,
      "num_tokens": 18493428.0,
      "step": 9450
    },
    {
      "entropy": 6.277664709091186,
      "epoch": 0.9451691907832259,
      "grad_norm": 1.7265625,
      "learning_rate": 0.0004919566212884481,
      "loss": 6.0481,
      "mean_token_accuracy": 0.1487953968346119,
      "num_tokens": 18502756.0,
      "step": 9455
    },
    {
      "entropy": 6.22528600692749,
      "epoch": 0.9456690158444544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004919471613677921,
      "loss": 6.1189,
      "mean_token_accuracy": 0.14318000227212907,
      "num_tokens": 18512576.0,
      "step": 9460
    },
    {
      "entropy": 6.190670919418335,
      "epoch": 0.946168840905683,
      "grad_norm": 1.5,
      "learning_rate": 0.0004919376959888015,
      "loss": 5.9705,
      "mean_token_accuracy": 0.15100648999214172,
      "num_tokens": 18522271.0,
      "step": 9465
    },
    {
      "entropy": 6.161213350296021,
      "epoch": 0.9466686659669116,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004919282251517142,
      "loss": 5.9892,
      "mean_token_accuracy": 0.15205149948596955,
      "num_tokens": 18532738.0,
      "step": 9470
    },
    {
      "entropy": 6.263359498977661,
      "epoch": 0.9471684910281402,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004919187488567686,
      "loss": 6.2077,
      "mean_token_accuracy": 0.15216834098100662,
      "num_tokens": 18543132.0,
      "step": 9475
    },
    {
      "entropy": 6.20970311164856,
      "epoch": 0.9476683160893687,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004919092671042031,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15883567780256272,
      "num_tokens": 18551777.0,
      "step": 9480
    },
    {
      "entropy": 6.12543568611145,
      "epoch": 0.9481681411505973,
      "grad_norm": 1.375,
      "learning_rate": 0.0004918997798942562,
      "loss": 6.0368,
      "mean_token_accuracy": 0.15105368942022324,
      "num_tokens": 18561137.0,
      "step": 9485
    },
    {
      "entropy": 6.173322629928589,
      "epoch": 0.9486679662118258,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004918902872271668,
      "loss": 6.0903,
      "mean_token_accuracy": 0.14883806481957434,
      "num_tokens": 18571917.0,
      "step": 9490
    },
    {
      "entropy": 6.2400469303131105,
      "epoch": 0.9491677912730544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004918807891031734,
      "loss": 6.0595,
      "mean_token_accuracy": 0.1494263380765915,
      "num_tokens": 18581670.0,
      "step": 9495
    },
    {
      "entropy": 6.134951543807984,
      "epoch": 0.949667616334283,
      "grad_norm": 1.25,
      "learning_rate": 0.0004918712855225154,
      "loss": 5.9995,
      "mean_token_accuracy": 0.15328548699617386,
      "num_tokens": 18590140.0,
      "step": 9500
    },
    {
      "entropy": 6.06361198425293,
      "epoch": 0.9501674413955116,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004918617764854314,
      "loss": 5.8939,
      "mean_token_accuracy": 0.15935043692588807,
      "num_tokens": 18599496.0,
      "step": 9505
    },
    {
      "entropy": 6.1570947647094725,
      "epoch": 0.9506672664567402,
      "grad_norm": 1.265625,
      "learning_rate": 0.000491852261992161,
      "loss": 6.0564,
      "mean_token_accuracy": 0.15687580555677413,
      "num_tokens": 18609415.0,
      "step": 9510
    },
    {
      "entropy": 6.242318725585937,
      "epoch": 0.9511670915179687,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004918427420429435,
      "loss": 6.0616,
      "mean_token_accuracy": 0.15219078809022904,
      "num_tokens": 18619215.0,
      "step": 9515
    },
    {
      "entropy": 6.300957918167114,
      "epoch": 0.9516669165791973,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004918332166380185,
      "loss": 6.138,
      "mean_token_accuracy": 0.1504904732108116,
      "num_tokens": 18629381.0,
      "step": 9520
    },
    {
      "entropy": 6.209245109558106,
      "epoch": 0.9521667416404258,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004918236857776254,
      "loss": 6.0784,
      "mean_token_accuracy": 0.15259324312210082,
      "num_tokens": 18639049.0,
      "step": 9525
    },
    {
      "entropy": 6.220205688476563,
      "epoch": 0.9526665667016544,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004918141494620042,
      "loss": 6.0904,
      "mean_token_accuracy": 0.14841213822364807,
      "num_tokens": 18649096.0,
      "step": 9530
    },
    {
      "entropy": 6.207163238525391,
      "epoch": 0.953166391762883,
      "grad_norm": 1.25,
      "learning_rate": 0.0004918046076913949,
      "loss": 6.0389,
      "mean_token_accuracy": 0.15804794430732727,
      "num_tokens": 18658747.0,
      "step": 9535
    },
    {
      "entropy": 6.269297075271607,
      "epoch": 0.9536662168241116,
      "grad_norm": 1.375,
      "learning_rate": 0.0004917950604660374,
      "loss": 6.1209,
      "mean_token_accuracy": 0.1494915597140789,
      "num_tokens": 18668538.0,
      "step": 9540
    },
    {
      "entropy": 6.236788749694824,
      "epoch": 0.9541660418853402,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000491785507786172,
      "loss": 6.0802,
      "mean_token_accuracy": 0.1503964349627495,
      "num_tokens": 18679241.0,
      "step": 9545
    },
    {
      "entropy": 6.182567882537842,
      "epoch": 0.9546658669465687,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004917759496520388,
      "loss": 6.0761,
      "mean_token_accuracy": 0.14858175814151764,
      "num_tokens": 18689405.0,
      "step": 9550
    },
    {
      "entropy": 6.2947179794311525,
      "epoch": 0.9551656920077972,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004917663860638787,
      "loss": 6.1246,
      "mean_token_accuracy": 0.14366439804434777,
      "num_tokens": 18697989.0,
      "step": 9555
    },
    {
      "entropy": 6.196341228485108,
      "epoch": 0.9556655170690258,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004917568170219321,
      "loss": 6.0231,
      "mean_token_accuracy": 0.15023713409900666,
      "num_tokens": 18706992.0,
      "step": 9560
    },
    {
      "entropy": 6.121190166473388,
      "epoch": 0.9561653421302544,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004917472425264397,
      "loss": 6.0769,
      "mean_token_accuracy": 0.14825137630105018,
      "num_tokens": 18716742.0,
      "step": 9565
    },
    {
      "entropy": 6.245248031616211,
      "epoch": 0.956665167191483,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004917376625776425,
      "loss": 6.1791,
      "mean_token_accuracy": 0.14539650827646255,
      "num_tokens": 18725955.0,
      "step": 9570
    },
    {
      "entropy": 6.290267753601074,
      "epoch": 0.9571649922527116,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004917280771757814,
      "loss": 6.0837,
      "mean_token_accuracy": 0.15159521251916885,
      "num_tokens": 18735199.0,
      "step": 9575
    },
    {
      "entropy": 6.254014158248902,
      "epoch": 0.9576648173139402,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004917184863210977,
      "loss": 6.113,
      "mean_token_accuracy": 0.14517153799533844,
      "num_tokens": 18745457.0,
      "step": 9580
    },
    {
      "entropy": 6.246565961837769,
      "epoch": 0.9581646423751687,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004917088900138326,
      "loss": 6.0962,
      "mean_token_accuracy": 0.14720577746629715,
      "num_tokens": 18753869.0,
      "step": 9585
    },
    {
      "entropy": 6.22707691192627,
      "epoch": 0.9586644674363972,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004916992882542276,
      "loss": 6.0687,
      "mean_token_accuracy": 0.1583326369524002,
      "num_tokens": 18763022.0,
      "step": 9590
    },
    {
      "entropy": 6.213897800445556,
      "epoch": 0.9591642924976258,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004916896810425244,
      "loss": 5.9927,
      "mean_token_accuracy": 0.16006032824516297,
      "num_tokens": 18772827.0,
      "step": 9595
    },
    {
      "entropy": 6.199007415771485,
      "epoch": 0.9596641175588544,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004916800683789645,
      "loss": 6.0228,
      "mean_token_accuracy": 0.15236202478408814,
      "num_tokens": 18782272.0,
      "step": 9600
    },
    {
      "entropy": 6.207870101928711,
      "epoch": 0.960163942620083,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004916704502637897,
      "loss": 6.0557,
      "mean_token_accuracy": 0.1502628892660141,
      "num_tokens": 18792125.0,
      "step": 9605
    },
    {
      "entropy": 6.167756366729736,
      "epoch": 0.9606637676813116,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004916608266972422,
      "loss": 6.0183,
      "mean_token_accuracy": 0.15662460178136825,
      "num_tokens": 18801498.0,
      "step": 9610
    },
    {
      "entropy": 6.232539939880371,
      "epoch": 0.9611635927425402,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004916511976795642,
      "loss": 6.1399,
      "mean_token_accuracy": 0.14556847885251045,
      "num_tokens": 18811904.0,
      "step": 9615
    },
    {
      "entropy": 6.228086853027344,
      "epoch": 0.9616634178037686,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004916415632109977,
      "loss": 6.0872,
      "mean_token_accuracy": 0.1518653929233551,
      "num_tokens": 18821388.0,
      "step": 9620
    },
    {
      "entropy": 6.276173448562622,
      "epoch": 0.9621632428649972,
      "grad_norm": 1.5,
      "learning_rate": 0.0004916319232917852,
      "loss": 6.1232,
      "mean_token_accuracy": 0.15319905579090118,
      "num_tokens": 18831392.0,
      "step": 9625
    },
    {
      "entropy": 6.218940687179566,
      "epoch": 0.9626630679262258,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004916222779221694,
      "loss": 6.0388,
      "mean_token_accuracy": 0.14967302829027176,
      "num_tokens": 18841211.0,
      "step": 9630
    },
    {
      "entropy": 6.274160099029541,
      "epoch": 0.9631628929874544,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004916126271023926,
      "loss": 6.0684,
      "mean_token_accuracy": 0.150314125418663,
      "num_tokens": 18851033.0,
      "step": 9635
    },
    {
      "entropy": 6.222961330413819,
      "epoch": 0.963662718048683,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000491602970832698,
      "loss": 6.1083,
      "mean_token_accuracy": 0.15259940624237062,
      "num_tokens": 18861439.0,
      "step": 9640
    },
    {
      "entropy": 6.2386480331420895,
      "epoch": 0.9641625431099116,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004915933091133284,
      "loss": 6.0958,
      "mean_token_accuracy": 0.1450348451733589,
      "num_tokens": 18872056.0,
      "step": 9645
    },
    {
      "entropy": 6.251325702667236,
      "epoch": 0.9646623681711401,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004915836419445268,
      "loss": 6.0046,
      "mean_token_accuracy": 0.15753925889730452,
      "num_tokens": 18881540.0,
      "step": 9650
    },
    {
      "entropy": 6.1729137897491455,
      "epoch": 0.9651621932323686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004915739693265365,
      "loss": 6.0299,
      "mean_token_accuracy": 0.1475259095430374,
      "num_tokens": 18890466.0,
      "step": 9655
    },
    {
      "entropy": 6.224126148223877,
      "epoch": 0.9656620182935972,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004915642912596009,
      "loss": 6.0909,
      "mean_token_accuracy": 0.1462048664689064,
      "num_tokens": 18901397.0,
      "step": 9660
    },
    {
      "entropy": 6.237620115280151,
      "epoch": 0.9661618433548258,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915546077439634,
      "loss": 6.0207,
      "mean_token_accuracy": 0.15306469202041625,
      "num_tokens": 18911357.0,
      "step": 9665
    },
    {
      "entropy": 6.0991614818572994,
      "epoch": 0.9666616684160544,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004915449187798677,
      "loss": 5.9653,
      "mean_token_accuracy": 0.16138627082109452,
      "num_tokens": 18920901.0,
      "step": 9670
    },
    {
      "entropy": 6.228220987319946,
      "epoch": 0.967161493477283,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004915352243675576,
      "loss": 6.0893,
      "mean_token_accuracy": 0.1472497321665287,
      "num_tokens": 18931012.0,
      "step": 9675
    },
    {
      "entropy": 6.243789434432983,
      "epoch": 0.9676613185385116,
      "grad_norm": 1.359375,
      "learning_rate": 0.000491525524507277,
      "loss": 6.146,
      "mean_token_accuracy": 0.14576415121555328,
      "num_tokens": 18941222.0,
      "step": 9680
    },
    {
      "entropy": 6.2972959041595455,
      "epoch": 0.96816114359974,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004915158191992699,
      "loss": 6.1301,
      "mean_token_accuracy": 0.14853508695960044,
      "num_tokens": 18951073.0,
      "step": 9685
    },
    {
      "entropy": 6.285033798217773,
      "epoch": 0.9686609686609686,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004915061084437805,
      "loss": 6.1261,
      "mean_token_accuracy": 0.146405328810215,
      "num_tokens": 18960996.0,
      "step": 9690
    },
    {
      "entropy": 6.212119197845459,
      "epoch": 0.9691607937221972,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004914963922410531,
      "loss": 6.1467,
      "mean_token_accuracy": 0.14198032170534133,
      "num_tokens": 18970451.0,
      "step": 9695
    },
    {
      "entropy": 6.143320941925049,
      "epoch": 0.9696606187834258,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004914866705913322,
      "loss": 6.02,
      "mean_token_accuracy": 0.14886799156665803,
      "num_tokens": 18979781.0,
      "step": 9700
    },
    {
      "entropy": 6.24599256515503,
      "epoch": 0.9701604438446544,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004914769434948623,
      "loss": 6.0806,
      "mean_token_accuracy": 0.15258603692054748,
      "num_tokens": 18989663.0,
      "step": 9705
    },
    {
      "entropy": 6.2854218006134035,
      "epoch": 0.970660268905883,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004914672109518883,
      "loss": 6.1984,
      "mean_token_accuracy": 0.14810571521520616,
      "num_tokens": 18999613.0,
      "step": 9710
    },
    {
      "entropy": 6.167517614364624,
      "epoch": 0.9711600939671116,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000491457472962655,
      "loss": 5.9968,
      "mean_token_accuracy": 0.15997470319271087,
      "num_tokens": 19008976.0,
      "step": 9715
    },
    {
      "entropy": 6.117216634750366,
      "epoch": 0.97165991902834,
      "grad_norm": 1.25,
      "learning_rate": 0.0004914477295274073,
      "loss": 6.0121,
      "mean_token_accuracy": 0.15582980513572692,
      "num_tokens": 19019013.0,
      "step": 9720
    },
    {
      "entropy": 6.274588346481323,
      "epoch": 0.9721597440895686,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004914379806463905,
      "loss": 6.1162,
      "mean_token_accuracy": 0.15064358934760094,
      "num_tokens": 19029252.0,
      "step": 9725
    },
    {
      "entropy": 6.259511709213257,
      "epoch": 0.9726595691507972,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004914282263198498,
      "loss": 6.0622,
      "mean_token_accuracy": 0.1498521536588669,
      "num_tokens": 19038563.0,
      "step": 9730
    },
    {
      "entropy": 6.177741575241089,
      "epoch": 0.9731593942120258,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004914184665480305,
      "loss": 6.0845,
      "mean_token_accuracy": 0.14762043505907058,
      "num_tokens": 19048568.0,
      "step": 9735
    },
    {
      "entropy": 6.141620254516601,
      "epoch": 0.9736592192732544,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004914087013311784,
      "loss": 5.8567,
      "mean_token_accuracy": 0.166812152415514,
      "num_tokens": 19057527.0,
      "step": 9740
    },
    {
      "entropy": 6.2267848491668705,
      "epoch": 0.974159044334483,
      "grad_norm": 1.25,
      "learning_rate": 0.0004913989306695389,
      "loss": 6.0143,
      "mean_token_accuracy": 0.15289444401860236,
      "num_tokens": 19067372.0,
      "step": 9745
    },
    {
      "entropy": 6.1202271461486815,
      "epoch": 0.9746588693957114,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004913891545633582,
      "loss": 6.0186,
      "mean_token_accuracy": 0.1590340331196785,
      "num_tokens": 19077521.0,
      "step": 9750
    },
    {
      "entropy": 6.189548635482788,
      "epoch": 0.97515869445694,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000491379373012882,
      "loss": 5.9832,
      "mean_token_accuracy": 0.15376351699233054,
      "num_tokens": 19086674.0,
      "step": 9755
    },
    {
      "entropy": 6.284615325927734,
      "epoch": 0.9756585195181686,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004913695860183563,
      "loss": 6.1874,
      "mean_token_accuracy": 0.1400657922029495,
      "num_tokens": 19097586.0,
      "step": 9760
    },
    {
      "entropy": 6.257413387298584,
      "epoch": 0.9761583445793972,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004913597935800276,
      "loss": 6.1936,
      "mean_token_accuracy": 0.14045121744275094,
      "num_tokens": 19107763.0,
      "step": 9765
    },
    {
      "entropy": 6.207984066009521,
      "epoch": 0.9766581696406258,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004913499956981422,
      "loss": 5.9862,
      "mean_token_accuracy": 0.16010566800832748,
      "num_tokens": 19117931.0,
      "step": 9770
    },
    {
      "entropy": 6.195990562438965,
      "epoch": 0.9771579947018544,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004913401923729465,
      "loss": 6.0577,
      "mean_token_accuracy": 0.1509033925831318,
      "num_tokens": 19127707.0,
      "step": 9775
    },
    {
      "entropy": 6.177084016799927,
      "epoch": 0.977657819763083,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004913303836046873,
      "loss": 5.9847,
      "mean_token_accuracy": 0.15120234787464143,
      "num_tokens": 19137633.0,
      "step": 9780
    },
    {
      "entropy": 6.242629718780518,
      "epoch": 0.9781576448243114,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004913205693936112,
      "loss": 6.1016,
      "mean_token_accuracy": 0.15185408145189286,
      "num_tokens": 19146247.0,
      "step": 9785
    },
    {
      "entropy": 6.125142765045166,
      "epoch": 0.97865746988554,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004913107497399653,
      "loss": 6.0063,
      "mean_token_accuracy": 0.16843177229166031,
      "num_tokens": 19156225.0,
      "step": 9790
    },
    {
      "entropy": 6.180765914916992,
      "epoch": 0.9791572949467686,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004913009246439966,
      "loss": 5.9738,
      "mean_token_accuracy": 0.1527572140097618,
      "num_tokens": 19165763.0,
      "step": 9795
    },
    {
      "entropy": 6.167466640472412,
      "epoch": 0.9796571200079972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912910941059525,
      "loss": 5.9964,
      "mean_token_accuracy": 0.1560740739107132,
      "num_tokens": 19174116.0,
      "step": 9800
    },
    {
      "entropy": 6.239066648483276,
      "epoch": 0.9801569450692258,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004912812581260799,
      "loss": 6.1829,
      "mean_token_accuracy": 0.14594337865710258,
      "num_tokens": 19184416.0,
      "step": 9805
    },
    {
      "entropy": 6.24222469329834,
      "epoch": 0.9806567701304544,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004912714167046265,
      "loss": 6.1351,
      "mean_token_accuracy": 0.14393481314182283,
      "num_tokens": 19193356.0,
      "step": 9810
    },
    {
      "entropy": 6.190090370178223,
      "epoch": 0.9811565951916829,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004912615698418399,
      "loss": 5.9466,
      "mean_token_accuracy": 0.1673352375626564,
      "num_tokens": 19204257.0,
      "step": 9815
    },
    {
      "entropy": 6.276273345947265,
      "epoch": 0.9816564202529114,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004912517175379679,
      "loss": 6.0751,
      "mean_token_accuracy": 0.15363384038209915,
      "num_tokens": 19213889.0,
      "step": 9820
    },
    {
      "entropy": 6.282799339294433,
      "epoch": 0.98215624531414,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004912418597932584,
      "loss": 6.1386,
      "mean_token_accuracy": 0.1493024557828903,
      "num_tokens": 19224008.0,
      "step": 9825
    },
    {
      "entropy": 6.279945421218872,
      "epoch": 0.9826560703753686,
      "grad_norm": 1.25,
      "learning_rate": 0.0004912319966079592,
      "loss": 6.1309,
      "mean_token_accuracy": 0.14636823534965515,
      "num_tokens": 19234247.0,
      "step": 9830
    },
    {
      "entropy": 6.324870014190674,
      "epoch": 0.9831558954365972,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004912221279823187,
      "loss": 6.159,
      "mean_token_accuracy": 0.1421165309846401,
      "num_tokens": 19244838.0,
      "step": 9835
    },
    {
      "entropy": 6.228650856018066,
      "epoch": 0.9836557204978258,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004912122539165851,
      "loss": 6.065,
      "mean_token_accuracy": 0.15061584413051604,
      "num_tokens": 19254989.0,
      "step": 9840
    },
    {
      "entropy": 6.194226551055908,
      "epoch": 0.9841555455590544,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004912023744110068,
      "loss": 6.1121,
      "mean_token_accuracy": 0.14618947580456734,
      "num_tokens": 19264519.0,
      "step": 9845
    },
    {
      "entropy": 6.2256019592285154,
      "epoch": 0.9846553706202829,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004911924894658324,
      "loss": 6.0729,
      "mean_token_accuracy": 0.15336726009845733,
      "num_tokens": 19274153.0,
      "step": 9850
    },
    {
      "entropy": 6.14973931312561,
      "epoch": 0.9851551956815114,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004911825990813107,
      "loss": 5.9945,
      "mean_token_accuracy": 0.1522758722305298,
      "num_tokens": 19283383.0,
      "step": 9855
    },
    {
      "entropy": 6.156365537643433,
      "epoch": 0.98565502074274,
      "grad_norm": 1.25,
      "learning_rate": 0.0004911727032576903,
      "loss": 5.9788,
      "mean_token_accuracy": 0.15503705590963363,
      "num_tokens": 19292758.0,
      "step": 9860
    },
    {
      "entropy": 6.197921466827393,
      "epoch": 0.9861548458039686,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004911628019952203,
      "loss": 6.0867,
      "mean_token_accuracy": 0.14506686329841614,
      "num_tokens": 19302339.0,
      "step": 9865
    },
    {
      "entropy": 6.172068691253662,
      "epoch": 0.9866546708651972,
      "grad_norm": 2.609375,
      "learning_rate": 0.0004911528952941499,
      "loss": 5.9458,
      "mean_token_accuracy": 0.15001685172319412,
      "num_tokens": 19312560.0,
      "step": 9870
    },
    {
      "entropy": 6.235568523406982,
      "epoch": 0.9871544959264258,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004911429831547283,
      "loss": 5.9889,
      "mean_token_accuracy": 0.1545190468430519,
      "num_tokens": 19322796.0,
      "step": 9875
    },
    {
      "entropy": 6.1740562438964846,
      "epoch": 0.9876543209876543,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004911330655772047,
      "loss": 6.0821,
      "mean_token_accuracy": 0.1500004954636097,
      "num_tokens": 19332719.0,
      "step": 9880
    },
    {
      "entropy": 6.169285869598388,
      "epoch": 0.9881541460488829,
      "grad_norm": 1.34375,
      "learning_rate": 0.000491123142561829,
      "loss": 6.0799,
      "mean_token_accuracy": 0.14822375178337097,
      "num_tokens": 19342226.0,
      "step": 9885
    },
    {
      "entropy": 6.127970027923584,
      "epoch": 0.9886539711101114,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004911132141088505,
      "loss": 6.074,
      "mean_token_accuracy": 0.14566567093133925,
      "num_tokens": 19351933.0,
      "step": 9890
    },
    {
      "entropy": 6.223727083206176,
      "epoch": 0.98915379617134,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004911032802185192,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15183516442775727,
      "num_tokens": 19359900.0,
      "step": 9895
    },
    {
      "entropy": 6.234736013412475,
      "epoch": 0.9896536212325686,
      "grad_norm": 1.28125,
      "learning_rate": 0.000491093340891085,
      "loss": 6.0762,
      "mean_token_accuracy": 0.14417059123516082,
      "num_tokens": 19369139.0,
      "step": 9900
    },
    {
      "entropy": 6.166970491409302,
      "epoch": 0.9901534462937972,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004910833961267979,
      "loss": 5.9997,
      "mean_token_accuracy": 0.14931093007326127,
      "num_tokens": 19378765.0,
      "step": 9905
    },
    {
      "entropy": 6.233784627914429,
      "epoch": 0.9906532713550258,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004910734459259081,
      "loss": 6.1026,
      "mean_token_accuracy": 0.14605162888765336,
      "num_tokens": 19389001.0,
      "step": 9910
    },
    {
      "entropy": 6.249694681167602,
      "epoch": 0.9911530964162543,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004910634902886662,
      "loss": 6.0221,
      "mean_token_accuracy": 0.1571980208158493,
      "num_tokens": 19397990.0,
      "step": 9915
    },
    {
      "entropy": 6.126401376724243,
      "epoch": 0.9916529214774829,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004910535292153224,
      "loss": 5.9721,
      "mean_token_accuracy": 0.15890015438199043,
      "num_tokens": 19408060.0,
      "step": 9920
    },
    {
      "entropy": 6.119165897369385,
      "epoch": 0.9921527465387114,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004910435627061273,
      "loss": 5.9842,
      "mean_token_accuracy": 0.1566011816263199,
      "num_tokens": 19417675.0,
      "step": 9925
    },
    {
      "entropy": 6.2198375225067135,
      "epoch": 0.99265257159994,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004910335907613318,
      "loss": 6.0072,
      "mean_token_accuracy": 0.15505679994821547,
      "num_tokens": 19426433.0,
      "step": 9930
    },
    {
      "entropy": 6.180503702163696,
      "epoch": 0.9931523966611686,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004910236133811867,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15755116045475007,
      "num_tokens": 19435459.0,
      "step": 9935
    },
    {
      "entropy": 6.173477220535278,
      "epoch": 0.9936522217223972,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004910136305659432,
      "loss": 6.0392,
      "mean_token_accuracy": 0.15312758535146714,
      "num_tokens": 19445329.0,
      "step": 9940
    },
    {
      "entropy": 6.181215286254883,
      "epoch": 0.9941520467836257,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004910036423158523,
      "loss": 5.989,
      "mean_token_accuracy": 0.15353483185172082,
      "num_tokens": 19454417.0,
      "step": 9945
    },
    {
      "entropy": 6.172397804260254,
      "epoch": 0.9946518718448543,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004909936486311653,
      "loss": 6.0775,
      "mean_token_accuracy": 0.14898218661546708,
      "num_tokens": 19464274.0,
      "step": 9950
    },
    {
      "entropy": 6.226012134552002,
      "epoch": 0.9951516969060828,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004909836495121336,
      "loss": 6.0546,
      "mean_token_accuracy": 0.15275216549634935,
      "num_tokens": 19472455.0,
      "step": 9955
    },
    {
      "entropy": 6.241779804229736,
      "epoch": 0.9956515219673114,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004909736449590091,
      "loss": 6.1489,
      "mean_token_accuracy": 0.14590268433094025,
      "num_tokens": 19483320.0,
      "step": 9960
    },
    {
      "entropy": 6.1918559074401855,
      "epoch": 0.99615134702854,
      "grad_norm": 1.296875,
      "learning_rate": 0.000490963634972043,
      "loss": 6.0144,
      "mean_token_accuracy": 0.150690495967865,
      "num_tokens": 19493644.0,
      "step": 9965
    },
    {
      "entropy": 6.199136543273926,
      "epoch": 0.9966511720897686,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004909536195514875,
      "loss": 5.949,
      "mean_token_accuracy": 0.1585733525454998,
      "num_tokens": 19503035.0,
      "step": 9970
    },
    {
      "entropy": 6.23398642539978,
      "epoch": 0.9971509971509972,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004909435986975946,
      "loss": 6.0852,
      "mean_token_accuracy": 0.15468785911798477,
      "num_tokens": 19512254.0,
      "step": 9975
    },
    {
      "entropy": 6.181322050094605,
      "epoch": 0.9976508222122257,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004909335724106162,
      "loss": 5.9885,
      "mean_token_accuracy": 0.16040295213460923,
      "num_tokens": 19523114.0,
      "step": 9980
    },
    {
      "entropy": 6.229870271682739,
      "epoch": 0.9981506472734543,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004909235406908047,
      "loss": 6.0654,
      "mean_token_accuracy": 0.14656102210283278,
      "num_tokens": 19532940.0,
      "step": 9985
    },
    {
      "entropy": 6.110600519180298,
      "epoch": 0.9986504723346828,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004909135035384126,
      "loss": 6.0512,
      "mean_token_accuracy": 0.1576165422797203,
      "num_tokens": 19542807.0,
      "step": 9990
    },
    {
      "entropy": 6.186435508728027,
      "epoch": 0.9991502973959114,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004909034609536922,
      "loss": 5.9404,
      "mean_token_accuracy": 0.15428706333041192,
      "num_tokens": 19552099.0,
      "step": 9995
    },
    {
      "entropy": 6.187509489059448,
      "epoch": 0.99965012245714,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004908934129368963,
      "loss": 6.0076,
      "mean_token_accuracy": 0.15655447989702226,
      "num_tokens": 19561709.0,
      "step": 10000
    },
    {
      "entropy": 6.147636466556126,
      "epoch": 1.0000999650122457,
      "grad_norm": 1.25,
      "learning_rate": 0.0004908833594882777,
      "loss": 6.1405,
      "mean_token_accuracy": 0.15090219262573454,
      "num_tokens": 19571739.0,
      "step": 10005
    },
    {
      "entropy": 6.214311313629151,
      "epoch": 1.0005997900734742,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004908733006080893,
      "loss": 5.9604,
      "mean_token_accuracy": 0.1582156628370285,
      "num_tokens": 19582084.0,
      "step": 10010
    },
    {
      "entropy": 6.165178966522217,
      "epoch": 1.001099615134703,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004908632362965842,
      "loss": 5.9341,
      "mean_token_accuracy": 0.15487467050552367,
      "num_tokens": 19591889.0,
      "step": 10015
    },
    {
      "entropy": 6.2151100635528564,
      "epoch": 1.0015994401959314,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004908531665540158,
      "loss": 6.0484,
      "mean_token_accuracy": 0.1532239392399788,
      "num_tokens": 19601998.0,
      "step": 10020
    },
    {
      "entropy": 6.160410356521607,
      "epoch": 1.00209926525716,
      "grad_norm": 1.328125,
      "learning_rate": 0.000490843091380637,
      "loss": 5.858,
      "mean_token_accuracy": 0.16190654635429383,
      "num_tokens": 19611347.0,
      "step": 10025
    },
    {
      "entropy": 6.257710123062134,
      "epoch": 1.0025990903183886,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004908330107767018,
      "loss": 6.0238,
      "mean_token_accuracy": 0.15361922532320021,
      "num_tokens": 19621522.0,
      "step": 10030
    },
    {
      "entropy": 6.14965968132019,
      "epoch": 1.003098915379617,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004908229247424633,
      "loss": 5.8657,
      "mean_token_accuracy": 0.16673783361911773,
      "num_tokens": 19630815.0,
      "step": 10035
    },
    {
      "entropy": 6.1763228416442875,
      "epoch": 1.0035987404408457,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004908128332781758,
      "loss": 5.9648,
      "mean_token_accuracy": 0.1563599541783333,
      "num_tokens": 19640811.0,
      "step": 10040
    },
    {
      "entropy": 6.113300132751465,
      "epoch": 1.0040985655020742,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004908027363840928,
      "loss": 5.9356,
      "mean_token_accuracy": 0.15947846919298173,
      "num_tokens": 19651335.0,
      "step": 10045
    },
    {
      "entropy": 6.128355693817139,
      "epoch": 1.004598390563303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004907926340604686,
      "loss": 5.8745,
      "mean_token_accuracy": 0.15726337879896163,
      "num_tokens": 19660242.0,
      "step": 10050
    },
    {
      "entropy": 6.181172132492065,
      "epoch": 1.0050982156245314,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004907825263075573,
      "loss": 5.9719,
      "mean_token_accuracy": 0.14894978404045106,
      "num_tokens": 19670013.0,
      "step": 10055
    },
    {
      "entropy": 6.139958143234253,
      "epoch": 1.00559804068576,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004907724131256129,
      "loss": 5.8677,
      "mean_token_accuracy": 0.16356950700283052,
      "num_tokens": 19679479.0,
      "step": 10060
    },
    {
      "entropy": 6.144499588012695,
      "epoch": 1.0060978657469886,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004907622945148903,
      "loss": 5.9158,
      "mean_token_accuracy": 0.15895372331142427,
      "num_tokens": 19688603.0,
      "step": 10065
    },
    {
      "entropy": 6.122173833847046,
      "epoch": 1.006597690808217,
      "grad_norm": 1.125,
      "learning_rate": 0.0004907521704756438,
      "loss": 5.904,
      "mean_token_accuracy": 0.14928135201334952,
      "num_tokens": 19698375.0,
      "step": 10070
    },
    {
      "entropy": 6.16668291091919,
      "epoch": 1.0070975158694457,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004907420410081284,
      "loss": 5.9506,
      "mean_token_accuracy": 0.15356743335723877,
      "num_tokens": 19708911.0,
      "step": 10075
    },
    {
      "entropy": 6.131812238693238,
      "epoch": 1.0075973409306742,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004907319061125986,
      "loss": 5.9125,
      "mean_token_accuracy": 0.15422869324684144,
      "num_tokens": 19717733.0,
      "step": 10080
    },
    {
      "entropy": 6.146936845779419,
      "epoch": 1.008097165991903,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004907217657893096,
      "loss": 5.9246,
      "mean_token_accuracy": 0.1555325761437416,
      "num_tokens": 19727192.0,
      "step": 10085
    },
    {
      "entropy": 6.125443983078003,
      "epoch": 1.0085969910531314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004907116200385165,
      "loss": 5.9066,
      "mean_token_accuracy": 0.15680666267871857,
      "num_tokens": 19737937.0,
      "step": 10090
    },
    {
      "entropy": 6.1574780464172365,
      "epoch": 1.00909681611436,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004907014688604745,
      "loss": 5.8755,
      "mean_token_accuracy": 0.15403746962547302,
      "num_tokens": 19748270.0,
      "step": 10095
    },
    {
      "entropy": 6.055540752410889,
      "epoch": 1.0095966411755886,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004906913122554391,
      "loss": 5.785,
      "mean_token_accuracy": 0.1718839794397354,
      "num_tokens": 19758328.0,
      "step": 10100
    },
    {
      "entropy": 6.041609239578247,
      "epoch": 1.010096466236817,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004906811502236657,
      "loss": 5.8763,
      "mean_token_accuracy": 0.15874550640583038,
      "num_tokens": 19767780.0,
      "step": 10105
    },
    {
      "entropy": 6.164412021636963,
      "epoch": 1.0105962912980457,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004906709827654101,
      "loss": 6.0277,
      "mean_token_accuracy": 0.14890106469392778,
      "num_tokens": 19777202.0,
      "step": 10110
    },
    {
      "entropy": 6.127697563171386,
      "epoch": 1.0110961163592742,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004906608098809281,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15777188315987586,
      "num_tokens": 19788341.0,
      "step": 10115
    },
    {
      "entropy": 6.209414720535278,
      "epoch": 1.011595941420503,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004906506315704756,
      "loss": 5.9956,
      "mean_token_accuracy": 0.15578036308288573,
      "num_tokens": 19797511.0,
      "step": 10120
    },
    {
      "entropy": 6.216711091995239,
      "epoch": 1.0120957664817314,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004906404478343087,
      "loss": 5.9775,
      "mean_token_accuracy": 0.15336853340268136,
      "num_tokens": 19807639.0,
      "step": 10125
    },
    {
      "entropy": 6.28025598526001,
      "epoch": 1.0125955915429599,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004906302586726837,
      "loss": 6.0724,
      "mean_token_accuracy": 0.14119457602500915,
      "num_tokens": 19818379.0,
      "step": 10130
    },
    {
      "entropy": 6.158285284042359,
      "epoch": 1.0130954166041886,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004906200640858567,
      "loss": 5.9238,
      "mean_token_accuracy": 0.1623810574412346,
      "num_tokens": 19828706.0,
      "step": 10135
    },
    {
      "entropy": 6.16143159866333,
      "epoch": 1.013595241665417,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004906098640740846,
      "loss": 5.9267,
      "mean_token_accuracy": 0.16458344906568528,
      "num_tokens": 19838361.0,
      "step": 10140
    },
    {
      "entropy": 6.168706655502319,
      "epoch": 1.0140950667266457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004905996586376236,
      "loss": 5.9882,
      "mean_token_accuracy": 0.1580349773168564,
      "num_tokens": 19848205.0,
      "step": 10145
    },
    {
      "entropy": 6.149914503097534,
      "epoch": 1.0145948917878742,
      "grad_norm": 1.375,
      "learning_rate": 0.0004905894477767308,
      "loss": 6.0313,
      "mean_token_accuracy": 0.1549482248723507,
      "num_tokens": 19858490.0,
      "step": 10150
    },
    {
      "entropy": 6.14423098564148,
      "epoch": 1.015094716849103,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004905792314916629,
      "loss": 5.9811,
      "mean_token_accuracy": 0.1590231716632843,
      "num_tokens": 19867551.0,
      "step": 10155
    },
    {
      "entropy": 6.1879993915557865,
      "epoch": 1.0155945419103314,
      "grad_norm": 1.375,
      "learning_rate": 0.0004905690097826769,
      "loss": 6.0125,
      "mean_token_accuracy": 0.15192604959011077,
      "num_tokens": 19877581.0,
      "step": 10160
    },
    {
      "entropy": 6.287931060791015,
      "epoch": 1.0160943669715599,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004905587826500303,
      "loss": 6.0577,
      "mean_token_accuracy": 0.15124099403619767,
      "num_tokens": 19887827.0,
      "step": 10165
    },
    {
      "entropy": 6.198428535461426,
      "epoch": 1.0165941920327886,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004905485500939801,
      "loss": 5.9433,
      "mean_token_accuracy": 0.15472912341356276,
      "num_tokens": 19897395.0,
      "step": 10170
    },
    {
      "entropy": 6.1713724613189695,
      "epoch": 1.017094017094017,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004905383121147838,
      "loss": 5.9413,
      "mean_token_accuracy": 0.15247184932231903,
      "num_tokens": 19906601.0,
      "step": 10175
    },
    {
      "entropy": 6.178199195861817,
      "epoch": 1.0175938421552457,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004905280687126991,
      "loss": 5.8922,
      "mean_token_accuracy": 0.163296677172184,
      "num_tokens": 19916422.0,
      "step": 10180
    },
    {
      "entropy": 6.222864723205566,
      "epoch": 1.0180936672164742,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004905178198879837,
      "loss": 6.018,
      "mean_token_accuracy": 0.14862206801772118,
      "num_tokens": 19927182.0,
      "step": 10185
    },
    {
      "entropy": 6.1625419616699215,
      "epoch": 1.018593492277703,
      "grad_norm": 1.9140625,
      "learning_rate": 0.0004905075656408953,
      "loss": 5.9987,
      "mean_token_accuracy": 0.14578914418816566,
      "num_tokens": 19936537.0,
      "step": 10190
    },
    {
      "entropy": 6.200917482376099,
      "epoch": 1.0190933173389314,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000490497305971692,
      "loss": 5.9958,
      "mean_token_accuracy": 0.1549546502530575,
      "num_tokens": 19946315.0,
      "step": 10195
    },
    {
      "entropy": 6.184721279144287,
      "epoch": 1.0195931424001599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000490487040880632,
      "loss": 5.8883,
      "mean_token_accuracy": 0.15741394013166427,
      "num_tokens": 19956145.0,
      "step": 10200
    },
    {
      "entropy": 6.202844524383545,
      "epoch": 1.0200929674613886,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004904767703679735,
      "loss": 6.0178,
      "mean_token_accuracy": 0.14768111780285836,
      "num_tokens": 19966890.0,
      "step": 10205
    },
    {
      "entropy": 6.194277238845825,
      "epoch": 1.020592792522617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004904664944339748,
      "loss": 5.9757,
      "mean_token_accuracy": 0.1512325443327427,
      "num_tokens": 19976156.0,
      "step": 10210
    },
    {
      "entropy": 6.201154279708862,
      "epoch": 1.0210926175838457,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004904562130788947,
      "loss": 5.8991,
      "mean_token_accuracy": 0.16157799661159516,
      "num_tokens": 19985851.0,
      "step": 10215
    },
    {
      "entropy": 6.1232868194580075,
      "epoch": 1.0215924426450742,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004904459263029916,
      "loss": 5.9988,
      "mean_token_accuracy": 0.1600301131606102,
      "num_tokens": 19996531.0,
      "step": 10220
    },
    {
      "entropy": 6.187566947937012,
      "epoch": 1.0220922677063027,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004904356341065245,
      "loss": 5.8594,
      "mean_token_accuracy": 0.15702765434980392,
      "num_tokens": 20006050.0,
      "step": 10225
    },
    {
      "entropy": 6.199140501022339,
      "epoch": 1.0225920927675314,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004904253364897521,
      "loss": 5.9988,
      "mean_token_accuracy": 0.14996725916862488,
      "num_tokens": 20015997.0,
      "step": 10230
    },
    {
      "entropy": 6.206694078445435,
      "epoch": 1.0230919178287599,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004904150334529339,
      "loss": 6.1179,
      "mean_token_accuracy": 0.14577429443597795,
      "num_tokens": 20026682.0,
      "step": 10235
    },
    {
      "entropy": 6.186622428894043,
      "epoch": 1.0235917428899886,
      "grad_norm": 2.234375,
      "learning_rate": 0.0004904047249963289,
      "loss": 5.9273,
      "mean_token_accuracy": 0.15516939610242844,
      "num_tokens": 20036166.0,
      "step": 10240
    },
    {
      "entropy": 6.131216526031494,
      "epoch": 1.024091567951217,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004903944111201964,
      "loss": 5.8999,
      "mean_token_accuracy": 0.16437097936868666,
      "num_tokens": 20045545.0,
      "step": 10245
    },
    {
      "entropy": 6.2051472663879395,
      "epoch": 1.0245913930124457,
      "grad_norm": 1.375,
      "learning_rate": 0.0004903840918247958,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15454999953508378,
      "num_tokens": 20055072.0,
      "step": 10250
    },
    {
      "entropy": 6.251996040344238,
      "epoch": 1.0250912180736742,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000490373767110387,
      "loss": 6.0258,
      "mean_token_accuracy": 0.14834926277399063,
      "num_tokens": 20064227.0,
      "step": 10255
    },
    {
      "entropy": 6.1686817646026615,
      "epoch": 1.0255910431349027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004903634369772296,
      "loss": 5.9727,
      "mean_token_accuracy": 0.158766470849514,
      "num_tokens": 20074833.0,
      "step": 10260
    },
    {
      "entropy": 6.166867446899414,
      "epoch": 1.0260908681961314,
      "grad_norm": 1.25,
      "learning_rate": 0.0004903531014255835,
      "loss": 5.9593,
      "mean_token_accuracy": 0.1468280076980591,
      "num_tokens": 20084340.0,
      "step": 10265
    },
    {
      "entropy": 6.248073768615723,
      "epoch": 1.0265906932573599,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004903427604557089,
      "loss": 5.9467,
      "mean_token_accuracy": 0.15468152165412902,
      "num_tokens": 20094372.0,
      "step": 10270
    },
    {
      "entropy": 6.214684915542603,
      "epoch": 1.0270905183185886,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004903324140678658,
      "loss": 5.9069,
      "mean_token_accuracy": 0.1550743706524372,
      "num_tokens": 20104015.0,
      "step": 10275
    },
    {
      "entropy": 6.157267093658447,
      "epoch": 1.027590343379817,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004903220622623146,
      "loss": 5.9366,
      "mean_token_accuracy": 0.1536530524492264,
      "num_tokens": 20113570.0,
      "step": 10280
    },
    {
      "entropy": 6.144119644165039,
      "epoch": 1.0280901684410457,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004903117050393157,
      "loss": 5.9728,
      "mean_token_accuracy": 0.15331557244062424,
      "num_tokens": 20124147.0,
      "step": 10285
    },
    {
      "entropy": 6.165179967880249,
      "epoch": 1.0285899935022742,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004903013423991296,
      "loss": 5.9965,
      "mean_token_accuracy": 0.14884453564882277,
      "num_tokens": 20134310.0,
      "step": 10290
    },
    {
      "entropy": 6.175594902038574,
      "epoch": 1.0290898185635027,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004902909743420173,
      "loss": 5.9209,
      "mean_token_accuracy": 0.15928234159946442,
      "num_tokens": 20143894.0,
      "step": 10295
    },
    {
      "entropy": 6.178833293914795,
      "epoch": 1.0295896436247314,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004902806008682394,
      "loss": 5.9891,
      "mean_token_accuracy": 0.15750854164361955,
      "num_tokens": 20154862.0,
      "step": 10300
    },
    {
      "entropy": 6.1593873500823975,
      "epoch": 1.0300894686859599,
      "grad_norm": 1.4921875,
      "learning_rate": 0.000490270221978057,
      "loss": 5.8979,
      "mean_token_accuracy": 0.15869831740856172,
      "num_tokens": 20164109.0,
      "step": 10305
    },
    {
      "entropy": 6.223573780059814,
      "epoch": 1.0305892937471886,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004902598376717312,
      "loss": 5.9608,
      "mean_token_accuracy": 0.14981644973158836,
      "num_tokens": 20173971.0,
      "step": 10310
    },
    {
      "entropy": 6.16736626625061,
      "epoch": 1.031089118808417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004902494479495233,
      "loss": 6.012,
      "mean_token_accuracy": 0.14979782402515412,
      "num_tokens": 20183683.0,
      "step": 10315
    },
    {
      "entropy": 6.160792303085327,
      "epoch": 1.0315889438696457,
      "grad_norm": 1.375,
      "learning_rate": 0.0004902390528116946,
      "loss": 5.9017,
      "mean_token_accuracy": 0.15736511200666428,
      "num_tokens": 20192945.0,
      "step": 10320
    },
    {
      "entropy": 6.186724424362183,
      "epoch": 1.0320887689308742,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004902286522585068,
      "loss": 5.9689,
      "mean_token_accuracy": 0.1567423090338707,
      "num_tokens": 20202919.0,
      "step": 10325
    },
    {
      "entropy": 6.137691164016724,
      "epoch": 1.0325885939921027,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004902182462902215,
      "loss": 5.889,
      "mean_token_accuracy": 0.16214252412319183,
      "num_tokens": 20212377.0,
      "step": 10330
    },
    {
      "entropy": 6.138012838363648,
      "epoch": 1.0330884190533314,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004902078349071004,
      "loss": 5.8658,
      "mean_token_accuracy": 0.16125884354114534,
      "num_tokens": 20221083.0,
      "step": 10335
    },
    {
      "entropy": 6.157989549636841,
      "epoch": 1.0335882441145599,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004901974181094057,
      "loss": 6.0032,
      "mean_token_accuracy": 0.15755846947431565,
      "num_tokens": 20229577.0,
      "step": 10340
    },
    {
      "entropy": 6.204761075973511,
      "epoch": 1.0340880691757885,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004901869958973991,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16452007442712785,
      "num_tokens": 20238400.0,
      "step": 10345
    },
    {
      "entropy": 6.0796619892120365,
      "epoch": 1.034587894237017,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004901765682713432,
      "loss": 6.0086,
      "mean_token_accuracy": 0.15354873538017272,
      "num_tokens": 20248479.0,
      "step": 10350
    },
    {
      "entropy": 6.1550952911376955,
      "epoch": 1.0350877192982457,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004901661352315003,
      "loss": 5.9677,
      "mean_token_accuracy": 0.15186137706041336,
      "num_tokens": 20258268.0,
      "step": 10355
    },
    {
      "entropy": 6.199358510971069,
      "epoch": 1.0355875443594742,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004901556967781327,
      "loss": 5.9001,
      "mean_token_accuracy": 0.16515364050865172,
      "num_tokens": 20267924.0,
      "step": 10360
    },
    {
      "entropy": 6.216310834884643,
      "epoch": 1.0360873694207027,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004901452529115031,
      "loss": 5.9963,
      "mean_token_accuracy": 0.14871933683753014,
      "num_tokens": 20278173.0,
      "step": 10365
    },
    {
      "entropy": 6.076262331008911,
      "epoch": 1.0365871944819314,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004901348036318743,
      "loss": 5.8917,
      "mean_token_accuracy": 0.15932832062244415,
      "num_tokens": 20287557.0,
      "step": 10370
    },
    {
      "entropy": 6.1769074440002445,
      "epoch": 1.0370870195431598,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004901243489395093,
      "loss": 6.0269,
      "mean_token_accuracy": 0.15256206318736076,
      "num_tokens": 20297517.0,
      "step": 10375
    },
    {
      "entropy": 6.19257378578186,
      "epoch": 1.0375868446043885,
      "grad_norm": 1.296875,
      "learning_rate": 0.000490113888834671,
      "loss": 5.85,
      "mean_token_accuracy": 0.16008726358413697,
      "num_tokens": 20308029.0,
      "step": 10380
    },
    {
      "entropy": 6.060464477539062,
      "epoch": 1.038086669665617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004901034233176227,
      "loss": 5.8334,
      "mean_token_accuracy": 0.16678026169538498,
      "num_tokens": 20318962.0,
      "step": 10385
    },
    {
      "entropy": 6.1899237632751465,
      "epoch": 1.0385864947268455,
      "grad_norm": 1.625,
      "learning_rate": 0.0004900929523886274,
      "loss": 5.97,
      "mean_token_accuracy": 0.14651992917060852,
      "num_tokens": 20328729.0,
      "step": 10390
    },
    {
      "entropy": 6.220776700973511,
      "epoch": 1.0390863197880742,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000490082476047949,
      "loss": 5.9933,
      "mean_token_accuracy": 0.1497042030096054,
      "num_tokens": 20339002.0,
      "step": 10395
    },
    {
      "entropy": 6.180819177627564,
      "epoch": 1.0395861448493027,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004900719942958506,
      "loss": 5.9198,
      "mean_token_accuracy": 0.16357310861349106,
      "num_tokens": 20347869.0,
      "step": 10400
    },
    {
      "entropy": 6.1000371932983395,
      "epoch": 1.0400859699105314,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004900615071325963,
      "loss": 5.8023,
      "mean_token_accuracy": 0.16311781257390975,
      "num_tokens": 20356609.0,
      "step": 10405
    },
    {
      "entropy": 6.078080701828003,
      "epoch": 1.0405857949717598,
      "grad_norm": 1.125,
      "learning_rate": 0.00049005101455845,
      "loss": 5.894,
      "mean_token_accuracy": 0.16752987056970597,
      "num_tokens": 20368793.0,
      "step": 10410
    },
    {
      "entropy": 6.09317102432251,
      "epoch": 1.0410856200329885,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004900405165736753,
      "loss": 5.8483,
      "mean_token_accuracy": 0.1591524973511696,
      "num_tokens": 20379525.0,
      "step": 10415
    },
    {
      "entropy": 6.207449197769165,
      "epoch": 1.041585445094217,
      "grad_norm": 1.25,
      "learning_rate": 0.0004900300131785365,
      "loss": 5.9523,
      "mean_token_accuracy": 0.15846727937459945,
      "num_tokens": 20388253.0,
      "step": 10420
    },
    {
      "entropy": 6.118882846832276,
      "epoch": 1.0420852701554455,
      "grad_norm": 1.265625,
      "learning_rate": 0.000490019504373298,
      "loss": 5.9553,
      "mean_token_accuracy": 0.1522550530731678,
      "num_tokens": 20398086.0,
      "step": 10425
    },
    {
      "entropy": 6.122468042373657,
      "epoch": 1.0425850952166742,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004900089901582241,
      "loss": 6.0355,
      "mean_token_accuracy": 0.1552769273519516,
      "num_tokens": 20408113.0,
      "step": 10430
    },
    {
      "entropy": 6.203159952163697,
      "epoch": 1.0430849202779027,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004899984705335792,
      "loss": 6.0069,
      "mean_token_accuracy": 0.1545163795351982,
      "num_tokens": 20417717.0,
      "step": 10435
    },
    {
      "entropy": 6.14851222038269,
      "epoch": 1.0435847453391314,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004899879454996282,
      "loss": 5.8783,
      "mean_token_accuracy": 0.158889739215374,
      "num_tokens": 20428307.0,
      "step": 10440
    },
    {
      "entropy": 6.063283061981201,
      "epoch": 1.0440845704003598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004899774150566358,
      "loss": 5.908,
      "mean_token_accuracy": 0.1610788583755493,
      "num_tokens": 20437493.0,
      "step": 10445
    },
    {
      "entropy": 6.132338953018189,
      "epoch": 1.0445843954615885,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000489966879204867,
      "loss": 5.9505,
      "mean_token_accuracy": 0.155977825820446,
      "num_tokens": 20448218.0,
      "step": 10450
    },
    {
      "entropy": 6.242360305786133,
      "epoch": 1.045084220522817,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004899563379445869,
      "loss": 5.9753,
      "mean_token_accuracy": 0.15821472853422164,
      "num_tokens": 20457478.0,
      "step": 10455
    },
    {
      "entropy": 6.180446910858154,
      "epoch": 1.0455840455840455,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004899457912760605,
      "loss": 5.8657,
      "mean_token_accuracy": 0.16103392392396926,
      "num_tokens": 20466776.0,
      "step": 10460
    },
    {
      "entropy": 6.099647903442383,
      "epoch": 1.0460838706452742,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004899352391995533,
      "loss": 5.9161,
      "mean_token_accuracy": 0.15760088115930557,
      "num_tokens": 20476461.0,
      "step": 10465
    },
    {
      "entropy": 6.081489419937133,
      "epoch": 1.0465836957065027,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004899246817153309,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15264288932085038,
      "num_tokens": 20487022.0,
      "step": 10470
    },
    {
      "entropy": 6.1263350486755375,
      "epoch": 1.0470835207677314,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004899141188236587,
      "loss": 5.8985,
      "mean_token_accuracy": 0.15767640322446824,
      "num_tokens": 20496326.0,
      "step": 10475
    },
    {
      "entropy": 6.2036134719848635,
      "epoch": 1.0475833458289598,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004899035505248027,
      "loss": 5.9145,
      "mean_token_accuracy": 0.15668063312768937,
      "num_tokens": 20504933.0,
      "step": 10480
    },
    {
      "entropy": 6.120710277557373,
      "epoch": 1.0480831708901883,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004898929768190286,
      "loss": 5.9022,
      "mean_token_accuracy": 0.1550515353679657,
      "num_tokens": 20513931.0,
      "step": 10485
    },
    {
      "entropy": 6.077323150634766,
      "epoch": 1.048582995951417,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004898823977066024,
      "loss": 5.8552,
      "mean_token_accuracy": 0.16517774388194084,
      "num_tokens": 20523176.0,
      "step": 10490
    },
    {
      "entropy": 6.139301681518555,
      "epoch": 1.0490828210126455,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004898718131877906,
      "loss": 5.9591,
      "mean_token_accuracy": 0.15306072384119035,
      "num_tokens": 20533049.0,
      "step": 10495
    },
    {
      "entropy": 6.1143545627594,
      "epoch": 1.0495826460738742,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004898612232628591,
      "loss": 5.8422,
      "mean_token_accuracy": 0.15934127420186997,
      "num_tokens": 20542132.0,
      "step": 10500
    },
    {
      "entropy": 6.1658447742462155,
      "epoch": 1.0500824711351027,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004898506279320745,
      "loss": 5.9459,
      "mean_token_accuracy": 0.15497103929519654,
      "num_tokens": 20551779.0,
      "step": 10505
    },
    {
      "entropy": 6.154783296585083,
      "epoch": 1.0505822961963314,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004898400271957034,
      "loss": 5.9955,
      "mean_token_accuracy": 0.1522693380713463,
      "num_tokens": 20560548.0,
      "step": 10510
    },
    {
      "entropy": 6.157649755477905,
      "epoch": 1.0510821212575598,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004898294210540126,
      "loss": 6.0489,
      "mean_token_accuracy": 0.15239737778902054,
      "num_tokens": 20570909.0,
      "step": 10515
    },
    {
      "entropy": 6.240548419952392,
      "epoch": 1.0515819463187883,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004898188095072688,
      "loss": 5.9728,
      "mean_token_accuracy": 0.14969491511583327,
      "num_tokens": 20580367.0,
      "step": 10520
    },
    {
      "entropy": 6.20021014213562,
      "epoch": 1.052081771380017,
      "grad_norm": 1.21875,
      "learning_rate": 0.000489808192555739,
      "loss": 5.8877,
      "mean_token_accuracy": 0.15834127366542816,
      "num_tokens": 20589653.0,
      "step": 10525
    },
    {
      "entropy": 6.113036441802978,
      "epoch": 1.0525815964412455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004897975701996903,
      "loss": 5.9109,
      "mean_token_accuracy": 0.15765326619148254,
      "num_tokens": 20599530.0,
      "step": 10530
    },
    {
      "entropy": 6.0761730670928955,
      "epoch": 1.0530814215024742,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004897869424393902,
      "loss": 5.8967,
      "mean_token_accuracy": 0.15822335332632065,
      "num_tokens": 20607989.0,
      "step": 10535
    },
    {
      "entropy": 6.134182596206665,
      "epoch": 1.0535812465637027,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004897763092751057,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15557971596717834,
      "num_tokens": 20618315.0,
      "step": 10540
    },
    {
      "entropy": 6.110230112075806,
      "epoch": 1.0540810716249314,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004897656707071045,
      "loss": 5.9613,
      "mean_token_accuracy": 0.15172860473394395,
      "num_tokens": 20627554.0,
      "step": 10545
    },
    {
      "entropy": 6.116087198257446,
      "epoch": 1.0545808966861598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004897550267356545,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15402805656194687,
      "num_tokens": 20636841.0,
      "step": 10550
    },
    {
      "entropy": 6.125463628768921,
      "epoch": 1.0550807217473883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004897443773610231,
      "loss": 5.8714,
      "mean_token_accuracy": 0.1593158110976219,
      "num_tokens": 20647129.0,
      "step": 10555
    },
    {
      "entropy": 6.190079593658448,
      "epoch": 1.055580546808617,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004897337225834785,
      "loss": 5.9771,
      "mean_token_accuracy": 0.14718882888555526,
      "num_tokens": 20657971.0,
      "step": 10560
    },
    {
      "entropy": 6.164877271652221,
      "epoch": 1.0560803718698455,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004897230624032888,
      "loss": 5.9546,
      "mean_token_accuracy": 0.15360582023859023,
      "num_tokens": 20666266.0,
      "step": 10565
    },
    {
      "entropy": 6.191629362106323,
      "epoch": 1.0565801969310742,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004897123968207219,
      "loss": 5.9373,
      "mean_token_accuracy": 0.15611499547958374,
      "num_tokens": 20676454.0,
      "step": 10570
    },
    {
      "entropy": 6.147012376785279,
      "epoch": 1.0570800219923027,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004897017258360465,
      "loss": 6.0234,
      "mean_token_accuracy": 0.15055552497506142,
      "num_tokens": 20687333.0,
      "step": 10575
    },
    {
      "entropy": 6.109683704376221,
      "epoch": 1.0575798470535314,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004896910494495308,
      "loss": 5.8385,
      "mean_token_accuracy": 0.16457239985466005,
      "num_tokens": 20696530.0,
      "step": 10580
    },
    {
      "entropy": 6.157804250717163,
      "epoch": 1.0580796721147598,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004896803676614436,
      "loss": 6.0031,
      "mean_token_accuracy": 0.15323769450187683,
      "num_tokens": 20706674.0,
      "step": 10585
    },
    {
      "entropy": 6.066685962677002,
      "epoch": 1.0585794971759883,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004896696804720536,
      "loss": 5.8219,
      "mean_token_accuracy": 0.1718673124909401,
      "num_tokens": 20715695.0,
      "step": 10590
    },
    {
      "entropy": 6.163532495498657,
      "epoch": 1.059079322237217,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004896589878816297,
      "loss": 6.0013,
      "mean_token_accuracy": 0.15940355211496354,
      "num_tokens": 20725763.0,
      "step": 10595
    },
    {
      "entropy": 6.115515184402466,
      "epoch": 1.0595791472984455,
      "grad_norm": 2.265625,
      "learning_rate": 0.0004896482898904409,
      "loss": 5.9239,
      "mean_token_accuracy": 0.15567436814308167,
      "num_tokens": 20734944.0,
      "step": 10600
    },
    {
      "entropy": 6.126058673858642,
      "epoch": 1.0600789723596742,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004896375864987563,
      "loss": 5.8959,
      "mean_token_accuracy": 0.1661020740866661,
      "num_tokens": 20744951.0,
      "step": 10605
    },
    {
      "entropy": 6.125064563751221,
      "epoch": 1.0605787974209027,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004896268777068453,
      "loss": 5.8637,
      "mean_token_accuracy": 0.16403547376394273,
      "num_tokens": 20754937.0,
      "step": 10610
    },
    {
      "entropy": 6.165239095687866,
      "epoch": 1.0610786224821314,
      "grad_norm": 1.25,
      "learning_rate": 0.0004896161635149773,
      "loss": 5.9564,
      "mean_token_accuracy": 0.14778674244880677,
      "num_tokens": 20767373.0,
      "step": 10615
    },
    {
      "entropy": 6.1061114311218265,
      "epoch": 1.0615784475433598,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004896054439234217,
      "loss": 5.8972,
      "mean_token_accuracy": 0.15883555710315705,
      "num_tokens": 20777647.0,
      "step": 10620
    },
    {
      "entropy": 6.103294754028321,
      "epoch": 1.0620782726045883,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004895947189324486,
      "loss": 5.9224,
      "mean_token_accuracy": 0.16352221071720124,
      "num_tokens": 20788598.0,
      "step": 10625
    },
    {
      "entropy": 6.101770830154419,
      "epoch": 1.062578097665817,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004895839885423273,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16014006584882737,
      "num_tokens": 20798586.0,
      "step": 10630
    },
    {
      "entropy": 6.115962743759155,
      "epoch": 1.0630779227270455,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004895732527533283,
      "loss": 5.8752,
      "mean_token_accuracy": 0.16609516739845276,
      "num_tokens": 20807447.0,
      "step": 10635
    },
    {
      "entropy": 6.08459095954895,
      "epoch": 1.0635777477882742,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004895625115657215,
      "loss": 5.9168,
      "mean_token_accuracy": 0.1622459352016449,
      "num_tokens": 20817889.0,
      "step": 10640
    },
    {
      "entropy": 6.129696464538574,
      "epoch": 1.0640775728495027,
      "grad_norm": 1.34375,
      "learning_rate": 0.000489551764979777,
      "loss": 5.9092,
      "mean_token_accuracy": 0.15434760004281997,
      "num_tokens": 20827858.0,
      "step": 10645
    },
    {
      "entropy": 6.117703628540039,
      "epoch": 1.0645773979107314,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004895410129957653,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1598484195768833,
      "num_tokens": 20837114.0,
      "step": 10650
    },
    {
      "entropy": 6.186951684951782,
      "epoch": 1.0650772229719598,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004895302556139569,
      "loss": 5.9688,
      "mean_token_accuracy": 0.15327855944633484,
      "num_tokens": 20846979.0,
      "step": 10655
    },
    {
      "entropy": 6.123814678192138,
      "epoch": 1.0655770480331883,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004895194928346226,
      "loss": 5.9925,
      "mean_token_accuracy": 0.15588875859975815,
      "num_tokens": 20856831.0,
      "step": 10660
    },
    {
      "entropy": 6.055693435668945,
      "epoch": 1.066076873094417,
      "grad_norm": 2.359375,
      "learning_rate": 0.000489508724658033,
      "loss": 5.8917,
      "mean_token_accuracy": 0.16461400240659713,
      "num_tokens": 20866203.0,
      "step": 10665
    },
    {
      "entropy": 6.1558702945709225,
      "epoch": 1.0665766981556455,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000489497951084459,
      "loss": 5.8926,
      "mean_token_accuracy": 0.15563634037971497,
      "num_tokens": 20875813.0,
      "step": 10670
    },
    {
      "entropy": 6.209543323516845,
      "epoch": 1.0670765232168742,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004894871721141718,
      "loss": 5.9519,
      "mean_token_accuracy": 0.15867811739444732,
      "num_tokens": 20884773.0,
      "step": 10675
    },
    {
      "entropy": 6.081042146682739,
      "epoch": 1.0675763482781027,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004894763877474426,
      "loss": 5.8898,
      "mean_token_accuracy": 0.16398087441921233,
      "num_tokens": 20894305.0,
      "step": 10680
    },
    {
      "entropy": 6.073525714874267,
      "epoch": 1.0680761733393311,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004894655979845427,
      "loss": 5.9745,
      "mean_token_accuracy": 0.15942955762147903,
      "num_tokens": 20903849.0,
      "step": 10685
    },
    {
      "entropy": 6.120926427841186,
      "epoch": 1.0685759984005598,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004894548028257434,
      "loss": 5.8537,
      "mean_token_accuracy": 0.16132117360830306,
      "num_tokens": 20913380.0,
      "step": 10690
    },
    {
      "entropy": 6.097256088256836,
      "epoch": 1.0690758234617883,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004894440022713165,
      "loss": 5.9502,
      "mean_token_accuracy": 0.1568446233868599,
      "num_tokens": 20923921.0,
      "step": 10695
    },
    {
      "entropy": 6.138236141204834,
      "epoch": 1.069575648523017,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004894331963215337,
      "loss": 5.8744,
      "mean_token_accuracy": 0.15862390846014024,
      "num_tokens": 20934613.0,
      "step": 10700
    },
    {
      "entropy": 6.113357877731323,
      "epoch": 1.0700754735842455,
      "grad_norm": 1.296875,
      "learning_rate": 0.000489422384976667,
      "loss": 5.926,
      "mean_token_accuracy": 0.15920139849185944,
      "num_tokens": 20944052.0,
      "step": 10705
    },
    {
      "entropy": 6.126824426651001,
      "epoch": 1.0705752986454742,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004894115682369881,
      "loss": 5.9411,
      "mean_token_accuracy": 0.15480325222015381,
      "num_tokens": 20953406.0,
      "step": 10710
    },
    {
      "entropy": 6.194420385360718,
      "epoch": 1.0710751237067027,
      "grad_norm": 2.5625,
      "learning_rate": 0.0004894007461027693,
      "loss": 6.0491,
      "mean_token_accuracy": 0.15091044902801515,
      "num_tokens": 20963425.0,
      "step": 10715
    },
    {
      "entropy": 6.178598880767822,
      "epoch": 1.0715749487679311,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004893899185742828,
      "loss": 5.9963,
      "mean_token_accuracy": 0.15100168585777282,
      "num_tokens": 20973777.0,
      "step": 10720
    },
    {
      "entropy": 6.125496101379395,
      "epoch": 1.0720747738291598,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004893790856518014,
      "loss": 5.8597,
      "mean_token_accuracy": 0.15923932567238808,
      "num_tokens": 20983418.0,
      "step": 10725
    },
    {
      "entropy": 6.130977535247803,
      "epoch": 1.0725745988903883,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004893682473355972,
      "loss": 5.8473,
      "mean_token_accuracy": 0.15785391330718995,
      "num_tokens": 20993051.0,
      "step": 10730
    },
    {
      "entropy": 6.081547784805298,
      "epoch": 1.073074423951617,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004893574036259432,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1653227314352989,
      "num_tokens": 21003439.0,
      "step": 10735
    },
    {
      "entropy": 6.167814683914185,
      "epoch": 1.0735742490128455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004893465545231119,
      "loss": 5.9135,
      "mean_token_accuracy": 0.16147751212120057,
      "num_tokens": 21013242.0,
      "step": 10740
    },
    {
      "entropy": 6.130961227416992,
      "epoch": 1.074074074074074,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004893357000273765,
      "loss": 5.9237,
      "mean_token_accuracy": 0.16037783026695251,
      "num_tokens": 21022268.0,
      "step": 10745
    },
    {
      "entropy": 6.143378162384034,
      "epoch": 1.0745738991353027,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048932484013901,
      "loss": 5.9653,
      "mean_token_accuracy": 0.15658110827207566,
      "num_tokens": 21032130.0,
      "step": 10750
    },
    {
      "entropy": 6.0662007331848145,
      "epoch": 1.0750737241965311,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004893139748582857,
      "loss": 5.8159,
      "mean_token_accuracy": 0.16519955098628997,
      "num_tokens": 21041966.0,
      "step": 10755
    },
    {
      "entropy": 6.034940147399903,
      "epoch": 1.0755735492577598,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000489303104185477,
      "loss": 5.803,
      "mean_token_accuracy": 0.16406726092100143,
      "num_tokens": 21051270.0,
      "step": 10760
    },
    {
      "entropy": 6.187984228134155,
      "epoch": 1.0760733743189883,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004892922281208573,
      "loss": 6.0062,
      "mean_token_accuracy": 0.15241029858589172,
      "num_tokens": 21060926.0,
      "step": 10765
    },
    {
      "entropy": 6.244573545455933,
      "epoch": 1.076573199380217,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004892813466647002,
      "loss": 6.0579,
      "mean_token_accuracy": 0.14731452465057374,
      "num_tokens": 21071367.0,
      "step": 10770
    },
    {
      "entropy": 6.10287299156189,
      "epoch": 1.0770730244414455,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004892704598172797,
      "loss": 5.9167,
      "mean_token_accuracy": 0.1556888461112976,
      "num_tokens": 21082775.0,
      "step": 10775
    },
    {
      "entropy": 6.177744340896607,
      "epoch": 1.077572849502674,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004892595675788696,
      "loss": 6.0095,
      "mean_token_accuracy": 0.15000844076275827,
      "num_tokens": 21092018.0,
      "step": 10780
    },
    {
      "entropy": 6.177650499343872,
      "epoch": 1.0780726745639027,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004892486699497439,
      "loss": 5.9097,
      "mean_token_accuracy": 0.15728323087096213,
      "num_tokens": 21101246.0,
      "step": 10785
    },
    {
      "entropy": 6.1128153800964355,
      "epoch": 1.0785724996251311,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004892377669301766,
      "loss": 5.958,
      "mean_token_accuracy": 0.1544174887239933,
      "num_tokens": 21110816.0,
      "step": 10790
    },
    {
      "entropy": 6.112386798858642,
      "epoch": 1.0790723246863598,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004892268585204423,
      "loss": 5.9314,
      "mean_token_accuracy": 0.15691734850406647,
      "num_tokens": 21120228.0,
      "step": 10795
    },
    {
      "entropy": 6.019856023788452,
      "epoch": 1.0795721497475883,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004892159447208154,
      "loss": 5.8292,
      "mean_token_accuracy": 0.16394349187612534,
      "num_tokens": 21129921.0,
      "step": 10800
    },
    {
      "entropy": 6.138217449188232,
      "epoch": 1.080071974808817,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004892050255315705,
      "loss": 6.0044,
      "mean_token_accuracy": 0.1535131722688675,
      "num_tokens": 21139605.0,
      "step": 10805
    },
    {
      "entropy": 6.071519994735718,
      "epoch": 1.0805717998700455,
      "grad_norm": 1.5,
      "learning_rate": 0.0004891941009529822,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1676989421248436,
      "num_tokens": 21149004.0,
      "step": 10810
    },
    {
      "entropy": 6.19970121383667,
      "epoch": 1.081071624931274,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004891831709853255,
      "loss": 5.9457,
      "mean_token_accuracy": 0.14905372858047486,
      "num_tokens": 21159013.0,
      "step": 10815
    },
    {
      "entropy": 6.138840961456299,
      "epoch": 1.0815714499925027,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004891722356288752,
      "loss": 5.88,
      "mean_token_accuracy": 0.1590219870209694,
      "num_tokens": 21169432.0,
      "step": 10820
    },
    {
      "entropy": 6.075982475280762,
      "epoch": 1.0820712750537311,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004891612948839068,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16560973972082138,
      "num_tokens": 21179672.0,
      "step": 10825
    },
    {
      "entropy": 6.06056957244873,
      "epoch": 1.0825711001149598,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004891503487506951,
      "loss": 5.8978,
      "mean_token_accuracy": 0.16178446263074875,
      "num_tokens": 21189562.0,
      "step": 10830
    },
    {
      "entropy": 6.20071120262146,
      "epoch": 1.0830709251761883,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004891393972295159,
      "loss": 6.038,
      "mean_token_accuracy": 0.1428327202796936,
      "num_tokens": 21199538.0,
      "step": 10835
    },
    {
      "entropy": 6.1935553550720215,
      "epoch": 1.083570750237417,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004891284403206444,
      "loss": 5.9849,
      "mean_token_accuracy": 0.155917414277792,
      "num_tokens": 21209355.0,
      "step": 10840
    },
    {
      "entropy": 6.099454545974732,
      "epoch": 1.0840705752986455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004891174780243566,
      "loss": 5.8697,
      "mean_token_accuracy": 0.16033392995595933,
      "num_tokens": 21218268.0,
      "step": 10845
    },
    {
      "entropy": 6.1465202331542965,
      "epoch": 1.084570400359874,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000489106510340928,
      "loss": 5.9918,
      "mean_token_accuracy": 0.15531508922576903,
      "num_tokens": 21228376.0,
      "step": 10850
    },
    {
      "entropy": 6.220686721801758,
      "epoch": 1.0850702254211027,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004890955372706348,
      "loss": 5.9884,
      "mean_token_accuracy": 0.15212095379829407,
      "num_tokens": 21237781.0,
      "step": 10855
    },
    {
      "entropy": 6.116746234893799,
      "epoch": 1.0855700504823311,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004890845588137531,
      "loss": 5.8298,
      "mean_token_accuracy": 0.1613258957862854,
      "num_tokens": 21247096.0,
      "step": 10860
    },
    {
      "entropy": 6.115295934677124,
      "epoch": 1.0860698755435598,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004890735749705587,
      "loss": 6.0153,
      "mean_token_accuracy": 0.15539822578430176,
      "num_tokens": 21258645.0,
      "step": 10865
    },
    {
      "entropy": 6.161628198623657,
      "epoch": 1.0865697006047883,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004890625857413283,
      "loss": 5.9407,
      "mean_token_accuracy": 0.15730203241109847,
      "num_tokens": 21268634.0,
      "step": 10870
    },
    {
      "entropy": 6.155200004577637,
      "epoch": 1.087069525666017,
      "grad_norm": 1.25,
      "learning_rate": 0.0004890515911263385,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1557600937783718,
      "num_tokens": 21278996.0,
      "step": 10875
    },
    {
      "entropy": 5.9977256774902346,
      "epoch": 1.0875693507272455,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004890405911258656,
      "loss": 5.7303,
      "mean_token_accuracy": 0.17373621612787246,
      "num_tokens": 21288689.0,
      "step": 10880
    },
    {
      "entropy": 6.110237455368042,
      "epoch": 1.088069175788474,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004890295857401866,
      "loss": 5.9598,
      "mean_token_accuracy": 0.15061759799718857,
      "num_tokens": 21298542.0,
      "step": 10885
    },
    {
      "entropy": 6.144878768920899,
      "epoch": 1.0885690008497027,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004890185749695783,
      "loss": 5.906,
      "mean_token_accuracy": 0.15705824345350267,
      "num_tokens": 21308035.0,
      "step": 10890
    },
    {
      "entropy": 6.221332359313965,
      "epoch": 1.0890688259109311,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004890075588143176,
      "loss": 6.0395,
      "mean_token_accuracy": 0.1469349555671215,
      "num_tokens": 21317898.0,
      "step": 10895
    },
    {
      "entropy": 6.158019113540649,
      "epoch": 1.0895686509721598,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004889965372746819,
      "loss": 5.9194,
      "mean_token_accuracy": 0.15325864255428315,
      "num_tokens": 21327347.0,
      "step": 10900
    },
    {
      "entropy": 6.153690147399902,
      "epoch": 1.0900684760333883,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004889855103509484,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15160580798983575,
      "num_tokens": 21336592.0,
      "step": 10905
    },
    {
      "entropy": 6.146178150177002,
      "epoch": 1.090568301094617,
      "grad_norm": 1.125,
      "learning_rate": 0.0004889744780433944,
      "loss": 5.9992,
      "mean_token_accuracy": 0.15149357169866562,
      "num_tokens": 21345764.0,
      "step": 10910
    },
    {
      "entropy": 6.1833874702453615,
      "epoch": 1.0910681261558455,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004889634403522976,
      "loss": 5.9173,
      "mean_token_accuracy": 0.1578316032886505,
      "num_tokens": 21354423.0,
      "step": 10915
    },
    {
      "entropy": 6.12634916305542,
      "epoch": 1.091567951217074,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004889523972779359,
      "loss": 5.944,
      "mean_token_accuracy": 0.16468725353479385,
      "num_tokens": 21364407.0,
      "step": 10920
    },
    {
      "entropy": 6.190716123580932,
      "epoch": 1.0920677762783026,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004889413488205868,
      "loss": 6.0535,
      "mean_token_accuracy": 0.15155917033553123,
      "num_tokens": 21374642.0,
      "step": 10925
    },
    {
      "entropy": 6.089123868942261,
      "epoch": 1.0925676013395311,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004889302949805283,
      "loss": 5.8216,
      "mean_token_accuracy": 0.1710789069533348,
      "num_tokens": 21384437.0,
      "step": 10930
    },
    {
      "entropy": 6.11716103553772,
      "epoch": 1.0930674264007598,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004889192357580388,
      "loss": 5.8411,
      "mean_token_accuracy": 0.16206925809383393,
      "num_tokens": 21393924.0,
      "step": 10935
    },
    {
      "entropy": 6.043091726303101,
      "epoch": 1.0935672514619883,
      "grad_norm": 1.375,
      "learning_rate": 0.0004889081711533964,
      "loss": 5.8672,
      "mean_token_accuracy": 0.16775585114955902,
      "num_tokens": 21402924.0,
      "step": 10940
    },
    {
      "entropy": 6.1635995388031,
      "epoch": 1.0940670765232168,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004888971011668793,
      "loss": 6.0171,
      "mean_token_accuracy": 0.14716267064213753,
      "num_tokens": 21412422.0,
      "step": 10945
    },
    {
      "entropy": 6.181364631652832,
      "epoch": 1.0945669015844455,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004888860257987662,
      "loss": 5.9795,
      "mean_token_accuracy": 0.15764326080679894,
      "num_tokens": 21421830.0,
      "step": 10950
    },
    {
      "entropy": 6.143895864486694,
      "epoch": 1.095066726645674,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004888749450493358,
      "loss": 5.9125,
      "mean_token_accuracy": 0.16008587554097176,
      "num_tokens": 21431602.0,
      "step": 10955
    },
    {
      "entropy": 6.135023593902588,
      "epoch": 1.0955665517069026,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004888638589188669,
      "loss": 5.9572,
      "mean_token_accuracy": 0.15589238554239274,
      "num_tokens": 21441110.0,
      "step": 10960
    },
    {
      "entropy": 6.194552850723267,
      "epoch": 1.0960663767681311,
      "grad_norm": 2.40625,
      "learning_rate": 0.0004888527674076382,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1514907792210579,
      "num_tokens": 21450626.0,
      "step": 10965
    },
    {
      "entropy": 6.130096864700318,
      "epoch": 1.0965662018293598,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000488841670515929,
      "loss": 5.8471,
      "mean_token_accuracy": 0.16127452552318572,
      "num_tokens": 21461174.0,
      "step": 10970
    },
    {
      "entropy": 6.110013437271118,
      "epoch": 1.0970660268905883,
      "grad_norm": 1.25,
      "learning_rate": 0.0004888305682440183,
      "loss": 5.9107,
      "mean_token_accuracy": 0.15942692905664443,
      "num_tokens": 21470895.0,
      "step": 10975
    },
    {
      "entropy": 6.081044864654541,
      "epoch": 1.0975658519518168,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004888194605921855,
      "loss": 5.8517,
      "mean_token_accuracy": 0.16124730408191681,
      "num_tokens": 21480192.0,
      "step": 10980
    },
    {
      "entropy": 6.172085857391357,
      "epoch": 1.0980656770130455,
      "grad_norm": 1.25,
      "learning_rate": 0.0004888083475607102,
      "loss": 5.9504,
      "mean_token_accuracy": 0.1547415539622307,
      "num_tokens": 21489418.0,
      "step": 10985
    },
    {
      "entropy": 6.195443487167358,
      "epoch": 1.098565502074274,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004887972291498718,
      "loss": 6.0021,
      "mean_token_accuracy": 0.15475521832704545,
      "num_tokens": 21499228.0,
      "step": 10990
    },
    {
      "entropy": 6.194277811050415,
      "epoch": 1.0990653271355026,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004887861053599503,
      "loss": 5.8674,
      "mean_token_accuracy": 0.15728126466274261,
      "num_tokens": 21509051.0,
      "step": 10995
    },
    {
      "entropy": 6.1151832103729244,
      "epoch": 1.0995651521967311,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004887749761912252,
      "loss": 5.9378,
      "mean_token_accuracy": 0.15539976209402084,
      "num_tokens": 21518853.0,
      "step": 11000
    },
    {
      "entropy": 6.168104791641236,
      "epoch": 1.1000649772579596,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0004887638416439768,
      "loss": 6.0083,
      "mean_token_accuracy": 0.1579429641366005,
      "num_tokens": 21528547.0,
      "step": 11005
    },
    {
      "entropy": 6.228917646408081,
      "epoch": 1.1005648023191883,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004887527017184851,
      "loss": 5.9796,
      "mean_token_accuracy": 0.15040475875139236,
      "num_tokens": 21538166.0,
      "step": 11010
    },
    {
      "entropy": 6.192111921310425,
      "epoch": 1.1010646273804168,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004887415564150305,
      "loss": 5.9326,
      "mean_token_accuracy": 0.15772496163845062,
      "num_tokens": 21548212.0,
      "step": 11015
    },
    {
      "entropy": 6.191812181472779,
      "epoch": 1.1015644524416455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004887304057338934,
      "loss": 5.9243,
      "mean_token_accuracy": 0.15140566676855088,
      "num_tokens": 21558809.0,
      "step": 11020
    },
    {
      "entropy": 6.0244838237762455,
      "epoch": 1.102064277502874,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000488719249675354,
      "loss": 5.8215,
      "mean_token_accuracy": 0.1684559941291809,
      "num_tokens": 21567923.0,
      "step": 11025
    },
    {
      "entropy": 6.051117420196533,
      "epoch": 1.1025641025641026,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004887080882396936,
      "loss": 5.9449,
      "mean_token_accuracy": 0.16105058640241623,
      "num_tokens": 21578051.0,
      "step": 11030
    },
    {
      "entropy": 6.105144405364991,
      "epoch": 1.1030639276253311,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004886969214271925,
      "loss": 5.9608,
      "mean_token_accuracy": 0.16272468864917755,
      "num_tokens": 21586950.0,
      "step": 11035
    },
    {
      "entropy": 6.1967620849609375,
      "epoch": 1.1035637526865596,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004886857492381319,
      "loss": 5.953,
      "mean_token_accuracy": 0.15561981797218322,
      "num_tokens": 21596577.0,
      "step": 11040
    },
    {
      "entropy": 6.191910457611084,
      "epoch": 1.1040635777477883,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004886745716727928,
      "loss": 5.9874,
      "mean_token_accuracy": 0.15815995931625365,
      "num_tokens": 21606474.0,
      "step": 11045
    },
    {
      "entropy": 6.085840225219727,
      "epoch": 1.1045634028090168,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004886633887314565,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16914485394954681,
      "num_tokens": 21616277.0,
      "step": 11050
    },
    {
      "entropy": 6.156643676757812,
      "epoch": 1.1050632278702455,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004886522004144044,
      "loss": 5.981,
      "mean_token_accuracy": 0.1498487636446953,
      "num_tokens": 21626096.0,
      "step": 11055
    },
    {
      "entropy": 6.199607944488525,
      "epoch": 1.105563052931474,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004886410067219178,
      "loss": 6.0119,
      "mean_token_accuracy": 0.1499502770602703,
      "num_tokens": 21635295.0,
      "step": 11060
    },
    {
      "entropy": 6.224227237701416,
      "epoch": 1.1060628779927026,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004886298076542785,
      "loss": 6.023,
      "mean_token_accuracy": 0.1452885553240776,
      "num_tokens": 21645900.0,
      "step": 11065
    },
    {
      "entropy": 6.121536016464233,
      "epoch": 1.1065627030539311,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004886186032117682,
      "loss": 5.9212,
      "mean_token_accuracy": 0.15657880008220673,
      "num_tokens": 21654757.0,
      "step": 11070
    },
    {
      "entropy": 6.091936302185059,
      "epoch": 1.1070625281151596,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004886073933946688,
      "loss": 5.8579,
      "mean_token_accuracy": 0.16425320208072663,
      "num_tokens": 21665253.0,
      "step": 11075
    },
    {
      "entropy": 6.04825987815857,
      "epoch": 1.1075623531763883,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004885961782032624,
      "loss": 5.7777,
      "mean_token_accuracy": 0.17008589953184128,
      "num_tokens": 21674345.0,
      "step": 11080
    },
    {
      "entropy": 6.104118824005127,
      "epoch": 1.1080621782376168,
      "grad_norm": 1.515625,
      "learning_rate": 0.000488584957637831,
      "loss": 5.9636,
      "mean_token_accuracy": 0.15971079617738723,
      "num_tokens": 21683110.0,
      "step": 11085
    },
    {
      "entropy": 6.0610589504241945,
      "epoch": 1.1085620032988455,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004885737316986571,
      "loss": 5.8419,
      "mean_token_accuracy": 0.16379538476467131,
      "num_tokens": 21692141.0,
      "step": 11090
    },
    {
      "entropy": 6.2004670143127445,
      "epoch": 1.109061828360074,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000488562500386023,
      "loss": 5.9557,
      "mean_token_accuracy": 0.15822781026363372,
      "num_tokens": 21700945.0,
      "step": 11095
    },
    {
      "entropy": 6.076145982742309,
      "epoch": 1.1095616534213026,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004885512637002115,
      "loss": 5.8555,
      "mean_token_accuracy": 0.17093067169189452,
      "num_tokens": 21711630.0,
      "step": 11100
    },
    {
      "entropy": 6.213993406295776,
      "epoch": 1.1100614784825311,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004885400216415051,
      "loss": 6.0415,
      "mean_token_accuracy": 0.14727294221520423,
      "num_tokens": 21721673.0,
      "step": 11105
    },
    {
      "entropy": 6.225103759765625,
      "epoch": 1.1105613035437596,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004885287742101865,
      "loss": 5.9774,
      "mean_token_accuracy": 0.14789579063653946,
      "num_tokens": 21731655.0,
      "step": 11110
    },
    {
      "entropy": 6.115630149841309,
      "epoch": 1.1110611286049883,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004885175214065392,
      "loss": 5.9681,
      "mean_token_accuracy": 0.16154656708240508,
      "num_tokens": 21741465.0,
      "step": 11115
    },
    {
      "entropy": 6.132784795761109,
      "epoch": 1.1115609536662168,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004885062632308457,
      "loss": 5.9053,
      "mean_token_accuracy": 0.16261844485998153,
      "num_tokens": 21750313.0,
      "step": 11120
    },
    {
      "entropy": 6.22063512802124,
      "epoch": 1.1120607787274455,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004884949996833898,
      "loss": 6.0262,
      "mean_token_accuracy": 0.14515237063169478,
      "num_tokens": 21759842.0,
      "step": 11125
    },
    {
      "entropy": 6.170238447189331,
      "epoch": 1.112560603788674,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004884837307644545,
      "loss": 5.8768,
      "mean_token_accuracy": 0.16237102448940277,
      "num_tokens": 21770126.0,
      "step": 11130
    },
    {
      "entropy": 6.054769802093506,
      "epoch": 1.1130604288499026,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004884724564743236,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16065026372671126,
      "num_tokens": 21779878.0,
      "step": 11135
    },
    {
      "entropy": 6.156855869293213,
      "epoch": 1.1135602539111311,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004884611768132805,
      "loss": 6.0,
      "mean_token_accuracy": 0.15262884944677352,
      "num_tokens": 21790971.0,
      "step": 11140
    },
    {
      "entropy": 6.19208836555481,
      "epoch": 1.1140600789723596,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004884498917816092,
      "loss": 5.9994,
      "mean_token_accuracy": 0.147748701274395,
      "num_tokens": 21800733.0,
      "step": 11145
    },
    {
      "entropy": 6.155978870391846,
      "epoch": 1.1145599040335883,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004884386013795935,
      "loss": 5.9083,
      "mean_token_accuracy": 0.16489729285240173,
      "num_tokens": 21809840.0,
      "step": 11150
    },
    {
      "entropy": 6.069447326660156,
      "epoch": 1.1150597290948168,
      "grad_norm": 1.6015625,
      "learning_rate": 0.0004884273056075175,
      "loss": 5.9508,
      "mean_token_accuracy": 0.1577574774622917,
      "num_tokens": 21819539.0,
      "step": 11155
    },
    {
      "entropy": 6.198151683807373,
      "epoch": 1.1155595541560455,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004884160044656654,
      "loss": 5.9289,
      "mean_token_accuracy": 0.16398066580295562,
      "num_tokens": 21829488.0,
      "step": 11160
    },
    {
      "entropy": 6.24658784866333,
      "epoch": 1.116059379217274,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004884046979543217,
      "loss": 5.9779,
      "mean_token_accuracy": 0.15356515273451804,
      "num_tokens": 21839244.0,
      "step": 11165
    },
    {
      "entropy": 6.168745231628418,
      "epoch": 1.1165592042785026,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004883933860737706,
      "loss": 6.0692,
      "mean_token_accuracy": 0.15458495691418647,
      "num_tokens": 21850102.0,
      "step": 11170
    },
    {
      "entropy": 6.22074327468872,
      "epoch": 1.1170590293397311,
      "grad_norm": 1.25,
      "learning_rate": 0.0004883820688242968,
      "loss": 5.9601,
      "mean_token_accuracy": 0.15867770090699196,
      "num_tokens": 21860409.0,
      "step": 11175
    },
    {
      "entropy": 6.1002233028411865,
      "epoch": 1.1175588544009596,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004883707462061851,
      "loss": 5.8247,
      "mean_token_accuracy": 0.15803707242012024,
      "num_tokens": 21870055.0,
      "step": 11180
    },
    {
      "entropy": 6.133212614059448,
      "epoch": 1.1180586794621883,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004883594182197203,
      "loss": 5.9826,
      "mean_token_accuracy": 0.15488905608654022,
      "num_tokens": 21879103.0,
      "step": 11185
    },
    {
      "entropy": 6.144361162185669,
      "epoch": 1.1185585045234168,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004883480848651875,
      "loss": 6.0008,
      "mean_token_accuracy": 0.15601723194122313,
      "num_tokens": 21890140.0,
      "step": 11190
    },
    {
      "entropy": 6.299938154220581,
      "epoch": 1.1190583295846455,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004883367461428716,
      "loss": 6.1177,
      "mean_token_accuracy": 0.14664446488022803,
      "num_tokens": 21900917.0,
      "step": 11195
    },
    {
      "entropy": 6.238546514511109,
      "epoch": 1.119558154645874,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883254020530582,
      "loss": 5.953,
      "mean_token_accuracy": 0.15080673694610597,
      "num_tokens": 21910476.0,
      "step": 11200
    },
    {
      "entropy": 6.125157070159912,
      "epoch": 1.1200579797071024,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004883140525960327,
      "loss": 5.9289,
      "mean_token_accuracy": 0.15645034462213517,
      "num_tokens": 21920219.0,
      "step": 11205
    },
    {
      "entropy": 6.07569260597229,
      "epoch": 1.1205578047683311,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004883026977720803,
      "loss": 5.9009,
      "mean_token_accuracy": 0.1592889681458473,
      "num_tokens": 21930592.0,
      "step": 11210
    },
    {
      "entropy": 6.2165083408355715,
      "epoch": 1.1210576298295596,
      "grad_norm": 1.234375,
      "learning_rate": 0.000488291337581487,
      "loss": 6.1128,
      "mean_token_accuracy": 0.14356039613485336,
      "num_tokens": 21939785.0,
      "step": 11215
    },
    {
      "entropy": 6.225708532333374,
      "epoch": 1.1215574548907883,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004882799720245385,
      "loss": 5.9138,
      "mean_token_accuracy": 0.15952995121479036,
      "num_tokens": 21948243.0,
      "step": 11220
    },
    {
      "entropy": 6.20636134147644,
      "epoch": 1.1220572799520168,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004882686011015208,
      "loss": 5.9704,
      "mean_token_accuracy": 0.15272683650255203,
      "num_tokens": 21958523.0,
      "step": 11225
    },
    {
      "entropy": 6.110539436340332,
      "epoch": 1.1225571050132455,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048825722481272006,
      "loss": 5.962,
      "mean_token_accuracy": 0.15418931245803832,
      "num_tokens": 21968685.0,
      "step": 11230
    },
    {
      "entropy": 6.15864405632019,
      "epoch": 1.123056930074474,
      "grad_norm": 1.6484375,
      "learning_rate": 0.0004882458431584223,
      "loss": 5.9776,
      "mean_token_accuracy": 0.14996535331010818,
      "num_tokens": 21979896.0,
      "step": 11235
    },
    {
      "entropy": 6.238428544998169,
      "epoch": 1.1235567551357024,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004882344561389141,
      "loss": 5.9549,
      "mean_token_accuracy": 0.15391377434134484,
      "num_tokens": 21989513.0,
      "step": 11240
    },
    {
      "entropy": 6.1046394348144535,
      "epoch": 1.124056580196931,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048822306375448176,
      "loss": 5.904,
      "mean_token_accuracy": 0.15924802124500276,
      "num_tokens": 21999238.0,
      "step": 11245
    },
    {
      "entropy": 6.123053979873657,
      "epoch": 1.1245564052581596,
      "grad_norm": 1.375,
      "learning_rate": 0.0004882116660054121,
      "loss": 5.93,
      "mean_token_accuracy": 0.1566139817237854,
      "num_tokens": 22008992.0,
      "step": 11250
    },
    {
      "entropy": 6.166887712478638,
      "epoch": 1.1250562303193883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004882002628919917,
      "loss": 5.908,
      "mean_token_accuracy": 0.15392929315567017,
      "num_tokens": 22018333.0,
      "step": 11255
    },
    {
      "entropy": 6.1368818283081055,
      "epoch": 1.1255560553806168,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004881888544145076,
      "loss": 5.9434,
      "mean_token_accuracy": 0.1624426931142807,
      "num_tokens": 22028003.0,
      "step": 11260
    },
    {
      "entropy": 6.174270677566528,
      "epoch": 1.1260558804418452,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048817744057324673,
      "loss": 5.988,
      "mean_token_accuracy": 0.15294573158025743,
      "num_tokens": 22038318.0,
      "step": 11265
    },
    {
      "entropy": 6.154047346115112,
      "epoch": 1.126555705503074,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048816602136849644,
      "loss": 5.9185,
      "mean_token_accuracy": 0.1574176475405693,
      "num_tokens": 22048180.0,
      "step": 11270
    },
    {
      "entropy": 6.137226724624634,
      "epoch": 1.1270555305643024,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048815459680054375,
      "loss": 5.956,
      "mean_token_accuracy": 0.14919542372226716,
      "num_tokens": 22057831.0,
      "step": 11275
    },
    {
      "entropy": 6.097395324707032,
      "epoch": 1.127555355625531,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048814316686967635,
      "loss": 5.855,
      "mean_token_accuracy": 0.1623263344168663,
      "num_tokens": 22068569.0,
      "step": 11280
    },
    {
      "entropy": 6.203636169433594,
      "epoch": 1.1280551806867596,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004881317315761817,
      "loss": 6.0236,
      "mean_token_accuracy": 0.1530197262763977,
      "num_tokens": 22079267.0,
      "step": 11285
    },
    {
      "entropy": 6.136734914779663,
      "epoch": 1.1285550057479883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048812029092034747,
      "loss": 5.8664,
      "mean_token_accuracy": 0.1613774538040161,
      "num_tokens": 22089469.0,
      "step": 11290
    },
    {
      "entropy": 6.101068639755249,
      "epoch": 1.1290548308092168,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048810884490246164,
      "loss": 5.9207,
      "mean_token_accuracy": 0.1657998040318489,
      "num_tokens": 22099148.0,
      "step": 11295
    },
    {
      "entropy": 6.120045280456543,
      "epoch": 1.1295546558704452,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000488097393522812,
      "loss": 5.8885,
      "mean_token_accuracy": 0.16285810470581055,
      "num_tokens": 22107750.0,
      "step": 11300
    },
    {
      "entropy": 6.177941370010376,
      "epoch": 1.130054480931674,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004880859367816868,
      "loss": 6.0177,
      "mean_token_accuracy": 0.15005035623908042,
      "num_tokens": 22118336.0,
      "step": 11305
    },
    {
      "entropy": 6.214954471588134,
      "epoch": 1.1305543059929024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004880744746793744,
      "loss": 5.9228,
      "mean_token_accuracy": 0.1561885416507721,
      "num_tokens": 22128166.0,
      "step": 11310
    },
    {
      "entropy": 6.169269800186157,
      "epoch": 1.131054131054131,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048806300721616295,
      "loss": 5.913,
      "mean_token_accuracy": 0.15371521562337875,
      "num_tokens": 22136810.0,
      "step": 11315
    },
    {
      "entropy": 6.0631739616394045,
      "epoch": 1.1315539561153596,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048805153439234107,
      "loss": 5.878,
      "mean_token_accuracy": 0.16770315319299697,
      "num_tokens": 22146398.0,
      "step": 11320
    },
    {
      "entropy": 6.1555359840393065,
      "epoch": 1.1320537811765883,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004880400562081974,
      "loss": 5.9181,
      "mean_token_accuracy": 0.1562088444828987,
      "num_tokens": 22155161.0,
      "step": 11325
    },
    {
      "entropy": 6.109787702560425,
      "epoch": 1.1325536062378168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048802857266402075,
      "loss": 5.9685,
      "mean_token_accuracy": 0.16022876799106597,
      "num_tokens": 22166486.0,
      "step": 11330
    },
    {
      "entropy": 6.076881980895996,
      "epoch": 1.1330534312990452,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0004880170837601,
      "loss": 5.8722,
      "mean_token_accuracy": 0.16129899621009827,
      "num_tokens": 22176487.0,
      "step": 11335
    },
    {
      "entropy": 6.137448167800903,
      "epoch": 1.133553256360274,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004880055894967242,
      "loss": 5.9489,
      "mean_token_accuracy": 0.15399421602487565,
      "num_tokens": 22186513.0,
      "step": 11340
    },
    {
      "entropy": 6.175388669967651,
      "epoch": 1.1340530814215024,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048799408987418264,
      "loss": 5.9451,
      "mean_token_accuracy": 0.1530529133975506,
      "num_tokens": 22197325.0,
      "step": 11345
    },
    {
      "entropy": 6.235398578643799,
      "epoch": 1.134552906482731,
      "grad_norm": 1.375,
      "learning_rate": 0.00048798258489276457,
      "loss": 5.9558,
      "mean_token_accuracy": 0.15394756197929382,
      "num_tokens": 22207160.0,
      "step": 11350
    },
    {
      "entropy": 6.2171226978302006,
      "epoch": 1.1350527315439596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004879710745527595,
      "loss": 5.9886,
      "mean_token_accuracy": 0.1499202013015747,
      "num_tokens": 22216533.0,
      "step": 11355
    },
    {
      "entropy": 6.168427658081055,
      "epoch": 1.1355525566051883,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048795955885445696,
      "loss": 5.9375,
      "mean_token_accuracy": 0.154099303483963,
      "num_tokens": 22226186.0,
      "step": 11360
    },
    {
      "entropy": 6.132285690307617,
      "epoch": 1.1360523816664168,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004879480377981467,
      "loss": 5.9062,
      "mean_token_accuracy": 0.1605280488729477,
      "num_tokens": 22236315.0,
      "step": 11365
    },
    {
      "entropy": 6.150118064880371,
      "epoch": 1.1365522067276452,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048793651138411866,
      "loss": 5.8645,
      "mean_token_accuracy": 0.16115637719631196,
      "num_tokens": 22245250.0,
      "step": 11370
    },
    {
      "entropy": 6.224485445022583,
      "epoch": 1.137052031788874,
      "grad_norm": 1.90625,
      "learning_rate": 0.00048792497961266273,
      "loss": 6.0292,
      "mean_token_accuracy": 0.1485135167837143,
      "num_tokens": 22255093.0,
      "step": 11375
    },
    {
      "entropy": 6.039935493469239,
      "epoch": 1.1375518568501024,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048791344248406916,
      "loss": 5.8264,
      "mean_token_accuracy": 0.15991712212562562,
      "num_tokens": 22264680.0,
      "step": 11380
    },
    {
      "entropy": 6.128985977172851,
      "epoch": 1.138051681911331,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048790189999862814,
      "loss": 5.9895,
      "mean_token_accuracy": 0.1472291514277458,
      "num_tokens": 22273804.0,
      "step": 11385
    },
    {
      "entropy": 6.181646013259888,
      "epoch": 1.1385515069725596,
      "grad_norm": 1.671875,
      "learning_rate": 0.00048789035215663013,
      "loss": 5.9326,
      "mean_token_accuracy": 0.15447390973567962,
      "num_tokens": 22284302.0,
      "step": 11390
    },
    {
      "entropy": 6.137444877624512,
      "epoch": 1.1390513320337883,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048787879895836563,
      "loss": 5.867,
      "mean_token_accuracy": 0.16312607228755951,
      "num_tokens": 22293439.0,
      "step": 11395
    },
    {
      "entropy": 6.1159912109375,
      "epoch": 1.1395511570950168,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004878672404041253,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15982389748096465,
      "num_tokens": 22303005.0,
      "step": 11400
    },
    {
      "entropy": 6.085371494293213,
      "epoch": 1.1400509821562452,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048785567649419994,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16179968118667604,
      "num_tokens": 22311962.0,
      "step": 11405
    },
    {
      "entropy": 6.048926877975464,
      "epoch": 1.140550807217474,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004878441072288806,
      "loss": 5.95,
      "mean_token_accuracy": 0.14932782649993898,
      "num_tokens": 22320998.0,
      "step": 11410
    },
    {
      "entropy": 6.135735702514649,
      "epoch": 1.1410506322787024,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048783253260845826,
      "loss": 5.9274,
      "mean_token_accuracy": 0.15540642365813256,
      "num_tokens": 22330977.0,
      "step": 11415
    },
    {
      "entropy": 6.1888429641723635,
      "epoch": 1.141550457339931,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004878209526332242,
      "loss": 5.9255,
      "mean_token_accuracy": 0.15925299525260925,
      "num_tokens": 22341287.0,
      "step": 11420
    },
    {
      "entropy": 6.119518089294433,
      "epoch": 1.1420502824011596,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048780936730346967,
      "loss": 5.8915,
      "mean_token_accuracy": 0.15947517156600952,
      "num_tokens": 22350507.0,
      "step": 11425
    },
    {
      "entropy": 6.142720794677734,
      "epoch": 1.1425501074623883,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048779777661948627,
      "loss": 5.9904,
      "mean_token_accuracy": 0.14859276562929152,
      "num_tokens": 22361202.0,
      "step": 11430
    },
    {
      "entropy": 6.217611837387085,
      "epoch": 1.1430499325236168,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004877861805815655,
      "loss": 6.0351,
      "mean_token_accuracy": 0.15991177707910537,
      "num_tokens": 22370824.0,
      "step": 11435
    },
    {
      "entropy": 6.117256784439087,
      "epoch": 1.1435497575848452,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004877745791899993,
      "loss": 5.8993,
      "mean_token_accuracy": 0.16042656004428862,
      "num_tokens": 22380857.0,
      "step": 11440
    },
    {
      "entropy": 6.186435127258301,
      "epoch": 1.144049582646074,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048776297244507936,
      "loss": 5.9117,
      "mean_token_accuracy": 0.15356415063142775,
      "num_tokens": 22391354.0,
      "step": 11445
    },
    {
      "entropy": 6.0872776985168455,
      "epoch": 1.1445494077073024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048775136034709784,
      "loss": 5.8421,
      "mean_token_accuracy": 0.15969178453087807,
      "num_tokens": 22400656.0,
      "step": 11450
    },
    {
      "entropy": 6.148261451721192,
      "epoch": 1.145049232768531,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048773974289634676,
      "loss": 5.9217,
      "mean_token_accuracy": 0.15257085114717484,
      "num_tokens": 22410887.0,
      "step": 11455
    },
    {
      "entropy": 6.164778852462769,
      "epoch": 1.1455490578297596,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004877281200931185,
      "loss": 5.9273,
      "mean_token_accuracy": 0.15148293375968933,
      "num_tokens": 22420370.0,
      "step": 11460
    },
    {
      "entropy": 6.0964478015899655,
      "epoch": 1.1460488828909883,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048771649193770554,
      "loss": 5.8139,
      "mean_token_accuracy": 0.16508594900369644,
      "num_tokens": 22430746.0,
      "step": 11465
    },
    {
      "entropy": 6.139988899230957,
      "epoch": 1.1465487079522168,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004877048584304004,
      "loss": 5.9214,
      "mean_token_accuracy": 0.15477398037910461,
      "num_tokens": 22440812.0,
      "step": 11470
    },
    {
      "entropy": 6.2163482189178465,
      "epoch": 1.1470485330134452,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048769321957149557,
      "loss": 6.0627,
      "mean_token_accuracy": 0.14305093809962272,
      "num_tokens": 22450460.0,
      "step": 11475
    },
    {
      "entropy": 6.15441632270813,
      "epoch": 1.147548358074674,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004876815753612843,
      "loss": 5.9191,
      "mean_token_accuracy": 0.15828651040792466,
      "num_tokens": 22460660.0,
      "step": 11480
    },
    {
      "entropy": 6.117127704620361,
      "epoch": 1.1480481831359024,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004876699258000592,
      "loss": 5.8484,
      "mean_token_accuracy": 0.16004034876823425,
      "num_tokens": 22469238.0,
      "step": 11485
    },
    {
      "entropy": 6.089229393005371,
      "epoch": 1.148548008197131,
      "grad_norm": 1.375,
      "learning_rate": 0.0004876582708881135,
      "loss": 5.8919,
      "mean_token_accuracy": 0.1520943135023117,
      "num_tokens": 22479685.0,
      "step": 11490
    },
    {
      "entropy": 6.101541519165039,
      "epoch": 1.1490478332583596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004876466106257404,
      "loss": 5.8746,
      "mean_token_accuracy": 0.15702069252729417,
      "num_tokens": 22489770.0,
      "step": 11495
    },
    {
      "entropy": 6.227074670791626,
      "epoch": 1.149547658319588,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048763494501323333,
      "loss": 6.0422,
      "mean_token_accuracy": 0.14994479939341546,
      "num_tokens": 22499957.0,
      "step": 11500
    },
    {
      "entropy": 6.0912751197814945,
      "epoch": 1.1500474833808167,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048762327405088573,
      "loss": 5.8444,
      "mean_token_accuracy": 0.16363071501255036,
      "num_tokens": 22508863.0,
      "step": 11505
    },
    {
      "entropy": 6.194194602966308,
      "epoch": 1.1505473084420452,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004876115977389913,
      "loss": 5.9218,
      "mean_token_accuracy": 0.1516316130757332,
      "num_tokens": 22517563.0,
      "step": 11510
    },
    {
      "entropy": 6.063374948501587,
      "epoch": 1.151047133503274,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048759991607784375,
      "loss": 5.7861,
      "mean_token_accuracy": 0.17359053045511247,
      "num_tokens": 22527907.0,
      "step": 11515
    },
    {
      "entropy": 6.033289194107056,
      "epoch": 1.1515469585645024,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048758822906773706,
      "loss": 5.8387,
      "mean_token_accuracy": 0.1652185320854187,
      "num_tokens": 22537373.0,
      "step": 11520
    },
    {
      "entropy": 6.139695024490356,
      "epoch": 1.1520467836257309,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048757653670896523,
      "loss": 5.9776,
      "mean_token_accuracy": 0.1563122510910034,
      "num_tokens": 22547851.0,
      "step": 11525
    },
    {
      "entropy": 6.2897301197052,
      "epoch": 1.1525466086869596,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004875648390018224,
      "loss": 5.9603,
      "mean_token_accuracy": 0.1579339623451233,
      "num_tokens": 22558127.0,
      "step": 11530
    },
    {
      "entropy": 6.13180570602417,
      "epoch": 1.153046433748188,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000487553135946603,
      "loss": 5.9848,
      "mean_token_accuracy": 0.15117980912327766,
      "num_tokens": 22568151.0,
      "step": 11535
    },
    {
      "entropy": 6.138003778457642,
      "epoch": 1.1535462588094167,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048754142754360135,
      "loss": 6.0592,
      "mean_token_accuracy": 0.1499926522374153,
      "num_tokens": 22579772.0,
      "step": 11540
    },
    {
      "entropy": 6.16640157699585,
      "epoch": 1.1540460838706452,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004875297137931121,
      "loss": 5.9397,
      "mean_token_accuracy": 0.15832630544900894,
      "num_tokens": 22589551.0,
      "step": 11545
    },
    {
      "entropy": 6.166927719116211,
      "epoch": 1.154545908931874,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004875179946954298,
      "loss": 5.9474,
      "mean_token_accuracy": 0.1573859468102455,
      "num_tokens": 22599139.0,
      "step": 11550
    },
    {
      "entropy": 6.225337028503418,
      "epoch": 1.1550457339931024,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048750627025084955,
      "loss": 6.0752,
      "mean_token_accuracy": 0.14671857059001922,
      "num_tokens": 22609413.0,
      "step": 11555
    },
    {
      "entropy": 6.2118078708648685,
      "epoch": 1.1555455590543309,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004874945404596662,
      "loss": 6.0739,
      "mean_token_accuracy": 0.149736175686121,
      "num_tokens": 22619395.0,
      "step": 11560
    },
    {
      "entropy": 6.119500970840454,
      "epoch": 1.1560453841155596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048748280532217485,
      "loss": 5.873,
      "mean_token_accuracy": 0.16254231929779053,
      "num_tokens": 22629623.0,
      "step": 11565
    },
    {
      "entropy": 6.106204700469971,
      "epoch": 1.156545209176788,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004874710648386709,
      "loss": 5.9019,
      "mean_token_accuracy": 0.1501620218157768,
      "num_tokens": 22639735.0,
      "step": 11570
    },
    {
      "entropy": 6.126580715179443,
      "epoch": 1.1570450342380167,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004874593190094496,
      "loss": 5.8246,
      "mean_token_accuracy": 0.15912945568561554,
      "num_tokens": 22648563.0,
      "step": 11575
    },
    {
      "entropy": 6.089101839065552,
      "epoch": 1.1575448592992452,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004874475678348065,
      "loss": 5.8588,
      "mean_token_accuracy": 0.16259310096502305,
      "num_tokens": 22657137.0,
      "step": 11580
    },
    {
      "entropy": 6.10340051651001,
      "epoch": 1.158044684360474,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048743581131503716,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15471033602952958,
      "num_tokens": 22667987.0,
      "step": 11585
    },
    {
      "entropy": 6.0946660995483395,
      "epoch": 1.1585445094217024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004874240494504376,
      "loss": 5.9803,
      "mean_token_accuracy": 0.1506967782974243,
      "num_tokens": 22678830.0,
      "step": 11590
    },
    {
      "entropy": 6.251873636245728,
      "epoch": 1.1590443344829309,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048741228224130357,
      "loss": 5.9691,
      "mean_token_accuracy": 0.15562665909528733,
      "num_tokens": 22690568.0,
      "step": 11595
    },
    {
      "entropy": 6.205672645568848,
      "epoch": 1.1595441595441596,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048740050968793116,
      "loss": 5.9795,
      "mean_token_accuracy": 0.1545748621225357,
      "num_tokens": 22700150.0,
      "step": 11600
    },
    {
      "entropy": 6.030290508270264,
      "epoch": 1.160043984605388,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004873887317906166,
      "loss": 5.8807,
      "mean_token_accuracy": 0.15282610505819322,
      "num_tokens": 22709365.0,
      "step": 11605
    },
    {
      "entropy": 6.236346769332886,
      "epoch": 1.1605438096666167,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048737694854965617,
      "loss": 6.0389,
      "mean_token_accuracy": 0.15078128203749658,
      "num_tokens": 22718636.0,
      "step": 11610
    },
    {
      "entropy": 6.249845027923584,
      "epoch": 1.1610436347278452,
      "grad_norm": 1.75,
      "learning_rate": 0.00048736515996534644,
      "loss": 6.0214,
      "mean_token_accuracy": 0.15172371566295623,
      "num_tokens": 22728713.0,
      "step": 11615
    },
    {
      "entropy": 6.0499663829803465,
      "epoch": 1.161543459789074,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004873533660379839,
      "loss": 5.7814,
      "mean_token_accuracy": 0.16726926118135452,
      "num_tokens": 22739061.0,
      "step": 11620
    },
    {
      "entropy": 6.195918989181519,
      "epoch": 1.1620432848503024,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048734156676786525,
      "loss": 5.9825,
      "mean_token_accuracy": 0.14829165786504744,
      "num_tokens": 22748752.0,
      "step": 11625
    },
    {
      "entropy": 6.10856523513794,
      "epoch": 1.1625431099115309,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004873297621552875,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16870082318782806,
      "num_tokens": 22758695.0,
      "step": 11630
    },
    {
      "entropy": 6.213428735733032,
      "epoch": 1.1630429349727596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048731795220054755,
      "loss": 6.0129,
      "mean_token_accuracy": 0.15455391854047776,
      "num_tokens": 22768948.0,
      "step": 11635
    },
    {
      "entropy": 6.1464197635650635,
      "epoch": 1.163542760033988,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004873061369039425,
      "loss": 5.9615,
      "mean_token_accuracy": 0.15550061464309692,
      "num_tokens": 22779446.0,
      "step": 11640
    },
    {
      "entropy": 6.070646715164185,
      "epoch": 1.1640425850952167,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004872943162657697,
      "loss": 5.8804,
      "mean_token_accuracy": 0.15846378356218338,
      "num_tokens": 22789654.0,
      "step": 11645
    },
    {
      "entropy": 6.178193664550781,
      "epoch": 1.1645424101564452,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004872824902863265,
      "loss": 5.988,
      "mean_token_accuracy": 0.15519210696220398,
      "num_tokens": 22798938.0,
      "step": 11650
    },
    {
      "entropy": 6.205599403381347,
      "epoch": 1.165042235217674,
      "grad_norm": 1.25,
      "learning_rate": 0.0004872706589659106,
      "loss": 5.9633,
      "mean_token_accuracy": 0.15547027438879013,
      "num_tokens": 22808279.0,
      "step": 11655
    },
    {
      "entropy": 6.109768390655518,
      "epoch": 1.1655420602789024,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048725882230481947,
      "loss": 5.8922,
      "mean_token_accuracy": 0.16145075261592864,
      "num_tokens": 22817378.0,
      "step": 11660
    },
    {
      "entropy": 6.267738389968872,
      "epoch": 1.1660418853401309,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048724698030335096,
      "loss": 5.994,
      "mean_token_accuracy": 0.14961753487586976,
      "num_tokens": 22828652.0,
      "step": 11665
    },
    {
      "entropy": 6.145955181121826,
      "epoch": 1.1665417104013596,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004872351329618031,
      "loss": 5.947,
      "mean_token_accuracy": 0.16081907153129577,
      "num_tokens": 22839156.0,
      "step": 11670
    },
    {
      "entropy": 6.184478807449341,
      "epoch": 1.167041535462588,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048722328028047386,
      "loss": 6.0175,
      "mean_token_accuracy": 0.15380825847387314,
      "num_tokens": 22849004.0,
      "step": 11675
    },
    {
      "entropy": 6.144978284835815,
      "epoch": 1.1675413605238167,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00048721142225966146,
      "loss": 5.9943,
      "mean_token_accuracy": 0.14920471459627152,
      "num_tokens": 22858984.0,
      "step": 11680
    },
    {
      "entropy": 6.121554899215698,
      "epoch": 1.1680411855850452,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004871995588996644,
      "loss": 5.8825,
      "mean_token_accuracy": 0.16054070591926575,
      "num_tokens": 22869149.0,
      "step": 11685
    },
    {
      "entropy": 6.213811826705933,
      "epoch": 1.168541010646274,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004871876902007809,
      "loss": 6.0419,
      "mean_token_accuracy": 0.15133528038859367,
      "num_tokens": 22878522.0,
      "step": 11690
    },
    {
      "entropy": 6.180964803695678,
      "epoch": 1.1690408357075024,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048717581616330985,
      "loss": 5.9237,
      "mean_token_accuracy": 0.15596554428339005,
      "num_tokens": 22888356.0,
      "step": 11695
    },
    {
      "entropy": 6.160168790817261,
      "epoch": 1.1695406607687309,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048716393678754975,
      "loss": 5.9325,
      "mean_token_accuracy": 0.15649837106466294,
      "num_tokens": 22897551.0,
      "step": 11700
    },
    {
      "entropy": 6.201762914657593,
      "epoch": 1.1700404858299596,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004871520520737996,
      "loss": 5.8999,
      "mean_token_accuracy": 0.1590888246893883,
      "num_tokens": 22906480.0,
      "step": 11705
    },
    {
      "entropy": 6.117210483551025,
      "epoch": 1.170540310891188,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004871401620223585,
      "loss": 5.806,
      "mean_token_accuracy": 0.1601024404168129,
      "num_tokens": 22916544.0,
      "step": 11710
    },
    {
      "entropy": 6.051312208175659,
      "epoch": 1.1710401359524167,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004871282666335255,
      "loss": 5.838,
      "mean_token_accuracy": 0.16646503210067748,
      "num_tokens": 22926083.0,
      "step": 11715
    },
    {
      "entropy": 6.225358390808106,
      "epoch": 1.1715399610136452,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048711636590759985,
      "loss": 5.98,
      "mean_token_accuracy": 0.1536261223256588,
      "num_tokens": 22936127.0,
      "step": 11720
    },
    {
      "entropy": 6.183586120605469,
      "epoch": 1.172039786074874,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048710445984488106,
      "loss": 5.9121,
      "mean_token_accuracy": 0.15770825743675232,
      "num_tokens": 22946579.0,
      "step": 11725
    },
    {
      "entropy": 6.152203941345215,
      "epoch": 1.1725396111361024,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004870925484456686,
      "loss": 5.9291,
      "mean_token_accuracy": 0.15968383252620696,
      "num_tokens": 22956286.0,
      "step": 11730
    },
    {
      "entropy": 6.0786479949951175,
      "epoch": 1.1730394361973309,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004870806317102622,
      "loss": 5.9533,
      "mean_token_accuracy": 0.1588772267103195,
      "num_tokens": 22965625.0,
      "step": 11735
    },
    {
      "entropy": 6.232479047775269,
      "epoch": 1.1735392612585596,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004870687096389617,
      "loss": 6.0103,
      "mean_token_accuracy": 0.15224907100200652,
      "num_tokens": 22975853.0,
      "step": 11740
    },
    {
      "entropy": 6.203512573242188,
      "epoch": 1.174039086319788,
      "grad_norm": 1.4375,
      "learning_rate": 0.000487056782232067,
      "loss": 5.9744,
      "mean_token_accuracy": 0.15265572369098662,
      "num_tokens": 22985928.0,
      "step": 11745
    },
    {
      "entropy": 6.139250183105469,
      "epoch": 1.1745389113810167,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048704484948987824,
      "loss": 5.9478,
      "mean_token_accuracy": 0.15957126766443253,
      "num_tokens": 22995222.0,
      "step": 11750
    },
    {
      "entropy": 6.131657457351684,
      "epoch": 1.1750387364422452,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004870329114126956,
      "loss": 5.9021,
      "mean_token_accuracy": 0.1558387815952301,
      "num_tokens": 23005735.0,
      "step": 11755
    },
    {
      "entropy": 6.14085283279419,
      "epoch": 1.1755385615034737,
      "grad_norm": 1.5,
      "learning_rate": 0.0004870209680008196,
      "loss": 5.8862,
      "mean_token_accuracy": 0.15256378352642058,
      "num_tokens": 23016198.0,
      "step": 11760
    },
    {
      "entropy": 6.154154443740845,
      "epoch": 1.1760383865647024,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004870090192545505,
      "loss": 5.9631,
      "mean_token_accuracy": 0.15193866193294525,
      "num_tokens": 23025927.0,
      "step": 11765
    },
    {
      "entropy": 6.123627805709839,
      "epoch": 1.1765382116259309,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000486997065174189,
      "loss": 5.7699,
      "mean_token_accuracy": 0.16785600930452346,
      "num_tokens": 23034626.0,
      "step": 11770
    },
    {
      "entropy": 6.05954794883728,
      "epoch": 1.1770380366871596,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048698510576003593,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15829506069421767,
      "num_tokens": 23045773.0,
      "step": 11775
    },
    {
      "entropy": 6.139459848403931,
      "epoch": 1.177537861748388,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048697314101239217,
      "loss": 5.8639,
      "mean_token_accuracy": 0.1559164971113205,
      "num_tokens": 23056803.0,
      "step": 11780
    },
    {
      "entropy": 6.170506858825684,
      "epoch": 1.1780376868096165,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004869611709315587,
      "loss": 5.9219,
      "mean_token_accuracy": 0.15472777485847472,
      "num_tokens": 23066513.0,
      "step": 11785
    },
    {
      "entropy": 6.162024259567261,
      "epoch": 1.1785375118708452,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048694919551783665,
      "loss": 5.9584,
      "mean_token_accuracy": 0.15121255367994307,
      "num_tokens": 23077309.0,
      "step": 11790
    },
    {
      "entropy": 6.02949275970459,
      "epoch": 1.1790373369320737,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048693721477152755,
      "loss": 5.901,
      "mean_token_accuracy": 0.1658010944724083,
      "num_tokens": 23088292.0,
      "step": 11795
    },
    {
      "entropy": 6.150199127197266,
      "epoch": 1.1795371619933024,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048692522869293246,
      "loss": 5.8983,
      "mean_token_accuracy": 0.15488027185201644,
      "num_tokens": 23098744.0,
      "step": 11800
    },
    {
      "entropy": 6.1170916080474855,
      "epoch": 1.1800369870545309,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004869132372823533,
      "loss": 5.8707,
      "mean_token_accuracy": 0.16533601582050322,
      "num_tokens": 23108223.0,
      "step": 11805
    },
    {
      "entropy": 6.151248598098755,
      "epoch": 1.1805368121157596,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048690124054009157,
      "loss": 5.9663,
      "mean_token_accuracy": 0.15317804142832755,
      "num_tokens": 23118595.0,
      "step": 11810
    },
    {
      "entropy": 6.114837217330932,
      "epoch": 1.181036637176988,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004868892384664491,
      "loss": 5.8062,
      "mean_token_accuracy": 0.16875326484441758,
      "num_tokens": 23127640.0,
      "step": 11815
    },
    {
      "entropy": 6.148391914367676,
      "epoch": 1.1815364622382165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048687723106172805,
      "loss": 5.9303,
      "mean_token_accuracy": 0.1543321892619133,
      "num_tokens": 23138409.0,
      "step": 11820
    },
    {
      "entropy": 6.133092403411865,
      "epoch": 1.1820362872994452,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048686521832623026,
      "loss": 5.9285,
      "mean_token_accuracy": 0.15853227823972701,
      "num_tokens": 23148427.0,
      "step": 11825
    },
    {
      "entropy": 6.107066297531128,
      "epoch": 1.1825361123606737,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004868532002602581,
      "loss": 5.9234,
      "mean_token_accuracy": 0.16216319501399995,
      "num_tokens": 23157873.0,
      "step": 11830
    },
    {
      "entropy": 6.0920837879180905,
      "epoch": 1.1830359374219024,
      "grad_norm": 1.34375,
      "learning_rate": 0.000486841176864114,
      "loss": 5.849,
      "mean_token_accuracy": 0.1616566389799118,
      "num_tokens": 23166488.0,
      "step": 11835
    },
    {
      "entropy": 6.093384170532227,
      "epoch": 1.1835357624831309,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048682914813810035,
      "loss": 5.8084,
      "mean_token_accuracy": 0.16753440648317336,
      "num_tokens": 23176548.0,
      "step": 11840
    },
    {
      "entropy": 6.080803108215332,
      "epoch": 1.1840355875443596,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048681711408251986,
      "loss": 5.8934,
      "mean_token_accuracy": 0.15499286949634553,
      "num_tokens": 23186269.0,
      "step": 11845
    },
    {
      "entropy": 6.100447750091552,
      "epoch": 1.184535412605588,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048680507469767533,
      "loss": 5.849,
      "mean_token_accuracy": 0.1630789592862129,
      "num_tokens": 23197181.0,
      "step": 11850
    },
    {
      "entropy": 6.143448066711426,
      "epoch": 1.1850352376668165,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004867930299838696,
      "loss": 5.8386,
      "mean_token_accuracy": 0.16788185089826585,
      "num_tokens": 23207391.0,
      "step": 11855
    },
    {
      "entropy": 6.276797437667847,
      "epoch": 1.1855350627280452,
      "grad_norm": 1.640625,
      "learning_rate": 0.00048678097994140566,
      "loss": 6.0885,
      "mean_token_accuracy": 0.15211521089076996,
      "num_tokens": 23217510.0,
      "step": 11860
    },
    {
      "entropy": 6.032604551315307,
      "epoch": 1.1860348877892737,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048676892457058687,
      "loss": 5.8109,
      "mean_token_accuracy": 0.1659929484128952,
      "num_tokens": 23226746.0,
      "step": 11865
    },
    {
      "entropy": 6.103962135314942,
      "epoch": 1.1865347128505024,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004867568638717164,
      "loss": 5.9666,
      "mean_token_accuracy": 0.15252401754260064,
      "num_tokens": 23236333.0,
      "step": 11870
    },
    {
      "entropy": 6.103334569931031,
      "epoch": 1.1870345379117309,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048674479784509766,
      "loss": 5.7674,
      "mean_token_accuracy": 0.16894800812005997,
      "num_tokens": 23245071.0,
      "step": 11875
    },
    {
      "entropy": 6.095103979110718,
      "epoch": 1.1875343629729596,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048673272649103435,
      "loss": 5.8701,
      "mean_token_accuracy": 0.161318339407444,
      "num_tokens": 23254225.0,
      "step": 11880
    },
    {
      "entropy": 6.12320728302002,
      "epoch": 1.188034188034188,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048672064980983013,
      "loss": 5.865,
      "mean_token_accuracy": 0.1625265136361122,
      "num_tokens": 23264822.0,
      "step": 11885
    },
    {
      "entropy": 6.174591064453125,
      "epoch": 1.1885340130954165,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048670856780178874,
      "loss": 6.0039,
      "mean_token_accuracy": 0.15414587706327437,
      "num_tokens": 23274559.0,
      "step": 11890
    },
    {
      "entropy": 6.1506813049316404,
      "epoch": 1.1890338381566452,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048669648046721436,
      "loss": 5.9232,
      "mean_token_accuracy": 0.16194604858756065,
      "num_tokens": 23284431.0,
      "step": 11895
    },
    {
      "entropy": 6.149059772491455,
      "epoch": 1.1895336632178737,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000486684387806411,
      "loss": 5.9773,
      "mean_token_accuracy": 0.15229665786027907,
      "num_tokens": 23294814.0,
      "step": 11900
    },
    {
      "entropy": 6.188577795028687,
      "epoch": 1.1900334882791024,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048667228981968285,
      "loss": 5.9174,
      "mean_token_accuracy": 0.15563267469406128,
      "num_tokens": 23304701.0,
      "step": 11905
    },
    {
      "entropy": 6.264931344985962,
      "epoch": 1.1905333133403309,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048666018650733446,
      "loss": 5.9555,
      "mean_token_accuracy": 0.15236850306391717,
      "num_tokens": 23314335.0,
      "step": 11910
    },
    {
      "entropy": 6.162019920349121,
      "epoch": 1.1910331384015596,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004866480778696702,
      "loss": 6.0068,
      "mean_token_accuracy": 0.15393581837415696,
      "num_tokens": 23324591.0,
      "step": 11915
    },
    {
      "entropy": 6.080836963653565,
      "epoch": 1.191532963462788,
      "grad_norm": 1.25,
      "learning_rate": 0.00048663596390699477,
      "loss": 5.9042,
      "mean_token_accuracy": 0.16133854836225509,
      "num_tokens": 23334682.0,
      "step": 11920
    },
    {
      "entropy": 6.169084978103638,
      "epoch": 1.1920327885240165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004866238446196129,
      "loss": 5.9182,
      "mean_token_accuracy": 0.15725234448909758,
      "num_tokens": 23343943.0,
      "step": 11925
    },
    {
      "entropy": 6.130331420898438,
      "epoch": 1.1925326135852452,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004866117200078296,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15819316059350969,
      "num_tokens": 23354031.0,
      "step": 11930
    },
    {
      "entropy": 6.205503606796265,
      "epoch": 1.1930324386464737,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048659959007194985,
      "loss": 5.9935,
      "mean_token_accuracy": 0.15126484185457229,
      "num_tokens": 23363922.0,
      "step": 11935
    },
    {
      "entropy": 6.136963558197022,
      "epoch": 1.1935322637077024,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048658745481227884,
      "loss": 5.9696,
      "mean_token_accuracy": 0.14811476692557335,
      "num_tokens": 23372552.0,
      "step": 11940
    },
    {
      "entropy": 6.1538183212280275,
      "epoch": 1.1940320887689309,
      "grad_norm": 1.296875,
      "learning_rate": 0.000486575314229122,
      "loss": 5.8947,
      "mean_token_accuracy": 0.16712473183870316,
      "num_tokens": 23383454.0,
      "step": 11945
    },
    {
      "entropy": 6.187349033355713,
      "epoch": 1.1945319138301596,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004865631683227846,
      "loss": 5.9292,
      "mean_token_accuracy": 0.15424201488494874,
      "num_tokens": 23393429.0,
      "step": 11950
    },
    {
      "entropy": 6.104176139831543,
      "epoch": 1.195031738891388,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00048655101709357243,
      "loss": 5.8826,
      "mean_token_accuracy": 0.15551133453845978,
      "num_tokens": 23403742.0,
      "step": 11955
    },
    {
      "entropy": 6.0717264175415036,
      "epoch": 1.1955315639526165,
      "grad_norm": 1.5,
      "learning_rate": 0.000486538860541791,
      "loss": 5.8342,
      "mean_token_accuracy": 0.16000038087368013,
      "num_tokens": 23412875.0,
      "step": 11960
    },
    {
      "entropy": 6.141061210632325,
      "epoch": 1.1960313890138452,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048652669866774636,
      "loss": 5.9461,
      "mean_token_accuracy": 0.15706803500652314,
      "num_tokens": 23422838.0,
      "step": 11965
    },
    {
      "entropy": 6.2078554153442385,
      "epoch": 1.1965312140750737,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004865145314717443,
      "loss": 5.8965,
      "mean_token_accuracy": 0.15291520208120346,
      "num_tokens": 23431407.0,
      "step": 11970
    },
    {
      "entropy": 6.105554056167603,
      "epoch": 1.1970310391363024,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048650235895409115,
      "loss": 5.882,
      "mean_token_accuracy": 0.1668234571814537,
      "num_tokens": 23441580.0,
      "step": 11975
    },
    {
      "entropy": 6.088036966323853,
      "epoch": 1.1975308641975309,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004864901811150931,
      "loss": 5.9639,
      "mean_token_accuracy": 0.15468232110142707,
      "num_tokens": 23451508.0,
      "step": 11980
    },
    {
      "entropy": 6.15867166519165,
      "epoch": 1.1980306892587596,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004864779979550564,
      "loss": 5.9558,
      "mean_token_accuracy": 0.156481471657753,
      "num_tokens": 23461928.0,
      "step": 11985
    },
    {
      "entropy": 6.185863590240478,
      "epoch": 1.198530514319988,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004864658094742878,
      "loss": 5.8818,
      "mean_token_accuracy": 0.16178420633077623,
      "num_tokens": 23471501.0,
      "step": 11990
    },
    {
      "entropy": 6.22927508354187,
      "epoch": 1.1990303393812165,
      "grad_norm": 1.453125,
      "learning_rate": 0.00048645361567309375,
      "loss": 6.0367,
      "mean_token_accuracy": 0.1582635261118412,
      "num_tokens": 23481478.0,
      "step": 11995
    },
    {
      "entropy": 6.180087614059448,
      "epoch": 1.1995301644424452,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004864414165517812,
      "loss": 5.9858,
      "mean_token_accuracy": 0.15466733351349832,
      "num_tokens": 23491253.0,
      "step": 12000
    },
    {
      "epoch": 1.1995301644424452,
      "eval_entropy": 5.923490237341932,
      "eval_loss": 6.036754608154297,
      "eval_mean_token_accuracy": 0.16042811631977835,
      "eval_num_tokens": 23491253.0,
      "eval_runtime": 23.8632,
      "eval_samples_per_second": 1300.958,
      "eval_steps_per_second": 162.635,
      "step": 12000
    },
    {
      "entropy": 6.143440246582031,
      "epoch": 1.2000299895036737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000486429212110657,
      "loss": 5.974,
      "mean_token_accuracy": 0.1541808433830738,
      "num_tokens": 23500753.0,
      "step": 12005
    },
    {
      "entropy": 6.202025556564331,
      "epoch": 1.2005298145649024,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004864170023500282,
      "loss": 5.9989,
      "mean_token_accuracy": 0.14976682364940644,
      "num_tokens": 23509837.0,
      "step": 12010
    },
    {
      "entropy": 6.141349983215332,
      "epoch": 1.2010296396261309,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004864047872702021,
      "loss": 5.8879,
      "mean_token_accuracy": 0.1568389616906643,
      "num_tokens": 23519926.0,
      "step": 12015
    },
    {
      "entropy": 6.107258701324463,
      "epoch": 1.2015294646873593,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004863925668714859,
      "loss": 6.0677,
      "mean_token_accuracy": 0.14842357486486435,
      "num_tokens": 23531165.0,
      "step": 12020
    },
    {
      "entropy": 6.17024507522583,
      "epoch": 1.202029289748588,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004863803411541871,
      "loss": 5.9056,
      "mean_token_accuracy": 0.1597069025039673,
      "num_tokens": 23541696.0,
      "step": 12025
    },
    {
      "entropy": 6.201007604598999,
      "epoch": 1.2025291148098165,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004863681101186134,
      "loss": 5.9277,
      "mean_token_accuracy": 0.15260101705789567,
      "num_tokens": 23552382.0,
      "step": 12030
    },
    {
      "entropy": 6.144422149658203,
      "epoch": 1.2030289398710452,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004863558737650723,
      "loss": 5.9991,
      "mean_token_accuracy": 0.15438851416110994,
      "num_tokens": 23561444.0,
      "step": 12035
    },
    {
      "entropy": 6.103683614730835,
      "epoch": 1.2035287649322737,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004863436320938719,
      "loss": 5.9105,
      "mean_token_accuracy": 0.15528724938631058,
      "num_tokens": 23570985.0,
      "step": 12040
    },
    {
      "entropy": 6.043612432479859,
      "epoch": 1.2040285899935022,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048633138510532003,
      "loss": 5.8322,
      "mean_token_accuracy": 0.1686202332377434,
      "num_tokens": 23581291.0,
      "step": 12045
    },
    {
      "entropy": 6.210238885879517,
      "epoch": 1.2045284150547308,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048631913279972497,
      "loss": 5.9939,
      "mean_token_accuracy": 0.15565201342105867,
      "num_tokens": 23591087.0,
      "step": 12050
    },
    {
      "entropy": 6.146289730072022,
      "epoch": 1.2050282401159593,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048630687517739484,
      "loss": 6.0009,
      "mean_token_accuracy": 0.14873012602329255,
      "num_tokens": 23601635.0,
      "step": 12055
    },
    {
      "entropy": 6.22243070602417,
      "epoch": 1.205528065177188,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004862946122386381,
      "loss": 5.9478,
      "mean_token_accuracy": 0.1590518295764923,
      "num_tokens": 23612523.0,
      "step": 12060
    },
    {
      "entropy": 6.1395985126495365,
      "epoch": 1.2060278902384165,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048628234398376334,
      "loss": 5.9532,
      "mean_token_accuracy": 0.1569541722536087,
      "num_tokens": 23622258.0,
      "step": 12065
    },
    {
      "entropy": 6.03805103302002,
      "epoch": 1.2065277152996452,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048627007041307914,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16446440666913986,
      "num_tokens": 23631198.0,
      "step": 12070
    },
    {
      "entropy": 6.157372617721558,
      "epoch": 1.2070275403608737,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048625779152689435,
      "loss": 5.9542,
      "mean_token_accuracy": 0.15221483558416365,
      "num_tokens": 23641052.0,
      "step": 12075
    },
    {
      "entropy": 6.1856677532196045,
      "epoch": 1.2075273654221022,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004862455073255179,
      "loss": 5.9067,
      "mean_token_accuracy": 0.15828054696321486,
      "num_tokens": 23650785.0,
      "step": 12080
    },
    {
      "entropy": 6.110441732406616,
      "epoch": 1.2080271904833308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048623321780925883,
      "loss": 5.9419,
      "mean_token_accuracy": 0.15526666194200517,
      "num_tokens": 23660548.0,
      "step": 12085
    },
    {
      "entropy": 6.033561134338379,
      "epoch": 1.2085270155445593,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048622092297842637,
      "loss": 5.8067,
      "mean_token_accuracy": 0.16373466402292253,
      "num_tokens": 23669979.0,
      "step": 12090
    },
    {
      "entropy": 6.174818086624145,
      "epoch": 1.209026840605788,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004862086228333298,
      "loss": 5.9806,
      "mean_token_accuracy": 0.1489489808678627,
      "num_tokens": 23679740.0,
      "step": 12095
    },
    {
      "entropy": 6.223101806640625,
      "epoch": 1.2095266656670165,
      "grad_norm": 1.375,
      "learning_rate": 0.00048619631737427865,
      "loss": 5.9713,
      "mean_token_accuracy": 0.1611980602145195,
      "num_tokens": 23689688.0,
      "step": 12100
    },
    {
      "entropy": 6.164903545379639,
      "epoch": 1.2100264907282452,
      "grad_norm": 1.671875,
      "learning_rate": 0.0004861840066015825,
      "loss": 5.88,
      "mean_token_accuracy": 0.1636110857129097,
      "num_tokens": 23699550.0,
      "step": 12105
    },
    {
      "entropy": 6.152115392684936,
      "epoch": 1.2105263157894737,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048617169051555113,
      "loss": 5.9662,
      "mean_token_accuracy": 0.15651750415563584,
      "num_tokens": 23709114.0,
      "step": 12110
    },
    {
      "entropy": 6.182860279083252,
      "epoch": 1.2110261408507021,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004861593691164942,
      "loss": 5.9575,
      "mean_token_accuracy": 0.15342798531055452,
      "num_tokens": 23718758.0,
      "step": 12115
    },
    {
      "entropy": 6.204360389709473,
      "epoch": 1.2115259659119308,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004861470424047221,
      "loss": 5.9055,
      "mean_token_accuracy": 0.16160933673381805,
      "num_tokens": 23728175.0,
      "step": 12120
    },
    {
      "entropy": 5.968841123580932,
      "epoch": 1.2120257909731593,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004861347103805446,
      "loss": 5.7477,
      "mean_token_accuracy": 0.1743275284767151,
      "num_tokens": 23738308.0,
      "step": 12125
    },
    {
      "entropy": 6.171118879318238,
      "epoch": 1.212525616034388,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004861223730442722,
      "loss": 5.9012,
      "mean_token_accuracy": 0.1530655413866043,
      "num_tokens": 23747867.0,
      "step": 12130
    },
    {
      "entropy": 6.090215253829956,
      "epoch": 1.2130254410956165,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048611003039621516,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1588980495929718,
      "num_tokens": 23758347.0,
      "step": 12135
    },
    {
      "entropy": 6.1678403377532955,
      "epoch": 1.2135252661568452,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004860976824366841,
      "loss": 5.9379,
      "mean_token_accuracy": 0.16392111480236055,
      "num_tokens": 23767847.0,
      "step": 12140
    },
    {
      "entropy": 6.174919557571411,
      "epoch": 1.2140250912180737,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048608532916598964,
      "loss": 5.9708,
      "mean_token_accuracy": 0.15442277640104293,
      "num_tokens": 23777098.0,
      "step": 12145
    },
    {
      "entropy": 6.163359689712524,
      "epoch": 1.2145249162793021,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048607297058444267,
      "loss": 5.963,
      "mean_token_accuracy": 0.16118056029081346,
      "num_tokens": 23787332.0,
      "step": 12150
    },
    {
      "entropy": 6.127607679367065,
      "epoch": 1.2150247413405308,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004860606066923541,
      "loss": 5.8777,
      "mean_token_accuracy": 0.16466124951839448,
      "num_tokens": 23796263.0,
      "step": 12155
    },
    {
      "entropy": 6.14617223739624,
      "epoch": 1.2155245664017593,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048604823749003485,
      "loss": 5.911,
      "mean_token_accuracy": 0.15743542313575745,
      "num_tokens": 23806051.0,
      "step": 12160
    },
    {
      "entropy": 6.106583786010742,
      "epoch": 1.216024391462988,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0004860358629777963,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16498946845531465,
      "num_tokens": 23814925.0,
      "step": 12165
    },
    {
      "entropy": 6.083284664154053,
      "epoch": 1.2165242165242165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004860234831559498,
      "loss": 5.8052,
      "mean_token_accuracy": 0.16558174788951874,
      "num_tokens": 23824785.0,
      "step": 12170
    },
    {
      "entropy": 6.172229862213134,
      "epoch": 1.2170240415854452,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048601109802480676,
      "loss": 5.9929,
      "mean_token_accuracy": 0.1472548358142376,
      "num_tokens": 23835102.0,
      "step": 12175
    },
    {
      "entropy": 6.176716375350952,
      "epoch": 1.2175238666466737,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048599870758467875,
      "loss": 5.9464,
      "mean_token_accuracy": 0.15777206867933274,
      "num_tokens": 23845718.0,
      "step": 12180
    },
    {
      "entropy": 6.131068325042724,
      "epoch": 1.2180236917079021,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048598631183587753,
      "loss": 5.8817,
      "mean_token_accuracy": 0.16344125270843507,
      "num_tokens": 23855078.0,
      "step": 12185
    },
    {
      "entropy": 6.175488567352295,
      "epoch": 1.2185235167691308,
      "grad_norm": 1.59375,
      "learning_rate": 0.00048597391077871504,
      "loss": 5.955,
      "mean_token_accuracy": 0.15968553274869918,
      "num_tokens": 23864287.0,
      "step": 12190
    },
    {
      "entropy": 6.2001941204071045,
      "epoch": 1.2190233418303593,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048596150441350324,
      "loss": 6.0496,
      "mean_token_accuracy": 0.14788421392440795,
      "num_tokens": 23874477.0,
      "step": 12195
    },
    {
      "entropy": 6.1178265571594235,
      "epoch": 1.219523166891588,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004859490927405543,
      "loss": 5.8163,
      "mean_token_accuracy": 0.1505747690796852,
      "num_tokens": 23884208.0,
      "step": 12200
    },
    {
      "entropy": 6.2836266040802,
      "epoch": 1.2200229919528165,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004859366757601804,
      "loss": 6.0522,
      "mean_token_accuracy": 0.15309324264526367,
      "num_tokens": 23894226.0,
      "step": 12205
    },
    {
      "entropy": 6.093864154815674,
      "epoch": 1.2205228170140452,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048592425347269405,
      "loss": 5.8641,
      "mean_token_accuracy": 0.16805485039949417,
      "num_tokens": 23904238.0,
      "step": 12210
    },
    {
      "entropy": 6.091812467575073,
      "epoch": 1.2210226420752737,
      "grad_norm": 1.546875,
      "learning_rate": 0.00048591182587840777,
      "loss": 5.8723,
      "mean_token_accuracy": 0.16053334921598433,
      "num_tokens": 23914611.0,
      "step": 12215
    },
    {
      "entropy": 6.144492053985596,
      "epoch": 1.2215224671365021,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004858993929776342,
      "loss": 5.9985,
      "mean_token_accuracy": 0.14830951690673827,
      "num_tokens": 23924270.0,
      "step": 12220
    },
    {
      "entropy": 6.143527507781982,
      "epoch": 1.2220222921977308,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048588695477068617,
      "loss": 5.896,
      "mean_token_accuracy": 0.15724143162369728,
      "num_tokens": 23932677.0,
      "step": 12225
    },
    {
      "entropy": 6.2154967308044435,
      "epoch": 1.2225221172589593,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048587451125787665,
      "loss": 5.9286,
      "mean_token_accuracy": 0.15063874423503876,
      "num_tokens": 23943259.0,
      "step": 12230
    },
    {
      "entropy": 6.198149538040161,
      "epoch": 1.223021942320188,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004858620624395187,
      "loss": 5.9793,
      "mean_token_accuracy": 0.15282555669546127,
      "num_tokens": 23953162.0,
      "step": 12235
    },
    {
      "entropy": 6.253126049041748,
      "epoch": 1.2235217673814165,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048584960831592547,
      "loss": 6.0202,
      "mean_token_accuracy": 0.14977033585309982,
      "num_tokens": 23961577.0,
      "step": 12240
    },
    {
      "entropy": 6.126883125305175,
      "epoch": 1.2240215924426452,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048583714888741035,
      "loss": 5.8278,
      "mean_token_accuracy": 0.16924876421689988,
      "num_tokens": 23969691.0,
      "step": 12245
    },
    {
      "entropy": 6.038387107849121,
      "epoch": 1.2245214175038737,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048582468415428687,
      "loss": 5.9055,
      "mean_token_accuracy": 0.15907573848962783,
      "num_tokens": 23980753.0,
      "step": 12250
    },
    {
      "entropy": 6.183107614517212,
      "epoch": 1.2250212425651021,
      "grad_norm": 1.5,
      "learning_rate": 0.00048581221411686857,
      "loss": 5.9774,
      "mean_token_accuracy": 0.15522108525037764,
      "num_tokens": 23991841.0,
      "step": 12255
    },
    {
      "entropy": 6.180950736999511,
      "epoch": 1.2255210676263308,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048579973877546916,
      "loss": 5.8767,
      "mean_token_accuracy": 0.16092007905244826,
      "num_tokens": 24000749.0,
      "step": 12260
    },
    {
      "entropy": 6.246607065200806,
      "epoch": 1.2260208926875593,
      "grad_norm": 1.7578125,
      "learning_rate": 0.00048578725813040263,
      "loss": 6.0764,
      "mean_token_accuracy": 0.1493350088596344,
      "num_tokens": 24010055.0,
      "step": 12265
    },
    {
      "entropy": 6.1432219505310055,
      "epoch": 1.226520717748788,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004857747721819829,
      "loss": 5.9425,
      "mean_token_accuracy": 0.15949156284332275,
      "num_tokens": 24019307.0,
      "step": 12270
    },
    {
      "entropy": 6.118947696685791,
      "epoch": 1.2270205428100165,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004857622809305241,
      "loss": 5.7952,
      "mean_token_accuracy": 0.16963599920272826,
      "num_tokens": 24028913.0,
      "step": 12275
    },
    {
      "entropy": 6.136103773117066,
      "epoch": 1.2275203678712452,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048574978437634057,
      "loss": 5.8192,
      "mean_token_accuracy": 0.16369618177413942,
      "num_tokens": 24038357.0,
      "step": 12280
    },
    {
      "entropy": 6.144922304153442,
      "epoch": 1.2280201929324737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004857372825197468,
      "loss": 5.885,
      "mean_token_accuracy": 0.16259193122386933,
      "num_tokens": 24048199.0,
      "step": 12285
    },
    {
      "entropy": 6.149617767333984,
      "epoch": 1.2285200179937021,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004857247753610571,
      "loss": 5.9656,
      "mean_token_accuracy": 0.16008811444044113,
      "num_tokens": 24058311.0,
      "step": 12290
    },
    {
      "entropy": 6.178594255447388,
      "epoch": 1.2290198430549308,
      "grad_norm": 1.34375,
      "learning_rate": 0.00048571226290058635,
      "loss": 5.9345,
      "mean_token_accuracy": 0.15250544100999833,
      "num_tokens": 24068471.0,
      "step": 12295
    },
    {
      "entropy": 6.168077993392944,
      "epoch": 1.2295196681161593,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004856997451386492,
      "loss": 5.9128,
      "mean_token_accuracy": 0.15999604612588883,
      "num_tokens": 24078118.0,
      "step": 12300
    },
    {
      "entropy": 6.174386119842529,
      "epoch": 1.2300194931773878,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004856872220755607,
      "loss": 5.9231,
      "mean_token_accuracy": 0.16301810443401338,
      "num_tokens": 24088517.0,
      "step": 12305
    },
    {
      "entropy": 6.2022782325744625,
      "epoch": 1.2305193182386165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.000485674693711636,
      "loss": 6.0139,
      "mean_token_accuracy": 0.1541484698653221,
      "num_tokens": 24097421.0,
      "step": 12310
    },
    {
      "entropy": 6.082457399368286,
      "epoch": 1.231019143299845,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004856621600471901,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16490086764097214,
      "num_tokens": 24106531.0,
      "step": 12315
    },
    {
      "entropy": 6.133380794525147,
      "epoch": 1.2315189683610737,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048564962108253853,
      "loss": 5.8929,
      "mean_token_accuracy": 0.16117538064718245,
      "num_tokens": 24115246.0,
      "step": 12320
    },
    {
      "entropy": 6.212507724761963,
      "epoch": 1.2320187934223021,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004856370768179967,
      "loss": 5.9846,
      "mean_token_accuracy": 0.15108444541692734,
      "num_tokens": 24126163.0,
      "step": 12325
    },
    {
      "entropy": 6.148311281204224,
      "epoch": 1.2325186184835308,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048562452725388024,
      "loss": 5.9421,
      "mean_token_accuracy": 0.15411113798618317,
      "num_tokens": 24135722.0,
      "step": 12330
    },
    {
      "entropy": 6.039162063598633,
      "epoch": 1.2330184435447593,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004856119723905048,
      "loss": 5.757,
      "mean_token_accuracy": 0.16728441715240477,
      "num_tokens": 24145278.0,
      "step": 12335
    },
    {
      "entropy": 6.136285829544067,
      "epoch": 1.2335182686059878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004855994122281864,
      "loss": 5.933,
      "mean_token_accuracy": 0.15703438371419906,
      "num_tokens": 24154710.0,
      "step": 12340
    },
    {
      "entropy": 6.157153034210205,
      "epoch": 1.2340180936672165,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048558684676724083,
      "loss": 5.9662,
      "mean_token_accuracy": 0.14584285765886307,
      "num_tokens": 24164373.0,
      "step": 12345
    },
    {
      "entropy": 6.114428377151489,
      "epoch": 1.234517918728445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048557427600798454,
      "loss": 5.8033,
      "mean_token_accuracy": 0.16437434107065202,
      "num_tokens": 24173085.0,
      "step": 12350
    },
    {
      "entropy": 6.133389234542847,
      "epoch": 1.2350177437896737,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004855616999507336,
      "loss": 5.922,
      "mean_token_accuracy": 0.1623062863945961,
      "num_tokens": 24182936.0,
      "step": 12355
    },
    {
      "entropy": 6.206312608718872,
      "epoch": 1.2355175688509021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048554911859580445,
      "loss": 5.9178,
      "mean_token_accuracy": 0.1538863644003868,
      "num_tokens": 24192407.0,
      "step": 12360
    },
    {
      "entropy": 6.263148021697998,
      "epoch": 1.2360173939121308,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004855365319435137,
      "loss": 5.9299,
      "mean_token_accuracy": 0.1604176089167595,
      "num_tokens": 24201770.0,
      "step": 12365
    },
    {
      "entropy": 6.0705146312713625,
      "epoch": 1.2365172189733593,
      "grad_norm": 1.25,
      "learning_rate": 0.000485523939994178,
      "loss": 5.8659,
      "mean_token_accuracy": 0.16046635955572128,
      "num_tokens": 24211287.0,
      "step": 12370
    },
    {
      "entropy": 6.066037034988403,
      "epoch": 1.2370170440345878,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004855113427481141,
      "loss": 5.8959,
      "mean_token_accuracy": 0.15989235490560533,
      "num_tokens": 24221277.0,
      "step": 12375
    },
    {
      "entropy": 6.163655948638916,
      "epoch": 1.2375168690958165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000485498740205639,
      "loss": 5.8121,
      "mean_token_accuracy": 0.16204496324062348,
      "num_tokens": 24231272.0,
      "step": 12380
    },
    {
      "entropy": 6.160795259475708,
      "epoch": 1.238016694157045,
      "grad_norm": 1.796875,
      "learning_rate": 0.00048548613236706976,
      "loss": 5.947,
      "mean_token_accuracy": 0.1588428571820259,
      "num_tokens": 24242073.0,
      "step": 12385
    },
    {
      "entropy": 5.980325698852539,
      "epoch": 1.2385165192182737,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004854735192327236,
      "loss": 5.7687,
      "mean_token_accuracy": 0.1730379968881607,
      "num_tokens": 24251928.0,
      "step": 12390
    },
    {
      "entropy": 6.052261543273926,
      "epoch": 1.2390163442795021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048546090080291783,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16469375938177108,
      "num_tokens": 24261146.0,
      "step": 12395
    },
    {
      "entropy": 6.176646041870117,
      "epoch": 1.2395161693407308,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048544827707797,
      "loss": 6.0087,
      "mean_token_accuracy": 0.15118223130702974,
      "num_tokens": 24271675.0,
      "step": 12400
    },
    {
      "entropy": 6.197798395156861,
      "epoch": 1.2400159944019593,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048543564805819764,
      "loss": 5.8749,
      "mean_token_accuracy": 0.15736810266971588,
      "num_tokens": 24280795.0,
      "step": 12405
    },
    {
      "entropy": 6.131864166259765,
      "epoch": 1.2405158194631878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004854230137439185,
      "loss": 5.9011,
      "mean_token_accuracy": 0.15908429622650147,
      "num_tokens": 24291204.0,
      "step": 12410
    },
    {
      "entropy": 6.114142274856567,
      "epoch": 1.2410156445244165,
      "grad_norm": 1.375,
      "learning_rate": 0.00048541037413545046,
      "loss": 5.8798,
      "mean_token_accuracy": 0.16358888447284697,
      "num_tokens": 24300926.0,
      "step": 12415
    },
    {
      "entropy": 6.085081577301025,
      "epoch": 1.241515469585645,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004853977292331116,
      "loss": 5.7744,
      "mean_token_accuracy": 0.1617984801530838,
      "num_tokens": 24310319.0,
      "step": 12420
    },
    {
      "entropy": 6.037611579895019,
      "epoch": 1.2420152946468737,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004853850790372201,
      "loss": 5.9363,
      "mean_token_accuracy": 0.15590060651302337,
      "num_tokens": 24321671.0,
      "step": 12425
    },
    {
      "entropy": 6.136835527420044,
      "epoch": 1.2425151197081021,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048537242354809403,
      "loss": 5.9856,
      "mean_token_accuracy": 0.1503479741513729,
      "num_tokens": 24330275.0,
      "step": 12430
    },
    {
      "entropy": 6.193158340454102,
      "epoch": 1.2430149447693308,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048535976276605186,
      "loss": 5.9108,
      "mean_token_accuracy": 0.15676185339689255,
      "num_tokens": 24339753.0,
      "step": 12435
    },
    {
      "entropy": 6.115321302413941,
      "epoch": 1.2435147698305593,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004853470966914123,
      "loss": 5.9097,
      "mean_token_accuracy": 0.15569395571947098,
      "num_tokens": 24349517.0,
      "step": 12440
    },
    {
      "entropy": 6.04531421661377,
      "epoch": 1.2440145948917878,
      "grad_norm": 1.375,
      "learning_rate": 0.0004853344253244939,
      "loss": 5.7996,
      "mean_token_accuracy": 0.16397407799959182,
      "num_tokens": 24358036.0,
      "step": 12445
    },
    {
      "entropy": 6.0908716201782225,
      "epoch": 1.2445144199530165,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048532174866561543,
      "loss": 5.7881,
      "mean_token_accuracy": 0.16707671880722047,
      "num_tokens": 24367488.0,
      "step": 12450
    },
    {
      "entropy": 6.2450621128082275,
      "epoch": 1.245014245014245,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048530906671509597,
      "loss": 6.0684,
      "mean_token_accuracy": 0.14787414222955703,
      "num_tokens": 24377053.0,
      "step": 12455
    },
    {
      "entropy": 6.197347784042359,
      "epoch": 1.2455140700754737,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048529637947325444,
      "loss": 5.9664,
      "mean_token_accuracy": 0.14754286259412766,
      "num_tokens": 24387245.0,
      "step": 12460
    },
    {
      "entropy": 6.228249740600586,
      "epoch": 1.2460138951367021,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004852836869404101,
      "loss": 5.9739,
      "mean_token_accuracy": 0.15413450449705124,
      "num_tokens": 24395901.0,
      "step": 12465
    },
    {
      "entropy": 6.145458030700683,
      "epoch": 1.2465137201979308,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004852709891168823,
      "loss": 5.865,
      "mean_token_accuracy": 0.16206178367137908,
      "num_tokens": 24405148.0,
      "step": 12470
    },
    {
      "entropy": 6.069533491134644,
      "epoch": 1.2470135452591593,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004852582860029906,
      "loss": 5.8347,
      "mean_token_accuracy": 0.16046560406684876,
      "num_tokens": 24415406.0,
      "step": 12475
    },
    {
      "entropy": 6.0369627475738525,
      "epoch": 1.2475133703203878,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048524557759905445,
      "loss": 5.8139,
      "mean_token_accuracy": 0.1653281643986702,
      "num_tokens": 24425702.0,
      "step": 12480
    },
    {
      "entropy": 6.174066209793091,
      "epoch": 1.2480131953816165,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004852328639053937,
      "loss": 5.9245,
      "mean_token_accuracy": 0.1507974810898304,
      "num_tokens": 24435702.0,
      "step": 12485
    },
    {
      "entropy": 6.197006750106811,
      "epoch": 1.248513020442845,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004852201449223282,
      "loss": 5.9521,
      "mean_token_accuracy": 0.1494345098733902,
      "num_tokens": 24445545.0,
      "step": 12490
    },
    {
      "entropy": 6.1598365783691404,
      "epoch": 1.2490128455040737,
      "grad_norm": 1.40625,
      "learning_rate": 0.000485207420650178,
      "loss": 5.9819,
      "mean_token_accuracy": 0.15773148462176323,
      "num_tokens": 24455269.0,
      "step": 12495
    },
    {
      "entropy": 6.160503768920899,
      "epoch": 1.2495126705653021,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004851946910892631,
      "loss": 5.929,
      "mean_token_accuracy": 0.16134475618600846,
      "num_tokens": 24465563.0,
      "step": 12500
    },
    {
      "entropy": 6.097579622268677,
      "epoch": 1.2500124956265308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004851819562399039,
      "loss": 5.9698,
      "mean_token_accuracy": 0.15840654373168944,
      "num_tokens": 24475837.0,
      "step": 12505
    },
    {
      "entropy": 6.24086709022522,
      "epoch": 1.2505123206877593,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004851692161024207,
      "loss": 6.0171,
      "mean_token_accuracy": 0.14837371334433555,
      "num_tokens": 24486384.0,
      "step": 12510
    },
    {
      "entropy": 6.227885293960571,
      "epoch": 1.2510121457489878,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048515647067713425,
      "loss": 6.0304,
      "mean_token_accuracy": 0.15465719103813172,
      "num_tokens": 24495840.0,
      "step": 12515
    },
    {
      "entropy": 6.112557983398437,
      "epoch": 1.2515119708102165,
      "grad_norm": 1.375,
      "learning_rate": 0.000485143719964365,
      "loss": 5.8764,
      "mean_token_accuracy": 0.15842797607183456,
      "num_tokens": 24505849.0,
      "step": 12520
    },
    {
      "entropy": 6.109187746047974,
      "epoch": 1.252011795871445,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004851309639644338,
      "loss": 5.9793,
      "mean_token_accuracy": 0.15028330609202384,
      "num_tokens": 24517440.0,
      "step": 12525
    },
    {
      "entropy": 6.118728399276733,
      "epoch": 1.2525116209326734,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004851182026776616,
      "loss": 5.8453,
      "mean_token_accuracy": 0.16685607880353928,
      "num_tokens": 24527602.0,
      "step": 12530
    },
    {
      "entropy": 6.097123289108277,
      "epoch": 1.2530114459939021,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004851054361043696,
      "loss": 5.8136,
      "mean_token_accuracy": 0.16851051151752472,
      "num_tokens": 24536770.0,
      "step": 12535
    },
    {
      "entropy": 6.124136161804199,
      "epoch": 1.2535112710551308,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048509266424487873,
      "loss": 5.933,
      "mean_token_accuracy": 0.1538613349199295,
      "num_tokens": 24546230.0,
      "step": 12540
    },
    {
      "entropy": 6.050775623321533,
      "epoch": 1.2540110961163593,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004850798870995106,
      "loss": 5.8765,
      "mean_token_accuracy": 0.15772414803504944,
      "num_tokens": 24555130.0,
      "step": 12545
    },
    {
      "entropy": 6.1660980701446535,
      "epoch": 1.2545109211775878,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048506710466858657,
      "loss": 5.9823,
      "mean_token_accuracy": 0.15560240149497986,
      "num_tokens": 24565167.0,
      "step": 12550
    },
    {
      "entropy": 6.138290357589722,
      "epoch": 1.2550107462388165,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048505431695242823,
      "loss": 5.9677,
      "mean_token_accuracy": 0.1618862971663475,
      "num_tokens": 24574545.0,
      "step": 12555
    },
    {
      "entropy": 6.170036697387696,
      "epoch": 1.255510571300045,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048504152395135726,
      "loss": 5.9833,
      "mean_token_accuracy": 0.15542703121900558,
      "num_tokens": 24585050.0,
      "step": 12560
    },
    {
      "entropy": 6.256245422363281,
      "epoch": 1.2560103963612734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048502872566569564,
      "loss": 6.0113,
      "mean_token_accuracy": 0.14865442514419555,
      "num_tokens": 24596334.0,
      "step": 12565
    },
    {
      "entropy": 6.15869107246399,
      "epoch": 1.2565102214225021,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048501592209576535,
      "loss": 5.9101,
      "mean_token_accuracy": 0.15852668583393098,
      "num_tokens": 24605947.0,
      "step": 12570
    },
    {
      "entropy": 6.0848160743713375,
      "epoch": 1.2570100464837308,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004850031132418885,
      "loss": 5.7945,
      "mean_token_accuracy": 0.17117061614990234,
      "num_tokens": 24614336.0,
      "step": 12575
    },
    {
      "entropy": 6.109485626220703,
      "epoch": 1.2575098715449593,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048499029910438734,
      "loss": 5.8887,
      "mean_token_accuracy": 0.16080235689878464,
      "num_tokens": 24624906.0,
      "step": 12580
    },
    {
      "entropy": 6.215049362182617,
      "epoch": 1.2580096966061878,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048497747968358424,
      "loss": 5.9384,
      "mean_token_accuracy": 0.15699917152523996,
      "num_tokens": 24634127.0,
      "step": 12585
    },
    {
      "entropy": 6.06918158531189,
      "epoch": 1.2585095216674165,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004849646549798019,
      "loss": 5.8211,
      "mean_token_accuracy": 0.1637175738811493,
      "num_tokens": 24642451.0,
      "step": 12590
    },
    {
      "entropy": 6.116307544708252,
      "epoch": 1.259009346728645,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048495182499336275,
      "loss": 5.8379,
      "mean_token_accuracy": 0.16285798996686934,
      "num_tokens": 24651229.0,
      "step": 12595
    },
    {
      "entropy": 6.146953392028808,
      "epoch": 1.2595091717898734,
      "grad_norm": 1.5625,
      "learning_rate": 0.00048493898972458985,
      "loss": 6.0355,
      "mean_token_accuracy": 0.15328094512224197,
      "num_tokens": 24660435.0,
      "step": 12600
    },
    {
      "entropy": 6.131228876113892,
      "epoch": 1.2600089968511021,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004849261491738059,
      "loss": 5.9337,
      "mean_token_accuracy": 0.15446446985006332,
      "num_tokens": 24669254.0,
      "step": 12605
    },
    {
      "entropy": 6.13405065536499,
      "epoch": 1.2605088219123306,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048491330334133405,
      "loss": 5.8254,
      "mean_token_accuracy": 0.1619556024670601,
      "num_tokens": 24677943.0,
      "step": 12610
    },
    {
      "entropy": 6.216641044616699,
      "epoch": 1.2610086469735593,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00048490045222749744,
      "loss": 5.9961,
      "mean_token_accuracy": 0.149625363945961,
      "num_tokens": 24686678.0,
      "step": 12615
    },
    {
      "entropy": 6.142766618728638,
      "epoch": 1.2615084720347878,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048488759583261955,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16586627811193466,
      "num_tokens": 24696417.0,
      "step": 12620
    },
    {
      "entropy": 6.1422553062438965,
      "epoch": 1.2620082970960165,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004848747341570238,
      "loss": 5.9693,
      "mean_token_accuracy": 0.15504771694540978,
      "num_tokens": 24705624.0,
      "step": 12625
    },
    {
      "entropy": 6.149065208435059,
      "epoch": 1.262508122157245,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048486186720103364,
      "loss": 5.8893,
      "mean_token_accuracy": 0.15257293730974197,
      "num_tokens": 24715705.0,
      "step": 12630
    },
    {
      "entropy": 6.116073226928711,
      "epoch": 1.2630079472184734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000484848994964973,
      "loss": 5.8561,
      "mean_token_accuracy": 0.16290032714605332,
      "num_tokens": 24725358.0,
      "step": 12635
    },
    {
      "entropy": 6.09953031539917,
      "epoch": 1.2635077722797021,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048483611744916556,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1561470717191696,
      "num_tokens": 24735457.0,
      "step": 12640
    },
    {
      "entropy": 6.258626937866211,
      "epoch": 1.2640075973409306,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048482323465393545,
      "loss": 5.9732,
      "mean_token_accuracy": 0.1545448236167431,
      "num_tokens": 24745391.0,
      "step": 12645
    },
    {
      "entropy": 6.153474807739258,
      "epoch": 1.2645074224021593,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004848103465796067,
      "loss": 5.897,
      "mean_token_accuracy": 0.16451151072978973,
      "num_tokens": 24754578.0,
      "step": 12650
    },
    {
      "entropy": 6.087050724029541,
      "epoch": 1.2650072474633878,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004847974532265037,
      "loss": 5.9306,
      "mean_token_accuracy": 0.1613159030675888,
      "num_tokens": 24764115.0,
      "step": 12655
    },
    {
      "entropy": 6.140824604034424,
      "epoch": 1.2655070725246165,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004847845545949507,
      "loss": 5.8993,
      "mean_token_accuracy": 0.15882621258497237,
      "num_tokens": 24773247.0,
      "step": 12660
    },
    {
      "entropy": 6.101389408111572,
      "epoch": 1.266006897585845,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048477165068527237,
      "loss": 5.8787,
      "mean_token_accuracy": 0.16493503898382186,
      "num_tokens": 24783139.0,
      "step": 12665
    },
    {
      "entropy": 6.213921165466308,
      "epoch": 1.2665067226470734,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048475874149779313,
      "loss": 5.9355,
      "mean_token_accuracy": 0.1574448376893997,
      "num_tokens": 24793591.0,
      "step": 12670
    },
    {
      "entropy": 6.1698345184326175,
      "epoch": 1.2670065477083021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000484745827032838,
      "loss": 5.8554,
      "mean_token_accuracy": 0.16278644055128097,
      "num_tokens": 24803340.0,
      "step": 12675
    },
    {
      "entropy": 6.071834707260132,
      "epoch": 1.2675063727695306,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048473290729073194,
      "loss": 5.7862,
      "mean_token_accuracy": 0.16299667656421662,
      "num_tokens": 24812133.0,
      "step": 12680
    },
    {
      "entropy": 6.071779394149781,
      "epoch": 1.2680061978307593,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004847199822717998,
      "loss": 5.9166,
      "mean_token_accuracy": 0.15988943725824356,
      "num_tokens": 24822443.0,
      "step": 12685
    },
    {
      "entropy": 6.088105487823486,
      "epoch": 1.2685060228919878,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004847070519763668,
      "loss": 5.8529,
      "mean_token_accuracy": 0.17268768846988677,
      "num_tokens": 24832390.0,
      "step": 12690
    },
    {
      "entropy": 6.130566692352295,
      "epoch": 1.2690058479532165,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004846941164047585,
      "loss": 5.8429,
      "mean_token_accuracy": 0.16656181067228318,
      "num_tokens": 24841704.0,
      "step": 12695
    },
    {
      "entropy": 6.235428762435913,
      "epoch": 1.269505673014445,
      "grad_norm": 1.171875,
      "learning_rate": 0.00048468117555730013,
      "loss": 5.9967,
      "mean_token_accuracy": 0.15421323403716086,
      "num_tokens": 24852459.0,
      "step": 12700
    },
    {
      "entropy": 6.137821531295776,
      "epoch": 1.2700054980756734,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004846682294343173,
      "loss": 5.9582,
      "mean_token_accuracy": 0.15802925378084182,
      "num_tokens": 24861758.0,
      "step": 12705
    },
    {
      "entropy": 6.263102197647095,
      "epoch": 1.2705053231369021,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004846552780361358,
      "loss": 5.994,
      "mean_token_accuracy": 0.1518670491874218,
      "num_tokens": 24871855.0,
      "step": 12710
    },
    {
      "entropy": 6.159356212615966,
      "epoch": 1.2710051481981306,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048464232136308146,
      "loss": 5.8889,
      "mean_token_accuracy": 0.16126847863197327,
      "num_tokens": 24880602.0,
      "step": 12715
    },
    {
      "entropy": 6.095026016235352,
      "epoch": 1.2715049732593593,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004846293594154802,
      "loss": 5.9184,
      "mean_token_accuracy": 0.15892921686172484,
      "num_tokens": 24891032.0,
      "step": 12720
    },
    {
      "entropy": 6.124080944061279,
      "epoch": 1.2720047983205878,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004846163921936583,
      "loss": 5.8954,
      "mean_token_accuracy": 0.15868026465177537,
      "num_tokens": 24900892.0,
      "step": 12725
    },
    {
      "entropy": 6.210001373291016,
      "epoch": 1.2725046233818165,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048460341969794183,
      "loss": 5.9724,
      "mean_token_accuracy": 0.15081714242696762,
      "num_tokens": 24910576.0,
      "step": 12730
    },
    {
      "entropy": 6.112710809707641,
      "epoch": 1.273004448443045,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048459044192865726,
      "loss": 5.9639,
      "mean_token_accuracy": 0.1581423833966255,
      "num_tokens": 24920861.0,
      "step": 12735
    },
    {
      "entropy": 6.187848234176636,
      "epoch": 1.2735042735042734,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004845774588861311,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15202307403087617,
      "num_tokens": 24930539.0,
      "step": 12740
    },
    {
      "entropy": 6.106684780120849,
      "epoch": 1.2740040985655021,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048456447057069004,
      "loss": 5.9644,
      "mean_token_accuracy": 0.15325247421860694,
      "num_tokens": 24939737.0,
      "step": 12745
    },
    {
      "entropy": 6.174986457824707,
      "epoch": 1.2745039236267306,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048455147698266084,
      "loss": 6.0029,
      "mean_token_accuracy": 0.1542637139558792,
      "num_tokens": 24949316.0,
      "step": 12750
    },
    {
      "entropy": 6.1558287143707275,
      "epoch": 1.2750037486879593,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004845384781223703,
      "loss": 5.852,
      "mean_token_accuracy": 0.16335819512605668,
      "num_tokens": 24959454.0,
      "step": 12755
    },
    {
      "entropy": 6.12948055267334,
      "epoch": 1.2755035737491878,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048452547399014566,
      "loss": 5.9352,
      "mean_token_accuracy": 0.15802226662635804,
      "num_tokens": 24969091.0,
      "step": 12760
    },
    {
      "entropy": 6.145900726318359,
      "epoch": 1.2760033988104165,
      "grad_norm": 1.3125,
      "learning_rate": 0.000484512464586314,
      "loss": 5.7846,
      "mean_token_accuracy": 0.16441865712404252,
      "num_tokens": 24979239.0,
      "step": 12765
    },
    {
      "entropy": 6.2058042049407955,
      "epoch": 1.276503223871645,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048449944991120264,
      "loss": 5.9834,
      "mean_token_accuracy": 0.1546609953045845,
      "num_tokens": 24989274.0,
      "step": 12770
    },
    {
      "entropy": 6.18181848526001,
      "epoch": 1.2770030489328734,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004844864299651391,
      "loss": 5.8697,
      "mean_token_accuracy": 0.15770785361528397,
      "num_tokens": 24997873.0,
      "step": 12775
    },
    {
      "entropy": 6.043127346038818,
      "epoch": 1.2775028739941021,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048447340474845076,
      "loss": 5.862,
      "mean_token_accuracy": 0.16047936379909516,
      "num_tokens": 25007829.0,
      "step": 12780
    },
    {
      "entropy": 6.05285062789917,
      "epoch": 1.2780026990553306,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048446037426146555,
      "loss": 5.8513,
      "mean_token_accuracy": 0.1603124648332596,
      "num_tokens": 25016795.0,
      "step": 12785
    },
    {
      "entropy": 6.138433170318604,
      "epoch": 1.278502524116559,
      "grad_norm": 1.25,
      "learning_rate": 0.0004844473385045112,
      "loss": 5.9159,
      "mean_token_accuracy": 0.15596069470047952,
      "num_tokens": 25026012.0,
      "step": 12790
    },
    {
      "entropy": 6.093439102172852,
      "epoch": 1.2790023491777878,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004844342974779157,
      "loss": 5.8291,
      "mean_token_accuracy": 0.1692409783601761,
      "num_tokens": 25035353.0,
      "step": 12795
    },
    {
      "entropy": 6.1351806163787845,
      "epoch": 1.2795021742390165,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048442125118200707,
      "loss": 5.9162,
      "mean_token_accuracy": 0.15810562670230865,
      "num_tokens": 25045301.0,
      "step": 12800
    },
    {
      "entropy": 6.173824310302734,
      "epoch": 1.280001999300245,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004844081996171137,
      "loss": 5.9589,
      "mean_token_accuracy": 0.15624310076236725,
      "num_tokens": 25055576.0,
      "step": 12805
    },
    {
      "entropy": 6.0311055183410645,
      "epoch": 1.2805018243614734,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000484395142783564,
      "loss": 5.8121,
      "mean_token_accuracy": 0.16766391843557357,
      "num_tokens": 25064077.0,
      "step": 12810
    },
    {
      "entropy": 6.017964506149292,
      "epoch": 1.2810016494227021,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004843820806816863,
      "loss": 5.8144,
      "mean_token_accuracy": 0.1673479273915291,
      "num_tokens": 25072908.0,
      "step": 12815
    },
    {
      "entropy": 6.138772106170654,
      "epoch": 1.2815014744839306,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004843690133118094,
      "loss": 5.9236,
      "mean_token_accuracy": 0.15627482235431672,
      "num_tokens": 25082919.0,
      "step": 12820
    },
    {
      "entropy": 6.2176624774932865,
      "epoch": 1.282001299545159,
      "grad_norm": 1.21875,
      "learning_rate": 0.000484355940674262,
      "loss": 5.9598,
      "mean_token_accuracy": 0.1600956827402115,
      "num_tokens": 25092520.0,
      "step": 12825
    },
    {
      "entropy": 6.11890435218811,
      "epoch": 1.2825011246063878,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048434286276937294,
      "loss": 5.952,
      "mean_token_accuracy": 0.15088521540164948,
      "num_tokens": 25102638.0,
      "step": 12830
    },
    {
      "entropy": 6.05759973526001,
      "epoch": 1.2830009496676165,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004843297795974713,
      "loss": 5.9399,
      "mean_token_accuracy": 0.15684837549924852,
      "num_tokens": 25112011.0,
      "step": 12835
    },
    {
      "entropy": 6.182925748825073,
      "epoch": 1.283500774728845,
      "grad_norm": 1.375,
      "learning_rate": 0.0004843166911588864,
      "loss": 5.9875,
      "mean_token_accuracy": 0.15050630867481232,
      "num_tokens": 25122051.0,
      "step": 12840
    },
    {
      "entropy": 6.237850284576416,
      "epoch": 1.2840005997900734,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004843035974539473,
      "loss": 5.9483,
      "mean_token_accuracy": 0.15074261724948884,
      "num_tokens": 25131928.0,
      "step": 12845
    },
    {
      "entropy": 6.022640514373779,
      "epoch": 1.2845004248513021,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048429049848298364,
      "loss": 5.7247,
      "mean_token_accuracy": 0.1703232541680336,
      "num_tokens": 25140204.0,
      "step": 12850
    },
    {
      "entropy": 6.0534802913665775,
      "epoch": 1.2850002499125306,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048427739424632485,
      "loss": 5.8285,
      "mean_token_accuracy": 0.16151807755231856,
      "num_tokens": 25149445.0,
      "step": 12855
    },
    {
      "entropy": 6.1226630210876465,
      "epoch": 1.285500074973759,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048426428474430055,
      "loss": 5.8684,
      "mean_token_accuracy": 0.16000229269266128,
      "num_tokens": 25159311.0,
      "step": 12860
    },
    {
      "entropy": 6.05877947807312,
      "epoch": 1.2859999000349878,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048425116997724083,
      "loss": 5.8156,
      "mean_token_accuracy": 0.1679415859282017,
      "num_tokens": 25168880.0,
      "step": 12865
    },
    {
      "entropy": 6.10956826210022,
      "epoch": 1.2864997250962165,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048423804994547555,
      "loss": 5.8511,
      "mean_token_accuracy": 0.1589955359697342,
      "num_tokens": 25177812.0,
      "step": 12870
    },
    {
      "entropy": 6.012914180755615,
      "epoch": 1.286999550157445,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048422492464933464,
      "loss": 5.8141,
      "mean_token_accuracy": 0.16360452622175217,
      "num_tokens": 25187831.0,
      "step": 12875
    },
    {
      "entropy": 6.074532222747803,
      "epoch": 1.2874993752186734,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048421179408914857,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16218030005693435,
      "num_tokens": 25198164.0,
      "step": 12880
    },
    {
      "entropy": 6.217318916320801,
      "epoch": 1.2879992002799021,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048419865826524757,
      "loss": 5.9544,
      "mean_token_accuracy": 0.15810884088277816,
      "num_tokens": 25208740.0,
      "step": 12885
    },
    {
      "entropy": 6.136310720443726,
      "epoch": 1.2884990253411306,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048418551717796217,
      "loss": 5.8559,
      "mean_token_accuracy": 0.16279749274253846,
      "num_tokens": 25218475.0,
      "step": 12890
    },
    {
      "entropy": 6.1293775081634525,
      "epoch": 1.288998850402359,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048417237082762295,
      "loss": 5.9791,
      "mean_token_accuracy": 0.15591106414794922,
      "num_tokens": 25227943.0,
      "step": 12895
    },
    {
      "entropy": 5.933111667633057,
      "epoch": 1.2894986754635878,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048415921921456074,
      "loss": 5.7717,
      "mean_token_accuracy": 0.16989419162273406,
      "num_tokens": 25237131.0,
      "step": 12900
    },
    {
      "entropy": 6.131748151779175,
      "epoch": 1.2899985005248162,
      "grad_norm": 1.421875,
      "learning_rate": 0.00048414606233910634,
      "loss": 5.8482,
      "mean_token_accuracy": 0.16231559514999389,
      "num_tokens": 25245983.0,
      "step": 12905
    },
    {
      "entropy": 6.080750846862793,
      "epoch": 1.290498325586045,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004841329002015909,
      "loss": 5.8689,
      "mean_token_accuracy": 0.1597699150443077,
      "num_tokens": 25255155.0,
      "step": 12910
    },
    {
      "entropy": 6.093892669677734,
      "epoch": 1.2909981506472734,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048411973280234553,
      "loss": 5.8688,
      "mean_token_accuracy": 0.15933849513530732,
      "num_tokens": 25265207.0,
      "step": 12915
    },
    {
      "entropy": 6.13886079788208,
      "epoch": 1.291497975708502,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048410656014170135,
      "loss": 5.8283,
      "mean_token_accuracy": 0.16692849695682527,
      "num_tokens": 25273995.0,
      "step": 12920
    },
    {
      "entropy": 6.0460280418396,
      "epoch": 1.2919978007697306,
      "grad_norm": 2.28125,
      "learning_rate": 0.00048409338221999,
      "loss": 5.813,
      "mean_token_accuracy": 0.1601512536406517,
      "num_tokens": 25284334.0,
      "step": 12925
    },
    {
      "entropy": 6.040530395507813,
      "epoch": 1.292497625830959,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048408019903754297,
      "loss": 5.8527,
      "mean_token_accuracy": 0.1636658266186714,
      "num_tokens": 25294360.0,
      "step": 12930
    },
    {
      "entropy": 6.108194208145141,
      "epoch": 1.2929974508921878,
      "grad_norm": 1.328125,
      "learning_rate": 0.000484067010594692,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16477513164281846,
      "num_tokens": 25302869.0,
      "step": 12935
    },
    {
      "entropy": 6.073296117782593,
      "epoch": 1.2934972759534162,
      "grad_norm": 3.390625,
      "learning_rate": 0.00048405381689176877,
      "loss": 5.8498,
      "mean_token_accuracy": 0.16637372151017188,
      "num_tokens": 25313825.0,
      "step": 12940
    },
    {
      "entropy": 6.115132999420166,
      "epoch": 1.293997101014645,
      "grad_norm": 1.375,
      "learning_rate": 0.00048404061792910527,
      "loss": 5.9137,
      "mean_token_accuracy": 0.15858078002929688,
      "num_tokens": 25323212.0,
      "step": 12945
    },
    {
      "entropy": 6.204974508285522,
      "epoch": 1.2944969260758734,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048402741370703363,
      "loss": 5.9992,
      "mean_token_accuracy": 0.15611710101366044,
      "num_tokens": 25333037.0,
      "step": 12950
    },
    {
      "entropy": 6.165543413162231,
      "epoch": 1.294996751137102,
      "grad_norm": 1.25,
      "learning_rate": 0.0004840142042258861,
      "loss": 5.9156,
      "mean_token_accuracy": 0.1604437455534935,
      "num_tokens": 25342810.0,
      "step": 12955
    },
    {
      "entropy": 6.158083486557007,
      "epoch": 1.2954965761983306,
      "grad_norm": 1.078125,
      "learning_rate": 0.000484000989485995,
      "loss": 5.9863,
      "mean_token_accuracy": 0.15187346041202546,
      "num_tokens": 25353119.0,
      "step": 12960
    },
    {
      "entropy": 6.149922847747803,
      "epoch": 1.295996401259559,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048398776948769265,
      "loss": 5.8807,
      "mean_token_accuracy": 0.16456405818462372,
      "num_tokens": 25363450.0,
      "step": 12965
    },
    {
      "entropy": 6.064756059646607,
      "epoch": 1.2964962263207878,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004839745442313119,
      "loss": 5.836,
      "mean_token_accuracy": 0.16085902005434036,
      "num_tokens": 25372830.0,
      "step": 12970
    },
    {
      "entropy": 6.139484786987305,
      "epoch": 1.2969960513820162,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048396131371718535,
      "loss": 5.9098,
      "mean_token_accuracy": 0.15242418497800828,
      "num_tokens": 25382102.0,
      "step": 12975
    },
    {
      "entropy": 6.1773934841156,
      "epoch": 1.297495876443245,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048394807794564586,
      "loss": 5.9621,
      "mean_token_accuracy": 0.15555364340543748,
      "num_tokens": 25390817.0,
      "step": 12980
    },
    {
      "entropy": 6.145787382125855,
      "epoch": 1.2979957015044734,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004839348369170266,
      "loss": 5.9251,
      "mean_token_accuracy": 0.1562415450811386,
      "num_tokens": 25399958.0,
      "step": 12985
    },
    {
      "entropy": 6.147863912582397,
      "epoch": 1.298495526565702,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048392159063166055,
      "loss": 5.9144,
      "mean_token_accuracy": 0.156745782494545,
      "num_tokens": 25410870.0,
      "step": 12990
    },
    {
      "entropy": 6.208051347732544,
      "epoch": 1.2989953516269306,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004839083390898811,
      "loss": 5.9616,
      "mean_token_accuracy": 0.15137394964694978,
      "num_tokens": 25420906.0,
      "step": 12995
    },
    {
      "entropy": 6.1741392612457275,
      "epoch": 1.299495176688159,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004838950822920215,
      "loss": 5.8808,
      "mean_token_accuracy": 0.15629660785198213,
      "num_tokens": 25430046.0,
      "step": 13000
    },
    {
      "entropy": 6.042098808288574,
      "epoch": 1.2999950017493878,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004838818202384155,
      "loss": 5.8352,
      "mean_token_accuracy": 0.1678431436419487,
      "num_tokens": 25438785.0,
      "step": 13005
    },
    {
      "entropy": 6.103204965591431,
      "epoch": 1.3004948268106162,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048386855292939664,
      "loss": 5.9232,
      "mean_token_accuracy": 0.1591215431690216,
      "num_tokens": 25448815.0,
      "step": 13010
    },
    {
      "entropy": 6.16128945350647,
      "epoch": 1.300994651871845,
      "grad_norm": 1.125,
      "learning_rate": 0.0004838552803652986,
      "loss": 5.8716,
      "mean_token_accuracy": 0.15830889195203782,
      "num_tokens": 25458528.0,
      "step": 13015
    },
    {
      "entropy": 6.137110948562622,
      "epoch": 1.3014944769330734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048384200254645557,
      "loss": 5.8004,
      "mean_token_accuracy": 0.16168108433485032,
      "num_tokens": 25467678.0,
      "step": 13020
    },
    {
      "entropy": 6.017528963088989,
      "epoch": 1.301994301994302,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048382871947320145,
      "loss": 5.7896,
      "mean_token_accuracy": 0.17351060658693312,
      "num_tokens": 25478601.0,
      "step": 13025
    },
    {
      "entropy": 6.0995360851287845,
      "epoch": 1.3024941270555306,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004838154311458705,
      "loss": 5.8752,
      "mean_token_accuracy": 0.15749110281467438,
      "num_tokens": 25489439.0,
      "step": 13030
    },
    {
      "entropy": 6.068431997299195,
      "epoch": 1.302993952116759,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048380213756479704,
      "loss": 5.8835,
      "mean_token_accuracy": 0.16281150579452514,
      "num_tokens": 25499976.0,
      "step": 13035
    },
    {
      "entropy": 6.164010763168335,
      "epoch": 1.3034937771779878,
      "grad_norm": 1.5,
      "learning_rate": 0.00048378883873031553,
      "loss": 6.0211,
      "mean_token_accuracy": 0.15788150876760482,
      "num_tokens": 25509737.0,
      "step": 13040
    },
    {
      "entropy": 6.211325693130493,
      "epoch": 1.3039936022392162,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004837755346427605,
      "loss": 5.9352,
      "mean_token_accuracy": 0.15355114936828612,
      "num_tokens": 25519130.0,
      "step": 13045
    },
    {
      "entropy": 6.178356313705445,
      "epoch": 1.3044934273004447,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048376222530246684,
      "loss": 5.9339,
      "mean_token_accuracy": 0.15717733800411224,
      "num_tokens": 25528625.0,
      "step": 13050
    },
    {
      "entropy": 6.137736892700195,
      "epoch": 1.3049932523616734,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048374891070976925,
      "loss": 5.9609,
      "mean_token_accuracy": 0.1493631862103939,
      "num_tokens": 25537912.0,
      "step": 13055
    },
    {
      "entropy": 6.1544262886047365,
      "epoch": 1.305493077422902,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004837355908650028,
      "loss": 5.9404,
      "mean_token_accuracy": 0.15519590377807618,
      "num_tokens": 25548622.0,
      "step": 13060
    },
    {
      "entropy": 6.00636248588562,
      "epoch": 1.3059929024841306,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004837222657685026,
      "loss": 5.7061,
      "mean_token_accuracy": 0.17257911413908006,
      "num_tokens": 25558322.0,
      "step": 13065
    },
    {
      "entropy": 6.144831800460816,
      "epoch": 1.306492727545359,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048370893542060383,
      "loss": 5.9349,
      "mean_token_accuracy": 0.15338888317346572,
      "num_tokens": 25568586.0,
      "step": 13070
    },
    {
      "entropy": 6.2042272090911865,
      "epoch": 1.3069925526065878,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004836955998216421,
      "loss": 5.9268,
      "mean_token_accuracy": 0.15211068466305733,
      "num_tokens": 25577405.0,
      "step": 13075
    },
    {
      "entropy": 6.11801323890686,
      "epoch": 1.3074923776678162,
      "grad_norm": 1.46875,
      "learning_rate": 0.00048368225897195264,
      "loss": 5.7577,
      "mean_token_accuracy": 0.16974805295467377,
      "num_tokens": 25587719.0,
      "step": 13080
    },
    {
      "entropy": 6.094621801376343,
      "epoch": 1.3079922027290447,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048366891287187133,
      "loss": 5.9015,
      "mean_token_accuracy": 0.16151498407125472,
      "num_tokens": 25597334.0,
      "step": 13085
    },
    {
      "entropy": 6.273237609863282,
      "epoch": 1.3084920277902734,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004836555615217337,
      "loss": 6.063,
      "mean_token_accuracy": 0.1484041228890419,
      "num_tokens": 25606684.0,
      "step": 13090
    },
    {
      "entropy": 6.100587940216064,
      "epoch": 1.308991852851502,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004836422049218761,
      "loss": 5.7917,
      "mean_token_accuracy": 0.173512265086174,
      "num_tokens": 25615808.0,
      "step": 13095
    },
    {
      "entropy": 6.010607576370239,
      "epoch": 1.3094916779127306,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048362884307263416,
      "loss": 5.8064,
      "mean_token_accuracy": 0.16788436025381087,
      "num_tokens": 25626444.0,
      "step": 13100
    },
    {
      "entropy": 6.135896348953247,
      "epoch": 1.309991502973959,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048361547597434426,
      "loss": 5.863,
      "mean_token_accuracy": 0.15651350393891333,
      "num_tokens": 25635730.0,
      "step": 13105
    },
    {
      "entropy": 6.183680772781372,
      "epoch": 1.3104913280351878,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048360210362734257,
      "loss": 5.8935,
      "mean_token_accuracy": 0.1595741964876652,
      "num_tokens": 25645938.0,
      "step": 13110
    },
    {
      "entropy": 6.127643585205078,
      "epoch": 1.3109911530964162,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004835887260319657,
      "loss": 5.9768,
      "mean_token_accuracy": 0.15532899796962737,
      "num_tokens": 25656135.0,
      "step": 13115
    },
    {
      "entropy": 6.193100643157959,
      "epoch": 1.3114909781576447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004835753431885502,
      "loss": 5.9862,
      "mean_token_accuracy": 0.1517387479543686,
      "num_tokens": 25666134.0,
      "step": 13120
    },
    {
      "entropy": 6.205765151977539,
      "epoch": 1.3119908032188734,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048356195509743265,
      "loss": 6.0211,
      "mean_token_accuracy": 0.14750407189130782,
      "num_tokens": 25675716.0,
      "step": 13125
    },
    {
      "entropy": 6.161862325668335,
      "epoch": 1.312490628280102,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00048354856175895,
      "loss": 5.8508,
      "mean_token_accuracy": 0.15966537594795227,
      "num_tokens": 25685814.0,
      "step": 13130
    },
    {
      "entropy": 6.139588689804077,
      "epoch": 1.3129904533413306,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004835351631734392,
      "loss": 5.8502,
      "mean_token_accuracy": 0.1668509155511856,
      "num_tokens": 25694995.0,
      "step": 13135
    },
    {
      "entropy": 6.161866331100464,
      "epoch": 1.313490278402559,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004835217593412374,
      "loss": 5.9214,
      "mean_token_accuracy": 0.15764110907912254,
      "num_tokens": 25705041.0,
      "step": 13140
    },
    {
      "entropy": 6.22493896484375,
      "epoch": 1.3139901034637878,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004835083502626818,
      "loss": 5.9205,
      "mean_token_accuracy": 0.15436616092920302,
      "num_tokens": 25715830.0,
      "step": 13145
    },
    {
      "entropy": 6.123371934890747,
      "epoch": 1.3144899285250162,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00048349493593810973,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15670751929283142,
      "num_tokens": 25725357.0,
      "step": 13150
    },
    {
      "entropy": 6.122776222229004,
      "epoch": 1.3149897535862447,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004834815163678587,
      "loss": 5.9302,
      "mean_token_accuracy": 0.16070552319288253,
      "num_tokens": 25735720.0,
      "step": 13155
    },
    {
      "entropy": 6.154614114761353,
      "epoch": 1.3154895786474734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004834680915522664,
      "loss": 5.9087,
      "mean_token_accuracy": 0.15416300743818284,
      "num_tokens": 25745728.0,
      "step": 13160
    },
    {
      "entropy": 6.177586507797241,
      "epoch": 1.3159894037087019,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004834546614916705,
      "loss": 5.8928,
      "mean_token_accuracy": 0.1595405623316765,
      "num_tokens": 25755917.0,
      "step": 13165
    },
    {
      "entropy": 6.1583672046661375,
      "epoch": 1.3164892287699306,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048344122618640905,
      "loss": 5.9844,
      "mean_token_accuracy": 0.1551694765686989,
      "num_tokens": 25766195.0,
      "step": 13170
    },
    {
      "entropy": 6.165318727493286,
      "epoch": 1.316989053831159,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004834277856368199,
      "loss": 5.8217,
      "mean_token_accuracy": 0.16466738730669023,
      "num_tokens": 25774134.0,
      "step": 13175
    },
    {
      "entropy": 6.073446655273438,
      "epoch": 1.3174888788923877,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048341433984324127,
      "loss": 5.8802,
      "mean_token_accuracy": 0.16324811279773713,
      "num_tokens": 25783377.0,
      "step": 13180
    },
    {
      "entropy": 6.016869354248047,
      "epoch": 1.3179887039536162,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048340088880601163,
      "loss": 5.7209,
      "mean_token_accuracy": 0.16514964103698732,
      "num_tokens": 25793562.0,
      "step": 13185
    },
    {
      "entropy": 6.117653751373291,
      "epoch": 1.3184885290148447,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004833874325254691,
      "loss": 5.9567,
      "mean_token_accuracy": 0.16125948578119279,
      "num_tokens": 25803299.0,
      "step": 13190
    },
    {
      "entropy": 6.188000154495239,
      "epoch": 1.3189883540760734,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048337397100195245,
      "loss": 5.9752,
      "mean_token_accuracy": 0.15082984566688537,
      "num_tokens": 25813053.0,
      "step": 13195
    },
    {
      "entropy": 6.19323320388794,
      "epoch": 1.3194881791373019,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004833605042358003,
      "loss": 5.9666,
      "mean_token_accuracy": 0.15669853687286378,
      "num_tokens": 25822311.0,
      "step": 13200
    },
    {
      "entropy": 6.1460075855255125,
      "epoch": 1.3199880041985306,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004833470322273515,
      "loss": 5.9845,
      "mean_token_accuracy": 0.1451917141675949,
      "num_tokens": 25832473.0,
      "step": 13205
    },
    {
      "entropy": 6.170310354232788,
      "epoch": 1.320487829259759,
      "grad_norm": 1.25,
      "learning_rate": 0.00048333355497694506,
      "loss": 5.8529,
      "mean_token_accuracy": 0.16258705407381058,
      "num_tokens": 25840919.0,
      "step": 13210
    },
    {
      "entropy": 6.120567417144775,
      "epoch": 1.3209876543209877,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004833200724849199,
      "loss": 5.8678,
      "mean_token_accuracy": 0.16181413531303407,
      "num_tokens": 25850517.0,
      "step": 13215
    },
    {
      "entropy": 6.221341276168824,
      "epoch": 1.3214874793822162,
      "grad_norm": 1.53125,
      "learning_rate": 0.00048330658475161526,
      "loss": 6.0339,
      "mean_token_accuracy": 0.14737141653895378,
      "num_tokens": 25861580.0,
      "step": 13220
    },
    {
      "entropy": 6.172426557540893,
      "epoch": 1.3219873044434447,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004832930917773706,
      "loss": 5.9548,
      "mean_token_accuracy": 0.1602742850780487,
      "num_tokens": 25872383.0,
      "step": 13225
    },
    {
      "entropy": 6.122541618347168,
      "epoch": 1.3224871295046734,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048327959356252544,
      "loss": 5.9422,
      "mean_token_accuracy": 0.1594276785850525,
      "num_tokens": 25883016.0,
      "step": 13230
    },
    {
      "entropy": 6.149798822402954,
      "epoch": 1.3229869545659019,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004832660901074192,
      "loss": 5.9223,
      "mean_token_accuracy": 0.15740931481122972,
      "num_tokens": 25892810.0,
      "step": 13235
    },
    {
      "entropy": 6.1663611888885494,
      "epoch": 1.3234867796271306,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048325258141239174,
      "loss": 5.9102,
      "mean_token_accuracy": 0.1587192162871361,
      "num_tokens": 25902548.0,
      "step": 13240
    },
    {
      "entropy": 6.130265474319458,
      "epoch": 1.323986604688359,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004832390674777829,
      "loss": 5.8827,
      "mean_token_accuracy": 0.1631936974823475,
      "num_tokens": 25911572.0,
      "step": 13245
    },
    {
      "entropy": 6.061541223526001,
      "epoch": 1.3244864297495877,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048322554830393274,
      "loss": 5.7852,
      "mean_token_accuracy": 0.16264218240976333,
      "num_tokens": 25921226.0,
      "step": 13250
    },
    {
      "entropy": 6.1363156795501705,
      "epoch": 1.3249862548108162,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004832120238911813,
      "loss": 5.8991,
      "mean_token_accuracy": 0.15605154782533645,
      "num_tokens": 25930920.0,
      "step": 13255
    },
    {
      "entropy": 6.175198698043824,
      "epoch": 1.3254860798720447,
      "grad_norm": 1.125,
      "learning_rate": 0.000483198494239869,
      "loss": 5.843,
      "mean_token_accuracy": 0.1654557093977928,
      "num_tokens": 25941042.0,
      "step": 13260
    },
    {
      "entropy": 6.206301498413086,
      "epoch": 1.3259859049332734,
      "grad_norm": 1.375,
      "learning_rate": 0.00048318495935033615,
      "loss": 6.0046,
      "mean_token_accuracy": 0.14853819608688354,
      "num_tokens": 25950696.0,
      "step": 13265
    },
    {
      "entropy": 6.160824394226074,
      "epoch": 1.3264857299945019,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048317141922292326,
      "loss": 5.8926,
      "mean_token_accuracy": 0.16059210747480393,
      "num_tokens": 25960034.0,
      "step": 13270
    },
    {
      "entropy": 6.13079719543457,
      "epoch": 1.3269855550557306,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004831578738579711,
      "loss": 5.9003,
      "mean_token_accuracy": 0.1688828647136688,
      "num_tokens": 25969264.0,
      "step": 13275
    },
    {
      "entropy": 6.177458381652832,
      "epoch": 1.327485380116959,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004831443232558203,
      "loss": 6.018,
      "mean_token_accuracy": 0.15124542117118836,
      "num_tokens": 25979046.0,
      "step": 13280
    },
    {
      "entropy": 6.111147260665893,
      "epoch": 1.3279852051781877,
      "grad_norm": 1.15625,
      "learning_rate": 0.00048313076741681186,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1578248620033264,
      "num_tokens": 25988797.0,
      "step": 13285
    },
    {
      "entropy": 6.1148851871490475,
      "epoch": 1.3284850302394162,
      "grad_norm": 1.484375,
      "learning_rate": 0.00048311720634128697,
      "loss": 5.7741,
      "mean_token_accuracy": 0.1661039859056473,
      "num_tokens": 25997586.0,
      "step": 13290
    },
    {
      "entropy": 6.1335609436035154,
      "epoch": 1.3289848553006447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004831036400295867,
      "loss": 5.9629,
      "mean_token_accuracy": 0.150631383061409,
      "num_tokens": 26007369.0,
      "step": 13295
    },
    {
      "entropy": 6.121255350112915,
      "epoch": 1.3294846803618734,
      "grad_norm": 1.25,
      "learning_rate": 0.00048309006848205237,
      "loss": 5.8906,
      "mean_token_accuracy": 0.16120503693819047,
      "num_tokens": 26017063.0,
      "step": 13300
    },
    {
      "entropy": 6.146492958068848,
      "epoch": 1.3299845054231019,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004830764916990255,
      "loss": 5.8487,
      "mean_token_accuracy": 0.16345795392990112,
      "num_tokens": 26026555.0,
      "step": 13305
    },
    {
      "entropy": 6.218799638748169,
      "epoch": 1.3304843304843303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004830629096808476,
      "loss": 5.9354,
      "mean_token_accuracy": 0.1593967080116272,
      "num_tokens": 26036698.0,
      "step": 13310
    },
    {
      "entropy": 6.13630781173706,
      "epoch": 1.330984155545559,
      "grad_norm": 1.375,
      "learning_rate": 0.0004830493224278605,
      "loss": 5.8822,
      "mean_token_accuracy": 0.15833362936973572,
      "num_tokens": 26046462.0,
      "step": 13315
    },
    {
      "entropy": 6.08898286819458,
      "epoch": 1.3314839806067877,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004830357299404059,
      "loss": 5.902,
      "mean_token_accuracy": 0.16499090641736985,
      "num_tokens": 26056180.0,
      "step": 13320
    },
    {
      "entropy": 6.157568025588989,
      "epoch": 1.3319838056680162,
      "grad_norm": 1.25,
      "learning_rate": 0.0004830221322188259,
      "loss": 5.9192,
      "mean_token_accuracy": 0.1544218897819519,
      "num_tokens": 26065621.0,
      "step": 13325
    },
    {
      "entropy": 6.1957679271698,
      "epoch": 1.3324836307292447,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004830085292634626,
      "loss": 5.921,
      "mean_token_accuracy": 0.15938180685043335,
      "num_tokens": 26074640.0,
      "step": 13330
    },
    {
      "entropy": 6.123087739944458,
      "epoch": 1.3329834557904734,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048299492107465816,
      "loss": 5.8815,
      "mean_token_accuracy": 0.15639126747846605,
      "num_tokens": 26084221.0,
      "step": 13335
    },
    {
      "entropy": 6.138051128387451,
      "epoch": 1.3334832808517019,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004829813076527551,
      "loss": 5.9231,
      "mean_token_accuracy": 0.16323902383446692,
      "num_tokens": 26095065.0,
      "step": 13340
    },
    {
      "entropy": 6.140139770507813,
      "epoch": 1.3339831059129303,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048296768899809577,
      "loss": 5.833,
      "mean_token_accuracy": 0.17097852975130082,
      "num_tokens": 26104978.0,
      "step": 13345
    },
    {
      "entropy": 6.043597555160522,
      "epoch": 1.334482930974159,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048295406511102303,
      "loss": 5.8248,
      "mean_token_accuracy": 0.16585729271173477,
      "num_tokens": 26115023.0,
      "step": 13350
    },
    {
      "entropy": 6.142004013061523,
      "epoch": 1.3349827560353877,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004829404359918795,
      "loss": 5.9434,
      "mean_token_accuracy": 0.1579558677971363,
      "num_tokens": 26124533.0,
      "step": 13355
    },
    {
      "entropy": 6.186449956893921,
      "epoch": 1.3354825810966162,
      "grad_norm": 1.5859375,
      "learning_rate": 0.000482926801641008,
      "loss": 5.8489,
      "mean_token_accuracy": 0.16535816490650176,
      "num_tokens": 26134466.0,
      "step": 13360
    },
    {
      "entropy": 6.1876969814300535,
      "epoch": 1.3359824061578447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048291316205875174,
      "loss": 5.9446,
      "mean_token_accuracy": 0.16021082252264024,
      "num_tokens": 26145710.0,
      "step": 13365
    },
    {
      "entropy": 6.114179897308349,
      "epoch": 1.3364822312190734,
      "grad_norm": 1.25,
      "learning_rate": 0.00048289951724545384,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16406906694173812,
      "num_tokens": 26154393.0,
      "step": 13370
    },
    {
      "entropy": 6.012562322616577,
      "epoch": 1.3369820562803019,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048288586720145757,
      "loss": 5.7725,
      "mean_token_accuracy": 0.17339322566986085,
      "num_tokens": 26164230.0,
      "step": 13375
    },
    {
      "entropy": 6.1056078433990475,
      "epoch": 1.3374818813415303,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004828722119271064,
      "loss": 5.8889,
      "mean_token_accuracy": 0.15901896953582764,
      "num_tokens": 26174224.0,
      "step": 13380
    },
    {
      "entropy": 6.1055690288543705,
      "epoch": 1.337981706402759,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00048285855142274375,
      "loss": 5.8669,
      "mean_token_accuracy": 0.16952089965343475,
      "num_tokens": 26183447.0,
      "step": 13385
    },
    {
      "entropy": 6.197963666915894,
      "epoch": 1.3384815314639877,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004828448856887135,
      "loss": 5.984,
      "mean_token_accuracy": 0.15778826624155046,
      "num_tokens": 26193479.0,
      "step": 13390
    },
    {
      "entropy": 6.184249401092529,
      "epoch": 1.3389813565252162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004828312147253595,
      "loss": 5.9366,
      "mean_token_accuracy": 0.15459681898355485,
      "num_tokens": 26202493.0,
      "step": 13395
    },
    {
      "entropy": 6.151545810699463,
      "epoch": 1.3394811815864447,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004828175385330255,
      "loss": 5.9501,
      "mean_token_accuracy": 0.1563641160726547,
      "num_tokens": 26212133.0,
      "step": 13400
    },
    {
      "entropy": 6.089784860610962,
      "epoch": 1.3399810066476734,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004828038571120558,
      "loss": 5.908,
      "mean_token_accuracy": 0.16656311452388764,
      "num_tokens": 26222905.0,
      "step": 13405
    },
    {
      "entropy": 6.181836986541748,
      "epoch": 1.3404808317089019,
      "grad_norm": 1.25,
      "learning_rate": 0.00048279017046279446,
      "loss": 5.9437,
      "mean_token_accuracy": 0.15331728756427765,
      "num_tokens": 26234273.0,
      "step": 13410
    },
    {
      "entropy": 6.1181422710418705,
      "epoch": 1.3409806567701303,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00048277647858558596,
      "loss": 5.8464,
      "mean_token_accuracy": 0.16213321983814238,
      "num_tokens": 26243965.0,
      "step": 13415
    },
    {
      "entropy": 6.042984962463379,
      "epoch": 1.341480481831359,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048276278148077465,
      "loss": 5.8979,
      "mean_token_accuracy": 0.16166059225797652,
      "num_tokens": 26253401.0,
      "step": 13420
    },
    {
      "entropy": 6.006541061401367,
      "epoch": 1.3419803068925875,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048274907914870526,
      "loss": 5.8344,
      "mean_token_accuracy": 0.16181168258190154,
      "num_tokens": 26262481.0,
      "step": 13425
    },
    {
      "entropy": 6.20740008354187,
      "epoch": 1.3424801319538162,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048273537158972246,
      "loss": 5.9998,
      "mean_token_accuracy": 0.15424103438854217,
      "num_tokens": 26271153.0,
      "step": 13430
    },
    {
      "entropy": 6.218473768234253,
      "epoch": 1.3429799570150447,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048272165880417124,
      "loss": 6.051,
      "mean_token_accuracy": 0.14814718887209893,
      "num_tokens": 26281047.0,
      "step": 13435
    },
    {
      "entropy": 6.148539924621582,
      "epoch": 1.3434797820762734,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048270794079239643,
      "loss": 5.9644,
      "mean_token_accuracy": 0.1517001546919346,
      "num_tokens": 26291029.0,
      "step": 13440
    },
    {
      "entropy": 6.12006721496582,
      "epoch": 1.3439796071375019,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048269421755474336,
      "loss": 5.898,
      "mean_token_accuracy": 0.15874358937144278,
      "num_tokens": 26300637.0,
      "step": 13445
    },
    {
      "entropy": 6.1137724876403805,
      "epoch": 1.3444794321987303,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004826804890915572,
      "loss": 5.9562,
      "mean_token_accuracy": 0.1522716410458088,
      "num_tokens": 26309867.0,
      "step": 13450
    },
    {
      "entropy": 6.126278400421143,
      "epoch": 1.344979257259959,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004826667554031834,
      "loss": 5.8497,
      "mean_token_accuracy": 0.16445547044277192,
      "num_tokens": 26318691.0,
      "step": 13455
    },
    {
      "entropy": 6.222006034851074,
      "epoch": 1.3454790823211875,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004826530164899674,
      "loss": 5.9651,
      "mean_token_accuracy": 0.15411202758550643,
      "num_tokens": 26327294.0,
      "step": 13460
    },
    {
      "entropy": 6.145077037811279,
      "epoch": 1.3459789073824162,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048263927235225505,
      "loss": 5.9377,
      "mean_token_accuracy": 0.15691346377134324,
      "num_tokens": 26339965.0,
      "step": 13465
    },
    {
      "entropy": 6.161531686782837,
      "epoch": 1.3464787324436447,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004826255229903919,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16356169581413268,
      "num_tokens": 26350447.0,
      "step": 13470
    },
    {
      "entropy": 6.129216861724854,
      "epoch": 1.3469785575048734,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048261176840472416,
      "loss": 5.9228,
      "mean_token_accuracy": 0.153837464004755,
      "num_tokens": 26359641.0,
      "step": 13475
    },
    {
      "entropy": 6.126151371002197,
      "epoch": 1.3474783825661019,
      "grad_norm": 1.21875,
      "learning_rate": 0.00048259800859559764,
      "loss": 6.0393,
      "mean_token_accuracy": 0.15073853433132173,
      "num_tokens": 26369161.0,
      "step": 13480
    },
    {
      "entropy": 6.204946708679199,
      "epoch": 1.3479782076273303,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004825842435633587,
      "loss": 5.9165,
      "mean_token_accuracy": 0.16021182984113694,
      "num_tokens": 26379592.0,
      "step": 13485
    },
    {
      "entropy": 6.14502911567688,
      "epoch": 1.348478032688559,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048257047330835354,
      "loss": 5.8646,
      "mean_token_accuracy": 0.15866025686264038,
      "num_tokens": 26389164.0,
      "step": 13490
    },
    {
      "entropy": 6.120100498199463,
      "epoch": 1.3489778577497875,
      "grad_norm": 1.203125,
      "learning_rate": 0.00048255669783092876,
      "loss": 5.9396,
      "mean_token_accuracy": 0.15557236820459366,
      "num_tokens": 26399574.0,
      "step": 13495
    },
    {
      "entropy": 6.127181482315064,
      "epoch": 1.3494776828110162,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048254291713143083,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16242548823356628,
      "num_tokens": 26410240.0,
      "step": 13500
    },
    {
      "entropy": 6.13079867362976,
      "epoch": 1.3499775078722447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004825291312102066,
      "loss": 5.817,
      "mean_token_accuracy": 0.16208799332380294,
      "num_tokens": 26419763.0,
      "step": 13505
    },
    {
      "entropy": 6.068390893936157,
      "epoch": 1.3504773329334734,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004825153400676027,
      "loss": 5.9472,
      "mean_token_accuracy": 0.15317082926630973,
      "num_tokens": 26429370.0,
      "step": 13510
    },
    {
      "entropy": 6.115858173370361,
      "epoch": 1.3509771579947019,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004825015437039663,
      "loss": 5.8695,
      "mean_token_accuracy": 0.16802127212285994,
      "num_tokens": 26438425.0,
      "step": 13515
    },
    {
      "entropy": 6.176099061965942,
      "epoch": 1.3514769830559303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004824877421196444,
      "loss": 5.9228,
      "mean_token_accuracy": 0.15681560933589936,
      "num_tokens": 26448039.0,
      "step": 13520
    },
    {
      "entropy": 6.085469961166382,
      "epoch": 1.351976808117159,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004824739353149844,
      "loss": 5.8621,
      "mean_token_accuracy": 0.15977695286273957,
      "num_tokens": 26457682.0,
      "step": 13525
    },
    {
      "entropy": 6.013343858718872,
      "epoch": 1.3524766331783875,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048246012329033355,
      "loss": 5.7869,
      "mean_token_accuracy": 0.16719080209732057,
      "num_tokens": 26466292.0,
      "step": 13530
    },
    {
      "entropy": 6.2077422618865965,
      "epoch": 1.3529764582396162,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048244630604603935,
      "loss": 6.0495,
      "mean_token_accuracy": 0.15241734161973,
      "num_tokens": 26476880.0,
      "step": 13535
    },
    {
      "entropy": 6.21451096534729,
      "epoch": 1.3534762833008447,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048243248358244945,
      "loss": 5.9171,
      "mean_token_accuracy": 0.15523341447114944,
      "num_tokens": 26486132.0,
      "step": 13540
    },
    {
      "entropy": 6.1377252578735355,
      "epoch": 1.3539761083620734,
      "grad_norm": 1.5625,
      "learning_rate": 0.0004824186558999117,
      "loss": 5.8306,
      "mean_token_accuracy": 0.15467707961797714,
      "num_tokens": 26495647.0,
      "step": 13545
    },
    {
      "entropy": 5.986305570602417,
      "epoch": 1.3544759334233019,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048240482299877386,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16964656263589858,
      "num_tokens": 26506048.0,
      "step": 13550
    },
    {
      "entropy": 6.1818756580352785,
      "epoch": 1.3549757584845303,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004823909848793841,
      "loss": 5.9594,
      "mean_token_accuracy": 0.16166492104530333,
      "num_tokens": 26515460.0,
      "step": 13555
    },
    {
      "entropy": 6.2131218910217285,
      "epoch": 1.355475583545759,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048237714154209053,
      "loss": 5.9196,
      "mean_token_accuracy": 0.15648221224546432,
      "num_tokens": 26526106.0,
      "step": 13560
    },
    {
      "entropy": 6.131097030639649,
      "epoch": 1.3559754086069875,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004823632929872415,
      "loss": 5.8207,
      "mean_token_accuracy": 0.1589489161968231,
      "num_tokens": 26535582.0,
      "step": 13565
    },
    {
      "entropy": 6.098101186752319,
      "epoch": 1.3564752336682162,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048234943921518524,
      "loss": 5.8195,
      "mean_token_accuracy": 0.1609871968626976,
      "num_tokens": 26544811.0,
      "step": 13570
    },
    {
      "entropy": 6.164801979064942,
      "epoch": 1.3569750587294447,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004823355802262705,
      "loss": 5.9777,
      "mean_token_accuracy": 0.15619554072618486,
      "num_tokens": 26555340.0,
      "step": 13575
    },
    {
      "entropy": 6.186211824417114,
      "epoch": 1.3574748837906734,
      "grad_norm": 1.25,
      "learning_rate": 0.0004823217160208459,
      "loss": 5.873,
      "mean_token_accuracy": 0.1559571385383606,
      "num_tokens": 26564876.0,
      "step": 13580
    },
    {
      "entropy": 6.0705554485321045,
      "epoch": 1.3579747088519019,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004823078465992604,
      "loss": 5.8572,
      "mean_token_accuracy": 0.15852668285369872,
      "num_tokens": 26573967.0,
      "step": 13585
    },
    {
      "entropy": 6.1822394847869875,
      "epoch": 1.3584745339131303,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004822939719618627,
      "loss": 5.8806,
      "mean_token_accuracy": 0.16476380378007888,
      "num_tokens": 26583753.0,
      "step": 13590
    },
    {
      "entropy": 6.117038154602051,
      "epoch": 1.358974358974359,
      "grad_norm": 1.859375,
      "learning_rate": 0.000482280092109002,
      "loss": 5.8986,
      "mean_token_accuracy": 0.16375560015439988,
      "num_tokens": 26593214.0,
      "step": 13595
    },
    {
      "entropy": 6.095485734939575,
      "epoch": 1.3594741840355875,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048226620704102757,
      "loss": 5.9455,
      "mean_token_accuracy": 0.16200030893087386,
      "num_tokens": 26603257.0,
      "step": 13600
    },
    {
      "entropy": 6.0923604488372805,
      "epoch": 1.359974009096816,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004822523167582886,
      "loss": 5.8467,
      "mean_token_accuracy": 0.1613400548696518,
      "num_tokens": 26612483.0,
      "step": 13605
    },
    {
      "entropy": 6.151202249526977,
      "epoch": 1.3604738341580447,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004822384212611348,
      "loss": 5.9554,
      "mean_token_accuracy": 0.1553657755255699,
      "num_tokens": 26623116.0,
      "step": 13610
    },
    {
      "entropy": 6.163904523849487,
      "epoch": 1.3609736592192734,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048222452054991563,
      "loss": 5.8039,
      "mean_token_accuracy": 0.16505444794893265,
      "num_tokens": 26632648.0,
      "step": 13615
    },
    {
      "entropy": 6.08295841217041,
      "epoch": 1.3614734842805019,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004822106146249808,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16474096477031708,
      "num_tokens": 26642208.0,
      "step": 13620
    },
    {
      "entropy": 6.147450351715088,
      "epoch": 1.3619733093417303,
      "grad_norm": 1.4375,
      "learning_rate": 0.00048219670348668026,
      "loss": 5.9276,
      "mean_token_accuracy": 0.1589716464281082,
      "num_tokens": 26651967.0,
      "step": 13625
    },
    {
      "entropy": 6.092786264419556,
      "epoch": 1.362473134402959,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00048218278713536394,
      "loss": 5.7735,
      "mean_token_accuracy": 0.1690451443195343,
      "num_tokens": 26661138.0,
      "step": 13630
    },
    {
      "entropy": 6.090594625473022,
      "epoch": 1.3629729594641875,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004821688655713821,
      "loss": 5.9106,
      "mean_token_accuracy": 0.1559771254658699,
      "num_tokens": 26671779.0,
      "step": 13635
    },
    {
      "entropy": 5.960031890869141,
      "epoch": 1.363472784525416,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048215493879508474,
      "loss": 5.6095,
      "mean_token_accuracy": 0.18487453162670137,
      "num_tokens": 26681030.0,
      "step": 13640
    },
    {
      "entropy": 6.125823640823365,
      "epoch": 1.3639726095866447,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004821410068068226,
      "loss": 5.9772,
      "mean_token_accuracy": 0.15146733373403548,
      "num_tokens": 26691609.0,
      "step": 13645
    },
    {
      "entropy": 6.205774259567261,
      "epoch": 1.3644724346478734,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004821270696069459,
      "loss": 6.0046,
      "mean_token_accuracy": 0.14849706292152404,
      "num_tokens": 26700874.0,
      "step": 13650
    },
    {
      "entropy": 6.098849725723267,
      "epoch": 1.3649722597091019,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048211312719580553,
      "loss": 5.8448,
      "mean_token_accuracy": 0.1648109495639801,
      "num_tokens": 26711234.0,
      "step": 13655
    },
    {
      "entropy": 6.128755187988281,
      "epoch": 1.3654720847703303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004820991795737521,
      "loss": 5.7828,
      "mean_token_accuracy": 0.17134784162044525,
      "num_tokens": 26720348.0,
      "step": 13660
    },
    {
      "entropy": 6.040027809143067,
      "epoch": 1.365971909831559,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048208522674113665,
      "loss": 5.8261,
      "mean_token_accuracy": 0.16448691338300706,
      "num_tokens": 26730868.0,
      "step": 13665
    },
    {
      "entropy": 6.110388898849488,
      "epoch": 1.3664717348927875,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004820712686983101,
      "loss": 5.9274,
      "mean_token_accuracy": 0.15922615081071853,
      "num_tokens": 26740093.0,
      "step": 13670
    },
    {
      "entropy": 6.13338074684143,
      "epoch": 1.366971559954016,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048205730544562385,
      "loss": 5.8972,
      "mean_token_accuracy": 0.16107486560940742,
      "num_tokens": 26749550.0,
      "step": 13675
    },
    {
      "entropy": 6.190288639068603,
      "epoch": 1.3674713850152447,
      "grad_norm": 1.6640625,
      "learning_rate": 0.000482043336983429,
      "loss": 5.9434,
      "mean_token_accuracy": 0.16200666427612304,
      "num_tokens": 26759416.0,
      "step": 13680
    },
    {
      "entropy": 6.056473922729492,
      "epoch": 1.3679712100764732,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048202936331207707,
      "loss": 5.8552,
      "mean_token_accuracy": 0.15675311386585236,
      "num_tokens": 26769782.0,
      "step": 13685
    },
    {
      "entropy": 6.152235078811645,
      "epoch": 1.3684710351377019,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004820153844319196,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16164583414793016,
      "num_tokens": 26779007.0,
      "step": 13690
    },
    {
      "entropy": 6.057853412628174,
      "epoch": 1.3689708601989303,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048200140034330844,
      "loss": 5.8405,
      "mean_token_accuracy": 0.16202525049448013,
      "num_tokens": 26787828.0,
      "step": 13695
    },
    {
      "entropy": 6.119669818878174,
      "epoch": 1.369470685260159,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048198741104659516,
      "loss": 5.7962,
      "mean_token_accuracy": 0.16255740225315093,
      "num_tokens": 26798190.0,
      "step": 13700
    },
    {
      "entropy": 6.097092819213867,
      "epoch": 1.3699705103213875,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000481973416542132,
      "loss": 5.8578,
      "mean_token_accuracy": 0.16069574058055877,
      "num_tokens": 26807312.0,
      "step": 13705
    },
    {
      "entropy": 6.0869701385498045,
      "epoch": 1.370470335382616,
      "grad_norm": 1.375,
      "learning_rate": 0.0004819594168302709,
      "loss": 5.8539,
      "mean_token_accuracy": 0.1703674912452698,
      "num_tokens": 26816477.0,
      "step": 13710
    },
    {
      "entropy": 5.965642833709717,
      "epoch": 1.3709701604438447,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004819454119113641,
      "loss": 5.7613,
      "mean_token_accuracy": 0.1646174281835556,
      "num_tokens": 26825826.0,
      "step": 13715
    },
    {
      "entropy": 6.087044191360474,
      "epoch": 1.3714699855050732,
      "grad_norm": 1.5078125,
      "learning_rate": 0.000481931401785764,
      "loss": 5.8514,
      "mean_token_accuracy": 0.16051580533385276,
      "num_tokens": 26834809.0,
      "step": 13720
    },
    {
      "entropy": 6.028216552734375,
      "epoch": 1.3719698105663019,
      "grad_norm": 1.25,
      "learning_rate": 0.0004819173864538231,
      "loss": 5.8068,
      "mean_token_accuracy": 0.16428762823343276,
      "num_tokens": 26845322.0,
      "step": 13725
    },
    {
      "entropy": 6.156345558166504,
      "epoch": 1.3724696356275303,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048190336591589394,
      "loss": 5.9821,
      "mean_token_accuracy": 0.14763540029525757,
      "num_tokens": 26854328.0,
      "step": 13730
    },
    {
      "entropy": 6.182986927032471,
      "epoch": 1.372969460688759,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004818893401723294,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15230980068445205,
      "num_tokens": 26863299.0,
      "step": 13735
    },
    {
      "entropy": 6.066908836364746,
      "epoch": 1.3734692857499875,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00048187530922348226,
      "loss": 5.8353,
      "mean_token_accuracy": 0.16558251082897185,
      "num_tokens": 26873461.0,
      "step": 13740
    },
    {
      "entropy": 6.071912574768066,
      "epoch": 1.373969110811216,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004818612730697056,
      "loss": 5.8759,
      "mean_token_accuracy": 0.16145392060279845,
      "num_tokens": 26883310.0,
      "step": 13745
    },
    {
      "entropy": 6.096778821945191,
      "epoch": 1.3744689358724447,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00048184723171135256,
      "loss": 5.8548,
      "mean_token_accuracy": 0.1662290424108505,
      "num_tokens": 26892009.0,
      "step": 13750
    },
    {
      "entropy": 6.097678089141846,
      "epoch": 1.3749687609336732,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048183318514877634,
      "loss": 5.8578,
      "mean_token_accuracy": 0.1552806481719017,
      "num_tokens": 26902246.0,
      "step": 13755
    },
    {
      "entropy": 6.165782499313354,
      "epoch": 1.3754685859949018,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004818191333823305,
      "loss": 5.9292,
      "mean_token_accuracy": 0.16147550344467163,
      "num_tokens": 26912912.0,
      "step": 13760
    },
    {
      "entropy": 6.195747137069702,
      "epoch": 1.3759684110561303,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004818050764123684,
      "loss": 5.8255,
      "mean_token_accuracy": 0.1638929709792137,
      "num_tokens": 26921431.0,
      "step": 13765
    },
    {
      "entropy": 6.1088663101196286,
      "epoch": 1.376468236117359,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004817910142392438,
      "loss": 5.8687,
      "mean_token_accuracy": 0.16177405864000322,
      "num_tokens": 26931990.0,
      "step": 13770
    },
    {
      "entropy": 6.126968431472778,
      "epoch": 1.3769680611785875,
      "grad_norm": 1.328125,
      "learning_rate": 0.00048177694686331047,
      "loss": 5.9213,
      "mean_token_accuracy": 0.16035618185997008,
      "num_tokens": 26942801.0,
      "step": 13775
    },
    {
      "entropy": 6.157605171203613,
      "epoch": 1.377467886239816,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004817628742849224,
      "loss": 5.9391,
      "mean_token_accuracy": 0.15881012231111527,
      "num_tokens": 26952896.0,
      "step": 13780
    },
    {
      "entropy": 6.166085433959961,
      "epoch": 1.3779677113010447,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00048174879650443364,
      "loss": 5.976,
      "mean_token_accuracy": 0.15370875224471092,
      "num_tokens": 26963170.0,
      "step": 13785
    },
    {
      "entropy": 6.185829544067383,
      "epoch": 1.3784675363622731,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048173471352219835,
      "loss": 5.9575,
      "mean_token_accuracy": 0.15716409757733346,
      "num_tokens": 26972937.0,
      "step": 13790
    },
    {
      "entropy": 6.240061950683594,
      "epoch": 1.3789673614235018,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004817206253385708,
      "loss": 5.9841,
      "mean_token_accuracy": 0.15869363397359848,
      "num_tokens": 26985115.0,
      "step": 13795
    },
    {
      "entropy": 6.163411951065063,
      "epoch": 1.3794671864847303,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00048170653195390555,
      "loss": 5.9306,
      "mean_token_accuracy": 0.1638138011097908,
      "num_tokens": 26993658.0,
      "step": 13800
    },
    {
      "entropy": 6.116409587860107,
      "epoch": 1.379967011545959,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004816924333685572,
      "loss": 5.9143,
      "mean_token_accuracy": 0.15303592756390572,
      "num_tokens": 27003877.0,
      "step": 13805
    },
    {
      "entropy": 6.0737035274505615,
      "epoch": 1.3804668366071875,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004816783295828804,
      "loss": 5.7906,
      "mean_token_accuracy": 0.17063479274511337,
      "num_tokens": 27012125.0,
      "step": 13810
    },
    {
      "entropy": 6.142703104019165,
      "epoch": 1.380966661668416,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00048166422059722996,
      "loss": 5.9892,
      "mean_token_accuracy": 0.1559419110417366,
      "num_tokens": 27022158.0,
      "step": 13815
    },
    {
      "entropy": 6.110621786117553,
      "epoch": 1.3814664867296447,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004816501064119609,
      "loss": 5.8921,
      "mean_token_accuracy": 0.16148971170186996,
      "num_tokens": 27031966.0,
      "step": 13820
    },
    {
      "entropy": 6.175729990005493,
      "epoch": 1.3819663117908731,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004816359870274284,
      "loss": 5.8612,
      "mean_token_accuracy": 0.16162631064653396,
      "num_tokens": 27040805.0,
      "step": 13825
    },
    {
      "entropy": 6.131179237365723,
      "epoch": 1.3824661368521018,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004816218624439876,
      "loss": 5.9535,
      "mean_token_accuracy": 0.15544168949127196,
      "num_tokens": 27049210.0,
      "step": 13830
    },
    {
      "entropy": 6.054516124725342,
      "epoch": 1.3829659619133303,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048160773266199384,
      "loss": 5.7928,
      "mean_token_accuracy": 0.16536794155836104,
      "num_tokens": 27058980.0,
      "step": 13835
    },
    {
      "entropy": 6.185720729827881,
      "epoch": 1.383465786974559,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004815935976818028,
      "loss": 5.9276,
      "mean_token_accuracy": 0.16246310174465178,
      "num_tokens": 27068539.0,
      "step": 13840
    },
    {
      "entropy": 6.14019546508789,
      "epoch": 1.3839656120357875,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00048157945750377005,
      "loss": 5.8319,
      "mean_token_accuracy": 0.16166296154260634,
      "num_tokens": 27077500.0,
      "step": 13845
    },
    {
      "entropy": 6.027490949630737,
      "epoch": 1.384465437097016,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004815653121282512,
      "loss": 5.7851,
      "mean_token_accuracy": 0.16689274460077286,
      "num_tokens": 27087039.0,
      "step": 13850
    },
    {
      "entropy": 6.0403172969818115,
      "epoch": 1.3849652621582447,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048155116155560227,
      "loss": 5.8654,
      "mean_token_accuracy": 0.1638965755701065,
      "num_tokens": 27096575.0,
      "step": 13855
    },
    {
      "entropy": 6.169073820114136,
      "epoch": 1.3854650872194731,
      "grad_norm": 1.625,
      "learning_rate": 0.00048153700578617924,
      "loss": 5.9574,
      "mean_token_accuracy": 0.1572210446000099,
      "num_tokens": 27106246.0,
      "step": 13860
    },
    {
      "entropy": 6.152325296401978,
      "epoch": 1.3859649122807016,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004815228448203384,
      "loss": 5.9308,
      "mean_token_accuracy": 0.16242442280054092,
      "num_tokens": 27115529.0,
      "step": 13865
    },
    {
      "entropy": 6.134983539581299,
      "epoch": 1.3864647373419303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004815086786584358,
      "loss": 5.8127,
      "mean_token_accuracy": 0.164150208234787,
      "num_tokens": 27125204.0,
      "step": 13870
    },
    {
      "entropy": 6.045771551132202,
      "epoch": 1.386964562403159,
      "grad_norm": 1.53125,
      "learning_rate": 0.000481494507300828,
      "loss": 5.8696,
      "mean_token_accuracy": 0.1608981430530548,
      "num_tokens": 27135053.0,
      "step": 13875
    },
    {
      "entropy": 6.191318941116333,
      "epoch": 1.3874643874643875,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004814803307478716,
      "loss": 5.9147,
      "mean_token_accuracy": 0.15449856221675873,
      "num_tokens": 27144863.0,
      "step": 13880
    },
    {
      "entropy": 6.096641921997071,
      "epoch": 1.387964212525616,
      "grad_norm": 1.6796875,
      "learning_rate": 0.00048146614899992313,
      "loss": 5.831,
      "mean_token_accuracy": 0.16545611321926118,
      "num_tokens": 27154178.0,
      "step": 13885
    },
    {
      "entropy": 6.1149076461792,
      "epoch": 1.3884640375868447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048145196205733954,
      "loss": 5.8078,
      "mean_token_accuracy": 0.162298721075058,
      "num_tokens": 27164218.0,
      "step": 13890
    },
    {
      "entropy": 6.118490076065063,
      "epoch": 1.3889638626480731,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048143776992047767,
      "loss": 5.8352,
      "mean_token_accuracy": 0.15735221058130264,
      "num_tokens": 27174255.0,
      "step": 13895
    },
    {
      "entropy": 6.157198476791382,
      "epoch": 1.3894636877093016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048142357258969465,
      "loss": 5.9438,
      "mean_token_accuracy": 0.15201016068458556,
      "num_tokens": 27184989.0,
      "step": 13900
    },
    {
      "entropy": 6.114120054244995,
      "epoch": 1.3899635127705303,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004814093700653477,
      "loss": 5.8477,
      "mean_token_accuracy": 0.1661065325140953,
      "num_tokens": 27196249.0,
      "step": 13905
    },
    {
      "entropy": 6.077417707443237,
      "epoch": 1.390463337831759,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048139516234779403,
      "loss": 5.7859,
      "mean_token_accuracy": 0.17097297757864,
      "num_tokens": 27204997.0,
      "step": 13910
    },
    {
      "entropy": 6.129012250900269,
      "epoch": 1.3909631628929875,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00048138094943739123,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16172044426202775,
      "num_tokens": 27214344.0,
      "step": 13915
    },
    {
      "entropy": 6.155151796340943,
      "epoch": 1.391462987954216,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00048136673133449687,
      "loss": 5.8783,
      "mean_token_accuracy": 0.15825018733739854,
      "num_tokens": 27223950.0,
      "step": 13920
    },
    {
      "entropy": 6.028068590164184,
      "epoch": 1.3919628130154447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004813525080394686,
      "loss": 5.6914,
      "mean_token_accuracy": 0.16767370253801345,
      "num_tokens": 27233371.0,
      "step": 13925
    },
    {
      "entropy": 6.125385904312134,
      "epoch": 1.3924626380766731,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004813382795526644,
      "loss": 5.9383,
      "mean_token_accuracy": 0.16219381392002105,
      "num_tokens": 27243413.0,
      "step": 13930
    },
    {
      "entropy": 6.167134809494018,
      "epoch": 1.3929624631379016,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004813240458744421,
      "loss": 5.8513,
      "mean_token_accuracy": 0.16141010373830794,
      "num_tokens": 27254117.0,
      "step": 13935
    },
    {
      "entropy": 6.086665201187134,
      "epoch": 1.3934622881991303,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00048130980700516,
      "loss": 5.9433,
      "mean_token_accuracy": 0.15809763520956038,
      "num_tokens": 27263202.0,
      "step": 13940
    },
    {
      "entropy": 6.082571744918823,
      "epoch": 1.3939621132603588,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004812955629451761,
      "loss": 5.8153,
      "mean_token_accuracy": 0.16160611659288407,
      "num_tokens": 27272830.0,
      "step": 13945
    },
    {
      "entropy": 6.1360499382019045,
      "epoch": 1.3944619383215875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004812813136948491,
      "loss": 5.8548,
      "mean_token_accuracy": 0.16546165868639945,
      "num_tokens": 27283567.0,
      "step": 13950
    },
    {
      "entropy": 6.102085256576538,
      "epoch": 1.394961763382816,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048126705925453716,
      "loss": 5.7733,
      "mean_token_accuracy": 0.16804579496383668,
      "num_tokens": 27292113.0,
      "step": 13955
    },
    {
      "entropy": 5.987590408325195,
      "epoch": 1.3954615884440447,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004812527996245992,
      "loss": 5.8333,
      "mean_token_accuracy": 0.16707413047552108,
      "num_tokens": 27301605.0,
      "step": 13960
    },
    {
      "entropy": 6.100281238555908,
      "epoch": 1.3959614135052731,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004812385348053938,
      "loss": 5.9296,
      "mean_token_accuracy": 0.1572559043765068,
      "num_tokens": 27312453.0,
      "step": 13965
    },
    {
      "entropy": 6.1346252918243405,
      "epoch": 1.3964612385665016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048122426479728,
      "loss": 5.9253,
      "mean_token_accuracy": 0.15773826390504836,
      "num_tokens": 27322641.0,
      "step": 13970
    },
    {
      "entropy": 6.124874591827393,
      "epoch": 1.3969610636277303,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004812099896006167,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17333748936653137,
      "num_tokens": 27332283.0,
      "step": 13975
    },
    {
      "entropy": 6.0636121273040775,
      "epoch": 1.3974608886889588,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004811957092157631,
      "loss": 5.8188,
      "mean_token_accuracy": 0.16680866628885269,
      "num_tokens": 27341203.0,
      "step": 13980
    },
    {
      "entropy": 6.058031845092773,
      "epoch": 1.3979607137501875,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048118142364307864,
      "loss": 5.9137,
      "mean_token_accuracy": 0.1649721771478653,
      "num_tokens": 27351263.0,
      "step": 13985
    },
    {
      "entropy": 6.090591335296631,
      "epoch": 1.398460538811416,
      "grad_norm": 1.375,
      "learning_rate": 0.0004811671328829225,
      "loss": 5.8341,
      "mean_token_accuracy": 0.164004909992218,
      "num_tokens": 27361001.0,
      "step": 13990
    },
    {
      "entropy": 6.139341259002686,
      "epoch": 1.3989603638726447,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048115283693565445,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16129040718078613,
      "num_tokens": 27370097.0,
      "step": 13995
    },
    {
      "entropy": 6.069870710372925,
      "epoch": 1.3994601889338731,
      "grad_norm": 1.4453125,
      "learning_rate": 0.000481138535801634,
      "loss": 5.8763,
      "mean_token_accuracy": 0.16523979902267455,
      "num_tokens": 27379795.0,
      "step": 14000
    },
    {
      "entropy": 6.0448414325714115,
      "epoch": 1.3999600139951016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000481124229481221,
      "loss": 5.8772,
      "mean_token_accuracy": 0.16665548384189605,
      "num_tokens": 27389515.0,
      "step": 14005
    },
    {
      "entropy": 6.130039167404175,
      "epoch": 1.4004598390563303,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004811099179747755,
      "loss": 5.7807,
      "mean_token_accuracy": 0.18001209795475007,
      "num_tokens": 27400177.0,
      "step": 14010
    },
    {
      "entropy": 6.171075868606567,
      "epoch": 1.4009596641175588,
      "grad_norm": 1.25,
      "learning_rate": 0.00048109560128265747,
      "loss": 5.8543,
      "mean_token_accuracy": 0.15925300270318984,
      "num_tokens": 27409381.0,
      "step": 14015
    },
    {
      "entropy": 6.114352178573609,
      "epoch": 1.4014594891787875,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004810812794052271,
      "loss": 5.9659,
      "mean_token_accuracy": 0.15845771729946137,
      "num_tokens": 27419897.0,
      "step": 14020
    },
    {
      "entropy": 6.222824811935425,
      "epoch": 1.401959314240016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004810669523428449,
      "loss": 6.0706,
      "mean_token_accuracy": 0.14748722389340402,
      "num_tokens": 27430693.0,
      "step": 14025
    },
    {
      "entropy": 6.264172220230103,
      "epoch": 1.4024591393012447,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000481052620095871,
      "loss": 6.0102,
      "mean_token_accuracy": 0.15318381488323213,
      "num_tokens": 27440834.0,
      "step": 14030
    },
    {
      "entropy": 6.085946130752563,
      "epoch": 1.4029589643624731,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004810382826646664,
      "loss": 5.7799,
      "mean_token_accuracy": 0.16960890740156173,
      "num_tokens": 27450166.0,
      "step": 14035
    },
    {
      "entropy": 6.0345203399658205,
      "epoch": 1.4034587894237016,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004810239400495915,
      "loss": 5.8715,
      "mean_token_accuracy": 0.15571456998586655,
      "num_tokens": 27459863.0,
      "step": 14040
    },
    {
      "entropy": 6.072359609603882,
      "epoch": 1.4039586144849303,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004810095922510073,
      "loss": 5.7952,
      "mean_token_accuracy": 0.1644686669111252,
      "num_tokens": 27469118.0,
      "step": 14045
    },
    {
      "entropy": 6.148534345626831,
      "epoch": 1.4044584395461588,
      "grad_norm": 1.40625,
      "learning_rate": 0.00048099523926927487,
      "loss": 5.8729,
      "mean_token_accuracy": 0.16246576830744744,
      "num_tokens": 27479550.0,
      "step": 14050
    },
    {
      "entropy": 6.142660236358642,
      "epoch": 1.4049582646073875,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004809808811047552,
      "loss": 5.9161,
      "mean_token_accuracy": 0.15946181863546371,
      "num_tokens": 27489625.0,
      "step": 14055
    },
    {
      "entropy": 6.239021968841553,
      "epoch": 1.405458089668616,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048096651775780943,
      "loss": 5.9611,
      "mean_token_accuracy": 0.1540763221681118,
      "num_tokens": 27501459.0,
      "step": 14060
    },
    {
      "entropy": 6.137648344039917,
      "epoch": 1.4059579147298447,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004809521492287992,
      "loss": 5.7912,
      "mean_token_accuracy": 0.16124285236001015,
      "num_tokens": 27511450.0,
      "step": 14065
    },
    {
      "entropy": 6.02143030166626,
      "epoch": 1.4064577397910731,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004809377755180858,
      "loss": 5.7558,
      "mean_token_accuracy": 0.1690579757094383,
      "num_tokens": 27520783.0,
      "step": 14070
    },
    {
      "entropy": 6.167116355895996,
      "epoch": 1.4069575648523016,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004809233966260311,
      "loss": 5.883,
      "mean_token_accuracy": 0.15916331931948663,
      "num_tokens": 27529433.0,
      "step": 14075
    },
    {
      "entropy": 6.105887413024902,
      "epoch": 1.4074573899135303,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004809090125529966,
      "loss": 5.8406,
      "mean_token_accuracy": 0.16154496520757675,
      "num_tokens": 27539278.0,
      "step": 14080
    },
    {
      "entropy": 6.119472408294678,
      "epoch": 1.4079572149747588,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00048089462329934437,
      "loss": 5.9151,
      "mean_token_accuracy": 0.16004838794469833,
      "num_tokens": 27548474.0,
      "step": 14085
    },
    {
      "entropy": 6.116377115249634,
      "epoch": 1.4084570400359875,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00048088022886543636,
      "loss": 5.9182,
      "mean_token_accuracy": 0.16162464618682862,
      "num_tokens": 27557757.0,
      "step": 14090
    },
    {
      "entropy": 6.134588813781738,
      "epoch": 1.408956865097216,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004808658292516348,
      "loss": 5.9702,
      "mean_token_accuracy": 0.15040709376335143,
      "num_tokens": 27567971.0,
      "step": 14095
    },
    {
      "entropy": 6.193719673156738,
      "epoch": 1.4094566901584447,
      "grad_norm": 1.265625,
      "learning_rate": 0.00048085142445830197,
      "loss": 5.9771,
      "mean_token_accuracy": 0.14814667254686356,
      "num_tokens": 27578135.0,
      "step": 14100
    },
    {
      "entropy": 6.172084617614746,
      "epoch": 1.4099565152196731,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00048083701448580027,
      "loss": 5.8458,
      "mean_token_accuracy": 0.16409674137830735,
      "num_tokens": 27587459.0,
      "step": 14105
    },
    {
      "entropy": 6.197263956069946,
      "epoch": 1.4104563402809016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048082259933449224,
      "loss": 5.9325,
      "mean_token_accuracy": 0.16178551763296128,
      "num_tokens": 27596494.0,
      "step": 14110
    },
    {
      "entropy": 6.053715705871582,
      "epoch": 1.4109561653421303,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004808081790047406,
      "loss": 5.7952,
      "mean_token_accuracy": 0.17472569346427919,
      "num_tokens": 27605482.0,
      "step": 14115
    },
    {
      "entropy": 6.076188707351685,
      "epoch": 1.4114559904033588,
      "grad_norm": 1.234375,
      "learning_rate": 0.00048079375349690807,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16015429869294168,
      "num_tokens": 27615386.0,
      "step": 14120
    },
    {
      "entropy": 6.175529336929321,
      "epoch": 1.4119558154645873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048077932281135775,
      "loss": 5.911,
      "mean_token_accuracy": 0.16377355605363847,
      "num_tokens": 27624834.0,
      "step": 14125
    },
    {
      "entropy": 6.1573728084564205,
      "epoch": 1.412455640525816,
      "grad_norm": 1.390625,
      "learning_rate": 0.00048076488694845257,
      "loss": 5.8231,
      "mean_token_accuracy": 0.1622598186135292,
      "num_tokens": 27634087.0,
      "step": 14130
    },
    {
      "entropy": 6.130871391296386,
      "epoch": 1.4129554655870447,
      "grad_norm": 1.1875,
      "learning_rate": 0.00048075044590855574,
      "loss": 5.7668,
      "mean_token_accuracy": 0.17290389388799668,
      "num_tokens": 27643739.0,
      "step": 14135
    },
    {
      "entropy": 6.098408794403076,
      "epoch": 1.4134552906482731,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048073599969203064,
      "loss": 5.9041,
      "mean_token_accuracy": 0.15937146097421645,
      "num_tokens": 27652594.0,
      "step": 14140
    },
    {
      "entropy": 6.1507586479187015,
      "epoch": 1.4139551157095016,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004807215482992408,
      "loss": 5.9333,
      "mean_token_accuracy": 0.15585235953330995,
      "num_tokens": 27662766.0,
      "step": 14145
    },
    {
      "entropy": 6.168141508102417,
      "epoch": 1.4144549407707303,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004807070917305497,
      "loss": 5.9181,
      "mean_token_accuracy": 0.15718501806259155,
      "num_tokens": 27673547.0,
      "step": 14150
    },
    {
      "entropy": 6.178928995132447,
      "epoch": 1.4149547658319588,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00048069262998632115,
      "loss": 5.8977,
      "mean_token_accuracy": 0.15978171080350875,
      "num_tokens": 27683056.0,
      "step": 14155
    },
    {
      "entropy": 6.128573846817017,
      "epoch": 1.4154545908931873,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004806781630669189,
      "loss": 5.8912,
      "mean_token_accuracy": 0.16052587777376176,
      "num_tokens": 27693049.0,
      "step": 14160
    },
    {
      "entropy": 6.171237134933472,
      "epoch": 1.415954415954416,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00048066369097270694,
      "loss": 5.8603,
      "mean_token_accuracy": 0.15596067160367966,
      "num_tokens": 27703712.0,
      "step": 14165
    },
    {
      "entropy": 6.182238960266114,
      "epoch": 1.4164542410156447,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004806492137040495,
      "loss": 5.8845,
      "mean_token_accuracy": 0.15887209475040437,
      "num_tokens": 27713460.0,
      "step": 14170
    },
    {
      "entropy": 6.197443103790283,
      "epoch": 1.4169540660768731,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00048063473126131086,
      "loss": 5.9246,
      "mean_token_accuracy": 0.16057073473930358,
      "num_tokens": 27722942.0,
      "step": 14175
    },
    {
      "entropy": 6.124551773071289,
      "epoch": 1.4174538911381016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004806202436448551,
      "loss": 5.9002,
      "mean_token_accuracy": 0.15448820739984512,
      "num_tokens": 27732738.0,
      "step": 14180
    },
    {
      "entropy": 6.177687072753907,
      "epoch": 1.4179537161993303,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004806057508550471,
      "loss": 5.9331,
      "mean_token_accuracy": 0.15591414794325828,
      "num_tokens": 27742169.0,
      "step": 14185
    },
    {
      "entropy": 6.170655965805054,
      "epoch": 1.4184535412605588,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004805912528922512,
      "loss": 5.889,
      "mean_token_accuracy": 0.1644885301589966,
      "num_tokens": 27752539.0,
      "step": 14190
    },
    {
      "entropy": 6.184833097457886,
      "epoch": 1.4189533663217873,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004805767497568323,
      "loss": 5.9516,
      "mean_token_accuracy": 0.15195757299661636,
      "num_tokens": 27763494.0,
      "step": 14195
    },
    {
      "entropy": 6.069012212753296,
      "epoch": 1.419453191383016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048056224144915534,
      "loss": 5.77,
      "mean_token_accuracy": 0.17369400560855866,
      "num_tokens": 27772884.0,
      "step": 14200
    },
    {
      "entropy": 6.046985912322998,
      "epoch": 1.4199530164442444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048054772796958517,
      "loss": 5.7834,
      "mean_token_accuracy": 0.17454491853713988,
      "num_tokens": 27782447.0,
      "step": 14205
    },
    {
      "entropy": 6.1596465587615965,
      "epoch": 1.4204528415054731,
      "grad_norm": 1.25,
      "learning_rate": 0.00048053320931848713,
      "loss": 5.855,
      "mean_token_accuracy": 0.16317107528448105,
      "num_tokens": 27790781.0,
      "step": 14210
    },
    {
      "entropy": 6.140770721435547,
      "epoch": 1.4209526665667016,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00048051868549622644,
      "loss": 5.7867,
      "mean_token_accuracy": 0.17033137679100036,
      "num_tokens": 27799781.0,
      "step": 14215
    },
    {
      "entropy": 6.013639974594116,
      "epoch": 1.4214524916279303,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00048050415650316853,
      "loss": 5.84,
      "mean_token_accuracy": 0.15950441211462021,
      "num_tokens": 27809067.0,
      "step": 14220
    },
    {
      "entropy": 6.1785845279693605,
      "epoch": 1.4219523166891588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004804896223396789,
      "loss": 5.9277,
      "mean_token_accuracy": 0.1541512832045555,
      "num_tokens": 27819194.0,
      "step": 14225
    },
    {
      "entropy": 6.195931100845337,
      "epoch": 1.4224521417503873,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004804750830061232,
      "loss": 5.8825,
      "mean_token_accuracy": 0.156583009660244,
      "num_tokens": 27830179.0,
      "step": 14230
    },
    {
      "entropy": 6.165843296051025,
      "epoch": 1.422951966811616,
      "grad_norm": 1.796875,
      "learning_rate": 0.0004804605385028673,
      "loss": 5.8679,
      "mean_token_accuracy": 0.15971438735723495,
      "num_tokens": 27840476.0,
      "step": 14235
    },
    {
      "entropy": 6.0903496742248535,
      "epoch": 1.4234517918728444,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004804459888302772,
      "loss": 5.7724,
      "mean_token_accuracy": 0.16487712115049363,
      "num_tokens": 27850247.0,
      "step": 14240
    },
    {
      "entropy": 6.083036851882935,
      "epoch": 1.4239516169340731,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00048043143398871874,
      "loss": 5.9114,
      "mean_token_accuracy": 0.15706351771950722,
      "num_tokens": 27860290.0,
      "step": 14245
    },
    {
      "entropy": 6.102430629730224,
      "epoch": 1.4244514419953016,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004804168739785583,
      "loss": 5.908,
      "mean_token_accuracy": 0.15358297079801558,
      "num_tokens": 27870506.0,
      "step": 14250
    },
    {
      "entropy": 6.250571966171265,
      "epoch": 1.4249512670565303,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004804023088001623,
      "loss": 5.9583,
      "mean_token_accuracy": 0.14947654753923417,
      "num_tokens": 27880679.0,
      "step": 14255
    },
    {
      "entropy": 6.113212585449219,
      "epoch": 1.4254510921177588,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004803877384538969,
      "loss": 5.8415,
      "mean_token_accuracy": 0.16503433585166932,
      "num_tokens": 27891532.0,
      "step": 14260
    },
    {
      "entropy": 5.879591083526611,
      "epoch": 1.4259509171789873,
      "grad_norm": 1.5390625,
      "learning_rate": 0.000480373162940129,
      "loss": 5.6479,
      "mean_token_accuracy": 0.18816069066524505,
      "num_tokens": 27902659.0,
      "step": 14265
    },
    {
      "entropy": 6.115726470947266,
      "epoch": 1.426450742240216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048035858225922506,
      "loss": 5.8454,
      "mean_token_accuracy": 0.16464049518108367,
      "num_tokens": 27912422.0,
      "step": 14270
    },
    {
      "entropy": 6.113480281829834,
      "epoch": 1.4269505673014444,
      "grad_norm": 1.953125,
      "learning_rate": 0.00048034399641155206,
      "loss": 5.8583,
      "mean_token_accuracy": 0.1586094781756401,
      "num_tokens": 27922898.0,
      "step": 14275
    },
    {
      "entropy": 6.1192607402801515,
      "epoch": 1.4274503923626731,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00048032940539747697,
      "loss": 5.878,
      "mean_token_accuracy": 0.16154274195432664,
      "num_tokens": 27932915.0,
      "step": 14280
    },
    {
      "entropy": 6.150334358215332,
      "epoch": 1.4279502174239016,
      "grad_norm": 1.296875,
      "learning_rate": 0.00048031480921736686,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16029138937592508,
      "num_tokens": 27942760.0,
      "step": 14285
    },
    {
      "entropy": 6.158548259735108,
      "epoch": 1.4284500424851303,
      "grad_norm": 1.203125,
      "learning_rate": 0.000480300207871589,
      "loss": 5.9191,
      "mean_token_accuracy": 0.15693411529064177,
      "num_tokens": 27952898.0,
      "step": 14290
    },
    {
      "entropy": 6.102324390411377,
      "epoch": 1.4289498675463588,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048028560136051076,
      "loss": 5.843,
      "mean_token_accuracy": 0.16362176239490508,
      "num_tokens": 27962292.0,
      "step": 14295
    },
    {
      "entropy": 6.128807067871094,
      "epoch": 1.4294496926075873,
      "grad_norm": 1.359375,
      "learning_rate": 0.00048027098968449954,
      "loss": 5.8937,
      "mean_token_accuracy": 0.15985726863145827,
      "num_tokens": 27972480.0,
      "step": 14300
    },
    {
      "entropy": 6.193449974060059,
      "epoch": 1.429949517668816,
      "grad_norm": 1.5,
      "learning_rate": 0.0004802563728439231,
      "loss": 5.9611,
      "mean_token_accuracy": 0.15251852124929427,
      "num_tokens": 27982224.0,
      "step": 14305
    },
    {
      "entropy": 6.17263240814209,
      "epoch": 1.4304493427300444,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004802417508391491,
      "loss": 5.9485,
      "mean_token_accuracy": 0.16087263971567153,
      "num_tokens": 27991512.0,
      "step": 14310
    },
    {
      "entropy": 6.111741542816162,
      "epoch": 1.4309491677912731,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004802271236705455,
      "loss": 5.7502,
      "mean_token_accuracy": 0.16769987046718599,
      "num_tokens": 28001392.0,
      "step": 14315
    },
    {
      "entropy": 6.005036163330078,
      "epoch": 1.4314489928525016,
      "grad_norm": 1.28125,
      "learning_rate": 0.00048021249133848025,
      "loss": 5.7397,
      "mean_token_accuracy": 0.1675102964043617,
      "num_tokens": 28009934.0,
      "step": 14320
    },
    {
      "entropy": 6.095283460617066,
      "epoch": 1.4319488179137303,
      "grad_norm": 1.3125,
      "learning_rate": 0.00048019785384332154,
      "loss": 5.8807,
      "mean_token_accuracy": 0.160319022834301,
      "num_tokens": 28019485.0,
      "step": 14325
    },
    {
      "entropy": 6.1149028778076175,
      "epoch": 1.4324486429749588,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004801832111854376,
      "loss": 5.885,
      "mean_token_accuracy": 0.15777227729558946,
      "num_tokens": 28028666.0,
      "step": 14330
    },
    {
      "entropy": 6.1116303443908695,
      "epoch": 1.4329484680361873,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004801685633651969,
      "loss": 5.8772,
      "mean_token_accuracy": 0.15705778151750566,
      "num_tokens": 28038677.0,
      "step": 14335
    },
    {
      "entropy": 6.186305904388428,
      "epoch": 1.433448293097416,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004801539103829679,
      "loss": 5.9708,
      "mean_token_accuracy": 0.14852447658777237,
      "num_tokens": 28048580.0,
      "step": 14340
    },
    {
      "entropy": 6.174959421157837,
      "epoch": 1.4339481181586444,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004801392522391193,
      "loss": 5.8875,
      "mean_token_accuracy": 0.15756209269165994,
      "num_tokens": 28058958.0,
      "step": 14345
    },
    {
      "entropy": 6.124604797363281,
      "epoch": 1.4344479432198731,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00048012458893401995,
      "loss": 5.8682,
      "mean_token_accuracy": 0.16091903001070024,
      "num_tokens": 28068861.0,
      "step": 14350
    },
    {
      "entropy": 6.0634050369262695,
      "epoch": 1.4349477682811016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004801099204680387,
      "loss": 5.9067,
      "mean_token_accuracy": 0.16454484760761262,
      "num_tokens": 28078638.0,
      "step": 14355
    },
    {
      "entropy": 6.133096504211426,
      "epoch": 1.4354475933423303,
      "grad_norm": 1.25,
      "learning_rate": 0.00048009524684154464,
      "loss": 5.8495,
      "mean_token_accuracy": 0.1610197439789772,
      "num_tokens": 28088549.0,
      "step": 14360
    },
    {
      "entropy": 6.251824569702149,
      "epoch": 1.4359474184035588,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004800805680549069,
      "loss": 5.9046,
      "mean_token_accuracy": 0.1600938022136688,
      "num_tokens": 28097743.0,
      "step": 14365
    },
    {
      "entropy": 6.07877197265625,
      "epoch": 1.4364472434647872,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00048006588410849497,
      "loss": 5.8207,
      "mean_token_accuracy": 0.16540497392416,
      "num_tokens": 28107187.0,
      "step": 14370
    },
    {
      "entropy": 6.06516923904419,
      "epoch": 1.436947068526016,
      "grad_norm": 1.171875,
      "learning_rate": 0.000480051195002678,
      "loss": 5.8026,
      "mean_token_accuracy": 0.16269998997449875,
      "num_tokens": 28117730.0,
      "step": 14375
    },
    {
      "entropy": 6.194702100753784,
      "epoch": 1.4374468935872444,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00048003650073782586,
      "loss": 5.9159,
      "mean_token_accuracy": 0.1575499728322029,
      "num_tokens": 28127803.0,
      "step": 14380
    },
    {
      "entropy": 6.178099060058594,
      "epoch": 1.437946718648473,
      "grad_norm": 1.28125,
      "learning_rate": 0.000480021801314308,
      "loss": 5.8602,
      "mean_token_accuracy": 0.16381236612796785,
      "num_tokens": 28137946.0,
      "step": 14385
    },
    {
      "entropy": 6.114961194992065,
      "epoch": 1.4384465437097016,
      "grad_norm": 1.125,
      "learning_rate": 0.0004800070967324944,
      "loss": 5.9403,
      "mean_token_accuracy": 0.15581810474395752,
      "num_tokens": 28147895.0,
      "step": 14390
    },
    {
      "entropy": 6.044751787185669,
      "epoch": 1.4389463687709303,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047999238699275506,
      "loss": 5.846,
      "mean_token_accuracy": 0.169302798807621,
      "num_tokens": 28157641.0,
      "step": 14395
    },
    {
      "entropy": 6.111778831481933,
      "epoch": 1.4394461938321588,
      "grad_norm": 1.25,
      "learning_rate": 0.00047997767209546,
      "loss": 5.8888,
      "mean_token_accuracy": 0.15626642853021622,
      "num_tokens": 28167320.0,
      "step": 14400
    },
    {
      "entropy": 6.120743179321289,
      "epoch": 1.4399460188933872,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004799629520409795,
      "loss": 5.8367,
      "mean_token_accuracy": 0.16600980311632157,
      "num_tokens": 28176697.0,
      "step": 14405
    },
    {
      "entropy": 6.133093070983887,
      "epoch": 1.440445843954616,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047994822682968373,
      "loss": 5.8924,
      "mean_token_accuracy": 0.16539354324340821,
      "num_tokens": 28186688.0,
      "step": 14410
    },
    {
      "entropy": 6.166804838180542,
      "epoch": 1.4409456690158444,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047993349646194343,
      "loss": 5.9894,
      "mean_token_accuracy": 0.15547190383076667,
      "num_tokens": 28197293.0,
      "step": 14415
    },
    {
      "entropy": 6.255355548858643,
      "epoch": 1.441445494077073,
      "grad_norm": 3.390625,
      "learning_rate": 0.00047991876093812904,
      "loss": 5.9856,
      "mean_token_accuracy": 0.15245133489370347,
      "num_tokens": 28207413.0,
      "step": 14420
    },
    {
      "entropy": 6.163256120681763,
      "epoch": 1.4419453191383016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047990402025861143,
      "loss": 5.8773,
      "mean_token_accuracy": 0.16162368953227996,
      "num_tokens": 28217575.0,
      "step": 14425
    },
    {
      "entropy": 6.011227178573608,
      "epoch": 1.4424451441995303,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004798892744237613,
      "loss": 5.7261,
      "mean_token_accuracy": 0.1700431749224663,
      "num_tokens": 28226460.0,
      "step": 14430
    },
    {
      "entropy": 6.04676079750061,
      "epoch": 1.4429449692607588,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047987452343394976,
      "loss": 5.8521,
      "mean_token_accuracy": 0.16183318942785263,
      "num_tokens": 28236689.0,
      "step": 14435
    },
    {
      "entropy": 6.108124876022339,
      "epoch": 1.4434447943219872,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047985976728954794,
      "loss": 5.865,
      "mean_token_accuracy": 0.16307731419801713,
      "num_tokens": 28247057.0,
      "step": 14440
    },
    {
      "entropy": 6.221473932266235,
      "epoch": 1.443944619383216,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004798450059909271,
      "loss": 5.8886,
      "mean_token_accuracy": 0.15970712304115295,
      "num_tokens": 28256475.0,
      "step": 14445
    },
    {
      "entropy": 6.117563152313233,
      "epoch": 1.4444444444444444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004798302395384586,
      "loss": 5.8484,
      "mean_token_accuracy": 0.1598137989640236,
      "num_tokens": 28266120.0,
      "step": 14450
    },
    {
      "entropy": 6.073667287826538,
      "epoch": 1.444944269505673,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047981546793251405,
      "loss": 5.8519,
      "mean_token_accuracy": 0.1594000518321991,
      "num_tokens": 28276325.0,
      "step": 14455
    },
    {
      "entropy": 6.102376651763916,
      "epoch": 1.4454440945669016,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000479800691173465,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16341444477438927,
      "num_tokens": 28286007.0,
      "step": 14460
    },
    {
      "entropy": 6.107060384750366,
      "epoch": 1.4459439196281303,
      "grad_norm": 1.25,
      "learning_rate": 0.0004797859092616832,
      "loss": 5.8466,
      "mean_token_accuracy": 0.15577509999275208,
      "num_tokens": 28296346.0,
      "step": 14465
    },
    {
      "entropy": 6.014606571197509,
      "epoch": 1.4464437446893588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047977112219754064,
      "loss": 5.7331,
      "mean_token_accuracy": 0.17575829103589058,
      "num_tokens": 28306137.0,
      "step": 14470
    },
    {
      "entropy": 6.142667531967163,
      "epoch": 1.4469435697505872,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004797563299814094,
      "loss": 5.9694,
      "mean_token_accuracy": 0.1518622189760208,
      "num_tokens": 28316369.0,
      "step": 14475
    },
    {
      "entropy": 6.1949272632598875,
      "epoch": 1.447443394811816,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047974153261366156,
      "loss": 5.8505,
      "mean_token_accuracy": 0.16391684710979462,
      "num_tokens": 28325338.0,
      "step": 14480
    },
    {
      "entropy": 6.084573125839233,
      "epoch": 1.4479432198730444,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004797267300946694,
      "loss": 5.8456,
      "mean_token_accuracy": 0.16104142665863036,
      "num_tokens": 28334929.0,
      "step": 14485
    },
    {
      "entropy": 6.142187595367432,
      "epoch": 1.448443044934273,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047971192242480544,
      "loss": 5.901,
      "mean_token_accuracy": 0.16370442360639573,
      "num_tokens": 28345157.0,
      "step": 14490
    },
    {
      "entropy": 6.08145055770874,
      "epoch": 1.4489428699955016,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004796971096044422,
      "loss": 5.7909,
      "mean_token_accuracy": 0.1635267600417137,
      "num_tokens": 28354277.0,
      "step": 14495
    },
    {
      "entropy": 6.078655195236206,
      "epoch": 1.44944269505673,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004796822916339523,
      "loss": 5.8034,
      "mean_token_accuracy": 0.16733310669660567,
      "num_tokens": 28364154.0,
      "step": 14500
    },
    {
      "entropy": 6.109423351287842,
      "epoch": 1.4499425201179588,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047966746851370865,
      "loss": 5.8813,
      "mean_token_accuracy": 0.16106073409318925,
      "num_tokens": 28375260.0,
      "step": 14505
    },
    {
      "entropy": 6.1359468460083,
      "epoch": 1.4504423451791872,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004796526402440841,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16572780460119246,
      "num_tokens": 28384359.0,
      "step": 14510
    },
    {
      "entropy": 6.103290605545044,
      "epoch": 1.450942170240416,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004796378068254518,
      "loss": 5.8745,
      "mean_token_accuracy": 0.16320171803236008,
      "num_tokens": 28394389.0,
      "step": 14515
    },
    {
      "entropy": 6.097010564804077,
      "epoch": 1.4514419953016444,
      "grad_norm": 1.296875,
      "learning_rate": 0.000479622968258185,
      "loss": 5.8711,
      "mean_token_accuracy": 0.15855445116758346,
      "num_tokens": 28403495.0,
      "step": 14520
    },
    {
      "entropy": 6.161702680587768,
      "epoch": 1.451941820362873,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004796081245426569,
      "loss": 5.9139,
      "mean_token_accuracy": 0.1594494789838791,
      "num_tokens": 28413132.0,
      "step": 14525
    },
    {
      "entropy": 6.191201543807983,
      "epoch": 1.4524416454241016,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047959327567924115,
      "loss": 5.8604,
      "mean_token_accuracy": 0.1627531960606575,
      "num_tokens": 28423815.0,
      "step": 14530
    },
    {
      "entropy": 6.180623960494995,
      "epoch": 1.45294147048533,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047957842166831103,
      "loss": 5.8272,
      "mean_token_accuracy": 0.15769873261451722,
      "num_tokens": 28433682.0,
      "step": 14535
    },
    {
      "entropy": 6.124186897277832,
      "epoch": 1.4534412955465588,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047956356251024066,
      "loss": 5.8287,
      "mean_token_accuracy": 0.1700795590877533,
      "num_tokens": 28442388.0,
      "step": 14540
    },
    {
      "entropy": 6.109618377685547,
      "epoch": 1.4539411206077872,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004795486982054036,
      "loss": 5.8494,
      "mean_token_accuracy": 0.16241691410541534,
      "num_tokens": 28452753.0,
      "step": 14545
    },
    {
      "entropy": 6.140168571472168,
      "epoch": 1.454440945669016,
      "grad_norm": 1.7578125,
      "learning_rate": 0.000479533828754174,
      "loss": 5.9442,
      "mean_token_accuracy": 0.1612556718289852,
      "num_tokens": 28462818.0,
      "step": 14550
    },
    {
      "entropy": 6.139218091964722,
      "epoch": 1.4549407707302444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004795189541569258,
      "loss": 5.8826,
      "mean_token_accuracy": 0.16183040887117386,
      "num_tokens": 28472208.0,
      "step": 14555
    },
    {
      "entropy": 6.171624279022216,
      "epoch": 1.455440595791473,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047950407441403343,
      "loss": 5.8696,
      "mean_token_accuracy": 0.16190688461065292,
      "num_tokens": 28480887.0,
      "step": 14560
    },
    {
      "entropy": 6.143547630310058,
      "epoch": 1.4559404208527016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004794891895258711,
      "loss": 5.9425,
      "mean_token_accuracy": 0.15840804129838942,
      "num_tokens": 28491805.0,
      "step": 14565
    },
    {
      "entropy": 6.12728443145752,
      "epoch": 1.45644024591393,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004794742994928134,
      "loss": 5.9062,
      "mean_token_accuracy": 0.16192419677972794,
      "num_tokens": 28502913.0,
      "step": 14570
    },
    {
      "entropy": 6.176183223724365,
      "epoch": 1.4569400709751588,
      "grad_norm": 1.4609375,
      "learning_rate": 0.000479459404315235,
      "loss": 5.8055,
      "mean_token_accuracy": 0.15854290276765823,
      "num_tokens": 28513143.0,
      "step": 14575
    },
    {
      "entropy": 6.164974021911621,
      "epoch": 1.4574398960363872,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047944450399351054,
      "loss": 5.8889,
      "mean_token_accuracy": 0.1610184833407402,
      "num_tokens": 28522211.0,
      "step": 14580
    },
    {
      "entropy": 6.1216721534729,
      "epoch": 1.457939721097616,
      "grad_norm": 2.03125,
      "learning_rate": 0.000479429598528015,
      "loss": 5.9144,
      "mean_token_accuracy": 0.15820422247052193,
      "num_tokens": 28531672.0,
      "step": 14585
    },
    {
      "entropy": 6.105369901657104,
      "epoch": 1.4584395461588444,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004794146879191233,
      "loss": 5.8392,
      "mean_token_accuracy": 0.16402121782302856,
      "num_tokens": 28541614.0,
      "step": 14590
    },
    {
      "entropy": 6.1976540088653564,
      "epoch": 1.458939371220073,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004793997721672107,
      "loss": 5.937,
      "mean_token_accuracy": 0.15269664153456688,
      "num_tokens": 28550353.0,
      "step": 14595
    },
    {
      "entropy": 6.115636348724365,
      "epoch": 1.4594391962813016,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047938485127265243,
      "loss": 5.8429,
      "mean_token_accuracy": 0.1628204792737961,
      "num_tokens": 28559772.0,
      "step": 14600
    },
    {
      "entropy": 6.129355955123901,
      "epoch": 1.45993902134253,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004793699252358239,
      "loss": 5.8759,
      "mean_token_accuracy": 0.15915580987930297,
      "num_tokens": 28569564.0,
      "step": 14605
    },
    {
      "entropy": 6.0863288879394535,
      "epoch": 1.4604388464037588,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004793549940571006,
      "loss": 5.8921,
      "mean_token_accuracy": 0.15911906957626343,
      "num_tokens": 28579473.0,
      "step": 14610
    },
    {
      "entropy": 6.149911642074585,
      "epoch": 1.4609386714649872,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047934005773685824,
      "loss": 5.8721,
      "mean_token_accuracy": 0.1584754019975662,
      "num_tokens": 28588407.0,
      "step": 14615
    },
    {
      "entropy": 6.164475393295288,
      "epoch": 1.461438496526216,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047932511627547266,
      "loss": 5.9879,
      "mean_token_accuracy": 0.1551612690091133,
      "num_tokens": 28599931.0,
      "step": 14620
    },
    {
      "entropy": 5.937235975265503,
      "epoch": 1.4619383215874444,
      "grad_norm": 1.7421875,
      "learning_rate": 0.0004793101696733197,
      "loss": 5.6486,
      "mean_token_accuracy": 0.1821054771542549,
      "num_tokens": 28609317.0,
      "step": 14625
    },
    {
      "entropy": 6.0786644458770756,
      "epoch": 1.4624381466486729,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047929521793077545,
      "loss": 5.9327,
      "mean_token_accuracy": 0.1560387820005417,
      "num_tokens": 28618248.0,
      "step": 14630
    },
    {
      "entropy": 6.163923358917236,
      "epoch": 1.4629379717099016,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004792802610482161,
      "loss": 5.9294,
      "mean_token_accuracy": 0.1587363637983799,
      "num_tokens": 28629046.0,
      "step": 14635
    },
    {
      "entropy": 6.292413330078125,
      "epoch": 1.46343779677113,
      "grad_norm": 1.25,
      "learning_rate": 0.0004792652990260179,
      "loss": 6.0115,
      "mean_token_accuracy": 0.14543681144714354,
      "num_tokens": 28640829.0,
      "step": 14640
    },
    {
      "entropy": 6.201502990722656,
      "epoch": 1.4639376218323585,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004792503318645574,
      "loss": 5.9258,
      "mean_token_accuracy": 0.15602559000253677,
      "num_tokens": 28650812.0,
      "step": 14645
    },
    {
      "entropy": 6.200658321380615,
      "epoch": 1.4644374468935872,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047923535956421104,
      "loss": 5.9195,
      "mean_token_accuracy": 0.15982192754745483,
      "num_tokens": 28661795.0,
      "step": 14650
    },
    {
      "entropy": 6.091787481307984,
      "epoch": 1.464937271954816,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047922038212535565,
      "loss": 5.7629,
      "mean_token_accuracy": 0.16758955866098404,
      "num_tokens": 28670729.0,
      "step": 14655
    },
    {
      "entropy": 6.02588906288147,
      "epoch": 1.4654370970160444,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047920539954836793,
      "loss": 5.7883,
      "mean_token_accuracy": 0.16938893347978592,
      "num_tokens": 28680469.0,
      "step": 14660
    },
    {
      "entropy": 6.076907157897949,
      "epoch": 1.4659369220772729,
      "grad_norm": 1.28125,
      "learning_rate": 0.000479190411833625,
      "loss": 5.8896,
      "mean_token_accuracy": 0.15583025217056273,
      "num_tokens": 28689698.0,
      "step": 14665
    },
    {
      "entropy": 6.178696012496948,
      "epoch": 1.4664367471385016,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047917541898150377,
      "loss": 5.8302,
      "mean_token_accuracy": 0.1650609165430069,
      "num_tokens": 28699861.0,
      "step": 14670
    },
    {
      "entropy": 6.1417553424835205,
      "epoch": 1.46693657219973,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047916042099238154,
      "loss": 5.8505,
      "mean_token_accuracy": 0.15783541575074195,
      "num_tokens": 28709370.0,
      "step": 14675
    },
    {
      "entropy": 6.146949434280396,
      "epoch": 1.4674363972609585,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047914541786663566,
      "loss": 6.1112,
      "mean_token_accuracy": 0.1627703696489334,
      "num_tokens": 28718123.0,
      "step": 14680
    },
    {
      "entropy": 6.2011024951934814,
      "epoch": 1.4679362223221872,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047913040960464363,
      "loss": 5.9502,
      "mean_token_accuracy": 0.15725143104791642,
      "num_tokens": 28727972.0,
      "step": 14685
    },
    {
      "entropy": 6.142420721054077,
      "epoch": 1.468436047383416,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047911539620678305,
      "loss": 5.8579,
      "mean_token_accuracy": 0.15449655279517174,
      "num_tokens": 28737988.0,
      "step": 14690
    },
    {
      "entropy": 6.09318356513977,
      "epoch": 1.4689358724446444,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004791003776734316,
      "loss": 5.9154,
      "mean_token_accuracy": 0.16092705279588698,
      "num_tokens": 28747852.0,
      "step": 14695
    },
    {
      "entropy": 6.039407682418823,
      "epoch": 1.4694356975058729,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047908535400496713,
      "loss": 5.8102,
      "mean_token_accuracy": 0.1638214409351349,
      "num_tokens": 28758652.0,
      "step": 14700
    },
    {
      "entropy": 6.081682538986206,
      "epoch": 1.4699355225671016,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047907032520176765,
      "loss": 5.7801,
      "mean_token_accuracy": 0.16654175221920015,
      "num_tokens": 28767603.0,
      "step": 14705
    },
    {
      "entropy": 6.12945146560669,
      "epoch": 1.47043534762833,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004790552912642113,
      "loss": 5.8589,
      "mean_token_accuracy": 0.16593801379203796,
      "num_tokens": 28777984.0,
      "step": 14710
    },
    {
      "entropy": 6.156704235076904,
      "epoch": 1.4709351726895585,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047904025219267643,
      "loss": 5.9108,
      "mean_token_accuracy": 0.15695432722568511,
      "num_tokens": 28789033.0,
      "step": 14715
    },
    {
      "entropy": 6.065527153015137,
      "epoch": 1.4714349977507872,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047902520798754125,
      "loss": 5.7902,
      "mean_token_accuracy": 0.17116821855306624,
      "num_tokens": 28797728.0,
      "step": 14720
    },
    {
      "entropy": 6.015084457397461,
      "epoch": 1.471934822812016,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004790101586491843,
      "loss": 5.8123,
      "mean_token_accuracy": 0.15724297985434532,
      "num_tokens": 28807505.0,
      "step": 14725
    },
    {
      "entropy": 6.157178544998169,
      "epoch": 1.4724346478732444,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004789951041779843,
      "loss": 5.921,
      "mean_token_accuracy": 0.16030739694833757,
      "num_tokens": 28817289.0,
      "step": 14730
    },
    {
      "entropy": 6.120894813537598,
      "epoch": 1.4729344729344729,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047898004457432,
      "loss": 5.8748,
      "mean_token_accuracy": 0.15950123518705367,
      "num_tokens": 28827697.0,
      "step": 14735
    },
    {
      "entropy": 6.227978038787842,
      "epoch": 1.4734342979957016,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004789649798385702,
      "loss": 5.9236,
      "mean_token_accuracy": 0.1580186054110527,
      "num_tokens": 28837427.0,
      "step": 14740
    },
    {
      "entropy": 6.114156007766724,
      "epoch": 1.47393412305693,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047894990997111403,
      "loss": 5.8696,
      "mean_token_accuracy": 0.15791167616844176,
      "num_tokens": 28845692.0,
      "step": 14745
    },
    {
      "entropy": 6.051042079925537,
      "epoch": 1.4744339481181585,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00047893483497233057,
      "loss": 5.7879,
      "mean_token_accuracy": 0.163599706441164,
      "num_tokens": 28854854.0,
      "step": 14750
    },
    {
      "entropy": 6.089698839187622,
      "epoch": 1.4749337731793872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004789197548425992,
      "loss": 5.8877,
      "mean_token_accuracy": 0.15952688157558442,
      "num_tokens": 28865711.0,
      "step": 14755
    },
    {
      "entropy": 6.0582763195037845,
      "epoch": 1.4754335982406157,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004789046695822992,
      "loss": 5.7507,
      "mean_token_accuracy": 0.17356730848550797,
      "num_tokens": 28874193.0,
      "step": 14760
    },
    {
      "entropy": 6.034773921966552,
      "epoch": 1.4759334233018444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047888957919181026,
      "loss": 5.8416,
      "mean_token_accuracy": 0.16772895008325578,
      "num_tokens": 28883652.0,
      "step": 14765
    },
    {
      "entropy": 6.120548915863037,
      "epoch": 1.4764332483630729,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004788744836715119,
      "loss": 5.8631,
      "mean_token_accuracy": 0.1631767362356186,
      "num_tokens": 28893040.0,
      "step": 14770
    },
    {
      "entropy": 6.154530000686646,
      "epoch": 1.4769330734243016,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004788593830217841,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16373732835054397,
      "num_tokens": 28903330.0,
      "step": 14775
    },
    {
      "entropy": 6.1097158908844,
      "epoch": 1.47743289848553,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047884427724300657,
      "loss": 5.8691,
      "mean_token_accuracy": 0.16185442358255386,
      "num_tokens": 28913032.0,
      "step": 14780
    },
    {
      "entropy": 6.077975559234619,
      "epoch": 1.4779327235467585,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004788291663355595,
      "loss": 5.8568,
      "mean_token_accuracy": 0.16055911481380464,
      "num_tokens": 28921298.0,
      "step": 14785
    },
    {
      "entropy": 6.066040134429931,
      "epoch": 1.4784325486079872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004788140502998231,
      "loss": 5.7869,
      "mean_token_accuracy": 0.16843023896217346,
      "num_tokens": 28930724.0,
      "step": 14790
    },
    {
      "entropy": 6.197900819778442,
      "epoch": 1.4789323736692157,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004787989291361776,
      "loss": 5.9734,
      "mean_token_accuracy": 0.14943530261516572,
      "num_tokens": 28941670.0,
      "step": 14795
    },
    {
      "entropy": 6.080639886856079,
      "epoch": 1.4794321987304444,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004787838028450035,
      "loss": 5.7802,
      "mean_token_accuracy": 0.1693911924958229,
      "num_tokens": 28950175.0,
      "step": 14800
    },
    {
      "entropy": 6.0849419116973875,
      "epoch": 1.4799320237916729,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047876867142668136,
      "loss": 5.7577,
      "mean_token_accuracy": 0.1665428712964058,
      "num_tokens": 28958410.0,
      "step": 14805
    },
    {
      "entropy": 6.072857189178467,
      "epoch": 1.4804318488529016,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004787535348815918,
      "loss": 5.702,
      "mean_token_accuracy": 0.1706013947725296,
      "num_tokens": 28967911.0,
      "step": 14810
    },
    {
      "entropy": 6.030284833908081,
      "epoch": 1.48093167391413,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047873839321011587,
      "loss": 5.8087,
      "mean_token_accuracy": 0.1703863710165024,
      "num_tokens": 28978112.0,
      "step": 14815
    },
    {
      "entropy": 6.074821853637696,
      "epoch": 1.4814314989753585,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004787232464126343,
      "loss": 5.8205,
      "mean_token_accuracy": 0.16091413646936417,
      "num_tokens": 28987960.0,
      "step": 14820
    },
    {
      "entropy": 6.067629337310791,
      "epoch": 1.4819313240365872,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004787080944895283,
      "loss": 5.8699,
      "mean_token_accuracy": 0.15592397302389144,
      "num_tokens": 28997000.0,
      "step": 14825
    },
    {
      "entropy": 6.155301761627197,
      "epoch": 1.4824311490978157,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047869293744117906,
      "loss": 5.8667,
      "mean_token_accuracy": 0.16108942925930023,
      "num_tokens": 29005725.0,
      "step": 14830
    },
    {
      "entropy": 6.047872829437256,
      "epoch": 1.4829309741590444,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047867777526796793,
      "loss": 5.8178,
      "mean_token_accuracy": 0.16866486817598342,
      "num_tokens": 29015168.0,
      "step": 14835
    },
    {
      "entropy": 6.126291656494141,
      "epoch": 1.4834307992202729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004786626079702764,
      "loss": 5.934,
      "mean_token_accuracy": 0.15931625217199324,
      "num_tokens": 29025170.0,
      "step": 14840
    },
    {
      "entropy": 6.1496683120727536,
      "epoch": 1.4839306242815016,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004786474355484861,
      "loss": 5.7995,
      "mean_token_accuracy": 0.16587658524513244,
      "num_tokens": 29034262.0,
      "step": 14845
    },
    {
      "entropy": 6.1078180313110355,
      "epoch": 1.48443044934273,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004786322580029786,
      "loss": 5.8885,
      "mean_token_accuracy": 0.15823860615491867,
      "num_tokens": 29043847.0,
      "step": 14850
    },
    {
      "entropy": 6.1236096858978275,
      "epoch": 1.4849302744039585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047861707533413593,
      "loss": 5.8695,
      "mean_token_accuracy": 0.1647449553012848,
      "num_tokens": 29053378.0,
      "step": 14855
    },
    {
      "entropy": 6.1678955078125,
      "epoch": 1.4854300994651872,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004786018875423401,
      "loss": 5.8567,
      "mean_token_accuracy": 0.16231548935174941,
      "num_tokens": 29063335.0,
      "step": 14860
    },
    {
      "entropy": 6.075611305236817,
      "epoch": 1.4859299245264157,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047858669462797317,
      "loss": 5.8508,
      "mean_token_accuracy": 0.16726845651865005,
      "num_tokens": 29073322.0,
      "step": 14865
    },
    {
      "entropy": 6.1054102897644045,
      "epoch": 1.4864297495876444,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047857149659141736,
      "loss": 5.8549,
      "mean_token_accuracy": 0.16108544170856476,
      "num_tokens": 29082618.0,
      "step": 14870
    },
    {
      "entropy": 6.100301456451416,
      "epoch": 1.4869295746488729,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004785562934330551,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1672778844833374,
      "num_tokens": 29091854.0,
      "step": 14875
    },
    {
      "entropy": 6.0595207691192625,
      "epoch": 1.4874293997101016,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047854108515326885,
      "loss": 5.7741,
      "mean_token_accuracy": 0.16831050366163253,
      "num_tokens": 29100816.0,
      "step": 14880
    },
    {
      "entropy": 6.1129515171051025,
      "epoch": 1.48792922477133,
      "grad_norm": 1.4453125,
      "learning_rate": 0.0004785258717524412,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16706235259771346,
      "num_tokens": 29110216.0,
      "step": 14885
    },
    {
      "entropy": 6.177827310562134,
      "epoch": 1.4884290498325585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047851065323095514,
      "loss": 5.8812,
      "mean_token_accuracy": 0.15973010212182998,
      "num_tokens": 29120910.0,
      "step": 14890
    },
    {
      "entropy": 6.128977346420288,
      "epoch": 1.4889288748937872,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004784954295891933,
      "loss": 5.9015,
      "mean_token_accuracy": 0.1597882926464081,
      "num_tokens": 29131163.0,
      "step": 14895
    },
    {
      "entropy": 6.067168426513672,
      "epoch": 1.4894286999550157,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047848020082753883,
      "loss": 5.7867,
      "mean_token_accuracy": 0.16889406889677047,
      "num_tokens": 29140549.0,
      "step": 14900
    },
    {
      "entropy": 6.057073497772217,
      "epoch": 1.4899285250162442,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004784649669463749,
      "loss": 5.7655,
      "mean_token_accuracy": 0.16541184931993486,
      "num_tokens": 29149884.0,
      "step": 14905
    },
    {
      "entropy": 6.090287542343139,
      "epoch": 1.4904283500774729,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047844972794608474,
      "loss": 5.921,
      "mean_token_accuracy": 0.16085264831781387,
      "num_tokens": 29158913.0,
      "step": 14910
    },
    {
      "entropy": 6.141239929199219,
      "epoch": 1.4909281751387016,
      "grad_norm": 1.375,
      "learning_rate": 0.00047843448382705185,
      "loss": 5.876,
      "mean_token_accuracy": 0.1552782490849495,
      "num_tokens": 29167548.0,
      "step": 14915
    },
    {
      "entropy": 6.1099042892456055,
      "epoch": 1.49142800019993,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047841923458965966,
      "loss": 5.8497,
      "mean_token_accuracy": 0.1688336357474327,
      "num_tokens": 29177155.0,
      "step": 14920
    },
    {
      "entropy": 6.115854930877686,
      "epoch": 1.4919278252611585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047840398023429176,
      "loss": 5.8877,
      "mean_token_accuracy": 0.16071407347917557,
      "num_tokens": 29185920.0,
      "step": 14925
    },
    {
      "entropy": 6.204611968994141,
      "epoch": 1.4924276503223872,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047838872076133213,
      "loss": 5.8504,
      "mean_token_accuracy": 0.16223606914281846,
      "num_tokens": 29195707.0,
      "step": 14930
    },
    {
      "entropy": 6.169075012207031,
      "epoch": 1.4929274753836157,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004783734561711647,
      "loss": 5.9022,
      "mean_token_accuracy": 0.15263044387102126,
      "num_tokens": 29206395.0,
      "step": 14935
    },
    {
      "entropy": 6.0841480731964115,
      "epoch": 1.4934273004448442,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047835818646417333,
      "loss": 5.8928,
      "mean_token_accuracy": 0.1655309572815895,
      "num_tokens": 29215903.0,
      "step": 14940
    },
    {
      "entropy": 6.181803750991821,
      "epoch": 1.4939271255060729,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004783429116407423,
      "loss": 5.9254,
      "mean_token_accuracy": 0.15845088362693788,
      "num_tokens": 29225768.0,
      "step": 14945
    },
    {
      "entropy": 6.159802579879761,
      "epoch": 1.4944269505673016,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047832763170125605,
      "loss": 5.8786,
      "mean_token_accuracy": 0.16108133345842363,
      "num_tokens": 29235612.0,
      "step": 14950
    },
    {
      "entropy": 6.128161811828614,
      "epoch": 1.49492677562853,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047831234664609887,
      "loss": 5.7647,
      "mean_token_accuracy": 0.1677384465932846,
      "num_tokens": 29244556.0,
      "step": 14955
    },
    {
      "entropy": 6.126919174194336,
      "epoch": 1.4954266006897585,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047829705647565535,
      "loss": 5.9576,
      "mean_token_accuracy": 0.15524016469717025,
      "num_tokens": 29254978.0,
      "step": 14960
    },
    {
      "entropy": 6.034291076660156,
      "epoch": 1.4959264257509872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004782817611903102,
      "loss": 5.7415,
      "mean_token_accuracy": 0.16946225315332414,
      "num_tokens": 29263339.0,
      "step": 14965
    },
    {
      "entropy": 6.062045955657959,
      "epoch": 1.4964262508122157,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047826646079044826,
      "loss": 5.7263,
      "mean_token_accuracy": 0.16392131000757218,
      "num_tokens": 29272707.0,
      "step": 14970
    },
    {
      "entropy": 6.097141981124878,
      "epoch": 1.4969260758734442,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047825115527645447,
      "loss": 5.8523,
      "mean_token_accuracy": 0.16153173446655272,
      "num_tokens": 29282292.0,
      "step": 14975
    },
    {
      "entropy": 6.084741592407227,
      "epoch": 1.4974259009346729,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004782358446487139,
      "loss": 5.9338,
      "mean_token_accuracy": 0.17100015804171562,
      "num_tokens": 29291289.0,
      "step": 14980
    },
    {
      "entropy": 6.066371965408325,
      "epoch": 1.4979257259959016,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047822052890761173,
      "loss": 5.7481,
      "mean_token_accuracy": 0.17149513810873032,
      "num_tokens": 29300166.0,
      "step": 14985
    },
    {
      "entropy": 6.044461393356324,
      "epoch": 1.49842555105713,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004782052080535334,
      "loss": 5.7504,
      "mean_token_accuracy": 0.1678796663880348,
      "num_tokens": 29308422.0,
      "step": 14990
    },
    {
      "entropy": 6.140214157104492,
      "epoch": 1.4989253761183585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004781898820868644,
      "loss": 5.9046,
      "mean_token_accuracy": 0.16995158940553665,
      "num_tokens": 29317699.0,
      "step": 14995
    },
    {
      "entropy": 6.178449964523315,
      "epoch": 1.4994252011795872,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004781745510079901,
      "loss": 5.9047,
      "mean_token_accuracy": 0.15471889525651933,
      "num_tokens": 29327711.0,
      "step": 15000
    },
    {
      "epoch": 1.4994252011795872,
      "eval_entropy": 5.969458627565655,
      "eval_loss": 5.963263988494873,
      "eval_mean_token_accuracy": 0.1646997789060181,
      "eval_num_tokens": 29327711.0,
      "eval_runtime": 23.8667,
      "eval_samples_per_second": 1300.765,
      "eval_steps_per_second": 162.611,
      "step": 15000
    },
    {
      "entropy": 6.077131271362305,
      "epoch": 1.4999250262408157,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047815921481729647,
      "loss": 5.7928,
      "mean_token_accuracy": 0.16737655997276307,
      "num_tokens": 29336540.0,
      "step": 15005
    },
    {
      "entropy": 6.190031099319458,
      "epoch": 1.5004248513020442,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004781438735151693,
      "loss": 5.9999,
      "mean_token_accuracy": 0.15030388236045839,
      "num_tokens": 29345512.0,
      "step": 15010
    },
    {
      "entropy": 6.182788753509522,
      "epoch": 1.5009246763632729,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047812852710199446,
      "loss": 5.9569,
      "mean_token_accuracy": 0.1508342981338501,
      "num_tokens": 29355726.0,
      "step": 15015
    },
    {
      "entropy": 6.116374826431274,
      "epoch": 1.5014245014245016,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004781131755781582,
      "loss": 5.7993,
      "mean_token_accuracy": 0.16601921916007994,
      "num_tokens": 29366570.0,
      "step": 15020
    },
    {
      "entropy": 6.0964476585388185,
      "epoch": 1.50192432648573,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004780978189440467,
      "loss": 5.9089,
      "mean_token_accuracy": 0.1565070003271103,
      "num_tokens": 29376356.0,
      "step": 15025
    },
    {
      "entropy": 6.175416374206543,
      "epoch": 1.5024241515469585,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047808245720004633,
      "loss": 5.9025,
      "mean_token_accuracy": 0.1557520419359207,
      "num_tokens": 29385246.0,
      "step": 15030
    },
    {
      "entropy": 6.104362535476684,
      "epoch": 1.5029239766081872,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047806709034654355,
      "loss": 5.7773,
      "mean_token_accuracy": 0.16742318123579025,
      "num_tokens": 29395352.0,
      "step": 15035
    },
    {
      "entropy": 6.1111784934997555,
      "epoch": 1.5034238016694157,
      "grad_norm": 1.5859375,
      "learning_rate": 0.0004780517183839251,
      "loss": 5.9145,
      "mean_token_accuracy": 0.15390866845846177,
      "num_tokens": 29404771.0,
      "step": 15040
    },
    {
      "entropy": 6.0870363235473635,
      "epoch": 1.5039236267306442,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004780363413125777,
      "loss": 5.819,
      "mean_token_accuracy": 0.1627956748008728,
      "num_tokens": 29413378.0,
      "step": 15045
    },
    {
      "entropy": 6.154175806045532,
      "epoch": 1.5044234517918729,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047802095913288814,
      "loss": 5.9103,
      "mean_token_accuracy": 0.1604010283946991,
      "num_tokens": 29424125.0,
      "step": 15050
    },
    {
      "entropy": 6.143818044662476,
      "epoch": 1.5049232768531016,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047800557184524345,
      "loss": 5.8603,
      "mean_token_accuracy": 0.15866938084363938,
      "num_tokens": 29435411.0,
      "step": 15055
    },
    {
      "entropy": 6.081011343002319,
      "epoch": 1.5054231019143298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004779901794500308,
      "loss": 5.9066,
      "mean_token_accuracy": 0.15717824548482895,
      "num_tokens": 29445515.0,
      "step": 15060
    },
    {
      "entropy": 6.181150102615357,
      "epoch": 1.5059229269755585,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047797478194763754,
      "loss": 5.8884,
      "mean_token_accuracy": 0.15744663774967194,
      "num_tokens": 29454864.0,
      "step": 15065
    },
    {
      "entropy": 6.15263876914978,
      "epoch": 1.5064227520367872,
      "grad_norm": 1.375,
      "learning_rate": 0.00047795937933845103,
      "loss": 5.9136,
      "mean_token_accuracy": 0.15776969417929648,
      "num_tokens": 29464443.0,
      "step": 15070
    },
    {
      "entropy": 6.153004360198975,
      "epoch": 1.5069225770980157,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004779439716228587,
      "loss": 5.909,
      "mean_token_accuracy": 0.15651096850633622,
      "num_tokens": 29474954.0,
      "step": 15075
    },
    {
      "entropy": 6.146423196792602,
      "epoch": 1.5074224021592442,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004779285588012482,
      "loss": 5.8769,
      "mean_token_accuracy": 0.16065066754817964,
      "num_tokens": 29484647.0,
      "step": 15080
    },
    {
      "entropy": 6.132674646377564,
      "epoch": 1.5079222272204729,
      "grad_norm": 1.25,
      "learning_rate": 0.00047791314087400745,
      "loss": 5.908,
      "mean_token_accuracy": 0.16260896474123002,
      "num_tokens": 29494782.0,
      "step": 15085
    },
    {
      "entropy": 6.086255264282227,
      "epoch": 1.5084220522817016,
      "grad_norm": 1.25,
      "learning_rate": 0.0004778977178415243,
      "loss": 5.8385,
      "mean_token_accuracy": 0.1640683740377426,
      "num_tokens": 29505311.0,
      "step": 15090
    },
    {
      "entropy": 6.116140699386596,
      "epoch": 1.5089218773429298,
      "grad_norm": 1.25,
      "learning_rate": 0.0004778822897041868,
      "loss": 5.8053,
      "mean_token_accuracy": 0.16211504340171815,
      "num_tokens": 29515192.0,
      "step": 15095
    },
    {
      "entropy": 6.181423807144165,
      "epoch": 1.5094217024041585,
      "grad_norm": 1.28125,
      "learning_rate": 0.000477866856462383,
      "loss": 5.868,
      "mean_token_accuracy": 0.16014138907194136,
      "num_tokens": 29524429.0,
      "step": 15100
    },
    {
      "entropy": 6.195126485824585,
      "epoch": 1.5099215274653872,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004778514181165014,
      "loss": 5.8018,
      "mean_token_accuracy": 0.1683039277791977,
      "num_tokens": 29532926.0,
      "step": 15105
    },
    {
      "entropy": 6.137876176834107,
      "epoch": 1.5104213525266157,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047783597466693025,
      "loss": 5.9055,
      "mean_token_accuracy": 0.1598990947008133,
      "num_tokens": 29542588.0,
      "step": 15110
    },
    {
      "entropy": 6.106893825531006,
      "epoch": 1.5109211775878442,
      "grad_norm": 1.953125,
      "learning_rate": 0.0004778205261140582,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16088825911283494,
      "num_tokens": 29551710.0,
      "step": 15115
    },
    {
      "entropy": 6.113062143325806,
      "epoch": 1.5114210026490729,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004778050724582739,
      "loss": 5.9339,
      "mean_token_accuracy": 0.1654717057943344,
      "num_tokens": 29561357.0,
      "step": 15120
    },
    {
      "entropy": 6.120569181442261,
      "epoch": 1.5119208277103013,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047778961369996623,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16532306969165803,
      "num_tokens": 29570060.0,
      "step": 15125
    },
    {
      "entropy": 6.170530080795288,
      "epoch": 1.5124206527715298,
      "grad_norm": 1.34375,
      "learning_rate": 0.000477774149839524,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16479945927858353,
      "num_tokens": 29580097.0,
      "step": 15130
    },
    {
      "entropy": 6.233632612228393,
      "epoch": 1.5129204778327585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047775868087733634,
      "loss": 6.0435,
      "mean_token_accuracy": 0.14772638231515883,
      "num_tokens": 29590290.0,
      "step": 15135
    },
    {
      "entropy": 6.144825124740601,
      "epoch": 1.5134203028939872,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047774320681379237,
      "loss": 5.8466,
      "mean_token_accuracy": 0.16631245240569115,
      "num_tokens": 29598839.0,
      "step": 15140
    },
    {
      "entropy": 6.1948157787323,
      "epoch": 1.5139201279552157,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004777277276492816,
      "loss": 5.9143,
      "mean_token_accuracy": 0.167092901468277,
      "num_tokens": 29609550.0,
      "step": 15145
    },
    {
      "entropy": 6.249425649642944,
      "epoch": 1.5144199530164442,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047771224338419326,
      "loss": 5.9996,
      "mean_token_accuracy": 0.1465565636754036,
      "num_tokens": 29619786.0,
      "step": 15150
    },
    {
      "entropy": 6.122082948684692,
      "epoch": 1.5149197780776729,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004776967540189171,
      "loss": 5.8043,
      "mean_token_accuracy": 0.16506993174552917,
      "num_tokens": 29630492.0,
      "step": 15155
    },
    {
      "entropy": 6.176985406875611,
      "epoch": 1.5154196031389013,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004776812595538428,
      "loss": 5.9099,
      "mean_token_accuracy": 0.15269644558429718,
      "num_tokens": 29641965.0,
      "step": 15160
    },
    {
      "entropy": 6.16685380935669,
      "epoch": 1.5159194282001298,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047766575998936013,
      "loss": 5.9204,
      "mean_token_accuracy": 0.15826703906059264,
      "num_tokens": 29651692.0,
      "step": 15165
    },
    {
      "entropy": 6.175631999969482,
      "epoch": 1.5164192532613585,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047765025532585907,
      "loss": 5.8679,
      "mean_token_accuracy": 0.1623964712023735,
      "num_tokens": 29660541.0,
      "step": 15170
    },
    {
      "entropy": 6.08281397819519,
      "epoch": 1.5169190783225872,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004776347455637298,
      "loss": 5.8633,
      "mean_token_accuracy": 0.16244495958089827,
      "num_tokens": 29670498.0,
      "step": 15175
    },
    {
      "entropy": 6.046838092803955,
      "epoch": 1.5174189033838157,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004776192307033624,
      "loss": 5.7586,
      "mean_token_accuracy": 0.17302726060152054,
      "num_tokens": 29678898.0,
      "step": 15180
    },
    {
      "entropy": 6.175777292251587,
      "epoch": 1.5179187284450442,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004776037107451473,
      "loss": 5.8981,
      "mean_token_accuracy": 0.15911915153265,
      "num_tokens": 29689723.0,
      "step": 15185
    },
    {
      "entropy": 6.110267448425293,
      "epoch": 1.5184185535062729,
      "grad_norm": 1.21875,
      "learning_rate": 0.000477588185689475,
      "loss": 5.7894,
      "mean_token_accuracy": 0.17264097929000854,
      "num_tokens": 29699246.0,
      "step": 15190
    },
    {
      "entropy": 6.060447120666504,
      "epoch": 1.5189183785675013,
      "grad_norm": 1.25,
      "learning_rate": 0.00047757265553673606,
      "loss": 5.7594,
      "mean_token_accuracy": 0.1698579266667366,
      "num_tokens": 29709125.0,
      "step": 15195
    },
    {
      "entropy": 6.082809066772461,
      "epoch": 1.5194182036287298,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047755712028732126,
      "loss": 5.7972,
      "mean_token_accuracy": 0.1643887847661972,
      "num_tokens": 29718953.0,
      "step": 15200
    },
    {
      "entropy": 6.1613805294036865,
      "epoch": 1.5199180286899585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004775415799416214,
      "loss": 5.8539,
      "mean_token_accuracy": 0.15922629535198213,
      "num_tokens": 29728764.0,
      "step": 15205
    },
    {
      "entropy": 6.029280471801758,
      "epoch": 1.5204178537511872,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004775260345000275,
      "loss": 5.8698,
      "mean_token_accuracy": 0.1581306904554367,
      "num_tokens": 29738033.0,
      "step": 15210
    },
    {
      "entropy": 6.008817434310913,
      "epoch": 1.5209176788124157,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047751048396293077,
      "loss": 5.7947,
      "mean_token_accuracy": 0.17406957894563674,
      "num_tokens": 29748450.0,
      "step": 15215
    },
    {
      "entropy": 6.227540063858032,
      "epoch": 1.5214175038736442,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047749492833072226,
      "loss": 5.927,
      "mean_token_accuracy": 0.1563795879483223,
      "num_tokens": 29757984.0,
      "step": 15220
    },
    {
      "entropy": 6.137864112854004,
      "epoch": 1.5219173289348729,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047747936760379354,
      "loss": 5.8629,
      "mean_token_accuracy": 0.15938941091299058,
      "num_tokens": 29768200.0,
      "step": 15225
    },
    {
      "entropy": 6.026132488250733,
      "epoch": 1.5224171539961013,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000477463801782536,
      "loss": 5.8362,
      "mean_token_accuracy": 0.16058966964483262,
      "num_tokens": 29778242.0,
      "step": 15230
    },
    {
      "entropy": 6.174581670761109,
      "epoch": 1.5229169790573298,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047744823086734136,
      "loss": 5.9735,
      "mean_token_accuracy": 0.152067668735981,
      "num_tokens": 29787636.0,
      "step": 15235
    },
    {
      "entropy": 6.198061227798462,
      "epoch": 1.5234168041185585,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047743265485860123,
      "loss": 5.803,
      "mean_token_accuracy": 0.16485027223825455,
      "num_tokens": 29796537.0,
      "step": 15240
    },
    {
      "entropy": 6.115128755569458,
      "epoch": 1.5239166291797872,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047741707375670766,
      "loss": 5.8569,
      "mean_token_accuracy": 0.16327260583639144,
      "num_tokens": 29805495.0,
      "step": 15245
    },
    {
      "entropy": 6.095377445220947,
      "epoch": 1.5244164542410157,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004774014875620526,
      "loss": 5.7601,
      "mean_token_accuracy": 0.16748379319906234,
      "num_tokens": 29815193.0,
      "step": 15250
    },
    {
      "entropy": 6.120562982559204,
      "epoch": 1.5249162793022442,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00047738589627502814,
      "loss": 5.88,
      "mean_token_accuracy": 0.16375370174646378,
      "num_tokens": 29825004.0,
      "step": 15255
    },
    {
      "entropy": 6.067328929901123,
      "epoch": 1.5254161043634729,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047737029989602665,
      "loss": 5.8905,
      "mean_token_accuracy": 0.1636359930038452,
      "num_tokens": 29834507.0,
      "step": 15260
    },
    {
      "entropy": 6.038512897491455,
      "epoch": 1.5259159294247013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004773546984254405,
      "loss": 5.7549,
      "mean_token_accuracy": 0.16233235597610474,
      "num_tokens": 29844397.0,
      "step": 15265
    },
    {
      "entropy": 6.152181100845337,
      "epoch": 1.5264157544859298,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004773390918636622,
      "loss": 5.8732,
      "mean_token_accuracy": 0.16565565168857574,
      "num_tokens": 29855702.0,
      "step": 15270
    },
    {
      "entropy": 6.093813753128051,
      "epoch": 1.5269155795471585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004773234802110844,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16872600615024566,
      "num_tokens": 29865806.0,
      "step": 15275
    },
    {
      "entropy": 6.040602731704712,
      "epoch": 1.5274154046083872,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004773078634681,
      "loss": 5.7703,
      "mean_token_accuracy": 0.17455331087112427,
      "num_tokens": 29874827.0,
      "step": 15280
    },
    {
      "entropy": 6.043024969100952,
      "epoch": 1.5279152296696157,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047729224163510167,
      "loss": 5.8484,
      "mean_token_accuracy": 0.16239284127950668,
      "num_tokens": 29883517.0,
      "step": 15285
    },
    {
      "entropy": 6.061124610900879,
      "epoch": 1.5284150547308442,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047727661471248263,
      "loss": 5.7637,
      "mean_token_accuracy": 0.16573414728045463,
      "num_tokens": 29893303.0,
      "step": 15290
    },
    {
      "entropy": 6.175179767608642,
      "epoch": 1.5289148797920729,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000477260982700636,
      "loss": 5.9868,
      "mean_token_accuracy": 0.1538536839187145,
      "num_tokens": 29904719.0,
      "step": 15295
    },
    {
      "entropy": 6.186687374114991,
      "epoch": 1.5294147048533013,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004772453455999552,
      "loss": 5.8728,
      "mean_token_accuracy": 0.1573803335428238,
      "num_tokens": 29914852.0,
      "step": 15300
    },
    {
      "entropy": 6.064591646194458,
      "epoch": 1.5299145299145298,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004772297034108334,
      "loss": 5.8224,
      "mean_token_accuracy": 0.16104813665151596,
      "num_tokens": 29924479.0,
      "step": 15305
    },
    {
      "entropy": 6.127867603302002,
      "epoch": 1.5304143549757585,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047721405613366434,
      "loss": 5.8689,
      "mean_token_accuracy": 0.15917230993509293,
      "num_tokens": 29933162.0,
      "step": 15310
    },
    {
      "entropy": 6.082047605514527,
      "epoch": 1.5309141800369872,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047719840376884173,
      "loss": 5.7877,
      "mean_token_accuracy": 0.1677139550447464,
      "num_tokens": 29942468.0,
      "step": 15315
    },
    {
      "entropy": 6.075866079330444,
      "epoch": 1.5314140050982155,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004771827463167593,
      "loss": 5.8769,
      "mean_token_accuracy": 0.1628609538078308,
      "num_tokens": 29951716.0,
      "step": 15320
    },
    {
      "entropy": 6.1270959854125975,
      "epoch": 1.5319138301594442,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004771670837778109,
      "loss": 5.9359,
      "mean_token_accuracy": 0.1540454462170601,
      "num_tokens": 29960233.0,
      "step": 15325
    },
    {
      "entropy": 6.099421834945678,
      "epoch": 1.5324136552206729,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047715141615239075,
      "loss": 5.867,
      "mean_token_accuracy": 0.16315412521362305,
      "num_tokens": 29968979.0,
      "step": 15330
    },
    {
      "entropy": 6.07310733795166,
      "epoch": 1.5329134802819013,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047713574344089305,
      "loss": 5.8146,
      "mean_token_accuracy": 0.16052270978689193,
      "num_tokens": 29978830.0,
      "step": 15335
    },
    {
      "entropy": 6.146570539474487,
      "epoch": 1.5334133053431298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.000477120065643712,
      "loss": 5.8325,
      "mean_token_accuracy": 0.16192905455827714,
      "num_tokens": 29987708.0,
      "step": 15340
    },
    {
      "entropy": 6.189952230453491,
      "epoch": 1.5339131304043585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004771043827612421,
      "loss": 5.8506,
      "mean_token_accuracy": 0.16452495157718658,
      "num_tokens": 29997351.0,
      "step": 15345
    },
    {
      "entropy": 6.081881093978882,
      "epoch": 1.5344129554655872,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047708869479387795,
      "loss": 5.8555,
      "mean_token_accuracy": 0.1639459475874901,
      "num_tokens": 30006412.0,
      "step": 15350
    },
    {
      "entropy": 6.13397250175476,
      "epoch": 1.5349127805268155,
      "grad_norm": 1.7734375,
      "learning_rate": 0.0004770730017420143,
      "loss": 5.8544,
      "mean_token_accuracy": 0.15548537522554398,
      "num_tokens": 30016629.0,
      "step": 15355
    },
    {
      "entropy": 6.0898314952850345,
      "epoch": 1.5354126055880442,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004770573036060458,
      "loss": 5.7203,
      "mean_token_accuracy": 0.17135403901338578,
      "num_tokens": 30026395.0,
      "step": 15360
    },
    {
      "entropy": 6.131185531616211,
      "epoch": 1.5359124306492729,
      "grad_norm": 1.25,
      "learning_rate": 0.0004770416003863676,
      "loss": 5.8888,
      "mean_token_accuracy": 0.16102334409952163,
      "num_tokens": 30037178.0,
      "step": 15365
    },
    {
      "entropy": 6.065256881713867,
      "epoch": 1.5364122557105013,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004770258920833748,
      "loss": 5.7779,
      "mean_token_accuracy": 0.1677483156323433,
      "num_tokens": 30046335.0,
      "step": 15370
    },
    {
      "entropy": 6.1482987880706785,
      "epoch": 1.5369120807717298,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047701017869746236,
      "loss": 5.8505,
      "mean_token_accuracy": 0.1594402588903904,
      "num_tokens": 30056574.0,
      "step": 15375
    },
    {
      "entropy": 6.088780450820923,
      "epoch": 1.5374119058329585,
      "grad_norm": 1.5625,
      "learning_rate": 0.00047699446022902596,
      "loss": 5.774,
      "mean_token_accuracy": 0.165068681538105,
      "num_tokens": 30067048.0,
      "step": 15380
    },
    {
      "entropy": 5.999118947982788,
      "epoch": 1.537911730894187,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047697873667846094,
      "loss": 5.7974,
      "mean_token_accuracy": 0.16534364521503447,
      "num_tokens": 30077177.0,
      "step": 15385
    },
    {
      "entropy": 6.134491586685181,
      "epoch": 1.5384115559554155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047696300804616276,
      "loss": 5.9004,
      "mean_token_accuracy": 0.16050038784742354,
      "num_tokens": 30086348.0,
      "step": 15390
    },
    {
      "entropy": 6.0851703643798825,
      "epoch": 1.5389113810166442,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004769472743325273,
      "loss": 5.7752,
      "mean_token_accuracy": 0.16414364129304887,
      "num_tokens": 30096721.0,
      "step": 15395
    },
    {
      "entropy": 6.0597583770751955,
      "epoch": 1.5394112060778729,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047693153553795046,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16537482291460037,
      "num_tokens": 30106797.0,
      "step": 15400
    },
    {
      "entropy": 6.086478900909424,
      "epoch": 1.5399110311391013,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004769157916628281,
      "loss": 5.8348,
      "mean_token_accuracy": 0.1604676753282547,
      "num_tokens": 30116643.0,
      "step": 15405
    },
    {
      "entropy": 6.09989366531372,
      "epoch": 1.5404108562003298,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047690004270755634,
      "loss": 5.7773,
      "mean_token_accuracy": 0.17121385931968688,
      "num_tokens": 30125336.0,
      "step": 15410
    },
    {
      "entropy": 5.986818647384643,
      "epoch": 1.5409106812615585,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004768842886725314,
      "loss": 5.8022,
      "mean_token_accuracy": 0.16936761885881424,
      "num_tokens": 30135238.0,
      "step": 15415
    },
    {
      "entropy": 6.145879030227661,
      "epoch": 1.541410506322787,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047686852955814986,
      "loss": 5.8219,
      "mean_token_accuracy": 0.16371650546789168,
      "num_tokens": 30144469.0,
      "step": 15420
    },
    {
      "entropy": 6.086048984527588,
      "epoch": 1.5419103313840155,
      "grad_norm": 1.296875,
      "learning_rate": 0.000476852765364808,
      "loss": 5.8479,
      "mean_token_accuracy": 0.15872230231761933,
      "num_tokens": 30154484.0,
      "step": 15425
    },
    {
      "entropy": 6.081093215942383,
      "epoch": 1.5424101564452442,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004768369960929025,
      "loss": 5.8312,
      "mean_token_accuracy": 0.1636373907327652,
      "num_tokens": 30163832.0,
      "step": 15430
    },
    {
      "entropy": 6.057599067687988,
      "epoch": 1.5429099815064728,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004768212217428301,
      "loss": 5.7698,
      "mean_token_accuracy": 0.1623804524540901,
      "num_tokens": 30173503.0,
      "step": 15435
    },
    {
      "entropy": 6.128013038635254,
      "epoch": 1.5434098065677013,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047680544231498766,
      "loss": 5.8694,
      "mean_token_accuracy": 0.16631856858730315,
      "num_tokens": 30183476.0,
      "step": 15440
    },
    {
      "entropy": 6.144038534164428,
      "epoch": 1.5439096316289298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004767896578097723,
      "loss": 5.8589,
      "mean_token_accuracy": 0.16868437975645065,
      "num_tokens": 30192552.0,
      "step": 15445
    },
    {
      "entropy": 6.209604787826538,
      "epoch": 1.5444094566901585,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004767738682275811,
      "loss": 5.9225,
      "mean_token_accuracy": 0.15715301781892776,
      "num_tokens": 30202796.0,
      "step": 15450
    },
    {
      "entropy": 6.133153104782105,
      "epoch": 1.544909281751387,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004767580735688112,
      "loss": 5.8254,
      "mean_token_accuracy": 0.16285106614232064,
      "num_tokens": 30211878.0,
      "step": 15455
    },
    {
      "entropy": 6.077563190460205,
      "epoch": 1.5454091068126155,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004767422738338601,
      "loss": 5.8755,
      "mean_token_accuracy": 0.15973365157842637,
      "num_tokens": 30221572.0,
      "step": 15460
    },
    {
      "entropy": 6.072950315475464,
      "epoch": 1.5459089318738441,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004767264690231253,
      "loss": 5.8215,
      "mean_token_accuracy": 0.16243474185466766,
      "num_tokens": 30230770.0,
      "step": 15465
    },
    {
      "entropy": 6.104480743408203,
      "epoch": 1.5464087569350728,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004767106591370045,
      "loss": 5.7707,
      "mean_token_accuracy": 0.17189311981201172,
      "num_tokens": 30241404.0,
      "step": 15470
    },
    {
      "entropy": 6.175734519958496,
      "epoch": 1.5469085819963013,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047669484417589537,
      "loss": 5.9095,
      "mean_token_accuracy": 0.15281098037958146,
      "num_tokens": 30251471.0,
      "step": 15475
    },
    {
      "entropy": 6.203527736663818,
      "epoch": 1.5474084070575298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047667902414019585,
      "loss": 5.8697,
      "mean_token_accuracy": 0.15795407965779304,
      "num_tokens": 30260962.0,
      "step": 15480
    },
    {
      "entropy": 6.153222322463989,
      "epoch": 1.5479082321187585,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000476663199030304,
      "loss": 5.9168,
      "mean_token_accuracy": 0.15842970907688142,
      "num_tokens": 30271959.0,
      "step": 15485
    },
    {
      "entropy": 6.013221788406372,
      "epoch": 1.548408057179987,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047664736884661804,
      "loss": 5.7757,
      "mean_token_accuracy": 0.16625786870718,
      "num_tokens": 30281134.0,
      "step": 15490
    },
    {
      "entropy": 6.005005979537964,
      "epoch": 1.5489078822412155,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047663153358953604,
      "loss": 5.7511,
      "mean_token_accuracy": 0.17048237919807435,
      "num_tokens": 30289989.0,
      "step": 15495
    },
    {
      "entropy": 6.120188283920288,
      "epoch": 1.5494077073024441,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004766156932594566,
      "loss": 5.8806,
      "mean_token_accuracy": 0.1587989792227745,
      "num_tokens": 30300416.0,
      "step": 15500
    },
    {
      "entropy": 6.169473457336426,
      "epoch": 1.5499075323636728,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047659984785677824,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16052213683724403,
      "num_tokens": 30310056.0,
      "step": 15505
    },
    {
      "entropy": 6.112693500518799,
      "epoch": 1.5504073574249013,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047658399738189955,
      "loss": 5.8686,
      "mean_token_accuracy": 0.16063059717416764,
      "num_tokens": 30319489.0,
      "step": 15510
    },
    {
      "entropy": 6.164441108703613,
      "epoch": 1.5509071824861298,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047656814183521937,
      "loss": 5.9358,
      "mean_token_accuracy": 0.15607595890760423,
      "num_tokens": 30328814.0,
      "step": 15515
    },
    {
      "entropy": 6.138902950286865,
      "epoch": 1.5514070075473585,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004765522812171366,
      "loss": 5.8031,
      "mean_token_accuracy": 0.1670185148715973,
      "num_tokens": 30339219.0,
      "step": 15520
    },
    {
      "entropy": 6.0118121147155765,
      "epoch": 1.551906832608587,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047653641552805023,
      "loss": 5.8179,
      "mean_token_accuracy": 0.1645095318555832,
      "num_tokens": 30349553.0,
      "step": 15525
    },
    {
      "entropy": 6.097389173507691,
      "epoch": 1.5524066576698154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004765205447683595,
      "loss": 5.876,
      "mean_token_accuracy": 0.155429208278656,
      "num_tokens": 30359782.0,
      "step": 15530
    },
    {
      "entropy": 6.129660320281983,
      "epoch": 1.5529064827310441,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004765046689384638,
      "loss": 5.8277,
      "mean_token_accuracy": 0.16279025226831437,
      "num_tokens": 30369635.0,
      "step": 15535
    },
    {
      "entropy": 6.070033359527588,
      "epoch": 1.5534063077922728,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004764887880387625,
      "loss": 5.7679,
      "mean_token_accuracy": 0.16790590286254883,
      "num_tokens": 30379453.0,
      "step": 15540
    },
    {
      "entropy": 6.1385540008544925,
      "epoch": 1.5539061328535013,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00047647290206965506,
      "loss": 5.8466,
      "mean_token_accuracy": 0.15930682942271232,
      "num_tokens": 30388989.0,
      "step": 15545
    },
    {
      "entropy": 6.1380664825439455,
      "epoch": 1.5544059579147298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047645701103154125,
      "loss": 5.8624,
      "mean_token_accuracy": 0.16486021429300307,
      "num_tokens": 30399401.0,
      "step": 15550
    },
    {
      "entropy": 6.185231351852417,
      "epoch": 1.5549057829759585,
      "grad_norm": 1.25,
      "learning_rate": 0.00047644111492482094,
      "loss": 5.8998,
      "mean_token_accuracy": 0.1483965940773487,
      "num_tokens": 30409795.0,
      "step": 15555
    },
    {
      "entropy": 6.179520320892334,
      "epoch": 1.555405608037187,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000476425213749894,
      "loss": 5.9365,
      "mean_token_accuracy": 0.15703604742884636,
      "num_tokens": 30419243.0,
      "step": 15560
    },
    {
      "entropy": 6.111941385269165,
      "epoch": 1.5559054330984154,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004764093075071605,
      "loss": 5.8799,
      "mean_token_accuracy": 0.1628809839487076,
      "num_tokens": 30429278.0,
      "step": 15565
    },
    {
      "entropy": 6.050267219543457,
      "epoch": 1.5564052581596441,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004763933961970206,
      "loss": 5.8253,
      "mean_token_accuracy": 0.16774676442146302,
      "num_tokens": 30438442.0,
      "step": 15570
    },
    {
      "entropy": 6.087157726287842,
      "epoch": 1.5569050832208728,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004763774798198748,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16341341584920882,
      "num_tokens": 30448779.0,
      "step": 15575
    },
    {
      "entropy": 6.136054420471192,
      "epoch": 1.5574049082821013,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004763615583761234,
      "loss": 5.8009,
      "mean_token_accuracy": 0.163629412651062,
      "num_tokens": 30458219.0,
      "step": 15580
    },
    {
      "entropy": 6.173776245117187,
      "epoch": 1.5579047333433298,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004763456318661669,
      "loss": 5.8506,
      "mean_token_accuracy": 0.1635434776544571,
      "num_tokens": 30467772.0,
      "step": 15585
    },
    {
      "entropy": 6.109755373001098,
      "epoch": 1.5584045584045585,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004763297002904062,
      "loss": 5.8096,
      "mean_token_accuracy": 0.16153863072395325,
      "num_tokens": 30477025.0,
      "step": 15590
    },
    {
      "entropy": 6.1238078594207765,
      "epoch": 1.558904383465787,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047631376364924195,
      "loss": 5.8759,
      "mean_token_accuracy": 0.1667680948972702,
      "num_tokens": 30486691.0,
      "step": 15595
    },
    {
      "entropy": 6.0402512550354,
      "epoch": 1.5594042085270154,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004762978219430753,
      "loss": 5.7792,
      "mean_token_accuracy": 0.16216624081134795,
      "num_tokens": 30497138.0,
      "step": 15600
    },
    {
      "entropy": 6.145529699325562,
      "epoch": 1.5599040335882441,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004762818751723072,
      "loss": 5.9476,
      "mean_token_accuracy": 0.1560802549123764,
      "num_tokens": 30507526.0,
      "step": 15605
    },
    {
      "entropy": 6.1236495018005375,
      "epoch": 1.5604038586494728,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047626592333733895,
      "loss": 5.7904,
      "mean_token_accuracy": 0.16258176118135453,
      "num_tokens": 30517823.0,
      "step": 15610
    },
    {
      "entropy": 6.093144178390503,
      "epoch": 1.560903683710701,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004762499664385719,
      "loss": 5.8179,
      "mean_token_accuracy": 0.16506240367889405,
      "num_tokens": 30527892.0,
      "step": 15615
    },
    {
      "entropy": 6.148361921310425,
      "epoch": 1.5614035087719298,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047623400447640733,
      "loss": 5.829,
      "mean_token_accuracy": 0.17027583867311477,
      "num_tokens": 30537329.0,
      "step": 15620
    },
    {
      "entropy": 6.107088327407837,
      "epoch": 1.5619033338331585,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047621803745124717,
      "loss": 5.8467,
      "mean_token_accuracy": 0.16183447688817978,
      "num_tokens": 30546406.0,
      "step": 15625
    },
    {
      "entropy": 6.158752822875977,
      "epoch": 1.562403158894387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047620206536349294,
      "loss": 5.9205,
      "mean_token_accuracy": 0.16163745671510696,
      "num_tokens": 30556646.0,
      "step": 15630
    },
    {
      "entropy": 6.104726314544678,
      "epoch": 1.5629029839556154,
      "grad_norm": 1.25,
      "learning_rate": 0.00047618608821354636,
      "loss": 5.7775,
      "mean_token_accuracy": 0.17397433370351792,
      "num_tokens": 30565984.0,
      "step": 15635
    },
    {
      "entropy": 6.158686113357544,
      "epoch": 1.5634028090168441,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004761701060018097,
      "loss": 5.8675,
      "mean_token_accuracy": 0.1569593667984009,
      "num_tokens": 30575257.0,
      "step": 15640
    },
    {
      "entropy": 6.080680274963379,
      "epoch": 1.5639026340780728,
      "grad_norm": 1.234375,
      "learning_rate": 0.000476154118728685,
      "loss": 5.8016,
      "mean_token_accuracy": 0.15917307734489441,
      "num_tokens": 30584248.0,
      "step": 15645
    },
    {
      "entropy": 6.128966856002807,
      "epoch": 1.564402459139301,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047613812639457425,
      "loss": 5.862,
      "mean_token_accuracy": 0.16283520311117172,
      "num_tokens": 30593875.0,
      "step": 15650
    },
    {
      "entropy": 6.112940788269043,
      "epoch": 1.5649022842005298,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004761221289998801,
      "loss": 5.7588,
      "mean_token_accuracy": 0.1657870039343834,
      "num_tokens": 30603696.0,
      "step": 15655
    },
    {
      "entropy": 6.117493677139282,
      "epoch": 1.5654021092617585,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004761061265450049,
      "loss": 5.7938,
      "mean_token_accuracy": 0.16458658874034882,
      "num_tokens": 30613451.0,
      "step": 15660
    },
    {
      "entropy": 6.093286371231079,
      "epoch": 1.565901934322987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047609011903035136,
      "loss": 5.8793,
      "mean_token_accuracy": 0.1588987961411476,
      "num_tokens": 30623567.0,
      "step": 15665
    },
    {
      "entropy": 6.12623143196106,
      "epoch": 1.5664017593842154,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047607410645632216,
      "loss": 5.8326,
      "mean_token_accuracy": 0.1615612030029297,
      "num_tokens": 30634828.0,
      "step": 15670
    },
    {
      "entropy": 6.197947978973389,
      "epoch": 1.5669015844454441,
      "grad_norm": 1.5546875,
      "learning_rate": 0.00047605808882332013,
      "loss": 5.9849,
      "mean_token_accuracy": 0.15303948819637297,
      "num_tokens": 30644552.0,
      "step": 15675
    },
    {
      "entropy": 6.170982313156128,
      "epoch": 1.5674014095066726,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004760420661317484,
      "loss": 5.9662,
      "mean_token_accuracy": 0.1577688455581665,
      "num_tokens": 30654874.0,
      "step": 15680
    },
    {
      "entropy": 6.173335075378418,
      "epoch": 1.567901234567901,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004760260383820099,
      "loss": 5.8174,
      "mean_token_accuracy": 0.1662250429391861,
      "num_tokens": 30665619.0,
      "step": 15685
    },
    {
      "entropy": 6.045621633529663,
      "epoch": 1.5684010596291298,
      "grad_norm": 1.3125,
      "learning_rate": 0.00047601000557450816,
      "loss": 5.802,
      "mean_token_accuracy": 0.16115155220031738,
      "num_tokens": 30675072.0,
      "step": 15690
    },
    {
      "entropy": 6.170194101333618,
      "epoch": 1.5689008846903585,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004759939677096463,
      "loss": 5.9146,
      "mean_token_accuracy": 0.15461870431900024,
      "num_tokens": 30684567.0,
      "step": 15695
    },
    {
      "entropy": 6.107315492630005,
      "epoch": 1.569400709751587,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004759779247878279,
      "loss": 5.8883,
      "mean_token_accuracy": 0.16287163645029068,
      "num_tokens": 30693808.0,
      "step": 15700
    },
    {
      "entropy": 5.993677473068237,
      "epoch": 1.5699005348128154,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047596187680945675,
      "loss": 5.7905,
      "mean_token_accuracy": 0.1765650823712349,
      "num_tokens": 30702895.0,
      "step": 15705
    },
    {
      "entropy": 6.179835367202759,
      "epoch": 1.5704003598740441,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004759458237749364,
      "loss": 5.9413,
      "mean_token_accuracy": 0.15825000479817392,
      "num_tokens": 30714582.0,
      "step": 15710
    },
    {
      "entropy": 6.155795621871948,
      "epoch": 1.5709001849352726,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004759297656846708,
      "loss": 5.8901,
      "mean_token_accuracy": 0.1566804453730583,
      "num_tokens": 30724415.0,
      "step": 15715
    },
    {
      "entropy": 6.09362096786499,
      "epoch": 1.571400009996501,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047591370253906415,
      "loss": 5.7779,
      "mean_token_accuracy": 0.16407259702682495,
      "num_tokens": 30733040.0,
      "step": 15720
    },
    {
      "entropy": 6.066727066040039,
      "epoch": 1.5718998350577298,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047589763433852027,
      "loss": 5.7973,
      "mean_token_accuracy": 0.16331228315830232,
      "num_tokens": 30743022.0,
      "step": 15725
    },
    {
      "entropy": 6.06973032951355,
      "epoch": 1.5723996601189585,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004758815610834437,
      "loss": 5.8029,
      "mean_token_accuracy": 0.1651538133621216,
      "num_tokens": 30753335.0,
      "step": 15730
    },
    {
      "entropy": 6.082103586196899,
      "epoch": 1.572899485180187,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047586548277423873,
      "loss": 5.8486,
      "mean_token_accuracy": 0.1571740135550499,
      "num_tokens": 30762863.0,
      "step": 15735
    },
    {
      "entropy": 6.121285057067871,
      "epoch": 1.5733993102414154,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047584939941130994,
      "loss": 5.8614,
      "mean_token_accuracy": 0.1668570101261139,
      "num_tokens": 30773090.0,
      "step": 15740
    },
    {
      "entropy": 6.121962404251098,
      "epoch": 1.5738991353026441,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004758333109950619,
      "loss": 5.8841,
      "mean_token_accuracy": 0.16276065558195113,
      "num_tokens": 30782589.0,
      "step": 15745
    },
    {
      "entropy": 6.153895664215088,
      "epoch": 1.5743989603638726,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047581721752589943,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16383911520242692,
      "num_tokens": 30791508.0,
      "step": 15750
    },
    {
      "entropy": 6.067975234985352,
      "epoch": 1.574898785425101,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004758011190042274,
      "loss": 5.7904,
      "mean_token_accuracy": 0.16405827254056932,
      "num_tokens": 30801133.0,
      "step": 15755
    },
    {
      "entropy": 6.057309293746949,
      "epoch": 1.5753986104863298,
      "grad_norm": 1.5234375,
      "learning_rate": 0.0004757850154304509,
      "loss": 5.7153,
      "mean_token_accuracy": 0.17136121839284896,
      "num_tokens": 30810100.0,
      "step": 15760
    },
    {
      "entropy": 6.072236824035644,
      "epoch": 1.5758984355475585,
      "grad_norm": 1.25,
      "learning_rate": 0.0004757689068049752,
      "loss": 5.7921,
      "mean_token_accuracy": 0.17318984419107436,
      "num_tokens": 30820118.0,
      "step": 15765
    },
    {
      "entropy": 6.068833875656128,
      "epoch": 1.576398260608787,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047575279312820533,
      "loss": 5.8262,
      "mean_token_accuracy": 0.15900918394327163,
      "num_tokens": 30830480.0,
      "step": 15770
    },
    {
      "entropy": 6.134298419952392,
      "epoch": 1.5768980856700154,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004757366744005469,
      "loss": 5.9333,
      "mean_token_accuracy": 0.15336010754108428,
      "num_tokens": 30840421.0,
      "step": 15775
    },
    {
      "entropy": 6.163825654983521,
      "epoch": 1.5773979107312441,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047572055062240537,
      "loss": 5.8268,
      "mean_token_accuracy": 0.15723081678152084,
      "num_tokens": 30849247.0,
      "step": 15780
    },
    {
      "entropy": 6.118316555023194,
      "epoch": 1.5778977357924726,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004757044217941865,
      "loss": 5.8135,
      "mean_token_accuracy": 0.16745300889015197,
      "num_tokens": 30858804.0,
      "step": 15785
    },
    {
      "entropy": 6.123394727706909,
      "epoch": 1.578397560853701,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047568828791629605,
      "loss": 5.8545,
      "mean_token_accuracy": 0.16265913397073745,
      "num_tokens": 30869150.0,
      "step": 15790
    },
    {
      "entropy": 6.0276207447052,
      "epoch": 1.5788973859149298,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004756721489891398,
      "loss": 5.7715,
      "mean_token_accuracy": 0.17174263596534728,
      "num_tokens": 30878671.0,
      "step": 15795
    },
    {
      "entropy": 6.165126132965088,
      "epoch": 1.5793972109761585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004756560050131239,
      "loss": 5.904,
      "mean_token_accuracy": 0.15668088793754578,
      "num_tokens": 30888064.0,
      "step": 15800
    },
    {
      "entropy": 6.138621664047241,
      "epoch": 1.579897036037387,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047563985598865466,
      "loss": 5.9125,
      "mean_token_accuracy": 0.16091287955641748,
      "num_tokens": 30898205.0,
      "step": 15805
    },
    {
      "entropy": 6.169529819488526,
      "epoch": 1.5803968610986154,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047562370191613816,
      "loss": 5.8743,
      "mean_token_accuracy": 0.1600824385881424,
      "num_tokens": 30907371.0,
      "step": 15810
    },
    {
      "entropy": 6.270576000213623,
      "epoch": 1.5808966861598441,
      "grad_norm": 1.25,
      "learning_rate": 0.0004756075427959811,
      "loss": 5.9795,
      "mean_token_accuracy": 0.1534248858690262,
      "num_tokens": 30917481.0,
      "step": 15815
    },
    {
      "entropy": 6.14884614944458,
      "epoch": 1.5813965112210726,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004755913786285898,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15823470056056976,
      "num_tokens": 30927375.0,
      "step": 15820
    },
    {
      "entropy": 6.048176908493042,
      "epoch": 1.581896336282301,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047557520941437093,
      "loss": 5.8092,
      "mean_token_accuracy": 0.16888693273067473,
      "num_tokens": 30935888.0,
      "step": 15825
    },
    {
      "entropy": 6.135527896881103,
      "epoch": 1.5823961613435298,
      "grad_norm": 1.25,
      "learning_rate": 0.00047555903515373156,
      "loss": 5.8523,
      "mean_token_accuracy": 0.1688093736767769,
      "num_tokens": 30946433.0,
      "step": 15830
    },
    {
      "entropy": 6.187525510787964,
      "epoch": 1.5828959864047585,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00047554285584707834,
      "loss": 5.8635,
      "mean_token_accuracy": 0.16482745110988617,
      "num_tokens": 30956951.0,
      "step": 15835
    },
    {
      "entropy": 6.019034337997437,
      "epoch": 1.583395811465987,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004755266714948186,
      "loss": 5.7279,
      "mean_token_accuracy": 0.1707964926958084,
      "num_tokens": 30967332.0,
      "step": 15840
    },
    {
      "entropy": 6.11483097076416,
      "epoch": 1.5838956365272154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004755104820973593,
      "loss": 5.9249,
      "mean_token_accuracy": 0.15547935962677,
      "num_tokens": 30977704.0,
      "step": 15845
    },
    {
      "entropy": 6.277363061904907,
      "epoch": 1.5843954615884441,
      "grad_norm": 1.515625,
      "learning_rate": 0.00047549428765510793,
      "loss": 5.9985,
      "mean_token_accuracy": 0.1495994433760643,
      "num_tokens": 30987916.0,
      "step": 15850
    },
    {
      "entropy": 6.1189018249511715,
      "epoch": 1.5848952866496726,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047547808816847186,
      "loss": 5.7789,
      "mean_token_accuracy": 0.17720956355333328,
      "num_tokens": 30997357.0,
      "step": 15855
    },
    {
      "entropy": 6.186125183105469,
      "epoch": 1.585395111710901,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004754618836378587,
      "loss": 5.9121,
      "mean_token_accuracy": 0.1553702786564827,
      "num_tokens": 31006113.0,
      "step": 15860
    },
    {
      "entropy": 6.058428144454956,
      "epoch": 1.5858949367721298,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004754456740636761,
      "loss": 5.7793,
      "mean_token_accuracy": 0.16829523891210557,
      "num_tokens": 31014952.0,
      "step": 15865
    },
    {
      "entropy": 6.244317150115966,
      "epoch": 1.5863947618333585,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047542945944633187,
      "loss": 5.9443,
      "mean_token_accuracy": 0.1488759696483612,
      "num_tokens": 31024934.0,
      "step": 15870
    },
    {
      "entropy": 6.117820978164673,
      "epoch": 1.5868945868945867,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047541323978623405,
      "loss": 5.7569,
      "mean_token_accuracy": 0.17177275717258453,
      "num_tokens": 31033934.0,
      "step": 15875
    },
    {
      "entropy": 6.119710493087768,
      "epoch": 1.5873944119558154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047539701508379073,
      "loss": 5.9298,
      "mean_token_accuracy": 0.16198905780911446,
      "num_tokens": 31043896.0,
      "step": 15880
    },
    {
      "entropy": 6.1023224830627445,
      "epoch": 1.5878942370170441,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004753807853394101,
      "loss": 5.8652,
      "mean_token_accuracy": 0.16319505870342255,
      "num_tokens": 31055103.0,
      "step": 15885
    },
    {
      "entropy": 6.190455389022827,
      "epoch": 1.5883940620782726,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047536455055350034,
      "loss": 5.9312,
      "mean_token_accuracy": 0.15753868818283082,
      "num_tokens": 31065478.0,
      "step": 15890
    },
    {
      "entropy": 6.125552320480347,
      "epoch": 1.588893887139501,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047534831072647016,
      "loss": 5.8255,
      "mean_token_accuracy": 0.16183240264654158,
      "num_tokens": 31074408.0,
      "step": 15895
    },
    {
      "entropy": 6.1248459815979,
      "epoch": 1.5893937122007298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047533206585872805,
      "loss": 5.8667,
      "mean_token_accuracy": 0.16485778540372847,
      "num_tokens": 31084223.0,
      "step": 15900
    },
    {
      "entropy": 6.053799343109131,
      "epoch": 1.5898935372619585,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004753158159506827,
      "loss": 5.7569,
      "mean_token_accuracy": 0.17025361359119415,
      "num_tokens": 31094133.0,
      "step": 15905
    },
    {
      "entropy": 6.161025905609131,
      "epoch": 1.5903933623231867,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004752995610027429,
      "loss": 5.862,
      "mean_token_accuracy": 0.16503244936466216,
      "num_tokens": 31104283.0,
      "step": 15910
    },
    {
      "entropy": 6.090030574798584,
      "epoch": 1.5908931873844154,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047528330101531777,
      "loss": 5.8276,
      "mean_token_accuracy": 0.15893703922629357,
      "num_tokens": 31114274.0,
      "step": 15915
    },
    {
      "entropy": 6.1467639923095705,
      "epoch": 1.5913930124456441,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047526703598881623,
      "loss": 5.8427,
      "mean_token_accuracy": 0.16548595130443572,
      "num_tokens": 31123680.0,
      "step": 15920
    },
    {
      "entropy": 6.103705930709839,
      "epoch": 1.5918928375068726,
      "grad_norm": 1.375,
      "learning_rate": 0.0004752507659236476,
      "loss": 5.7904,
      "mean_token_accuracy": 0.17176930457353592,
      "num_tokens": 31133520.0,
      "step": 15925
    },
    {
      "entropy": 5.973827838897705,
      "epoch": 1.592392662568101,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004752344908202213,
      "loss": 5.7382,
      "mean_token_accuracy": 0.17715881764888763,
      "num_tokens": 31142915.0,
      "step": 15930
    },
    {
      "entropy": 6.03070330619812,
      "epoch": 1.5928924876293298,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004752182106789467,
      "loss": 5.7589,
      "mean_token_accuracy": 0.1658123776316643,
      "num_tokens": 31152133.0,
      "step": 15935
    },
    {
      "entropy": 6.039987230300904,
      "epoch": 1.5933923126905583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047520192550023344,
      "loss": 5.7747,
      "mean_token_accuracy": 0.1715406596660614,
      "num_tokens": 31161666.0,
      "step": 15940
    },
    {
      "entropy": 6.054074478149414,
      "epoch": 1.5938921377517867,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004751856352844913,
      "loss": 5.8148,
      "mean_token_accuracy": 0.16528140008449554,
      "num_tokens": 31171591.0,
      "step": 15945
    },
    {
      "entropy": 6.212997913360596,
      "epoch": 1.5943919628130154,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047516934003213,
      "loss": 5.9348,
      "mean_token_accuracy": 0.15825078785419464,
      "num_tokens": 31181396.0,
      "step": 15950
    },
    {
      "entropy": 6.1554888725280765,
      "epoch": 1.5948917878742441,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047515303974355965,
      "loss": 5.8791,
      "mean_token_accuracy": 0.16116903126239776,
      "num_tokens": 31191926.0,
      "step": 15955
    },
    {
      "entropy": 6.1184299945831295,
      "epoch": 1.5953916129354726,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047513673441919036,
      "loss": 5.864,
      "mean_token_accuracy": 0.15740237087011338,
      "num_tokens": 31202809.0,
      "step": 15960
    },
    {
      "entropy": 6.139546871185303,
      "epoch": 1.595891437996701,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004751204240594323,
      "loss": 5.8326,
      "mean_token_accuracy": 0.16230722069740294,
      "num_tokens": 31211763.0,
      "step": 15965
    },
    {
      "entropy": 6.120927238464356,
      "epoch": 1.5963912630579298,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004751041086646959,
      "loss": 5.8468,
      "mean_token_accuracy": 0.15655516535043718,
      "num_tokens": 31220881.0,
      "step": 15970
    },
    {
      "entropy": 6.0632421493530275,
      "epoch": 1.5968910881191583,
      "grad_norm": 1.78125,
      "learning_rate": 0.0004750877882353915,
      "loss": 5.794,
      "mean_token_accuracy": 0.1687676802277565,
      "num_tokens": 31230740.0,
      "step": 15975
    },
    {
      "entropy": 6.018096780776977,
      "epoch": 1.5973909131803867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047507146277192997,
      "loss": 5.6552,
      "mean_token_accuracy": 0.1715314656496048,
      "num_tokens": 31239777.0,
      "step": 15980
    },
    {
      "entropy": 6.066577672958374,
      "epoch": 1.5978907382416154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004750551322747219,
      "loss": 5.8304,
      "mean_token_accuracy": 0.16165436059236526,
      "num_tokens": 31248574.0,
      "step": 15985
    },
    {
      "entropy": 6.11440601348877,
      "epoch": 1.5983905633028441,
      "grad_norm": 2.546875,
      "learning_rate": 0.0004750387967441781,
      "loss": 5.9047,
      "mean_token_accuracy": 0.16036269664764405,
      "num_tokens": 31258888.0,
      "step": 15990
    },
    {
      "entropy": 6.09486870765686,
      "epoch": 1.5988903883640726,
      "grad_norm": 2.25,
      "learning_rate": 0.00047502245618070976,
      "loss": 5.7846,
      "mean_token_accuracy": 0.17157715409994126,
      "num_tokens": 31269966.0,
      "step": 15995
    },
    {
      "entropy": 6.115769672393799,
      "epoch": 1.599390213425301,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047500611058472783,
      "loss": 5.8361,
      "mean_token_accuracy": 0.16435294970870018,
      "num_tokens": 31279746.0,
      "step": 16000
    },
    {
      "entropy": 6.044664716720581,
      "epoch": 1.5998900384865298,
      "grad_norm": 1.625,
      "learning_rate": 0.00047498975995664356,
      "loss": 5.8803,
      "mean_token_accuracy": 0.15991793125867843,
      "num_tokens": 31289677.0,
      "step": 16005
    },
    {
      "entropy": 6.062617492675781,
      "epoch": 1.6003898635477583,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004749734042968685,
      "loss": 5.894,
      "mean_token_accuracy": 0.1650659702718258,
      "num_tokens": 31299752.0,
      "step": 16010
    },
    {
      "entropy": 6.08680830001831,
      "epoch": 1.6008896886089867,
      "grad_norm": 1.296875,
      "learning_rate": 0.000474957043605814,
      "loss": 5.7699,
      "mean_token_accuracy": 0.16378215402364732,
      "num_tokens": 31308986.0,
      "step": 16015
    },
    {
      "entropy": 6.089492082595825,
      "epoch": 1.6013895136702154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047494067788389174,
      "loss": 5.7554,
      "mean_token_accuracy": 0.17349808067083358,
      "num_tokens": 31318848.0,
      "step": 16020
    },
    {
      "entropy": 6.136660051345825,
      "epoch": 1.6018893387314441,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047492430713151347,
      "loss": 5.8885,
      "mean_token_accuracy": 0.15980696976184844,
      "num_tokens": 31329397.0,
      "step": 16025
    },
    {
      "entropy": 6.219451570510865,
      "epoch": 1.6023891637926726,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004749079313490911,
      "loss": 5.9444,
      "mean_token_accuracy": 0.15631850808858871,
      "num_tokens": 31340440.0,
      "step": 16030
    },
    {
      "entropy": 6.1272295951843265,
      "epoch": 1.602888988853901,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047489155053703655,
      "loss": 5.8857,
      "mean_token_accuracy": 0.1611655905842781,
      "num_tokens": 31350418.0,
      "step": 16035
    },
    {
      "entropy": 6.070876932144165,
      "epoch": 1.6033888139151298,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004748751646957621,
      "loss": 5.8305,
      "mean_token_accuracy": 0.1658831998705864,
      "num_tokens": 31360714.0,
      "step": 16040
    },
    {
      "entropy": 6.147923231124878,
      "epoch": 1.6038886389763582,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004748587738256799,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16950891315937042,
      "num_tokens": 31369693.0,
      "step": 16045
    },
    {
      "entropy": 6.051037693023682,
      "epoch": 1.6043884640375867,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004748423779272024,
      "loss": 5.8123,
      "mean_token_accuracy": 0.16109999790787696,
      "num_tokens": 31380068.0,
      "step": 16050
    },
    {
      "entropy": 6.122638463973999,
      "epoch": 1.6048882890988154,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047482597700074204,
      "loss": 5.8566,
      "mean_token_accuracy": 0.1613924875855446,
      "num_tokens": 31389151.0,
      "step": 16055
    },
    {
      "entropy": 6.17003288269043,
      "epoch": 1.6053881141600441,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00047480957104671157,
      "loss": 5.9916,
      "mean_token_accuracy": 0.14955553263425828,
      "num_tokens": 31400260.0,
      "step": 16060
    },
    {
      "entropy": 6.2190080165863035,
      "epoch": 1.6058879392212726,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004747931600655237,
      "loss": 5.8873,
      "mean_token_accuracy": 0.15634534806013106,
      "num_tokens": 31410747.0,
      "step": 16065
    },
    {
      "entropy": 6.116169738769531,
      "epoch": 1.606387764282501,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047477674405759136,
      "loss": 5.8103,
      "mean_token_accuracy": 0.1668058693408966,
      "num_tokens": 31419230.0,
      "step": 16070
    },
    {
      "entropy": 6.0804932594299315,
      "epoch": 1.6068875893437298,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004747603230233275,
      "loss": 5.7983,
      "mean_token_accuracy": 0.16895988285541536,
      "num_tokens": 31428314.0,
      "step": 16075
    },
    {
      "entropy": 6.006974172592163,
      "epoch": 1.6073874144049582,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0004747438969631453,
      "loss": 5.7879,
      "mean_token_accuracy": 0.1638770133256912,
      "num_tokens": 31438141.0,
      "step": 16080
    },
    {
      "entropy": 6.0893206119537355,
      "epoch": 1.6078872394661867,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004747274658774581,
      "loss": 5.8444,
      "mean_token_accuracy": 0.1598839282989502,
      "num_tokens": 31448609.0,
      "step": 16085
    },
    {
      "entropy": 6.156227111816406,
      "epoch": 1.6083870645274154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004747110297666793,
      "loss": 5.8622,
      "mean_token_accuracy": 0.159651480615139,
      "num_tokens": 31458603.0,
      "step": 16090
    },
    {
      "entropy": 6.104046201705932,
      "epoch": 1.6088868895886441,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004746945886312223,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17387848496437072,
      "num_tokens": 31468917.0,
      "step": 16095
    },
    {
      "entropy": 6.064031267166138,
      "epoch": 1.6093867146498726,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004746781424715009,
      "loss": 5.8692,
      "mean_token_accuracy": 0.15919560045003892,
      "num_tokens": 31479061.0,
      "step": 16100
    },
    {
      "entropy": 6.134131193161011,
      "epoch": 1.609886539711101,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047466169128792873,
      "loss": 5.9379,
      "mean_token_accuracy": 0.15091839134693147,
      "num_tokens": 31488723.0,
      "step": 16105
    },
    {
      "entropy": 6.108094692230225,
      "epoch": 1.6103863647723298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004746452350809198,
      "loss": 5.8654,
      "mean_token_accuracy": 0.16363528221845627,
      "num_tokens": 31497715.0,
      "step": 16110
    },
    {
      "entropy": 6.101785278320312,
      "epoch": 1.6108861898335582,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047462877385088815,
      "loss": 5.8137,
      "mean_token_accuracy": 0.16396145522594452,
      "num_tokens": 31507571.0,
      "step": 16115
    },
    {
      "entropy": 6.03979721069336,
      "epoch": 1.6113860148947867,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047461230759824794,
      "loss": 5.8595,
      "mean_token_accuracy": 0.16097067594528197,
      "num_tokens": 31517582.0,
      "step": 16120
    },
    {
      "entropy": 6.1500955581665036,
      "epoch": 1.6118858399560154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047459583632341343,
      "loss": 5.9003,
      "mean_token_accuracy": 0.16386755257844926,
      "num_tokens": 31528225.0,
      "step": 16125
    },
    {
      "entropy": 6.1058070182800295,
      "epoch": 1.6123856650172441,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047457936002679905,
      "loss": 5.7112,
      "mean_token_accuracy": 0.1699998989701271,
      "num_tokens": 31537778.0,
      "step": 16130
    },
    {
      "entropy": 6.040977239608765,
      "epoch": 1.6128854900784724,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00047456287870881924,
      "loss": 5.8032,
      "mean_token_accuracy": 0.16930240243673325,
      "num_tokens": 31548431.0,
      "step": 16135
    },
    {
      "entropy": 6.05656476020813,
      "epoch": 1.613385315139701,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004745463923698888,
      "loss": 5.7655,
      "mean_token_accuracy": 0.1716919168829918,
      "num_tokens": 31557333.0,
      "step": 16140
    },
    {
      "entropy": 6.05495433807373,
      "epoch": 1.6138851402009298,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004745299010104226,
      "loss": 5.8011,
      "mean_token_accuracy": 0.1625152423977852,
      "num_tokens": 31566333.0,
      "step": 16145
    },
    {
      "entropy": 5.978433322906494,
      "epoch": 1.6143849652621582,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047451340463083527,
      "loss": 5.7504,
      "mean_token_accuracy": 0.17534879297018052,
      "num_tokens": 31575971.0,
      "step": 16150
    },
    {
      "entropy": 6.04327359199524,
      "epoch": 1.6148847903233867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047449690323154206,
      "loss": 5.7903,
      "mean_token_accuracy": 0.17007804214954375,
      "num_tokens": 31585396.0,
      "step": 16155
    },
    {
      "entropy": 6.105011796951294,
      "epoch": 1.6153846153846154,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00047448039681295814,
      "loss": 5.865,
      "mean_token_accuracy": 0.15810033455491065,
      "num_tokens": 31595800.0,
      "step": 16160
    },
    {
      "entropy": 6.128646326065064,
      "epoch": 1.6158844404458441,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047446388537549867,
      "loss": 5.8682,
      "mean_token_accuracy": 0.1624359592795372,
      "num_tokens": 31604792.0,
      "step": 16165
    },
    {
      "entropy": 6.107560110092163,
      "epoch": 1.6163842655070724,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004744473689195793,
      "loss": 5.8754,
      "mean_token_accuracy": 0.15935287103056908,
      "num_tokens": 31614952.0,
      "step": 16170
    },
    {
      "entropy": 6.064169597625733,
      "epoch": 1.616884090568301,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004744308474456154,
      "loss": 5.803,
      "mean_token_accuracy": 0.17217198014259338,
      "num_tokens": 31624635.0,
      "step": 16175
    },
    {
      "entropy": 6.041958904266357,
      "epoch": 1.6173839156295298,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004744143209540227,
      "loss": 5.7664,
      "mean_token_accuracy": 0.170831997692585,
      "num_tokens": 31634710.0,
      "step": 16180
    },
    {
      "entropy": 6.233108758926392,
      "epoch": 1.6178837406907582,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004743977894452169,
      "loss": 5.9275,
      "mean_token_accuracy": 0.15390657782554626,
      "num_tokens": 31644371.0,
      "step": 16185
    },
    {
      "entropy": 6.128524303436279,
      "epoch": 1.6183835657519867,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004743812529196142,
      "loss": 5.8071,
      "mean_token_accuracy": 0.16524098366498946,
      "num_tokens": 31654262.0,
      "step": 16190
    },
    {
      "entropy": 6.1295530796051025,
      "epoch": 1.6188833908132154,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004743647113776304,
      "loss": 5.8821,
      "mean_token_accuracy": 0.1575243815779686,
      "num_tokens": 31663209.0,
      "step": 16195
    },
    {
      "entropy": 6.019235229492187,
      "epoch": 1.619383215874444,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047434816481968175,
      "loss": 5.8218,
      "mean_token_accuracy": 0.16196248382329942,
      "num_tokens": 31672370.0,
      "step": 16200
    },
    {
      "entropy": 6.1180863857269285,
      "epoch": 1.6198830409356724,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047433161324618456,
      "loss": 5.8706,
      "mean_token_accuracy": 0.16636834293603897,
      "num_tokens": 31681886.0,
      "step": 16205
    },
    {
      "entropy": 6.158323907852173,
      "epoch": 1.620382865996901,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004743150566575553,
      "loss": 5.8064,
      "mean_token_accuracy": 0.17043220698833467,
      "num_tokens": 31691098.0,
      "step": 16210
    },
    {
      "entropy": 6.049533033370972,
      "epoch": 1.6208826910581298,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004742984950542105,
      "loss": 5.7884,
      "mean_token_accuracy": 0.1622710943222046,
      "num_tokens": 31700638.0,
      "step": 16215
    },
    {
      "entropy": 6.031154680252075,
      "epoch": 1.6213825161193582,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00047428192843656684,
      "loss": 5.8423,
      "mean_token_accuracy": 0.1616874635219574,
      "num_tokens": 31711326.0,
      "step": 16220
    },
    {
      "entropy": 6.0814954280853275,
      "epoch": 1.6218823411805867,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047426535680504105,
      "loss": 5.7973,
      "mean_token_accuracy": 0.17051303386688232,
      "num_tokens": 31721220.0,
      "step": 16225
    },
    {
      "entropy": 6.0870726108551025,
      "epoch": 1.6223821662418154,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047424878016005036,
      "loss": 5.8528,
      "mean_token_accuracy": 0.16325292289257048,
      "num_tokens": 31731190.0,
      "step": 16230
    },
    {
      "entropy": 6.165756750106811,
      "epoch": 1.622881991303044,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0004742321985020115,
      "loss": 5.8708,
      "mean_token_accuracy": 0.1659107655286789,
      "num_tokens": 31740761.0,
      "step": 16235
    },
    {
      "entropy": 6.115328359603882,
      "epoch": 1.6233818163642724,
      "grad_norm": 1.25,
      "learning_rate": 0.0004742156118313418,
      "loss": 5.8706,
      "mean_token_accuracy": 0.16334801465272902,
      "num_tokens": 31750038.0,
      "step": 16240
    },
    {
      "entropy": 6.146216678619385,
      "epoch": 1.623881641425501,
      "grad_norm": 1.140625,
      "learning_rate": 0.00047419902014845855,
      "loss": 5.8031,
      "mean_token_accuracy": 0.16566898673772812,
      "num_tokens": 31761574.0,
      "step": 16245
    },
    {
      "entropy": 6.0853657722473145,
      "epoch": 1.6243814664867298,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004741824234537793,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16479350328445436,
      "num_tokens": 31771980.0,
      "step": 16250
    },
    {
      "entropy": 6.070226764678955,
      "epoch": 1.6248812915479582,
      "grad_norm": 1.375,
      "learning_rate": 0.00047416582174772143,
      "loss": 5.8274,
      "mean_token_accuracy": 0.16199822574853898,
      "num_tokens": 31781242.0,
      "step": 16255
    },
    {
      "entropy": 6.202419281005859,
      "epoch": 1.6253811166091867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047414921503070276,
      "loss": 5.9052,
      "mean_token_accuracy": 0.1622503623366356,
      "num_tokens": 31792250.0,
      "step": 16260
    },
    {
      "entropy": 6.07946548461914,
      "epoch": 1.6258809416704154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047413260330314116,
      "loss": 5.7727,
      "mean_token_accuracy": 0.16607222110033035,
      "num_tokens": 31802512.0,
      "step": 16265
    },
    {
      "entropy": 6.038339757919312,
      "epoch": 1.626380766731644,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004741159865654544,
      "loss": 5.729,
      "mean_token_accuracy": 0.17349571436643602,
      "num_tokens": 31811957.0,
      "step": 16270
    },
    {
      "entropy": 6.101879644393921,
      "epoch": 1.6268805917928724,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047409936481806076,
      "loss": 5.8231,
      "mean_token_accuracy": 0.16853295117616654,
      "num_tokens": 31821101.0,
      "step": 16275
    },
    {
      "entropy": 6.130295038223267,
      "epoch": 1.627380416854101,
      "grad_norm": 1.375,
      "learning_rate": 0.0004740827380613782,
      "loss": 5.6867,
      "mean_token_accuracy": 0.16625203788280488,
      "num_tokens": 31831085.0,
      "step": 16280
    },
    {
      "entropy": 6.0749116897583,
      "epoch": 1.6278802419153298,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047406610629582526,
      "loss": 5.8454,
      "mean_token_accuracy": 0.1650839328765869,
      "num_tokens": 31840354.0,
      "step": 16285
    },
    {
      "entropy": 6.010754680633545,
      "epoch": 1.6283800669765582,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047404946952182034,
      "loss": 5.7837,
      "mean_token_accuracy": 0.16882811039686202,
      "num_tokens": 31849250.0,
      "step": 16290
    },
    {
      "entropy": 6.10916166305542,
      "epoch": 1.6288798920377867,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004740328277397819,
      "loss": 5.8323,
      "mean_token_accuracy": 0.16149095296859742,
      "num_tokens": 31859533.0,
      "step": 16295
    },
    {
      "entropy": 6.193563222885132,
      "epoch": 1.6293797170990154,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004740161809501288,
      "loss": 5.9022,
      "mean_token_accuracy": 0.15910781174898148,
      "num_tokens": 31868821.0,
      "step": 16300
    },
    {
      "entropy": 6.138738107681275,
      "epoch": 1.6298795421602439,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00047399952915327975,
      "loss": 5.8308,
      "mean_token_accuracy": 0.15946413725614547,
      "num_tokens": 31878084.0,
      "step": 16305
    },
    {
      "entropy": 6.075612592697143,
      "epoch": 1.6303793672214724,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004739828723496538,
      "loss": 5.7909,
      "mean_token_accuracy": 0.16261777132749558,
      "num_tokens": 31887634.0,
      "step": 16310
    },
    {
      "entropy": 6.019099235534668,
      "epoch": 1.630879192282701,
      "grad_norm": 1.25,
      "learning_rate": 0.00047396621053967,
      "loss": 5.7929,
      "mean_token_accuracy": 0.17053304314613343,
      "num_tokens": 31897871.0,
      "step": 16315
    },
    {
      "entropy": 6.119353866577148,
      "epoch": 1.6313790173439298,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047394954372374743,
      "loss": 5.8632,
      "mean_token_accuracy": 0.16035366654396058,
      "num_tokens": 31907960.0,
      "step": 16320
    },
    {
      "entropy": 6.119000482559204,
      "epoch": 1.6318788424051582,
      "grad_norm": 1.125,
      "learning_rate": 0.00047393287190230566,
      "loss": 5.8288,
      "mean_token_accuracy": 0.15797059535980223,
      "num_tokens": 31918550.0,
      "step": 16325
    },
    {
      "entropy": 6.154240369796753,
      "epoch": 1.6323786674663867,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004739161950757639,
      "loss": 5.9393,
      "mean_token_accuracy": 0.16078022569417955,
      "num_tokens": 31927941.0,
      "step": 16330
    },
    {
      "entropy": 6.169876337051392,
      "epoch": 1.6328784925276154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000473899513244542,
      "loss": 5.8912,
      "mean_token_accuracy": 0.16059373915195466,
      "num_tokens": 31938179.0,
      "step": 16335
    },
    {
      "entropy": 6.069890689849854,
      "epoch": 1.6333783175888439,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004738828264090594,
      "loss": 5.7495,
      "mean_token_accuracy": 0.1718252718448639,
      "num_tokens": 31947287.0,
      "step": 16340
    },
    {
      "entropy": 5.9992509365081785,
      "epoch": 1.6338781426500724,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047386613456973614,
      "loss": 5.7659,
      "mean_token_accuracy": 0.17057114690542222,
      "num_tokens": 31955953.0,
      "step": 16345
    },
    {
      "entropy": 6.052100753784179,
      "epoch": 1.634377967711301,
      "grad_norm": 2.359375,
      "learning_rate": 0.0004738494377269921,
      "loss": 5.8007,
      "mean_token_accuracy": 0.16357877850532532,
      "num_tokens": 31964874.0,
      "step": 16350
    },
    {
      "entropy": 6.133013296127319,
      "epoch": 1.6348777927725298,
      "grad_norm": 1.25,
      "learning_rate": 0.00047383273588124736,
      "loss": 5.8876,
      "mean_token_accuracy": 0.1596738412976265,
      "num_tokens": 31975865.0,
      "step": 16355
    },
    {
      "entropy": 6.079090547561646,
      "epoch": 1.6353776178337582,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004738160290329222,
      "loss": 5.7837,
      "mean_token_accuracy": 0.16713957935571672,
      "num_tokens": 31985229.0,
      "step": 16360
    },
    {
      "entropy": 6.038175058364868,
      "epoch": 1.6358774428949867,
      "grad_norm": 1.09375,
      "learning_rate": 0.00047379931718243684,
      "loss": 5.8908,
      "mean_token_accuracy": 0.16973757445812226,
      "num_tokens": 31995195.0,
      "step": 16365
    },
    {
      "entropy": 6.130444717407227,
      "epoch": 1.6363772679562154,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004737826003302118,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16212511360645293,
      "num_tokens": 32006053.0,
      "step": 16370
    },
    {
      "entropy": 6.15709753036499,
      "epoch": 1.6368770930174439,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047376587847666767,
      "loss": 5.8842,
      "mean_token_accuracy": 0.15770581513643264,
      "num_tokens": 32015025.0,
      "step": 16375
    },
    {
      "entropy": 6.025450086593628,
      "epoch": 1.6373769180786724,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004737491516222253,
      "loss": 5.7522,
      "mean_token_accuracy": 0.1685287207365036,
      "num_tokens": 32025116.0,
      "step": 16380
    },
    {
      "entropy": 6.083757925033569,
      "epoch": 1.637876743139901,
      "grad_norm": 1.46875,
      "learning_rate": 0.00047373241976730537,
      "loss": 5.8145,
      "mean_token_accuracy": 0.16748420745134354,
      "num_tokens": 32034271.0,
      "step": 16385
    },
    {
      "entropy": 6.079638719558716,
      "epoch": 1.6383765682011298,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004737156829123289,
      "loss": 5.7117,
      "mean_token_accuracy": 0.18040030896663667,
      "num_tokens": 32043139.0,
      "step": 16390
    },
    {
      "entropy": 6.068140888214112,
      "epoch": 1.638876393262358,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047369894105771693,
      "loss": 5.7701,
      "mean_token_accuracy": 0.1635582000017166,
      "num_tokens": 32052719.0,
      "step": 16395
    },
    {
      "entropy": 6.14244122505188,
      "epoch": 1.6393762183235867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047368219420389077,
      "loss": 5.8776,
      "mean_token_accuracy": 0.15964355021715165,
      "num_tokens": 32062430.0,
      "step": 16400
    },
    {
      "entropy": 6.086572074890137,
      "epoch": 1.6398760433848154,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004736654423512717,
      "loss": 5.7858,
      "mean_token_accuracy": 0.16606002449989318,
      "num_tokens": 32072396.0,
      "step": 16405
    },
    {
      "entropy": 6.102981662750244,
      "epoch": 1.6403758684460439,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004736486855002813,
      "loss": 5.8699,
      "mean_token_accuracy": 0.15601673871278762,
      "num_tokens": 32082200.0,
      "step": 16410
    },
    {
      "entropy": 6.092888641357422,
      "epoch": 1.6408756935072724,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000473631923651341,
      "loss": 5.7761,
      "mean_token_accuracy": 0.17181171476840973,
      "num_tokens": 32091670.0,
      "step": 16415
    },
    {
      "entropy": 6.11961088180542,
      "epoch": 1.641375518568501,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004736151568048726,
      "loss": 5.9227,
      "mean_token_accuracy": 0.15400823205709457,
      "num_tokens": 32102072.0,
      "step": 16420
    },
    {
      "entropy": 6.060141372680664,
      "epoch": 1.6418753436297298,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000473598384961298,
      "loss": 5.7312,
      "mean_token_accuracy": 0.16847492307424544,
      "num_tokens": 32111367.0,
      "step": 16425
    },
    {
      "entropy": 6.131742811203003,
      "epoch": 1.642375168690958,
      "grad_norm": 1.796875,
      "learning_rate": 0.00047358160812103925,
      "loss": 5.7966,
      "mean_token_accuracy": 0.16571346223354338,
      "num_tokens": 32121065.0,
      "step": 16430
    },
    {
      "entropy": 6.147887897491455,
      "epoch": 1.6428749937521867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004735648262845182,
      "loss": 5.8319,
      "mean_token_accuracy": 0.1572011500597,
      "num_tokens": 32130142.0,
      "step": 16435
    },
    {
      "entropy": 6.124507904052734,
      "epoch": 1.6433748188134154,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047354803945215727,
      "loss": 5.9434,
      "mean_token_accuracy": 0.16040437892079354,
      "num_tokens": 32140689.0,
      "step": 16440
    },
    {
      "entropy": 6.05601921081543,
      "epoch": 1.6438746438746439,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047353124762437865,
      "loss": 5.7124,
      "mean_token_accuracy": 0.1761801838874817,
      "num_tokens": 32150888.0,
      "step": 16445
    },
    {
      "entropy": 6.1184571266174315,
      "epoch": 1.6443744689358724,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004735144508016051,
      "loss": 5.9402,
      "mean_token_accuracy": 0.1562290295958519,
      "num_tokens": 32161188.0,
      "step": 16450
    },
    {
      "entropy": 6.075448417663575,
      "epoch": 1.644874293997101,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004734976489842589,
      "loss": 5.8078,
      "mean_token_accuracy": 0.1667831540107727,
      "num_tokens": 32171684.0,
      "step": 16455
    },
    {
      "entropy": 6.130577850341797,
      "epoch": 1.6453741190583295,
      "grad_norm": 1.578125,
      "learning_rate": 0.000473480842172763,
      "loss": 5.8897,
      "mean_token_accuracy": 0.15875401496887206,
      "num_tokens": 32181497.0,
      "step": 16460
    },
    {
      "entropy": 6.125845527648925,
      "epoch": 1.645873944119558,
      "grad_norm": 1.25,
      "learning_rate": 0.00047346403036754023,
      "loss": 5.7575,
      "mean_token_accuracy": 0.16723476499319076,
      "num_tokens": 32190653.0,
      "step": 16465
    },
    {
      "entropy": 6.089094638824463,
      "epoch": 1.6463737691807867,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047344721356901344,
      "loss": 5.8546,
      "mean_token_accuracy": 0.16362821757793428,
      "num_tokens": 32200367.0,
      "step": 16470
    },
    {
      "entropy": 6.094277286529541,
      "epoch": 1.6468735942420154,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004734303917776059,
      "loss": 5.9003,
      "mean_token_accuracy": 0.15828943699598313,
      "num_tokens": 32210187.0,
      "step": 16475
    },
    {
      "entropy": 6.172597503662109,
      "epoch": 1.6473734193032439,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047341356499374065,
      "loss": 5.802,
      "mean_token_accuracy": 0.16504286378622054,
      "num_tokens": 32218754.0,
      "step": 16480
    },
    {
      "entropy": 6.107746171951294,
      "epoch": 1.6478732443644724,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004733967332178412,
      "loss": 5.8169,
      "mean_token_accuracy": 0.1639198824763298,
      "num_tokens": 32228105.0,
      "step": 16485
    },
    {
      "entropy": 6.112506961822509,
      "epoch": 1.648373069425701,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004733798964503311,
      "loss": 5.8623,
      "mean_token_accuracy": 0.16502905786037445,
      "num_tokens": 32239303.0,
      "step": 16490
    },
    {
      "entropy": 6.032963657379151,
      "epoch": 1.6488728944869295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004733630546916338,
      "loss": 5.7112,
      "mean_token_accuracy": 0.17206179052591325,
      "num_tokens": 32248614.0,
      "step": 16495
    },
    {
      "entropy": 6.08461709022522,
      "epoch": 1.649372719548158,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047334620794217306,
      "loss": 5.87,
      "mean_token_accuracy": 0.16419921219348907,
      "num_tokens": 32257012.0,
      "step": 16500
    },
    {
      "entropy": 6.111689567565918,
      "epoch": 1.6498725446093867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004733293562023727,
      "loss": 5.7954,
      "mean_token_accuracy": 0.16484722197055818,
      "num_tokens": 32267990.0,
      "step": 16505
    },
    {
      "entropy": 6.118923711776733,
      "epoch": 1.6503723696706154,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047331249947265685,
      "loss": 5.7684,
      "mean_token_accuracy": 0.16947761923074722,
      "num_tokens": 32277519.0,
      "step": 16510
    },
    {
      "entropy": 6.054642391204834,
      "epoch": 1.6508721947318439,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004732956377534496,
      "loss": 5.7913,
      "mean_token_accuracy": 0.16124063730239868,
      "num_tokens": 32286749.0,
      "step": 16515
    },
    {
      "entropy": 6.148294067382812,
      "epoch": 1.6513720197930724,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000473278771045175,
      "loss": 5.9654,
      "mean_token_accuracy": 0.15079726427793502,
      "num_tokens": 32296344.0,
      "step": 16520
    },
    {
      "entropy": 6.158928918838501,
      "epoch": 1.651871844854301,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047326189934825764,
      "loss": 5.7998,
      "mean_token_accuracy": 0.16486818790435792,
      "num_tokens": 32305857.0,
      "step": 16525
    },
    {
      "entropy": 6.082451868057251,
      "epoch": 1.6523716699155295,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0004732450226631219,
      "loss": 5.8132,
      "mean_token_accuracy": 0.1678343653678894,
      "num_tokens": 32315265.0,
      "step": 16530
    },
    {
      "entropy": 6.2120177268981935,
      "epoch": 1.652871494976758,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047322814099019236,
      "loss": 5.9519,
      "mean_token_accuracy": 0.15277455300092696,
      "num_tokens": 32324724.0,
      "step": 16535
    },
    {
      "entropy": 6.134037351608276,
      "epoch": 1.6533713200379867,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004732112543298938,
      "loss": 5.8358,
      "mean_token_accuracy": 0.16458573937416077,
      "num_tokens": 32334958.0,
      "step": 16540
    },
    {
      "entropy": 6.098423910140991,
      "epoch": 1.6538711450992154,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004731943626826511,
      "loss": 5.9167,
      "mean_token_accuracy": 0.1607313871383667,
      "num_tokens": 32345153.0,
      "step": 16545
    },
    {
      "entropy": 6.114382123947143,
      "epoch": 1.6543709701604439,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047317746604888923,
      "loss": 5.7889,
      "mean_token_accuracy": 0.16762964725494384,
      "num_tokens": 32354692.0,
      "step": 16550
    },
    {
      "entropy": 6.151679182052613,
      "epoch": 1.6548707952216724,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047316056442903333,
      "loss": 5.8719,
      "mean_token_accuracy": 0.15983538776636125,
      "num_tokens": 32363614.0,
      "step": 16555
    },
    {
      "entropy": 6.182926034927368,
      "epoch": 1.655370620282901,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0004731436578235086,
      "loss": 5.8349,
      "mean_token_accuracy": 0.1624493904411793,
      "num_tokens": 32373139.0,
      "step": 16560
    },
    {
      "entropy": 6.159624338150024,
      "epoch": 1.6558704453441295,
      "grad_norm": 1.484375,
      "learning_rate": 0.00047312674623274046,
      "loss": 5.8551,
      "mean_token_accuracy": 0.1513068586587906,
      "num_tokens": 32383624.0,
      "step": 16565
    },
    {
      "entropy": 6.051584959030151,
      "epoch": 1.656370270405358,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004731098296571543,
      "loss": 5.6903,
      "mean_token_accuracy": 0.16859373450279236,
      "num_tokens": 32393221.0,
      "step": 16570
    },
    {
      "entropy": 6.115183734893799,
      "epoch": 1.6568700954665867,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047309290809717584,
      "loss": 5.8881,
      "mean_token_accuracy": 0.16096863597631456,
      "num_tokens": 32402908.0,
      "step": 16575
    },
    {
      "entropy": 6.0643336296081545,
      "epoch": 1.6573699205278154,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047307598155323083,
      "loss": 5.8268,
      "mean_token_accuracy": 0.15815944820642472,
      "num_tokens": 32413597.0,
      "step": 16580
    },
    {
      "entropy": 6.12480993270874,
      "epoch": 1.6578697455890439,
      "grad_norm": 1.625,
      "learning_rate": 0.000473059050025745,
      "loss": 5.8401,
      "mean_token_accuracy": 0.16443217694759368,
      "num_tokens": 32423948.0,
      "step": 16585
    },
    {
      "entropy": 6.095758581161499,
      "epoch": 1.6583695706502724,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004730421135151445,
      "loss": 5.805,
      "mean_token_accuracy": 0.16168527454137802,
      "num_tokens": 32433176.0,
      "step": 16590
    },
    {
      "entropy": 6.007092761993408,
      "epoch": 1.658869395711501,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004730251720218554,
      "loss": 5.7785,
      "mean_token_accuracy": 0.16693072766065598,
      "num_tokens": 32442868.0,
      "step": 16595
    },
    {
      "entropy": 6.0971451759338375,
      "epoch": 1.6593692207727295,
      "grad_norm": 1.4375,
      "learning_rate": 0.00047300822554630393,
      "loss": 5.8154,
      "mean_token_accuracy": 0.16690618693828582,
      "num_tokens": 32451966.0,
      "step": 16600
    },
    {
      "entropy": 6.170837116241455,
      "epoch": 1.659869045833958,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004729912740889164,
      "loss": 5.8276,
      "mean_token_accuracy": 0.17084040492773056,
      "num_tokens": 32462442.0,
      "step": 16605
    },
    {
      "entropy": 6.145010709762573,
      "epoch": 1.6603688708951867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047297431765011945,
      "loss": 5.7589,
      "mean_token_accuracy": 0.17135907113552093,
      "num_tokens": 32471709.0,
      "step": 16610
    },
    {
      "entropy": 6.089041137695313,
      "epoch": 1.6608686959564154,
      "grad_norm": 1.1875,
      "learning_rate": 0.00047295735623033953,
      "loss": 5.8368,
      "mean_token_accuracy": 0.16932764202356337,
      "num_tokens": 32482147.0,
      "step": 16615
    },
    {
      "entropy": 6.061543416976929,
      "epoch": 1.6613685210176439,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004729403898300035,
      "loss": 5.7421,
      "mean_token_accuracy": 0.1719137504696846,
      "num_tokens": 32492626.0,
      "step": 16620
    },
    {
      "entropy": 6.079841184616089,
      "epoch": 1.6618683460788723,
      "grad_norm": 1.25,
      "learning_rate": 0.00047292341844953813,
      "loss": 5.7884,
      "mean_token_accuracy": 0.17396891117095947,
      "num_tokens": 32502887.0,
      "step": 16625
    },
    {
      "entropy": 6.1855528354644775,
      "epoch": 1.662368171140101,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004729064420893706,
      "loss": 5.8915,
      "mean_token_accuracy": 0.1603488102555275,
      "num_tokens": 32512860.0,
      "step": 16630
    },
    {
      "entropy": 6.199327802658081,
      "epoch": 1.6628679962013295,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004728894607499278,
      "loss": 5.9473,
      "mean_token_accuracy": 0.1513063997030258,
      "num_tokens": 32523103.0,
      "step": 16635
    },
    {
      "entropy": 6.055564737319946,
      "epoch": 1.663367821262558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004728724744316371,
      "loss": 5.7512,
      "mean_token_accuracy": 0.16596733629703522,
      "num_tokens": 32532940.0,
      "step": 16640
    },
    {
      "entropy": 6.176367378234863,
      "epoch": 1.6638676463237867,
      "grad_norm": 1.4140625,
      "learning_rate": 0.00047285548313492594,
      "loss": 5.8776,
      "mean_token_accuracy": 0.162559212744236,
      "num_tokens": 32543503.0,
      "step": 16645
    },
    {
      "entropy": 6.143779659271241,
      "epoch": 1.6643674713850154,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004728384868602217,
      "loss": 5.7509,
      "mean_token_accuracy": 0.17365916669368744,
      "num_tokens": 32552025.0,
      "step": 16650
    },
    {
      "entropy": 6.0741345405578615,
      "epoch": 1.6648672964462436,
      "grad_norm": 1.3828125,
      "learning_rate": 0.000472821485607952,
      "loss": 5.7635,
      "mean_token_accuracy": 0.16234611719846725,
      "num_tokens": 32561262.0,
      "step": 16655
    },
    {
      "entropy": 6.082498216629029,
      "epoch": 1.6653671215074723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004728044793785446,
      "loss": 5.8501,
      "mean_token_accuracy": 0.16009418666362762,
      "num_tokens": 32572067.0,
      "step": 16660
    },
    {
      "entropy": 6.176837968826294,
      "epoch": 1.665866946568701,
      "grad_norm": 1.40625,
      "learning_rate": 0.00047278746817242754,
      "loss": 5.8361,
      "mean_token_accuracy": 0.17092930376529694,
      "num_tokens": 32582033.0,
      "step": 16665
    },
    {
      "entropy": 6.129619789123535,
      "epoch": 1.6663667716299295,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004727704519900286,
      "loss": 5.821,
      "mean_token_accuracy": 0.16560350358486176,
      "num_tokens": 32592338.0,
      "step": 16670
    },
    {
      "entropy": 6.105005979537964,
      "epoch": 1.666866596691158,
      "grad_norm": 1.328125,
      "learning_rate": 0.000472753430831776,
      "loss": 5.7835,
      "mean_token_accuracy": 0.16761494502425195,
      "num_tokens": 32602310.0,
      "step": 16675
    },
    {
      "entropy": 6.198736333847046,
      "epoch": 1.6673664217523867,
      "grad_norm": 1.421875,
      "learning_rate": 0.00047273640469809786,
      "loss": 5.9063,
      "mean_token_accuracy": 0.15738597139716148,
      "num_tokens": 32612632.0,
      "step": 16680
    },
    {
      "entropy": 6.135696268081665,
      "epoch": 1.6678662468136154,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004727193735894227,
      "loss": 5.801,
      "mean_token_accuracy": 0.16541199535131454,
      "num_tokens": 32622455.0,
      "step": 16685
    },
    {
      "entropy": 6.028338527679443,
      "epoch": 1.6683660718748436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047270233750617907,
      "loss": 5.7874,
      "mean_token_accuracy": 0.16524736434221268,
      "num_tokens": 32631854.0,
      "step": 16690
    },
    {
      "entropy": 6.0832243919372555,
      "epoch": 1.6688658969360723,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047268529644879545,
      "loss": 5.8577,
      "mean_token_accuracy": 0.16892827898263932,
      "num_tokens": 32641532.0,
      "step": 16695
    },
    {
      "entropy": 6.140908575057983,
      "epoch": 1.669365721997301,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047266825041770066,
      "loss": 5.8326,
      "mean_token_accuracy": 0.1667509511113167,
      "num_tokens": 32650748.0,
      "step": 16700
    },
    {
      "entropy": 6.127052593231201,
      "epoch": 1.6698655470585295,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004726511994133235,
      "loss": 5.8444,
      "mean_token_accuracy": 0.16549401581287385,
      "num_tokens": 32660828.0,
      "step": 16705
    },
    {
      "entropy": 6.076432943344116,
      "epoch": 1.670365372119758,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047263414343609307,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1624479979276657,
      "num_tokens": 32670951.0,
      "step": 16710
    },
    {
      "entropy": 6.120549201965332,
      "epoch": 1.6708651971809867,
      "grad_norm": 1.8203125,
      "learning_rate": 0.00047261708248643846,
      "loss": 5.8464,
      "mean_token_accuracy": 0.16010861843824387,
      "num_tokens": 32680977.0,
      "step": 16715
    },
    {
      "entropy": 6.190199565887451,
      "epoch": 1.6713650222422152,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004726000165647888,
      "loss": 5.8372,
      "mean_token_accuracy": 0.16532304733991623,
      "num_tokens": 32690004.0,
      "step": 16720
    },
    {
      "entropy": 6.128579092025757,
      "epoch": 1.6718648473034436,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004725829456715737,
      "loss": 5.8185,
      "mean_token_accuracy": 0.16368092298507692,
      "num_tokens": 32699862.0,
      "step": 16725
    },
    {
      "entropy": 6.052625274658203,
      "epoch": 1.6723646723646723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004725658698072225,
      "loss": 5.7984,
      "mean_token_accuracy": 0.16279423832893372,
      "num_tokens": 32708732.0,
      "step": 16730
    },
    {
      "entropy": 6.120071268081665,
      "epoch": 1.672864497425901,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004725487889721648,
      "loss": 5.9455,
      "mean_token_accuracy": 0.1501711994409561,
      "num_tokens": 32719117.0,
      "step": 16735
    },
    {
      "entropy": 6.128191041946411,
      "epoch": 1.6733643224871295,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004725317031668304,
      "loss": 5.8025,
      "mean_token_accuracy": 0.1618262618780136,
      "num_tokens": 32728357.0,
      "step": 16740
    },
    {
      "entropy": 6.159316539764404,
      "epoch": 1.673864147548358,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047251461239164916,
      "loss": 5.8362,
      "mean_token_accuracy": 0.16226603239774703,
      "num_tokens": 32738249.0,
      "step": 16745
    },
    {
      "entropy": 6.007031726837158,
      "epoch": 1.6743639726095867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004724975166470512,
      "loss": 5.752,
      "mean_token_accuracy": 0.17386188358068466,
      "num_tokens": 32747880.0,
      "step": 16750
    },
    {
      "entropy": 6.0817413330078125,
      "epoch": 1.6748637976708152,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004724804159334665,
      "loss": 5.769,
      "mean_token_accuracy": 0.1614336758852005,
      "num_tokens": 32758215.0,
      "step": 16755
    },
    {
      "entropy": 6.197255706787109,
      "epoch": 1.6753636227320436,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004724633102513253,
      "loss": 5.9019,
      "mean_token_accuracy": 0.15056714415550232,
      "num_tokens": 32767426.0,
      "step": 16760
    },
    {
      "entropy": 6.071948909759522,
      "epoch": 1.6758634477932723,
      "grad_norm": 1.6640625,
      "learning_rate": 0.0004724461996010581,
      "loss": 5.7827,
      "mean_token_accuracy": 0.16054512411355973,
      "num_tokens": 32776507.0,
      "step": 16765
    },
    {
      "entropy": 6.003354883193969,
      "epoch": 1.676363272854501,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004724290839830953,
      "loss": 5.7898,
      "mean_token_accuracy": 0.17102356851100922,
      "num_tokens": 32787386.0,
      "step": 16770
    },
    {
      "entropy": 6.089016151428223,
      "epoch": 1.6768630979157295,
      "grad_norm": 1.25,
      "learning_rate": 0.0004724119633978675,
      "loss": 5.8269,
      "mean_token_accuracy": 0.1615647107362747,
      "num_tokens": 32797245.0,
      "step": 16775
    },
    {
      "entropy": 6.1575757503509525,
      "epoch": 1.677362922976958,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047239483784580554,
      "loss": 5.8628,
      "mean_token_accuracy": 0.16592977792024613,
      "num_tokens": 32807256.0,
      "step": 16780
    },
    {
      "entropy": 6.14875054359436,
      "epoch": 1.6778627480381867,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004723777073273403,
      "loss": 5.8556,
      "mean_token_accuracy": 0.16490956842899324,
      "num_tokens": 32816818.0,
      "step": 16785
    },
    {
      "entropy": 6.059776020050049,
      "epoch": 1.6783625730994152,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004723605718429026,
      "loss": 5.8898,
      "mean_token_accuracy": 0.16015442162752153,
      "num_tokens": 32826275.0,
      "step": 16790
    },
    {
      "entropy": 6.066444969177246,
      "epoch": 1.6788623981606436,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004723434313929238,
      "loss": 5.8203,
      "mean_token_accuracy": 0.16815488338470458,
      "num_tokens": 32836703.0,
      "step": 16795
    },
    {
      "entropy": 6.039139986038208,
      "epoch": 1.6793622232218723,
      "grad_norm": 1.25,
      "learning_rate": 0.00047232628597783507,
      "loss": 5.738,
      "mean_token_accuracy": 0.1705315738916397,
      "num_tokens": 32847745.0,
      "step": 16800
    },
    {
      "entropy": 6.149748468399048,
      "epoch": 1.679862048283101,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047230913559806773,
      "loss": 5.8004,
      "mean_token_accuracy": 0.169490747153759,
      "num_tokens": 32857123.0,
      "step": 16805
    },
    {
      "entropy": 6.027956295013428,
      "epoch": 1.6803618733443295,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004722919802540533,
      "loss": 5.744,
      "mean_token_accuracy": 0.168426750600338,
      "num_tokens": 32866337.0,
      "step": 16810
    },
    {
      "entropy": 6.076097917556763,
      "epoch": 1.680861698405558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004722748199462234,
      "loss": 5.8847,
      "mean_token_accuracy": 0.167574006319046,
      "num_tokens": 32877525.0,
      "step": 16815
    },
    {
      "entropy": 6.144401693344117,
      "epoch": 1.6813615234667867,
      "grad_norm": 2.125,
      "learning_rate": 0.0004722576546750098,
      "loss": 5.8454,
      "mean_token_accuracy": 0.16448385864496232,
      "num_tokens": 32888165.0,
      "step": 16820
    },
    {
      "entropy": 6.177789354324341,
      "epoch": 1.6818613485280152,
      "grad_norm": 1.9296875,
      "learning_rate": 0.00047224048444084434,
      "loss": 5.8995,
      "mean_token_accuracy": 0.1598668083548546,
      "num_tokens": 32898541.0,
      "step": 16825
    },
    {
      "entropy": 6.044449424743652,
      "epoch": 1.6823611735892436,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047222330924415906,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17765017002820968,
      "num_tokens": 32907796.0,
      "step": 16830
    },
    {
      "entropy": 6.085809135437012,
      "epoch": 1.6828609986504723,
      "grad_norm": 1.8125,
      "learning_rate": 0.0004722061290853862,
      "loss": 5.7493,
      "mean_token_accuracy": 0.1701370507478714,
      "num_tokens": 32917067.0,
      "step": 16835
    },
    {
      "entropy": 6.1288896083831785,
      "epoch": 1.683360823711701,
      "grad_norm": 1.390625,
      "learning_rate": 0.00047218894396495765,
      "loss": 5.8013,
      "mean_token_accuracy": 0.17118319720029831,
      "num_tokens": 32927039.0,
      "step": 16840
    },
    {
      "entropy": 6.09756555557251,
      "epoch": 1.6838606487729295,
      "grad_norm": 1.9609375,
      "learning_rate": 0.0004721717538833061,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16473134905099868,
      "num_tokens": 32937574.0,
      "step": 16845
    },
    {
      "entropy": 6.164473009109497,
      "epoch": 1.684360473834158,
      "grad_norm": 1.21875,
      "learning_rate": 0.000472154558840864,
      "loss": 5.8837,
      "mean_token_accuracy": 0.1601118713617325,
      "num_tokens": 32948401.0,
      "step": 16850
    },
    {
      "entropy": 6.107227993011475,
      "epoch": 1.6848602988953867,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000472137358838064,
      "loss": 5.7815,
      "mean_token_accuracy": 0.1681626707315445,
      "num_tokens": 32957359.0,
      "step": 16855
    },
    {
      "entropy": 6.053154563903808,
      "epoch": 1.6853601239566152,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004721201538753387,
      "loss": 5.7155,
      "mean_token_accuracy": 0.1769217073917389,
      "num_tokens": 32966340.0,
      "step": 16860
    },
    {
      "entropy": 6.070557594299316,
      "epoch": 1.6858599490178436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047210294395312103,
      "loss": 5.7846,
      "mean_token_accuracy": 0.1664064809679985,
      "num_tokens": 32976376.0,
      "step": 16865
    },
    {
      "entropy": 6.070236921310425,
      "epoch": 1.6863597740790723,
      "grad_norm": 1.375,
      "learning_rate": 0.000472085729071844,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1732776328921318,
      "num_tokens": 32985680.0,
      "step": 16870
    },
    {
      "entropy": 6.019136095046997,
      "epoch": 1.686859599140301,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047206850923194076,
      "loss": 5.7855,
      "mean_token_accuracy": 0.17065980285406113,
      "num_tokens": 32995557.0,
      "step": 16875
    },
    {
      "entropy": 6.122857189178466,
      "epoch": 1.6873594242015295,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047205128443384456,
      "loss": 5.7808,
      "mean_token_accuracy": 0.16405878216028214,
      "num_tokens": 33004655.0,
      "step": 16880
    },
    {
      "entropy": 6.105445384979248,
      "epoch": 1.687859249262758,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004720340546779887,
      "loss": 5.8267,
      "mean_token_accuracy": 0.16379288583993912,
      "num_tokens": 33014621.0,
      "step": 16885
    },
    {
      "entropy": 6.09962739944458,
      "epoch": 1.6883590743239867,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004720168199648068,
      "loss": 5.7648,
      "mean_token_accuracy": 0.17362350672483445,
      "num_tokens": 33024390.0,
      "step": 16890
    },
    {
      "entropy": 6.108802318572998,
      "epoch": 1.6888588993852152,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00047199958029473237,
      "loss": 5.918,
      "mean_token_accuracy": 0.16664462685585021,
      "num_tokens": 33035351.0,
      "step": 16895
    },
    {
      "entropy": 6.053422307968139,
      "epoch": 1.6893587244464436,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047198233566819915,
      "loss": 5.6958,
      "mean_token_accuracy": 0.1765718713402748,
      "num_tokens": 33044148.0,
      "step": 16900
    },
    {
      "entropy": 6.077435779571533,
      "epoch": 1.6898585495076723,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004719650860856411,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16806502342224122,
      "num_tokens": 33053772.0,
      "step": 16905
    },
    {
      "entropy": 6.152287483215332,
      "epoch": 1.690358374568901,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047194783154749217,
      "loss": 5.8031,
      "mean_token_accuracy": 0.17019152045249938,
      "num_tokens": 33064422.0,
      "step": 16910
    },
    {
      "entropy": 6.167243528366089,
      "epoch": 1.6908581996301293,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004719305720541865,
      "loss": 5.9626,
      "mean_token_accuracy": 0.15155426114797593,
      "num_tokens": 33075235.0,
      "step": 16915
    },
    {
      "entropy": 6.083862066268921,
      "epoch": 1.691358024691358,
      "grad_norm": 1.359375,
      "learning_rate": 0.00047191330760615823,
      "loss": 5.8079,
      "mean_token_accuracy": 0.16891692876815795,
      "num_tokens": 33085489.0,
      "step": 16920
    },
    {
      "entropy": 6.070099592208862,
      "epoch": 1.6918578497525867,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047189603820384186,
      "loss": 5.8438,
      "mean_token_accuracy": 0.16291453540325165,
      "num_tokens": 33094932.0,
      "step": 16925
    },
    {
      "entropy": 6.048491668701172,
      "epoch": 1.6923576748138152,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047187876384767187,
      "loss": 5.7728,
      "mean_token_accuracy": 0.1714451014995575,
      "num_tokens": 33104224.0,
      "step": 16930
    },
    {
      "entropy": 6.092360401153565,
      "epoch": 1.6928574998750436,
      "grad_norm": 1.203125,
      "learning_rate": 0.00047186148453808274,
      "loss": 5.798,
      "mean_token_accuracy": 0.16460594683885574,
      "num_tokens": 33114039.0,
      "step": 16935
    },
    {
      "entropy": 6.109842920303345,
      "epoch": 1.6933573249362723,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004718442002755094,
      "loss": 5.832,
      "mean_token_accuracy": 0.16622547954320907,
      "num_tokens": 33123274.0,
      "step": 16940
    },
    {
      "entropy": 6.113402414321899,
      "epoch": 1.693857149997501,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004718269110603866,
      "loss": 5.839,
      "mean_token_accuracy": 0.16477811634540557,
      "num_tokens": 33134259.0,
      "step": 16945
    },
    {
      "entropy": 6.131065893173218,
      "epoch": 1.6943569750587293,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047180961689314937,
      "loss": 5.7975,
      "mean_token_accuracy": 0.16725666373968123,
      "num_tokens": 33144662.0,
      "step": 16950
    },
    {
      "entropy": 6.084068727493286,
      "epoch": 1.694856800119958,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004717923177742329,
      "loss": 5.8005,
      "mean_token_accuracy": 0.16392777413129805,
      "num_tokens": 33154074.0,
      "step": 16955
    },
    {
      "entropy": 6.058197975158691,
      "epoch": 1.6953566251811867,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047177501370407223,
      "loss": 5.7888,
      "mean_token_accuracy": 0.16668004542589188,
      "num_tokens": 33163973.0,
      "step": 16960
    },
    {
      "entropy": 6.122508478164673,
      "epoch": 1.6958564502424152,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047175770468310294,
      "loss": 5.8824,
      "mean_token_accuracy": 0.15859579145908356,
      "num_tokens": 33173607.0,
      "step": 16965
    },
    {
      "entropy": 6.091284894943238,
      "epoch": 1.6963562753036436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047174039071176037,
      "loss": 5.739,
      "mean_token_accuracy": 0.1705292895436287,
      "num_tokens": 33183104.0,
      "step": 16970
    },
    {
      "entropy": 6.096176481246948,
      "epoch": 1.6968561003648723,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047172307179048025,
      "loss": 5.7554,
      "mean_token_accuracy": 0.16481644958257674,
      "num_tokens": 33191066.0,
      "step": 16975
    },
    {
      "entropy": 6.067199325561523,
      "epoch": 1.697355925426101,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004717057479196982,
      "loss": 5.806,
      "mean_token_accuracy": 0.16632929742336272,
      "num_tokens": 33201499.0,
      "step": 16980
    },
    {
      "entropy": 6.027872371673584,
      "epoch": 1.6978557504873293,
      "grad_norm": 1.125,
      "learning_rate": 0.0004716884190998502,
      "loss": 5.7427,
      "mean_token_accuracy": 0.1696314424276352,
      "num_tokens": 33211989.0,
      "step": 16985
    },
    {
      "entropy": 6.147678279876709,
      "epoch": 1.698355575548558,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004716710853313721,
      "loss": 5.8695,
      "mean_token_accuracy": 0.1669941633939743,
      "num_tokens": 33221715.0,
      "step": 16990
    },
    {
      "entropy": 6.126942300796509,
      "epoch": 1.6988554006097867,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047165374661470023,
      "loss": 5.8463,
      "mean_token_accuracy": 0.15994910150766373,
      "num_tokens": 33232812.0,
      "step": 16995
    },
    {
      "entropy": 6.171700191497803,
      "epoch": 1.6993552256710152,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047163640295027064,
      "loss": 5.7589,
      "mean_token_accuracy": 0.16988738775253295,
      "num_tokens": 33242147.0,
      "step": 17000
    },
    {
      "entropy": 6.075820827484131,
      "epoch": 1.6998550507322436,
      "grad_norm": 1.25,
      "learning_rate": 0.00047161905433851974,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16917648911476135,
      "num_tokens": 33251433.0,
      "step": 17005
    },
    {
      "entropy": 6.049834823608398,
      "epoch": 1.7003548757934723,
      "grad_norm": 1.125,
      "learning_rate": 0.00047160170077988403,
      "loss": 5.7222,
      "mean_token_accuracy": 0.16617857962846755,
      "num_tokens": 33261338.0,
      "step": 17010
    },
    {
      "entropy": 6.071846532821655,
      "epoch": 1.7008547008547008,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004715843422748001,
      "loss": 5.8493,
      "mean_token_accuracy": 0.16450188010931016,
      "num_tokens": 33271312.0,
      "step": 17015
    },
    {
      "entropy": 6.1309826374053955,
      "epoch": 1.7013545259159293,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047156697882370474,
      "loss": 5.8194,
      "mean_token_accuracy": 0.16155257672071457,
      "num_tokens": 33280525.0,
      "step": 17020
    },
    {
      "entropy": 6.080914115905761,
      "epoch": 1.701854350977158,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004715496104270348,
      "loss": 5.8035,
      "mean_token_accuracy": 0.17069359570741655,
      "num_tokens": 33289968.0,
      "step": 17025
    },
    {
      "entropy": 6.017856454849243,
      "epoch": 1.7023541760383867,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004715322370852272,
      "loss": 5.8047,
      "mean_token_accuracy": 0.1706988990306854,
      "num_tokens": 33299262.0,
      "step": 17030
    },
    {
      "entropy": 6.0879889011383055,
      "epoch": 1.7028540010996152,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00047151485879871907,
      "loss": 5.812,
      "mean_token_accuracy": 0.16554296612739564,
      "num_tokens": 33309433.0,
      "step": 17035
    },
    {
      "entropy": 6.059764623641968,
      "epoch": 1.7033538261608436,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047149747556794763,
      "loss": 5.7409,
      "mean_token_accuracy": 0.16620809584856033,
      "num_tokens": 33318587.0,
      "step": 17040
    },
    {
      "entropy": 6.047638750076294,
      "epoch": 1.7038536512220723,
      "grad_norm": 1.265625,
      "learning_rate": 0.00047148008739335035,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16197993457317353,
      "num_tokens": 33327487.0,
      "step": 17045
    },
    {
      "entropy": 6.10042371749878,
      "epoch": 1.7043534762833008,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004714626942753646,
      "loss": 5.8804,
      "mean_token_accuracy": 0.1574838176369667,
      "num_tokens": 33337151.0,
      "step": 17050
    },
    {
      "entropy": 6.182776737213135,
      "epoch": 1.7048533013445293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000471445296214428,
      "loss": 5.7532,
      "mean_token_accuracy": 0.1659430220723152,
      "num_tokens": 33346547.0,
      "step": 17055
    },
    {
      "entropy": 6.05530195236206,
      "epoch": 1.705353126405758,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047142789321097834,
      "loss": 5.7753,
      "mean_token_accuracy": 0.16415545344352722,
      "num_tokens": 33355277.0,
      "step": 17060
    },
    {
      "entropy": 6.0681627750396725,
      "epoch": 1.7058529514669867,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004714104852654534,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1717139959335327,
      "num_tokens": 33364340.0,
      "step": 17065
    },
    {
      "entropy": 6.117710971832276,
      "epoch": 1.7063527765282152,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004713930723782912,
      "loss": 5.7856,
      "mean_token_accuracy": 0.1688816323876381,
      "num_tokens": 33374302.0,
      "step": 17070
    },
    {
      "entropy": 6.074854707717895,
      "epoch": 1.7068526015894436,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047137565454992983,
      "loss": 5.8634,
      "mean_token_accuracy": 0.16269135922193528,
      "num_tokens": 33384098.0,
      "step": 17075
    },
    {
      "entropy": 6.051084852218628,
      "epoch": 1.7073524266506723,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00047135823178080745,
      "loss": 5.7422,
      "mean_token_accuracy": 0.16914181411266327,
      "num_tokens": 33393032.0,
      "step": 17080
    },
    {
      "entropy": 5.997193670272827,
      "epoch": 1.7078522517119008,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004713408040713626,
      "loss": 5.7375,
      "mean_token_accuracy": 0.16791051626205444,
      "num_tokens": 33402533.0,
      "step": 17085
    },
    {
      "entropy": 6.049596071243286,
      "epoch": 1.7083520767731293,
      "grad_norm": 1.171875,
      "learning_rate": 0.00047132337142203365,
      "loss": 5.8346,
      "mean_token_accuracy": 0.16434696763753892,
      "num_tokens": 33412826.0,
      "step": 17090
    },
    {
      "entropy": 6.127618503570557,
      "epoch": 1.708851901834358,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004713059338332591,
      "loss": 5.8077,
      "mean_token_accuracy": 0.16659756898880004,
      "num_tokens": 33422485.0,
      "step": 17095
    },
    {
      "entropy": 6.10188570022583,
      "epoch": 1.7093517268955867,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047128849130547774,
      "loss": 5.6968,
      "mean_token_accuracy": 0.17069557309150696,
      "num_tokens": 33432575.0,
      "step": 17100
    },
    {
      "entropy": 6.041983032226563,
      "epoch": 1.7098515519568152,
      "grad_norm": 1.59375,
      "learning_rate": 0.00047127104383912856,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16690985560417176,
      "num_tokens": 33442893.0,
      "step": 17105
    },
    {
      "entropy": 6.048993539810181,
      "epoch": 1.7103513770180436,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004712535914346504,
      "loss": 5.8862,
      "mean_token_accuracy": 0.16201318055391312,
      "num_tokens": 33452080.0,
      "step": 17110
    },
    {
      "entropy": 6.079095411300659,
      "epoch": 1.7108512020792723,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047123613409248233,
      "loss": 5.7377,
      "mean_token_accuracy": 0.16574988067150115,
      "num_tokens": 33461721.0,
      "step": 17115
    },
    {
      "entropy": 6.179989290237427,
      "epoch": 1.7113510271405008,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047121867181306365,
      "loss": 5.8546,
      "mean_token_accuracy": 0.15359147489070893,
      "num_tokens": 33470731.0,
      "step": 17120
    },
    {
      "entropy": 6.10111403465271,
      "epoch": 1.7118508522017293,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004712012045968337,
      "loss": 5.8246,
      "mean_token_accuracy": 0.16825323402881623,
      "num_tokens": 33481000.0,
      "step": 17125
    },
    {
      "entropy": 6.023765420913696,
      "epoch": 1.712350677262958,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047118373244423197,
      "loss": 5.803,
      "mean_token_accuracy": 0.1672479271888733,
      "num_tokens": 33489535.0,
      "step": 17130
    },
    {
      "entropy": 6.099935293197632,
      "epoch": 1.7128505023241867,
      "grad_norm": 1.328125,
      "learning_rate": 0.00047116625535569787,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16362534165382386,
      "num_tokens": 33499746.0,
      "step": 17135
    },
    {
      "entropy": 6.135937976837158,
      "epoch": 1.7133503273854152,
      "grad_norm": 1.546875,
      "learning_rate": 0.00047114877333167133,
      "loss": 5.7979,
      "mean_token_accuracy": 0.17299656420946122,
      "num_tokens": 33508526.0,
      "step": 17140
    },
    {
      "entropy": 6.1000889301300045,
      "epoch": 1.7138501524466436,
      "grad_norm": 1.453125,
      "learning_rate": 0.00047113128637259216,
      "loss": 5.7867,
      "mean_token_accuracy": 0.16638194173574447,
      "num_tokens": 33518675.0,
      "step": 17145
    },
    {
      "entropy": 6.118789005279541,
      "epoch": 1.7143499775078723,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004711137944789003,
      "loss": 5.8749,
      "mean_token_accuracy": 0.1637983128428459,
      "num_tokens": 33528132.0,
      "step": 17150
    },
    {
      "entropy": 6.059586048126221,
      "epoch": 1.7148498025691008,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047109629765103577,
      "loss": 5.7609,
      "mean_token_accuracy": 0.1721641793847084,
      "num_tokens": 33537259.0,
      "step": 17155
    },
    {
      "entropy": 5.99711127281189,
      "epoch": 1.7153496276303293,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047107879588943887,
      "loss": 5.7641,
      "mean_token_accuracy": 0.17076666802167892,
      "num_tokens": 33546189.0,
      "step": 17160
    },
    {
      "entropy": 6.14252724647522,
      "epoch": 1.715849452691558,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047106128919455,
      "loss": 5.8826,
      "mean_token_accuracy": 0.15790692418813707,
      "num_tokens": 33555392.0,
      "step": 17165
    },
    {
      "entropy": 6.164712142944336,
      "epoch": 1.7163492777527867,
      "grad_norm": 1.703125,
      "learning_rate": 0.0004710437775668094,
      "loss": 5.8828,
      "mean_token_accuracy": 0.1638158604502678,
      "num_tokens": 33564766.0,
      "step": 17170
    },
    {
      "entropy": 6.089115571975708,
      "epoch": 1.7168491028140151,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004710262610066579,
      "loss": 5.879,
      "mean_token_accuracy": 0.16215286105871202,
      "num_tokens": 33574680.0,
      "step": 17175
    },
    {
      "entropy": 6.128041410446167,
      "epoch": 1.7173489278752436,
      "grad_norm": 1.25,
      "learning_rate": 0.00047100873951453617,
      "loss": 5.809,
      "mean_token_accuracy": 0.16774974912405013,
      "num_tokens": 33583068.0,
      "step": 17180
    },
    {
      "entropy": 6.102362394332886,
      "epoch": 1.7178487529364723,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00047099121309088487,
      "loss": 5.805,
      "mean_token_accuracy": 0.1657631665468216,
      "num_tokens": 33592838.0,
      "step": 17185
    },
    {
      "entropy": 6.097935724258423,
      "epoch": 1.7183485779977008,
      "grad_norm": 1.34375,
      "learning_rate": 0.00047097368173614515,
      "loss": 5.8124,
      "mean_token_accuracy": 0.1675238326191902,
      "num_tokens": 33603627.0,
      "step": 17190
    },
    {
      "entropy": 6.06316442489624,
      "epoch": 1.7188484030589293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00047095614545075807,
      "loss": 5.8475,
      "mean_token_accuracy": 0.16252281665802001,
      "num_tokens": 33614604.0,
      "step": 17195
    },
    {
      "entropy": 6.076611614227295,
      "epoch": 1.719348228120158,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004709386042351647,
      "loss": 5.7443,
      "mean_token_accuracy": 0.1678536355495453,
      "num_tokens": 33624604.0,
      "step": 17200
    },
    {
      "entropy": 6.209040832519531,
      "epoch": 1.7198480531813867,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004709210580898065,
      "loss": 5.8582,
      "mean_token_accuracy": 0.15945647582411765,
      "num_tokens": 33634842.0,
      "step": 17205
    },
    {
      "entropy": 6.187971305847168,
      "epoch": 1.720347878242615,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004709035070151249,
      "loss": 5.8775,
      "mean_token_accuracy": 0.15866751596331596,
      "num_tokens": 33644829.0,
      "step": 17210
    },
    {
      "entropy": 6.138267183303833,
      "epoch": 1.7208477033038436,
      "grad_norm": 1.375,
      "learning_rate": 0.0004708859510115614,
      "loss": 5.9231,
      "mean_token_accuracy": 0.15832079946994781,
      "num_tokens": 33653952.0,
      "step": 17215
    },
    {
      "entropy": 6.106216812133789,
      "epoch": 1.7213475283650723,
      "grad_norm": 2.109375,
      "learning_rate": 0.0004708683900795579,
      "loss": 5.926,
      "mean_token_accuracy": 0.16010939031839372,
      "num_tokens": 33664014.0,
      "step": 17220
    },
    {
      "entropy": 6.128833389282226,
      "epoch": 1.7218473534263008,
      "grad_norm": 1.15625,
      "learning_rate": 0.00047085082421955603,
      "loss": 5.7987,
      "mean_token_accuracy": 0.16811731308698655,
      "num_tokens": 33674575.0,
      "step": 17225
    },
    {
      "entropy": 6.087554693222046,
      "epoch": 1.7223471784875293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047083325343199786,
      "loss": 5.8394,
      "mean_token_accuracy": 0.16705252528190612,
      "num_tokens": 33684431.0,
      "step": 17230
    },
    {
      "entropy": 6.155085802078247,
      "epoch": 1.722847003548758,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004708156777173254,
      "loss": 5.8851,
      "mean_token_accuracy": 0.15852571576833724,
      "num_tokens": 33693502.0,
      "step": 17235
    },
    {
      "entropy": 6.081858777999878,
      "epoch": 1.7233468286099867,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004707980970759809,
      "loss": 5.8556,
      "mean_token_accuracy": 0.1609411358833313,
      "num_tokens": 33703034.0,
      "step": 17240
    },
    {
      "entropy": 6.094609165191651,
      "epoch": 1.723846653671215,
      "grad_norm": 1.28125,
      "learning_rate": 0.00047078051150840666,
      "loss": 5.8141,
      "mean_token_accuracy": 0.16953291669487952,
      "num_tokens": 33712429.0,
      "step": 17245
    },
    {
      "entropy": 6.093194818496704,
      "epoch": 1.7243464787324436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004707629210150451,
      "loss": 5.8713,
      "mean_token_accuracy": 0.16456263065338134,
      "num_tokens": 33724254.0,
      "step": 17250
    },
    {
      "entropy": 5.949827098846436,
      "epoch": 1.7248463037936723,
      "grad_norm": 1.65625,
      "learning_rate": 0.0004707453255963388,
      "loss": 5.5892,
      "mean_token_accuracy": 0.1814504086971283,
      "num_tokens": 33735706.0,
      "step": 17255
    },
    {
      "entropy": 6.118174266815186,
      "epoch": 1.7253461288549008,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00047072772525273053,
      "loss": 5.8351,
      "mean_token_accuracy": 0.16584513038396836,
      "num_tokens": 33745683.0,
      "step": 17260
    },
    {
      "entropy": 6.1469803810119625,
      "epoch": 1.7258459539161293,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00047071011998466304,
      "loss": 5.8531,
      "mean_token_accuracy": 0.15809971541166307,
      "num_tokens": 33756776.0,
      "step": 17265
    },
    {
      "entropy": 6.142675971984863,
      "epoch": 1.726345778977358,
      "grad_norm": 1.25,
      "learning_rate": 0.00047069250979257934,
      "loss": 5.741,
      "mean_token_accuracy": 0.1699790582060814,
      "num_tokens": 33766004.0,
      "step": 17270
    },
    {
      "entropy": 6.140097332000733,
      "epoch": 1.7268456040385864,
      "grad_norm": 1.375,
      "learning_rate": 0.00047067489467692236,
      "loss": 5.9064,
      "mean_token_accuracy": 0.15822670310735704,
      "num_tokens": 33774709.0,
      "step": 17275
    },
    {
      "entropy": 6.192679452896118,
      "epoch": 1.727345429099815,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004706572746381354,
      "loss": 5.8593,
      "mean_token_accuracy": 0.1668548360466957,
      "num_tokens": 33785032.0,
      "step": 17280
    },
    {
      "entropy": 6.110248613357544,
      "epoch": 1.7278452541610436,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00047063964967666177,
      "loss": 5.7834,
      "mean_token_accuracy": 0.16075980588793753,
      "num_tokens": 33795320.0,
      "step": 17285
    },
    {
      "entropy": 6.150143146514893,
      "epoch": 1.7283450792222723,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004706220197929449,
      "loss": 5.9064,
      "mean_token_accuracy": 0.16197581589221954,
      "num_tokens": 33805764.0,
      "step": 17290
    },
    {
      "entropy": 6.121720314025879,
      "epoch": 1.7288449042835008,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004706043849874284,
      "loss": 5.8775,
      "mean_token_accuracy": 0.15801021754741668,
      "num_tokens": 33816045.0,
      "step": 17295
    },
    {
      "entropy": 6.1265195369720455,
      "epoch": 1.7293447293447293,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047058674526055585,
      "loss": 5.8246,
      "mean_token_accuracy": 0.16347893178462983,
      "num_tokens": 33825192.0,
      "step": 17300
    },
    {
      "entropy": 6.10821213722229,
      "epoch": 1.729844554405958,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047056910061277115,
      "loss": 5.8292,
      "mean_token_accuracy": 0.1663743495941162,
      "num_tokens": 33835034.0,
      "step": 17305
    },
    {
      "entropy": 6.131993627548217,
      "epoch": 1.7303443794671864,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004705514510445182,
      "loss": 5.8692,
      "mean_token_accuracy": 0.15888613164424897,
      "num_tokens": 33845225.0,
      "step": 17310
    },
    {
      "entropy": 6.114203786849975,
      "epoch": 1.730844204528415,
      "grad_norm": 1.3984375,
      "learning_rate": 0.000470533796556241,
      "loss": 5.829,
      "mean_token_accuracy": 0.16580835953354836,
      "num_tokens": 33855303.0,
      "step": 17315
    },
    {
      "entropy": 6.120185613632202,
      "epoch": 1.7313440295896436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004705161371483839,
      "loss": 5.8538,
      "mean_token_accuracy": 0.16225065439939498,
      "num_tokens": 33865336.0,
      "step": 17320
    },
    {
      "entropy": 6.173816585540772,
      "epoch": 1.7318438546508723,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00047049847282139095,
      "loss": 5.8786,
      "mean_token_accuracy": 0.15986034870147706,
      "num_tokens": 33875707.0,
      "step": 17325
    },
    {
      "entropy": 6.184801006317139,
      "epoch": 1.7323436797121008,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047048080357570685,
      "loss": 5.9423,
      "mean_token_accuracy": 0.16085763573646544,
      "num_tokens": 33886649.0,
      "step": 17330
    },
    {
      "entropy": 6.129288959503174,
      "epoch": 1.7328435047733293,
      "grad_norm": 1.3125,
      "learning_rate": 0.000470463129411776,
      "loss": 5.8659,
      "mean_token_accuracy": 0.15990834236145018,
      "num_tokens": 33896815.0,
      "step": 17335
    },
    {
      "entropy": 6.103807210922241,
      "epoch": 1.733343329834558,
      "grad_norm": 1.25,
      "learning_rate": 0.0004704454503300431,
      "loss": 5.837,
      "mean_token_accuracy": 0.16209650188684463,
      "num_tokens": 33907481.0,
      "step": 17340
    },
    {
      "entropy": 6.137941980361939,
      "epoch": 1.7338431548957864,
      "grad_norm": 1.5625,
      "learning_rate": 0.000470427766330953,
      "loss": 5.8189,
      "mean_token_accuracy": 0.16745028048753738,
      "num_tokens": 33917011.0,
      "step": 17345
    },
    {
      "entropy": 5.918174743652344,
      "epoch": 1.734342979957015,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004704100774149506,
      "loss": 5.649,
      "mean_token_accuracy": 0.1769773095846176,
      "num_tokens": 33926528.0,
      "step": 17350
    },
    {
      "entropy": 6.08259596824646,
      "epoch": 1.7348428050182436,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00047039238358248086,
      "loss": 5.8913,
      "mean_token_accuracy": 0.1613145127892494,
      "num_tokens": 33936207.0,
      "step": 17355
    },
    {
      "entropy": 6.063930368423462,
      "epoch": 1.7353426300794723,
      "grad_norm": 1.421875,
      "learning_rate": 0.000470374684833989,
      "loss": 5.8709,
      "mean_token_accuracy": 0.1652979999780655,
      "num_tokens": 33946632.0,
      "step": 17360
    },
    {
      "entropy": 6.165855836868286,
      "epoch": 1.7358424551407008,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004703569811699205,
      "loss": 5.87,
      "mean_token_accuracy": 0.16124885827302932,
      "num_tokens": 33956469.0,
      "step": 17365
    },
    {
      "entropy": 6.159896421432495,
      "epoch": 1.7363422802019293,
      "grad_norm": 1.234375,
      "learning_rate": 0.00047033927259072046,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16769057959318162,
      "num_tokens": 33965699.0,
      "step": 17370
    },
    {
      "entropy": 6.063975095748901,
      "epoch": 1.736842105263158,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00047032155909683464,
      "loss": 5.7511,
      "mean_token_accuracy": 0.16884975731372834,
      "num_tokens": 33975108.0,
      "step": 17375
    },
    {
      "entropy": 6.140924406051636,
      "epoch": 1.7373419303243864,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004703038406887087,
      "loss": 5.9143,
      "mean_token_accuracy": 0.15361512675881386,
      "num_tokens": 33986321.0,
      "step": 17380
    },
    {
      "entropy": 6.215884828567505,
      "epoch": 1.737841755385615,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047028611736678837,
      "loss": 5.902,
      "mean_token_accuracy": 0.1550668701529503,
      "num_tokens": 33996038.0,
      "step": 17385
    },
    {
      "entropy": 6.079052972793579,
      "epoch": 1.7383415804468436,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004702683891315196,
      "loss": 5.7881,
      "mean_token_accuracy": 0.1749996766448021,
      "num_tokens": 34004773.0,
      "step": 17390
    },
    {
      "entropy": 6.098669767379761,
      "epoch": 1.7388414055080723,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00047025065598334834,
      "loss": 5.8275,
      "mean_token_accuracy": 0.16623601466417312,
      "num_tokens": 34014249.0,
      "step": 17395
    },
    {
      "entropy": 6.146162128448486,
      "epoch": 1.7393412305693008,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004702329179227208,
      "loss": 5.9293,
      "mean_token_accuracy": 0.15506404191255568,
      "num_tokens": 34025553.0,
      "step": 17400
    },
    {
      "entropy": 6.120491933822632,
      "epoch": 1.7398410556305293,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047021517495008335,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16255650371313096,
      "num_tokens": 34036536.0,
      "step": 17405
    },
    {
      "entropy": 6.135387849807739,
      "epoch": 1.740340880691758,
      "grad_norm": 1.21875,
      "learning_rate": 0.00047019742706588233,
      "loss": 5.8566,
      "mean_token_accuracy": 0.160465969145298,
      "num_tokens": 34046353.0,
      "step": 17410
    },
    {
      "entropy": 6.155795001983643,
      "epoch": 1.7408407057529864,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004701796742705642,
      "loss": 5.7585,
      "mean_token_accuracy": 0.1741766318678856,
      "num_tokens": 34056172.0,
      "step": 17415
    },
    {
      "entropy": 6.104709577560425,
      "epoch": 1.741340530814215,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047016191656457573,
      "loss": 5.7558,
      "mean_token_accuracy": 0.16570465713739396,
      "num_tokens": 34066313.0,
      "step": 17420
    },
    {
      "entropy": 6.090389537811279,
      "epoch": 1.7418403558754436,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00047014415394836374,
      "loss": 5.7737,
      "mean_token_accuracy": 0.16618573516607285,
      "num_tokens": 34075594.0,
      "step": 17425
    },
    {
      "entropy": 6.071141147613526,
      "epoch": 1.7423401809366723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004701263864223749,
      "loss": 5.8024,
      "mean_token_accuracy": 0.1702573597431183,
      "num_tokens": 34085619.0,
      "step": 17430
    },
    {
      "entropy": 6.121971178054809,
      "epoch": 1.7428400059979008,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004701086139870565,
      "loss": 5.8134,
      "mean_token_accuracy": 0.1667136251926422,
      "num_tokens": 34094626.0,
      "step": 17435
    },
    {
      "entropy": 6.129219961166382,
      "epoch": 1.7433398310591293,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004700908366428556,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16450730562210084,
      "num_tokens": 34104950.0,
      "step": 17440
    },
    {
      "entropy": 6.206182146072388,
      "epoch": 1.743839656120358,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004700730543902194,
      "loss": 5.9536,
      "mean_token_accuracy": 0.15412802696228028,
      "num_tokens": 34114407.0,
      "step": 17445
    },
    {
      "entropy": 6.203389883041382,
      "epoch": 1.7443394811815864,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00047005526722959537,
      "loss": 5.8689,
      "mean_token_accuracy": 0.15363965630531312,
      "num_tokens": 34124842.0,
      "step": 17450
    },
    {
      "entropy": 6.147044563293457,
      "epoch": 1.744839306242815,
      "grad_norm": 1.296875,
      "learning_rate": 0.00047003747516143095,
      "loss": 5.7374,
      "mean_token_accuracy": 0.16998422741889954,
      "num_tokens": 34134265.0,
      "step": 17455
    },
    {
      "entropy": 6.167797803878784,
      "epoch": 1.7453391313040436,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00047001967818617385,
      "loss": 5.8827,
      "mean_token_accuracy": 0.15722391605377198,
      "num_tokens": 34143915.0,
      "step": 17460
    },
    {
      "entropy": 6.110768032073975,
      "epoch": 1.7458389563652723,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004700018763042719,
      "loss": 5.8518,
      "mean_token_accuracy": 0.16741512715816498,
      "num_tokens": 34154869.0,
      "step": 17465
    },
    {
      "entropy": 6.035772609710693,
      "epoch": 1.7463387814265006,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004699840695161728,
      "loss": 5.728,
      "mean_token_accuracy": 0.1715422511100769,
      "num_tokens": 34164494.0,
      "step": 17470
    },
    {
      "entropy": 6.122415018081665,
      "epoch": 1.7468386064877293,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004699662578223248,
      "loss": 5.8095,
      "mean_token_accuracy": 0.16024689227342606,
      "num_tokens": 34173986.0,
      "step": 17475
    },
    {
      "entropy": 6.200676488876343,
      "epoch": 1.747338431548958,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046994844122317587,
      "loss": 6.0051,
      "mean_token_accuracy": 0.15448372662067414,
      "num_tokens": 34183129.0,
      "step": 17480
    },
    {
      "entropy": 6.051246452331543,
      "epoch": 1.7478382566101864,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046993061971917433,
      "loss": 5.7499,
      "mean_token_accuracy": 0.16709147840738298,
      "num_tokens": 34192559.0,
      "step": 17485
    },
    {
      "entropy": 6.142779779434204,
      "epoch": 1.748338081671415,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046991279331076865,
      "loss": 5.8654,
      "mean_token_accuracy": 0.15936491936445235,
      "num_tokens": 34202599.0,
      "step": 17490
    },
    {
      "entropy": 6.144532918930054,
      "epoch": 1.7488379067326436,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004698949619984071,
      "loss": 5.9022,
      "mean_token_accuracy": 0.1548503190279007,
      "num_tokens": 34212883.0,
      "step": 17495
    },
    {
      "entropy": 6.110394287109375,
      "epoch": 1.7493377317938723,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004698771257825385,
      "loss": 5.8026,
      "mean_token_accuracy": 0.17068080604076385,
      "num_tokens": 34222019.0,
      "step": 17500
    },
    {
      "entropy": 6.060295724868775,
      "epoch": 1.7498375568551006,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004698592846636116,
      "loss": 5.7364,
      "mean_token_accuracy": 0.17127123475074768,
      "num_tokens": 34231817.0,
      "step": 17505
    },
    {
      "entropy": 6.148530435562134,
      "epoch": 1.7503373819163293,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004698414386420752,
      "loss": 5.8636,
      "mean_token_accuracy": 0.15936776995658875,
      "num_tokens": 34242035.0,
      "step": 17510
    },
    {
      "entropy": 6.110805654525757,
      "epoch": 1.750837206977558,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004698235877183784,
      "loss": 5.8102,
      "mean_token_accuracy": 0.16732075661420823,
      "num_tokens": 34251728.0,
      "step": 17515
    },
    {
      "entropy": 6.0559672832489015,
      "epoch": 1.7513370320387864,
      "grad_norm": 1.828125,
      "learning_rate": 0.00046980573189297017,
      "loss": 5.8425,
      "mean_token_accuracy": 0.1604541838169098,
      "num_tokens": 34262036.0,
      "step": 17520
    },
    {
      "entropy": 6.1391417503356935,
      "epoch": 1.751836857100015,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004697878711662999,
      "loss": 5.8544,
      "mean_token_accuracy": 0.16002709716558455,
      "num_tokens": 34272335.0,
      "step": 17525
    },
    {
      "entropy": 6.163509321212769,
      "epoch": 1.7523366821612436,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046977000553881696,
      "loss": 5.831,
      "mean_token_accuracy": 0.15586317479610443,
      "num_tokens": 34282931.0,
      "step": 17530
    },
    {
      "entropy": 6.143906164169311,
      "epoch": 1.752836507222472,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046975213501097076,
      "loss": 5.7887,
      "mean_token_accuracy": 0.17016445100307465,
      "num_tokens": 34293716.0,
      "step": 17535
    },
    {
      "entropy": 6.161886692047119,
      "epoch": 1.7533363322837006,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004697342595832109,
      "loss": 5.9131,
      "mean_token_accuracy": 0.1552520751953125,
      "num_tokens": 34303411.0,
      "step": 17540
    },
    {
      "entropy": 6.221400785446167,
      "epoch": 1.7538361573449293,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046971637925598725,
      "loss": 5.8225,
      "mean_token_accuracy": 0.1632259339094162,
      "num_tokens": 34313092.0,
      "step": 17545
    },
    {
      "entropy": 6.123753547668457,
      "epoch": 1.754335982406158,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004696984940297496,
      "loss": 5.7545,
      "mean_token_accuracy": 0.16910040527582168,
      "num_tokens": 34321909.0,
      "step": 17550
    },
    {
      "entropy": 6.154235935211181,
      "epoch": 1.7548358074673864,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000469680603904948,
      "loss": 5.9011,
      "mean_token_accuracy": 0.16492167115211487,
      "num_tokens": 34330213.0,
      "step": 17555
    },
    {
      "entropy": 6.157170152664184,
      "epoch": 1.755335632528615,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046966270888203237,
      "loss": 5.8303,
      "mean_token_accuracy": 0.16324958801269532,
      "num_tokens": 34339731.0,
      "step": 17560
    },
    {
      "entropy": 6.066625118255615,
      "epoch": 1.7558354575898436,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004696448089614532,
      "loss": 5.7503,
      "mean_token_accuracy": 0.16717460006475449,
      "num_tokens": 34348620.0,
      "step": 17565
    },
    {
      "entropy": 6.0737700939178465,
      "epoch": 1.756335282651072,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004696269041436606,
      "loss": 5.7443,
      "mean_token_accuracy": 0.16805184334516526,
      "num_tokens": 34358292.0,
      "step": 17570
    },
    {
      "entropy": 6.123239040374756,
      "epoch": 1.7568351077123006,
      "grad_norm": 1.25,
      "learning_rate": 0.0004696089944291052,
      "loss": 5.8677,
      "mean_token_accuracy": 0.16359450817108154,
      "num_tokens": 34367379.0,
      "step": 17575
    },
    {
      "entropy": 6.216329050064087,
      "epoch": 1.7573349327735293,
      "grad_norm": 1.25,
      "learning_rate": 0.0004695910798182377,
      "loss": 5.8701,
      "mean_token_accuracy": 0.15662339627742766,
      "num_tokens": 34376422.0,
      "step": 17580
    },
    {
      "entropy": 6.1055644989013675,
      "epoch": 1.757834757834758,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004695731603115086,
      "loss": 5.9132,
      "mean_token_accuracy": 0.16396692991256714,
      "num_tokens": 34386509.0,
      "step": 17585
    },
    {
      "entropy": 6.12147479057312,
      "epoch": 1.7583345828959864,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046955523590936884,
      "loss": 5.9044,
      "mean_token_accuracy": 0.15868469327688217,
      "num_tokens": 34396070.0,
      "step": 17590
    },
    {
      "entropy": 6.158386850357056,
      "epoch": 1.758834407957215,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004695373066122694,
      "loss": 5.8085,
      "mean_token_accuracy": 0.16521250903606416,
      "num_tokens": 34405707.0,
      "step": 17595
    },
    {
      "entropy": 6.11987042427063,
      "epoch": 1.7593342330184436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004695193724206615,
      "loss": 5.8118,
      "mean_token_accuracy": 0.16511878073215486,
      "num_tokens": 34415621.0,
      "step": 17600
    },
    {
      "entropy": 6.160628461837769,
      "epoch": 1.759834058079672,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046950143333499613,
      "loss": 5.8717,
      "mean_token_accuracy": 0.16161109805107116,
      "num_tokens": 34425115.0,
      "step": 17605
    },
    {
      "entropy": 6.172643852233887,
      "epoch": 1.7603338831409006,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046948348935572476,
      "loss": 5.8498,
      "mean_token_accuracy": 0.1610669359564781,
      "num_tokens": 34434164.0,
      "step": 17610
    },
    {
      "entropy": 6.016178941726684,
      "epoch": 1.7608337082021293,
      "grad_norm": 1.125,
      "learning_rate": 0.0004694655404832988,
      "loss": 5.7039,
      "mean_token_accuracy": 0.1701914981007576,
      "num_tokens": 34443620.0,
      "step": 17615
    },
    {
      "entropy": 6.139363574981689,
      "epoch": 1.761333533263358,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046944758671817,
      "loss": 5.8242,
      "mean_token_accuracy": 0.16067755073308945,
      "num_tokens": 34452831.0,
      "step": 17620
    },
    {
      "entropy": 6.053916883468628,
      "epoch": 1.7618333583245864,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004694296280607899,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1711435243487358,
      "num_tokens": 34462484.0,
      "step": 17625
    },
    {
      "entropy": 6.018930578231812,
      "epoch": 1.762333183385815,
      "grad_norm": 2.09375,
      "learning_rate": 0.0004694116645116104,
      "loss": 5.7483,
      "mean_token_accuracy": 0.17276942431926728,
      "num_tokens": 34471546.0,
      "step": 17630
    },
    {
      "entropy": 6.164613103866577,
      "epoch": 1.7628330084470436,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004693936960710834,
      "loss": 5.8649,
      "mean_token_accuracy": 0.16016077995300293,
      "num_tokens": 34482384.0,
      "step": 17635
    },
    {
      "entropy": 6.112231874465943,
      "epoch": 1.763332833508272,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046937572273966104,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17694345116615295,
      "num_tokens": 34491897.0,
      "step": 17640
    },
    {
      "entropy": 6.088321304321289,
      "epoch": 1.7638326585695006,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046935774451779556,
      "loss": 5.8104,
      "mean_token_accuracy": 0.1603526785969734,
      "num_tokens": 34501709.0,
      "step": 17645
    },
    {
      "entropy": 6.054780149459839,
      "epoch": 1.7643324836307293,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004693397614059392,
      "loss": 5.7468,
      "mean_token_accuracy": 0.17082723826169968,
      "num_tokens": 34510560.0,
      "step": 17650
    },
    {
      "entropy": 6.097449588775635,
      "epoch": 1.764832308691958,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004693217734045445,
      "loss": 5.801,
      "mean_token_accuracy": 0.16515442430973054,
      "num_tokens": 34519363.0,
      "step": 17655
    },
    {
      "entropy": 6.052739000320434,
      "epoch": 1.7653321337531864,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046930378051406393,
      "loss": 5.8073,
      "mean_token_accuracy": 0.1674882709980011,
      "num_tokens": 34530335.0,
      "step": 17660
    },
    {
      "entropy": 6.061106824874878,
      "epoch": 1.765831958814415,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046928578273495026,
      "loss": 5.7271,
      "mean_token_accuracy": 0.17058241218328477,
      "num_tokens": 34540603.0,
      "step": 17665
    },
    {
      "entropy": 6.103881072998047,
      "epoch": 1.7663317838756436,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004692677800676563,
      "loss": 5.7477,
      "mean_token_accuracy": 0.17275501787662506,
      "num_tokens": 34550765.0,
      "step": 17670
    },
    {
      "entropy": 6.150298738479615,
      "epoch": 1.766831608936872,
      "grad_norm": 1.40625,
      "learning_rate": 0.00046924977251263495,
      "loss": 5.7736,
      "mean_token_accuracy": 0.1703992635011673,
      "num_tokens": 34560330.0,
      "step": 17675
    },
    {
      "entropy": 6.060427570343018,
      "epoch": 1.7673314339981006,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046923176007033937,
      "loss": 5.7953,
      "mean_token_accuracy": 0.16737726777791978,
      "num_tokens": 34569356.0,
      "step": 17680
    },
    {
      "entropy": 6.056428289413452,
      "epoch": 1.7678312590593293,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046921374274122265,
      "loss": 5.7551,
      "mean_token_accuracy": 0.17020260095596312,
      "num_tokens": 34578613.0,
      "step": 17685
    },
    {
      "entropy": 6.035226583480835,
      "epoch": 1.768331084120558,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046919572052573814,
      "loss": 5.6696,
      "mean_token_accuracy": 0.17857344448566437,
      "num_tokens": 34587394.0,
      "step": 17690
    },
    {
      "entropy": 6.068990278244018,
      "epoch": 1.7688309091817864,
      "grad_norm": 1.25,
      "learning_rate": 0.00046917769342433925,
      "loss": 5.793,
      "mean_token_accuracy": 0.17222824394702912,
      "num_tokens": 34598707.0,
      "step": 17695
    },
    {
      "entropy": 6.142226934432983,
      "epoch": 1.769330734243015,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004691596614374796,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16175631284713746,
      "num_tokens": 34608270.0,
      "step": 17700
    },
    {
      "entropy": 6.106780385971069,
      "epoch": 1.7698305593042436,
      "grad_norm": 1.125,
      "learning_rate": 0.0004691416245656128,
      "loss": 5.8349,
      "mean_token_accuracy": 0.1698239229619503,
      "num_tokens": 34618533.0,
      "step": 17705
    },
    {
      "entropy": 6.14064130783081,
      "epoch": 1.770330384365472,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046912358280919267,
      "loss": 5.893,
      "mean_token_accuracy": 0.15824734717607497,
      "num_tokens": 34628158.0,
      "step": 17710
    },
    {
      "entropy": 6.126226282119751,
      "epoch": 1.7708302094267006,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046910553616867314,
      "loss": 5.747,
      "mean_token_accuracy": 0.17432289123535155,
      "num_tokens": 34638046.0,
      "step": 17715
    },
    {
      "entropy": 6.153215456008911,
      "epoch": 1.7713300344879293,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004690874846445083,
      "loss": 5.8691,
      "mean_token_accuracy": 0.1553115203976631,
      "num_tokens": 34647847.0,
      "step": 17720
    },
    {
      "entropy": 6.090858125686646,
      "epoch": 1.771829859549158,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004690694282371523,
      "loss": 5.7608,
      "mean_token_accuracy": 0.1657584860920906,
      "num_tokens": 34658830.0,
      "step": 17725
    },
    {
      "entropy": 6.024606037139892,
      "epoch": 1.7723296846103862,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004690513669470594,
      "loss": 5.7304,
      "mean_token_accuracy": 0.17838093042373657,
      "num_tokens": 34668679.0,
      "step": 17730
    },
    {
      "entropy": 6.096363973617554,
      "epoch": 1.772829509671615,
      "grad_norm": 1.3125,
      "learning_rate": 0.000469033300774684,
      "loss": 5.8054,
      "mean_token_accuracy": 0.16779908537864685,
      "num_tokens": 34678443.0,
      "step": 17735
    },
    {
      "entropy": 6.193962907791137,
      "epoch": 1.7733293347328436,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004690152297204808,
      "loss": 5.8379,
      "mean_token_accuracy": 0.167942114174366,
      "num_tokens": 34688348.0,
      "step": 17740
    },
    {
      "entropy": 5.902188539505005,
      "epoch": 1.773829159794072,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046899715378490427,
      "loss": 5.5149,
      "mean_token_accuracy": 0.19103506654500962,
      "num_tokens": 34698402.0,
      "step": 17745
    },
    {
      "entropy": 6.014567852020264,
      "epoch": 1.7743289848553006,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046897907296840927,
      "loss": 5.7573,
      "mean_token_accuracy": 0.1705679178237915,
      "num_tokens": 34707761.0,
      "step": 17750
    },
    {
      "entropy": 6.117718362808228,
      "epoch": 1.7748288099165292,
      "grad_norm": 1.25,
      "learning_rate": 0.0004689609872714507,
      "loss": 5.8774,
      "mean_token_accuracy": 0.16225721389055253,
      "num_tokens": 34717934.0,
      "step": 17755
    },
    {
      "entropy": 6.188833284378052,
      "epoch": 1.775328634977758,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046894289669448366,
      "loss": 5.8685,
      "mean_token_accuracy": 0.16039598286151885,
      "num_tokens": 34727486.0,
      "step": 17760
    },
    {
      "entropy": 6.145923137664795,
      "epoch": 1.7758284600389862,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00046892480123796314,
      "loss": 5.7747,
      "mean_token_accuracy": 0.1702690452337265,
      "num_tokens": 34738080.0,
      "step": 17765
    },
    {
      "entropy": 6.066918277740479,
      "epoch": 1.776328285100215,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004689067009023446,
      "loss": 5.8222,
      "mean_token_accuracy": 0.1629923477768898,
      "num_tokens": 34747595.0,
      "step": 17770
    },
    {
      "entropy": 6.045388078689575,
      "epoch": 1.7768281101614436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004688885956880834,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17143936604261398,
      "num_tokens": 34756818.0,
      "step": 17775
    },
    {
      "entropy": 6.07952880859375,
      "epoch": 1.777327935222672,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004688704855956349,
      "loss": 5.7755,
      "mean_token_accuracy": 0.16656854301691054,
      "num_tokens": 34767263.0,
      "step": 17780
    },
    {
      "entropy": 6.171099138259888,
      "epoch": 1.7778277602839005,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046885237062545493,
      "loss": 5.8643,
      "mean_token_accuracy": 0.15952659249305726,
      "num_tokens": 34776767.0,
      "step": 17785
    },
    {
      "entropy": 6.116524124145508,
      "epoch": 1.7783275853451292,
      "grad_norm": 1.125,
      "learning_rate": 0.00046883425077799923,
      "loss": 5.7904,
      "mean_token_accuracy": 0.1651027724146843,
      "num_tokens": 34787587.0,
      "step": 17790
    },
    {
      "entropy": 6.0904669761657715,
      "epoch": 1.7788274104063577,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004688161260537236,
      "loss": 5.8666,
      "mean_token_accuracy": 0.1545224145054817,
      "num_tokens": 34798586.0,
      "step": 17795
    },
    {
      "entropy": 6.118031311035156,
      "epoch": 1.7793272354675862,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004687979964530842,
      "loss": 5.8177,
      "mean_token_accuracy": 0.16262117102742196,
      "num_tokens": 34808469.0,
      "step": 17800
    },
    {
      "entropy": 6.153189706802368,
      "epoch": 1.779827060528815,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000468779861976537,
      "loss": 5.7937,
      "mean_token_accuracy": 0.16816478669643403,
      "num_tokens": 34817933.0,
      "step": 17805
    },
    {
      "entropy": 6.056660223007202,
      "epoch": 1.7803268855900436,
      "grad_norm": 1.5234375,
      "learning_rate": 0.00046876172262453843,
      "loss": 5.7472,
      "mean_token_accuracy": 0.16604597568511964,
      "num_tokens": 34827370.0,
      "step": 17810
    },
    {
      "entropy": 6.091251468658447,
      "epoch": 1.780826710651272,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046874357839754463,
      "loss": 5.7994,
      "mean_token_accuracy": 0.16718368828296662,
      "num_tokens": 34835521.0,
      "step": 17815
    },
    {
      "entropy": 6.190127658843994,
      "epoch": 1.7813265357125005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046872542929601234,
      "loss": 6.0719,
      "mean_token_accuracy": 0.15558838620781898,
      "num_tokens": 34846151.0,
      "step": 17820
    },
    {
      "entropy": 6.146645355224609,
      "epoch": 1.7818263607737292,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046870727532039816,
      "loss": 5.7238,
      "mean_token_accuracy": 0.16631383150815965,
      "num_tokens": 34855536.0,
      "step": 17825
    },
    {
      "entropy": 6.13318772315979,
      "epoch": 1.7823261858349577,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004686891164711586,
      "loss": 5.7935,
      "mean_token_accuracy": 0.16497173458337783,
      "num_tokens": 34865428.0,
      "step": 17830
    },
    {
      "entropy": 6.043006753921508,
      "epoch": 1.7828260108961862,
      "grad_norm": 1.375,
      "learning_rate": 0.00046867095274875085,
      "loss": 5.7425,
      "mean_token_accuracy": 0.163699933886528,
      "num_tokens": 34875273.0,
      "step": 17835
    },
    {
      "entropy": 6.0623523712158205,
      "epoch": 1.783325835957415,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046865278415363174,
      "loss": 5.848,
      "mean_token_accuracy": 0.1607326626777649,
      "num_tokens": 34885438.0,
      "step": 17840
    },
    {
      "entropy": 6.199893140792847,
      "epoch": 1.7838256610186436,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004686346106862584,
      "loss": 5.8631,
      "mean_token_accuracy": 0.1620113119482994,
      "num_tokens": 34895245.0,
      "step": 17845
    },
    {
      "entropy": 6.114214897155762,
      "epoch": 1.784325486079872,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004686164323470881,
      "loss": 5.7451,
      "mean_token_accuracy": 0.16725882589817048,
      "num_tokens": 34905028.0,
      "step": 17850
    },
    {
      "entropy": 6.126498985290527,
      "epoch": 1.7848253111411005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004685982491365781,
      "loss": 5.8867,
      "mean_token_accuracy": 0.16192000955343247,
      "num_tokens": 34915756.0,
      "step": 17855
    },
    {
      "entropy": 6.1313621520996096,
      "epoch": 1.7853251362023292,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046858006105518615,
      "loss": 5.7787,
      "mean_token_accuracy": 0.16696716845035553,
      "num_tokens": 34925369.0,
      "step": 17860
    },
    {
      "entropy": 6.100793695449829,
      "epoch": 1.7858249612635577,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004685618681033695,
      "loss": 5.8422,
      "mean_token_accuracy": 0.16141773760318756,
      "num_tokens": 34935693.0,
      "step": 17865
    },
    {
      "entropy": 6.122704315185547,
      "epoch": 1.7863247863247862,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004685436702815861,
      "loss": 5.8239,
      "mean_token_accuracy": 0.16642831712961198,
      "num_tokens": 34944582.0,
      "step": 17870
    },
    {
      "entropy": 6.135226249694824,
      "epoch": 1.786824611386015,
      "grad_norm": 2.34375,
      "learning_rate": 0.0004685254675902938,
      "loss": 5.8046,
      "mean_token_accuracy": 0.16437610238790512,
      "num_tokens": 34953762.0,
      "step": 17875
    },
    {
      "entropy": 6.123812866210938,
      "epoch": 1.7873244364472436,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004685072600299505,
      "loss": 5.8689,
      "mean_token_accuracy": 0.16340502351522446,
      "num_tokens": 34963003.0,
      "step": 17880
    },
    {
      "entropy": 6.083636713027954,
      "epoch": 1.787824261508472,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004684890476010144,
      "loss": 5.834,
      "mean_token_accuracy": 0.1611725464463234,
      "num_tokens": 34972968.0,
      "step": 17885
    },
    {
      "entropy": 6.140351486206055,
      "epoch": 1.7883240865697005,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004684708303039435,
      "loss": 5.9016,
      "mean_token_accuracy": 0.16559814512729645,
      "num_tokens": 34983050.0,
      "step": 17890
    },
    {
      "entropy": 6.200511360168457,
      "epoch": 1.7888239116309292,
      "grad_norm": 1.328125,
      "learning_rate": 0.00046845260813919645,
      "loss": 5.8233,
      "mean_token_accuracy": 0.16159948706626892,
      "num_tokens": 34992192.0,
      "step": 17895
    },
    {
      "entropy": 6.083561420440674,
      "epoch": 1.7893237366921577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00046843438110723143,
      "loss": 5.7742,
      "mean_token_accuracy": 0.169467294216156,
      "num_tokens": 35001135.0,
      "step": 17900
    },
    {
      "entropy": 6.073339605331421,
      "epoch": 1.7898235617533862,
      "grad_norm": 2.5625,
      "learning_rate": 0.00046841614920850724,
      "loss": 5.8377,
      "mean_token_accuracy": 0.162664508074522,
      "num_tokens": 35011389.0,
      "step": 17905
    },
    {
      "entropy": 6.131946611404419,
      "epoch": 1.790323386814615,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046839791244348243,
      "loss": 5.8616,
      "mean_token_accuracy": 0.16711240857839585,
      "num_tokens": 35021752.0,
      "step": 17910
    },
    {
      "entropy": 6.096583032608033,
      "epoch": 1.7908232118758436,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046837967081261606,
      "loss": 5.7851,
      "mean_token_accuracy": 0.16533706933259965,
      "num_tokens": 35031878.0,
      "step": 17915
    },
    {
      "entropy": 6.075601434707641,
      "epoch": 1.791323036937072,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004683614243163668,
      "loss": 5.7289,
      "mean_token_accuracy": 0.17105406820774077,
      "num_tokens": 35040995.0,
      "step": 17920
    },
    {
      "entropy": 6.0946441173553465,
      "epoch": 1.7918228619983005,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004683431729551939,
      "loss": 5.7598,
      "mean_token_accuracy": 0.1701200619339943,
      "num_tokens": 35050680.0,
      "step": 17925
    },
    {
      "entropy": 6.097586345672608,
      "epoch": 1.7923226870595292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004683249167295565,
      "loss": 5.7676,
      "mean_token_accuracy": 0.16994592994451524,
      "num_tokens": 35059910.0,
      "step": 17930
    },
    {
      "entropy": 6.054046154022217,
      "epoch": 1.7928225121207577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000468306655639914,
      "loss": 5.7356,
      "mean_token_accuracy": 0.17317907586693765,
      "num_tokens": 35069711.0,
      "step": 17935
    },
    {
      "entropy": 6.125696516036987,
      "epoch": 1.7933223371819862,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004682883896867257,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16578530818223952,
      "num_tokens": 35078711.0,
      "step": 17940
    },
    {
      "entropy": 6.069156980514526,
      "epoch": 1.793822162243215,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004682701188704513,
      "loss": 5.7909,
      "mean_token_accuracy": 0.17187907844781875,
      "num_tokens": 35087632.0,
      "step": 17945
    },
    {
      "entropy": 6.0654350280761715,
      "epoch": 1.7943219873044436,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004682518431915505,
      "loss": 5.792,
      "mean_token_accuracy": 0.17536998689174652,
      "num_tokens": 35097257.0,
      "step": 17950
    },
    {
      "entropy": 6.129639291763306,
      "epoch": 1.794821812365672,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046823356265048293,
      "loss": 5.8335,
      "mean_token_accuracy": 0.16681813597679138,
      "num_tokens": 35107734.0,
      "step": 17955
    },
    {
      "entropy": 6.13929295539856,
      "epoch": 1.7953216374269005,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004682152772477087,
      "loss": 5.7772,
      "mean_token_accuracy": 0.16905079185962676,
      "num_tokens": 35117052.0,
      "step": 17960
    },
    {
      "entropy": 6.114010906219482,
      "epoch": 1.7958214624881292,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046819698698368785,
      "loss": 5.8251,
      "mean_token_accuracy": 0.16201669871807098,
      "num_tokens": 35127345.0,
      "step": 17965
    },
    {
      "entropy": 6.073897933959961,
      "epoch": 1.7963212875493577,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046817869185888044,
      "loss": 5.7265,
      "mean_token_accuracy": 0.16584499627351762,
      "num_tokens": 35137216.0,
      "step": 17970
    },
    {
      "entropy": 6.116776180267334,
      "epoch": 1.7968211126105862,
      "grad_norm": 1.8203125,
      "learning_rate": 0.0004681603918737469,
      "loss": 5.8016,
      "mean_token_accuracy": 0.17316877245903015,
      "num_tokens": 35146433.0,
      "step": 17975
    },
    {
      "entropy": 6.055818176269531,
      "epoch": 1.7973209376718149,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046814208702874766,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16329140216112137,
      "num_tokens": 35155841.0,
      "step": 17980
    },
    {
      "entropy": 6.06991057395935,
      "epoch": 1.7978207627330436,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046812377732434316,
      "loss": 5.8724,
      "mean_token_accuracy": 0.1664571315050125,
      "num_tokens": 35165962.0,
      "step": 17985
    },
    {
      "entropy": 6.119021606445313,
      "epoch": 1.7983205877942718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046810546276099415,
      "loss": 5.8412,
      "mean_token_accuracy": 0.16151943281292916,
      "num_tokens": 35175887.0,
      "step": 17990
    },
    {
      "entropy": 6.173126745223999,
      "epoch": 1.7988204128555005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004680871433391613,
      "loss": 5.8131,
      "mean_token_accuracy": 0.16589714735746383,
      "num_tokens": 35185763.0,
      "step": 17995
    },
    {
      "entropy": 6.090228176116943,
      "epoch": 1.7993202379167292,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046806881905930565,
      "loss": 5.8007,
      "mean_token_accuracy": 0.17068539410829545,
      "num_tokens": 35195452.0,
      "step": 18000
    },
    {
      "epoch": 1.7993202379167292,
      "eval_entropy": 5.861933311952877,
      "eval_loss": 5.897674560546875,
      "eval_mean_token_accuracy": 0.16906350237840279,
      "eval_num_tokens": 35195452.0,
      "eval_runtime": 23.9861,
      "eval_samples_per_second": 1294.293,
      "eval_steps_per_second": 161.802,
      "step": 18000
    },
    {
      "entropy": 6.105839586257934,
      "epoch": 1.7998200629779577,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046805048992188823,
      "loss": 5.8644,
      "mean_token_accuracy": 0.16288941502571105,
      "num_tokens": 35205261.0,
      "step": 18005
    },
    {
      "entropy": 6.133154010772705,
      "epoch": 1.8003198880391862,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046803215592737014,
      "loss": 5.8433,
      "mean_token_accuracy": 0.16536638587713243,
      "num_tokens": 35214965.0,
      "step": 18010
    },
    {
      "entropy": 6.096221542358398,
      "epoch": 1.8008197131004149,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046801381707621266,
      "loss": 5.8013,
      "mean_token_accuracy": 0.16576366424560546,
      "num_tokens": 35224371.0,
      "step": 18015
    },
    {
      "entropy": 6.037886095046997,
      "epoch": 1.8013195381616436,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046799547336887727,
      "loss": 5.7216,
      "mean_token_accuracy": 0.1736925557255745,
      "num_tokens": 35234282.0,
      "step": 18020
    },
    {
      "entropy": 6.122465515136719,
      "epoch": 1.8018193632228718,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004679771248058254,
      "loss": 5.7429,
      "mean_token_accuracy": 0.17538924217224122,
      "num_tokens": 35243112.0,
      "step": 18025
    },
    {
      "entropy": 6.089845323562622,
      "epoch": 1.8023191882841005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004679587713875187,
      "loss": 5.7837,
      "mean_token_accuracy": 0.15819660127162932,
      "num_tokens": 35253398.0,
      "step": 18030
    },
    {
      "entropy": 6.068023061752319,
      "epoch": 1.8028190133453292,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000467940413114419,
      "loss": 5.7023,
      "mean_token_accuracy": 0.17255635261535646,
      "num_tokens": 35263359.0,
      "step": 18035
    },
    {
      "entropy": 6.145987033843994,
      "epoch": 1.8033188384065577,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004679220499869881,
      "loss": 5.8895,
      "mean_token_accuracy": 0.15931747555732728,
      "num_tokens": 35272500.0,
      "step": 18040
    },
    {
      "entropy": 6.202304124832153,
      "epoch": 1.8038186634677862,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004679036820056881,
      "loss": 5.8608,
      "mean_token_accuracy": 0.15133578553795815,
      "num_tokens": 35282415.0,
      "step": 18045
    },
    {
      "entropy": 6.172774791717529,
      "epoch": 1.8043184885290149,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046788530917098105,
      "loss": 5.808,
      "mean_token_accuracy": 0.16334254890680314,
      "num_tokens": 35291567.0,
      "step": 18050
    },
    {
      "entropy": 6.101656436920166,
      "epoch": 1.8048183135902434,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004678669314833293,
      "loss": 5.9229,
      "mean_token_accuracy": 0.15782424211502075,
      "num_tokens": 35301397.0,
      "step": 18055
    },
    {
      "entropy": 6.087209606170655,
      "epoch": 1.8053181386514718,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046784854894319525,
      "loss": 5.752,
      "mean_token_accuracy": 0.17305241525173187,
      "num_tokens": 35311296.0,
      "step": 18060
    },
    {
      "entropy": 6.131953477859497,
      "epoch": 1.8058179637127005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046783016155104123,
      "loss": 5.844,
      "mean_token_accuracy": 0.17097675651311875,
      "num_tokens": 35322819.0,
      "step": 18065
    },
    {
      "entropy": 6.119858932495117,
      "epoch": 1.8063177887739292,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046781176930732997,
      "loss": 5.7318,
      "mean_token_accuracy": 0.17132087498903276,
      "num_tokens": 35331374.0,
      "step": 18070
    },
    {
      "entropy": 6.049019622802734,
      "epoch": 1.8068176138351577,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004677933722125243,
      "loss": 5.8277,
      "mean_token_accuracy": 0.16618005335330963,
      "num_tokens": 35341040.0,
      "step": 18075
    },
    {
      "entropy": 6.151770782470703,
      "epoch": 1.8073174388963862,
      "grad_norm": 1.375,
      "learning_rate": 0.0004677749702670869,
      "loss": 5.852,
      "mean_token_accuracy": 0.15483629703521729,
      "num_tokens": 35351387.0,
      "step": 18080
    },
    {
      "entropy": 6.152039098739624,
      "epoch": 1.8078172639576149,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004677565634714809,
      "loss": 5.8553,
      "mean_token_accuracy": 0.15843773186206817,
      "num_tokens": 35361864.0,
      "step": 18085
    },
    {
      "entropy": 6.059840106964112,
      "epoch": 1.8083170890188434,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004677381518261693,
      "loss": 5.7478,
      "mean_token_accuracy": 0.1783355876803398,
      "num_tokens": 35370539.0,
      "step": 18090
    },
    {
      "entropy": 6.052991008758545,
      "epoch": 1.8088169140800718,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046771973533161543,
      "loss": 5.7621,
      "mean_token_accuracy": 0.17573347836732864,
      "num_tokens": 35381851.0,
      "step": 18095
    },
    {
      "entropy": 6.130247116088867,
      "epoch": 1.8093167391413005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046770131398828264,
      "loss": 5.773,
      "mean_token_accuracy": 0.17053579837083815,
      "num_tokens": 35391440.0,
      "step": 18100
    },
    {
      "entropy": 6.139250469207764,
      "epoch": 1.8098165642025292,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046768288779663436,
      "loss": 5.8232,
      "mean_token_accuracy": 0.16423137187957765,
      "num_tokens": 35401370.0,
      "step": 18105
    },
    {
      "entropy": 6.155709218978882,
      "epoch": 1.8103163892637577,
      "grad_norm": 1.25,
      "learning_rate": 0.0004676644567571342,
      "loss": 5.8104,
      "mean_token_accuracy": 0.16085606962442398,
      "num_tokens": 35411015.0,
      "step": 18110
    },
    {
      "entropy": 6.176739311218261,
      "epoch": 1.8108162143249862,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046764602087024596,
      "loss": 5.8205,
      "mean_token_accuracy": 0.16111122071743011,
      "num_tokens": 35421741.0,
      "step": 18115
    },
    {
      "entropy": 6.208624267578125,
      "epoch": 1.8113160393862149,
      "grad_norm": 1.125,
      "learning_rate": 0.0004676275801364332,
      "loss": 5.9434,
      "mean_token_accuracy": 0.1547234460711479,
      "num_tokens": 35432899.0,
      "step": 18120
    },
    {
      "entropy": 6.056198310852051,
      "epoch": 1.8118158644474434,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004676091345561603,
      "loss": 5.6935,
      "mean_token_accuracy": 0.17496990412473679,
      "num_tokens": 35442220.0,
      "step": 18125
    },
    {
      "entropy": 6.025050115585327,
      "epoch": 1.8123156895086718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000467590684129891,
      "loss": 5.742,
      "mean_token_accuracy": 0.1747843787074089,
      "num_tokens": 35452193.0,
      "step": 18130
    },
    {
      "entropy": 6.109678840637207,
      "epoch": 1.8128155145699005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046757222885808973,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1708253502845764,
      "num_tokens": 35462152.0,
      "step": 18135
    },
    {
      "entropy": 6.111292505264283,
      "epoch": 1.8133153396311292,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046755376874122075,
      "loss": 5.7891,
      "mean_token_accuracy": 0.16616717875003814,
      "num_tokens": 35472210.0,
      "step": 18140
    },
    {
      "entropy": 6.133566665649414,
      "epoch": 1.8138151646923577,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004675353037797485,
      "loss": 5.8598,
      "mean_token_accuracy": 0.15686623901128768,
      "num_tokens": 35482238.0,
      "step": 18145
    },
    {
      "entropy": 6.081112289428711,
      "epoch": 1.8143149897535862,
      "grad_norm": 1.25,
      "learning_rate": 0.0004675168339741375,
      "loss": 5.837,
      "mean_token_accuracy": 0.16125814765691757,
      "num_tokens": 35492042.0,
      "step": 18150
    },
    {
      "entropy": 6.1606358051300045,
      "epoch": 1.8148148148148149,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004674983593248525,
      "loss": 5.8976,
      "mean_token_accuracy": 0.15734914243221282,
      "num_tokens": 35501892.0,
      "step": 18155
    },
    {
      "entropy": 6.168277168273926,
      "epoch": 1.8153146398760434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046747987983235836,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16236926913261412,
      "num_tokens": 35512472.0,
      "step": 18160
    },
    {
      "entropy": 6.132704877853394,
      "epoch": 1.8158144649372718,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00046746139549711993,
      "loss": 5.7211,
      "mean_token_accuracy": 0.17130089551210403,
      "num_tokens": 35521467.0,
      "step": 18165
    },
    {
      "entropy": 6.031136560440063,
      "epoch": 1.8163142899985005,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004674429063196023,
      "loss": 5.8204,
      "mean_token_accuracy": 0.16831044256687164,
      "num_tokens": 35531584.0,
      "step": 18170
    },
    {
      "entropy": 6.088880014419556,
      "epoch": 1.8168141150597292,
      "grad_norm": 1.4453125,
      "learning_rate": 0.00046742441230027067,
      "loss": 5.675,
      "mean_token_accuracy": 0.1743665799498558,
      "num_tokens": 35541687.0,
      "step": 18175
    },
    {
      "entropy": 6.0870436668396,
      "epoch": 1.8173139401209577,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004674059134395904,
      "loss": 5.78,
      "mean_token_accuracy": 0.16289244294166566,
      "num_tokens": 35551310.0,
      "step": 18180
    },
    {
      "entropy": 6.093764114379883,
      "epoch": 1.8178137651821862,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004673874097380268,
      "loss": 5.7475,
      "mean_token_accuracy": 0.16740995943546294,
      "num_tokens": 35560356.0,
      "step": 18185
    },
    {
      "entropy": 6.096996879577636,
      "epoch": 1.8183135902434149,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004673689011960455,
      "loss": 5.7268,
      "mean_token_accuracy": 0.17150569558143616,
      "num_tokens": 35570633.0,
      "step": 18190
    },
    {
      "entropy": 6.140740251541137,
      "epoch": 1.8188134153046434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046735038781411204,
      "loss": 5.9291,
      "mean_token_accuracy": 0.15552953779697418,
      "num_tokens": 35581341.0,
      "step": 18195
    },
    {
      "entropy": 6.017335748672485,
      "epoch": 1.8193132403658718,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004673318695926925,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1844051957130432,
      "num_tokens": 35592301.0,
      "step": 18200
    },
    {
      "entropy": 6.048006963729859,
      "epoch": 1.8198130654271005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004673133465322525,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17337230890989302,
      "num_tokens": 35601782.0,
      "step": 18205
    },
    {
      "entropy": 6.022095012664795,
      "epoch": 1.8203128904883292,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004672948186332582,
      "loss": 5.6806,
      "mean_token_accuracy": 0.17057230770587922,
      "num_tokens": 35611307.0,
      "step": 18210
    },
    {
      "entropy": 6.112345218658447,
      "epoch": 1.8208127155495577,
      "grad_norm": 1.3984375,
      "learning_rate": 0.00046727628589617574,
      "loss": 5.8542,
      "mean_token_accuracy": 0.16126495748758315,
      "num_tokens": 35621358.0,
      "step": 18215
    },
    {
      "entropy": 6.074323749542236,
      "epoch": 1.8213125406107862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046725774832147147,
      "loss": 5.7665,
      "mean_token_accuracy": 0.16569652855396272,
      "num_tokens": 35630976.0,
      "step": 18220
    },
    {
      "entropy": 6.05137882232666,
      "epoch": 1.8218123656720149,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046723920590961167,
      "loss": 5.7369,
      "mean_token_accuracy": 0.16888122856616974,
      "num_tokens": 35640126.0,
      "step": 18225
    },
    {
      "entropy": 6.103144645690918,
      "epoch": 1.8223121907332434,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004672206586610629,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16719338297843933,
      "num_tokens": 35649282.0,
      "step": 18230
    },
    {
      "entropy": 6.075000667572022,
      "epoch": 1.8228120157944718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046720210657629185,
      "loss": 5.7131,
      "mean_token_accuracy": 0.17080237716436386,
      "num_tokens": 35658284.0,
      "step": 18235
    },
    {
      "entropy": 6.087240552902221,
      "epoch": 1.8233118408557005,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046718354965576527,
      "loss": 5.8742,
      "mean_token_accuracy": 0.159806552529335,
      "num_tokens": 35668355.0,
      "step": 18240
    },
    {
      "entropy": 6.096940708160401,
      "epoch": 1.8238116659169292,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046716498789995,
      "loss": 5.8315,
      "mean_token_accuracy": 0.1575819134712219,
      "num_tokens": 35677582.0,
      "step": 18245
    },
    {
      "entropy": 6.1221763610839846,
      "epoch": 1.8243114909781575,
      "grad_norm": 1.25,
      "learning_rate": 0.00046714642130931306,
      "loss": 5.8503,
      "mean_token_accuracy": 0.1635047286748886,
      "num_tokens": 35687691.0,
      "step": 18250
    },
    {
      "entropy": 6.086605978012085,
      "epoch": 1.8248113160393862,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00046712784988432173,
      "loss": 5.821,
      "mean_token_accuracy": 0.16328749060630798,
      "num_tokens": 35697796.0,
      "step": 18255
    },
    {
      "entropy": 6.1273908615112305,
      "epoch": 1.8253111411006149,
      "grad_norm": 1.375,
      "learning_rate": 0.000467109273625443,
      "loss": 5.742,
      "mean_token_accuracy": 0.17108754366636275,
      "num_tokens": 35706795.0,
      "step": 18260
    },
    {
      "entropy": 6.140839529037476,
      "epoch": 1.8258109661618434,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004670906925331444,
      "loss": 5.8461,
      "mean_token_accuracy": 0.16877455860376359,
      "num_tokens": 35717260.0,
      "step": 18265
    },
    {
      "entropy": 6.097375726699829,
      "epoch": 1.8263107912230718,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046707210660789343,
      "loss": 5.8592,
      "mean_token_accuracy": 0.16497842818498612,
      "num_tokens": 35726671.0,
      "step": 18270
    },
    {
      "entropy": 6.155775165557861,
      "epoch": 1.8268106162843005,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004670535158501577,
      "loss": 5.857,
      "mean_token_accuracy": 0.16165828555822373,
      "num_tokens": 35736672.0,
      "step": 18275
    },
    {
      "entropy": 6.153855037689209,
      "epoch": 1.8273104413455292,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046703492026040495,
      "loss": 5.7648,
      "mean_token_accuracy": 0.1625993475317955,
      "num_tokens": 35746291.0,
      "step": 18280
    },
    {
      "entropy": 6.041607856750488,
      "epoch": 1.8278102664067575,
      "grad_norm": 2.734375,
      "learning_rate": 0.00046701631983910296,
      "loss": 5.7417,
      "mean_token_accuracy": 0.16345332488417624,
      "num_tokens": 35755683.0,
      "step": 18285
    },
    {
      "entropy": 6.09064416885376,
      "epoch": 1.8283100914679862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046699771458671974,
      "loss": 5.8019,
      "mean_token_accuracy": 0.1636059269309044,
      "num_tokens": 35765870.0,
      "step": 18290
    },
    {
      "entropy": 5.998325967788697,
      "epoch": 1.8288099165292149,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046697910450372356,
      "loss": 5.6444,
      "mean_token_accuracy": 0.18049484938383104,
      "num_tokens": 35777055.0,
      "step": 18295
    },
    {
      "entropy": 6.031801319122314,
      "epoch": 1.8293097415904434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046696048959058236,
      "loss": 5.6892,
      "mean_token_accuracy": 0.17302531450986863,
      "num_tokens": 35786846.0,
      "step": 18300
    },
    {
      "entropy": 6.0700554847717285,
      "epoch": 1.8298095666516718,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004669418698477647,
      "loss": 5.7071,
      "mean_token_accuracy": 0.17461602240800858,
      "num_tokens": 35795471.0,
      "step": 18305
    },
    {
      "entropy": 6.039705896377564,
      "epoch": 1.8303093917129005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046692324527573907,
      "loss": 5.6621,
      "mean_token_accuracy": 0.1750674292445183,
      "num_tokens": 35804559.0,
      "step": 18310
    },
    {
      "entropy": 6.03839955329895,
      "epoch": 1.830809216774129,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004669046158749738,
      "loss": 5.7898,
      "mean_token_accuracy": 0.16533504724502562,
      "num_tokens": 35813545.0,
      "step": 18315
    },
    {
      "entropy": 6.034802579879761,
      "epoch": 1.8313090418353575,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046688598164593793,
      "loss": 5.8058,
      "mean_token_accuracy": 0.1689276859164238,
      "num_tokens": 35822740.0,
      "step": 18320
    },
    {
      "entropy": 6.063875293731689,
      "epoch": 1.8318088668965862,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004668673425891001,
      "loss": 5.8279,
      "mean_token_accuracy": 0.16060149073600768,
      "num_tokens": 35832593.0,
      "step": 18325
    },
    {
      "entropy": 6.15786828994751,
      "epoch": 1.8323086919578149,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046684869870492927,
      "loss": 5.7982,
      "mean_token_accuracy": 0.16117842346429825,
      "num_tokens": 35841399.0,
      "step": 18330
    },
    {
      "entropy": 6.0871964454650875,
      "epoch": 1.8328085170190433,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046683004999389455,
      "loss": 5.8193,
      "mean_token_accuracy": 0.166938915848732,
      "num_tokens": 35850099.0,
      "step": 18335
    },
    {
      "entropy": 6.107338857650757,
      "epoch": 1.8333083420802718,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004668113964564651,
      "loss": 5.8472,
      "mean_token_accuracy": 0.16298332512378694,
      "num_tokens": 35860398.0,
      "step": 18340
    },
    {
      "entropy": 6.089983081817627,
      "epoch": 1.8338081671415005,
      "grad_norm": 2.0,
      "learning_rate": 0.0004667927380931104,
      "loss": 5.7529,
      "mean_token_accuracy": 0.1697830006480217,
      "num_tokens": 35870390.0,
      "step": 18345
    },
    {
      "entropy": 6.075755929946899,
      "epoch": 1.834307992202729,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004667740749042997,
      "loss": 5.7084,
      "mean_token_accuracy": 0.1685636967420578,
      "num_tokens": 35881633.0,
      "step": 18350
    },
    {
      "entropy": 6.078665542602539,
      "epoch": 1.8348078172639575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046675540689050253,
      "loss": 5.8463,
      "mean_token_accuracy": 0.16277169138193132,
      "num_tokens": 35891904.0,
      "step": 18355
    },
    {
      "entropy": 6.185885715484619,
      "epoch": 1.8353076423251862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046673673405218876,
      "loss": 6.0559,
      "mean_token_accuracy": 0.15096188187599183,
      "num_tokens": 35901712.0,
      "step": 18360
    },
    {
      "entropy": 6.166149616241455,
      "epoch": 1.8358074673864149,
      "grad_norm": 1.25,
      "learning_rate": 0.0004667180563898281,
      "loss": 5.939,
      "mean_token_accuracy": 0.15502797663211823,
      "num_tokens": 35911314.0,
      "step": 18365
    },
    {
      "entropy": 6.229225587844849,
      "epoch": 1.8363072924476433,
      "grad_norm": 1.25,
      "learning_rate": 0.00046669937390389043,
      "loss": 5.8834,
      "mean_token_accuracy": 0.16151377111673354,
      "num_tokens": 35921135.0,
      "step": 18370
    },
    {
      "entropy": 6.2072821140289305,
      "epoch": 1.8368071175088718,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046668068659484584,
      "loss": 5.8552,
      "mean_token_accuracy": 0.1617009609937668,
      "num_tokens": 35930784.0,
      "step": 18375
    },
    {
      "entropy": 6.102618980407715,
      "epoch": 1.8373069425701005,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046666199446316454,
      "loss": 5.7425,
      "mean_token_accuracy": 0.1639061763882637,
      "num_tokens": 35940372.0,
      "step": 18380
    },
    {
      "entropy": 6.1054603099823,
      "epoch": 1.837806767631329,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046664329750931675,
      "loss": 5.7198,
      "mean_token_accuracy": 0.16964344680309296,
      "num_tokens": 35949507.0,
      "step": 18385
    },
    {
      "entropy": 6.124234867095947,
      "epoch": 1.8383065926925575,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000466624595733773,
      "loss": 5.7312,
      "mean_token_accuracy": 0.16554549485445022,
      "num_tokens": 35958981.0,
      "step": 18390
    },
    {
      "entropy": 6.1013140201568605,
      "epoch": 1.8388064177537862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004666058891370036,
      "loss": 5.7648,
      "mean_token_accuracy": 0.16905977427959443,
      "num_tokens": 35967842.0,
      "step": 18395
    },
    {
      "entropy": 6.056987285614014,
      "epoch": 1.8393062428150149,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004665871777194794,
      "loss": 5.7674,
      "mean_token_accuracy": 0.16577101796865462,
      "num_tokens": 35977669.0,
      "step": 18400
    },
    {
      "entropy": 6.092403697967529,
      "epoch": 1.8398060678762433,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046656846148167107,
      "loss": 5.7807,
      "mean_token_accuracy": 0.16774290651082993,
      "num_tokens": 35987815.0,
      "step": 18405
    },
    {
      "entropy": 6.183163356781006,
      "epoch": 1.8403058929374718,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046654974042404957,
      "loss": 5.9192,
      "mean_token_accuracy": 0.15566480308771133,
      "num_tokens": 35997831.0,
      "step": 18410
    },
    {
      "entropy": 6.127860116958618,
      "epoch": 1.8408057179987005,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046653101454708593,
      "loss": 5.792,
      "mean_token_accuracy": 0.1609821692109108,
      "num_tokens": 36006822.0,
      "step": 18415
    },
    {
      "entropy": 6.052264022827148,
      "epoch": 1.841305543059929,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004665122838512512,
      "loss": 5.6801,
      "mean_token_accuracy": 0.17176989167928697,
      "num_tokens": 36015648.0,
      "step": 18420
    },
    {
      "entropy": 6.137369775772095,
      "epoch": 1.8418053681211575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004664935483370167,
      "loss": 5.9226,
      "mean_token_accuracy": 0.15824417024850845,
      "num_tokens": 36025713.0,
      "step": 18425
    },
    {
      "entropy": 6.1446503639221195,
      "epoch": 1.8423051931823862,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004664748080048537,
      "loss": 5.7328,
      "mean_token_accuracy": 0.17061990052461623,
      "num_tokens": 36035054.0,
      "step": 18430
    },
    {
      "entropy": 6.092377662658691,
      "epoch": 1.8428050182436149,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046645606285523395,
      "loss": 5.7862,
      "mean_token_accuracy": 0.169942969083786,
      "num_tokens": 36045449.0,
      "step": 18435
    },
    {
      "entropy": 6.138909912109375,
      "epoch": 1.8433048433048433,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046643731288862884,
      "loss": 5.8266,
      "mean_token_accuracy": 0.1622672498226166,
      "num_tokens": 36055269.0,
      "step": 18440
    },
    {
      "entropy": 6.212170028686524,
      "epoch": 1.8438046683660718,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004664185581055103,
      "loss": 5.9326,
      "mean_token_accuracy": 0.1618528373539448,
      "num_tokens": 36064791.0,
      "step": 18445
    },
    {
      "entropy": 6.112911081314087,
      "epoch": 1.8443044934273005,
      "grad_norm": 1.7421875,
      "learning_rate": 0.00046639979850634995,
      "loss": 5.7716,
      "mean_token_accuracy": 0.1615307152271271,
      "num_tokens": 36075583.0,
      "step": 18450
    },
    {
      "entropy": 6.145718288421631,
      "epoch": 1.844804318488529,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046638103409162,
      "loss": 5.8473,
      "mean_token_accuracy": 0.16469219326972961,
      "num_tokens": 36084664.0,
      "step": 18455
    },
    {
      "entropy": 6.16836314201355,
      "epoch": 1.8453041435497575,
      "grad_norm": 1.4765625,
      "learning_rate": 0.00046636226486179244,
      "loss": 5.7805,
      "mean_token_accuracy": 0.17457810640335084,
      "num_tokens": 36094184.0,
      "step": 18460
    },
    {
      "entropy": 6.187822675704956,
      "epoch": 1.8458039686109862,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004663434908173396,
      "loss": 5.7754,
      "mean_token_accuracy": 0.17000445276498793,
      "num_tokens": 36103148.0,
      "step": 18465
    },
    {
      "entropy": 6.136875057220459,
      "epoch": 1.8463037936722149,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004663247119587338,
      "loss": 5.7998,
      "mean_token_accuracy": 0.16611856520175933,
      "num_tokens": 36114268.0,
      "step": 18470
    },
    {
      "entropy": 6.129911613464356,
      "epoch": 1.8468036187334433,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004663059282864474,
      "loss": 5.7838,
      "mean_token_accuracy": 0.1645222395658493,
      "num_tokens": 36124690.0,
      "step": 18475
    },
    {
      "entropy": 6.096324825286866,
      "epoch": 1.8473034437946718,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004662871398009531,
      "loss": 5.78,
      "mean_token_accuracy": 0.16634131222963333,
      "num_tokens": 36134628.0,
      "step": 18480
    },
    {
      "entropy": 6.103436660766602,
      "epoch": 1.8478032688559005,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004662683465027235,
      "loss": 5.8313,
      "mean_token_accuracy": 0.1665041282773018,
      "num_tokens": 36144919.0,
      "step": 18485
    },
    {
      "entropy": 6.083250522613525,
      "epoch": 1.848303093917129,
      "grad_norm": 1.125,
      "learning_rate": 0.00046624954839223165,
      "loss": 5.7356,
      "mean_token_accuracy": 0.168557745218277,
      "num_tokens": 36154081.0,
      "step": 18490
    },
    {
      "entropy": 6.003883600234985,
      "epoch": 1.8488029189783575,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046623074546995024,
      "loss": 5.7159,
      "mean_token_accuracy": 0.17256578356027602,
      "num_tokens": 36164318.0,
      "step": 18495
    },
    {
      "entropy": 6.084289598464966,
      "epoch": 1.8493027440395862,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046621193773635256,
      "loss": 5.8357,
      "mean_token_accuracy": 0.16844997107982634,
      "num_tokens": 36173361.0,
      "step": 18500
    },
    {
      "entropy": 6.026060676574707,
      "epoch": 1.8498025691008149,
      "grad_norm": 1.40625,
      "learning_rate": 0.0004661931251919118,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17317914664745332,
      "num_tokens": 36183133.0,
      "step": 18505
    },
    {
      "entropy": 6.079961919784546,
      "epoch": 1.8503023941620433,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046617430783710105,
      "loss": 5.7123,
      "mean_token_accuracy": 0.17176562249660493,
      "num_tokens": 36193041.0,
      "step": 18510
    },
    {
      "entropy": 6.0608189582824705,
      "epoch": 1.8508022192232718,
      "grad_norm": 1.21875,
      "learning_rate": 0.000466155485672394,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17790914028882981,
      "num_tokens": 36202306.0,
      "step": 18515
    },
    {
      "entropy": 6.071835327148437,
      "epoch": 1.8513020442845005,
      "grad_norm": 1.46875,
      "learning_rate": 0.0004661366586982641,
      "loss": 5.7086,
      "mean_token_accuracy": 0.16952946037054062,
      "num_tokens": 36212342.0,
      "step": 18520
    },
    {
      "entropy": 6.106742095947266,
      "epoch": 1.851801869345729,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004661178269151851,
      "loss": 5.7914,
      "mean_token_accuracy": 0.1648620620369911,
      "num_tokens": 36221181.0,
      "step": 18525
    },
    {
      "entropy": 6.134703159332275,
      "epoch": 1.8523016944069575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046609899032363074,
      "loss": 5.7942,
      "mean_token_accuracy": 0.17102007418870926,
      "num_tokens": 36230813.0,
      "step": 18530
    },
    {
      "entropy": 6.103135681152343,
      "epoch": 1.8528015194681862,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004660801489240749,
      "loss": 5.876,
      "mean_token_accuracy": 0.16575756371021272,
      "num_tokens": 36242315.0,
      "step": 18535
    },
    {
      "entropy": 6.129006624221802,
      "epoch": 1.8533013445294149,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046606130271699175,
      "loss": 5.7499,
      "mean_token_accuracy": 0.17253651171922685,
      "num_tokens": 36253665.0,
      "step": 18540
    },
    {
      "entropy": 6.1301744937896725,
      "epoch": 1.8538011695906431,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004660424517028553,
      "loss": 5.934,
      "mean_token_accuracy": 0.1596855565905571,
      "num_tokens": 36263514.0,
      "step": 18545
    },
    {
      "entropy": 6.052659749984741,
      "epoch": 1.8543009946518718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046602359588214,
      "loss": 5.7784,
      "mean_token_accuracy": 0.16314636170864105,
      "num_tokens": 36273103.0,
      "step": 18550
    },
    {
      "entropy": 6.138967132568359,
      "epoch": 1.8548008197131005,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004660047352553202,
      "loss": 5.7852,
      "mean_token_accuracy": 0.1708177551627159,
      "num_tokens": 36283079.0,
      "step": 18555
    },
    {
      "entropy": 6.172525358200073,
      "epoch": 1.855300644774329,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046598586982287034,
      "loss": 5.7928,
      "mean_token_accuracy": 0.16079241782426834,
      "num_tokens": 36292405.0,
      "step": 18560
    },
    {
      "entropy": 6.068410634994507,
      "epoch": 1.8558004698355575,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046596699958526515,
      "loss": 5.7205,
      "mean_token_accuracy": 0.17415057122707367,
      "num_tokens": 36301516.0,
      "step": 18565
    },
    {
      "entropy": 6.092334651947022,
      "epoch": 1.8563002948967862,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046594812454297944,
      "loss": 5.8532,
      "mean_token_accuracy": 0.16395721435546876,
      "num_tokens": 36311414.0,
      "step": 18570
    },
    {
      "entropy": 6.238265085220337,
      "epoch": 1.8568001199580149,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465929244696488,
      "loss": 5.93,
      "mean_token_accuracy": 0.1524401932954788,
      "num_tokens": 36321521.0,
      "step": 18575
    },
    {
      "entropy": 6.16213436126709,
      "epoch": 1.8572999450192431,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004659103600462659,
      "loss": 5.849,
      "mean_token_accuracy": 0.16322463303804396,
      "num_tokens": 36331115.0,
      "step": 18580
    },
    {
      "entropy": 6.07241907119751,
      "epoch": 1.8577997700804718,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004658914705927883,
      "loss": 5.8547,
      "mean_token_accuracy": 0.16690070927143097,
      "num_tokens": 36341131.0,
      "step": 18585
    },
    {
      "entropy": 6.022227096557617,
      "epoch": 1.8582995951417005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004658725763365304,
      "loss": 5.6996,
      "mean_token_accuracy": 0.17466237992048264,
      "num_tokens": 36350378.0,
      "step": 18590
    },
    {
      "entropy": 6.046438407897949,
      "epoch": 1.858799420202929,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004658536772779676,
      "loss": 5.7458,
      "mean_token_accuracy": 0.16925732642412186,
      "num_tokens": 36359706.0,
      "step": 18595
    },
    {
      "entropy": 6.076794099807739,
      "epoch": 1.8592992452641575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004658347734175754,
      "loss": 5.8235,
      "mean_token_accuracy": 0.16911087185144424,
      "num_tokens": 36370068.0,
      "step": 18600
    },
    {
      "entropy": 6.105840444564819,
      "epoch": 1.8597990703253862,
      "grad_norm": 1.140625,
      "learning_rate": 0.00046581586475582946,
      "loss": 5.768,
      "mean_token_accuracy": 0.16597312390804292,
      "num_tokens": 36380400.0,
      "step": 18605
    },
    {
      "entropy": 6.097484397888183,
      "epoch": 1.8602988953866146,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004657969512932054,
      "loss": 5.781,
      "mean_token_accuracy": 0.16529478132724762,
      "num_tokens": 36390114.0,
      "step": 18610
    },
    {
      "entropy": 6.085397386550904,
      "epoch": 1.8607987204478431,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004657780330301791,
      "loss": 5.7142,
      "mean_token_accuracy": 0.17287175953388215,
      "num_tokens": 36399389.0,
      "step": 18615
    },
    {
      "entropy": 6.097321081161499,
      "epoch": 1.8612985455090718,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046575910996722657,
      "loss": 5.7248,
      "mean_token_accuracy": 0.16774627119302749,
      "num_tokens": 36408512.0,
      "step": 18620
    },
    {
      "entropy": 6.1216953754425045,
      "epoch": 1.8617983705703005,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000465740182104824,
      "loss": 5.9128,
      "mean_token_accuracy": 0.15857245475053788,
      "num_tokens": 36419049.0,
      "step": 18625
    },
    {
      "entropy": 6.070342063903809,
      "epoch": 1.862298195631529,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046572124944344745,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17045060843229293,
      "num_tokens": 36428561.0,
      "step": 18630
    },
    {
      "entropy": 6.096560049057007,
      "epoch": 1.8627980206927575,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046570231198357344,
      "loss": 5.845,
      "mean_token_accuracy": 0.1630481779575348,
      "num_tokens": 36438346.0,
      "step": 18635
    },
    {
      "entropy": 6.10845627784729,
      "epoch": 1.8632978457539862,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004656833697256783,
      "loss": 5.7338,
      "mean_token_accuracy": 0.1745445817708969,
      "num_tokens": 36447756.0,
      "step": 18640
    },
    {
      "entropy": 6.075251960754395,
      "epoch": 1.8637976708152146,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004656644226702386,
      "loss": 5.7395,
      "mean_token_accuracy": 0.17425116747617722,
      "num_tokens": 36457784.0,
      "step": 18645
    },
    {
      "entropy": 5.981507015228272,
      "epoch": 1.8642974958764431,
      "grad_norm": 1.34375,
      "learning_rate": 0.00046564547081773115,
      "loss": 5.7156,
      "mean_token_accuracy": 0.17283906787633896,
      "num_tokens": 36467952.0,
      "step": 18650
    },
    {
      "entropy": 6.193004608154297,
      "epoch": 1.8647973209376718,
      "grad_norm": 1.25,
      "learning_rate": 0.0004656265141686326,
      "loss": 5.9245,
      "mean_token_accuracy": 0.1537487730383873,
      "num_tokens": 36478595.0,
      "step": 18655
    },
    {
      "entropy": 6.0586381435394285,
      "epoch": 1.8652971459989005,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046560755272342014,
      "loss": 5.6508,
      "mean_token_accuracy": 0.1800670802593231,
      "num_tokens": 36487866.0,
      "step": 18660
    },
    {
      "entropy": 6.111362266540527,
      "epoch": 1.865796971060129,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004655885864825706,
      "loss": 5.7607,
      "mean_token_accuracy": 0.1629126936197281,
      "num_tokens": 36497828.0,
      "step": 18665
    },
    {
      "entropy": 6.075566673278809,
      "epoch": 1.8662967961213575,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004655696154465613,
      "loss": 5.7492,
      "mean_token_accuracy": 0.16650566458702087,
      "num_tokens": 36508361.0,
      "step": 18670
    },
    {
      "entropy": 6.049298715591431,
      "epoch": 1.8667966211825862,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046555063961586956,
      "loss": 5.6802,
      "mean_token_accuracy": 0.17370931506156922,
      "num_tokens": 36518458.0,
      "step": 18675
    },
    {
      "entropy": 6.047168207168579,
      "epoch": 1.8672964462438146,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004655316589909727,
      "loss": 5.7561,
      "mean_token_accuracy": 0.16698964387178422,
      "num_tokens": 36527952.0,
      "step": 18680
    },
    {
      "entropy": 6.062076091766357,
      "epoch": 1.867796271305043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004655126735723484,
      "loss": 5.7971,
      "mean_token_accuracy": 0.16839283257722854,
      "num_tokens": 36537231.0,
      "step": 18685
    },
    {
      "entropy": 6.141802597045898,
      "epoch": 1.8682960963662718,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046549368336047415,
      "loss": 5.8468,
      "mean_token_accuracy": 0.16757193952798843,
      "num_tokens": 36546524.0,
      "step": 18690
    },
    {
      "entropy": 6.080959224700928,
      "epoch": 1.8687959214275005,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004654746883558279,
      "loss": 5.7305,
      "mean_token_accuracy": 0.17053497731685638,
      "num_tokens": 36557064.0,
      "step": 18695
    },
    {
      "entropy": 6.028773832321167,
      "epoch": 1.869295746488729,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004654556885588875,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17342813163995743,
      "num_tokens": 36565986.0,
      "step": 18700
    },
    {
      "entropy": 6.128507804870606,
      "epoch": 1.8697955715499575,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046543668397013103,
      "loss": 5.8362,
      "mean_token_accuracy": 0.16342108696699142,
      "num_tokens": 36575646.0,
      "step": 18705
    },
    {
      "entropy": 6.037174797058105,
      "epoch": 1.8702953966111862,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004654176745900365,
      "loss": 5.7318,
      "mean_token_accuracy": 0.17030182033777236,
      "num_tokens": 36585472.0,
      "step": 18710
    },
    {
      "entropy": 6.100129556655884,
      "epoch": 1.8707952216724146,
      "grad_norm": 1.359375,
      "learning_rate": 0.00046539866041908233,
      "loss": 5.7323,
      "mean_token_accuracy": 0.17398363798856736,
      "num_tokens": 36594309.0,
      "step": 18715
    },
    {
      "entropy": 6.126501846313476,
      "epoch": 1.871295046733643,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046537964145774687,
      "loss": 5.8495,
      "mean_token_accuracy": 0.16039469689130784,
      "num_tokens": 36604027.0,
      "step": 18720
    },
    {
      "entropy": 6.078078556060791,
      "epoch": 1.8717948717948718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004653606177065086,
      "loss": 5.7372,
      "mean_token_accuracy": 0.1676819607615471,
      "num_tokens": 36613613.0,
      "step": 18725
    },
    {
      "entropy": 6.124748706817627,
      "epoch": 1.8722946968561005,
      "grad_norm": 1.5703125,
      "learning_rate": 0.0004653415891658462,
      "loss": 5.8102,
      "mean_token_accuracy": 0.17030222862958908,
      "num_tokens": 36624049.0,
      "step": 18730
    },
    {
      "entropy": 6.12185263633728,
      "epoch": 1.872794521917329,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046532255583623837,
      "loss": 5.7799,
      "mean_token_accuracy": 0.16656329482793808,
      "num_tokens": 36634815.0,
      "step": 18735
    },
    {
      "entropy": 6.140682506561279,
      "epoch": 1.8732943469785575,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000465303517718164,
      "loss": 5.8106,
      "mean_token_accuracy": 0.16147739291191102,
      "num_tokens": 36644701.0,
      "step": 18740
    },
    {
      "entropy": 6.033550548553467,
      "epoch": 1.8737941720397862,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004652844748121021,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17370669096708297,
      "num_tokens": 36653581.0,
      "step": 18745
    },
    {
      "entropy": 6.042649221420288,
      "epoch": 1.8742939971010146,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004652654271185318,
      "loss": 5.7928,
      "mean_token_accuracy": 0.1653350293636322,
      "num_tokens": 36662644.0,
      "step": 18750
    },
    {
      "entropy": 6.103317832946777,
      "epoch": 1.874793822162243,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046524637463793233,
      "loss": 5.798,
      "mean_token_accuracy": 0.16485292911529542,
      "num_tokens": 36673260.0,
      "step": 18755
    },
    {
      "entropy": 6.147920370101929,
      "epoch": 1.8752936472234718,
      "grad_norm": 1.1875,
      "learning_rate": 0.000465227317370783,
      "loss": 5.8545,
      "mean_token_accuracy": 0.16623802483081818,
      "num_tokens": 36682899.0,
      "step": 18760
    },
    {
      "entropy": 6.075865650177002,
      "epoch": 1.8757934722847005,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046520825531756326,
      "loss": 5.7828,
      "mean_token_accuracy": 0.1667063608765602,
      "num_tokens": 36693190.0,
      "step": 18765
    },
    {
      "entropy": 6.035709047317505,
      "epoch": 1.876293297345929,
      "grad_norm": 1.25,
      "learning_rate": 0.0004651891884787528,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17097610086202622,
      "num_tokens": 36701263.0,
      "step": 18770
    },
    {
      "entropy": 6.069144153594971,
      "epoch": 1.8767931224071575,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046517011685483136,
      "loss": 5.7853,
      "mean_token_accuracy": 0.16177940219640732,
      "num_tokens": 36710622.0,
      "step": 18775
    },
    {
      "entropy": 6.090852212905884,
      "epoch": 1.8772929474683862,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046515104044627867,
      "loss": 5.7864,
      "mean_token_accuracy": 0.16395394802093505,
      "num_tokens": 36720510.0,
      "step": 18780
    },
    {
      "entropy": 6.030605506896973,
      "epoch": 1.8777927725296146,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004651319592535747,
      "loss": 5.8133,
      "mean_token_accuracy": 0.16223518401384354,
      "num_tokens": 36730570.0,
      "step": 18785
    },
    {
      "entropy": 6.120025396347046,
      "epoch": 1.878292597590843,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004651128732771996,
      "loss": 5.7616,
      "mean_token_accuracy": 0.16895679086446763,
      "num_tokens": 36739834.0,
      "step": 18790
    },
    {
      "entropy": 6.080505084991455,
      "epoch": 1.8787924226520718,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046509378251763346,
      "loss": 5.6523,
      "mean_token_accuracy": 0.17071826308965682,
      "num_tokens": 36750193.0,
      "step": 18795
    },
    {
      "entropy": 6.12220311164856,
      "epoch": 1.8792922477133005,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004650746869753567,
      "loss": 5.772,
      "mean_token_accuracy": 0.1605657160282135,
      "num_tokens": 36759253.0,
      "step": 18800
    },
    {
      "entropy": 6.136216640472412,
      "epoch": 1.8797920727745288,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004650555866508498,
      "loss": 5.8389,
      "mean_token_accuracy": 0.16491588205099106,
      "num_tokens": 36768841.0,
      "step": 18805
    },
    {
      "entropy": 6.093802881240845,
      "epoch": 1.8802918978357575,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046503648154459315,
      "loss": 5.904,
      "mean_token_accuracy": 0.15969534516334533,
      "num_tokens": 36780146.0,
      "step": 18810
    },
    {
      "entropy": 6.149350309371949,
      "epoch": 1.8807917228969862,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004650173716570676,
      "loss": 5.7935,
      "mean_token_accuracy": 0.1646161824464798,
      "num_tokens": 36789948.0,
      "step": 18815
    },
    {
      "entropy": 6.088639926910401,
      "epoch": 1.8812915479582146,
      "grad_norm": 1.375,
      "learning_rate": 0.00046499825698875384,
      "loss": 5.7006,
      "mean_token_accuracy": 0.17286859899759294,
      "num_tokens": 36799496.0,
      "step": 18820
    },
    {
      "entropy": 6.065236139297485,
      "epoch": 1.881791373019443,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046497913754013284,
      "loss": 5.7896,
      "mean_token_accuracy": 0.1652892604470253,
      "num_tokens": 36809905.0,
      "step": 18825
    },
    {
      "entropy": 6.074496936798096,
      "epoch": 1.8822911980806718,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004649600133116856,
      "loss": 5.7672,
      "mean_token_accuracy": 0.16325655877590178,
      "num_tokens": 36819855.0,
      "step": 18830
    },
    {
      "entropy": 6.11353554725647,
      "epoch": 1.8827910231419005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046494088430389335,
      "loss": 5.7943,
      "mean_token_accuracy": 0.16365567594766617,
      "num_tokens": 36830456.0,
      "step": 18835
    },
    {
      "entropy": 5.9790726661682125,
      "epoch": 1.8832908482031288,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004649217505172373,
      "loss": 5.7246,
      "mean_token_accuracy": 0.1732257917523384,
      "num_tokens": 36839914.0,
      "step": 18840
    },
    {
      "entropy": 6.0738489627838135,
      "epoch": 1.8837906732643575,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004649026119521989,
      "loss": 5.6665,
      "mean_token_accuracy": 0.17092276513576507,
      "num_tokens": 36848644.0,
      "step": 18845
    },
    {
      "entropy": 6.0418110370635985,
      "epoch": 1.8842904983255861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046488346860925955,
      "loss": 5.7595,
      "mean_token_accuracy": 0.17318593263626098,
      "num_tokens": 36857490.0,
      "step": 18850
    },
    {
      "entropy": 6.158822107315063,
      "epoch": 1.8847903233868146,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046486432048890104,
      "loss": 5.9476,
      "mean_token_accuracy": 0.15752012878656388,
      "num_tokens": 36867617.0,
      "step": 18855
    },
    {
      "entropy": 6.085655403137207,
      "epoch": 1.885290148448043,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004648451675916051,
      "loss": 5.7951,
      "mean_token_accuracy": 0.16660839021205903,
      "num_tokens": 36876626.0,
      "step": 18860
    },
    {
      "entropy": 6.133948707580567,
      "epoch": 1.8857899735092718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046482600991785365,
      "loss": 5.8324,
      "mean_token_accuracy": 0.1593191958963871,
      "num_tokens": 36887693.0,
      "step": 18865
    },
    {
      "entropy": 6.188600635528564,
      "epoch": 1.8862897985705003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046480684746812856,
      "loss": 5.8352,
      "mean_token_accuracy": 0.1673351749777794,
      "num_tokens": 36897817.0,
      "step": 18870
    },
    {
      "entropy": 6.090949964523316,
      "epoch": 1.8867896236317288,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046478768024291205,
      "loss": 5.7639,
      "mean_token_accuracy": 0.1682679161429405,
      "num_tokens": 36908154.0,
      "step": 18875
    },
    {
      "entropy": 6.0715961933135985,
      "epoch": 1.8872894486929574,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046476850824268633,
      "loss": 5.7626,
      "mean_token_accuracy": 0.1669762521982193,
      "num_tokens": 36917748.0,
      "step": 18880
    },
    {
      "entropy": 6.0552294731140135,
      "epoch": 1.8877892737541861,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004647493314679338,
      "loss": 5.7772,
      "mean_token_accuracy": 0.17492305785417556,
      "num_tokens": 36927128.0,
      "step": 18885
    },
    {
      "entropy": 6.039291477203369,
      "epoch": 1.8882890988154146,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000464730149919137,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17666176557540894,
      "num_tokens": 36936822.0,
      "step": 18890
    },
    {
      "entropy": 6.166332101821899,
      "epoch": 1.888788923876643,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004647109635967783,
      "loss": 5.8238,
      "mean_token_accuracy": 0.16208354383707047,
      "num_tokens": 36946781.0,
      "step": 18895
    },
    {
      "entropy": 6.078222799301147,
      "epoch": 1.8892887489378718,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004646917725013406,
      "loss": 5.7449,
      "mean_token_accuracy": 0.17030898183584214,
      "num_tokens": 36956998.0,
      "step": 18900
    },
    {
      "entropy": 6.105315685272217,
      "epoch": 1.8897885739991003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004646725766333068,
      "loss": 5.7804,
      "mean_token_accuracy": 0.16810195446014403,
      "num_tokens": 36966138.0,
      "step": 18905
    },
    {
      "entropy": 6.039903783798218,
      "epoch": 1.8902883990603288,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046465337599315966,
      "loss": 5.7562,
      "mean_token_accuracy": 0.17383481562137604,
      "num_tokens": 36976282.0,
      "step": 18910
    },
    {
      "entropy": 6.151530170440674,
      "epoch": 1.8907882241215574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046463417058138247,
      "loss": 5.8076,
      "mean_token_accuracy": 0.16685803681612016,
      "num_tokens": 36986795.0,
      "step": 18915
    },
    {
      "entropy": 6.017239570617676,
      "epoch": 1.8912880491827861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004646149603984583,
      "loss": 5.7734,
      "mean_token_accuracy": 0.16728876978158952,
      "num_tokens": 36997334.0,
      "step": 18920
    },
    {
      "entropy": 6.038569355010987,
      "epoch": 1.8917878742440146,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004645957454448706,
      "loss": 5.7026,
      "mean_token_accuracy": 0.17006948590278625,
      "num_tokens": 37006978.0,
      "step": 18925
    },
    {
      "entropy": 6.0481446266174315,
      "epoch": 1.892287699305243,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004645765257211026,
      "loss": 5.7242,
      "mean_token_accuracy": 0.17828518599271775,
      "num_tokens": 37016821.0,
      "step": 18930
    },
    {
      "entropy": 6.051163148880005,
      "epoch": 1.8927875243664718,
      "grad_norm": 1.125,
      "learning_rate": 0.00046455730122763804,
      "loss": 5.7235,
      "mean_token_accuracy": 0.1725027248263359,
      "num_tokens": 37026596.0,
      "step": 18935
    },
    {
      "entropy": 6.071281147003174,
      "epoch": 1.8932873494277003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004645380719649606,
      "loss": 5.753,
      "mean_token_accuracy": 0.16666456311941147,
      "num_tokens": 37035899.0,
      "step": 18940
    },
    {
      "entropy": 6.077476072311401,
      "epoch": 1.8937871744889287,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046451883793355406,
      "loss": 5.8492,
      "mean_token_accuracy": 0.1599900372326374,
      "num_tokens": 37045747.0,
      "step": 18945
    },
    {
      "entropy": 6.088874101638794,
      "epoch": 1.8942869995501574,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046449959913390227,
      "loss": 5.8121,
      "mean_token_accuracy": 0.1613006740808487,
      "num_tokens": 37056673.0,
      "step": 18950
    },
    {
      "entropy": 6.0792169094085695,
      "epoch": 1.8947868246113861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046448035556648936,
      "loss": 5.7857,
      "mean_token_accuracy": 0.16731211990118028,
      "num_tokens": 37065375.0,
      "step": 18955
    },
    {
      "entropy": 6.117296123504639,
      "epoch": 1.8952866496726146,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046446110723179945,
      "loss": 5.7348,
      "mean_token_accuracy": 0.17575096040964128,
      "num_tokens": 37074088.0,
      "step": 18960
    },
    {
      "entropy": 6.024017429351806,
      "epoch": 1.895786474733843,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004644418541303169,
      "loss": 5.7419,
      "mean_token_accuracy": 0.16408663541078566,
      "num_tokens": 37085418.0,
      "step": 18965
    },
    {
      "entropy": 6.117145347595215,
      "epoch": 1.8962862997950718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046442259626252593,
      "loss": 5.8237,
      "mean_token_accuracy": 0.16434903889894487,
      "num_tokens": 37095029.0,
      "step": 18970
    },
    {
      "entropy": 6.1155232906341555,
      "epoch": 1.8967861248563003,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046440333362891123,
      "loss": 5.732,
      "mean_token_accuracy": 0.17899542003870011,
      "num_tokens": 37104398.0,
      "step": 18975
    },
    {
      "entropy": 6.187513732910157,
      "epoch": 1.8972859499175287,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004643840662299574,
      "loss": 5.8447,
      "mean_token_accuracy": 0.16368479281663895,
      "num_tokens": 37114546.0,
      "step": 18980
    },
    {
      "entropy": 6.042089605331421,
      "epoch": 1.8977857749787574,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004643647940661492,
      "loss": 5.7175,
      "mean_token_accuracy": 0.1786101996898651,
      "num_tokens": 37124113.0,
      "step": 18985
    },
    {
      "entropy": 6.0935453414917,
      "epoch": 1.8982856000399861,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046434551713797146,
      "loss": 5.7891,
      "mean_token_accuracy": 0.16713183224201203,
      "num_tokens": 37133661.0,
      "step": 18990
    },
    {
      "entropy": 6.105218839645386,
      "epoch": 1.8987854251012146,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004643262354459093,
      "loss": 5.8121,
      "mean_token_accuracy": 0.16666805893182754,
      "num_tokens": 37143987.0,
      "step": 18995
    },
    {
      "entropy": 6.182411479949951,
      "epoch": 1.899285250162443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046430694899044774,
      "loss": 5.8779,
      "mean_token_accuracy": 0.16336581334471703,
      "num_tokens": 37153227.0,
      "step": 19000
    },
    {
      "entropy": 6.154174709320069,
      "epoch": 1.8997850752236718,
      "grad_norm": 1.21875,
      "learning_rate": 0.000464287657772072,
      "loss": 5.8333,
      "mean_token_accuracy": 0.16556011885404587,
      "num_tokens": 37162524.0,
      "step": 19005
    },
    {
      "entropy": 6.06543869972229,
      "epoch": 1.9002849002849003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004642683617912676,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17216623276472093,
      "num_tokens": 37172729.0,
      "step": 19010
    },
    {
      "entropy": 6.039314031600952,
      "epoch": 1.9007847253461287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046424906104851994,
      "loss": 5.718,
      "mean_token_accuracy": 0.17060794085264205,
      "num_tokens": 37182561.0,
      "step": 19015
    },
    {
      "entropy": 6.141635990142822,
      "epoch": 1.9012845504073574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046422975554431453,
      "loss": 5.8134,
      "mean_token_accuracy": 0.16745396703481674,
      "num_tokens": 37192089.0,
      "step": 19020
    },
    {
      "entropy": 6.0860066413879395,
      "epoch": 1.9017843754685861,
      "grad_norm": 1.25,
      "learning_rate": 0.00046421044527913713,
      "loss": 5.7596,
      "mean_token_accuracy": 0.1681295409798622,
      "num_tokens": 37200739.0,
      "step": 19025
    },
    {
      "entropy": 6.11237940788269,
      "epoch": 1.9022842005298146,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004641911302534737,
      "loss": 5.8063,
      "mean_token_accuracy": 0.16357001662254333,
      "num_tokens": 37209507.0,
      "step": 19030
    },
    {
      "entropy": 6.072798156738282,
      "epoch": 1.902784025591043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046417181046781004,
      "loss": 5.7648,
      "mean_token_accuracy": 0.17092804610729218,
      "num_tokens": 37219252.0,
      "step": 19035
    },
    {
      "entropy": 6.075875186920166,
      "epoch": 1.9032838506522718,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004641524859226324,
      "loss": 5.7191,
      "mean_token_accuracy": 0.17003142535686494,
      "num_tokens": 37229925.0,
      "step": 19040
    },
    {
      "entropy": 6.006577968597412,
      "epoch": 1.9037836757135003,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046413315661842676,
      "loss": 5.7709,
      "mean_token_accuracy": 0.17010821849107743,
      "num_tokens": 37239513.0,
      "step": 19045
    },
    {
      "entropy": 6.093907403945923,
      "epoch": 1.9042835007747287,
      "grad_norm": 1.125,
      "learning_rate": 0.0004641138225556797,
      "loss": 5.7381,
      "mean_token_accuracy": 0.17301941514015198,
      "num_tokens": 37249143.0,
      "step": 19050
    },
    {
      "entropy": 6.126611280441284,
      "epoch": 1.9047833258359574,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004640944837348774,
      "loss": 5.7397,
      "mean_token_accuracy": 0.1657972514629364,
      "num_tokens": 37258834.0,
      "step": 19055
    },
    {
      "entropy": 6.088424158096314,
      "epoch": 1.9052831508971861,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046407514015650667,
      "loss": 5.8048,
      "mean_token_accuracy": 0.1649680331349373,
      "num_tokens": 37269554.0,
      "step": 19060
    },
    {
      "entropy": 6.117116069793701,
      "epoch": 1.9057829759584144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046405579182105393,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16499934643507003,
      "num_tokens": 37278842.0,
      "step": 19065
    },
    {
      "entropy": 6.035621976852417,
      "epoch": 1.906282801019643,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0004640364387290062,
      "loss": 5.7368,
      "mean_token_accuracy": 0.17661378234624864,
      "num_tokens": 37289056.0,
      "step": 19070
    },
    {
      "entropy": 6.041466236114502,
      "epoch": 1.9067826260808718,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046401708088085036,
      "loss": 5.7654,
      "mean_token_accuracy": 0.17235675901174546,
      "num_tokens": 37298823.0,
      "step": 19075
    },
    {
      "entropy": 6.128154611587524,
      "epoch": 1.9072824511421003,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046399771827707335,
      "loss": 5.8947,
      "mean_token_accuracy": 0.16330674067139625,
      "num_tokens": 37310065.0,
      "step": 19080
    },
    {
      "entropy": 6.1373449802398685,
      "epoch": 1.9077822762033287,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046397835091816235,
      "loss": 5.7815,
      "mean_token_accuracy": 0.16305789053440095,
      "num_tokens": 37320856.0,
      "step": 19085
    },
    {
      "entropy": 6.117679166793823,
      "epoch": 1.9082821012645574,
      "grad_norm": 1.046875,
      "learning_rate": 0.00046395897880460474,
      "loss": 5.7914,
      "mean_token_accuracy": 0.17044728845357895,
      "num_tokens": 37330705.0,
      "step": 19090
    },
    {
      "entropy": 6.041719961166382,
      "epoch": 1.9087819263257861,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046393960193688783,
      "loss": 5.7887,
      "mean_token_accuracy": 0.17170997709035873,
      "num_tokens": 37339450.0,
      "step": 19095
    },
    {
      "entropy": 6.013335561752319,
      "epoch": 1.9092817513870144,
      "grad_norm": 1.09375,
      "learning_rate": 0.00046392022031549907,
      "loss": 5.7649,
      "mean_token_accuracy": 0.1751130223274231,
      "num_tokens": 37349173.0,
      "step": 19100
    },
    {
      "entropy": 6.113027095794678,
      "epoch": 1.909781576448243,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004639008339409263,
      "loss": 5.8711,
      "mean_token_accuracy": 0.15661650896072388,
      "num_tokens": 37359246.0,
      "step": 19105
    },
    {
      "entropy": 6.167144727706909,
      "epoch": 1.9102814015094718,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004638814428136571,
      "loss": 5.839,
      "mean_token_accuracy": 0.16727360785007478,
      "num_tokens": 37368056.0,
      "step": 19110
    },
    {
      "entropy": 6.2536163330078125,
      "epoch": 1.9107812265707003,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004638620469341794,
      "loss": 6.0299,
      "mean_token_accuracy": 0.15128510296344758,
      "num_tokens": 37377997.0,
      "step": 19115
    },
    {
      "entropy": 6.076793003082275,
      "epoch": 1.9112810516319287,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004638426463029812,
      "loss": 5.7175,
      "mean_token_accuracy": 0.17027703076601028,
      "num_tokens": 37387365.0,
      "step": 19120
    },
    {
      "entropy": 6.069569301605225,
      "epoch": 1.9117808766931574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004638232409205506,
      "loss": 5.7645,
      "mean_token_accuracy": 0.1706135168671608,
      "num_tokens": 37396649.0,
      "step": 19125
    },
    {
      "entropy": 6.111767292022705,
      "epoch": 1.912280701754386,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046380383078737584,
      "loss": 5.8847,
      "mean_token_accuracy": 0.1548004314303398,
      "num_tokens": 37407576.0,
      "step": 19130
    },
    {
      "entropy": 6.119776773452759,
      "epoch": 1.9127805268156144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046378441590394525,
      "loss": 5.7956,
      "mean_token_accuracy": 0.16432779282331467,
      "num_tokens": 37419270.0,
      "step": 19135
    },
    {
      "entropy": 6.1692718982696535,
      "epoch": 1.913280351876843,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004637649962707474,
      "loss": 5.8315,
      "mean_token_accuracy": 0.15893989354372023,
      "num_tokens": 37429155.0,
      "step": 19140
    },
    {
      "entropy": 6.0986082553863525,
      "epoch": 1.9137801769380718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046374557188827084,
      "loss": 5.8245,
      "mean_token_accuracy": 0.1676712229847908,
      "num_tokens": 37439952.0,
      "step": 19145
    },
    {
      "entropy": 6.105557060241699,
      "epoch": 1.9142800019993003,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004637261427570042,
      "loss": 5.7664,
      "mean_token_accuracy": 0.16891346126794815,
      "num_tokens": 37449537.0,
      "step": 19150
    },
    {
      "entropy": 6.082227325439453,
      "epoch": 1.9147798270605287,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004637067088774364,
      "loss": 5.8031,
      "mean_token_accuracy": 0.1688046395778656,
      "num_tokens": 37460202.0,
      "step": 19155
    },
    {
      "entropy": 6.060595464706421,
      "epoch": 1.9152796521217574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046368727025005634,
      "loss": 5.7382,
      "mean_token_accuracy": 0.17145115584135057,
      "num_tokens": 37469715.0,
      "step": 19160
    },
    {
      "entropy": 6.085369205474853,
      "epoch": 1.915779477182986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046366782687535315,
      "loss": 5.8157,
      "mean_token_accuracy": 0.1623654454946518,
      "num_tokens": 37478495.0,
      "step": 19165
    },
    {
      "entropy": 6.141348552703858,
      "epoch": 1.9162793022442144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000463648378753816,
      "loss": 5.8522,
      "mean_token_accuracy": 0.16692102402448655,
      "num_tokens": 37488125.0,
      "step": 19170
    },
    {
      "entropy": 6.052528238296508,
      "epoch": 1.916779127305443,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046362892588593406,
      "loss": 5.6938,
      "mean_token_accuracy": 0.17493156492710113,
      "num_tokens": 37498097.0,
      "step": 19175
    },
    {
      "entropy": 6.071595907211304,
      "epoch": 1.9172789523666718,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046360946827219703,
      "loss": 5.8315,
      "mean_token_accuracy": 0.16990201622247697,
      "num_tokens": 37507713.0,
      "step": 19180
    },
    {
      "entropy": 6.10126256942749,
      "epoch": 1.9177787774279003,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635900059130942,
      "loss": 5.8645,
      "mean_token_accuracy": 0.16321151852607726,
      "num_tokens": 37517862.0,
      "step": 19185
    },
    {
      "entropy": 6.0855896949768065,
      "epoch": 1.9182786024891287,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004635705388091154,
      "loss": 5.784,
      "mean_token_accuracy": 0.16281318962574004,
      "num_tokens": 37528988.0,
      "step": 19190
    },
    {
      "entropy": 6.123889636993408,
      "epoch": 1.9187784275503574,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004635510669607504,
      "loss": 5.8935,
      "mean_token_accuracy": 0.16121604889631272,
      "num_tokens": 37538467.0,
      "step": 19195
    },
    {
      "entropy": 6.134555768966675,
      "epoch": 1.919278252611586,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000463531590368489,
      "loss": 5.7353,
      "mean_token_accuracy": 0.16187530755996704,
      "num_tokens": 37547986.0,
      "step": 19200
    },
    {
      "entropy": 6.101825904846192,
      "epoch": 1.9197780776728144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046351210903282136,
      "loss": 5.8592,
      "mean_token_accuracy": 0.1670591041445732,
      "num_tokens": 37557403.0,
      "step": 19205
    },
    {
      "entropy": 6.134406423568725,
      "epoch": 1.920277902734043,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046349262295423754,
      "loss": 5.8555,
      "mean_token_accuracy": 0.16069888770580293,
      "num_tokens": 37568504.0,
      "step": 19210
    },
    {
      "entropy": 6.062772655487061,
      "epoch": 1.9207777277952718,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046347313213322786,
      "loss": 5.7636,
      "mean_token_accuracy": 0.17281778156757355,
      "num_tokens": 37577816.0,
      "step": 19215
    },
    {
      "entropy": 6.050261497497559,
      "epoch": 1.9212775528565003,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004634536365702826,
      "loss": 5.7184,
      "mean_token_accuracy": 0.1725157156586647,
      "num_tokens": 37587315.0,
      "step": 19220
    },
    {
      "entropy": 6.193058490753174,
      "epoch": 1.9217773779177287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046343413626589243,
      "loss": 5.8543,
      "mean_token_accuracy": 0.15970131307840346,
      "num_tokens": 37597708.0,
      "step": 19225
    },
    {
      "entropy": 6.057876300811768,
      "epoch": 1.9222772029789574,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004634146312205478,
      "loss": 5.6911,
      "mean_token_accuracy": 0.17779562324285508,
      "num_tokens": 37608093.0,
      "step": 19230
    },
    {
      "entropy": 6.017205905914307,
      "epoch": 1.922777028040186,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046339512143473956,
      "loss": 5.7173,
      "mean_token_accuracy": 0.16652730405330657,
      "num_tokens": 37617742.0,
      "step": 19235
    },
    {
      "entropy": 6.091655206680298,
      "epoch": 1.9232768531014144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046337560690895856,
      "loss": 5.7701,
      "mean_token_accuracy": 0.17033307552337645,
      "num_tokens": 37626999.0,
      "step": 19240
    },
    {
      "entropy": 6.101268243789673,
      "epoch": 1.923776678162643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046335608764369574,
      "loss": 5.8902,
      "mean_token_accuracy": 0.1596827283501625,
      "num_tokens": 37638142.0,
      "step": 19245
    },
    {
      "entropy": 6.0942178726196286,
      "epoch": 1.9242765032238718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046333656363944226,
      "loss": 5.7839,
      "mean_token_accuracy": 0.1665309727191925,
      "num_tokens": 37647373.0,
      "step": 19250
    },
    {
      "entropy": 6.026559448242187,
      "epoch": 1.9247763282851003,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046331703489668925,
      "loss": 5.7332,
      "mean_token_accuracy": 0.17193515747785568,
      "num_tokens": 37658677.0,
      "step": 19255
    },
    {
      "entropy": 6.036960935592651,
      "epoch": 1.9252761533463287,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00046329750141592814,
      "loss": 5.6445,
      "mean_token_accuracy": 0.173639877140522,
      "num_tokens": 37668344.0,
      "step": 19260
    },
    {
      "entropy": 6.037993574142456,
      "epoch": 1.9257759784075574,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004632779631976503,
      "loss": 5.7591,
      "mean_token_accuracy": 0.1612626925110817,
      "num_tokens": 37678000.0,
      "step": 19265
    },
    {
      "entropy": 6.104062128067016,
      "epoch": 1.926275803468786,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046325842024234735,
      "loss": 5.8894,
      "mean_token_accuracy": 0.15996284782886505,
      "num_tokens": 37688404.0,
      "step": 19270
    },
    {
      "entropy": 6.186236000061035,
      "epoch": 1.9267756285300144,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046323887255051105,
      "loss": 5.834,
      "mean_token_accuracy": 0.16601556837558745,
      "num_tokens": 37697699.0,
      "step": 19275
    },
    {
      "entropy": 6.128203916549682,
      "epoch": 1.927275453591243,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00046321932012263317,
      "loss": 5.7183,
      "mean_token_accuracy": 0.16982564330101013,
      "num_tokens": 37707261.0,
      "step": 19280
    },
    {
      "entropy": 6.034725522994995,
      "epoch": 1.9277752786524718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004631997629592055,
      "loss": 5.7077,
      "mean_token_accuracy": 0.1732149377465248,
      "num_tokens": 37716500.0,
      "step": 19285
    },
    {
      "entropy": 5.94683346748352,
      "epoch": 1.9282751037137003,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004631802010607204,
      "loss": 5.6675,
      "mean_token_accuracy": 0.17550969868898392,
      "num_tokens": 37725881.0,
      "step": 19290
    },
    {
      "entropy": 6.003698444366455,
      "epoch": 1.9287749287749287,
      "grad_norm": 1.375,
      "learning_rate": 0.00046316063442766975,
      "loss": 5.7643,
      "mean_token_accuracy": 0.17130634784698487,
      "num_tokens": 37736233.0,
      "step": 19295
    },
    {
      "entropy": 6.118240880966186,
      "epoch": 1.9292747538361574,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046314106306054596,
      "loss": 5.8024,
      "mean_token_accuracy": 0.16201290041208266,
      "num_tokens": 37747230.0,
      "step": 19300
    },
    {
      "entropy": 6.108151483535766,
      "epoch": 1.929774578897386,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046312148695984147,
      "loss": 5.7781,
      "mean_token_accuracy": 0.16905944645404816,
      "num_tokens": 37758470.0,
      "step": 19305
    },
    {
      "entropy": 6.193495512008667,
      "epoch": 1.9302744039586144,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004631019061260488,
      "loss": 5.902,
      "mean_token_accuracy": 0.16572934985160828,
      "num_tokens": 37767912.0,
      "step": 19310
    },
    {
      "entropy": 6.095262813568115,
      "epoch": 1.930774229019843,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004630823205596605,
      "loss": 5.7327,
      "mean_token_accuracy": 0.17267486602067947,
      "num_tokens": 37776847.0,
      "step": 19315
    },
    {
      "entropy": 6.104116201400757,
      "epoch": 1.9312740540810718,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046306273026116944,
      "loss": 5.8301,
      "mean_token_accuracy": 0.16072382628917695,
      "num_tokens": 37786960.0,
      "step": 19320
    },
    {
      "entropy": 6.04428129196167,
      "epoch": 1.9317738791423,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004630431352310685,
      "loss": 5.8073,
      "mean_token_accuracy": 0.16352638304233552,
      "num_tokens": 37797007.0,
      "step": 19325
    },
    {
      "entropy": 6.0544764518737795,
      "epoch": 1.9322737042035287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046302353546985066,
      "loss": 5.7236,
      "mean_token_accuracy": 0.1736273318529129,
      "num_tokens": 37806877.0,
      "step": 19330
    },
    {
      "entropy": 6.091245126724243,
      "epoch": 1.9327735292647574,
      "grad_norm": 1.21875,
      "learning_rate": 0.000463003930978009,
      "loss": 5.7147,
      "mean_token_accuracy": 0.16820840686559677,
      "num_tokens": 37815513.0,
      "step": 19335
    },
    {
      "entropy": 6.044360828399658,
      "epoch": 1.933273354325986,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004629843217560369,
      "loss": 5.7264,
      "mean_token_accuracy": 0.17193932384252547,
      "num_tokens": 37825240.0,
      "step": 19340
    },
    {
      "entropy": 6.142752981185913,
      "epoch": 1.9337731793872144,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004629647078044276,
      "loss": 5.7829,
      "mean_token_accuracy": 0.1690118968486786,
      "num_tokens": 37834333.0,
      "step": 19345
    },
    {
      "entropy": 6.110094308853149,
      "epoch": 1.934273004448443,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004629450891236746,
      "loss": 5.7513,
      "mean_token_accuracy": 0.17380376905202866,
      "num_tokens": 37844548.0,
      "step": 19350
    },
    {
      "entropy": 6.084015226364135,
      "epoch": 1.9347728295096718,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004629254657142715,
      "loss": 5.7454,
      "mean_token_accuracy": 0.1717337504029274,
      "num_tokens": 37853995.0,
      "step": 19355
    },
    {
      "entropy": 6.138976335525513,
      "epoch": 1.9352726545709,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004629058375767121,
      "loss": 5.8263,
      "mean_token_accuracy": 0.16265249401330947,
      "num_tokens": 37863472.0,
      "step": 19360
    },
    {
      "entropy": 6.130089807510376,
      "epoch": 1.9357724796321287,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046288620471149014,
      "loss": 5.777,
      "mean_token_accuracy": 0.17358509451150894,
      "num_tokens": 37872690.0,
      "step": 19365
    },
    {
      "entropy": 6.146674728393554,
      "epoch": 1.9362723046933574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046286656711909966,
      "loss": 5.8249,
      "mean_token_accuracy": 0.16314180791378022,
      "num_tokens": 37882296.0,
      "step": 19370
    },
    {
      "entropy": 6.088172340393067,
      "epoch": 1.936772129754586,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00046284692480003467,
      "loss": 5.7649,
      "mean_token_accuracy": 0.16607498824596406,
      "num_tokens": 37891612.0,
      "step": 19375
    },
    {
      "entropy": 6.097758626937866,
      "epoch": 1.9372719548158144,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046282727775478935,
      "loss": 5.7304,
      "mean_token_accuracy": 0.17140215933322905,
      "num_tokens": 37901107.0,
      "step": 19380
    },
    {
      "entropy": 6.091467189788818,
      "epoch": 1.937771779877043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004628076259838581,
      "loss": 5.9028,
      "mean_token_accuracy": 0.1621468558907509,
      "num_tokens": 37912020.0,
      "step": 19385
    },
    {
      "entropy": 6.037467622756958,
      "epoch": 1.9382716049382716,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046278796948773525,
      "loss": 5.7536,
      "mean_token_accuracy": 0.17437849044799805,
      "num_tokens": 37922681.0,
      "step": 19390
    },
    {
      "entropy": 6.124039649963379,
      "epoch": 1.9387714299995,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046276830826691553,
      "loss": 5.7924,
      "mean_token_accuracy": 0.1664620965719223,
      "num_tokens": 37934322.0,
      "step": 19395
    },
    {
      "entropy": 6.036565446853638,
      "epoch": 1.9392712550607287,
      "grad_norm": 1.0625,
      "learning_rate": 0.00046274864232189334,
      "loss": 5.7667,
      "mean_token_accuracy": 0.17003995478153228,
      "num_tokens": 37945028.0,
      "step": 19400
    },
    {
      "entropy": 6.036530685424805,
      "epoch": 1.9397710801219574,
      "grad_norm": 1.203125,
      "learning_rate": 0.00046272897165316367,
      "loss": 5.6758,
      "mean_token_accuracy": 0.1754847690463066,
      "num_tokens": 37955055.0,
      "step": 19405
    },
    {
      "entropy": 6.131713819503784,
      "epoch": 1.940270905183186,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004627092962612215,
      "loss": 5.8484,
      "mean_token_accuracy": 0.16468506157398224,
      "num_tokens": 37966018.0,
      "step": 19410
    },
    {
      "entropy": 6.098749351501465,
      "epoch": 1.9407707302444144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004626896161465616,
      "loss": 5.7561,
      "mean_token_accuracy": 0.17312566936016083,
      "num_tokens": 37975541.0,
      "step": 19415
    },
    {
      "entropy": 6.1257905006408695,
      "epoch": 1.941270555305643,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046266993130967925,
      "loss": 5.8022,
      "mean_token_accuracy": 0.16003427356481553,
      "num_tokens": 37986312.0,
      "step": 19420
    },
    {
      "entropy": 6.069368553161621,
      "epoch": 1.9417703803668716,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004626502417510698,
      "loss": 5.7794,
      "mean_token_accuracy": 0.1684298425912857,
      "num_tokens": 37994717.0,
      "step": 19425
    },
    {
      "entropy": 6.094864273071289,
      "epoch": 1.9422702054281,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046263054747122857,
      "loss": 5.8606,
      "mean_token_accuracy": 0.16345838755369185,
      "num_tokens": 38004282.0,
      "step": 19430
    },
    {
      "entropy": 6.127002096176147,
      "epoch": 1.9427700304893287,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004626108484706509,
      "loss": 5.7889,
      "mean_token_accuracy": 0.1696387767791748,
      "num_tokens": 38013255.0,
      "step": 19435
    },
    {
      "entropy": 6.137980842590332,
      "epoch": 1.9432698555505574,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00046259114474983266,
      "loss": 5.7839,
      "mean_token_accuracy": 0.16523623019456862,
      "num_tokens": 38023041.0,
      "step": 19440
    },
    {
      "entropy": 6.115823078155517,
      "epoch": 1.943769680611786,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046257143630926953,
      "loss": 5.7663,
      "mean_token_accuracy": 0.16982554346323014,
      "num_tokens": 38032193.0,
      "step": 19445
    },
    {
      "entropy": 6.053232145309448,
      "epoch": 1.9442695056730144,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004625517231494573,
      "loss": 5.7732,
      "mean_token_accuracy": 0.1650831401348114,
      "num_tokens": 38041470.0,
      "step": 19450
    },
    {
      "entropy": 5.935981559753418,
      "epoch": 1.944769330734243,
      "grad_norm": 1.6328125,
      "learning_rate": 0.0004625320052708919,
      "loss": 5.6279,
      "mean_token_accuracy": 0.18212861120700835,
      "num_tokens": 38051997.0,
      "step": 19455
    },
    {
      "entropy": 6.043599557876587,
      "epoch": 1.9452691557954715,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004625122826740696,
      "loss": 5.7811,
      "mean_token_accuracy": 0.17515923380851744,
      "num_tokens": 38061418.0,
      "step": 19460
    },
    {
      "entropy": 6.059353303909302,
      "epoch": 1.9457689808567,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004624925553594865,
      "loss": 5.6931,
      "mean_token_accuracy": 0.1744915008544922,
      "num_tokens": 38070853.0,
      "step": 19465
    },
    {
      "entropy": 6.147614288330078,
      "epoch": 1.9462688059179287,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004624728233276389,
      "loss": 5.8105,
      "mean_token_accuracy": 0.16278544217348098,
      "num_tokens": 38080565.0,
      "step": 19470
    },
    {
      "entropy": 6.10917387008667,
      "epoch": 1.9467686309791574,
      "grad_norm": 1.484375,
      "learning_rate": 0.00046245308657902334,
      "loss": 5.7107,
      "mean_token_accuracy": 0.17540966868400573,
      "num_tokens": 38090899.0,
      "step": 19475
    },
    {
      "entropy": 6.0047252655029295,
      "epoch": 1.947268456040386,
      "grad_norm": 1.59375,
      "learning_rate": 0.0004624333451141364,
      "loss": 5.7209,
      "mean_token_accuracy": 0.17185274809598922,
      "num_tokens": 38100104.0,
      "step": 19480
    },
    {
      "entropy": 6.060470628738403,
      "epoch": 1.9477682811016144,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004624135989334746,
      "loss": 5.7864,
      "mean_token_accuracy": 0.16969332993030548,
      "num_tokens": 38109720.0,
      "step": 19485
    },
    {
      "entropy": 6.076017236709594,
      "epoch": 1.948268106162843,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046239384803753496,
      "loss": 5.8292,
      "mean_token_accuracy": 0.1675061360001564,
      "num_tokens": 38119703.0,
      "step": 19490
    },
    {
      "entropy": 6.097591257095337,
      "epoch": 1.9487679312240715,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046237409242681435,
      "loss": 5.676,
      "mean_token_accuracy": 0.17280753701925278,
      "num_tokens": 38129329.0,
      "step": 19495
    },
    {
      "entropy": 6.15480637550354,
      "epoch": 1.9492677562853,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004623543321018097,
      "loss": 5.746,
      "mean_token_accuracy": 0.16588402092456817,
      "num_tokens": 38138607.0,
      "step": 19500
    },
    {
      "entropy": 6.05337176322937,
      "epoch": 1.9497675813465287,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004623345670630184,
      "loss": 5.7615,
      "mean_token_accuracy": 0.16906457841396333,
      "num_tokens": 38148749.0,
      "step": 19505
    },
    {
      "entropy": 6.12797999382019,
      "epoch": 1.9502674064077574,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004623147973109375,
      "loss": 5.8449,
      "mean_token_accuracy": 0.16413751244544983,
      "num_tokens": 38158929.0,
      "step": 19510
    },
    {
      "entropy": 6.165810775756836,
      "epoch": 1.950767231468986,
      "grad_norm": 1.125,
      "learning_rate": 0.00046229502284606455,
      "loss": 5.7705,
      "mean_token_accuracy": 0.16275450140237807,
      "num_tokens": 38169966.0,
      "step": 19515
    },
    {
      "entropy": 6.1165846824646,
      "epoch": 1.9512670565302144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000462275243668897,
      "loss": 5.8142,
      "mean_token_accuracy": 0.1642487794160843,
      "num_tokens": 38180363.0,
      "step": 19520
    },
    {
      "entropy": 6.088879585266113,
      "epoch": 1.951766881591443,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046225545977993245,
      "loss": 5.7331,
      "mean_token_accuracy": 0.16924408823251724,
      "num_tokens": 38190802.0,
      "step": 19525
    },
    {
      "entropy": 6.100208044052124,
      "epoch": 1.9522667066526715,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004622356711796688,
      "loss": 5.7778,
      "mean_token_accuracy": 0.1664141148328781,
      "num_tokens": 38201183.0,
      "step": 19530
    },
    {
      "entropy": 6.144399213790893,
      "epoch": 1.9527665317139,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004622158778686038,
      "loss": 5.8619,
      "mean_token_accuracy": 0.15717450231313707,
      "num_tokens": 38210953.0,
      "step": 19535
    },
    {
      "entropy": 6.057804918289184,
      "epoch": 1.9532663567751287,
      "grad_norm": 1.8515625,
      "learning_rate": 0.00046219607984723554,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17376583367586135,
      "num_tokens": 38220335.0,
      "step": 19540
    },
    {
      "entropy": 6.079552555084229,
      "epoch": 1.9537661818363574,
      "grad_norm": 1.390625,
      "learning_rate": 0.00046217627711606205,
      "loss": 5.7738,
      "mean_token_accuracy": 0.16597875356674194,
      "num_tokens": 38230530.0,
      "step": 19545
    },
    {
      "entropy": 6.120511722564697,
      "epoch": 1.954266006897586,
      "grad_norm": 1.765625,
      "learning_rate": 0.00046215646967558155,
      "loss": 5.6996,
      "mean_token_accuracy": 0.1737459808588028,
      "num_tokens": 38241275.0,
      "step": 19550
    },
    {
      "entropy": 6.130249547958374,
      "epoch": 1.9547658319588144,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046213665752629246,
      "loss": 5.7239,
      "mean_token_accuracy": 0.17319964170455932,
      "num_tokens": 38250158.0,
      "step": 19555
    },
    {
      "entropy": 6.116637134552002,
      "epoch": 1.955265657020043,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004621168406686933,
      "loss": 5.9533,
      "mean_token_accuracy": 0.16190205216407777,
      "num_tokens": 38259172.0,
      "step": 19560
    },
    {
      "entropy": 6.0760369300842285,
      "epoch": 1.9557654820812715,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046209701910328245,
      "loss": 5.8027,
      "mean_token_accuracy": 0.16568537801504135,
      "num_tokens": 38268261.0,
      "step": 19565
    },
    {
      "entropy": 6.073251056671142,
      "epoch": 1.9562653071425,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004620771928305588,
      "loss": 5.7509,
      "mean_token_accuracy": 0.167454032599926,
      "num_tokens": 38277843.0,
      "step": 19570
    },
    {
      "entropy": 6.027876901626587,
      "epoch": 1.9567651322037287,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004620573618510211,
      "loss": 5.639,
      "mean_token_accuracy": 0.17330751121044158,
      "num_tokens": 38286878.0,
      "step": 19575
    },
    {
      "entropy": 6.062406778335571,
      "epoch": 1.9572649572649574,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046203752616516834,
      "loss": 5.6866,
      "mean_token_accuracy": 0.1700051724910736,
      "num_tokens": 38296289.0,
      "step": 19580
    },
    {
      "entropy": 6.098640966415405,
      "epoch": 1.9577647823261857,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00046201768577349957,
      "loss": 5.8481,
      "mean_token_accuracy": 0.1600051149725914,
      "num_tokens": 38305238.0,
      "step": 19585
    },
    {
      "entropy": 6.041635274887085,
      "epoch": 1.9582646073874144,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004619978406765139,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16940855979919434,
      "num_tokens": 38315079.0,
      "step": 19590
    },
    {
      "entropy": 6.140627145767212,
      "epoch": 1.958764432448643,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004619779908747107,
      "loss": 5.8395,
      "mean_token_accuracy": 0.16493187844753265,
      "num_tokens": 38324314.0,
      "step": 19595
    },
    {
      "entropy": 6.2062671184539795,
      "epoch": 1.9592642575098715,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004619581363685893,
      "loss": 5.84,
      "mean_token_accuracy": 0.16414041668176652,
      "num_tokens": 38334708.0,
      "step": 19600
    },
    {
      "entropy": 6.180793714523316,
      "epoch": 1.9597640825711,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004619382771586494,
      "loss": 5.8071,
      "mean_token_accuracy": 0.1629054255783558,
      "num_tokens": 38344800.0,
      "step": 19605
    },
    {
      "entropy": 5.990603971481323,
      "epoch": 1.9602639076323287,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004619184132453905,
      "loss": 5.7137,
      "mean_token_accuracy": 0.17569042444229127,
      "num_tokens": 38354778.0,
      "step": 19610
    },
    {
      "entropy": 6.115309858322144,
      "epoch": 1.9607637326935574,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004618985446293125,
      "loss": 5.7087,
      "mean_token_accuracy": 0.16678045094013214,
      "num_tokens": 38365032.0,
      "step": 19615
    },
    {
      "entropy": 6.01761646270752,
      "epoch": 1.9612635577547857,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046187867131091504,
      "loss": 5.729,
      "mean_token_accuracy": 0.1689993530511856,
      "num_tokens": 38375051.0,
      "step": 19620
    },
    {
      "entropy": 6.018592023849488,
      "epoch": 1.9617633828160144,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004618587932906984,
      "loss": 5.7402,
      "mean_token_accuracy": 0.16969837993383408,
      "num_tokens": 38384107.0,
      "step": 19625
    },
    {
      "entropy": 5.990976619720459,
      "epoch": 1.962263207877243,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046183891056916256,
      "loss": 5.6158,
      "mean_token_accuracy": 0.18667572289705275,
      "num_tokens": 38394054.0,
      "step": 19630
    },
    {
      "entropy": 6.12450680732727,
      "epoch": 1.9627630329384715,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004618190231468079,
      "loss": 5.7915,
      "mean_token_accuracy": 0.16688738763332367,
      "num_tokens": 38403001.0,
      "step": 19635
    },
    {
      "entropy": 5.969665384292602,
      "epoch": 1.9632628579997,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046179913102413456,
      "loss": 5.6702,
      "mean_token_accuracy": 0.1773045003414154,
      "num_tokens": 38412752.0,
      "step": 19640
    },
    {
      "entropy": 6.188271236419678,
      "epoch": 1.9637626830609287,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004617792342016433,
      "loss": 5.8873,
      "mean_token_accuracy": 0.16569602787494658,
      "num_tokens": 38423523.0,
      "step": 19645
    },
    {
      "entropy": 6.276740407943725,
      "epoch": 1.9642625081221572,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004617593326798344,
      "loss": 5.821,
      "mean_token_accuracy": 0.16507180035114288,
      "num_tokens": 38433062.0,
      "step": 19650
    },
    {
      "entropy": 6.122944736480713,
      "epoch": 1.9647623331833857,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004617394264592089,
      "loss": 5.8988,
      "mean_token_accuracy": 0.15464505851268767,
      "num_tokens": 38443008.0,
      "step": 19655
    },
    {
      "entropy": 6.046458578109741,
      "epoch": 1.9652621582446144,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004617195155402674,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1765754222869873,
      "num_tokens": 38451805.0,
      "step": 19660
    },
    {
      "entropy": 6.161144733428955,
      "epoch": 1.965761983305843,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00046169959992351103,
      "loss": 5.82,
      "mean_token_accuracy": 0.16266558468341827,
      "num_tokens": 38461826.0,
      "step": 19665
    },
    {
      "entropy": 6.078861856460572,
      "epoch": 1.9662618083670715,
      "grad_norm": 1.28125,
      "learning_rate": 0.00046167967960944066,
      "loss": 5.7499,
      "mean_token_accuracy": 0.16985134184360504,
      "num_tokens": 38471746.0,
      "step": 19670
    },
    {
      "entropy": 6.127320909500122,
      "epoch": 1.9667616334283,
      "grad_norm": 1.25,
      "learning_rate": 0.0004616597545985577,
      "loss": 5.722,
      "mean_token_accuracy": 0.17380165457725524,
      "num_tokens": 38480749.0,
      "step": 19675
    },
    {
      "entropy": 6.065641450881958,
      "epoch": 1.9672614584895287,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004616398248913632,
      "loss": 5.786,
      "mean_token_accuracy": 0.16722732335329055,
      "num_tokens": 38491327.0,
      "step": 19680
    },
    {
      "entropy": 6.1317830085754395,
      "epoch": 1.9677612835507572,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046161989048835886,
      "loss": 5.8404,
      "mean_token_accuracy": 0.16861453652381897,
      "num_tokens": 38500573.0,
      "step": 19685
    },
    {
      "entropy": 6.092833852767944,
      "epoch": 1.9682611086119857,
      "grad_norm": 1.296875,
      "learning_rate": 0.00046159995139004594,
      "loss": 5.8249,
      "mean_token_accuracy": 0.16412762701511383,
      "num_tokens": 38510156.0,
      "step": 19690
    },
    {
      "entropy": 6.138033199310303,
      "epoch": 1.9687609336732144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004615800075969264,
      "loss": 5.7571,
      "mean_token_accuracy": 0.17079517841339112,
      "num_tokens": 38519804.0,
      "step": 19695
    },
    {
      "entropy": 6.183056497573853,
      "epoch": 1.969260758734443,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00046156005910950187,
      "loss": 5.8424,
      "mean_token_accuracy": 0.16714582741260528,
      "num_tokens": 38530332.0,
      "step": 19700
    },
    {
      "entropy": 6.165865325927735,
      "epoch": 1.9697605837956715,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004615401059282742,
      "loss": 5.8354,
      "mean_token_accuracy": 0.16495805233716965,
      "num_tokens": 38540480.0,
      "step": 19705
    },
    {
      "entropy": 6.079359483718872,
      "epoch": 1.9702604088569,
      "grad_norm": 1.25,
      "learning_rate": 0.00046152014805374546,
      "loss": 5.7679,
      "mean_token_accuracy": 0.17205318063497543,
      "num_tokens": 38549644.0,
      "step": 19710
    },
    {
      "entropy": 6.117202997207642,
      "epoch": 1.9707602339181287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004615001854864179,
      "loss": 5.8644,
      "mean_token_accuracy": 0.1603493645787239,
      "num_tokens": 38560018.0,
      "step": 19715
    },
    {
      "entropy": 6.1183442115783695,
      "epoch": 1.9712600589793572,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046148021822679347,
      "loss": 5.7009,
      "mean_token_accuracy": 0.1753981053829193,
      "num_tokens": 38569584.0,
      "step": 19720
    },
    {
      "entropy": 6.149140644073486,
      "epoch": 1.9717598840405857,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004614602462753749,
      "loss": 5.8307,
      "mean_token_accuracy": 0.16554740965366363,
      "num_tokens": 38579358.0,
      "step": 19725
    },
    {
      "entropy": 6.04987211227417,
      "epoch": 1.9722597091018144,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004614402696326645,
      "loss": 5.7048,
      "mean_token_accuracy": 0.17232066094875337,
      "num_tokens": 38588850.0,
      "step": 19730
    },
    {
      "entropy": 6.0556724071502686,
      "epoch": 1.972759534163043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046142028829916485,
      "loss": 5.8204,
      "mean_token_accuracy": 0.1696072205901146,
      "num_tokens": 38598972.0,
      "step": 19735
    },
    {
      "entropy": 6.245778369903564,
      "epoch": 1.9732593592242715,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004614003022753787,
      "loss": 5.916,
      "mean_token_accuracy": 0.16266273111104965,
      "num_tokens": 38609640.0,
      "step": 19740
    },
    {
      "entropy": 6.122527647018432,
      "epoch": 1.9737591842855,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004613803115618089,
      "loss": 5.8166,
      "mean_token_accuracy": 0.16626644283533096,
      "num_tokens": 38619487.0,
      "step": 19745
    },
    {
      "entropy": 6.176519298553467,
      "epoch": 1.9742590093467287,
      "grad_norm": 1.3125,
      "learning_rate": 0.00046136031615895846,
      "loss": 5.9143,
      "mean_token_accuracy": 0.15774578005075454,
      "num_tokens": 38629681.0,
      "step": 19750
    },
    {
      "entropy": 6.14497857093811,
      "epoch": 1.9747588344079572,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004613403160673305,
      "loss": 5.8507,
      "mean_token_accuracy": 0.16531630903482436,
      "num_tokens": 38640003.0,
      "step": 19755
    },
    {
      "entropy": 6.105613899230957,
      "epoch": 1.9752586594691857,
      "grad_norm": 1.125,
      "learning_rate": 0.000461320311287428,
      "loss": 5.7357,
      "mean_token_accuracy": 0.1658422216773033,
      "num_tokens": 38650691.0,
      "step": 19760
    },
    {
      "entropy": 6.088154983520508,
      "epoch": 1.9757584845304144,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046130030181975446,
      "loss": 5.7512,
      "mean_token_accuracy": 0.1677144318819046,
      "num_tokens": 38660144.0,
      "step": 19765
    },
    {
      "entropy": 6.086434268951416,
      "epoch": 1.976258309591643,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004612802876648133,
      "loss": 5.8268,
      "mean_token_accuracy": 0.16864923536777496,
      "num_tokens": 38670654.0,
      "step": 19770
    },
    {
      "entropy": 6.094899988174438,
      "epoch": 1.9767581346528715,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000461260268823108,
      "loss": 5.7444,
      "mean_token_accuracy": 0.16600856184959412,
      "num_tokens": 38681616.0,
      "step": 19775
    },
    {
      "entropy": 6.104507160186768,
      "epoch": 1.9772579597141,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046124024529514224,
      "loss": 5.6998,
      "mean_token_accuracy": 0.17901406437158585,
      "num_tokens": 38691586.0,
      "step": 19780
    },
    {
      "entropy": 6.201117324829101,
      "epoch": 1.9777577847753287,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004612202170814198,
      "loss": 5.8778,
      "mean_token_accuracy": 0.15796163380146028,
      "num_tokens": 38703320.0,
      "step": 19785
    },
    {
      "entropy": 6.10558443069458,
      "epoch": 1.9782576098365572,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004612001841824447,
      "loss": 5.7808,
      "mean_token_accuracy": 0.17703987658023834,
      "num_tokens": 38713318.0,
      "step": 19790
    },
    {
      "entropy": 6.079635334014893,
      "epoch": 1.9787574348977857,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00046118014659872085,
      "loss": 5.7935,
      "mean_token_accuracy": 0.1655510976910591,
      "num_tokens": 38723237.0,
      "step": 19795
    },
    {
      "entropy": 6.079729843139648,
      "epoch": 1.9792572599590144,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004611601043307525,
      "loss": 5.7387,
      "mean_token_accuracy": 0.17495503798127174,
      "num_tokens": 38732742.0,
      "step": 19800
    },
    {
      "entropy": 6.071866178512574,
      "epoch": 1.979757085020243,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00046114005737904376,
      "loss": 5.848,
      "mean_token_accuracy": 0.1593129187822342,
      "num_tokens": 38743259.0,
      "step": 19805
    },
    {
      "entropy": 6.139212274551392,
      "epoch": 1.9802569100814715,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000461120005744099,
      "loss": 5.806,
      "mean_token_accuracy": 0.16860250234603882,
      "num_tokens": 38753267.0,
      "step": 19810
    },
    {
      "entropy": 6.067881631851196,
      "epoch": 1.9807567351427,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046109994942642287,
      "loss": 5.6976,
      "mean_token_accuracy": 0.17632412910461426,
      "num_tokens": 38763897.0,
      "step": 19815
    },
    {
      "entropy": 6.075835657119751,
      "epoch": 1.9812565602039287,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004610798884265199,
      "loss": 5.7703,
      "mean_token_accuracy": 0.1623072437942028,
      "num_tokens": 38774864.0,
      "step": 19820
    },
    {
      "entropy": 6.076566743850708,
      "epoch": 1.9817563852651572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046105982274489483,
      "loss": 5.7284,
      "mean_token_accuracy": 0.17439104616641998,
      "num_tokens": 38784726.0,
      "step": 19825
    },
    {
      "entropy": 6.118537187576294,
      "epoch": 1.9822562103263857,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00046103975238205244,
      "loss": 5.8154,
      "mean_token_accuracy": 0.1627314418554306,
      "num_tokens": 38794649.0,
      "step": 19830
    },
    {
      "entropy": 6.163331747055054,
      "epoch": 1.9827560353876144,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004610196773384978,
      "loss": 5.8384,
      "mean_token_accuracy": 0.16460812240839004,
      "num_tokens": 38804567.0,
      "step": 19835
    },
    {
      "entropy": 6.13940200805664,
      "epoch": 1.983255860448843,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046099959761473593,
      "loss": 5.7931,
      "mean_token_accuracy": 0.16774245947599412,
      "num_tokens": 38813874.0,
      "step": 19840
    },
    {
      "entropy": 6.046881484985351,
      "epoch": 1.9837556855100713,
      "grad_norm": 1.25,
      "learning_rate": 0.00046097951321127207,
      "loss": 5.7882,
      "mean_token_accuracy": 0.17014411985874175,
      "num_tokens": 38822110.0,
      "step": 19845
    },
    {
      "entropy": 6.089656162261963,
      "epoch": 1.9842555105713,
      "grad_norm": 1.265625,
      "learning_rate": 0.00046095942412861145,
      "loss": 5.7835,
      "mean_token_accuracy": 0.1754411429166794,
      "num_tokens": 38832628.0,
      "step": 19850
    },
    {
      "entropy": 6.118313598632812,
      "epoch": 1.9847553356325287,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004609393303672596,
      "loss": 5.7039,
      "mean_token_accuracy": 0.17272548973560334,
      "num_tokens": 38841742.0,
      "step": 19855
    },
    {
      "entropy": 6.118927335739135,
      "epoch": 1.9852551606937572,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046091923192772205,
      "loss": 5.7405,
      "mean_token_accuracy": 0.16846216470003128,
      "num_tokens": 38851338.0,
      "step": 19860
    },
    {
      "entropy": 6.043152666091919,
      "epoch": 1.9857549857549857,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004608991288105045,
      "loss": 5.7225,
      "mean_token_accuracy": 0.1690527990460396,
      "num_tokens": 38861308.0,
      "step": 19865
    },
    {
      "entropy": 6.07279405593872,
      "epoch": 1.9862548108162144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046087902101611265,
      "loss": 5.7751,
      "mean_token_accuracy": 0.16783604472875596,
      "num_tokens": 38871911.0,
      "step": 19870
    },
    {
      "entropy": 6.001902341842651,
      "epoch": 1.986754635877443,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004608589085450525,
      "loss": 5.6414,
      "mean_token_accuracy": 0.17456620931625366,
      "num_tokens": 38882330.0,
      "step": 19875
    },
    {
      "entropy": 6.071153259277343,
      "epoch": 1.9872544609386713,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00046083879139783,
      "loss": 5.7115,
      "mean_token_accuracy": 0.1743754267692566,
      "num_tokens": 38892080.0,
      "step": 19880
    },
    {
      "entropy": 6.06404390335083,
      "epoch": 1.9877542859999,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004608186695749514,
      "loss": 5.7055,
      "mean_token_accuracy": 0.17080520540475846,
      "num_tokens": 38901190.0,
      "step": 19885
    },
    {
      "entropy": 6.103047037124634,
      "epoch": 1.9882541110611287,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004607985430769228,
      "loss": 5.9175,
      "mean_token_accuracy": 0.16123861074447632,
      "num_tokens": 38910900.0,
      "step": 19890
    },
    {
      "entropy": 6.076416921615601,
      "epoch": 1.9887539361223572,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004607784119042507,
      "loss": 5.7215,
      "mean_token_accuracy": 0.1716232880949974,
      "num_tokens": 38919988.0,
      "step": 19895
    },
    {
      "entropy": 6.0732134819030765,
      "epoch": 1.9892537611835857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004607582760574416,
      "loss": 5.742,
      "mean_token_accuracy": 0.17206894010305404,
      "num_tokens": 38930288.0,
      "step": 19900
    },
    {
      "entropy": 6.038689708709716,
      "epoch": 1.9897535862448144,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000460738135537002,
      "loss": 5.5809,
      "mean_token_accuracy": 0.1829849511384964,
      "num_tokens": 38938920.0,
      "step": 19905
    },
    {
      "entropy": 6.054684591293335,
      "epoch": 1.9902534113060428,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046071799034343874,
      "loss": 5.7936,
      "mean_token_accuracy": 0.16589105725288392,
      "num_tokens": 38948453.0,
      "step": 19910
    },
    {
      "entropy": 6.179603099822998,
      "epoch": 1.9907532363672713,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004606978404772587,
      "loss": 5.9081,
      "mean_token_accuracy": 0.15770720317959785,
      "num_tokens": 38958421.0,
      "step": 19915
    },
    {
      "entropy": 6.0551900386810305,
      "epoch": 1.9912530614285,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046067768593896865,
      "loss": 5.6875,
      "mean_token_accuracy": 0.16920602023601533,
      "num_tokens": 38967641.0,
      "step": 19920
    },
    {
      "entropy": 6.142995452880859,
      "epoch": 1.9917528864897287,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004606575267290759,
      "loss": 5.7631,
      "mean_token_accuracy": 0.17873106598854066,
      "num_tokens": 38977601.0,
      "step": 19925
    },
    {
      "entropy": 6.00747275352478,
      "epoch": 1.9922527115509572,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004606373628480875,
      "loss": 5.7839,
      "mean_token_accuracy": 0.1697892889380455,
      "num_tokens": 38987398.0,
      "step": 19930
    },
    {
      "entropy": 6.034529066085815,
      "epoch": 1.9927525366121857,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004606171942965108,
      "loss": 5.7455,
      "mean_token_accuracy": 0.17028876692056655,
      "num_tokens": 38997831.0,
      "step": 19935
    },
    {
      "entropy": 6.124624872207642,
      "epoch": 1.9932523616734144,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004605970210748533,
      "loss": 5.897,
      "mean_token_accuracy": 0.16648804843425752,
      "num_tokens": 39007667.0,
      "step": 19940
    },
    {
      "entropy": 6.091839027404785,
      "epoch": 1.9937521867346428,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004605768431836225,
      "loss": 5.7033,
      "mean_token_accuracy": 0.17937210351228713,
      "num_tokens": 39018192.0,
      "step": 19945
    },
    {
      "entropy": 6.087166023254395,
      "epoch": 1.9942520117958713,
      "grad_norm": 1.21875,
      "learning_rate": 0.00046055666062332605,
      "loss": 5.6988,
      "mean_token_accuracy": 0.17755381911993026,
      "num_tokens": 39027440.0,
      "step": 19950
    },
    {
      "entropy": 5.9695531845092775,
      "epoch": 1.9947518368571,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00046053647339447184,
      "loss": 5.6387,
      "mean_token_accuracy": 0.1793017014861107,
      "num_tokens": 39037078.0,
      "step": 19955
    },
    {
      "entropy": 6.067123937606811,
      "epoch": 1.9952516619183287,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046051628149756767,
      "loss": 5.7946,
      "mean_token_accuracy": 0.1627612829208374,
      "num_tokens": 39046462.0,
      "step": 19960
    },
    {
      "entropy": 6.147229480743408,
      "epoch": 1.9957514869795572,
      "grad_norm": 1.515625,
      "learning_rate": 0.0004604960849331216,
      "loss": 5.7485,
      "mean_token_accuracy": 0.1696407288312912,
      "num_tokens": 39057004.0,
      "step": 19965
    },
    {
      "entropy": 6.160590553283692,
      "epoch": 1.9962513120407857,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004604758837016417,
      "loss": 5.8509,
      "mean_token_accuracy": 0.1670793414115906,
      "num_tokens": 39066612.0,
      "step": 19970
    },
    {
      "entropy": 6.094274997711182,
      "epoch": 1.9967511371020144,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004604556778036363,
      "loss": 5.8169,
      "mean_token_accuracy": 0.16017839908599854,
      "num_tokens": 39076702.0,
      "step": 19975
    },
    {
      "entropy": 5.988117361068726,
      "epoch": 1.9972509621632428,
      "grad_norm": 1.1875,
      "learning_rate": 0.00046043546723961384,
      "loss": 5.6355,
      "mean_token_accuracy": 0.18264233618974685,
      "num_tokens": 39087110.0,
      "step": 19980
    },
    {
      "entropy": 6.06049656867981,
      "epoch": 1.9977507872244713,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00046041525201008266,
      "loss": 5.7887,
      "mean_token_accuracy": 0.1631213590502739,
      "num_tokens": 39097204.0,
      "step": 19985
    },
    {
      "entropy": 6.058638858795166,
      "epoch": 1.9982506122857,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046039503211555154,
      "loss": 5.7362,
      "mean_token_accuracy": 0.17046991139650344,
      "num_tokens": 39106469.0,
      "step": 19990
    },
    {
      "entropy": 6.087457847595215,
      "epoch": 1.9987504373469287,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046037480755652913,
      "loss": 5.7535,
      "mean_token_accuracy": 0.17235818952322007,
      "num_tokens": 39116304.0,
      "step": 19995
    },
    {
      "entropy": 6.098694944381714,
      "epoch": 1.9992502624081572,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046035457833352414,
      "loss": 5.8235,
      "mean_token_accuracy": 0.15795450061559677,
      "num_tokens": 39125573.0,
      "step": 20000
    },
    {
      "entropy": 6.10334358215332,
      "epoch": 1.9997500874693857,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00046033434444704577,
      "loss": 5.7384,
      "mean_token_accuracy": 0.17256668508052825,
      "num_tokens": 39134876.0,
      "step": 20005
    },
    {
      "entropy": 6.07966152826945,
      "epoch": 2.0001999300244915,
      "grad_norm": 1.109375,
      "learning_rate": 0.00046031410589760294,
      "loss": 5.6833,
      "mean_token_accuracy": 0.17534273697270286,
      "num_tokens": 39143851.0,
      "step": 20010
    },
    {
      "entropy": 5.9369086742401125,
      "epoch": 2.00069975508572,
      "grad_norm": 1.234375,
      "learning_rate": 0.00046029386268570485,
      "loss": 5.5608,
      "mean_token_accuracy": 0.1846308782696724,
      "num_tokens": 39152748.0,
      "step": 20015
    },
    {
      "entropy": 6.044854688644409,
      "epoch": 2.0011995801469484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004602736148118609,
      "loss": 5.7063,
      "mean_token_accuracy": 0.164373679459095,
      "num_tokens": 39162097.0,
      "step": 20020
    },
    {
      "entropy": 6.060443258285522,
      "epoch": 2.001699405208177,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046025336227658047,
      "loss": 5.7107,
      "mean_token_accuracy": 0.1759570062160492,
      "num_tokens": 39171629.0,
      "step": 20025
    },
    {
      "entropy": 6.197247076034546,
      "epoch": 2.002199230269406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00046023310508037316,
      "loss": 5.8245,
      "mean_token_accuracy": 0.15730675905942917,
      "num_tokens": 39181368.0,
      "step": 20030
    },
    {
      "entropy": 6.053267621994019,
      "epoch": 2.002699055330634,
      "grad_norm": 1.4921875,
      "learning_rate": 0.00046021284322374847,
      "loss": 5.6174,
      "mean_token_accuracy": 0.1808698907494545,
      "num_tokens": 39190492.0,
      "step": 20035
    },
    {
      "entropy": 6.103080463409424,
      "epoch": 2.003198880391863,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004601925767072164,
      "loss": 5.7196,
      "mean_token_accuracy": 0.16754649132490157,
      "num_tokens": 39200915.0,
      "step": 20040
    },
    {
      "entropy": 6.083680534362793,
      "epoch": 2.0036987054530915,
      "grad_norm": 1.4296875,
      "learning_rate": 0.00046017230553128675,
      "loss": 5.6521,
      "mean_token_accuracy": 0.17290466278791428,
      "num_tokens": 39210370.0,
      "step": 20045
    },
    {
      "entropy": 6.167730808258057,
      "epoch": 2.00419853051432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00046015202969646953,
      "loss": 5.7501,
      "mean_token_accuracy": 0.1718434602022171,
      "num_tokens": 39221731.0,
      "step": 20050
    },
    {
      "entropy": 6.210087442398072,
      "epoch": 2.0046983555755484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004601317492032749,
      "loss": 5.7194,
      "mean_token_accuracy": 0.1693051241338253,
      "num_tokens": 39232951.0,
      "step": 20055
    },
    {
      "entropy": 6.136258745193482,
      "epoch": 2.005198180636777,
      "grad_norm": 1.09375,
      "learning_rate": 0.000460111464052213,
      "loss": 5.7497,
      "mean_token_accuracy": 0.17018657624721528,
      "num_tokens": 39242433.0,
      "step": 20060
    },
    {
      "entropy": 6.090251541137695,
      "epoch": 2.005698005698006,
      "grad_norm": 1.171875,
      "learning_rate": 0.00046009117424379445,
      "loss": 5.7346,
      "mean_token_accuracy": 0.16759042739868163,
      "num_tokens": 39253496.0,
      "step": 20065
    },
    {
      "entropy": 6.044763469696045,
      "epoch": 2.006197830759234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00046007087977852947,
      "loss": 5.6881,
      "mean_token_accuracy": 0.17675623744726182,
      "num_tokens": 39263108.0,
      "step": 20070
    },
    {
      "entropy": 6.0318274974823,
      "epoch": 2.0066976558204628,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00046005058065692883,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1760835811495781,
      "num_tokens": 39272331.0,
      "step": 20075
    },
    {
      "entropy": 6.106259775161743,
      "epoch": 2.0071974808816915,
      "grad_norm": 1.15625,
      "learning_rate": 0.00046003027687950316,
      "loss": 5.6893,
      "mean_token_accuracy": 0.1708327889442444,
      "num_tokens": 39282387.0,
      "step": 20080
    },
    {
      "entropy": 6.064089441299439,
      "epoch": 2.00769730594292,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004600099684467634,
      "loss": 5.6986,
      "mean_token_accuracy": 0.17566324919462203,
      "num_tokens": 39293212.0,
      "step": 20085
    },
    {
      "entropy": 6.135978364944458,
      "epoch": 2.0081971310041484,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004599896553592204,
      "loss": 5.7901,
      "mean_token_accuracy": 0.1670550897717476,
      "num_tokens": 39302728.0,
      "step": 20090
    },
    {
      "entropy": 6.077429533004761,
      "epoch": 2.008696956065377,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004599693376173853,
      "loss": 5.644,
      "mean_token_accuracy": 0.1726323589682579,
      "num_tokens": 39313079.0,
      "step": 20095
    },
    {
      "entropy": 6.1456183910369875,
      "epoch": 2.009196781126606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004599490152217692,
      "loss": 5.7739,
      "mean_token_accuracy": 0.1617901310324669,
      "num_tokens": 39323563.0,
      "step": 20100
    },
    {
      "entropy": 6.075893926620483,
      "epoch": 2.009696606187834,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045992868817288355,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1734946548938751,
      "num_tokens": 39334541.0,
      "step": 20105
    },
    {
      "entropy": 6.112001514434814,
      "epoch": 2.0101964312490628,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004599083564712396,
      "loss": 5.7168,
      "mean_token_accuracy": 0.1722011536359787,
      "num_tokens": 39344848.0,
      "step": 20110
    },
    {
      "entropy": 6.111793804168701,
      "epoch": 2.0106962563102915,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045988802011734896,
      "loss": 5.6985,
      "mean_token_accuracy": 0.1607961818575859,
      "num_tokens": 39354830.0,
      "step": 20115
    },
    {
      "entropy": 6.141810035705566,
      "epoch": 2.01119608137152,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045986767911172343,
      "loss": 5.7622,
      "mean_token_accuracy": 0.1690826505422592,
      "num_tokens": 39363838.0,
      "step": 20120
    },
    {
      "entropy": 6.077844095230103,
      "epoch": 2.0116959064327484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045984733345487465,
      "loss": 5.6629,
      "mean_token_accuracy": 0.16918161362409592,
      "num_tokens": 39373549.0,
      "step": 20125
    },
    {
      "entropy": 6.097675895690918,
      "epoch": 2.012195731493977,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045982698314731447,
      "loss": 5.6599,
      "mean_token_accuracy": 0.17269637882709504,
      "num_tokens": 39383747.0,
      "step": 20130
    },
    {
      "entropy": 6.076425790786743,
      "epoch": 2.012695556555206,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045980662818955495,
      "loss": 5.7331,
      "mean_token_accuracy": 0.1705927088856697,
      "num_tokens": 39394283.0,
      "step": 20135
    },
    {
      "entropy": 6.098881340026855,
      "epoch": 2.013195381616434,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045978626858210833,
      "loss": 5.6155,
      "mean_token_accuracy": 0.179286889731884,
      "num_tokens": 39404777.0,
      "step": 20140
    },
    {
      "entropy": 6.064482641220093,
      "epoch": 2.0136952066776628,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004597659043254866,
      "loss": 5.6014,
      "mean_token_accuracy": 0.17342885434627534,
      "num_tokens": 39414749.0,
      "step": 20145
    },
    {
      "entropy": 6.0715758323669435,
      "epoch": 2.0141950317388915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004597455354202023,
      "loss": 5.6823,
      "mean_token_accuracy": 0.17355004101991653,
      "num_tokens": 39424185.0,
      "step": 20150
    },
    {
      "entropy": 6.068203067779541,
      "epoch": 2.01469485680012,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004597251618667679,
      "loss": 5.7475,
      "mean_token_accuracy": 0.17275481671094894,
      "num_tokens": 39433442.0,
      "step": 20155
    },
    {
      "entropy": 6.062960910797119,
      "epoch": 2.0151946818613484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045970478366569594,
      "loss": 5.7048,
      "mean_token_accuracy": 0.17705900967121124,
      "num_tokens": 39443373.0,
      "step": 20160
    },
    {
      "entropy": 6.13445348739624,
      "epoch": 2.015694506922577,
      "grad_norm": 1.25,
      "learning_rate": 0.00045968440081749917,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17833202481269836,
      "num_tokens": 39452293.0,
      "step": 20165
    },
    {
      "entropy": 6.0796795845031735,
      "epoch": 2.016194331983806,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004596640133226904,
      "loss": 5.6671,
      "mean_token_accuracy": 0.17232218235731125,
      "num_tokens": 39463063.0,
      "step": 20170
    },
    {
      "entropy": 6.071743392944336,
      "epoch": 2.016694157045034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045964362118178253,
      "loss": 5.6806,
      "mean_token_accuracy": 0.1720048189163208,
      "num_tokens": 39473349.0,
      "step": 20175
    },
    {
      "entropy": 6.085145044326782,
      "epoch": 2.0171939821062628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004596232243952887,
      "loss": 5.6813,
      "mean_token_accuracy": 0.1641944542527199,
      "num_tokens": 39485106.0,
      "step": 20180
    },
    {
      "entropy": 6.0731587409973145,
      "epoch": 2.0176938071674915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.000459602822963722,
      "loss": 5.7663,
      "mean_token_accuracy": 0.1687481477856636,
      "num_tokens": 39494785.0,
      "step": 20185
    },
    {
      "entropy": 5.9859678745269775,
      "epoch": 2.01819363222872,
      "grad_norm": 1.5078125,
      "learning_rate": 0.00045958241688759583,
      "loss": 5.5661,
      "mean_token_accuracy": 0.18221739828586578,
      "num_tokens": 39503846.0,
      "step": 20190
    },
    {
      "entropy": 6.098060655593872,
      "epoch": 2.0186934572899484,
      "grad_norm": 1.125,
      "learning_rate": 0.00045956200616742356,
      "loss": 5.7405,
      "mean_token_accuracy": 0.16870652586221696,
      "num_tokens": 39515207.0,
      "step": 20195
    },
    {
      "entropy": 6.080890750885009,
      "epoch": 2.019193282351177,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004595415908037187,
      "loss": 5.6704,
      "mean_token_accuracy": 0.1687462255358696,
      "num_tokens": 39524698.0,
      "step": 20200
    },
    {
      "entropy": 6.061633777618408,
      "epoch": 2.019693107412406,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004595211707969948,
      "loss": 5.6096,
      "mean_token_accuracy": 0.18250036090612412,
      "num_tokens": 39534485.0,
      "step": 20205
    },
    {
      "entropy": 6.044603872299194,
      "epoch": 2.020192932473634,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004595007461477658,
      "loss": 5.6204,
      "mean_token_accuracy": 0.18820348232984543,
      "num_tokens": 39544039.0,
      "step": 20210
    },
    {
      "entropy": 6.050969362258911,
      "epoch": 2.0206927575348628,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045948031685654557,
      "loss": 5.6757,
      "mean_token_accuracy": 0.17225667983293533,
      "num_tokens": 39553139.0,
      "step": 20215
    },
    {
      "entropy": 6.127994871139526,
      "epoch": 2.0211925825960915,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004594598829238479,
      "loss": 5.7273,
      "mean_token_accuracy": 0.16636809706687927,
      "num_tokens": 39562385.0,
      "step": 20220
    },
    {
      "entropy": 6.051290798187256,
      "epoch": 2.0216924076573197,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004594394443501872,
      "loss": 5.6203,
      "mean_token_accuracy": 0.18062713295221328,
      "num_tokens": 39571855.0,
      "step": 20225
    },
    {
      "entropy": 6.049437141418457,
      "epoch": 2.0221922327185484,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004594190011360775,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17510585635900497,
      "num_tokens": 39581230.0,
      "step": 20230
    },
    {
      "entropy": 6.140601778030396,
      "epoch": 2.022692057779777,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004593985532820331,
      "loss": 5.7482,
      "mean_token_accuracy": 0.16142921894788742,
      "num_tokens": 39590730.0,
      "step": 20235
    },
    {
      "entropy": 6.10253529548645,
      "epoch": 2.023191882841006,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004593781007885686,
      "loss": 5.6854,
      "mean_token_accuracy": 0.1811680018901825,
      "num_tokens": 39600110.0,
      "step": 20240
    },
    {
      "entropy": 6.097172355651855,
      "epoch": 2.023691707902234,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004593576436561985,
      "loss": 5.7687,
      "mean_token_accuracy": 0.16142215430736542,
      "num_tokens": 39610379.0,
      "step": 20245
    },
    {
      "entropy": 6.06792197227478,
      "epoch": 2.0241915329634628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045933718188543763,
      "loss": 5.7312,
      "mean_token_accuracy": 0.1643009901046753,
      "num_tokens": 39619857.0,
      "step": 20250
    },
    {
      "entropy": 6.089860916137695,
      "epoch": 2.0246913580246915,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004593167154768006,
      "loss": 5.6547,
      "mean_token_accuracy": 0.1776014819741249,
      "num_tokens": 39630049.0,
      "step": 20255
    },
    {
      "entropy": 6.103456592559814,
      "epoch": 2.0251911830859197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004592962444308024,
      "loss": 5.7334,
      "mean_token_accuracy": 0.16641913950443268,
      "num_tokens": 39638794.0,
      "step": 20260
    },
    {
      "entropy": 6.075905466079712,
      "epoch": 2.0256910081471484,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004592757687479582,
      "loss": 5.5968,
      "mean_token_accuracy": 0.17797008007764817,
      "num_tokens": 39648261.0,
      "step": 20265
    },
    {
      "entropy": 6.028750467300415,
      "epoch": 2.026190833208377,
      "grad_norm": 1.15625,
      "learning_rate": 0.000459255288428783,
      "loss": 5.5731,
      "mean_token_accuracy": 0.18388448506593705,
      "num_tokens": 39659217.0,
      "step": 20270
    },
    {
      "entropy": 5.884848642349243,
      "epoch": 2.026690658269606,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004592348034737922,
      "loss": 5.4446,
      "mean_token_accuracy": 0.1976071774959564,
      "num_tokens": 39669535.0,
      "step": 20275
    },
    {
      "entropy": 6.093022775650025,
      "epoch": 2.027190483330834,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004592143138835012,
      "loss": 5.7302,
      "mean_token_accuracy": 0.16700634509325027,
      "num_tokens": 39678640.0,
      "step": 20280
    },
    {
      "entropy": 6.0413614273071286,
      "epoch": 2.0276903083920628,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004591938196584254,
      "loss": 5.6573,
      "mean_token_accuracy": 0.17724614292383195,
      "num_tokens": 39688000.0,
      "step": 20285
    },
    {
      "entropy": 6.063565349578857,
      "epoch": 2.0281901334532915,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045917332079908055,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17249096930027008,
      "num_tokens": 39697778.0,
      "step": 20290
    },
    {
      "entropy": 6.024125480651856,
      "epoch": 2.0286899585145197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004591528173059823,
      "loss": 5.5859,
      "mean_token_accuracy": 0.17791785448789596,
      "num_tokens": 39707842.0,
      "step": 20295
    },
    {
      "entropy": 6.093098831176758,
      "epoch": 2.0291897835757484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004591323091796465,
      "loss": 5.7263,
      "mean_token_accuracy": 0.16548821777105333,
      "num_tokens": 39716465.0,
      "step": 20300
    },
    {
      "entropy": 6.062314224243164,
      "epoch": 2.029689608636977,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004591117964205893,
      "loss": 5.6093,
      "mean_token_accuracy": 0.17624322026968003,
      "num_tokens": 39725862.0,
      "step": 20305
    },
    {
      "entropy": 6.005319499969483,
      "epoch": 2.030189433698206,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045909127902932657,
      "loss": 5.7069,
      "mean_token_accuracy": 0.16486865431070327,
      "num_tokens": 39737042.0,
      "step": 20310
    },
    {
      "entropy": 6.070501899719238,
      "epoch": 2.030689258759434,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004590707570063747,
      "loss": 5.6758,
      "mean_token_accuracy": 0.1676729291677475,
      "num_tokens": 39747849.0,
      "step": 20315
    },
    {
      "entropy": 6.089812278747559,
      "epoch": 2.0311890838206628,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045905023035224993,
      "loss": 5.6628,
      "mean_token_accuracy": 0.17013980001211165,
      "num_tokens": 39757544.0,
      "step": 20320
    },
    {
      "entropy": 6.048898983001709,
      "epoch": 2.0316889088818915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045902969906746875,
      "loss": 5.6062,
      "mean_token_accuracy": 0.17718302309513093,
      "num_tokens": 39766127.0,
      "step": 20325
    },
    {
      "entropy": 6.109204912185669,
      "epoch": 2.0321887339431197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045900916315254765,
      "loss": 5.7584,
      "mean_token_accuracy": 0.1678036108613014,
      "num_tokens": 39775711.0,
      "step": 20330
    },
    {
      "entropy": 6.112554836273193,
      "epoch": 2.0326885590043484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045898862260800337,
      "loss": 5.7691,
      "mean_token_accuracy": 0.1748104766011238,
      "num_tokens": 39786132.0,
      "step": 20335
    },
    {
      "entropy": 6.1353082180023195,
      "epoch": 2.033188384065577,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045896807743435263,
      "loss": 5.754,
      "mean_token_accuracy": 0.15817032307386397,
      "num_tokens": 39796196.0,
      "step": 20340
    },
    {
      "entropy": 6.099129962921142,
      "epoch": 2.033688209126806,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004589475276321125,
      "loss": 5.6337,
      "mean_token_accuracy": 0.17562458515167237,
      "num_tokens": 39805912.0,
      "step": 20345
    },
    {
      "entropy": 6.051821613311768,
      "epoch": 2.034188034188034,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045892697320179987,
      "loss": 5.6072,
      "mean_token_accuracy": 0.18414118736982346,
      "num_tokens": 39814990.0,
      "step": 20350
    },
    {
      "entropy": 6.000114870071411,
      "epoch": 2.0346878592492628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004589064141439319,
      "loss": 5.6977,
      "mean_token_accuracy": 0.17812246233224868,
      "num_tokens": 39823877.0,
      "step": 20355
    },
    {
      "entropy": 5.956132793426514,
      "epoch": 2.0351876843104915,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004588858504590259,
      "loss": 5.5648,
      "mean_token_accuracy": 0.1835416540503502,
      "num_tokens": 39832784.0,
      "step": 20360
    },
    {
      "entropy": 6.116368389129638,
      "epoch": 2.0356875093717197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004588652821475992,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1691375583410263,
      "num_tokens": 39842388.0,
      "step": 20365
    },
    {
      "entropy": 6.052047777175903,
      "epoch": 2.0361873344329484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045884470921016935,
      "loss": 5.6353,
      "mean_token_accuracy": 0.18222550451755523,
      "num_tokens": 39851657.0,
      "step": 20370
    },
    {
      "entropy": 6.099119186401367,
      "epoch": 2.036687159494177,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045882413164725395,
      "loss": 5.71,
      "mean_token_accuracy": 0.17035361528396606,
      "num_tokens": 39861941.0,
      "step": 20375
    },
    {
      "entropy": 6.033570241928101,
      "epoch": 2.037186984555406,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045880354945937056,
      "loss": 5.6375,
      "mean_token_accuracy": 0.17889701277017594,
      "num_tokens": 39872000.0,
      "step": 20380
    },
    {
      "entropy": 6.063420724868775,
      "epoch": 2.037686809616634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004587829626470373,
      "loss": 5.6991,
      "mean_token_accuracy": 0.1693100854754448,
      "num_tokens": 39882281.0,
      "step": 20385
    },
    {
      "entropy": 6.086519765853882,
      "epoch": 2.0381866346778628,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045876237121077196,
      "loss": 5.698,
      "mean_token_accuracy": 0.17771842032670976,
      "num_tokens": 39892445.0,
      "step": 20390
    },
    {
      "entropy": 6.054373359680175,
      "epoch": 2.0386864597390915,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004587417751510926,
      "loss": 5.6419,
      "mean_token_accuracy": 0.1791992887854576,
      "num_tokens": 39902077.0,
      "step": 20395
    },
    {
      "entropy": 6.0784351348876955,
      "epoch": 2.0391862848003197,
      "grad_norm": 1.25,
      "learning_rate": 0.00045872117446851753,
      "loss": 5.726,
      "mean_token_accuracy": 0.16811724305152892,
      "num_tokens": 39912930.0,
      "step": 20400
    },
    {
      "entropy": 6.08008508682251,
      "epoch": 2.0396861098615484,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004587005691635649,
      "loss": 5.649,
      "mean_token_accuracy": 0.1756546229124069,
      "num_tokens": 39921884.0,
      "step": 20405
    },
    {
      "entropy": 6.153205537796021,
      "epoch": 2.040185934922777,
      "grad_norm": 1.125,
      "learning_rate": 0.00045867995923675326,
      "loss": 5.7405,
      "mean_token_accuracy": 0.1669187918305397,
      "num_tokens": 39931589.0,
      "step": 20410
    },
    {
      "entropy": 6.148994731903076,
      "epoch": 2.040685759984006,
      "grad_norm": 1.25,
      "learning_rate": 0.0004586593446886011,
      "loss": 5.7655,
      "mean_token_accuracy": 0.16430536061525344,
      "num_tokens": 39941624.0,
      "step": 20415
    },
    {
      "entropy": 6.049576616287231,
      "epoch": 2.041185585045234,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004586387255196271,
      "loss": 5.6356,
      "mean_token_accuracy": 0.1761637657880783,
      "num_tokens": 39951541.0,
      "step": 20420
    },
    {
      "entropy": 6.088350582122803,
      "epoch": 2.0416854101064628,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045861810173035,
      "loss": 5.719,
      "mean_token_accuracy": 0.1736340641975403,
      "num_tokens": 39962729.0,
      "step": 20425
    },
    {
      "entropy": 6.116737270355225,
      "epoch": 2.0421852351676915,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045859747332128883,
      "loss": 5.7186,
      "mean_token_accuracy": 0.17478419095277786,
      "num_tokens": 39972193.0,
      "step": 20430
    },
    {
      "entropy": 6.080168676376343,
      "epoch": 2.0426850602289197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004585768402929623,
      "loss": 5.6938,
      "mean_token_accuracy": 0.1645065277814865,
      "num_tokens": 39980813.0,
      "step": 20435
    },
    {
      "entropy": 6.050087308883667,
      "epoch": 2.0431848852901484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004585562026458898,
      "loss": 5.6639,
      "mean_token_accuracy": 0.18184519857168197,
      "num_tokens": 39989147.0,
      "step": 20440
    },
    {
      "entropy": 6.016243600845337,
      "epoch": 2.043684710351377,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045853556038059044,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1791103884577751,
      "num_tokens": 40000091.0,
      "step": 20445
    },
    {
      "entropy": 6.150309991836548,
      "epoch": 2.0441845354126054,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045851491349758366,
      "loss": 5.7664,
      "mean_token_accuracy": 0.17039090096950532,
      "num_tokens": 40009443.0,
      "step": 20450
    },
    {
      "entropy": 6.126755285263061,
      "epoch": 2.044684360473834,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00045849426199738884,
      "loss": 5.7182,
      "mean_token_accuracy": 0.1728259116411209,
      "num_tokens": 40020877.0,
      "step": 20455
    },
    {
      "entropy": 6.087894105911255,
      "epoch": 2.0451841855350628,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004584736058805256,
      "loss": 5.744,
      "mean_token_accuracy": 0.17785179913043975,
      "num_tokens": 40030952.0,
      "step": 20460
    },
    {
      "entropy": 6.082538700103759,
      "epoch": 2.0456840105962915,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004584529451475137,
      "loss": 5.6682,
      "mean_token_accuracy": 0.1715752512216568,
      "num_tokens": 40039995.0,
      "step": 20465
    },
    {
      "entropy": 6.035723543167114,
      "epoch": 2.0461838356575197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045843227979887286,
      "loss": 5.6007,
      "mean_token_accuracy": 0.18163315206766129,
      "num_tokens": 40049346.0,
      "step": 20470
    },
    {
      "entropy": 6.066049814224243,
      "epoch": 2.0466836607187484,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045841160983512303,
      "loss": 5.7074,
      "mean_token_accuracy": 0.17245981991291046,
      "num_tokens": 40058371.0,
      "step": 20475
    },
    {
      "entropy": 6.040920639038086,
      "epoch": 2.047183485779977,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004583909352567843,
      "loss": 5.6259,
      "mean_token_accuracy": 0.17842479944229125,
      "num_tokens": 40067504.0,
      "step": 20480
    },
    {
      "entropy": 6.065737533569336,
      "epoch": 2.0476833108412054,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004583702560643769,
      "loss": 5.6799,
      "mean_token_accuracy": 0.18423683643341066,
      "num_tokens": 40077551.0,
      "step": 20485
    },
    {
      "entropy": 6.101685094833374,
      "epoch": 2.048183135902434,
      "grad_norm": 1.109375,
      "learning_rate": 0.000458349572258421,
      "loss": 5.7025,
      "mean_token_accuracy": 0.16679072380065918,
      "num_tokens": 40087896.0,
      "step": 20490
    },
    {
      "entropy": 6.076264047622681,
      "epoch": 2.0486829609636628,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045832888383943706,
      "loss": 5.6418,
      "mean_token_accuracy": 0.1748429447412491,
      "num_tokens": 40097197.0,
      "step": 20495
    },
    {
      "entropy": 6.140564823150635,
      "epoch": 2.0491827860248915,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004583081908079455,
      "loss": 5.7148,
      "mean_token_accuracy": 0.17130834907293319,
      "num_tokens": 40108068.0,
      "step": 20500
    },
    {
      "entropy": 6.032321453094482,
      "epoch": 2.0496826110861197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004582874931644671,
      "loss": 5.6661,
      "mean_token_accuracy": 0.17062364071607589,
      "num_tokens": 40116905.0,
      "step": 20505
    },
    {
      "entropy": 6.02406120300293,
      "epoch": 2.0501824361473484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045826679090952253,
      "loss": 5.655,
      "mean_token_accuracy": 0.17141062766313553,
      "num_tokens": 40126611.0,
      "step": 20510
    },
    {
      "entropy": 6.011526775360108,
      "epoch": 2.050682261208577,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045824608404363267,
      "loss": 5.5519,
      "mean_token_accuracy": 0.18007354587316513,
      "num_tokens": 40136557.0,
      "step": 20515
    },
    {
      "entropy": 6.0919411182403564,
      "epoch": 2.0511820862698054,
      "grad_norm": 1.125,
      "learning_rate": 0.0004582253725673184,
      "loss": 5.6757,
      "mean_token_accuracy": 0.168178927898407,
      "num_tokens": 40146637.0,
      "step": 20520
    },
    {
      "entropy": 6.115887594223023,
      "epoch": 2.051681911331034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000458204656481101,
      "loss": 5.6158,
      "mean_token_accuracy": 0.17620432525873184,
      "num_tokens": 40156763.0,
      "step": 20525
    },
    {
      "entropy": 6.031459331512451,
      "epoch": 2.0521817363922628,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004581839357855015,
      "loss": 5.6685,
      "mean_token_accuracy": 0.17504208236932756,
      "num_tokens": 40167024.0,
      "step": 20530
    },
    {
      "entropy": 6.060496807098389,
      "epoch": 2.0526815614534915,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045816321048104143,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17103224098682404,
      "num_tokens": 40176977.0,
      "step": 20535
    },
    {
      "entropy": 6.014176702499389,
      "epoch": 2.0531813865147197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004581424805682421,
      "loss": 5.5804,
      "mean_token_accuracy": 0.17368256747722627,
      "num_tokens": 40187820.0,
      "step": 20540
    },
    {
      "entropy": 6.095091485977173,
      "epoch": 2.0536812115759484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000458121746047625,
      "loss": 5.6584,
      "mean_token_accuracy": 0.17084798067808152,
      "num_tokens": 40197518.0,
      "step": 20545
    },
    {
      "entropy": 6.092929887771606,
      "epoch": 2.054181036637177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004581010069197119,
      "loss": 5.7502,
      "mean_token_accuracy": 0.16642735227942468,
      "num_tokens": 40208360.0,
      "step": 20550
    },
    {
      "entropy": 6.074043464660645,
      "epoch": 2.0546808616984054,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045808026318502464,
      "loss": 5.7579,
      "mean_token_accuracy": 0.1720088705420494,
      "num_tokens": 40219526.0,
      "step": 20555
    },
    {
      "entropy": 6.126330280303955,
      "epoch": 2.055180686759634,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004580595148440851,
      "loss": 5.6867,
      "mean_token_accuracy": 0.17019727379083632,
      "num_tokens": 40230661.0,
      "step": 20560
    },
    {
      "entropy": 6.094132995605468,
      "epoch": 2.0556805118208628,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004580387618974152,
      "loss": 5.783,
      "mean_token_accuracy": 0.16729024052619934,
      "num_tokens": 40241675.0,
      "step": 20565
    },
    {
      "entropy": 6.07023344039917,
      "epoch": 2.0561803368820915,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045801800434553715,
      "loss": 5.6173,
      "mean_token_accuracy": 0.17420373260974883,
      "num_tokens": 40251275.0,
      "step": 20570
    },
    {
      "entropy": 6.123404836654663,
      "epoch": 2.0566801619433197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004579972421889732,
      "loss": 5.7044,
      "mean_token_accuracy": 0.17232297360897064,
      "num_tokens": 40260975.0,
      "step": 20575
    },
    {
      "entropy": 6.043794631958008,
      "epoch": 2.0571799870045484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004579764754282458,
      "loss": 5.6563,
      "mean_token_accuracy": 0.174043932557106,
      "num_tokens": 40271617.0,
      "step": 20580
    },
    {
      "entropy": 6.061738729476929,
      "epoch": 2.057679812065777,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045795570406387726,
      "loss": 5.6105,
      "mean_token_accuracy": 0.1701284572482109,
      "num_tokens": 40281365.0,
      "step": 20585
    },
    {
      "entropy": 6.060124444961548,
      "epoch": 2.0581796371270054,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004579349280963904,
      "loss": 5.6719,
      "mean_token_accuracy": 0.1702652871608734,
      "num_tokens": 40291050.0,
      "step": 20590
    },
    {
      "entropy": 6.034000015258789,
      "epoch": 2.058679462188234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045791414752630765,
      "loss": 5.5467,
      "mean_token_accuracy": 0.1794648513197899,
      "num_tokens": 40299911.0,
      "step": 20595
    },
    {
      "entropy": 6.04473123550415,
      "epoch": 2.0591792872494628,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004578933623541522,
      "loss": 5.6544,
      "mean_token_accuracy": 0.1734946697950363,
      "num_tokens": 40310860.0,
      "step": 20600
    },
    {
      "entropy": 6.055109548568725,
      "epoch": 2.0596791123106915,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045787257258044666,
      "loss": 5.7275,
      "mean_token_accuracy": 0.17559722810983658,
      "num_tokens": 40320045.0,
      "step": 20605
    },
    {
      "entropy": 6.0331666469573975,
      "epoch": 2.0601789373719197,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045785177820571434,
      "loss": 5.7003,
      "mean_token_accuracy": 0.17166504561901091,
      "num_tokens": 40329082.0,
      "step": 20610
    },
    {
      "entropy": 6.0015363693237305,
      "epoch": 2.0606787624331484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045783097923047833,
      "loss": 5.6048,
      "mean_token_accuracy": 0.1797710731625557,
      "num_tokens": 40338161.0,
      "step": 20615
    },
    {
      "entropy": 6.099999618530274,
      "epoch": 2.061178587494377,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045781017565526194,
      "loss": 5.7244,
      "mean_token_accuracy": 0.16695560663938522,
      "num_tokens": 40348316.0,
      "step": 20620
    },
    {
      "entropy": 6.111133146286011,
      "epoch": 2.0616784125556054,
      "grad_norm": 1.890625,
      "learning_rate": 0.00045778936748058855,
      "loss": 5.6857,
      "mean_token_accuracy": 0.16966263204813004,
      "num_tokens": 40358112.0,
      "step": 20625
    },
    {
      "entropy": 6.052626705169677,
      "epoch": 2.062178237616834,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045776855470698163,
      "loss": 5.6514,
      "mean_token_accuracy": 0.18248414546251296,
      "num_tokens": 40368670.0,
      "step": 20630
    },
    {
      "entropy": 6.0200357913970945,
      "epoch": 2.0626780626780628,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00045774773733496505,
      "loss": 5.642,
      "mean_token_accuracy": 0.17645278573036194,
      "num_tokens": 40378082.0,
      "step": 20635
    },
    {
      "entropy": 6.1242743015289305,
      "epoch": 2.0631778877392914,
      "grad_norm": 1.125,
      "learning_rate": 0.0004577269153650624,
      "loss": 5.7033,
      "mean_token_accuracy": 0.16984138488769532,
      "num_tokens": 40387679.0,
      "step": 20640
    },
    {
      "entropy": 6.11882209777832,
      "epoch": 2.0636777128005197,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004577060887977975,
      "loss": 5.643,
      "mean_token_accuracy": 0.17401585280895232,
      "num_tokens": 40397599.0,
      "step": 20645
    },
    {
      "entropy": 6.154007768630981,
      "epoch": 2.0641775378617484,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045768525763369443,
      "loss": 5.6711,
      "mean_token_accuracy": 0.17306867688894273,
      "num_tokens": 40407163.0,
      "step": 20650
    },
    {
      "entropy": 5.991796255111694,
      "epoch": 2.064677362922977,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045766442187327735,
      "loss": 5.6359,
      "mean_token_accuracy": 0.176875801384449,
      "num_tokens": 40416489.0,
      "step": 20655
    },
    {
      "entropy": 6.033590316772461,
      "epoch": 2.0651771879842054,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045764358151707043,
      "loss": 5.7186,
      "mean_token_accuracy": 0.17028831988573073,
      "num_tokens": 40425193.0,
      "step": 20660
    },
    {
      "entropy": 6.072287845611572,
      "epoch": 2.065677013045434,
      "grad_norm": 1.125,
      "learning_rate": 0.0004576227365655979,
      "loss": 5.7355,
      "mean_token_accuracy": 0.17852805256843568,
      "num_tokens": 40434025.0,
      "step": 20665
    },
    {
      "entropy": 6.074138450622558,
      "epoch": 2.0661768381066627,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004576018870193844,
      "loss": 5.7308,
      "mean_token_accuracy": 0.17524137943983079,
      "num_tokens": 40443010.0,
      "step": 20670
    },
    {
      "entropy": 6.081421518325806,
      "epoch": 2.0666766631678914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004575810328789543,
      "loss": 5.7027,
      "mean_token_accuracy": 0.16707088500261308,
      "num_tokens": 40452673.0,
      "step": 20675
    },
    {
      "entropy": 6.027284145355225,
      "epoch": 2.0671764882291197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004575601741448325,
      "loss": 5.6809,
      "mean_token_accuracy": 0.17248480916023254,
      "num_tokens": 40461747.0,
      "step": 20680
    },
    {
      "entropy": 6.0459771156311035,
      "epoch": 2.0676763132903484,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045753931081754366,
      "loss": 5.6335,
      "mean_token_accuracy": 0.17251685112714768,
      "num_tokens": 40470674.0,
      "step": 20685
    },
    {
      "entropy": 6.066454362869263,
      "epoch": 2.068176138351577,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004575184428976127,
      "loss": 5.6342,
      "mean_token_accuracy": 0.17397847175598144,
      "num_tokens": 40480287.0,
      "step": 20690
    },
    {
      "entropy": 6.016861343383789,
      "epoch": 2.0686759634128054,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004574975703855647,
      "loss": 5.5704,
      "mean_token_accuracy": 0.1805051937699318,
      "num_tokens": 40489990.0,
      "step": 20695
    },
    {
      "entropy": 6.059269809722901,
      "epoch": 2.069175788474034,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045747669328192477,
      "loss": 5.695,
      "mean_token_accuracy": 0.17664320319890975,
      "num_tokens": 40500999.0,
      "step": 20700
    },
    {
      "entropy": 6.093976259231567,
      "epoch": 2.0696756135352627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004574558115872183,
      "loss": 5.6917,
      "mean_token_accuracy": 0.17347504943609238,
      "num_tokens": 40509607.0,
      "step": 20705
    },
    {
      "entropy": 5.961733627319336,
      "epoch": 2.0701754385964914,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045743492530197046,
      "loss": 5.5768,
      "mean_token_accuracy": 0.17583061009645462,
      "num_tokens": 40518312.0,
      "step": 20710
    },
    {
      "entropy": 6.011999702453613,
      "epoch": 2.0706752636577197,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004574140344267069,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17529899030923843,
      "num_tokens": 40528015.0,
      "step": 20715
    },
    {
      "entropy": 6.058765983581543,
      "epoch": 2.0711750887189484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004573931389619531,
      "loss": 5.7077,
      "mean_token_accuracy": 0.16676236540079117,
      "num_tokens": 40537557.0,
      "step": 20720
    },
    {
      "entropy": 6.077595520019531,
      "epoch": 2.071674913780177,
      "grad_norm": 1.125,
      "learning_rate": 0.00045737223890823494,
      "loss": 5.6475,
      "mean_token_accuracy": 0.1656878873705864,
      "num_tokens": 40547360.0,
      "step": 20725
    },
    {
      "entropy": 6.156688928604126,
      "epoch": 2.0721747388414054,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045735133426607813,
      "loss": 5.8437,
      "mean_token_accuracy": 0.1636424332857132,
      "num_tokens": 40558156.0,
      "step": 20730
    },
    {
      "entropy": 6.09121470451355,
      "epoch": 2.072674563902634,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045733042503600865,
      "loss": 5.7087,
      "mean_token_accuracy": 0.16733210682868957,
      "num_tokens": 40568732.0,
      "step": 20735
    },
    {
      "entropy": 5.977545738220215,
      "epoch": 2.0731743889638627,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045730951121855255,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18073829412460327,
      "num_tokens": 40579093.0,
      "step": 20740
    },
    {
      "entropy": 6.138778591156006,
      "epoch": 2.0736742140250914,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004572885928142362,
      "loss": 5.8127,
      "mean_token_accuracy": 0.1644661545753479,
      "num_tokens": 40589327.0,
      "step": 20745
    },
    {
      "entropy": 6.098407125473022,
      "epoch": 2.0741740390863197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004572676698235857,
      "loss": 5.6642,
      "mean_token_accuracy": 0.1729595184326172,
      "num_tokens": 40600054.0,
      "step": 20750
    },
    {
      "entropy": 6.138379526138306,
      "epoch": 2.0746738641475484,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004572467422471275,
      "loss": 5.6953,
      "mean_token_accuracy": 0.1763329640030861,
      "num_tokens": 40609286.0,
      "step": 20755
    },
    {
      "entropy": 6.103570127487183,
      "epoch": 2.075173689208777,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045722581008538816,
      "loss": 5.7968,
      "mean_token_accuracy": 0.17277758866548537,
      "num_tokens": 40618150.0,
      "step": 20760
    },
    {
      "entropy": 6.036206436157227,
      "epoch": 2.0756735142700053,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004572048733388944,
      "loss": 5.5505,
      "mean_token_accuracy": 0.19003332108259202,
      "num_tokens": 40627842.0,
      "step": 20765
    },
    {
      "entropy": 6.081304216384888,
      "epoch": 2.076173339331234,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045718393200817284,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1789113014936447,
      "num_tokens": 40637767.0,
      "step": 20770
    },
    {
      "entropy": 6.029018545150757,
      "epoch": 2.0766731643924627,
      "grad_norm": 1.125,
      "learning_rate": 0.00045716298609375044,
      "loss": 5.7488,
      "mean_token_accuracy": 0.16704170554876327,
      "num_tokens": 40648611.0,
      "step": 20775
    },
    {
      "entropy": 6.110093402862549,
      "epoch": 2.077172989453691,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045714203559615423,
      "loss": 5.7212,
      "mean_token_accuracy": 0.1734963595867157,
      "num_tokens": 40659664.0,
      "step": 20780
    },
    {
      "entropy": 6.069572019577026,
      "epoch": 2.0776728145149197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004571210805159112,
      "loss": 5.6359,
      "mean_token_accuracy": 0.17799576967954636,
      "num_tokens": 40669145.0,
      "step": 20785
    },
    {
      "entropy": 6.088476705551147,
      "epoch": 2.0781726395761484,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004571001208535487,
      "loss": 5.6395,
      "mean_token_accuracy": 0.17895632982254028,
      "num_tokens": 40678250.0,
      "step": 20790
    },
    {
      "entropy": 6.043583011627197,
      "epoch": 2.078672464637377,
      "grad_norm": 1.078125,
      "learning_rate": 0.000457079156609594,
      "loss": 5.637,
      "mean_token_accuracy": 0.17629288882017136,
      "num_tokens": 40687590.0,
      "step": 20795
    },
    {
      "entropy": 5.894083452224732,
      "epoch": 2.0791722896986053,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004570581877845746,
      "loss": 5.5875,
      "mean_token_accuracy": 0.17559274286031723,
      "num_tokens": 40697452.0,
      "step": 20800
    },
    {
      "entropy": 6.08178973197937,
      "epoch": 2.079672114759834,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000457037214379018,
      "loss": 5.7252,
      "mean_token_accuracy": 0.17359966188669204,
      "num_tokens": 40707608.0,
      "step": 20805
    },
    {
      "entropy": 6.127851343154907,
      "epoch": 2.0801719398210627,
      "grad_norm": 1.5390625,
      "learning_rate": 0.00045701623639345204,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17511545568704606,
      "num_tokens": 40717369.0,
      "step": 20810
    },
    {
      "entropy": 6.026893377304077,
      "epoch": 2.080671764882291,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045699525382840423,
      "loss": 5.58,
      "mean_token_accuracy": 0.18460170328617095,
      "num_tokens": 40725899.0,
      "step": 20815
    },
    {
      "entropy": 6.0821774959564205,
      "epoch": 2.0811715899435197,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045697426668440277,
      "loss": 5.7551,
      "mean_token_accuracy": 0.1632165417075157,
      "num_tokens": 40735252.0,
      "step": 20820
    },
    {
      "entropy": 6.131407403945923,
      "epoch": 2.0816714150047484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045695327496197557,
      "loss": 5.762,
      "mean_token_accuracy": 0.16980428248643875,
      "num_tokens": 40745828.0,
      "step": 20825
    },
    {
      "entropy": 6.075436592102051,
      "epoch": 2.082171240065977,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004569322786616508,
      "loss": 5.6601,
      "mean_token_accuracy": 0.17247687429189681,
      "num_tokens": 40756388.0,
      "step": 20830
    },
    {
      "entropy": 6.015303325653076,
      "epoch": 2.0826710651272053,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004569112777839567,
      "loss": 5.5425,
      "mean_token_accuracy": 0.17467869967222213,
      "num_tokens": 40765914.0,
      "step": 20835
    },
    {
      "entropy": 6.032530689239502,
      "epoch": 2.083170890188434,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045689027232942165,
      "loss": 5.604,
      "mean_token_accuracy": 0.17408095598220824,
      "num_tokens": 40775015.0,
      "step": 20840
    },
    {
      "entropy": 5.932141447067261,
      "epoch": 2.0836707152496627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004568692622985742,
      "loss": 5.5926,
      "mean_token_accuracy": 0.17582179009914398,
      "num_tokens": 40783743.0,
      "step": 20845
    },
    {
      "entropy": 6.046247673034668,
      "epoch": 2.084170540310891,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00045684824769194284,
      "loss": 5.7199,
      "mean_token_accuracy": 0.17527076750993728,
      "num_tokens": 40793973.0,
      "step": 20850
    },
    {
      "entropy": 6.114718008041382,
      "epoch": 2.0846703653721197,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004568272285100564,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17568481117486953,
      "num_tokens": 40803800.0,
      "step": 20855
    },
    {
      "entropy": 6.15155062675476,
      "epoch": 2.0851701904333484,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004568062047534437,
      "loss": 5.7017,
      "mean_token_accuracy": 0.17210457772016524,
      "num_tokens": 40812791.0,
      "step": 20860
    },
    {
      "entropy": 6.057941818237305,
      "epoch": 2.085670015494577,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004567851764226337,
      "loss": 5.7054,
      "mean_token_accuracy": 0.1779686003923416,
      "num_tokens": 40822498.0,
      "step": 20865
    },
    {
      "entropy": 6.019943380355835,
      "epoch": 2.0861698405558053,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045676414351815533,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1813590943813324,
      "num_tokens": 40831965.0,
      "step": 20870
    },
    {
      "entropy": 6.062142324447632,
      "epoch": 2.086669665617034,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004567431060405379,
      "loss": 5.6634,
      "mean_token_accuracy": 0.16894571781158446,
      "num_tokens": 40840521.0,
      "step": 20875
    },
    {
      "entropy": 6.044579792022705,
      "epoch": 2.0871694906782627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004567220639903108,
      "loss": 5.7728,
      "mean_token_accuracy": 0.17107580453157425,
      "num_tokens": 40850935.0,
      "step": 20880
    },
    {
      "entropy": 6.056697034835816,
      "epoch": 2.087669315739491,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045670101736800325,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17859412133693695,
      "num_tokens": 40860662.0,
      "step": 20885
    },
    {
      "entropy": 5.920725727081299,
      "epoch": 2.0881691408007197,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045667996617414485,
      "loss": 5.4648,
      "mean_token_accuracy": 0.18851417452096939,
      "num_tokens": 40870818.0,
      "step": 20890
    },
    {
      "entropy": 6.07517557144165,
      "epoch": 2.0886689658619484,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004566589104092653,
      "loss": 5.8042,
      "mean_token_accuracy": 0.1679603025317192,
      "num_tokens": 40880842.0,
      "step": 20895
    },
    {
      "entropy": 6.1298644065856935,
      "epoch": 2.089168790923177,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004566378500738943,
      "loss": 5.7685,
      "mean_token_accuracy": 0.17060995250940322,
      "num_tokens": 40892070.0,
      "step": 20900
    },
    {
      "entropy": 6.01724591255188,
      "epoch": 2.0896686159844053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045661678516856164,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18394319713115692,
      "num_tokens": 40901588.0,
      "step": 20905
    },
    {
      "entropy": 6.113409805297851,
      "epoch": 2.090168441045634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004565957156937975,
      "loss": 5.6849,
      "mean_token_accuracy": 0.1786071166396141,
      "num_tokens": 40909961.0,
      "step": 20910
    },
    {
      "entropy": 6.089390420913697,
      "epoch": 2.0906682661068627,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004565746416501318,
      "loss": 5.6795,
      "mean_token_accuracy": 0.17709096074104308,
      "num_tokens": 40919528.0,
      "step": 20915
    },
    {
      "entropy": 6.065116119384766,
      "epoch": 2.091168091168091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045655356303809494,
      "loss": 5.6762,
      "mean_token_accuracy": 0.17108058482408522,
      "num_tokens": 40929975.0,
      "step": 20920
    },
    {
      "entropy": 6.07010440826416,
      "epoch": 2.0916679162293197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045653247985821707,
      "loss": 5.6288,
      "mean_token_accuracy": 0.17442016303539276,
      "num_tokens": 40939952.0,
      "step": 20925
    },
    {
      "entropy": 6.039519453048706,
      "epoch": 2.0921677412905484,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045651139211102875,
      "loss": 5.5859,
      "mean_token_accuracy": 0.18164878636598586,
      "num_tokens": 40948935.0,
      "step": 20930
    },
    {
      "entropy": 5.984440660476684,
      "epoch": 2.092667566351777,
      "grad_norm": 1.125,
      "learning_rate": 0.00045649029979706054,
      "loss": 5.6282,
      "mean_token_accuracy": 0.17706005573272704,
      "num_tokens": 40958169.0,
      "step": 20935
    },
    {
      "entropy": 6.0194236755371096,
      "epoch": 2.0931673914130053,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045646920291684303,
      "loss": 5.6527,
      "mean_token_accuracy": 0.17727464139461518,
      "num_tokens": 40967920.0,
      "step": 20940
    },
    {
      "entropy": 6.047277736663818,
      "epoch": 2.093667216474234,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004564481014709072,
      "loss": 5.6477,
      "mean_token_accuracy": 0.1758141800761223,
      "num_tokens": 40977494.0,
      "step": 20945
    },
    {
      "entropy": 6.131130838394165,
      "epoch": 2.0941670415354627,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004564269954597836,
      "loss": 5.7313,
      "mean_token_accuracy": 0.16574118584394454,
      "num_tokens": 40987838.0,
      "step": 20950
    },
    {
      "entropy": 6.024343872070313,
      "epoch": 2.094666866596691,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045640588488400364,
      "loss": 5.5447,
      "mean_token_accuracy": 0.17981948107481002,
      "num_tokens": 40997308.0,
      "step": 20955
    },
    {
      "entropy": 5.9922689437866214,
      "epoch": 2.0951666916579197,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004563847697440983,
      "loss": 5.6863,
      "mean_token_accuracy": 0.17248110771179198,
      "num_tokens": 41008159.0,
      "step": 20960
    },
    {
      "entropy": 6.0828855514526365,
      "epoch": 2.0956665167191484,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045636365004059876,
      "loss": 5.7214,
      "mean_token_accuracy": 0.17098451554775237,
      "num_tokens": 41018224.0,
      "step": 20965
    },
    {
      "entropy": 6.058098173141479,
      "epoch": 2.0961663417803766,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045634252577403644,
      "loss": 5.6094,
      "mean_token_accuracy": 0.17631808072328567,
      "num_tokens": 41026995.0,
      "step": 20970
    },
    {
      "entropy": 6.0191734790802,
      "epoch": 2.0966661668416053,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045632139694494294,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1857874482870102,
      "num_tokens": 41036243.0,
      "step": 20975
    },
    {
      "entropy": 6.04973349571228,
      "epoch": 2.097165991902834,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004563002635538496,
      "loss": 5.6575,
      "mean_token_accuracy": 0.1716335192322731,
      "num_tokens": 41046196.0,
      "step": 20980
    },
    {
      "entropy": 6.042449283599853,
      "epoch": 2.0976658169640627,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004562791256012884,
      "loss": 5.6514,
      "mean_token_accuracy": 0.18357468396425247,
      "num_tokens": 41057053.0,
      "step": 20985
    },
    {
      "entropy": 6.151152467727661,
      "epoch": 2.098165642025291,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004562579830877911,
      "loss": 5.8007,
      "mean_token_accuracy": 0.1707794949412346,
      "num_tokens": 41067349.0,
      "step": 20990
    },
    {
      "entropy": 6.079155588150025,
      "epoch": 2.0986654670865197,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004562368360138894,
      "loss": 5.7135,
      "mean_token_accuracy": 0.1743532732129097,
      "num_tokens": 41077327.0,
      "step": 20995
    },
    {
      "entropy": 6.058238410949707,
      "epoch": 2.0991652921477484,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004562156843801157,
      "loss": 5.6412,
      "mean_token_accuracy": 0.17634079456329346,
      "num_tokens": 41086569.0,
      "step": 21000
    },
    {
      "epoch": 2.0991652921477484,
      "eval_entropy": 5.882166661358593,
      "eval_loss": 5.825972557067871,
      "eval_mean_token_accuracy": 0.17458519192795968,
      "eval_num_tokens": 41086569.0,
      "eval_runtime": 23.7,
      "eval_samples_per_second": 1309.918,
      "eval_steps_per_second": 163.756,
      "step": 21000
    },
    {
      "entropy": 6.038561630249023,
      "epoch": 2.0996651172089766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045619452818700194,
      "loss": 5.6254,
      "mean_token_accuracy": 0.17201472967863082,
      "num_tokens": 41095170.0,
      "step": 21005
    },
    {
      "entropy": 6.069289588928223,
      "epoch": 2.1001649422702053,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045617336743508056,
      "loss": 5.6093,
      "mean_token_accuracy": 0.18301283419132233,
      "num_tokens": 41104778.0,
      "step": 21010
    },
    {
      "entropy": 5.994983339309693,
      "epoch": 2.100664767331434,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004561522021248837,
      "loss": 5.5414,
      "mean_token_accuracy": 0.18350038081407546,
      "num_tokens": 41114133.0,
      "step": 21015
    },
    {
      "entropy": 6.073039722442627,
      "epoch": 2.1011645923926627,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004561310322569442,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17736288011074067,
      "num_tokens": 41125052.0,
      "step": 21020
    },
    {
      "entropy": 6.0408008098602295,
      "epoch": 2.101664417453891,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045610985783179454,
      "loss": 5.6597,
      "mean_token_accuracy": 0.17129295617341994,
      "num_tokens": 41134160.0,
      "step": 21025
    },
    {
      "entropy": 6.0434144020080565,
      "epoch": 2.1021642425151197,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004560886788499674,
      "loss": 5.6042,
      "mean_token_accuracy": 0.18064468950033188,
      "num_tokens": 41144729.0,
      "step": 21030
    },
    {
      "entropy": 6.074617004394531,
      "epoch": 2.1026640675763484,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045606749531199573,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17435962110757827,
      "num_tokens": 41154935.0,
      "step": 21035
    },
    {
      "entropy": 6.071571254730225,
      "epoch": 2.1031638926375766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045604630721841243,
      "loss": 5.6392,
      "mean_token_accuracy": 0.17616469264030457,
      "num_tokens": 41164644.0,
      "step": 21040
    },
    {
      "entropy": 6.012654066085815,
      "epoch": 2.1036637176988053,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045602511456975064,
      "loss": 5.5906,
      "mean_token_accuracy": 0.18359473198652268,
      "num_tokens": 41174656.0,
      "step": 21045
    },
    {
      "entropy": 6.058246707916259,
      "epoch": 2.104163542760034,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004560039173665436,
      "loss": 5.68,
      "mean_token_accuracy": 0.17186628580093383,
      "num_tokens": 41184631.0,
      "step": 21050
    },
    {
      "entropy": 6.080548048019409,
      "epoch": 2.1046633678212627,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004559827156093245,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17425207942724227,
      "num_tokens": 41195157.0,
      "step": 21055
    },
    {
      "entropy": 6.11781940460205,
      "epoch": 2.105163192882491,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004559615092986268,
      "loss": 5.698,
      "mean_token_accuracy": 0.17153821289539337,
      "num_tokens": 41205566.0,
      "step": 21060
    },
    {
      "entropy": 6.016005277633667,
      "epoch": 2.1056630179437197,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045594029843498425,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17525962144136428,
      "num_tokens": 41214194.0,
      "step": 21065
    },
    {
      "entropy": 6.090572309494019,
      "epoch": 2.1061628430049484,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045591908301893027,
      "loss": 5.7851,
      "mean_token_accuracy": 0.16877472698688506,
      "num_tokens": 41223965.0,
      "step": 21070
    },
    {
      "entropy": 6.07534704208374,
      "epoch": 2.1066626680661766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045589786305099875,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1765807792544365,
      "num_tokens": 41233374.0,
      "step": 21075
    },
    {
      "entropy": 6.109201526641845,
      "epoch": 2.1071624931274053,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004558766385317235,
      "loss": 5.7279,
      "mean_token_accuracy": 0.17084459066390992,
      "num_tokens": 41243365.0,
      "step": 21080
    },
    {
      "entropy": 6.11083345413208,
      "epoch": 2.107662318188634,
      "grad_norm": 1.375,
      "learning_rate": 0.00045585540946163855,
      "loss": 5.6485,
      "mean_token_accuracy": 0.18230974227190017,
      "num_tokens": 41253013.0,
      "step": 21085
    },
    {
      "entropy": 6.023306655883789,
      "epoch": 2.1081621432498627,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004558341758412781,
      "loss": 5.6248,
      "mean_token_accuracy": 0.1822210118174553,
      "num_tokens": 41261889.0,
      "step": 21090
    },
    {
      "entropy": 6.019154977798462,
      "epoch": 2.108661968311091,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004558129376711763,
      "loss": 5.6165,
      "mean_token_accuracy": 0.18100370615720748,
      "num_tokens": 41271777.0,
      "step": 21095
    },
    {
      "entropy": 6.100373458862305,
      "epoch": 2.1091617933723197,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004557916949518675,
      "loss": 5.7405,
      "mean_token_accuracy": 0.1698204904794693,
      "num_tokens": 41281840.0,
      "step": 21100
    },
    {
      "entropy": 6.152972984313965,
      "epoch": 2.1096616184335484,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004557704476838862,
      "loss": 5.7484,
      "mean_token_accuracy": 0.1689772129058838,
      "num_tokens": 41292200.0,
      "step": 21105
    },
    {
      "entropy": 6.13608684539795,
      "epoch": 2.1101614434947766,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045574919586776686,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16896164715290068,
      "num_tokens": 41301898.0,
      "step": 21110
    },
    {
      "entropy": 6.064538478851318,
      "epoch": 2.1106612685560053,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045572793950404434,
      "loss": 5.6585,
      "mean_token_accuracy": 0.1727173462510109,
      "num_tokens": 41311519.0,
      "step": 21115
    },
    {
      "entropy": 6.1098658561706545,
      "epoch": 2.111161093617234,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045570667859325337,
      "loss": 5.704,
      "mean_token_accuracy": 0.1702266439795494,
      "num_tokens": 41321955.0,
      "step": 21120
    },
    {
      "entropy": 6.149000453948974,
      "epoch": 2.1116609186784627,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045568541313592884,
      "loss": 5.7415,
      "mean_token_accuracy": 0.16554326713085174,
      "num_tokens": 41332467.0,
      "step": 21125
    },
    {
      "entropy": 6.1504301071167,
      "epoch": 2.112160743739691,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045566414313260586,
      "loss": 5.7806,
      "mean_token_accuracy": 0.16828834712505342,
      "num_tokens": 41342451.0,
      "step": 21130
    },
    {
      "entropy": 6.040538311004639,
      "epoch": 2.1126605688009197,
      "grad_norm": 1.25,
      "learning_rate": 0.0004556428685838194,
      "loss": 5.6328,
      "mean_token_accuracy": 0.17486201822757722,
      "num_tokens": 41352678.0,
      "step": 21135
    },
    {
      "entropy": 6.044945240020752,
      "epoch": 2.1131603938621484,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000455621589490105,
      "loss": 5.736,
      "mean_token_accuracy": 0.17222084999084472,
      "num_tokens": 41362584.0,
      "step": 21140
    },
    {
      "entropy": 6.038116836547852,
      "epoch": 2.1136602189233766,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004556003058519979,
      "loss": 5.7152,
      "mean_token_accuracy": 0.1729067698121071,
      "num_tokens": 41371988.0,
      "step": 21145
    },
    {
      "entropy": 6.114104843139648,
      "epoch": 2.1141600439846053,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045557901767003347,
      "loss": 5.6351,
      "mean_token_accuracy": 0.17364222109317778,
      "num_tokens": 41381281.0,
      "step": 21150
    },
    {
      "entropy": 6.009316253662109,
      "epoch": 2.114659869045834,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045555772494474745,
      "loss": 5.5501,
      "mean_token_accuracy": 0.18858337998390198,
      "num_tokens": 41389711.0,
      "step": 21155
    },
    {
      "entropy": 6.1277862071990965,
      "epoch": 2.1151596941070627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004555364276766756,
      "loss": 5.6845,
      "mean_token_accuracy": 0.17271229028701782,
      "num_tokens": 41398375.0,
      "step": 21160
    },
    {
      "entropy": 6.105295038223266,
      "epoch": 2.115659519168291,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004555151258663535,
      "loss": 5.7572,
      "mean_token_accuracy": 0.17001912444829942,
      "num_tokens": 41408372.0,
      "step": 21165
    },
    {
      "entropy": 6.012362098693847,
      "epoch": 2.1161593442295197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004554938195143175,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17710380405187606,
      "num_tokens": 41418231.0,
      "step": 21170
    },
    {
      "entropy": 6.080168533325195,
      "epoch": 2.1166591692907484,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004554725086211033,
      "loss": 5.6606,
      "mean_token_accuracy": 0.18289808183908463,
      "num_tokens": 41427646.0,
      "step": 21175
    },
    {
      "entropy": 6.072376918792725,
      "epoch": 2.1171589943519766,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045545119318724734,
      "loss": 5.7334,
      "mean_token_accuracy": 0.17166556119918824,
      "num_tokens": 41437352.0,
      "step": 21180
    },
    {
      "entropy": 6.094399785995483,
      "epoch": 2.1176588194132053,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004554298732132857,
      "loss": 5.692,
      "mean_token_accuracy": 0.17641753405332566,
      "num_tokens": 41446769.0,
      "step": 21185
    },
    {
      "entropy": 6.106627988815307,
      "epoch": 2.118158644474434,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004554085486997549,
      "loss": 5.7055,
      "mean_token_accuracy": 0.16869733482599258,
      "num_tokens": 41456252.0,
      "step": 21190
    },
    {
      "entropy": 6.039885520935059,
      "epoch": 2.1186584695356627,
      "grad_norm": 1.71875,
      "learning_rate": 0.0004553872196471914,
      "loss": 5.6246,
      "mean_token_accuracy": 0.17406915724277497,
      "num_tokens": 41465388.0,
      "step": 21195
    },
    {
      "entropy": 6.090377950668335,
      "epoch": 2.119158294596891,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045536588605613195,
      "loss": 5.694,
      "mean_token_accuracy": 0.16643404364585876,
      "num_tokens": 41474772.0,
      "step": 21200
    },
    {
      "entropy": 6.039781904220581,
      "epoch": 2.1196581196581197,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045534454792711324,
      "loss": 5.603,
      "mean_token_accuracy": 0.17496849000453948,
      "num_tokens": 41484767.0,
      "step": 21205
    },
    {
      "entropy": 6.067582893371582,
      "epoch": 2.1201579447193484,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045532320526067206,
      "loss": 5.6789,
      "mean_token_accuracy": 0.1758434608578682,
      "num_tokens": 41495305.0,
      "step": 21210
    },
    {
      "entropy": 6.033560848236084,
      "epoch": 2.1206577697805766,
      "grad_norm": 1.125,
      "learning_rate": 0.0004553018580573454,
      "loss": 5.6118,
      "mean_token_accuracy": 0.17580133974552153,
      "num_tokens": 41504427.0,
      "step": 21215
    },
    {
      "entropy": 6.0532478332519535,
      "epoch": 2.1211575948418053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045528050631767037,
      "loss": 5.7003,
      "mean_token_accuracy": 0.17309947460889816,
      "num_tokens": 41514607.0,
      "step": 21220
    },
    {
      "entropy": 6.046260166168213,
      "epoch": 2.121657419903034,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045525915004218426,
      "loss": 5.6756,
      "mean_token_accuracy": 0.1726166397333145,
      "num_tokens": 41524280.0,
      "step": 21225
    },
    {
      "entropy": 6.083261966705322,
      "epoch": 2.1221572449642627,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045523778923142435,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17175640165805817,
      "num_tokens": 41533059.0,
      "step": 21230
    },
    {
      "entropy": 6.124611663818359,
      "epoch": 2.122657070025491,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004552164238859279,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17251282483339309,
      "num_tokens": 41542876.0,
      "step": 21235
    },
    {
      "entropy": 6.074017333984375,
      "epoch": 2.1231568950867197,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045519505400623274,
      "loss": 5.7317,
      "mean_token_accuracy": 0.16437674164772034,
      "num_tokens": 41552337.0,
      "step": 21240
    },
    {
      "entropy": 6.119915533065796,
      "epoch": 2.1236567201479484,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045517367959287635,
      "loss": 5.833,
      "mean_token_accuracy": 0.163869471848011,
      "num_tokens": 41563190.0,
      "step": 21245
    },
    {
      "entropy": 6.042694473266602,
      "epoch": 2.1241565452091766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045515230064639657,
      "loss": 5.6087,
      "mean_token_accuracy": 0.18154323250055313,
      "num_tokens": 41573927.0,
      "step": 21250
    },
    {
      "entropy": 6.028223180770874,
      "epoch": 2.1246563702704053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045513091716733117,
      "loss": 5.4972,
      "mean_token_accuracy": 0.1968925639986992,
      "num_tokens": 41583828.0,
      "step": 21255
    },
    {
      "entropy": 5.955699491500854,
      "epoch": 2.125156195331634,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045510952915621834,
      "loss": 5.6213,
      "mean_token_accuracy": 0.17516648322343825,
      "num_tokens": 41593179.0,
      "step": 21260
    },
    {
      "entropy": 6.015944099426269,
      "epoch": 2.1256560203928627,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004550881366135961,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16823360472917556,
      "num_tokens": 41603366.0,
      "step": 21265
    },
    {
      "entropy": 6.111535692214966,
      "epoch": 2.126155845454091,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045506673954000265,
      "loss": 5.6875,
      "mean_token_accuracy": 0.17646982669830322,
      "num_tokens": 41612916.0,
      "step": 21270
    },
    {
      "entropy": 6.141942834854126,
      "epoch": 2.1266556705153197,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004550453379359763,
      "loss": 5.7415,
      "mean_token_accuracy": 0.16683694422245027,
      "num_tokens": 41622770.0,
      "step": 21275
    },
    {
      "entropy": 6.106043529510498,
      "epoch": 2.1271554955765484,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045502393180205564,
      "loss": 5.7095,
      "mean_token_accuracy": 0.16883511543273927,
      "num_tokens": 41632590.0,
      "step": 21280
    },
    {
      "entropy": 6.051552724838257,
      "epoch": 2.1276553206377766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004550025211387792,
      "loss": 5.7232,
      "mean_token_accuracy": 0.17379293441772461,
      "num_tokens": 41641932.0,
      "step": 21285
    },
    {
      "entropy": 6.12023549079895,
      "epoch": 2.1281551456990053,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004549811059466857,
      "loss": 5.7197,
      "mean_token_accuracy": 0.17084573060274125,
      "num_tokens": 41651915.0,
      "step": 21290
    },
    {
      "entropy": 5.985575532913208,
      "epoch": 2.128654970760234,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004549596862263138,
      "loss": 5.5457,
      "mean_token_accuracy": 0.18461011201143265,
      "num_tokens": 41661087.0,
      "step": 21295
    },
    {
      "entropy": 5.995650148391723,
      "epoch": 2.1291547958214627,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004549382619782025,
      "loss": 5.6979,
      "mean_token_accuracy": 0.17427356094121932,
      "num_tokens": 41670418.0,
      "step": 21300
    },
    {
      "entropy": 6.059004211425782,
      "epoch": 2.129654620882691,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004549168332028908,
      "loss": 5.6207,
      "mean_token_accuracy": 0.17695794701576234,
      "num_tokens": 41680209.0,
      "step": 21305
    },
    {
      "entropy": 6.144467449188232,
      "epoch": 2.1301544459439197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045489539990091796,
      "loss": 5.686,
      "mean_token_accuracy": 0.17054597437381744,
      "num_tokens": 41689933.0,
      "step": 21310
    },
    {
      "entropy": 6.010845708847046,
      "epoch": 2.1306542710051484,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045487396207282316,
      "loss": 5.57,
      "mean_token_accuracy": 0.17796564549207688,
      "num_tokens": 41699017.0,
      "step": 21315
    },
    {
      "entropy": 5.999052238464356,
      "epoch": 2.1311540960663766,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045485251971914565,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1733529269695282,
      "num_tokens": 41708375.0,
      "step": 21320
    },
    {
      "entropy": 6.008264350891113,
      "epoch": 2.1316539211276053,
      "grad_norm": 1.5078125,
      "learning_rate": 0.0004548310728404251,
      "loss": 5.6147,
      "mean_token_accuracy": 0.1765448659658432,
      "num_tokens": 41718269.0,
      "step": 21325
    },
    {
      "entropy": 6.068723869323731,
      "epoch": 2.132153746188834,
      "grad_norm": 1.140625,
      "learning_rate": 0.000454809621437201,
      "loss": 5.6971,
      "mean_token_accuracy": 0.16611239463090896,
      "num_tokens": 41728426.0,
      "step": 21330
    },
    {
      "entropy": 6.031113576889038,
      "epoch": 2.1326535712500623,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004547881655100131,
      "loss": 5.6616,
      "mean_token_accuracy": 0.16666532009840013,
      "num_tokens": 41738255.0,
      "step": 21335
    },
    {
      "entropy": 6.171995639801025,
      "epoch": 2.133153396311291,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004547667050594013,
      "loss": 5.7496,
      "mean_token_accuracy": 0.16798180043697358,
      "num_tokens": 41747668.0,
      "step": 21340
    },
    {
      "entropy": 6.111067533493042,
      "epoch": 2.1336532213725197,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045474524008590536,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17852789610624314,
      "num_tokens": 41757414.0,
      "step": 21345
    },
    {
      "entropy": 6.084634780883789,
      "epoch": 2.1341530464337484,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004547237705900655,
      "loss": 5.7104,
      "mean_token_accuracy": 0.17668609768152238,
      "num_tokens": 41765413.0,
      "step": 21350
    },
    {
      "entropy": 6.022200965881348,
      "epoch": 2.1346528714949766,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045470229657242176,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17315897792577745,
      "num_tokens": 41775951.0,
      "step": 21355
    },
    {
      "entropy": 6.012775135040283,
      "epoch": 2.1351526965562053,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00045468081803351447,
      "loss": 5.6313,
      "mean_token_accuracy": 0.17673833221197127,
      "num_tokens": 41785440.0,
      "step": 21360
    },
    {
      "entropy": 6.097571086883545,
      "epoch": 2.135652521617434,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004546593349738841,
      "loss": 5.6969,
      "mean_token_accuracy": 0.16910674721002578,
      "num_tokens": 41794757.0,
      "step": 21365
    },
    {
      "entropy": 6.0979584693908695,
      "epoch": 2.1361523466786623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000454637847394071,
      "loss": 5.7474,
      "mean_token_accuracy": 0.17688190191984177,
      "num_tokens": 41804491.0,
      "step": 21370
    },
    {
      "entropy": 6.053183078765869,
      "epoch": 2.136652171739891,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045461635529461597,
      "loss": 5.6048,
      "mean_token_accuracy": 0.18072144240140914,
      "num_tokens": 41813867.0,
      "step": 21375
    },
    {
      "entropy": 6.1013788223266605,
      "epoch": 2.1371519968011197,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545948586760596,
      "loss": 5.7765,
      "mean_token_accuracy": 0.16022187396883963,
      "num_tokens": 41823167.0,
      "step": 21380
    },
    {
      "entropy": 6.0598876953125,
      "epoch": 2.1376518218623484,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004545733575389428,
      "loss": 5.636,
      "mean_token_accuracy": 0.16958805471658706,
      "num_tokens": 41831757.0,
      "step": 21385
    },
    {
      "entropy": 6.07030553817749,
      "epoch": 2.1381516469235766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004545518518838066,
      "loss": 5.6047,
      "mean_token_accuracy": 0.18150582909584045,
      "num_tokens": 41841741.0,
      "step": 21390
    },
    {
      "entropy": 6.030051136016846,
      "epoch": 2.1386514719848053,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000454530341711192,
      "loss": 5.5965,
      "mean_token_accuracy": 0.1800922393798828,
      "num_tokens": 41851940.0,
      "step": 21395
    },
    {
      "entropy": 6.064763164520263,
      "epoch": 2.139151297046034,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004545088270216401,
      "loss": 5.6347,
      "mean_token_accuracy": 0.17788725942373276,
      "num_tokens": 41861240.0,
      "step": 21400
    },
    {
      "entropy": 6.030258703231811,
      "epoch": 2.1396511221072623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004544873078156924,
      "loss": 5.744,
      "mean_token_accuracy": 0.16949288100004195,
      "num_tokens": 41871513.0,
      "step": 21405
    },
    {
      "entropy": 6.146198272705078,
      "epoch": 2.140150947168491,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045446578409389016,
      "loss": 5.7288,
      "mean_token_accuracy": 0.17184216380119324,
      "num_tokens": 41880346.0,
      "step": 21410
    },
    {
      "entropy": 6.0823462963104244,
      "epoch": 2.1406507722297197,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045444425585677503,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17804159522056578,
      "num_tokens": 41889182.0,
      "step": 21415
    },
    {
      "entropy": 6.050590085983276,
      "epoch": 2.1411505972909484,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004544227231048886,
      "loss": 5.629,
      "mean_token_accuracy": 0.17439205795526505,
      "num_tokens": 41899108.0,
      "step": 21420
    },
    {
      "entropy": 5.98486909866333,
      "epoch": 2.1416504223521766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004544011858387726,
      "loss": 5.6554,
      "mean_token_accuracy": 0.17556255012750627,
      "num_tokens": 41908930.0,
      "step": 21425
    },
    {
      "entropy": 6.034679126739502,
      "epoch": 2.1421502474134053,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00045437964405896895,
      "loss": 5.6695,
      "mean_token_accuracy": 0.18038943707942962,
      "num_tokens": 41919043.0,
      "step": 21430
    },
    {
      "entropy": 6.023789930343628,
      "epoch": 2.142650072474634,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045435809776601954,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1843360811471939,
      "num_tokens": 41929241.0,
      "step": 21435
    },
    {
      "entropy": 6.022858667373657,
      "epoch": 2.1431498975358623,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004543365469604666,
      "loss": 5.6812,
      "mean_token_accuracy": 0.17405349165201187,
      "num_tokens": 41940829.0,
      "step": 21440
    },
    {
      "entropy": 6.066896533966064,
      "epoch": 2.143649722597091,
      "grad_norm": 1.125,
      "learning_rate": 0.0004543149916428523,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1710815668106079,
      "num_tokens": 41950137.0,
      "step": 21445
    },
    {
      "entropy": 6.075569152832031,
      "epoch": 2.1441495476583197,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004542934318137189,
      "loss": 5.6843,
      "mean_token_accuracy": 0.17403293251991273,
      "num_tokens": 41960089.0,
      "step": 21450
    },
    {
      "entropy": 6.09486985206604,
      "epoch": 2.1446493727195484,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004542718674736089,
      "loss": 5.642,
      "mean_token_accuracy": 0.17538643777370452,
      "num_tokens": 41970168.0,
      "step": 21455
    },
    {
      "entropy": 6.035608148574829,
      "epoch": 2.1451491977807766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542502986230649,
      "loss": 5.6309,
      "mean_token_accuracy": 0.1749051347374916,
      "num_tokens": 41979540.0,
      "step": 21460
    },
    {
      "entropy": 6.049019145965576,
      "epoch": 2.1456490228420053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004542287252626295,
      "loss": 5.7001,
      "mean_token_accuracy": 0.17646882086992263,
      "num_tokens": 41990090.0,
      "step": 21465
    },
    {
      "entropy": 6.076330757141113,
      "epoch": 2.146148847903234,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004542071473928455,
      "loss": 5.7345,
      "mean_token_accuracy": 0.1662833347916603,
      "num_tokens": 42000455.0,
      "step": 21470
    },
    {
      "entropy": 6.060870933532715,
      "epoch": 2.1466486729644623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004541855650142557,
      "loss": 5.7427,
      "mean_token_accuracy": 0.17234362959861754,
      "num_tokens": 42010767.0,
      "step": 21475
    },
    {
      "entropy": 6.084904146194458,
      "epoch": 2.147148498025691,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045416397812740323,
      "loss": 5.6438,
      "mean_token_accuracy": 0.17523292154073716,
      "num_tokens": 42020043.0,
      "step": 21480
    },
    {
      "entropy": 6.00084228515625,
      "epoch": 2.1476483230869197,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004541423867328312,
      "loss": 5.7186,
      "mean_token_accuracy": 0.17343796342611312,
      "num_tokens": 42030511.0,
      "step": 21485
    },
    {
      "entropy": 6.075480318069458,
      "epoch": 2.148148148148148,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004541207908310829,
      "loss": 5.6998,
      "mean_token_accuracy": 0.1739243283867836,
      "num_tokens": 42040051.0,
      "step": 21490
    },
    {
      "entropy": 6.060919570922851,
      "epoch": 2.1486479732093766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004540991904227015,
      "loss": 5.7126,
      "mean_token_accuracy": 0.16983721256256104,
      "num_tokens": 42050206.0,
      "step": 21495
    },
    {
      "entropy": 6.070143508911133,
      "epoch": 2.1491477982706053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045407758550823055,
      "loss": 5.6678,
      "mean_token_accuracy": 0.17248247414827347,
      "num_tokens": 42059420.0,
      "step": 21500
    },
    {
      "entropy": 6.090549564361572,
      "epoch": 2.149647623331834,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045405597608821365,
      "loss": 5.6482,
      "mean_token_accuracy": 0.17585248202085496,
      "num_tokens": 42068527.0,
      "step": 21505
    },
    {
      "entropy": 6.092186641693115,
      "epoch": 2.1501474483930623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004540343621631944,
      "loss": 5.6926,
      "mean_token_accuracy": 0.172063010931015,
      "num_tokens": 42078006.0,
      "step": 21510
    },
    {
      "entropy": 6.087264776229858,
      "epoch": 2.150647273454291,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004540127437337167,
      "loss": 5.7339,
      "mean_token_accuracy": 0.1607787787914276,
      "num_tokens": 42087535.0,
      "step": 21515
    },
    {
      "entropy": 6.0491667747497555,
      "epoch": 2.1511470985155197,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004539911208003245,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17441024631261826,
      "num_tokens": 42097259.0,
      "step": 21520
    },
    {
      "entropy": 6.0171784400939945,
      "epoch": 2.151646923576748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004539694933635617,
      "loss": 5.5874,
      "mean_token_accuracy": 0.18415868878364564,
      "num_tokens": 42107216.0,
      "step": 21525
    },
    {
      "entropy": 6.038366508483887,
      "epoch": 2.1521467486379766,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045394786142397253,
      "loss": 5.7698,
      "mean_token_accuracy": 0.16704384982585907,
      "num_tokens": 42116014.0,
      "step": 21530
    },
    {
      "entropy": 6.046961545944214,
      "epoch": 2.1526465736992053,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004539262249821012,
      "loss": 5.6526,
      "mean_token_accuracy": 0.1749047741293907,
      "num_tokens": 42125864.0,
      "step": 21535
    },
    {
      "entropy": 5.964919328689575,
      "epoch": 2.153146398760434,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045390458403849203,
      "loss": 5.5369,
      "mean_token_accuracy": 0.1877485528588295,
      "num_tokens": 42135048.0,
      "step": 21540
    },
    {
      "entropy": 6.035745573043823,
      "epoch": 2.1536462238216623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004538829385936896,
      "loss": 5.6397,
      "mean_token_accuracy": 0.1712017238140106,
      "num_tokens": 42144692.0,
      "step": 21545
    },
    {
      "entropy": 6.146078729629517,
      "epoch": 2.154146048882891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004538612886482385,
      "loss": 5.8196,
      "mean_token_accuracy": 0.1683724120259285,
      "num_tokens": 42155117.0,
      "step": 21550
    },
    {
      "entropy": 6.1041521549224855,
      "epoch": 2.1546458739441197,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004538396342026834,
      "loss": 5.7382,
      "mean_token_accuracy": 0.16777512282133103,
      "num_tokens": 42165428.0,
      "step": 21555
    },
    {
      "entropy": 5.991495561599732,
      "epoch": 2.155145699005348,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004538179752575691,
      "loss": 5.6269,
      "mean_token_accuracy": 0.17486499547958373,
      "num_tokens": 42175841.0,
      "step": 21560
    },
    {
      "entropy": 6.1455882549285885,
      "epoch": 2.1556455240665766,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004537963118134406,
      "loss": 5.6961,
      "mean_token_accuracy": 0.16670922040939332,
      "num_tokens": 42186450.0,
      "step": 21565
    },
    {
      "entropy": 6.063235950469971,
      "epoch": 2.1561453491278053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045377464387084287,
      "loss": 5.7367,
      "mean_token_accuracy": 0.16863785535097123,
      "num_tokens": 42195994.0,
      "step": 21570
    },
    {
      "entropy": 6.07303786277771,
      "epoch": 2.156645174189034,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004537529714303211,
      "loss": 5.6069,
      "mean_token_accuracy": 0.18417707681655884,
      "num_tokens": 42206662.0,
      "step": 21575
    },
    {
      "entropy": 5.952543258666992,
      "epoch": 2.1571449992502623,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045373129449242054,
      "loss": 5.5356,
      "mean_token_accuracy": 0.18109029680490493,
      "num_tokens": 42216598.0,
      "step": 21580
    },
    {
      "entropy": 5.993759059906006,
      "epoch": 2.157644824311491,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004537096130576866,
      "loss": 5.6259,
      "mean_token_accuracy": 0.17885473370552063,
      "num_tokens": 42225869.0,
      "step": 21585
    },
    {
      "entropy": 6.100814485549927,
      "epoch": 2.1581446493727197,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045368792712666486,
      "loss": 5.7933,
      "mean_token_accuracy": 0.1629875421524048,
      "num_tokens": 42234749.0,
      "step": 21590
    },
    {
      "entropy": 6.099698257446289,
      "epoch": 2.158644474433948,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004536662366999008,
      "loss": 5.6902,
      "mean_token_accuracy": 0.16453911662101744,
      "num_tokens": 42243503.0,
      "step": 21595
    },
    {
      "entropy": 6.105169677734375,
      "epoch": 2.1591442994951766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004536445417779402,
      "loss": 5.754,
      "mean_token_accuracy": 0.16610672622919082,
      "num_tokens": 42253888.0,
      "step": 21600
    },
    {
      "entropy": 6.126466178894043,
      "epoch": 2.1596441245564053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045362284236132895,
      "loss": 5.7468,
      "mean_token_accuracy": 0.17409078329801558,
      "num_tokens": 42264052.0,
      "step": 21605
    },
    {
      "entropy": 6.096316385269165,
      "epoch": 2.160143949617634,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045360113845061283,
      "loss": 5.7181,
      "mean_token_accuracy": 0.17177004963159562,
      "num_tokens": 42273749.0,
      "step": 21610
    },
    {
      "entropy": 6.035578680038452,
      "epoch": 2.1606437746788623,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045357943004633813,
      "loss": 5.6655,
      "mean_token_accuracy": 0.17622694075107576,
      "num_tokens": 42282209.0,
      "step": 21615
    },
    {
      "entropy": 6.045993614196777,
      "epoch": 2.161143599740091,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00045355771714905086,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17359004467725753,
      "num_tokens": 42291941.0,
      "step": 21620
    },
    {
      "entropy": 6.119871616363525,
      "epoch": 2.1616434248013197,
      "grad_norm": 1.234375,
      "learning_rate": 0.00045353599975929746,
      "loss": 5.7483,
      "mean_token_accuracy": 0.16945235282182694,
      "num_tokens": 42300630.0,
      "step": 21625
    },
    {
      "entropy": 6.050996446609497,
      "epoch": 2.162143249862548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004535142778776241,
      "loss": 5.6004,
      "mean_token_accuracy": 0.17956664264202118,
      "num_tokens": 42309973.0,
      "step": 21630
    },
    {
      "entropy": 6.085428714752197,
      "epoch": 2.1626430749237766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004534925515045776,
      "loss": 5.7972,
      "mean_token_accuracy": 0.16310292780399321,
      "num_tokens": 42319880.0,
      "step": 21635
    },
    {
      "entropy": 6.053393840789795,
      "epoch": 2.1631428999850053,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045347082064070435,
      "loss": 5.5702,
      "mean_token_accuracy": 0.1814407840371132,
      "num_tokens": 42329554.0,
      "step": 21640
    },
    {
      "entropy": 5.996116352081299,
      "epoch": 2.163642725046234,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004534490852865512,
      "loss": 5.5562,
      "mean_token_accuracy": 0.1787356838583946,
      "num_tokens": 42339118.0,
      "step": 21645
    },
    {
      "entropy": 6.041047239303589,
      "epoch": 2.1641425501074623,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00045342734544266493,
      "loss": 5.6073,
      "mean_token_accuracy": 0.1787823960185051,
      "num_tokens": 42349115.0,
      "step": 21650
    },
    {
      "entropy": 6.042277240753174,
      "epoch": 2.164642375168691,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004534056011095926,
      "loss": 5.6155,
      "mean_token_accuracy": 0.1813841089606285,
      "num_tokens": 42357772.0,
      "step": 21655
    },
    {
      "entropy": 6.047502231597901,
      "epoch": 2.1651422002299197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045338385228788124,
      "loss": 5.6538,
      "mean_token_accuracy": 0.17192450314760208,
      "num_tokens": 42366908.0,
      "step": 21660
    },
    {
      "entropy": 5.915102577209472,
      "epoch": 2.165642025291148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00045336209897807813,
      "loss": 5.5405,
      "mean_token_accuracy": 0.19141213297843934,
      "num_tokens": 42377143.0,
      "step": 21665
    },
    {
      "entropy": 6.0820136070251465,
      "epoch": 2.1661418503523766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045334034118073047,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17062043845653535,
      "num_tokens": 42386815.0,
      "step": 21670
    },
    {
      "entropy": 6.051664161682129,
      "epoch": 2.1666416754136053,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045331857889638566,
      "loss": 5.6241,
      "mean_token_accuracy": 0.18142822086811067,
      "num_tokens": 42395821.0,
      "step": 21675
    },
    {
      "entropy": 6.006423664093018,
      "epoch": 2.167141500474834,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004532968121255914,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18069135397672653,
      "num_tokens": 42406757.0,
      "step": 21680
    },
    {
      "entropy": 6.0146812915802,
      "epoch": 2.1676413255360623,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045327504086889513,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1748896837234497,
      "num_tokens": 42417288.0,
      "step": 21685
    },
    {
      "entropy": 5.96293535232544,
      "epoch": 2.168141150597291,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045325326512684466,
      "loss": 5.5034,
      "mean_token_accuracy": 0.1824643313884735,
      "num_tokens": 42427614.0,
      "step": 21690
    },
    {
      "entropy": 6.097332859039307,
      "epoch": 2.1686409756585197,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045323148489998807,
      "loss": 5.7804,
      "mean_token_accuracy": 0.16856229901313782,
      "num_tokens": 42437475.0,
      "step": 21695
    },
    {
      "entropy": 6.029554843902588,
      "epoch": 2.169140800719748,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000453209700188873,
      "loss": 5.5803,
      "mean_token_accuracy": 0.1804562211036682,
      "num_tokens": 42446654.0,
      "step": 21700
    },
    {
      "entropy": 6.031294155120849,
      "epoch": 2.1696406257809766,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045318791099404785,
      "loss": 5.6464,
      "mean_token_accuracy": 0.1702555425465107,
      "num_tokens": 42456952.0,
      "step": 21705
    },
    {
      "entropy": 6.072096157073974,
      "epoch": 2.1701404508422053,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045316611731606075,
      "loss": 5.6475,
      "mean_token_accuracy": 0.17707123458385468,
      "num_tokens": 42466133.0,
      "step": 21710
    },
    {
      "entropy": 6.107791233062744,
      "epoch": 2.170640275903434,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004531443191554598,
      "loss": 5.749,
      "mean_token_accuracy": 0.17108369767665862,
      "num_tokens": 42476068.0,
      "step": 21715
    },
    {
      "entropy": 6.119190216064453,
      "epoch": 2.1711401009646623,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004531225165127938,
      "loss": 5.6598,
      "mean_token_accuracy": 0.17543508112430573,
      "num_tokens": 42485914.0,
      "step": 21720
    },
    {
      "entropy": 5.973965787887574,
      "epoch": 2.171639926025891,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045310070938861107,
      "loss": 5.6613,
      "mean_token_accuracy": 0.17104327976703643,
      "num_tokens": 42495563.0,
      "step": 21725
    },
    {
      "entropy": 6.06595811843872,
      "epoch": 2.1721397510871197,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045307889778346035,
      "loss": 5.6858,
      "mean_token_accuracy": 0.1727832719683647,
      "num_tokens": 42505441.0,
      "step": 21730
    },
    {
      "entropy": 6.166973495483399,
      "epoch": 2.172639576148348,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004530570816978903,
      "loss": 5.7371,
      "mean_token_accuracy": 0.16956319212913512,
      "num_tokens": 42515636.0,
      "step": 21735
    },
    {
      "entropy": 6.118170261383057,
      "epoch": 2.1731394012095766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045303526113245,
      "loss": 5.7163,
      "mean_token_accuracy": 0.17003063559532167,
      "num_tokens": 42525143.0,
      "step": 21740
    },
    {
      "entropy": 6.00959792137146,
      "epoch": 2.1736392262708053,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004530134360876882,
      "loss": 5.6382,
      "mean_token_accuracy": 0.18087038844823838,
      "num_tokens": 42534591.0,
      "step": 21745
    },
    {
      "entropy": 6.140269708633423,
      "epoch": 2.174139051332034,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004529916065641543,
      "loss": 5.7908,
      "mean_token_accuracy": 0.16796426475048065,
      "num_tokens": 42544956.0,
      "step": 21750
    },
    {
      "entropy": 6.065380811691284,
      "epoch": 2.1746388763932623,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004529697725623973,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18077627718448638,
      "num_tokens": 42554383.0,
      "step": 21755
    },
    {
      "entropy": 6.06083345413208,
      "epoch": 2.175138701454491,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045294793408296656,
      "loss": 5.6102,
      "mean_token_accuracy": 0.18747687637805938,
      "num_tokens": 42563703.0,
      "step": 21760
    },
    {
      "entropy": 5.98844542503357,
      "epoch": 2.1756385265157197,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004529260911264117,
      "loss": 5.5972,
      "mean_token_accuracy": 0.17997730523347855,
      "num_tokens": 42572225.0,
      "step": 21765
    },
    {
      "entropy": 6.012959289550781,
      "epoch": 2.176138351576948,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004529042436932822,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1820403143763542,
      "num_tokens": 42582053.0,
      "step": 21770
    },
    {
      "entropy": 6.003145790100097,
      "epoch": 2.1766381766381766,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045288239178412754,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1743568629026413,
      "num_tokens": 42590682.0,
      "step": 21775
    },
    {
      "entropy": 6.055115175247193,
      "epoch": 2.1771380016994053,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004528605353994978,
      "loss": 5.7,
      "mean_token_accuracy": 0.1711456596851349,
      "num_tokens": 42598769.0,
      "step": 21780
    },
    {
      "entropy": 6.0031250476837155,
      "epoch": 2.177637826760634,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004528386745399428,
      "loss": 5.5836,
      "mean_token_accuracy": 0.1783204570412636,
      "num_tokens": 42608126.0,
      "step": 21785
    },
    {
      "entropy": 6.088155937194824,
      "epoch": 2.1781376518218623,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00045281680920601235,
      "loss": 5.762,
      "mean_token_accuracy": 0.1736542358994484,
      "num_tokens": 42617602.0,
      "step": 21790
    },
    {
      "entropy": 6.050712013244629,
      "epoch": 2.178637476883091,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004527949393982569,
      "loss": 5.6037,
      "mean_token_accuracy": 0.1778274416923523,
      "num_tokens": 42627316.0,
      "step": 21795
    },
    {
      "entropy": 6.087167263031006,
      "epoch": 2.1791373019443196,
      "grad_norm": 1.109375,
      "learning_rate": 0.00045277306511722645,
      "loss": 5.6423,
      "mean_token_accuracy": 0.17012511640787126,
      "num_tokens": 42636872.0,
      "step": 21800
    },
    {
      "entropy": 6.067049264907837,
      "epoch": 2.179637127005548,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004527511863634714,
      "loss": 5.6464,
      "mean_token_accuracy": 0.17870114594697953,
      "num_tokens": 42645754.0,
      "step": 21805
    },
    {
      "entropy": 5.989529705047607,
      "epoch": 2.1801369520667766,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004527293031375422,
      "loss": 5.6379,
      "mean_token_accuracy": 0.17647577822208405,
      "num_tokens": 42655219.0,
      "step": 21810
    },
    {
      "entropy": 6.03337516784668,
      "epoch": 2.1806367771280053,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004527074154399896,
      "loss": 5.6597,
      "mean_token_accuracy": 0.1794843003153801,
      "num_tokens": 42664398.0,
      "step": 21815
    },
    {
      "entropy": 6.040105485916138,
      "epoch": 2.181136602189234,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00045268552327136415,
      "loss": 5.7148,
      "mean_token_accuracy": 0.1770220786333084,
      "num_tokens": 42674513.0,
      "step": 21820
    },
    {
      "entropy": 6.102819490432739,
      "epoch": 2.1816364272504623,
      "grad_norm": 1.359375,
      "learning_rate": 0.00045266362663221656,
      "loss": 5.682,
      "mean_token_accuracy": 0.18245375156402588,
      "num_tokens": 42684379.0,
      "step": 21825
    },
    {
      "entropy": 5.9896996974945065,
      "epoch": 2.182136252311691,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00045264172552309786,
      "loss": 5.617,
      "mean_token_accuracy": 0.17842446714639665,
      "num_tokens": 42694020.0,
      "step": 21830
    },
    {
      "entropy": 6.031555032730102,
      "epoch": 2.1826360773729196,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004526198199445591,
      "loss": 5.6058,
      "mean_token_accuracy": 0.17737309634685516,
      "num_tokens": 42703882.0,
      "step": 21835
    },
    {
      "entropy": 6.072951364517212,
      "epoch": 2.183135902434148,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045259790989715133,
      "loss": 5.6411,
      "mean_token_accuracy": 0.17604389190673828,
      "num_tokens": 42713148.0,
      "step": 21840
    },
    {
      "entropy": 6.154191017150879,
      "epoch": 2.1836357274953766,
      "grad_norm": 1.421875,
      "learning_rate": 0.00045257599538142587,
      "loss": 5.7632,
      "mean_token_accuracy": 0.1739290773868561,
      "num_tokens": 42722177.0,
      "step": 21845
    },
    {
      "entropy": 6.093825149536133,
      "epoch": 2.1841355525566053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045255407639793407,
      "loss": 5.7231,
      "mean_token_accuracy": 0.17204727828502656,
      "num_tokens": 42732454.0,
      "step": 21850
    },
    {
      "entropy": 6.1231378555297855,
      "epoch": 2.1846353776178336,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004525321529472274,
      "loss": 5.7708,
      "mean_token_accuracy": 0.16729852557182312,
      "num_tokens": 42742357.0,
      "step": 21855
    },
    {
      "entropy": 6.071427679061889,
      "epoch": 2.1851352026790622,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045251022502985737,
      "loss": 5.6321,
      "mean_token_accuracy": 0.17326045632362366,
      "num_tokens": 42751703.0,
      "step": 21860
    },
    {
      "entropy": 6.072061824798584,
      "epoch": 2.185635027740291,
      "grad_norm": 1.125,
      "learning_rate": 0.0004524882926463758,
      "loss": 5.6588,
      "mean_token_accuracy": 0.17383660823106767,
      "num_tokens": 42762198.0,
      "step": 21865
    },
    {
      "entropy": 6.076790046691895,
      "epoch": 2.1861348528015196,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004524663557973345,
      "loss": 5.7216,
      "mean_token_accuracy": 0.16407134830951692,
      "num_tokens": 42772348.0,
      "step": 21870
    },
    {
      "entropy": 6.0131713390350345,
      "epoch": 2.186634677862748,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045244441448328535,
      "loss": 5.6383,
      "mean_token_accuracy": 0.17364995628595353,
      "num_tokens": 42782592.0,
      "step": 21875
    },
    {
      "entropy": 6.058275270462036,
      "epoch": 2.1871345029239766,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045242246870478035,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17835331857204437,
      "num_tokens": 42792452.0,
      "step": 21880
    },
    {
      "entropy": 6.032591390609741,
      "epoch": 2.1876343279852053,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004524005184623717,
      "loss": 5.6441,
      "mean_token_accuracy": 0.181292162835598,
      "num_tokens": 42801545.0,
      "step": 21885
    },
    {
      "entropy": 6.095549440383911,
      "epoch": 2.1881341530464335,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004523785637566117,
      "loss": 5.6282,
      "mean_token_accuracy": 0.1765915721654892,
      "num_tokens": 42810504.0,
      "step": 21890
    },
    {
      "entropy": 6.028067255020142,
      "epoch": 2.1886339781076622,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004523566045880527,
      "loss": 5.6495,
      "mean_token_accuracy": 0.17564905136823655,
      "num_tokens": 42820226.0,
      "step": 21895
    },
    {
      "entropy": 6.12649974822998,
      "epoch": 2.189133803168891,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045233464095724706,
      "loss": 5.6859,
      "mean_token_accuracy": 0.1660838395357132,
      "num_tokens": 42829707.0,
      "step": 21900
    },
    {
      "entropy": 6.051554250717163,
      "epoch": 2.1896336282301196,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045231267286474755,
      "loss": 5.6806,
      "mean_token_accuracy": 0.16902831345796585,
      "num_tokens": 42840078.0,
      "step": 21905
    },
    {
      "entropy": 6.048523044586181,
      "epoch": 2.190133453291348,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004522907003111068,
      "loss": 5.6068,
      "mean_token_accuracy": 0.1871507376432419,
      "num_tokens": 42849998.0,
      "step": 21910
    },
    {
      "entropy": 6.015602684020996,
      "epoch": 2.1906332783525766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004522687232968776,
      "loss": 5.5743,
      "mean_token_accuracy": 0.18271636962890625,
      "num_tokens": 42859290.0,
      "step": 21915
    },
    {
      "entropy": 5.969098281860352,
      "epoch": 2.1911331034138053,
      "grad_norm": 1.296875,
      "learning_rate": 0.00045224674182261306,
      "loss": 5.6356,
      "mean_token_accuracy": 0.17439277172088624,
      "num_tokens": 42867945.0,
      "step": 21920
    },
    {
      "entropy": 5.945876359939575,
      "epoch": 2.1916329284750335,
      "grad_norm": 1.125,
      "learning_rate": 0.00045222475588886597,
      "loss": 5.5249,
      "mean_token_accuracy": 0.18588104993104934,
      "num_tokens": 42877527.0,
      "step": 21925
    },
    {
      "entropy": 6.010125303268433,
      "epoch": 2.1921327535362622,
      "grad_norm": 1.34375,
      "learning_rate": 0.00045220276549618973,
      "loss": 5.6445,
      "mean_token_accuracy": 0.18437591195106506,
      "num_tokens": 42887049.0,
      "step": 21930
    },
    {
      "entropy": 6.057902860641479,
      "epoch": 2.192632578597491,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00045218077064513744,
      "loss": 5.5971,
      "mean_token_accuracy": 0.17969248294830323,
      "num_tokens": 42896177.0,
      "step": 21935
    },
    {
      "entropy": 6.071481370925904,
      "epoch": 2.1931324036587196,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004521587713362626,
      "loss": 5.7207,
      "mean_token_accuracy": 0.1680138699710369,
      "num_tokens": 42906244.0,
      "step": 21940
    },
    {
      "entropy": 5.9518914222717285,
      "epoch": 2.193632228719948,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004521367675701187,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17892550975084304,
      "num_tokens": 42915776.0,
      "step": 21945
    },
    {
      "entropy": 6.081822299957276,
      "epoch": 2.1941320537811766,
      "grad_norm": 1.4375,
      "learning_rate": 0.0004521147593472592,
      "loss": 5.6183,
      "mean_token_accuracy": 0.18416496366262436,
      "num_tokens": 42926411.0,
      "step": 21950
    },
    {
      "entropy": 6.002219915390015,
      "epoch": 2.1946318788424053,
      "grad_norm": 1.265625,
      "learning_rate": 0.00045209274666823795,
      "loss": 5.503,
      "mean_token_accuracy": 0.18526443094015121,
      "num_tokens": 42935540.0,
      "step": 21955
    },
    {
      "entropy": 5.992690324783325,
      "epoch": 2.1951317039036335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004520707295336088,
      "loss": 5.6486,
      "mean_token_accuracy": 0.17101467698812484,
      "num_tokens": 42944796.0,
      "step": 21960
    },
    {
      "entropy": 6.1037092208862305,
      "epoch": 2.1956315289648622,
      "grad_norm": 1.140625,
      "learning_rate": 0.00045204870794392574,
      "loss": 5.7388,
      "mean_token_accuracy": 0.1653115451335907,
      "num_tokens": 42954141.0,
      "step": 21965
    },
    {
      "entropy": 6.102824783325195,
      "epoch": 2.196131354026091,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045202668189974263,
      "loss": 5.7156,
      "mean_token_accuracy": 0.16959594190120697,
      "num_tokens": 42963201.0,
      "step": 21970
    },
    {
      "entropy": 6.013363027572632,
      "epoch": 2.1966311790873196,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004520046514016138,
      "loss": 5.6006,
      "mean_token_accuracy": 0.17957703769207,
      "num_tokens": 42973798.0,
      "step": 21975
    },
    {
      "entropy": 6.061620140075684,
      "epoch": 2.197131004148548,
      "grad_norm": 1.28125,
      "learning_rate": 0.00045198261645009347,
      "loss": 5.7663,
      "mean_token_accuracy": 0.16192597299814224,
      "num_tokens": 42983587.0,
      "step": 21980
    },
    {
      "entropy": 6.04732871055603,
      "epoch": 2.1976308292097766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000451960577045736,
      "loss": 5.7403,
      "mean_token_accuracy": 0.17270044833421708,
      "num_tokens": 42994802.0,
      "step": 21985
    },
    {
      "entropy": 6.122683429718018,
      "epoch": 2.1981306542710053,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000451938533189096,
      "loss": 5.6426,
      "mean_token_accuracy": 0.17314900308847428,
      "num_tokens": 43003427.0,
      "step": 21990
    },
    {
      "entropy": 5.98048505783081,
      "epoch": 2.1986304793322335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004519164848807281,
      "loss": 5.6045,
      "mean_token_accuracy": 0.17703433483839034,
      "num_tokens": 43013268.0,
      "step": 21995
    },
    {
      "entropy": 6.058509778976441,
      "epoch": 2.1991303043934622,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045189443212118686,
      "loss": 5.6306,
      "mean_token_accuracy": 0.17500248402357102,
      "num_tokens": 43022389.0,
      "step": 22000
    },
    {
      "entropy": 6.067094945907593,
      "epoch": 2.199630129454691,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045187237491102727,
      "loss": 5.6247,
      "mean_token_accuracy": 0.17964363247156143,
      "num_tokens": 43033168.0,
      "step": 22005
    },
    {
      "entropy": 6.042067241668701,
      "epoch": 2.200129954515919,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045185031325080426,
      "loss": 5.7038,
      "mean_token_accuracy": 0.17055662125349044,
      "num_tokens": 43043002.0,
      "step": 22010
    },
    {
      "entropy": 6.057109022140503,
      "epoch": 2.200629779577148,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00045182824714107284,
      "loss": 5.6746,
      "mean_token_accuracy": 0.17811395078897477,
      "num_tokens": 43053474.0,
      "step": 22015
    },
    {
      "entropy": 6.086128807067871,
      "epoch": 2.2011296046383766,
      "grad_norm": 1.3671875,
      "learning_rate": 0.00045180617658238827,
      "loss": 5.6753,
      "mean_token_accuracy": 0.17516648173332214,
      "num_tokens": 43063821.0,
      "step": 22020
    },
    {
      "entropy": 6.052876806259155,
      "epoch": 2.2016294296996053,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004517841015753057,
      "loss": 5.6556,
      "mean_token_accuracy": 0.17615651786327363,
      "num_tokens": 43072981.0,
      "step": 22025
    },
    {
      "entropy": 5.9960111618042,
      "epoch": 2.2021292547608335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004517620221203807,
      "loss": 5.6252,
      "mean_token_accuracy": 0.18304435908794403,
      "num_tokens": 43082992.0,
      "step": 22030
    },
    {
      "entropy": 6.038987398147583,
      "epoch": 2.2026290798220622,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045173993821816873,
      "loss": 5.6987,
      "mean_token_accuracy": 0.17546250224113463,
      "num_tokens": 43092678.0,
      "step": 22035
    },
    {
      "entropy": 5.9675976753234865,
      "epoch": 2.203128904883291,
      "grad_norm": 1.3828125,
      "learning_rate": 0.0004517178498692253,
      "loss": 5.5898,
      "mean_token_accuracy": 0.18891261368989945,
      "num_tokens": 43102672.0,
      "step": 22040
    },
    {
      "entropy": 6.034510374069214,
      "epoch": 2.203628729944519,
      "grad_norm": 1.1875,
      "learning_rate": 0.00045169575707410637,
      "loss": 5.668,
      "mean_token_accuracy": 0.17588375508785248,
      "num_tokens": 43112747.0,
      "step": 22045
    },
    {
      "entropy": 6.027397108078003,
      "epoch": 2.204128555005748,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00045167365983336756,
      "loss": 5.6327,
      "mean_token_accuracy": 0.17682866603136063,
      "num_tokens": 43122370.0,
      "step": 22050
    },
    {
      "entropy": 6.027716875076294,
      "epoch": 2.2046283800669766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00045165155814756497,
      "loss": 5.6475,
      "mean_token_accuracy": 0.18118370026350022,
      "num_tokens": 43131538.0,
      "step": 22055
    },
    {
      "entropy": 6.1900177001953125,
      "epoch": 2.2051282051282053,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045162945201725457,
      "loss": 5.7977,
      "mean_token_accuracy": 0.16386715173721314,
      "num_tokens": 43141669.0,
      "step": 22060
    },
    {
      "entropy": 6.091287612915039,
      "epoch": 2.2056280301894335,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004516073414429927,
      "loss": 5.7579,
      "mean_token_accuracy": 0.17163655459880828,
      "num_tokens": 43151456.0,
      "step": 22065
    },
    {
      "entropy": 6.039238405227661,
      "epoch": 2.2061278552506622,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004515852264253356,
      "loss": 5.655,
      "mean_token_accuracy": 0.17147247195243837,
      "num_tokens": 43160491.0,
      "step": 22070
    },
    {
      "entropy": 6.064098501205445,
      "epoch": 2.206627680311891,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045156310696483953,
      "loss": 5.6425,
      "mean_token_accuracy": 0.17336027473211288,
      "num_tokens": 43170695.0,
      "step": 22075
    },
    {
      "entropy": 6.086498212814331,
      "epoch": 2.207127505373119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004515409830620612,
      "loss": 5.7496,
      "mean_token_accuracy": 0.16989077776670455,
      "num_tokens": 43180542.0,
      "step": 22080
    },
    {
      "entropy": 6.020513248443604,
      "epoch": 2.207627330434348,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004515188547175571,
      "loss": 5.6684,
      "mean_token_accuracy": 0.17119597792625427,
      "num_tokens": 43190739.0,
      "step": 22085
    },
    {
      "entropy": 6.06134352684021,
      "epoch": 2.2081271554955766,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000451496721931884,
      "loss": 5.7726,
      "mean_token_accuracy": 0.17430623024702072,
      "num_tokens": 43201808.0,
      "step": 22090
    },
    {
      "entropy": 6.1003166198730465,
      "epoch": 2.2086269805568053,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004514745847055989,
      "loss": 5.6697,
      "mean_token_accuracy": 0.1713427871465683,
      "num_tokens": 43211061.0,
      "step": 22095
    },
    {
      "entropy": 6.03029842376709,
      "epoch": 2.2091268056180335,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004514524430392586,
      "loss": 5.5639,
      "mean_token_accuracy": 0.18769593834877013,
      "num_tokens": 43220333.0,
      "step": 22100
    },
    {
      "entropy": 5.9788583278656,
      "epoch": 2.2096266306792622,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004514302969334202,
      "loss": 5.6228,
      "mean_token_accuracy": 0.1857643112540245,
      "num_tokens": 43230134.0,
      "step": 22105
    },
    {
      "entropy": 6.022599172592163,
      "epoch": 2.210126455740491,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045140814638864105,
      "loss": 5.6297,
      "mean_token_accuracy": 0.1791934221982956,
      "num_tokens": 43239216.0,
      "step": 22110
    },
    {
      "entropy": 6.027814674377441,
      "epoch": 2.210626280801719,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004513859914054782,
      "loss": 5.7432,
      "mean_token_accuracy": 0.16874404549598693,
      "num_tokens": 43249051.0,
      "step": 22115
    },
    {
      "entropy": 6.211530637741089,
      "epoch": 2.211126105862948,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00045136383198448926,
      "loss": 5.8083,
      "mean_token_accuracy": 0.16471776291728019,
      "num_tokens": 43259507.0,
      "step": 22120
    },
    {
      "entropy": 6.087680053710938,
      "epoch": 2.2116259309241766,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004513416681262317,
      "loss": 5.7268,
      "mean_token_accuracy": 0.1705778107047081,
      "num_tokens": 43269097.0,
      "step": 22125
    },
    {
      "entropy": 6.050832843780517,
      "epoch": 2.2121257559854053,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004513194998312631,
      "loss": 5.6249,
      "mean_token_accuracy": 0.17137878984212876,
      "num_tokens": 43278514.0,
      "step": 22130
    },
    {
      "entropy": 6.084692049026489,
      "epoch": 2.2126255810466335,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004512973271001413,
      "loss": 5.7518,
      "mean_token_accuracy": 0.16480341851711272,
      "num_tokens": 43288563.0,
      "step": 22135
    },
    {
      "entropy": 6.059920787811279,
      "epoch": 2.2131254061078622,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045127514993342413,
      "loss": 5.5771,
      "mean_token_accuracy": 0.17250026762485504,
      "num_tokens": 43298168.0,
      "step": 22140
    },
    {
      "entropy": 5.999436950683593,
      "epoch": 2.213625231169091,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004512529683316695,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17716497629880906,
      "num_tokens": 43308295.0,
      "step": 22145
    },
    {
      "entropy": 6.031063079833984,
      "epoch": 2.214125056230319,
      "grad_norm": 1.375,
      "learning_rate": 0.00045123078229543557,
      "loss": 5.6971,
      "mean_token_accuracy": 0.17268507033586503,
      "num_tokens": 43319283.0,
      "step": 22150
    },
    {
      "entropy": 6.122743129730225,
      "epoch": 2.214624881291548,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004512085918252805,
      "loss": 5.7194,
      "mean_token_accuracy": 0.1751079484820366,
      "num_tokens": 43329000.0,
      "step": 22155
    },
    {
      "entropy": 6.05804967880249,
      "epoch": 2.2151247063527766,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004511863969217625,
      "loss": 5.5654,
      "mean_token_accuracy": 0.1824063166975975,
      "num_tokens": 43338150.0,
      "step": 22160
    },
    {
      "entropy": 6.000379371643066,
      "epoch": 2.2156245314140053,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045116419758544024,
      "loss": 5.6247,
      "mean_token_accuracy": 0.18240000456571578,
      "num_tokens": 43348736.0,
      "step": 22165
    },
    {
      "entropy": 6.023730373382568,
      "epoch": 2.2161243564752335,
      "grad_norm": 1.125,
      "learning_rate": 0.00045114199381687207,
      "loss": 5.6509,
      "mean_token_accuracy": 0.18076959997415543,
      "num_tokens": 43359208.0,
      "step": 22170
    },
    {
      "entropy": 6.031237459182739,
      "epoch": 2.2166241815364622,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004511197856166166,
      "loss": 5.6298,
      "mean_token_accuracy": 0.17659648805856704,
      "num_tokens": 43368503.0,
      "step": 22175
    },
    {
      "entropy": 6.008730411529541,
      "epoch": 2.217124006597691,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004510975729852327,
      "loss": 5.6737,
      "mean_token_accuracy": 0.16624808460474014,
      "num_tokens": 43378778.0,
      "step": 22180
    },
    {
      "entropy": 6.053968477249145,
      "epoch": 2.217623831658919,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004510753559232792,
      "loss": 5.7113,
      "mean_token_accuracy": 0.17446492463350297,
      "num_tokens": 43389983.0,
      "step": 22185
    },
    {
      "entropy": 6.057756233215332,
      "epoch": 2.218123656720148,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045105313443131496,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17128798067569734,
      "num_tokens": 43400008.0,
      "step": 22190
    },
    {
      "entropy": 6.095964241027832,
      "epoch": 2.2186234817813766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004510309085098992,
      "loss": 5.6496,
      "mean_token_accuracy": 0.17622294872999192,
      "num_tokens": 43409626.0,
      "step": 22195
    },
    {
      "entropy": 6.112526512145996,
      "epoch": 2.2191233068426053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00045100867815959107,
      "loss": 5.6993,
      "mean_token_accuracy": 0.16991682052612306,
      "num_tokens": 43419277.0,
      "step": 22200
    },
    {
      "entropy": 5.987932395935059,
      "epoch": 2.2196231319038335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004509864433809499,
      "loss": 5.6579,
      "mean_token_accuracy": 0.17732749432325362,
      "num_tokens": 43429226.0,
      "step": 22205
    },
    {
      "entropy": 6.082731103897094,
      "epoch": 2.2201229569650622,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004509642041745351,
      "loss": 5.7274,
      "mean_token_accuracy": 0.16624752581119537,
      "num_tokens": 43438921.0,
      "step": 22210
    },
    {
      "entropy": 6.06203818321228,
      "epoch": 2.220622782026291,
      "grad_norm": 1.125,
      "learning_rate": 0.0004509419605409062,
      "loss": 5.6911,
      "mean_token_accuracy": 0.17973775118589402,
      "num_tokens": 43449450.0,
      "step": 22215
    },
    {
      "entropy": 6.113070154190064,
      "epoch": 2.221122607087519,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004509197124806229,
      "loss": 5.6665,
      "mean_token_accuracy": 0.17515143305063247,
      "num_tokens": 43458587.0,
      "step": 22220
    },
    {
      "entropy": 6.044864463806152,
      "epoch": 2.221622432148748,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00045089745999424486,
      "loss": 5.6795,
      "mean_token_accuracy": 0.17777097672224046,
      "num_tokens": 43467983.0,
      "step": 22225
    },
    {
      "entropy": 6.033739471435547,
      "epoch": 2.2221222572099766,
      "grad_norm": 1.171875,
      "learning_rate": 0.000450875203082332,
      "loss": 5.679,
      "mean_token_accuracy": 0.16325077116489412,
      "num_tokens": 43476887.0,
      "step": 22230
    },
    {
      "entropy": 6.052248191833496,
      "epoch": 2.2226220822712053,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004508529417454442,
      "loss": 5.6219,
      "mean_token_accuracy": 0.17426965534687042,
      "num_tokens": 43485871.0,
      "step": 22235
    },
    {
      "entropy": 6.069500207901001,
      "epoch": 2.2231219073324335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00045083067598414176,
      "loss": 5.6953,
      "mean_token_accuracy": 0.1676015317440033,
      "num_tokens": 43496062.0,
      "step": 22240
    },
    {
      "entropy": 6.101183891296387,
      "epoch": 2.2236217323936622,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045080840579898466,
      "loss": 5.742,
      "mean_token_accuracy": 0.16793988645076752,
      "num_tokens": 43505391.0,
      "step": 22245
    },
    {
      "entropy": 6.055177211761475,
      "epoch": 2.224121557454891,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004507861311905334,
      "loss": 5.6813,
      "mean_token_accuracy": 0.16928481608629226,
      "num_tokens": 43515500.0,
      "step": 22250
    },
    {
      "entropy": 5.991426086425781,
      "epoch": 2.224621382516119,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004507638521593482,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18511694371700288,
      "num_tokens": 43524493.0,
      "step": 22255
    },
    {
      "entropy": 6.1207671642303465,
      "epoch": 2.225121207577348,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004507415687059898,
      "loss": 5.6832,
      "mean_token_accuracy": 0.16908760368824005,
      "num_tokens": 43534075.0,
      "step": 22260
    },
    {
      "entropy": 6.06439437866211,
      "epoch": 2.2256210326385766,
      "grad_norm": 1.65625,
      "learning_rate": 0.00045071928083101874,
      "loss": 5.6634,
      "mean_token_accuracy": 0.17832424491643906,
      "num_tokens": 43543288.0,
      "step": 22265
    },
    {
      "entropy": 6.04383111000061,
      "epoch": 2.2261208576998053,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004506969885349957,
      "loss": 5.7062,
      "mean_token_accuracy": 0.17659382671117782,
      "num_tokens": 43553491.0,
      "step": 22270
    },
    {
      "entropy": 6.077741813659668,
      "epoch": 2.2266206827610335,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004506746918184818,
      "loss": 5.6866,
      "mean_token_accuracy": 0.17610663324594497,
      "num_tokens": 43563904.0,
      "step": 22275
    },
    {
      "entropy": 6.010642766952515,
      "epoch": 2.2271205078222622,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004506523906820378,
      "loss": 5.6852,
      "mean_token_accuracy": 0.17096028178930284,
      "num_tokens": 43573231.0,
      "step": 22280
    },
    {
      "entropy": 6.0773398876190186,
      "epoch": 2.227620332883491,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004506300851262248,
      "loss": 5.6692,
      "mean_token_accuracy": 0.16963550448417664,
      "num_tokens": 43584538.0,
      "step": 22285
    },
    {
      "entropy": 6.1204675197601315,
      "epoch": 2.228120157944719,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045060777515160404,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18678200989961624,
      "num_tokens": 43594665.0,
      "step": 22290
    },
    {
      "entropy": 6.068657684326172,
      "epoch": 2.228619983005948,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004505854607587369,
      "loss": 5.701,
      "mean_token_accuracy": 0.17283949255943298,
      "num_tokens": 43604681.0,
      "step": 22295
    },
    {
      "entropy": 5.9363751888275145,
      "epoch": 2.2291198080671766,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004505631419481847,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17675758004188538,
      "num_tokens": 43614186.0,
      "step": 22300
    },
    {
      "entropy": 6.103043937683106,
      "epoch": 2.2296196331284053,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004505408187205091,
      "loss": 5.6818,
      "mean_token_accuracy": 0.16714295446872712,
      "num_tokens": 43624202.0,
      "step": 22305
    },
    {
      "entropy": 6.124151372909546,
      "epoch": 2.2301194581896335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004505184910762716,
      "loss": 5.6987,
      "mean_token_accuracy": 0.17141597718000412,
      "num_tokens": 43634303.0,
      "step": 22310
    },
    {
      "entropy": 6.1126110553741455,
      "epoch": 2.2306192832508622,
      "grad_norm": 1.203125,
      "learning_rate": 0.00045049615901603406,
      "loss": 5.7158,
      "mean_token_accuracy": 0.1771593138575554,
      "num_tokens": 43644045.0,
      "step": 22315
    },
    {
      "entropy": 6.07891206741333,
      "epoch": 2.231119108312091,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004504738225403583,
      "loss": 5.7433,
      "mean_token_accuracy": 0.1725505292415619,
      "num_tokens": 43654220.0,
      "step": 22320
    },
    {
      "entropy": 6.061905336380005,
      "epoch": 2.231618933373319,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004504514816498064,
      "loss": 5.7029,
      "mean_token_accuracy": 0.17058081775903702,
      "num_tokens": 43665124.0,
      "step": 22325
    },
    {
      "entropy": 6.050745487213135,
      "epoch": 2.232118758434548,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045042913634494024,
      "loss": 5.5913,
      "mean_token_accuracy": 0.18154880702495574,
      "num_tokens": 43674029.0,
      "step": 22330
    },
    {
      "entropy": 6.035893964767456,
      "epoch": 2.2326185834957766,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004504067866263222,
      "loss": 5.6627,
      "mean_token_accuracy": 0.1772293195128441,
      "num_tokens": 43683019.0,
      "step": 22335
    },
    {
      "entropy": 6.034237861633301,
      "epoch": 2.2331184085570053,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004503844324945146,
      "loss": 5.5968,
      "mean_token_accuracy": 0.17474510669708251,
      "num_tokens": 43692762.0,
      "step": 22340
    },
    {
      "entropy": 6.0099798202514645,
      "epoch": 2.2336182336182335,
      "grad_norm": 1.125,
      "learning_rate": 0.0004503620739500797,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17870785146951676,
      "num_tokens": 43701899.0,
      "step": 22345
    },
    {
      "entropy": 6.057866430282592,
      "epoch": 2.2341180586794622,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004503397109935802,
      "loss": 5.6058,
      "mean_token_accuracy": 0.17802409082651138,
      "num_tokens": 43711233.0,
      "step": 22350
    },
    {
      "entropy": 5.924233722686767,
      "epoch": 2.234617883740691,
      "grad_norm": 1.25,
      "learning_rate": 0.0004503173436255787,
      "loss": 5.4983,
      "mean_token_accuracy": 0.19454552829265595,
      "num_tokens": 43721754.0,
      "step": 22355
    },
    {
      "entropy": 5.992033195495606,
      "epoch": 2.235117708801919,
      "grad_norm": 1.171875,
      "learning_rate": 0.000450294971846638,
      "loss": 5.6298,
      "mean_token_accuracy": 0.1832828179001808,
      "num_tokens": 43731114.0,
      "step": 22360
    },
    {
      "entropy": 6.117801284790039,
      "epoch": 2.235617533863148,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004502725956573208,
      "loss": 5.6639,
      "mean_token_accuracy": 0.17628454566001892,
      "num_tokens": 43741011.0,
      "step": 22365
    },
    {
      "entropy": 6.1154687881469725,
      "epoch": 2.2361173589243766,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004502502150581903,
      "loss": 5.755,
      "mean_token_accuracy": 0.17173747569322587,
      "num_tokens": 43750075.0,
      "step": 22370
    },
    {
      "entropy": 5.991245222091675,
      "epoch": 2.2366171839856053,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004502278300498094,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17549355030059816,
      "num_tokens": 43759826.0,
      "step": 22375
    },
    {
      "entropy": 6.048260307312011,
      "epoch": 2.2371170090468335,
      "grad_norm": 1.09375,
      "learning_rate": 0.00045020544063274147,
      "loss": 5.6371,
      "mean_token_accuracy": 0.17687432020902633,
      "num_tokens": 43769326.0,
      "step": 22380
    },
    {
      "entropy": 6.061201763153076,
      "epoch": 2.2376168341080622,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00045018304680754975,
      "loss": 5.601,
      "mean_token_accuracy": 0.17728709280490876,
      "num_tokens": 43778549.0,
      "step": 22385
    },
    {
      "entropy": 6.1114034175872805,
      "epoch": 2.238116659169291,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004501606485747975,
      "loss": 5.7746,
      "mean_token_accuracy": 0.16107561439275742,
      "num_tokens": 43789032.0,
      "step": 22390
    },
    {
      "entropy": 6.112794351577759,
      "epoch": 2.238616484230519,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004501382459350485,
      "loss": 5.7403,
      "mean_token_accuracy": 0.1763688802719116,
      "num_tokens": 43798246.0,
      "step": 22395
    },
    {
      "entropy": 6.072865629196167,
      "epoch": 2.239116309291748,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004501158388888663,
      "loss": 5.6858,
      "mean_token_accuracy": 0.17559813857078552,
      "num_tokens": 43807572.0,
      "step": 22400
    },
    {
      "entropy": 6.0150528907775875,
      "epoch": 2.2396161343529766,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004500934274368147,
      "loss": 5.6831,
      "mean_token_accuracy": 0.17309265583753586,
      "num_tokens": 43816912.0,
      "step": 22405
    },
    {
      "entropy": 6.002626657485962,
      "epoch": 2.240115959414205,
      "grad_norm": 1.21875,
      "learning_rate": 0.00045007101157945747,
      "loss": 5.6573,
      "mean_token_accuracy": 0.1720070406794548,
      "num_tokens": 43826386.0,
      "step": 22410
    },
    {
      "entropy": 6.072042751312256,
      "epoch": 2.2406157844754335,
      "grad_norm": 1.15625,
      "learning_rate": 0.00045004859131735867,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17359066307544707,
      "num_tokens": 43836456.0,
      "step": 22415
    },
    {
      "entropy": 6.003937244415283,
      "epoch": 2.2411156095366622,
      "grad_norm": 1.171875,
      "learning_rate": 0.00045002616665108233,
      "loss": 5.5564,
      "mean_token_accuracy": 0.18237012177705764,
      "num_tokens": 43846360.0,
      "step": 22420
    },
    {
      "entropy": 6.030667114257812,
      "epoch": 2.241615434597891,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004500037375811926,
      "loss": 5.6428,
      "mean_token_accuracy": 0.17764281779527663,
      "num_tokens": 43855999.0,
      "step": 22425
    },
    {
      "entropy": 6.039393281936645,
      "epoch": 2.242115259659119,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044998130410825395,
      "loss": 5.6272,
      "mean_token_accuracy": 0.17685568034648896,
      "num_tokens": 43865497.0,
      "step": 22430
    },
    {
      "entropy": 6.040842151641845,
      "epoch": 2.242615084720348,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004499588662328307,
      "loss": 5.6217,
      "mean_token_accuracy": 0.17351048439741135,
      "num_tokens": 43874450.0,
      "step": 22435
    },
    {
      "entropy": 6.021168184280396,
      "epoch": 2.2431149097815766,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004499364239554874,
      "loss": 5.6258,
      "mean_token_accuracy": 0.17424180656671523,
      "num_tokens": 43885552.0,
      "step": 22440
    },
    {
      "entropy": 6.119274854660034,
      "epoch": 2.243614734842805,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004499139772767887,
      "loss": 5.7228,
      "mean_token_accuracy": 0.1769503727555275,
      "num_tokens": 43894928.0,
      "step": 22445
    },
    {
      "entropy": 6.065196847915649,
      "epoch": 2.2441145599040335,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004498915261972993,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1837917298078537,
      "num_tokens": 43904841.0,
      "step": 22450
    },
    {
      "entropy": 6.001014518737793,
      "epoch": 2.244614384965262,
      "grad_norm": 1.125,
      "learning_rate": 0.00044986907071758417,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18431970626115798,
      "num_tokens": 43914690.0,
      "step": 22455
    },
    {
      "entropy": 5.923689556121826,
      "epoch": 2.245114210026491,
      "grad_norm": 1.125,
      "learning_rate": 0.0004498466108382082,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18531722128391265,
      "num_tokens": 43925237.0,
      "step": 22460
    },
    {
      "entropy": 6.070363712310791,
      "epoch": 2.245614035087719,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004498241465597365,
      "loss": 5.7041,
      "mean_token_accuracy": 0.17300845831632614,
      "num_tokens": 43934635.0,
      "step": 22465
    },
    {
      "entropy": 6.063670015335083,
      "epoch": 2.246113860148948,
      "grad_norm": 1.3671875,
      "learning_rate": 0.0004498016778827342,
      "loss": 5.658,
      "mean_token_accuracy": 0.18038448095321655,
      "num_tokens": 43943705.0,
      "step": 22470
    },
    {
      "entropy": 6.039931106567383,
      "epoch": 2.2466136852101766,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004497792048077667,
      "loss": 5.606,
      "mean_token_accuracy": 0.17851114869117737,
      "num_tokens": 43953559.0,
      "step": 22475
    },
    {
      "entropy": 6.0384022235870365,
      "epoch": 2.247113510271405,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004497567273353994,
      "loss": 5.6815,
      "mean_token_accuracy": 0.17560024708509445,
      "num_tokens": 43963940.0,
      "step": 22480
    },
    {
      "entropy": 6.031838512420654,
      "epoch": 2.2476133353326335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044973424546619774,
      "loss": 5.5714,
      "mean_token_accuracy": 0.18169790953397752,
      "num_tokens": 43973364.0,
      "step": 22485
    },
    {
      "entropy": 6.09574294090271,
      "epoch": 2.248113160393862,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004497117592007275,
      "loss": 5.6369,
      "mean_token_accuracy": 0.17298363894224167,
      "num_tokens": 43982531.0,
      "step": 22490
    },
    {
      "entropy": 6.096007823944092,
      "epoch": 2.248612985455091,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004496892685395543,
      "loss": 5.7946,
      "mean_token_accuracy": 0.16701375246047973,
      "num_tokens": 43994244.0,
      "step": 22495
    },
    {
      "entropy": 5.990036582946777,
      "epoch": 2.249112810516319,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004496667734832441,
      "loss": 5.62,
      "mean_token_accuracy": 0.17978659570217131,
      "num_tokens": 44003809.0,
      "step": 22500
    },
    {
      "entropy": 6.064924860000611,
      "epoch": 2.249612635577548,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044964427403236274,
      "loss": 5.6466,
      "mean_token_accuracy": 0.171216182410717,
      "num_tokens": 44014850.0,
      "step": 22505
    },
    {
      "entropy": 5.967130947113037,
      "epoch": 2.2501124606387766,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004496217701874765,
      "loss": 5.5284,
      "mean_token_accuracy": 0.1821651816368103,
      "num_tokens": 44025594.0,
      "step": 22510
    },
    {
      "entropy": 6.060387039184571,
      "epoch": 2.250612285700005,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004495992619491514,
      "loss": 5.7424,
      "mean_token_accuracy": 0.17245593369007112,
      "num_tokens": 44034681.0,
      "step": 22515
    },
    {
      "entropy": 6.0845990657806395,
      "epoch": 2.2511121107612335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004495767493179538,
      "loss": 5.6364,
      "mean_token_accuracy": 0.17254023402929305,
      "num_tokens": 44044149.0,
      "step": 22520
    },
    {
      "entropy": 6.091122341156006,
      "epoch": 2.251611935822462,
      "grad_norm": 1.25,
      "learning_rate": 0.0004495542322944501,
      "loss": 5.6824,
      "mean_token_accuracy": 0.17178913205862045,
      "num_tokens": 44054238.0,
      "step": 22525
    },
    {
      "entropy": 6.014432811737061,
      "epoch": 2.2521117608836905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044953171087920693,
      "loss": 5.7333,
      "mean_token_accuracy": 0.1670988455414772,
      "num_tokens": 44063826.0,
      "step": 22530
    },
    {
      "entropy": 6.1266919612884525,
      "epoch": 2.252611585944919,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004495091850727907,
      "loss": 5.7271,
      "mean_token_accuracy": 0.16965164989233017,
      "num_tokens": 44074588.0,
      "step": 22535
    },
    {
      "entropy": 6.040569496154785,
      "epoch": 2.253111411006148,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044948665487576833,
      "loss": 5.6567,
      "mean_token_accuracy": 0.17420101910829544,
      "num_tokens": 44084737.0,
      "step": 22540
    },
    {
      "entropy": 6.125871276855468,
      "epoch": 2.2536112360673766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044946412028870657,
      "loss": 5.6181,
      "mean_token_accuracy": 0.1773032248020172,
      "num_tokens": 44093542.0,
      "step": 22545
    },
    {
      "entropy": 6.097443246841431,
      "epoch": 2.254111061128605,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004494415813121725,
      "loss": 5.6876,
      "mean_token_accuracy": 0.16689204052090645,
      "num_tokens": 44104681.0,
      "step": 22550
    },
    {
      "entropy": 6.132251501083374,
      "epoch": 2.2546108861898335,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004494190379467331,
      "loss": 5.851,
      "mean_token_accuracy": 0.16538407951593398,
      "num_tokens": 44116567.0,
      "step": 22555
    },
    {
      "entropy": 6.028368520736694,
      "epoch": 2.255110711251062,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044939649019295564,
      "loss": 5.6478,
      "mean_token_accuracy": 0.1739356592297554,
      "num_tokens": 44125176.0,
      "step": 22560
    },
    {
      "entropy": 6.055741548538208,
      "epoch": 2.2556105363122905,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004493739380514073,
      "loss": 5.6763,
      "mean_token_accuracy": 0.17245809733867645,
      "num_tokens": 44136371.0,
      "step": 22565
    },
    {
      "entropy": 6.042793607711792,
      "epoch": 2.256110361373519,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004493513815226556,
      "loss": 5.6023,
      "mean_token_accuracy": 0.18558013886213304,
      "num_tokens": 44146456.0,
      "step": 22570
    },
    {
      "entropy": 5.99357647895813,
      "epoch": 2.256610186434748,
      "grad_norm": 1.171875,
      "learning_rate": 0.000449328820607268,
      "loss": 5.6262,
      "mean_token_accuracy": 0.18093568831682205,
      "num_tokens": 44155719.0,
      "step": 22575
    },
    {
      "entropy": 5.983214235305786,
      "epoch": 2.2571100114959766,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004493062553058121,
      "loss": 5.6326,
      "mean_token_accuracy": 0.18207505643367766,
      "num_tokens": 44165994.0,
      "step": 22580
    },
    {
      "entropy": 6.0403584957122805,
      "epoch": 2.257609836557205,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044928368561885566,
      "loss": 5.7575,
      "mean_token_accuracy": 0.17001302242279054,
      "num_tokens": 44175726.0,
      "step": 22585
    },
    {
      "entropy": 6.137831306457519,
      "epoch": 2.2581096616184335,
      "grad_norm": 1.25,
      "learning_rate": 0.00044926111154696655,
      "loss": 5.6798,
      "mean_token_accuracy": 0.17473064959049225,
      "num_tokens": 44185824.0,
      "step": 22590
    },
    {
      "entropy": 6.054876661300659,
      "epoch": 2.258609486679662,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004492385330907127,
      "loss": 5.6071,
      "mean_token_accuracy": 0.18138361126184463,
      "num_tokens": 44195817.0,
      "step": 22595
    },
    {
      "entropy": 5.9934288501739506,
      "epoch": 2.2591093117408905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004492159502506622,
      "loss": 5.6132,
      "mean_token_accuracy": 0.18111189156770707,
      "num_tokens": 44205885.0,
      "step": 22600
    },
    {
      "entropy": 6.147416305541992,
      "epoch": 2.259609136802119,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044919336302738324,
      "loss": 5.6737,
      "mean_token_accuracy": 0.17354893833398818,
      "num_tokens": 44215472.0,
      "step": 22605
    },
    {
      "entropy": 6.136277961730957,
      "epoch": 2.260108961863348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044917077142144407,
      "loss": 5.7015,
      "mean_token_accuracy": 0.17671042382717134,
      "num_tokens": 44225833.0,
      "step": 22610
    },
    {
      "entropy": 6.022605228424072,
      "epoch": 2.2606087869245766,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044914817543341315,
      "loss": 5.6009,
      "mean_token_accuracy": 0.18302031755447387,
      "num_tokens": 44236252.0,
      "step": 22615
    },
    {
      "entropy": 6.046173477172852,
      "epoch": 2.261108611985805,
      "grad_norm": 1.3125,
      "learning_rate": 0.0004491255750638589,
      "loss": 5.699,
      "mean_token_accuracy": 0.17131995856761933,
      "num_tokens": 44245387.0,
      "step": 22620
    },
    {
      "entropy": 6.143176078796387,
      "epoch": 2.2616084370470335,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044910297031335006,
      "loss": 5.7473,
      "mean_token_accuracy": 0.17100108414888382,
      "num_tokens": 44254578.0,
      "step": 22625
    },
    {
      "entropy": 6.032327365875244,
      "epoch": 2.262108262108262,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044908036118245523,
      "loss": 5.5612,
      "mean_token_accuracy": 0.18317103385925293,
      "num_tokens": 44263042.0,
      "step": 22630
    },
    {
      "entropy": 6.086282777786255,
      "epoch": 2.2626080871694905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004490577476717433,
      "loss": 5.6839,
      "mean_token_accuracy": 0.1689959242939949,
      "num_tokens": 44272068.0,
      "step": 22635
    },
    {
      "entropy": 5.9729146480560305,
      "epoch": 2.263107912230719,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004490351297817832,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18363638669252397,
      "num_tokens": 44281829.0,
      "step": 22640
    },
    {
      "entropy": 6.028397512435913,
      "epoch": 2.263607737291948,
      "grad_norm": 1.2734375,
      "learning_rate": 0.000449012507513144,
      "loss": 5.6401,
      "mean_token_accuracy": 0.17523443400859834,
      "num_tokens": 44290097.0,
      "step": 22645
    },
    {
      "entropy": 6.048000431060791,
      "epoch": 2.2641075623531766,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044898988086639497,
      "loss": 5.6505,
      "mean_token_accuracy": 0.17964818477630615,
      "num_tokens": 44300611.0,
      "step": 22650
    },
    {
      "entropy": 6.047449588775635,
      "epoch": 2.264607387414405,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004489672498421053,
      "loss": 5.6454,
      "mean_token_accuracy": 0.1749682679772377,
      "num_tokens": 44310324.0,
      "step": 22655
    },
    {
      "entropy": 5.909977197647095,
      "epoch": 2.2651072124756335,
      "grad_norm": 1.25,
      "learning_rate": 0.0004489446144408444,
      "loss": 5.4826,
      "mean_token_accuracy": 0.1874544396996498,
      "num_tokens": 44319750.0,
      "step": 22660
    },
    {
      "entropy": 6.0537059783935545,
      "epoch": 2.265607037536862,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004489219746631817,
      "loss": 5.6781,
      "mean_token_accuracy": 0.17433273494243623,
      "num_tokens": 44329092.0,
      "step": 22665
    },
    {
      "entropy": 6.021524953842163,
      "epoch": 2.2661068625980905,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044889933050968695,
      "loss": 5.5776,
      "mean_token_accuracy": 0.18377045691013336,
      "num_tokens": 44337608.0,
      "step": 22670
    },
    {
      "entropy": 6.076061153411866,
      "epoch": 2.266606687659319,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044887668198092975,
      "loss": 5.708,
      "mean_token_accuracy": 0.1752407044172287,
      "num_tokens": 44347600.0,
      "step": 22675
    },
    {
      "entropy": 6.008341360092163,
      "epoch": 2.267106512720548,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044885402907748,
      "loss": 5.6629,
      "mean_token_accuracy": 0.17800277024507521,
      "num_tokens": 44358187.0,
      "step": 22680
    },
    {
      "entropy": 6.099863767623901,
      "epoch": 2.2676063377817766,
      "grad_norm": 1.6171875,
      "learning_rate": 0.00044883137179990755,
      "loss": 5.6559,
      "mean_token_accuracy": 0.17946952134370803,
      "num_tokens": 44367563.0,
      "step": 22685
    },
    {
      "entropy": 6.127856492996216,
      "epoch": 2.268106162843005,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044880871014878253,
      "loss": 5.69,
      "mean_token_accuracy": 0.1662495568394661,
      "num_tokens": 44378123.0,
      "step": 22690
    },
    {
      "entropy": 6.054735565185547,
      "epoch": 2.2686059879042335,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044878604412467517,
      "loss": 5.6918,
      "mean_token_accuracy": 0.16861882358789443,
      "num_tokens": 44387753.0,
      "step": 22695
    },
    {
      "entropy": 6.02043924331665,
      "epoch": 2.269105812965462,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044876337372815555,
      "loss": 5.5982,
      "mean_token_accuracy": 0.17840524166822433,
      "num_tokens": 44397917.0,
      "step": 22700
    },
    {
      "entropy": 6.030861186981201,
      "epoch": 2.2696056380266905,
      "grad_norm": 1.25,
      "learning_rate": 0.00044874069895979417,
      "loss": 5.6628,
      "mean_token_accuracy": 0.18183663636445999,
      "num_tokens": 44407135.0,
      "step": 22705
    },
    {
      "entropy": 6.054340600967407,
      "epoch": 2.270105463087919,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004487180198201616,
      "loss": 5.7059,
      "mean_token_accuracy": 0.17559201121330262,
      "num_tokens": 44416814.0,
      "step": 22710
    },
    {
      "entropy": 6.046470832824707,
      "epoch": 2.270605288149148,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044869533630982833,
      "loss": 5.8059,
      "mean_token_accuracy": 0.16059192419052123,
      "num_tokens": 44427304.0,
      "step": 22715
    },
    {
      "entropy": 6.056251001358032,
      "epoch": 2.2711051132103766,
      "grad_norm": 1.5234375,
      "learning_rate": 0.000448672648429365,
      "loss": 5.6019,
      "mean_token_accuracy": 0.17755310088396073,
      "num_tokens": 44436813.0,
      "step": 22720
    },
    {
      "entropy": 5.99092469215393,
      "epoch": 2.271604938271605,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044864995617934265,
      "loss": 5.57,
      "mean_token_accuracy": 0.1836025357246399,
      "num_tokens": 44446603.0,
      "step": 22725
    },
    {
      "entropy": 6.0076549530029295,
      "epoch": 2.2721047633328335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000448627259560332,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17345665842294694,
      "num_tokens": 44456016.0,
      "step": 22730
    },
    {
      "entropy": 6.073874568939209,
      "epoch": 2.272604588394062,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044860455857290415,
      "loss": 5.6428,
      "mean_token_accuracy": 0.18114509731531142,
      "num_tokens": 44465384.0,
      "step": 22735
    },
    {
      "entropy": 6.0320093631744385,
      "epoch": 2.2731044134552905,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004485818532176304,
      "loss": 5.6798,
      "mean_token_accuracy": 0.16968780010938644,
      "num_tokens": 44475709.0,
      "step": 22740
    },
    {
      "entropy": 6.057575750350952,
      "epoch": 2.273604238516519,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004485591434950818,
      "loss": 5.7198,
      "mean_token_accuracy": 0.1708892598748207,
      "num_tokens": 44485602.0,
      "step": 22745
    },
    {
      "entropy": 6.038588523864746,
      "epoch": 2.274104063577748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004485364294058298,
      "loss": 5.6782,
      "mean_token_accuracy": 0.17668967992067336,
      "num_tokens": 44495121.0,
      "step": 22750
    },
    {
      "entropy": 6.043997526168823,
      "epoch": 2.2746038886389766,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044851371095044597,
      "loss": 5.5774,
      "mean_token_accuracy": 0.1805984854698181,
      "num_tokens": 44504336.0,
      "step": 22755
    },
    {
      "entropy": 6.116004180908203,
      "epoch": 2.275103713700205,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044849098812950174,
      "loss": 5.6975,
      "mean_token_accuracy": 0.17600071728229522,
      "num_tokens": 44514076.0,
      "step": 22760
    },
    {
      "entropy": 6.052427625656128,
      "epoch": 2.2756035387614335,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000448468260943569,
      "loss": 5.7678,
      "mean_token_accuracy": 0.16799335330724716,
      "num_tokens": 44523659.0,
      "step": 22765
    },
    {
      "entropy": 6.0135297775268555,
      "epoch": 2.276103363822662,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044844552939321937,
      "loss": 5.5985,
      "mean_token_accuracy": 0.18164594620466232,
      "num_tokens": 44534024.0,
      "step": 22770
    },
    {
      "entropy": 5.989290761947632,
      "epoch": 2.2766031888838905,
      "grad_norm": 1.125,
      "learning_rate": 0.0004484227934790248,
      "loss": 5.5706,
      "mean_token_accuracy": 0.18282771110534668,
      "num_tokens": 44543206.0,
      "step": 22775
    },
    {
      "entropy": 6.107430124282837,
      "epoch": 2.277103013945119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004484000532015575,
      "loss": 5.771,
      "mean_token_accuracy": 0.17538654059171677,
      "num_tokens": 44553670.0,
      "step": 22780
    },
    {
      "entropy": 5.995081090927124,
      "epoch": 2.277602839006348,
      "grad_norm": 1.125,
      "learning_rate": 0.00044837730856138935,
      "loss": 5.7173,
      "mean_token_accuracy": 0.16795311421155928,
      "num_tokens": 44564040.0,
      "step": 22785
    },
    {
      "entropy": 5.994439935684204,
      "epoch": 2.2781026640675766,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004483545595590928,
      "loss": 5.5701,
      "mean_token_accuracy": 0.1865209311246872,
      "num_tokens": 44573605.0,
      "step": 22790
    },
    {
      "entropy": 6.088307523727417,
      "epoch": 2.278602489128805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004483318061952401,
      "loss": 5.6206,
      "mean_token_accuracy": 0.17592047899961472,
      "num_tokens": 44584298.0,
      "step": 22795
    },
    {
      "entropy": 6.100736522674561,
      "epoch": 2.2791023141900335,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004483090484704038,
      "loss": 5.7448,
      "mean_token_accuracy": 0.16686265021562577,
      "num_tokens": 44594585.0,
      "step": 22800
    },
    {
      "entropy": 5.9258991241455075,
      "epoch": 2.279602139251262,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044828628638515647,
      "loss": 5.5247,
      "mean_token_accuracy": 0.18610540479421617,
      "num_tokens": 44604788.0,
      "step": 22805
    },
    {
      "entropy": 6.024156713485718,
      "epoch": 2.2801019643124905,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004482635199400707,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1761302724480629,
      "num_tokens": 44614259.0,
      "step": 22810
    },
    {
      "entropy": 6.011697101593017,
      "epoch": 2.280601789373719,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044824074913571936,
      "loss": 5.6142,
      "mean_token_accuracy": 0.17819089740514754,
      "num_tokens": 44624827.0,
      "step": 22815
    },
    {
      "entropy": 5.99537034034729,
      "epoch": 2.281101614434948,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044821797397267535,
      "loss": 5.5825,
      "mean_token_accuracy": 0.17829849272966386,
      "num_tokens": 44635741.0,
      "step": 22820
    },
    {
      "entropy": 6.11402735710144,
      "epoch": 2.2816014394961766,
      "grad_norm": 1.25,
      "learning_rate": 0.0004481951944515117,
      "loss": 5.7576,
      "mean_token_accuracy": 0.17041065394878388,
      "num_tokens": 44644935.0,
      "step": 22825
    },
    {
      "entropy": 5.931255054473877,
      "epoch": 2.282101264557405,
      "grad_norm": 1.6640625,
      "learning_rate": 0.00044817241057280155,
      "loss": 5.5465,
      "mean_token_accuracy": 0.18928992599248887,
      "num_tokens": 44654265.0,
      "step": 22830
    },
    {
      "entropy": 6.121523857116699,
      "epoch": 2.2826010896186335,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004481496223371181,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17326626628637315,
      "num_tokens": 44664461.0,
      "step": 22835
    },
    {
      "entropy": 6.0548779487609865,
      "epoch": 2.283100914679862,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044812682974503464,
      "loss": 5.5838,
      "mean_token_accuracy": 0.17743927389383315,
      "num_tokens": 44673774.0,
      "step": 22840
    },
    {
      "entropy": 6.0594401359558105,
      "epoch": 2.2836007397410905,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004481040327971248,
      "loss": 5.6763,
      "mean_token_accuracy": 0.1744630068540573,
      "num_tokens": 44683061.0,
      "step": 22845
    },
    {
      "entropy": 6.073571443557739,
      "epoch": 2.284100564802319,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044808123149396193,
      "loss": 5.6601,
      "mean_token_accuracy": 0.1713159441947937,
      "num_tokens": 44691874.0,
      "step": 22850
    },
    {
      "entropy": 6.0932690620422365,
      "epoch": 2.284600389863548,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044805842583611993,
      "loss": 5.7182,
      "mean_token_accuracy": 0.17390562891960143,
      "num_tokens": 44701796.0,
      "step": 22855
    },
    {
      "entropy": 6.129049491882324,
      "epoch": 2.2851002149247766,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044803561582417237,
      "loss": 5.7745,
      "mean_token_accuracy": 0.16220706105232238,
      "num_tokens": 44711749.0,
      "step": 22860
    },
    {
      "entropy": 6.03991117477417,
      "epoch": 2.285600039986005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044801280145869327,
      "loss": 5.5761,
      "mean_token_accuracy": 0.1820228949189186,
      "num_tokens": 44721507.0,
      "step": 22865
    },
    {
      "entropy": 6.081853246688842,
      "epoch": 2.2860998650472335,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004479899827402566,
      "loss": 5.7347,
      "mean_token_accuracy": 0.17001610845327378,
      "num_tokens": 44731178.0,
      "step": 22870
    },
    {
      "entropy": 6.036158466339112,
      "epoch": 2.286599690108462,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044796715966943654,
      "loss": 5.6276,
      "mean_token_accuracy": 0.17586903870105744,
      "num_tokens": 44741385.0,
      "step": 22875
    },
    {
      "entropy": 6.109137678146363,
      "epoch": 2.2870995151696905,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004479443322468072,
      "loss": 5.6844,
      "mean_token_accuracy": 0.17533629089593888,
      "num_tokens": 44751316.0,
      "step": 22880
    },
    {
      "entropy": 6.0589264869689945,
      "epoch": 2.287599340230919,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004479215004729429,
      "loss": 5.6745,
      "mean_token_accuracy": 0.1721330389380455,
      "num_tokens": 44760786.0,
      "step": 22885
    },
    {
      "entropy": 6.051329469680786,
      "epoch": 2.288099165292148,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004478986643484183,
      "loss": 5.6818,
      "mean_token_accuracy": 0.17198379039764405,
      "num_tokens": 44771203.0,
      "step": 22890
    },
    {
      "entropy": 6.110095500946045,
      "epoch": 2.2885989903533765,
      "grad_norm": 1.234375,
      "learning_rate": 0.00044787582387380764,
      "loss": 5.7243,
      "mean_token_accuracy": 0.16875131130218507,
      "num_tokens": 44780739.0,
      "step": 22895
    },
    {
      "entropy": 6.127801179885864,
      "epoch": 2.289098815414605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004478529790496858,
      "loss": 5.685,
      "mean_token_accuracy": 0.17555733472108842,
      "num_tokens": 44790534.0,
      "step": 22900
    },
    {
      "entropy": 6.0240319728851315,
      "epoch": 2.2895986404758335,
      "grad_norm": 1.4375,
      "learning_rate": 0.00044783012987662755,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17706696093082427,
      "num_tokens": 44799767.0,
      "step": 22905
    },
    {
      "entropy": 6.034433603286743,
      "epoch": 2.290098465537062,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004478072763552076,
      "loss": 5.7262,
      "mean_token_accuracy": 0.16661795675754548,
      "num_tokens": 44810431.0,
      "step": 22910
    },
    {
      "entropy": 6.021226930618286,
      "epoch": 2.2905982905982905,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004477844184860011,
      "loss": 5.6032,
      "mean_token_accuracy": 0.17970228791236878,
      "num_tokens": 44820605.0,
      "step": 22915
    },
    {
      "entropy": 6.147016906738282,
      "epoch": 2.291098115659519,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044776155626958305,
      "loss": 5.7142,
      "mean_token_accuracy": 0.1735147327184677,
      "num_tokens": 44829895.0,
      "step": 22920
    },
    {
      "entropy": 6.0952064990997314,
      "epoch": 2.291597940720748,
      "grad_norm": 1.0625,
      "learning_rate": 0.00044773868970652875,
      "loss": 5.7044,
      "mean_token_accuracy": 0.1727651759982109,
      "num_tokens": 44839162.0,
      "step": 22925
    },
    {
      "entropy": 6.04616060256958,
      "epoch": 2.2920977657819765,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044771581879741334,
      "loss": 5.4982,
      "mean_token_accuracy": 0.1821899652481079,
      "num_tokens": 44849395.0,
      "step": 22930
    },
    {
      "entropy": 6.056617736816406,
      "epoch": 2.292597590843205,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044769294354281253,
      "loss": 5.626,
      "mean_token_accuracy": 0.17853233963251114,
      "num_tokens": 44858695.0,
      "step": 22935
    },
    {
      "entropy": 6.082744932174682,
      "epoch": 2.2930974159044335,
      "grad_norm": 1.3359375,
      "learning_rate": 0.00044767006394330166,
      "loss": 5.7424,
      "mean_token_accuracy": 0.17392259687185288,
      "num_tokens": 44869447.0,
      "step": 22940
    },
    {
      "entropy": 5.926930236816406,
      "epoch": 2.293597240965662,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044764717999945644,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19093242585659026,
      "num_tokens": 44878155.0,
      "step": 22945
    },
    {
      "entropy": 6.031009197235107,
      "epoch": 2.2940970660268905,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044762429171185246,
      "loss": 5.6854,
      "mean_token_accuracy": 0.1685200423002243,
      "num_tokens": 44888465.0,
      "step": 22950
    },
    {
      "entropy": 6.042946767807007,
      "epoch": 2.294596891088119,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044760139908106583,
      "loss": 5.6342,
      "mean_token_accuracy": 0.17464472949504853,
      "num_tokens": 44898915.0,
      "step": 22955
    },
    {
      "entropy": 6.077787780761719,
      "epoch": 2.295096716149348,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004475785021076724,
      "loss": 5.5975,
      "mean_token_accuracy": 0.17435404285788536,
      "num_tokens": 44908543.0,
      "step": 22960
    },
    {
      "entropy": 6.005533456802368,
      "epoch": 2.2955965412105765,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044755560079224834,
      "loss": 5.6212,
      "mean_token_accuracy": 0.1799132212996483,
      "num_tokens": 44918429.0,
      "step": 22965
    },
    {
      "entropy": 6.042086029052735,
      "epoch": 2.296096366271805,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004475326951353697,
      "loss": 5.6031,
      "mean_token_accuracy": 0.17452608048915863,
      "num_tokens": 44927479.0,
      "step": 22970
    },
    {
      "entropy": 5.946207904815674,
      "epoch": 2.2965961913330335,
      "grad_norm": 1.25,
      "learning_rate": 0.00044750978513761276,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18469358682632447,
      "num_tokens": 44937317.0,
      "step": 22975
    },
    {
      "entropy": 6.104387998580933,
      "epoch": 2.297096016394262,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004474868707995541,
      "loss": 5.7639,
      "mean_token_accuracy": 0.1725318104028702,
      "num_tokens": 44947587.0,
      "step": 22980
    },
    {
      "entropy": 6.0311352729797365,
      "epoch": 2.2975958414554905,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044746395212177015,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1869001254439354,
      "num_tokens": 44956850.0,
      "step": 22985
    },
    {
      "entropy": 6.07925443649292,
      "epoch": 2.298095666516719,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044744102910483755,
      "loss": 5.6545,
      "mean_token_accuracy": 0.17090903967618942,
      "num_tokens": 44966310.0,
      "step": 22990
    },
    {
      "entropy": 6.066224527359009,
      "epoch": 2.298595491577948,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000447418101749333,
      "loss": 5.6935,
      "mean_token_accuracy": 0.17769431918859482,
      "num_tokens": 44975361.0,
      "step": 22995
    },
    {
      "entropy": 6.074802255630493,
      "epoch": 2.299095316639176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004473951700558334,
      "loss": 5.7141,
      "mean_token_accuracy": 0.16872006505727768,
      "num_tokens": 44985420.0,
      "step": 23000
    },
    {
      "entropy": 6.037393236160279,
      "epoch": 2.299595141700405,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044737223402491574,
      "loss": 5.6456,
      "mean_token_accuracy": 0.1764036923646927,
      "num_tokens": 44995622.0,
      "step": 23005
    },
    {
      "entropy": 6.055285406112671,
      "epoch": 2.3000949667616335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044734929365715685,
      "loss": 5.6896,
      "mean_token_accuracy": 0.18102693259716035,
      "num_tokens": 45006017.0,
      "step": 23010
    },
    {
      "entropy": 6.092122554779053,
      "epoch": 2.3005947918228618,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044732634895313424,
      "loss": 5.6673,
      "mean_token_accuracy": 0.172272028028965,
      "num_tokens": 45017024.0,
      "step": 23015
    },
    {
      "entropy": 6.077941942214966,
      "epoch": 2.3010946168840904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000447303399913425,
      "loss": 5.6595,
      "mean_token_accuracy": 0.1731354206800461,
      "num_tokens": 45026384.0,
      "step": 23020
    },
    {
      "entropy": 5.968791818618774,
      "epoch": 2.301594441945319,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044728044653860653,
      "loss": 5.5623,
      "mean_token_accuracy": 0.19222511500120162,
      "num_tokens": 45037144.0,
      "step": 23025
    },
    {
      "entropy": 5.995727682113648,
      "epoch": 2.302094267006548,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004472574888292564,
      "loss": 5.6928,
      "mean_token_accuracy": 0.171397802233696,
      "num_tokens": 45048272.0,
      "step": 23030
    },
    {
      "entropy": 6.069711589813233,
      "epoch": 2.302594092067776,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004472345267859521,
      "loss": 5.6627,
      "mean_token_accuracy": 0.17340897619724274,
      "num_tokens": 45057799.0,
      "step": 23035
    },
    {
      "entropy": 6.030992460250855,
      "epoch": 2.303093917129005,
      "grad_norm": 1.25,
      "learning_rate": 0.0004472115604092715,
      "loss": 5.6782,
      "mean_token_accuracy": 0.1706493079662323,
      "num_tokens": 45066786.0,
      "step": 23040
    },
    {
      "entropy": 6.05184383392334,
      "epoch": 2.3035937421902335,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044718858969979236,
      "loss": 5.6698,
      "mean_token_accuracy": 0.1775143101811409,
      "num_tokens": 45075425.0,
      "step": 23045
    },
    {
      "entropy": 6.152289581298828,
      "epoch": 2.3040935672514617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00044716561465809256,
      "loss": 5.7653,
      "mean_token_accuracy": 0.17674290537834167,
      "num_tokens": 45085470.0,
      "step": 23050
    },
    {
      "entropy": 6.186959457397461,
      "epoch": 2.3045933923126904,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044714263528475024,
      "loss": 5.9183,
      "mean_token_accuracy": 0.16066047847270964,
      "num_tokens": 45096093.0,
      "step": 23055
    },
    {
      "entropy": 6.104692220687866,
      "epoch": 2.305093217373919,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044711965158034345,
      "loss": 5.7125,
      "mean_token_accuracy": 0.16941499561071396,
      "num_tokens": 45106031.0,
      "step": 23060
    },
    {
      "entropy": 6.093301486968994,
      "epoch": 2.305593042435148,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004470966635454506,
      "loss": 5.6382,
      "mean_token_accuracy": 0.18020707964897156,
      "num_tokens": 45116211.0,
      "step": 23065
    },
    {
      "entropy": 6.026928949356079,
      "epoch": 2.306092867496376,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004470736711806499,
      "loss": 5.6904,
      "mean_token_accuracy": 0.1716480165719986,
      "num_tokens": 45126232.0,
      "step": 23070
    },
    {
      "entropy": 6.075389432907104,
      "epoch": 2.306592692557605,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044705067448651996,
      "loss": 5.5846,
      "mean_token_accuracy": 0.18015961050987245,
      "num_tokens": 45136388.0,
      "step": 23075
    },
    {
      "entropy": 6.026543712615966,
      "epoch": 2.3070925176188335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004470276734636393,
      "loss": 5.6703,
      "mean_token_accuracy": 0.1806821808218956,
      "num_tokens": 45145569.0,
      "step": 23080
    },
    {
      "entropy": 5.934084463119507,
      "epoch": 2.3075923426800617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044700466811258666,
      "loss": 5.6094,
      "mean_token_accuracy": 0.18117471784353256,
      "num_tokens": 45155172.0,
      "step": 23085
    },
    {
      "entropy": 6.014734601974487,
      "epoch": 2.3080921677412904,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004469816584339408,
      "loss": 5.5554,
      "mean_token_accuracy": 0.1887346997857094,
      "num_tokens": 45164537.0,
      "step": 23090
    },
    {
      "entropy": 6.095815229415893,
      "epoch": 2.308591992802519,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044695864442828066,
      "loss": 5.7353,
      "mean_token_accuracy": 0.1704954594373703,
      "num_tokens": 45174031.0,
      "step": 23095
    },
    {
      "entropy": 5.985543203353882,
      "epoch": 2.309091817863748,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044693562609618534,
      "loss": 5.6071,
      "mean_token_accuracy": 0.17279779314994811,
      "num_tokens": 45183211.0,
      "step": 23100
    },
    {
      "entropy": 6.0615095615386965,
      "epoch": 2.309591642924976,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004469126034382339,
      "loss": 5.7108,
      "mean_token_accuracy": 0.170160610973835,
      "num_tokens": 45193619.0,
      "step": 23105
    },
    {
      "entropy": 6.056368446350097,
      "epoch": 2.310091467986205,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004468895764550055,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18198743015527724,
      "num_tokens": 45203541.0,
      "step": 23110
    },
    {
      "entropy": 6.058747911453247,
      "epoch": 2.3105912930474335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004468665451470797,
      "loss": 5.7159,
      "mean_token_accuracy": 0.1720944419503212,
      "num_tokens": 45213231.0,
      "step": 23115
    },
    {
      "entropy": 6.123208856582641,
      "epoch": 2.3110911181086617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004468435095150358,
      "loss": 5.7044,
      "mean_token_accuracy": 0.1635023534297943,
      "num_tokens": 45222585.0,
      "step": 23120
    },
    {
      "entropy": 6.071886301040649,
      "epoch": 2.3115909431698904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044682046955945345,
      "loss": 5.7072,
      "mean_token_accuracy": 0.1736617550253868,
      "num_tokens": 45233068.0,
      "step": 23125
    },
    {
      "entropy": 6.020181560516358,
      "epoch": 2.312090768231119,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004467974252809123,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1786353513598442,
      "num_tokens": 45242643.0,
      "step": 23130
    },
    {
      "entropy": 6.0808085918426515,
      "epoch": 2.312590593292348,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044677437667999204,
      "loss": 5.6348,
      "mean_token_accuracy": 0.1746222198009491,
      "num_tokens": 45251870.0,
      "step": 23135
    },
    {
      "entropy": 6.0341767311096195,
      "epoch": 2.313090418353576,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044675132375727273,
      "loss": 5.6276,
      "mean_token_accuracy": 0.17861589342355727,
      "num_tokens": 45261209.0,
      "step": 23140
    },
    {
      "entropy": 6.052816724777221,
      "epoch": 2.313590243414805,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004467282665133343,
      "loss": 5.6929,
      "mean_token_accuracy": 0.17612568736076356,
      "num_tokens": 45270339.0,
      "step": 23145
    },
    {
      "entropy": 6.131492567062378,
      "epoch": 2.3140900684760335,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004467052049487569,
      "loss": 5.7529,
      "mean_token_accuracy": 0.1654794007539749,
      "num_tokens": 45281107.0,
      "step": 23150
    },
    {
      "entropy": 6.074229717254639,
      "epoch": 2.3145898935372617,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044668213906412074,
      "loss": 5.5786,
      "mean_token_accuracy": 0.17920369952917098,
      "num_tokens": 45289615.0,
      "step": 23155
    },
    {
      "entropy": 5.990182018280029,
      "epoch": 2.3150897185984904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044665906886000604,
      "loss": 5.581,
      "mean_token_accuracy": 0.18555006384849548,
      "num_tokens": 45299768.0,
      "step": 23160
    },
    {
      "entropy": 5.974020051956177,
      "epoch": 2.315589543659719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044663599433699346,
      "loss": 5.5439,
      "mean_token_accuracy": 0.18630737364292144,
      "num_tokens": 45310407.0,
      "step": 23165
    },
    {
      "entropy": 6.058645486831665,
      "epoch": 2.316089368720948,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044661291549566337,
      "loss": 5.6139,
      "mean_token_accuracy": 0.1766300991177559,
      "num_tokens": 45319993.0,
      "step": 23170
    },
    {
      "entropy": 6.014542484283448,
      "epoch": 2.316589193782176,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004465898323365965,
      "loss": 5.6885,
      "mean_token_accuracy": 0.1731710433959961,
      "num_tokens": 45329598.0,
      "step": 23175
    },
    {
      "entropy": 6.0917784690856935,
      "epoch": 2.317089018843405,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004465667448603736,
      "loss": 5.6617,
      "mean_token_accuracy": 0.17391647398471832,
      "num_tokens": 45339630.0,
      "step": 23180
    },
    {
      "entropy": 6.013296794891358,
      "epoch": 2.3175888439046335,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044654365306757553,
      "loss": 5.5955,
      "mean_token_accuracy": 0.175054694712162,
      "num_tokens": 45349441.0,
      "step": 23185
    },
    {
      "entropy": 6.096804523468018,
      "epoch": 2.3180886689658617,
      "grad_norm": 1.125,
      "learning_rate": 0.00044652055695878335,
      "loss": 5.7745,
      "mean_token_accuracy": 0.1690856099128723,
      "num_tokens": 45359858.0,
      "step": 23190
    },
    {
      "entropy": 6.104031658172607,
      "epoch": 2.3185884940270904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044649745653457806,
      "loss": 5.6602,
      "mean_token_accuracy": 0.17211828529834747,
      "num_tokens": 45369100.0,
      "step": 23195
    },
    {
      "entropy": 6.068248701095581,
      "epoch": 2.319088319088319,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044647435179554087,
      "loss": 5.6326,
      "mean_token_accuracy": 0.17745068818330764,
      "num_tokens": 45380514.0,
      "step": 23200
    },
    {
      "entropy": 6.068368721008301,
      "epoch": 2.319588144149548,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004464512427422531,
      "loss": 5.6981,
      "mean_token_accuracy": 0.16932753324508668,
      "num_tokens": 45390689.0,
      "step": 23205
    },
    {
      "entropy": 6.096634387969971,
      "epoch": 2.320087969210776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044642812937529624,
      "loss": 5.673,
      "mean_token_accuracy": 0.16878530085086824,
      "num_tokens": 45400436.0,
      "step": 23210
    },
    {
      "entropy": 6.123605108261108,
      "epoch": 2.320587794272005,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004464050116952517,
      "loss": 5.7407,
      "mean_token_accuracy": 0.17254570126533508,
      "num_tokens": 45409906.0,
      "step": 23215
    },
    {
      "entropy": 6.017495632171631,
      "epoch": 2.3210876193332335,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044638188970270114,
      "loss": 5.684,
      "mean_token_accuracy": 0.17094576358795166,
      "num_tokens": 45419869.0,
      "step": 23220
    },
    {
      "entropy": 6.0266517162322994,
      "epoch": 2.3215874443944617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044635876339822647,
      "loss": 5.6002,
      "mean_token_accuracy": 0.18555708527565,
      "num_tokens": 45430630.0,
      "step": 23225
    },
    {
      "entropy": 6.059256553649902,
      "epoch": 2.3220872694556904,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044633563278240933,
      "loss": 5.5225,
      "mean_token_accuracy": 0.18503057062625886,
      "num_tokens": 45439765.0,
      "step": 23230
    },
    {
      "entropy": 6.136633014678955,
      "epoch": 2.322587094516919,
      "grad_norm": 1.296875,
      "learning_rate": 0.00044631249785583173,
      "loss": 5.6563,
      "mean_token_accuracy": 0.17462363988161086,
      "num_tokens": 45448619.0,
      "step": 23235
    },
    {
      "entropy": 6.079303026199341,
      "epoch": 2.323086919578148,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004462893586190758,
      "loss": 5.654,
      "mean_token_accuracy": 0.1770624950528145,
      "num_tokens": 45456874.0,
      "step": 23240
    },
    {
      "entropy": 6.0508819103240965,
      "epoch": 2.323586744639376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044626621507272364,
      "loss": 5.7436,
      "mean_token_accuracy": 0.17352414429187774,
      "num_tokens": 45465971.0,
      "step": 23245
    },
    {
      "entropy": 6.079803848266602,
      "epoch": 2.324086569700605,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044624306721735763,
      "loss": 5.6945,
      "mean_token_accuracy": 0.16615247577428818,
      "num_tokens": 45475802.0,
      "step": 23250
    },
    {
      "entropy": 6.091593933105469,
      "epoch": 2.3245863947618335,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044621991505356014,
      "loss": 5.6665,
      "mean_token_accuracy": 0.18154877573251724,
      "num_tokens": 45486270.0,
      "step": 23255
    },
    {
      "entropy": 6.0929179191589355,
      "epoch": 2.3250862198230617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044619675858191353,
      "loss": 5.7012,
      "mean_token_accuracy": 0.1756749764084816,
      "num_tokens": 45495936.0,
      "step": 23260
    },
    {
      "entropy": 5.966797542572022,
      "epoch": 2.3255860448842904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004461735978030006,
      "loss": 5.4976,
      "mean_token_accuracy": 0.19037146121263504,
      "num_tokens": 45505719.0,
      "step": 23265
    },
    {
      "entropy": 6.058510208129883,
      "epoch": 2.326085869945519,
      "grad_norm": 1.265625,
      "learning_rate": 0.00044615043271740404,
      "loss": 5.7013,
      "mean_token_accuracy": 0.17487429082393646,
      "num_tokens": 45515322.0,
      "step": 23270
    },
    {
      "entropy": 6.03613543510437,
      "epoch": 2.326585695006748,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004461272633257065,
      "loss": 5.5906,
      "mean_token_accuracy": 0.18316141963005067,
      "num_tokens": 45524763.0,
      "step": 23275
    },
    {
      "entropy": 6.043449211120605,
      "epoch": 2.327085520067976,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004461040896284912,
      "loss": 5.6245,
      "mean_token_accuracy": 0.17957112193107605,
      "num_tokens": 45533140.0,
      "step": 23280
    },
    {
      "entropy": 6.0959900379180905,
      "epoch": 2.327585345129205,
      "grad_norm": 2.140625,
      "learning_rate": 0.000446080911626341,
      "loss": 5.6452,
      "mean_token_accuracy": 0.17920673862099648,
      "num_tokens": 45543162.0,
      "step": 23285
    },
    {
      "entropy": 6.095486307144165,
      "epoch": 2.3280851701904335,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004460577293198391,
      "loss": 5.7887,
      "mean_token_accuracy": 0.16426983773708342,
      "num_tokens": 45552863.0,
      "step": 23290
    },
    {
      "entropy": 6.0184712409973145,
      "epoch": 2.3285849952516617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044603454270956863,
      "loss": 5.6654,
      "mean_token_accuracy": 0.1710359513759613,
      "num_tokens": 45562177.0,
      "step": 23295
    },
    {
      "entropy": 6.142314291000366,
      "epoch": 2.3290848203128904,
      "grad_norm": 1.546875,
      "learning_rate": 0.0004460113517961132,
      "loss": 5.7151,
      "mean_token_accuracy": 0.17252847254276277,
      "num_tokens": 45572864.0,
      "step": 23300
    },
    {
      "entropy": 6.1144633293151855,
      "epoch": 2.329584645374119,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004459881565800561,
      "loss": 5.723,
      "mean_token_accuracy": 0.1723456934094429,
      "num_tokens": 45582778.0,
      "step": 23305
    },
    {
      "entropy": 5.982622575759888,
      "epoch": 2.330084470435348,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004459649570619809,
      "loss": 5.5622,
      "mean_token_accuracy": 0.17738457322120665,
      "num_tokens": 45591659.0,
      "step": 23310
    },
    {
      "entropy": 6.036458206176758,
      "epoch": 2.330584295496576,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004459417532424715,
      "loss": 5.8281,
      "mean_token_accuracy": 0.16232905983924867,
      "num_tokens": 45602215.0,
      "step": 23315
    },
    {
      "entropy": 6.069948625564575,
      "epoch": 2.331084120557805,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004459185451221116,
      "loss": 5.64,
      "mean_token_accuracy": 0.17492918223142623,
      "num_tokens": 45611680.0,
      "step": 23320
    },
    {
      "entropy": 5.968486595153808,
      "epoch": 2.3315839456190335,
      "grad_norm": 1.3671875,
      "learning_rate": 0.000445895332701485,
      "loss": 5.5416,
      "mean_token_accuracy": 0.18527524173259735,
      "num_tokens": 45621452.0,
      "step": 23325
    },
    {
      "entropy": 6.098551321029663,
      "epoch": 2.3320837706802617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044587211598117577,
      "loss": 5.7645,
      "mean_token_accuracy": 0.17103444188833236,
      "num_tokens": 45631594.0,
      "step": 23330
    },
    {
      "entropy": 6.060859966278076,
      "epoch": 2.3325835957414904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044584889496176815,
      "loss": 5.6591,
      "mean_token_accuracy": 0.17607979327440262,
      "num_tokens": 45642373.0,
      "step": 23335
    },
    {
      "entropy": 6.099746227264404,
      "epoch": 2.333083420802719,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004458256696438463,
      "loss": 5.6266,
      "mean_token_accuracy": 0.17169416546821595,
      "num_tokens": 45652851.0,
      "step": 23340
    },
    {
      "entropy": 5.88638596534729,
      "epoch": 2.333583245863948,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044580244002799454,
      "loss": 5.4824,
      "mean_token_accuracy": 0.1838074341416359,
      "num_tokens": 45662638.0,
      "step": 23345
    },
    {
      "entropy": 5.933127212524414,
      "epoch": 2.334083070925176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004457792061147973,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17635180354118346,
      "num_tokens": 45672252.0,
      "step": 23350
    },
    {
      "entropy": 6.09796175956726,
      "epoch": 2.334582895986405,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004457559679048392,
      "loss": 5.6641,
      "mean_token_accuracy": 0.17317719757556915,
      "num_tokens": 45682678.0,
      "step": 23355
    },
    {
      "entropy": 6.093754577636719,
      "epoch": 2.3350827210476335,
      "grad_norm": 1.21875,
      "learning_rate": 0.000445732725398705,
      "loss": 5.648,
      "mean_token_accuracy": 0.17362246066331863,
      "num_tokens": 45692454.0,
      "step": 23360
    },
    {
      "entropy": 6.082568883895874,
      "epoch": 2.3355825461088617,
      "grad_norm": 1.484375,
      "learning_rate": 0.0004457094785969792,
      "loss": 5.6679,
      "mean_token_accuracy": 0.1755481705069542,
      "num_tokens": 45701925.0,
      "step": 23365
    },
    {
      "entropy": 6.051724052429199,
      "epoch": 2.3360823711700904,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044568622750024693,
      "loss": 5.6876,
      "mean_token_accuracy": 0.17561267465353012,
      "num_tokens": 45711217.0,
      "step": 23370
    },
    {
      "entropy": 5.994741582870484,
      "epoch": 2.336582196231319,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004456629721090931,
      "loss": 5.6171,
      "mean_token_accuracy": 0.182993283867836,
      "num_tokens": 45720651.0,
      "step": 23375
    },
    {
      "entropy": 6.0729694843292235,
      "epoch": 2.337082021292548,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004456397124241028,
      "loss": 5.7621,
      "mean_token_accuracy": 0.1713467448949814,
      "num_tokens": 45730649.0,
      "step": 23380
    },
    {
      "entropy": 6.091461181640625,
      "epoch": 2.337581846353776,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004456164484458612,
      "loss": 5.6136,
      "mean_token_accuracy": 0.18295783400535584,
      "num_tokens": 45739862.0,
      "step": 23385
    },
    {
      "entropy": 6.056644058227539,
      "epoch": 2.338081671415005,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004455931801749537,
      "loss": 5.7024,
      "mean_token_accuracy": 0.1738718032836914,
      "num_tokens": 45750710.0,
      "step": 23390
    },
    {
      "entropy": 6.162650728225708,
      "epoch": 2.3385814964762335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044556990761196566,
      "loss": 5.7586,
      "mean_token_accuracy": 0.16935701668262482,
      "num_tokens": 45759725.0,
      "step": 23395
    },
    {
      "entropy": 6.1244056224823,
      "epoch": 2.3390813215374617,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044554663075748266,
      "loss": 5.7645,
      "mean_token_accuracy": 0.1709905281662941,
      "num_tokens": 45769143.0,
      "step": 23400
    },
    {
      "entropy": 6.073747158050537,
      "epoch": 2.3395811465986904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004455233496120903,
      "loss": 5.7238,
      "mean_token_accuracy": 0.1674390822649002,
      "num_tokens": 45779505.0,
      "step": 23405
    },
    {
      "entropy": 6.1019505023956295,
      "epoch": 2.340080971659919,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004455000641763743,
      "loss": 5.7306,
      "mean_token_accuracy": 0.16764169186353683,
      "num_tokens": 45789337.0,
      "step": 23410
    },
    {
      "entropy": 6.039696836471558,
      "epoch": 2.340580796721148,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044547677445092064,
      "loss": 5.6958,
      "mean_token_accuracy": 0.17590183168649673,
      "num_tokens": 45799385.0,
      "step": 23415
    },
    {
      "entropy": 6.061013412475586,
      "epoch": 2.341080621782376,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044545348043631525,
      "loss": 5.6846,
      "mean_token_accuracy": 0.16533951908349992,
      "num_tokens": 45810918.0,
      "step": 23420
    },
    {
      "entropy": 6.11961555480957,
      "epoch": 2.341580446843605,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044543018213314404,
      "loss": 5.6964,
      "mean_token_accuracy": 0.17520616203546524,
      "num_tokens": 45820910.0,
      "step": 23425
    },
    {
      "entropy": 6.007542085647583,
      "epoch": 2.3420802719048335,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00044540687954199335,
      "loss": 5.5858,
      "mean_token_accuracy": 0.17503385245800018,
      "num_tokens": 45830709.0,
      "step": 23430
    },
    {
      "entropy": 6.029560852050781,
      "epoch": 2.3425800969660617,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044538357266344936,
      "loss": 5.6114,
      "mean_token_accuracy": 0.17776171267032623,
      "num_tokens": 45840244.0,
      "step": 23435
    },
    {
      "entropy": 6.03725118637085,
      "epoch": 2.3430799220272904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004453602614980986,
      "loss": 5.6817,
      "mean_token_accuracy": 0.17812149077653885,
      "num_tokens": 45849643.0,
      "step": 23440
    },
    {
      "entropy": 6.149209594726562,
      "epoch": 2.343579747088519,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004453369460465274,
      "loss": 5.7045,
      "mean_token_accuracy": 0.17016686499118805,
      "num_tokens": 45858689.0,
      "step": 23445
    },
    {
      "entropy": 6.131392526626587,
      "epoch": 2.344079572149748,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004453136263093225,
      "loss": 5.7222,
      "mean_token_accuracy": 0.17099529057741164,
      "num_tokens": 45869538.0,
      "step": 23450
    },
    {
      "entropy": 6.103089237213135,
      "epoch": 2.344579397210976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004452903022870707,
      "loss": 5.7337,
      "mean_token_accuracy": 0.17047263830900192,
      "num_tokens": 45878248.0,
      "step": 23455
    },
    {
      "entropy": 6.012277889251709,
      "epoch": 2.3450792222722048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004452669739803586,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17705504447221757,
      "num_tokens": 45888682.0,
      "step": 23460
    },
    {
      "entropy": 6.064145803451538,
      "epoch": 2.3455790473334335,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004452436413897733,
      "loss": 5.6775,
      "mean_token_accuracy": 0.1767082393169403,
      "num_tokens": 45898690.0,
      "step": 23465
    },
    {
      "entropy": 6.054799509048462,
      "epoch": 2.3460788723946617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004452203045159018,
      "loss": 5.6442,
      "mean_token_accuracy": 0.17071681171655656,
      "num_tokens": 45907579.0,
      "step": 23470
    },
    {
      "entropy": 5.966411495208741,
      "epoch": 2.3465786974558904,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004451969633593312,
      "loss": 5.5936,
      "mean_token_accuracy": 0.1785627767443657,
      "num_tokens": 45917065.0,
      "step": 23475
    },
    {
      "entropy": 6.0560143947601315,
      "epoch": 2.347078522517119,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044517361792064883,
      "loss": 5.7529,
      "mean_token_accuracy": 0.16877295225858688,
      "num_tokens": 45927427.0,
      "step": 23480
    },
    {
      "entropy": 6.080617952346802,
      "epoch": 2.347578347578348,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044515026820044195,
      "loss": 5.5475,
      "mean_token_accuracy": 0.1825200453400612,
      "num_tokens": 45937453.0,
      "step": 23485
    },
    {
      "entropy": 5.972042369842529,
      "epoch": 2.348078172639576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004451269141992982,
      "loss": 5.5936,
      "mean_token_accuracy": 0.18557309359312057,
      "num_tokens": 45947285.0,
      "step": 23490
    },
    {
      "entropy": 6.111046600341797,
      "epoch": 2.3485779977008048,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044510355591780503,
      "loss": 5.7466,
      "mean_token_accuracy": 0.1698734149336815,
      "num_tokens": 45956413.0,
      "step": 23495
    },
    {
      "entropy": 6.082316207885742,
      "epoch": 2.3490778227620335,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004450801933565502,
      "loss": 5.6844,
      "mean_token_accuracy": 0.1754311889410019,
      "num_tokens": 45967243.0,
      "step": 23500
    },
    {
      "entropy": 5.981764507293701,
      "epoch": 2.3495776478232617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004450568265161214,
      "loss": 5.6382,
      "mean_token_accuracy": 0.17779497802257538,
      "num_tokens": 45976837.0,
      "step": 23505
    },
    {
      "entropy": 6.030059957504273,
      "epoch": 2.3500774728844904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044503345539710664,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17372239232063294,
      "num_tokens": 45986498.0,
      "step": 23510
    },
    {
      "entropy": 6.10572829246521,
      "epoch": 2.350577297945719,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004450100800000939,
      "loss": 5.6632,
      "mean_token_accuracy": 0.18146397918462753,
      "num_tokens": 45995490.0,
      "step": 23515
    },
    {
      "entropy": 6.042029571533203,
      "epoch": 2.3510771230069474,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004449867003256713,
      "loss": 5.5999,
      "mean_token_accuracy": 0.18052243888378144,
      "num_tokens": 46004765.0,
      "step": 23520
    },
    {
      "entropy": 6.018957185745239,
      "epoch": 2.351576948068176,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044496331637442697,
      "loss": 5.675,
      "mean_token_accuracy": 0.1732357159256935,
      "num_tokens": 46014666.0,
      "step": 23525
    },
    {
      "entropy": 5.996883296966553,
      "epoch": 2.3520767731294048,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004449399281469495,
      "loss": 5.6008,
      "mean_token_accuracy": 0.17808209955692292,
      "num_tokens": 46024104.0,
      "step": 23530
    },
    {
      "entropy": 5.9809380054473875,
      "epoch": 2.3525765981906335,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044491653564382704,
      "loss": 5.5779,
      "mean_token_accuracy": 0.18037257641553878,
      "num_tokens": 46034709.0,
      "step": 23535
    },
    {
      "entropy": 6.026202058792114,
      "epoch": 2.3530764232518617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004448931388656483,
      "loss": 5.7001,
      "mean_token_accuracy": 0.172713603079319,
      "num_tokens": 46044404.0,
      "step": 23540
    },
    {
      "entropy": 6.021583938598633,
      "epoch": 2.3535762483130904,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044486973781300196,
      "loss": 5.6373,
      "mean_token_accuracy": 0.17765775918960572,
      "num_tokens": 46054223.0,
      "step": 23545
    },
    {
      "entropy": 6.06899471282959,
      "epoch": 2.354076073374319,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044484633248647665,
      "loss": 5.6425,
      "mean_token_accuracy": 0.1729290023446083,
      "num_tokens": 46063491.0,
      "step": 23550
    },
    {
      "entropy": 6.041161394119262,
      "epoch": 2.3545758984355474,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044482292288666134,
      "loss": 5.6534,
      "mean_token_accuracy": 0.17445360124111176,
      "num_tokens": 46072552.0,
      "step": 23555
    },
    {
      "entropy": 5.917170667648316,
      "epoch": 2.355075723496776,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000444799509014145,
      "loss": 5.5246,
      "mean_token_accuracy": 0.19003575593233107,
      "num_tokens": 46083249.0,
      "step": 23560
    },
    {
      "entropy": 6.080554723739624,
      "epoch": 2.3555755485580048,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004447760908695167,
      "loss": 5.7154,
      "mean_token_accuracy": 0.17288369536399842,
      "num_tokens": 46091842.0,
      "step": 23565
    },
    {
      "entropy": 5.939648151397705,
      "epoch": 2.356075373619233,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044475266845336566,
      "loss": 5.5561,
      "mean_token_accuracy": 0.18651992529630662,
      "num_tokens": 46101453.0,
      "step": 23570
    },
    {
      "entropy": 6.119129037857055,
      "epoch": 2.3565751986804617,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004447292417662811,
      "loss": 5.8015,
      "mean_token_accuracy": 0.1650396004319191,
      "num_tokens": 46110969.0,
      "step": 23575
    },
    {
      "entropy": 6.068396186828613,
      "epoch": 2.3570750237416904,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004447058108088525,
      "loss": 5.5311,
      "mean_token_accuracy": 0.19118430018424987,
      "num_tokens": 46120846.0,
      "step": 23580
    },
    {
      "entropy": 6.022601556777954,
      "epoch": 2.357574848802919,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044468237558166945,
      "loss": 5.5581,
      "mean_token_accuracy": 0.18963632434606553,
      "num_tokens": 46130039.0,
      "step": 23585
    },
    {
      "entropy": 5.987316131591797,
      "epoch": 2.3580746738641474,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044465893608532146,
      "loss": 5.5913,
      "mean_token_accuracy": 0.18409784883260727,
      "num_tokens": 46139503.0,
      "step": 23590
    },
    {
      "entropy": 6.0680365562438965,
      "epoch": 2.358574498925376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044463549232039824,
      "loss": 5.664,
      "mean_token_accuracy": 0.17304683476686478,
      "num_tokens": 46148247.0,
      "step": 23595
    },
    {
      "entropy": 6.058181953430176,
      "epoch": 2.3590743239866048,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004446120442874898,
      "loss": 5.6648,
      "mean_token_accuracy": 0.18391262292861937,
      "num_tokens": 46157233.0,
      "step": 23600
    },
    {
      "entropy": 6.013462877273559,
      "epoch": 2.359574149047833,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004445885919871859,
      "loss": 5.6363,
      "mean_token_accuracy": 0.17553962469100953,
      "num_tokens": 46166621.0,
      "step": 23605
    },
    {
      "entropy": 6.07929220199585,
      "epoch": 2.3600739741090617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004445651354200766,
      "loss": 5.6941,
      "mean_token_accuracy": 0.17370732128620148,
      "num_tokens": 46176142.0,
      "step": 23610
    },
    {
      "entropy": 6.006615829467774,
      "epoch": 2.3605737991702904,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044454167458675225,
      "loss": 5.5535,
      "mean_token_accuracy": 0.17931206226348878,
      "num_tokens": 46185506.0,
      "step": 23615
    },
    {
      "entropy": 6.030964517593384,
      "epoch": 2.361073624231519,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00044451820948780295,
      "loss": 5.6805,
      "mean_token_accuracy": 0.1785634234547615,
      "num_tokens": 46195351.0,
      "step": 23620
    },
    {
      "entropy": 6.03352370262146,
      "epoch": 2.3615734492927474,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044449474012381903,
      "loss": 5.6557,
      "mean_token_accuracy": 0.1691276401281357,
      "num_tokens": 46205431.0,
      "step": 23625
    },
    {
      "entropy": 6.092788600921631,
      "epoch": 2.362073274353976,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044447126649539123,
      "loss": 5.6283,
      "mean_token_accuracy": 0.17407059222459792,
      "num_tokens": 46214731.0,
      "step": 23630
    },
    {
      "entropy": 6.021127605438233,
      "epoch": 2.3625730994152048,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004444477886031098,
      "loss": 5.6286,
      "mean_token_accuracy": 0.18202770352363587,
      "num_tokens": 46223951.0,
      "step": 23635
    },
    {
      "entropy": 6.059401559829712,
      "epoch": 2.363072924476433,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004444243064475657,
      "loss": 5.7296,
      "mean_token_accuracy": 0.16963689774274826,
      "num_tokens": 46234458.0,
      "step": 23640
    },
    {
      "entropy": 6.078516101837158,
      "epoch": 2.3635727495376617,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004444008200293496,
      "loss": 5.7014,
      "mean_token_accuracy": 0.1743922770023346,
      "num_tokens": 46244549.0,
      "step": 23645
    },
    {
      "entropy": 6.060221099853516,
      "epoch": 2.3640725745988904,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044437732934905256,
      "loss": 5.6601,
      "mean_token_accuracy": 0.17803427428007126,
      "num_tokens": 46254360.0,
      "step": 23650
    },
    {
      "entropy": 6.0557066917419435,
      "epoch": 2.364572399660119,
      "grad_norm": 1.125,
      "learning_rate": 0.0004443538344072654,
      "loss": 5.6134,
      "mean_token_accuracy": 0.17843423932790756,
      "num_tokens": 46264153.0,
      "step": 23655
    },
    {
      "entropy": 6.080674695968628,
      "epoch": 2.3650722247213474,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044433033520457937,
      "loss": 5.5948,
      "mean_token_accuracy": 0.17616735696792601,
      "num_tokens": 46272797.0,
      "step": 23660
    },
    {
      "entropy": 6.049421262741089,
      "epoch": 2.365572049782576,
      "grad_norm": 1.578125,
      "learning_rate": 0.0004443068317415857,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17518768459558487,
      "num_tokens": 46282515.0,
      "step": 23665
    },
    {
      "entropy": 6.059255838394165,
      "epoch": 2.3660718748438048,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00044428332401887566,
      "loss": 5.7291,
      "mean_token_accuracy": 0.1694168046116829,
      "num_tokens": 46292682.0,
      "step": 23670
    },
    {
      "entropy": 5.997803592681885,
      "epoch": 2.366571699905033,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004442598120370408,
      "loss": 5.5806,
      "mean_token_accuracy": 0.18882497996091843,
      "num_tokens": 46302598.0,
      "step": 23675
    },
    {
      "entropy": 6.025753688812256,
      "epoch": 2.3670715249662617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004442362957966726,
      "loss": 5.6006,
      "mean_token_accuracy": 0.18503525257110595,
      "num_tokens": 46313015.0,
      "step": 23680
    },
    {
      "entropy": 5.988218116760254,
      "epoch": 2.3675713500274904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044421277529836276,
      "loss": 5.6545,
      "mean_token_accuracy": 0.17723398208618163,
      "num_tokens": 46322450.0,
      "step": 23685
    },
    {
      "entropy": 6.058750104904175,
      "epoch": 2.368071175088719,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004441892505427031,
      "loss": 5.6378,
      "mean_token_accuracy": 0.17618710398674012,
      "num_tokens": 46332357.0,
      "step": 23690
    },
    {
      "entropy": 6.118291139602661,
      "epoch": 2.3685710001499474,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004441657215302853,
      "loss": 5.6914,
      "mean_token_accuracy": 0.17251085937023164,
      "num_tokens": 46341805.0,
      "step": 23695
    },
    {
      "entropy": 6.060851812362671,
      "epoch": 2.369070825211176,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044414218826170155,
      "loss": 5.6482,
      "mean_token_accuracy": 0.1757635071873665,
      "num_tokens": 46352814.0,
      "step": 23700
    },
    {
      "entropy": 6.0773509502410885,
      "epoch": 2.3695706502724048,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004441186507375439,
      "loss": 5.7075,
      "mean_token_accuracy": 0.1712249204516411,
      "num_tokens": 46362968.0,
      "step": 23705
    },
    {
      "entropy": 6.0236259460449215,
      "epoch": 2.370070475333633,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044409510895840446,
      "loss": 5.6867,
      "mean_token_accuracy": 0.17789150401949883,
      "num_tokens": 46373379.0,
      "step": 23710
    },
    {
      "entropy": 5.993005800247192,
      "epoch": 2.3705703003948617,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00044407156292487553,
      "loss": 5.5718,
      "mean_token_accuracy": 0.18568568974733352,
      "num_tokens": 46382788.0,
      "step": 23715
    },
    {
      "entropy": 6.117120409011841,
      "epoch": 2.3710701254560904,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004440480126375497,
      "loss": 5.691,
      "mean_token_accuracy": 0.17374709397554397,
      "num_tokens": 46392984.0,
      "step": 23720
    },
    {
      "entropy": 6.030602884292603,
      "epoch": 2.371569950517319,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004440244580970193,
      "loss": 5.5867,
      "mean_token_accuracy": 0.17945046126842498,
      "num_tokens": 46403387.0,
      "step": 23725
    },
    {
      "entropy": 5.985666131973266,
      "epoch": 2.3720697755785474,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00044400089930387703,
      "loss": 5.5932,
      "mean_token_accuracy": 0.17470182627439498,
      "num_tokens": 46413010.0,
      "step": 23730
    },
    {
      "entropy": 6.025519323348999,
      "epoch": 2.372569600639776,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00044397733625871566,
      "loss": 5.6714,
      "mean_token_accuracy": 0.17632315158843995,
      "num_tokens": 46423790.0,
      "step": 23735
    },
    {
      "entropy": 6.1154101371765135,
      "epoch": 2.3730694257010048,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044395376896212805,
      "loss": 5.6591,
      "mean_token_accuracy": 0.1698581576347351,
      "num_tokens": 46434210.0,
      "step": 23740
    },
    {
      "entropy": 5.984733247756958,
      "epoch": 2.373569250762233,
      "grad_norm": 1.203125,
      "learning_rate": 0.00044393019741470696,
      "loss": 5.5404,
      "mean_token_accuracy": 0.1842620477080345,
      "num_tokens": 46443747.0,
      "step": 23745
    },
    {
      "entropy": 5.987794208526611,
      "epoch": 2.3740690758234617,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004439066216170456,
      "loss": 5.5518,
      "mean_token_accuracy": 0.18251242190599443,
      "num_tokens": 46453829.0,
      "step": 23750
    },
    {
      "entropy": 6.052422761917114,
      "epoch": 2.3745689008846904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044388304156973713,
      "loss": 5.7057,
      "mean_token_accuracy": 0.17490700632333755,
      "num_tokens": 46463406.0,
      "step": 23755
    },
    {
      "entropy": 6.024632310867309,
      "epoch": 2.375068725945919,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044385945727337477,
      "loss": 5.643,
      "mean_token_accuracy": 0.1769236743450165,
      "num_tokens": 46473374.0,
      "step": 23760
    },
    {
      "entropy": 6.041012239456177,
      "epoch": 2.3755685510071474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000443835868728552,
      "loss": 5.6387,
      "mean_token_accuracy": 0.17929480820894242,
      "num_tokens": 46482881.0,
      "step": 23765
    },
    {
      "entropy": 6.013991594314575,
      "epoch": 2.376068376068376,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004438122759358621,
      "loss": 5.6557,
      "mean_token_accuracy": 0.17798885852098464,
      "num_tokens": 46493040.0,
      "step": 23770
    },
    {
      "entropy": 6.064805316925049,
      "epoch": 2.3765682011296048,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044378867889589884,
      "loss": 5.6473,
      "mean_token_accuracy": 0.17287222743034364,
      "num_tokens": 46503243.0,
      "step": 23775
    },
    {
      "entropy": 6.027049732208252,
      "epoch": 2.377068026190833,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004437650776092558,
      "loss": 5.5755,
      "mean_token_accuracy": 0.177457694709301,
      "num_tokens": 46512864.0,
      "step": 23780
    },
    {
      "entropy": 6.018921613693237,
      "epoch": 2.3775678512520617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044374147207652695,
      "loss": 5.6664,
      "mean_token_accuracy": 0.184862320125103,
      "num_tokens": 46522404.0,
      "step": 23785
    },
    {
      "entropy": 5.979655027389526,
      "epoch": 2.3780676763132904,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000443717862298306,
      "loss": 5.6395,
      "mean_token_accuracy": 0.1699163407087326,
      "num_tokens": 46532235.0,
      "step": 23790
    },
    {
      "entropy": 6.023217296600341,
      "epoch": 2.378567501374519,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00044369424827518704,
      "loss": 5.6075,
      "mean_token_accuracy": 0.17641149312257767,
      "num_tokens": 46542337.0,
      "step": 23795
    },
    {
      "entropy": 6.08636245727539,
      "epoch": 2.3790673264357474,
      "grad_norm": 1.328125,
      "learning_rate": 0.00044367063000776423,
      "loss": 5.693,
      "mean_token_accuracy": 0.17525937259197236,
      "num_tokens": 46551446.0,
      "step": 23800
    },
    {
      "entropy": 6.068008756637573,
      "epoch": 2.379567151496976,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004436470074966317,
      "loss": 5.6973,
      "mean_token_accuracy": 0.17796041965484619,
      "num_tokens": 46562289.0,
      "step": 23805
    },
    {
      "entropy": 5.950279426574707,
      "epoch": 2.3800669765582048,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000443623380742384,
      "loss": 5.5768,
      "mean_token_accuracy": 0.18046288788318635,
      "num_tokens": 46572135.0,
      "step": 23810
    },
    {
      "entropy": 6.028979873657226,
      "epoch": 2.380566801619433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004435997497456153,
      "loss": 5.6471,
      "mean_token_accuracy": 0.1761978641152382,
      "num_tokens": 46581810.0,
      "step": 23815
    },
    {
      "entropy": 5.9980076313018795,
      "epoch": 2.3810666266806617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004435761145069203,
      "loss": 5.5853,
      "mean_token_accuracy": 0.1780306115746498,
      "num_tokens": 46591886.0,
      "step": 23820
    },
    {
      "entropy": 6.063993883132935,
      "epoch": 2.3815664517418904,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044355247502689374,
      "loss": 5.7169,
      "mean_token_accuracy": 0.16958054155111313,
      "num_tokens": 46601632.0,
      "step": 23825
    },
    {
      "entropy": 6.036028575897217,
      "epoch": 2.382066276803119,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004435288313061302,
      "loss": 5.6419,
      "mean_token_accuracy": 0.18105792105197907,
      "num_tokens": 46611802.0,
      "step": 23830
    },
    {
      "entropy": 6.012950325012207,
      "epoch": 2.3825661018643474,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004435051833452246,
      "loss": 5.5597,
      "mean_token_accuracy": 0.18311626464128494,
      "num_tokens": 46621965.0,
      "step": 23835
    },
    {
      "entropy": 6.045510721206665,
      "epoch": 2.383065926925576,
      "grad_norm": 1.140625,
      "learning_rate": 0.000443481531144772,
      "loss": 5.6114,
      "mean_token_accuracy": 0.17995087504386903,
      "num_tokens": 46632518.0,
      "step": 23840
    },
    {
      "entropy": 5.967634677886963,
      "epoch": 2.3835657519868048,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004434578747053674,
      "loss": 5.5677,
      "mean_token_accuracy": 0.19097691476345063,
      "num_tokens": 46642393.0,
      "step": 23845
    },
    {
      "entropy": 5.991853332519531,
      "epoch": 2.384065577048033,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044343421402760606,
      "loss": 5.586,
      "mean_token_accuracy": 0.17931993007659913,
      "num_tokens": 46653025.0,
      "step": 23850
    },
    {
      "entropy": 6.072017431259155,
      "epoch": 2.3845654021092617,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004434105491120832,
      "loss": 5.6278,
      "mean_token_accuracy": 0.1773751497268677,
      "num_tokens": 46661655.0,
      "step": 23855
    },
    {
      "entropy": 6.100204181671143,
      "epoch": 2.3850652271704904,
      "grad_norm": 1.171875,
      "learning_rate": 0.00044338687995939424,
      "loss": 5.712,
      "mean_token_accuracy": 0.17500317394733428,
      "num_tokens": 46672392.0,
      "step": 23860
    },
    {
      "entropy": 6.084134340286255,
      "epoch": 2.385565052231719,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004433632065701348,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1716666743159294,
      "num_tokens": 46683099.0,
      "step": 23865
    },
    {
      "entropy": 6.029203462600708,
      "epoch": 2.3860648772929474,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004433395289449003,
      "loss": 5.606,
      "mean_token_accuracy": 0.18551905006170272,
      "num_tokens": 46692807.0,
      "step": 23870
    },
    {
      "entropy": 5.988838529586792,
      "epoch": 2.386564702354176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004433158470842867,
      "loss": 5.5945,
      "mean_token_accuracy": 0.17838135808706285,
      "num_tokens": 46703093.0,
      "step": 23875
    },
    {
      "entropy": 5.983804845809937,
      "epoch": 2.3870645274154048,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004432921609888896,
      "loss": 5.6113,
      "mean_token_accuracy": 0.17373797595500945,
      "num_tokens": 46712487.0,
      "step": 23880
    },
    {
      "entropy": 6.033302307128906,
      "epoch": 2.387564352476633,
      "grad_norm": 1.078125,
      "learning_rate": 0.00044326847065930507,
      "loss": 5.6611,
      "mean_token_accuracy": 0.17533704340457917,
      "num_tokens": 46723627.0,
      "step": 23885
    },
    {
      "entropy": 6.022245740890503,
      "epoch": 2.3880641775378617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044324477609612907,
      "loss": 5.5515,
      "mean_token_accuracy": 0.18532310128211976,
      "num_tokens": 46733130.0,
      "step": 23890
    },
    {
      "entropy": 6.052120161056519,
      "epoch": 2.3885640025990904,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004432210772999578,
      "loss": 5.7614,
      "mean_token_accuracy": 0.16833141595125198,
      "num_tokens": 46743380.0,
      "step": 23895
    },
    {
      "entropy": 6.046127891540527,
      "epoch": 2.389063827660319,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004431973742713876,
      "loss": 5.687,
      "mean_token_accuracy": 0.16911535561084748,
      "num_tokens": 46753420.0,
      "step": 23900
    },
    {
      "entropy": 6.038953065872192,
      "epoch": 2.3895636527215474,
      "grad_norm": 1.125,
      "learning_rate": 0.0004431736670110146,
      "loss": 5.6271,
      "mean_token_accuracy": 0.174855338037014,
      "num_tokens": 46763342.0,
      "step": 23905
    },
    {
      "entropy": 5.973513841629028,
      "epoch": 2.390063477782776,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044314995551943554,
      "loss": 5.5507,
      "mean_token_accuracy": 0.18543343544006347,
      "num_tokens": 46772300.0,
      "step": 23910
    },
    {
      "entropy": 6.060969829559326,
      "epoch": 2.3905633028440048,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004431262397972468,
      "loss": 5.5812,
      "mean_token_accuracy": 0.18429369777441024,
      "num_tokens": 46781778.0,
      "step": 23915
    },
    {
      "entropy": 5.917259454727173,
      "epoch": 2.391063127905233,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044310251984504517,
      "loss": 5.4622,
      "mean_token_accuracy": 0.19555078446865082,
      "num_tokens": 46792845.0,
      "step": 23920
    },
    {
      "entropy": 6.1373742580413815,
      "epoch": 2.3915629529664617,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00044307879566342736,
      "loss": 5.7644,
      "mean_token_accuracy": 0.16085453033447267,
      "num_tokens": 46802940.0,
      "step": 23925
    },
    {
      "entropy": 6.079683685302735,
      "epoch": 2.3920627780276904,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004430550672529903,
      "loss": 5.5957,
      "mean_token_accuracy": 0.18066017031669618,
      "num_tokens": 46812493.0,
      "step": 23930
    },
    {
      "entropy": 6.075826787948609,
      "epoch": 2.392562603088919,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000443031334614331,
      "loss": 5.6549,
      "mean_token_accuracy": 0.1781351834535599,
      "num_tokens": 46821102.0,
      "step": 23935
    },
    {
      "entropy": 5.996089172363281,
      "epoch": 2.3930624281501474,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004430075977480466,
      "loss": 5.6529,
      "mean_token_accuracy": 0.17873415499925613,
      "num_tokens": 46832717.0,
      "step": 23940
    },
    {
      "entropy": 5.986906671524048,
      "epoch": 2.393562253211376,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004429838566547343,
      "loss": 5.566,
      "mean_token_accuracy": 0.1789796307682991,
      "num_tokens": 46842756.0,
      "step": 23945
    },
    {
      "entropy": 6.038371801376343,
      "epoch": 2.3940620782726048,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004429601113349913,
      "loss": 5.6743,
      "mean_token_accuracy": 0.1719118744134903,
      "num_tokens": 46853114.0,
      "step": 23950
    },
    {
      "entropy": 6.095573997497558,
      "epoch": 2.394561903333833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044293636178941516,
      "loss": 5.6204,
      "mean_token_accuracy": 0.16998432576656342,
      "num_tokens": 46862286.0,
      "step": 23955
    },
    {
      "entropy": 6.045434904098511,
      "epoch": 2.3950617283950617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004429126080186034,
      "loss": 5.6711,
      "mean_token_accuracy": 0.172371569275856,
      "num_tokens": 46872155.0,
      "step": 23960
    },
    {
      "entropy": 5.956069231033325,
      "epoch": 2.3955615534562904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044288885002315354,
      "loss": 5.5962,
      "mean_token_accuracy": 0.18287757337093352,
      "num_tokens": 46881969.0,
      "step": 23965
    },
    {
      "entropy": 6.043852376937866,
      "epoch": 2.396061378517519,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044286508780366345,
      "loss": 5.6487,
      "mean_token_accuracy": 0.17964032888412476,
      "num_tokens": 46890855.0,
      "step": 23970
    },
    {
      "entropy": 6.035159111022949,
      "epoch": 2.3965612035787474,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044284132136073097,
      "loss": 5.5744,
      "mean_token_accuracy": 0.18571922928094864,
      "num_tokens": 46900803.0,
      "step": 23975
    },
    {
      "entropy": 5.959583377838134,
      "epoch": 2.397061028639976,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004428175506949541,
      "loss": 5.6322,
      "mean_token_accuracy": 0.18005622625350953,
      "num_tokens": 46910178.0,
      "step": 23980
    },
    {
      "entropy": 6.0496374607086185,
      "epoch": 2.3975608537012048,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044279377580693077,
      "loss": 5.6134,
      "mean_token_accuracy": 0.1770930290222168,
      "num_tokens": 46919860.0,
      "step": 23985
    },
    {
      "entropy": 6.066586685180664,
      "epoch": 2.398060678762433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004427699966972592,
      "loss": 5.5933,
      "mean_token_accuracy": 0.1857215568423271,
      "num_tokens": 46929570.0,
      "step": 23990
    },
    {
      "entropy": 6.009964466094971,
      "epoch": 2.3985605038236617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044274621336653765,
      "loss": 5.6076,
      "mean_token_accuracy": 0.17551508247852327,
      "num_tokens": 46939685.0,
      "step": 23995
    },
    {
      "entropy": 5.9474201679229735,
      "epoch": 2.3990603288848904,
      "grad_norm": 1.34375,
      "learning_rate": 0.00044272242581536457,
      "loss": 5.5342,
      "mean_token_accuracy": 0.19125236421823502,
      "num_tokens": 46949720.0,
      "step": 24000
    },
    {
      "epoch": 2.3990603288848904,
      "eval_entropy": 5.821133703277517,
      "eval_loss": 5.770229816436768,
      "eval_mean_token_accuracy": 0.17952073723004824,
      "eval_num_tokens": 46949720.0,
      "eval_runtime": 23.8002,
      "eval_samples_per_second": 1304.403,
      "eval_steps_per_second": 163.066,
      "step": 24000
    },
    {
      "entropy": 6.023554801940918,
      "epoch": 2.399560153946119,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044269863404433846,
      "loss": 5.616,
      "mean_token_accuracy": 0.1782217651605606,
      "num_tokens": 46958883.0,
      "step": 24005
    },
    {
      "entropy": 6.1249762058258055,
      "epoch": 2.4000599790073474,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004426748380540578,
      "loss": 5.6955,
      "mean_token_accuracy": 0.1685681402683258,
      "num_tokens": 46969146.0,
      "step": 24010
    },
    {
      "entropy": 6.022667503356933,
      "epoch": 2.400559804068576,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004426510378451215,
      "loss": 5.68,
      "mean_token_accuracy": 0.1747714638710022,
      "num_tokens": 46978250.0,
      "step": 24015
    },
    {
      "entropy": 6.0925352573394775,
      "epoch": 2.4010596291298048,
      "grad_norm": 1.28125,
      "learning_rate": 0.00044262723341812805,
      "loss": 5.6838,
      "mean_token_accuracy": 0.17414737939834596,
      "num_tokens": 46988413.0,
      "step": 24020
    },
    {
      "entropy": 6.142070150375366,
      "epoch": 2.401559454191033,
      "grad_norm": 1.125,
      "learning_rate": 0.0004426034247736766,
      "loss": 5.7541,
      "mean_token_accuracy": 0.16979750841856003,
      "num_tokens": 46998092.0,
      "step": 24025
    },
    {
      "entropy": 6.036782121658325,
      "epoch": 2.4020592792522617,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004425796119123662,
      "loss": 5.7009,
      "mean_token_accuracy": 0.1769142910838127,
      "num_tokens": 47006764.0,
      "step": 24030
    },
    {
      "entropy": 6.06001124382019,
      "epoch": 2.4025591043134904,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044255579483479577,
      "loss": 5.6317,
      "mean_token_accuracy": 0.17491981238126755,
      "num_tokens": 47016789.0,
      "step": 24035
    },
    {
      "entropy": 6.003628969192505,
      "epoch": 2.4030589293747187,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004425319735415647,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1821936547756195,
      "num_tokens": 47026984.0,
      "step": 24040
    },
    {
      "entropy": 6.049056816101074,
      "epoch": 2.4035587544359474,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004425081480332723,
      "loss": 5.6826,
      "mean_token_accuracy": 0.17453884333372116,
      "num_tokens": 47035893.0,
      "step": 24045
    },
    {
      "entropy": 5.957960844039917,
      "epoch": 2.404058579497176,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000442484318310518,
      "loss": 5.5663,
      "mean_token_accuracy": 0.1779484048485756,
      "num_tokens": 47045351.0,
      "step": 24050
    },
    {
      "entropy": 5.96492829322815,
      "epoch": 2.4045584045584047,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044246048437390133,
      "loss": 5.6585,
      "mean_token_accuracy": 0.1754602313041687,
      "num_tokens": 47055849.0,
      "step": 24055
    },
    {
      "entropy": 6.059024000167847,
      "epoch": 2.405058229619633,
      "grad_norm": 1.21875,
      "learning_rate": 0.000442436646224022,
      "loss": 5.654,
      "mean_token_accuracy": 0.1702026441693306,
      "num_tokens": 47065068.0,
      "step": 24060
    },
    {
      "entropy": 6.00704927444458,
      "epoch": 2.4055580546808617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044241280386147965,
      "loss": 5.5727,
      "mean_token_accuracy": 0.17969609647989274,
      "num_tokens": 47075000.0,
      "step": 24065
    },
    {
      "entropy": 5.9703278064727785,
      "epoch": 2.4060578797420904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044238895728687433,
      "loss": 5.6284,
      "mean_token_accuracy": 0.17608874142169953,
      "num_tokens": 47084711.0,
      "step": 24070
    },
    {
      "entropy": 5.983452367782593,
      "epoch": 2.4065577048033187,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004423651065008059,
      "loss": 5.5228,
      "mean_token_accuracy": 0.17741337418556213,
      "num_tokens": 47094466.0,
      "step": 24075
    },
    {
      "entropy": 6.017945051193237,
      "epoch": 2.4070575298645474,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004423412515038744,
      "loss": 5.6209,
      "mean_token_accuracy": 0.1752084717154503,
      "num_tokens": 47104388.0,
      "step": 24080
    },
    {
      "entropy": 6.038263463973999,
      "epoch": 2.407557354925776,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004423173922966802,
      "loss": 5.7139,
      "mean_token_accuracy": 0.17268013209104538,
      "num_tokens": 47115158.0,
      "step": 24085
    },
    {
      "entropy": 6.12706184387207,
      "epoch": 2.4080571799870043,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044229352887982325,
      "loss": 5.7324,
      "mean_token_accuracy": 0.16658879965543746,
      "num_tokens": 47126739.0,
      "step": 24090
    },
    {
      "entropy": 6.117573404312134,
      "epoch": 2.408557005048233,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044226966125390435,
      "loss": 5.6883,
      "mean_token_accuracy": 0.17589689791202545,
      "num_tokens": 47136586.0,
      "step": 24095
    },
    {
      "entropy": 5.983213567733765,
      "epoch": 2.4090568301094617,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004422457894195237,
      "loss": 5.5791,
      "mean_token_accuracy": 0.1839229181408882,
      "num_tokens": 47146099.0,
      "step": 24100
    },
    {
      "entropy": 5.973686552047729,
      "epoch": 2.4095566551706904,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004422219133772821,
      "loss": 5.6685,
      "mean_token_accuracy": 0.1767183482646942,
      "num_tokens": 47154165.0,
      "step": 24105
    },
    {
      "entropy": 6.013171672821045,
      "epoch": 2.4100564802319187,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004421980331277802,
      "loss": 5.6549,
      "mean_token_accuracy": 0.18122874498367308,
      "num_tokens": 47164060.0,
      "step": 24110
    },
    {
      "entropy": 6.078809547424316,
      "epoch": 2.4105563052931473,
      "grad_norm": 1.125,
      "learning_rate": 0.00044217414867161873,
      "loss": 5.7364,
      "mean_token_accuracy": 0.16950304061174393,
      "num_tokens": 47174256.0,
      "step": 24115
    },
    {
      "entropy": 6.004992580413818,
      "epoch": 2.411056130354376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004421502600093987,
      "loss": 5.6113,
      "mean_token_accuracy": 0.18138744533061982,
      "num_tokens": 47183553.0,
      "step": 24120
    },
    {
      "entropy": 6.049961948394776,
      "epoch": 2.4115559554156043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044212636714172115,
      "loss": 5.6571,
      "mean_token_accuracy": 0.17427137047052382,
      "num_tokens": 47194542.0,
      "step": 24125
    },
    {
      "entropy": 6.067128467559814,
      "epoch": 2.412055780476833,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004421024700691873,
      "loss": 5.6352,
      "mean_token_accuracy": 0.17796468883752822,
      "num_tokens": 47204418.0,
      "step": 24130
    },
    {
      "entropy": 5.968400239944458,
      "epoch": 2.4125556055380617,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004420785687923982,
      "loss": 5.5532,
      "mean_token_accuracy": 0.19123557358980178,
      "num_tokens": 47212727.0,
      "step": 24135
    },
    {
      "entropy": 5.9016228199005125,
      "epoch": 2.4130554305992904,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004420546633119554,
      "loss": 5.6544,
      "mean_token_accuracy": 0.18143839240074158,
      "num_tokens": 47222907.0,
      "step": 24140
    },
    {
      "entropy": 5.959653568267822,
      "epoch": 2.4135552556605187,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004420307536284602,
      "loss": 5.5747,
      "mean_token_accuracy": 0.1852305218577385,
      "num_tokens": 47233170.0,
      "step": 24145
    },
    {
      "entropy": 6.111540937423706,
      "epoch": 2.4140550807217473,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044200683974251424,
      "loss": 5.5922,
      "mean_token_accuracy": 0.18360538482666017,
      "num_tokens": 47243747.0,
      "step": 24150
    },
    {
      "entropy": 6.011303806304932,
      "epoch": 2.414554905782976,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004419829216547191,
      "loss": 5.6196,
      "mean_token_accuracy": 0.17668455243110656,
      "num_tokens": 47253372.0,
      "step": 24155
    },
    {
      "entropy": 5.973196268081665,
      "epoch": 2.4150547308442043,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004419589993656767,
      "loss": 5.6179,
      "mean_token_accuracy": 0.1835993528366089,
      "num_tokens": 47263190.0,
      "step": 24160
    },
    {
      "entropy": 6.081429672241211,
      "epoch": 2.415554555905433,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004419350728759888,
      "loss": 5.7623,
      "mean_token_accuracy": 0.17463454306125642,
      "num_tokens": 47273702.0,
      "step": 24165
    },
    {
      "entropy": 6.045127105712891,
      "epoch": 2.4160543809666617,
      "grad_norm": 1.1875,
      "learning_rate": 0.00044191114218625736,
      "loss": 5.5578,
      "mean_token_accuracy": 0.18675146847963334,
      "num_tokens": 47282613.0,
      "step": 24170
    },
    {
      "entropy": 5.959947490692139,
      "epoch": 2.4165542060278904,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004418872072970846,
      "loss": 5.5168,
      "mean_token_accuracy": 0.19066780805587769,
      "num_tokens": 47292515.0,
      "step": 24175
    },
    {
      "entropy": 6.107087469100952,
      "epoch": 2.4170540310891186,
      "grad_norm": 1.453125,
      "learning_rate": 0.0004418632682090727,
      "loss": 5.6433,
      "mean_token_accuracy": 0.1733883649110794,
      "num_tokens": 47303258.0,
      "step": 24180
    },
    {
      "entropy": 6.022756910324096,
      "epoch": 2.4175538561503473,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004418393249228238,
      "loss": 5.722,
      "mean_token_accuracy": 0.1668480381369591,
      "num_tokens": 47313544.0,
      "step": 24185
    },
    {
      "entropy": 6.138251066207886,
      "epoch": 2.418053681211576,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004418153774389405,
      "loss": 5.7529,
      "mean_token_accuracy": 0.16700546741485595,
      "num_tokens": 47325221.0,
      "step": 24190
    },
    {
      "entropy": 6.101276493072509,
      "epoch": 2.4185535062728043,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004417914257580252,
      "loss": 5.6652,
      "mean_token_accuracy": 0.1752559795975685,
      "num_tokens": 47335043.0,
      "step": 24195
    },
    {
      "entropy": 6.025561904907226,
      "epoch": 2.419053331334033,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004417674698806805,
      "loss": 5.6423,
      "mean_token_accuracy": 0.17508832067251207,
      "num_tokens": 47343572.0,
      "step": 24200
    },
    {
      "entropy": 6.066837739944458,
      "epoch": 2.4195531563952617,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004417435098075093,
      "loss": 5.7511,
      "mean_token_accuracy": 0.17415841072797775,
      "num_tokens": 47352944.0,
      "step": 24205
    },
    {
      "entropy": 6.01858344078064,
      "epoch": 2.4200529814564904,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004417195455391142,
      "loss": 5.6158,
      "mean_token_accuracy": 0.18179512321949004,
      "num_tokens": 47362614.0,
      "step": 24210
    },
    {
      "entropy": 5.991080331802368,
      "epoch": 2.4205528065177186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044169557707609826,
      "loss": 5.5978,
      "mean_token_accuracy": 0.17907897382974625,
      "num_tokens": 47372712.0,
      "step": 24215
    },
    {
      "entropy": 5.9922222137451175,
      "epoch": 2.4210526315789473,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004416716044190644,
      "loss": 5.6781,
      "mean_token_accuracy": 0.1747306242585182,
      "num_tokens": 47382570.0,
      "step": 24220
    },
    {
      "entropy": 6.020452880859375,
      "epoch": 2.421552456640176,
      "grad_norm": 1.28125,
      "learning_rate": 0.000441647627568616,
      "loss": 5.5937,
      "mean_token_accuracy": 0.1834121823310852,
      "num_tokens": 47392537.0,
      "step": 24225
    },
    {
      "entropy": 5.981452798843383,
      "epoch": 2.4220522817014043,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044162364652535613,
      "loss": 5.6188,
      "mean_token_accuracy": 0.1810839831829071,
      "num_tokens": 47403213.0,
      "step": 24230
    },
    {
      "entropy": 6.062117719650269,
      "epoch": 2.422552106762633,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004415996612898882,
      "loss": 5.6373,
      "mean_token_accuracy": 0.17094153612852098,
      "num_tokens": 47413832.0,
      "step": 24235
    },
    {
      "entropy": 5.9899232387542725,
      "epoch": 2.4230519318238617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044157567186281553,
      "loss": 5.6129,
      "mean_token_accuracy": 0.17884231060743333,
      "num_tokens": 47423051.0,
      "step": 24240
    },
    {
      "entropy": 6.047880506515503,
      "epoch": 2.4235517568850904,
      "grad_norm": 1.265625,
      "learning_rate": 0.000441551678244742,
      "loss": 5.7288,
      "mean_token_accuracy": 0.1723594382405281,
      "num_tokens": 47431755.0,
      "step": 24245
    },
    {
      "entropy": 6.098805952072143,
      "epoch": 2.4240515819463186,
      "grad_norm": 1.09375,
      "learning_rate": 0.000441527680436271,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17861958891153334,
      "num_tokens": 47441510.0,
      "step": 24250
    },
    {
      "entropy": 6.0698505401611325,
      "epoch": 2.4245514070075473,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044150367843800645,
      "loss": 5.661,
      "mean_token_accuracy": 0.173063001036644,
      "num_tokens": 47451559.0,
      "step": 24255
    },
    {
      "entropy": 6.0517051219940186,
      "epoch": 2.425051232068776,
      "grad_norm": 1.125,
      "learning_rate": 0.0004414796722505522,
      "loss": 5.6004,
      "mean_token_accuracy": 0.1733100339770317,
      "num_tokens": 47462068.0,
      "step": 24260
    },
    {
      "entropy": 6.039041900634766,
      "epoch": 2.4255510571300043,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004414556618745122,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18225948214530946,
      "num_tokens": 47470684.0,
      "step": 24265
    },
    {
      "entropy": 6.035745096206665,
      "epoch": 2.426050882191233,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044143164731049063,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18886494487524033,
      "num_tokens": 47480147.0,
      "step": 24270
    },
    {
      "entropy": 6.032062435150147,
      "epoch": 2.4265507072524617,
      "grad_norm": 1.4296875,
      "learning_rate": 0.0004414076285590916,
      "loss": 5.6584,
      "mean_token_accuracy": 0.17310948222875594,
      "num_tokens": 47489499.0,
      "step": 24275
    },
    {
      "entropy": 6.023316717147827,
      "epoch": 2.4270505323136904,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004413836056209195,
      "loss": 5.4621,
      "mean_token_accuracy": 0.19797069132328032,
      "num_tokens": 47500155.0,
      "step": 24280
    },
    {
      "entropy": 6.010325193405151,
      "epoch": 2.4275503573749186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00044135957849657866,
      "loss": 5.6054,
      "mean_token_accuracy": 0.17980749011039734,
      "num_tokens": 47510302.0,
      "step": 24285
    },
    {
      "entropy": 5.990495395660401,
      "epoch": 2.4280501824361473,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044133554718667363,
      "loss": 5.6401,
      "mean_token_accuracy": 0.17334047853946685,
      "num_tokens": 47519997.0,
      "step": 24290
    },
    {
      "entropy": 6.064727258682251,
      "epoch": 2.428550007497376,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044131151169180905,
      "loss": 5.6524,
      "mean_token_accuracy": 0.17154275625944138,
      "num_tokens": 47528417.0,
      "step": 24295
    },
    {
      "entropy": 6.037304973602295,
      "epoch": 2.4290498325586043,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044128747201258956,
      "loss": 5.5461,
      "mean_token_accuracy": 0.18078582882881164,
      "num_tokens": 47538461.0,
      "step": 24300
    },
    {
      "entropy": 6.078216314315796,
      "epoch": 2.429549657619833,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044126342814962015,
      "loss": 5.5851,
      "mean_token_accuracy": 0.1870872750878334,
      "num_tokens": 47547256.0,
      "step": 24305
    },
    {
      "entropy": 5.993676090240479,
      "epoch": 2.4300494826810617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00044123938010350575,
      "loss": 5.6511,
      "mean_token_accuracy": 0.17038078606128693,
      "num_tokens": 47556827.0,
      "step": 24310
    },
    {
      "entropy": 6.017264413833618,
      "epoch": 2.4305493077422904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044121532787485114,
      "loss": 5.5476,
      "mean_token_accuracy": 0.18001786023378372,
      "num_tokens": 47566600.0,
      "step": 24315
    },
    {
      "entropy": 6.026751613616943,
      "epoch": 2.4310491328035186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044119127146426167,
      "loss": 5.5607,
      "mean_token_accuracy": 0.17912608683109282,
      "num_tokens": 47575849.0,
      "step": 24320
    },
    {
      "entropy": 6.0941685199737545,
      "epoch": 2.4315489578647473,
      "grad_norm": 1.125,
      "learning_rate": 0.00044116721087234266,
      "loss": 5.7081,
      "mean_token_accuracy": 0.17261117100715637,
      "num_tokens": 47586417.0,
      "step": 24325
    },
    {
      "entropy": 6.074436140060425,
      "epoch": 2.432048782925976,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004411431460996994,
      "loss": 5.693,
      "mean_token_accuracy": 0.17241280749440194,
      "num_tokens": 47596917.0,
      "step": 24330
    },
    {
      "entropy": 6.095558738708496,
      "epoch": 2.4325486079872043,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004411190771469373,
      "loss": 5.7076,
      "mean_token_accuracy": 0.17015113383531572,
      "num_tokens": 47607231.0,
      "step": 24335
    },
    {
      "entropy": 6.0676453590393065,
      "epoch": 2.433048433048433,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044109500401466194,
      "loss": 5.662,
      "mean_token_accuracy": 0.1782022163271904,
      "num_tokens": 47616320.0,
      "step": 24340
    },
    {
      "entropy": 6.02717695236206,
      "epoch": 2.4335482581096617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004410709267034791,
      "loss": 5.6333,
      "mean_token_accuracy": 0.17748998254537582,
      "num_tokens": 47625391.0,
      "step": 24345
    },
    {
      "entropy": 6.038974952697754,
      "epoch": 2.4340480831708904,
      "grad_norm": 1.125,
      "learning_rate": 0.00044104684521399434,
      "loss": 5.6522,
      "mean_token_accuracy": 0.17077535390853882,
      "num_tokens": 47635474.0,
      "step": 24350
    },
    {
      "entropy": 6.072029876708984,
      "epoch": 2.4345479082321186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004410227595468138,
      "loss": 5.592,
      "mean_token_accuracy": 0.17991064339876175,
      "num_tokens": 47644860.0,
      "step": 24355
    },
    {
      "entropy": 5.967791271209717,
      "epoch": 2.4350477332933473,
      "grad_norm": 1.15625,
      "learning_rate": 0.00044099866970254327,
      "loss": 5.6328,
      "mean_token_accuracy": 0.18850301951169968,
      "num_tokens": 47654531.0,
      "step": 24360
    },
    {
      "entropy": 5.923406934738159,
      "epoch": 2.435547558354576,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044097457568178903,
      "loss": 5.5032,
      "mean_token_accuracy": 0.19691420793533326,
      "num_tokens": 47663110.0,
      "step": 24365
    },
    {
      "entropy": 5.975079917907715,
      "epoch": 2.4360473834158043,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00044095047748515706,
      "loss": 5.5316,
      "mean_token_accuracy": 0.18982431739568711,
      "num_tokens": 47672466.0,
      "step": 24370
    },
    {
      "entropy": 6.077485609054565,
      "epoch": 2.436547208477033,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044092637511325384,
      "loss": 5.6711,
      "mean_token_accuracy": 0.18393763452768325,
      "num_tokens": 47681949.0,
      "step": 24375
    },
    {
      "entropy": 6.039769887924194,
      "epoch": 2.4370470335382617,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004409022685666857,
      "loss": 5.6222,
      "mean_token_accuracy": 0.18226393908262253,
      "num_tokens": 47692515.0,
      "step": 24380
    },
    {
      "entropy": 6.054116439819336,
      "epoch": 2.4375468585994904,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004408781578460592,
      "loss": 5.743,
      "mean_token_accuracy": 0.17137934565544127,
      "num_tokens": 47701604.0,
      "step": 24385
    },
    {
      "entropy": 5.989136409759522,
      "epoch": 2.4380466836607186,
      "grad_norm": 1.046875,
      "learning_rate": 0.00044085404295198097,
      "loss": 5.5634,
      "mean_token_accuracy": 0.1819287955760956,
      "num_tokens": 47711567.0,
      "step": 24390
    },
    {
      "entropy": 6.096336841583252,
      "epoch": 2.4385465087219473,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004408299238850576,
      "loss": 5.7119,
      "mean_token_accuracy": 0.17520872950553895,
      "num_tokens": 47721575.0,
      "step": 24395
    },
    {
      "entropy": 6.104909086227417,
      "epoch": 2.439046333783176,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004408058006458961,
      "loss": 5.7093,
      "mean_token_accuracy": 0.17287303507328033,
      "num_tokens": 47731510.0,
      "step": 24400
    },
    {
      "entropy": 6.059074831008911,
      "epoch": 2.4395461588444043,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004407816732351032,
      "loss": 5.7224,
      "mean_token_accuracy": 0.17018707394599913,
      "num_tokens": 47741842.0,
      "step": 24405
    },
    {
      "entropy": 6.095255041122437,
      "epoch": 2.440045983905633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00044075754165328614,
      "loss": 5.7074,
      "mean_token_accuracy": 0.1715589165687561,
      "num_tokens": 47752074.0,
      "step": 24410
    },
    {
      "entropy": 6.097635793685913,
      "epoch": 2.4405458089668617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000440733405901052,
      "loss": 5.6519,
      "mean_token_accuracy": 0.17655315697193147,
      "num_tokens": 47760584.0,
      "step": 24415
    },
    {
      "entropy": 6.072246360778808,
      "epoch": 2.4410456340280904,
      "grad_norm": 1.140625,
      "learning_rate": 0.00044070926597900804,
      "loss": 5.6994,
      "mean_token_accuracy": 0.17952407151460648,
      "num_tokens": 47770744.0,
      "step": 24420
    },
    {
      "entropy": 6.089491271972657,
      "epoch": 2.4415454590893186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004406851218877615,
      "loss": 5.7073,
      "mean_token_accuracy": 0.17447005063295365,
      "num_tokens": 47780098.0,
      "step": 24425
    },
    {
      "entropy": 6.006195449829102,
      "epoch": 2.4420452841505473,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00044066097362792,
      "loss": 5.5999,
      "mean_token_accuracy": 0.17984318137168884,
      "num_tokens": 47789807.0,
      "step": 24430
    },
    {
      "entropy": 6.021993112564087,
      "epoch": 2.442545109211776,
      "grad_norm": 1.3359375,
      "learning_rate": 0.000440636821200091,
      "loss": 5.5779,
      "mean_token_accuracy": 0.1863965645432472,
      "num_tokens": 47798110.0,
      "step": 24435
    },
    {
      "entropy": 6.057878589630127,
      "epoch": 2.4430449342730043,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004406126646048822,
      "loss": 5.6768,
      "mean_token_accuracy": 0.17468643933534622,
      "num_tokens": 47809084.0,
      "step": 24440
    },
    {
      "entropy": 6.173198509216308,
      "epoch": 2.443544759334233,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004405885038429014,
      "loss": 5.6846,
      "mean_token_accuracy": 0.17383741289377214,
      "num_tokens": 47819694.0,
      "step": 24445
    },
    {
      "entropy": 6.054164743423462,
      "epoch": 2.4440445843954617,
      "grad_norm": 1.09375,
      "learning_rate": 0.00044056433891475645,
      "loss": 5.6718,
      "mean_token_accuracy": 0.17495127171278,
      "num_tokens": 47828924.0,
      "step": 24450
    },
    {
      "entropy": 5.928030681610108,
      "epoch": 2.4445444094566904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004405401698210553,
      "loss": 5.5149,
      "mean_token_accuracy": 0.18488841950893403,
      "num_tokens": 47838992.0,
      "step": 24455
    },
    {
      "entropy": 6.040327262878418,
      "epoch": 2.4450442345179186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004405159965624061,
      "loss": 5.5874,
      "mean_token_accuracy": 0.17767405062913894,
      "num_tokens": 47848682.0,
      "step": 24460
    },
    {
      "entropy": 6.055491209030151,
      "epoch": 2.4455440595791473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.000440491819139417,
      "loss": 5.6305,
      "mean_token_accuracy": 0.18010813593864441,
      "num_tokens": 47858574.0,
      "step": 24465
    },
    {
      "entropy": 5.984416532516479,
      "epoch": 2.446043884640376,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004404676375526963,
      "loss": 5.6774,
      "mean_token_accuracy": 0.18010955303907394,
      "num_tokens": 47868455.0,
      "step": 24470
    },
    {
      "entropy": 5.954157924652099,
      "epoch": 2.4465437097016043,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00044044345180285245,
      "loss": 5.6319,
      "mean_token_accuracy": 0.1780990481376648,
      "num_tokens": 47878718.0,
      "step": 24475
    },
    {
      "entropy": 6.057654142379761,
      "epoch": 2.447043534762833,
      "grad_norm": 1.203125,
      "learning_rate": 0.000440419261890494,
      "loss": 5.5676,
      "mean_token_accuracy": 0.1834159567952156,
      "num_tokens": 47887444.0,
      "step": 24480
    },
    {
      "entropy": 6.02215838432312,
      "epoch": 2.4475433598240617,
      "grad_norm": 1.375,
      "learning_rate": 0.00044039506781622936,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18781089335680007,
      "num_tokens": 47895935.0,
      "step": 24485
    },
    {
      "entropy": 6.000536489486694,
      "epoch": 2.4480431848852904,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004403708695806674,
      "loss": 5.6498,
      "mean_token_accuracy": 0.17686448991298676,
      "num_tokens": 47906552.0,
      "step": 24490
    },
    {
      "entropy": 5.973035669326782,
      "epoch": 2.4485430099465186,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004403466671844169,
      "loss": 5.5962,
      "mean_token_accuracy": 0.1853729099035263,
      "num_tokens": 47916691.0,
      "step": 24495
    },
    {
      "entropy": 6.054156684875489,
      "epoch": 2.4490428350077473,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004403224606280869,
      "loss": 5.6272,
      "mean_token_accuracy": 0.1815432071685791,
      "num_tokens": 47926579.0,
      "step": 24500
    },
    {
      "entropy": 6.010115003585815,
      "epoch": 2.449542660068976,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004402982499122862,
      "loss": 5.4852,
      "mean_token_accuracy": 0.18931954056024553,
      "num_tokens": 47935889.0,
      "step": 24505
    },
    {
      "entropy": 5.992005825042725,
      "epoch": 2.4500424851302043,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044027403503762416,
      "loss": 5.5912,
      "mean_token_accuracy": 0.17804358750581742,
      "num_tokens": 47944993.0,
      "step": 24510
    },
    {
      "entropy": 6.085336399078369,
      "epoch": 2.450542310191433,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004402498160047099,
      "loss": 5.6749,
      "mean_token_accuracy": 0.17265985608100892,
      "num_tokens": 47954984.0,
      "step": 24515
    },
    {
      "entropy": 6.0319281101226805,
      "epoch": 2.4510421352526617,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004402255928141528,
      "loss": 5.6535,
      "mean_token_accuracy": 0.1767471507191658,
      "num_tokens": 47964986.0,
      "step": 24520
    },
    {
      "entropy": 6.07125506401062,
      "epoch": 2.4515419603138904,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004402013654665623,
      "loss": 5.6733,
      "mean_token_accuracy": 0.17612204998731612,
      "num_tokens": 47975962.0,
      "step": 24525
    },
    {
      "entropy": 6.100185966491699,
      "epoch": 2.4520417853751186,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00044017713396254796,
      "loss": 5.7093,
      "mean_token_accuracy": 0.17525184899568558,
      "num_tokens": 47985159.0,
      "step": 24530
    },
    {
      "entropy": 5.986420392990112,
      "epoch": 2.4525416104363473,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00044015289830271945,
      "loss": 5.5696,
      "mean_token_accuracy": 0.18415430784225464,
      "num_tokens": 47993634.0,
      "step": 24535
    },
    {
      "entropy": 6.006438827514648,
      "epoch": 2.453041435497576,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00044012865848768644,
      "loss": 5.6127,
      "mean_token_accuracy": 0.17167626619338988,
      "num_tokens": 48003573.0,
      "step": 24540
    },
    {
      "entropy": 5.923998260498047,
      "epoch": 2.4535412605588043,
      "grad_norm": 1.21875,
      "learning_rate": 0.00044010441451805896,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1885721653699875,
      "num_tokens": 48013688.0,
      "step": 24545
    },
    {
      "entropy": 6.031884813308716,
      "epoch": 2.454041085620033,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004400801663944468,
      "loss": 5.483,
      "mean_token_accuracy": 0.18609380722045898,
      "num_tokens": 48022629.0,
      "step": 24550
    },
    {
      "entropy": 6.0448671817779545,
      "epoch": 2.4545409106812617,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00044005591411746023,
      "loss": 5.667,
      "mean_token_accuracy": 0.17622793316841126,
      "num_tokens": 48033495.0,
      "step": 24555
    },
    {
      "entropy": 6.011681175231933,
      "epoch": 2.4550407357424904,
      "grad_norm": 1.25,
      "learning_rate": 0.00044003165768770927,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18541127145290376,
      "num_tokens": 48043471.0,
      "step": 24560
    },
    {
      "entropy": 6.0248839378356935,
      "epoch": 2.4555405608037186,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004400073971058043,
      "loss": 5.6695,
      "mean_token_accuracy": 0.1759689912199974,
      "num_tokens": 48052811.0,
      "step": 24565
    },
    {
      "entropy": 6.05173807144165,
      "epoch": 2.4560403858649473,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043998313237235566,
      "loss": 5.6228,
      "mean_token_accuracy": 0.17885940074920653,
      "num_tokens": 48062826.0,
      "step": 24570
    },
    {
      "entropy": 6.036870765686035,
      "epoch": 2.456540210926176,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004399588634879739,
      "loss": 5.6236,
      "mean_token_accuracy": 0.17361117899417877,
      "num_tokens": 48072867.0,
      "step": 24575
    },
    {
      "entropy": 6.036831283569336,
      "epoch": 2.4570400359874043,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004399345904532695,
      "loss": 5.6224,
      "mean_token_accuracy": 0.17903644293546678,
      "num_tokens": 48082816.0,
      "step": 24580
    },
    {
      "entropy": 6.027196311950684,
      "epoch": 2.457539861048633,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004399103132688533,
      "loss": 5.6158,
      "mean_token_accuracy": 0.18493862748146056,
      "num_tokens": 48092715.0,
      "step": 24585
    },
    {
      "entropy": 6.027986145019531,
      "epoch": 2.4580396861098617,
      "grad_norm": 1.3125,
      "learning_rate": 0.00043988603193533606,
      "loss": 5.5989,
      "mean_token_accuracy": 0.17064955234527587,
      "num_tokens": 48101372.0,
      "step": 24590
    },
    {
      "entropy": 6.011467790603637,
      "epoch": 2.45853951117109,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004398617464533287,
      "loss": 5.6041,
      "mean_token_accuracy": 0.1853222742676735,
      "num_tokens": 48111894.0,
      "step": 24595
    },
    {
      "entropy": 6.098705005645752,
      "epoch": 2.4590393362323186,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004398374568234422,
      "loss": 5.7736,
      "mean_token_accuracy": 0.15889275968074798,
      "num_tokens": 48121480.0,
      "step": 24600
    },
    {
      "entropy": 6.068133974075318,
      "epoch": 2.4595391612935473,
      "grad_norm": 1.125,
      "learning_rate": 0.0004398131630462877,
      "loss": 5.6303,
      "mean_token_accuracy": 0.1769583359360695,
      "num_tokens": 48130404.0,
      "step": 24605
    },
    {
      "entropy": 6.1115068912506105,
      "epoch": 2.4600389863547756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043978886512247646,
      "loss": 5.6485,
      "mean_token_accuracy": 0.17188361585140227,
      "num_tokens": 48141218.0,
      "step": 24610
    },
    {
      "entropy": 6.069027471542358,
      "epoch": 2.4605388114160043,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004397645630526197,
      "loss": 5.6642,
      "mean_token_accuracy": 0.17690529823303222,
      "num_tokens": 48151817.0,
      "step": 24615
    },
    {
      "entropy": 6.0522377490997314,
      "epoch": 2.461038636477233,
      "grad_norm": 1.140625,
      "learning_rate": 0.000439740256837329,
      "loss": 5.5717,
      "mean_token_accuracy": 0.18075603991746902,
      "num_tokens": 48161044.0,
      "step": 24620
    },
    {
      "entropy": 6.024737167358398,
      "epoch": 2.4615384615384617,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004397159464772158,
      "loss": 5.6604,
      "mean_token_accuracy": 0.17505499869585037,
      "num_tokens": 48170139.0,
      "step": 24625
    },
    {
      "entropy": 5.963724851608276,
      "epoch": 2.46203828659969,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043969163197289174,
      "loss": 5.6393,
      "mean_token_accuracy": 0.18492036759853364,
      "num_tokens": 48179418.0,
      "step": 24630
    },
    {
      "entropy": 6.067753028869629,
      "epoch": 2.4625381116609186,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004396673133249687,
      "loss": 5.6147,
      "mean_token_accuracy": 0.18043505996465684,
      "num_tokens": 48188877.0,
      "step": 24635
    },
    {
      "entropy": 6.102846431732178,
      "epoch": 2.4630379367221473,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004396429905340584,
      "loss": 5.7009,
      "mean_token_accuracy": 0.1701342761516571,
      "num_tokens": 48198969.0,
      "step": 24640
    },
    {
      "entropy": 6.16652626991272,
      "epoch": 2.4635377617833756,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043961866360077266,
      "loss": 5.769,
      "mean_token_accuracy": 0.17000649720430375,
      "num_tokens": 48209814.0,
      "step": 24645
    },
    {
      "entropy": 6.020314359664917,
      "epoch": 2.4640375868446043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004395943325257239,
      "loss": 5.6321,
      "mean_token_accuracy": 0.18322097659111022,
      "num_tokens": 48218810.0,
      "step": 24650
    },
    {
      "entropy": 6.022289562225342,
      "epoch": 2.464537411905833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043956999730952403,
      "loss": 5.6225,
      "mean_token_accuracy": 0.18003946393728257,
      "num_tokens": 48228484.0,
      "step": 24655
    },
    {
      "entropy": 6.049333715438843,
      "epoch": 2.4650372369670617,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043954565795278536,
      "loss": 5.6731,
      "mean_token_accuracy": 0.17862130105495452,
      "num_tokens": 48239029.0,
      "step": 24660
    },
    {
      "entropy": 6.068674850463867,
      "epoch": 2.46553706202829,
      "grad_norm": 1.125,
      "learning_rate": 0.0004395213144561203,
      "loss": 5.6008,
      "mean_token_accuracy": 0.1798049822449684,
      "num_tokens": 48249448.0,
      "step": 24665
    },
    {
      "entropy": 6.053813886642456,
      "epoch": 2.4660368870895186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004394969668201413,
      "loss": 5.5956,
      "mean_token_accuracy": 0.18056170791387557,
      "num_tokens": 48258631.0,
      "step": 24670
    },
    {
      "entropy": 5.948708200454712,
      "epoch": 2.4665367121507473,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004394726150454609,
      "loss": 5.4987,
      "mean_token_accuracy": 0.18710066825151445,
      "num_tokens": 48266800.0,
      "step": 24675
    },
    {
      "entropy": 6.0495470523834225,
      "epoch": 2.4670365372119756,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004394482591326918,
      "loss": 5.6737,
      "mean_token_accuracy": 0.18270571678876876,
      "num_tokens": 48277566.0,
      "step": 24680
    },
    {
      "entropy": 6.032958269119263,
      "epoch": 2.4675363622732043,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043942389908244693,
      "loss": 5.6128,
      "mean_token_accuracy": 0.17813760042190552,
      "num_tokens": 48287033.0,
      "step": 24685
    },
    {
      "entropy": 6.002897691726685,
      "epoch": 2.468036187334433,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000439399534895339,
      "loss": 5.6117,
      "mean_token_accuracy": 0.17471453696489334,
      "num_tokens": 48296627.0,
      "step": 24690
    },
    {
      "entropy": 6.0418625831604,
      "epoch": 2.4685360123956617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004393751665719811,
      "loss": 5.5941,
      "mean_token_accuracy": 0.17930218875408171,
      "num_tokens": 48305940.0,
      "step": 24695
    },
    {
      "entropy": 5.9872547626495365,
      "epoch": 2.46903583745689,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004393507941129863,
      "loss": 5.5723,
      "mean_token_accuracy": 0.17916447520256043,
      "num_tokens": 48315785.0,
      "step": 24700
    },
    {
      "entropy": 6.009989261627197,
      "epoch": 2.4695356625181186,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004393264175189678,
      "loss": 5.6241,
      "mean_token_accuracy": 0.18514046519994737,
      "num_tokens": 48325803.0,
      "step": 24705
    },
    {
      "entropy": 6.067319440841675,
      "epoch": 2.4700354875793473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043930203679053885,
      "loss": 5.6131,
      "mean_token_accuracy": 0.1813179448246956,
      "num_tokens": 48334860.0,
      "step": 24710
    },
    {
      "entropy": 6.01424880027771,
      "epoch": 2.4705353126405756,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043927765192831305,
      "loss": 5.6605,
      "mean_token_accuracy": 0.17389233708381652,
      "num_tokens": 48344939.0,
      "step": 24715
    },
    {
      "entropy": 6.0112604141235355,
      "epoch": 2.4710351377018043,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004392532629329038,
      "loss": 5.6332,
      "mean_token_accuracy": 0.17982484698295592,
      "num_tokens": 48354493.0,
      "step": 24720
    },
    {
      "entropy": 5.997668075561523,
      "epoch": 2.471534962763033,
      "grad_norm": 1.125,
      "learning_rate": 0.00043922886980492465,
      "loss": 5.6065,
      "mean_token_accuracy": 0.18695901334285736,
      "num_tokens": 48364650.0,
      "step": 24725
    },
    {
      "entropy": 6.048045063018799,
      "epoch": 2.4720347878242617,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043920447254498947,
      "loss": 5.688,
      "mean_token_accuracy": 0.17523399293422698,
      "num_tokens": 48374373.0,
      "step": 24730
    },
    {
      "entropy": 6.0222100734710695,
      "epoch": 2.47253461288549,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004391800711537119,
      "loss": 5.5729,
      "mean_token_accuracy": 0.17724181413650514,
      "num_tokens": 48384488.0,
      "step": 24735
    },
    {
      "entropy": 6.087334823608399,
      "epoch": 2.4730344379467186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043915566563170605,
      "loss": 5.6657,
      "mean_token_accuracy": 0.1771863207221031,
      "num_tokens": 48395083.0,
      "step": 24740
    },
    {
      "entropy": 6.030396223068237,
      "epoch": 2.4735342630079473,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043913125597958594,
      "loss": 5.5626,
      "mean_token_accuracy": 0.18976465314626695,
      "num_tokens": 48404229.0,
      "step": 24745
    },
    {
      "entropy": 6.043383693695068,
      "epoch": 2.4740340880691756,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043910684219796564,
      "loss": 5.6819,
      "mean_token_accuracy": 0.17328319996595382,
      "num_tokens": 48413521.0,
      "step": 24750
    },
    {
      "entropy": 5.9874022006988525,
      "epoch": 2.4745339131304043,
      "grad_norm": 1.25,
      "learning_rate": 0.00043908242428745937,
      "loss": 5.6048,
      "mean_token_accuracy": 0.18587420433759688,
      "num_tokens": 48423229.0,
      "step": 24755
    },
    {
      "entropy": 6.070016384124756,
      "epoch": 2.475033738191633,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043905800224868153,
      "loss": 5.6689,
      "mean_token_accuracy": 0.17651266157627105,
      "num_tokens": 48432261.0,
      "step": 24760
    },
    {
      "entropy": 6.074644374847412,
      "epoch": 2.4755335632528617,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004390335760822466,
      "loss": 5.6087,
      "mean_token_accuracy": 0.1795772820711136,
      "num_tokens": 48441484.0,
      "step": 24765
    },
    {
      "entropy": 6.024156856536865,
      "epoch": 2.47603338831409,
      "grad_norm": 1.4765625,
      "learning_rate": 0.0004390091457887691,
      "loss": 5.5366,
      "mean_token_accuracy": 0.19025348275899887,
      "num_tokens": 48452055.0,
      "step": 24770
    },
    {
      "entropy": 5.909773588180542,
      "epoch": 2.4765332133753186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004389847113688637,
      "loss": 5.4366,
      "mean_token_accuracy": 0.19471009075641632,
      "num_tokens": 48460349.0,
      "step": 24775
    },
    {
      "entropy": 5.974616050720215,
      "epoch": 2.4770330384365473,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043896027282314516,
      "loss": 5.6217,
      "mean_token_accuracy": 0.17966757267713546,
      "num_tokens": 48470625.0,
      "step": 24780
    },
    {
      "entropy": 6.014792823791504,
      "epoch": 2.4775328634977756,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004389358301522283,
      "loss": 5.5832,
      "mean_token_accuracy": 0.18376269936561584,
      "num_tokens": 48480866.0,
      "step": 24785
    },
    {
      "entropy": 5.953444242477417,
      "epoch": 2.4780326885590043,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043891138335672825,
      "loss": 5.5592,
      "mean_token_accuracy": 0.18722262978553772,
      "num_tokens": 48490484.0,
      "step": 24790
    },
    {
      "entropy": 6.014924335479736,
      "epoch": 2.478532513620233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004388869324372599,
      "loss": 5.6145,
      "mean_token_accuracy": 0.1766123354434967,
      "num_tokens": 48501139.0,
      "step": 24795
    },
    {
      "entropy": 6.119218015670777,
      "epoch": 2.4790323386814617,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004388624773944385,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17633720189332963,
      "num_tokens": 48510793.0,
      "step": 24800
    },
    {
      "entropy": 6.121218824386597,
      "epoch": 2.47953216374269,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043883801822887933,
      "loss": 5.7866,
      "mean_token_accuracy": 0.1680760458111763,
      "num_tokens": 48521193.0,
      "step": 24805
    },
    {
      "entropy": 6.033409976959229,
      "epoch": 2.4800319888039186,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043881355494119773,
      "loss": 5.5879,
      "mean_token_accuracy": 0.18876389861106874,
      "num_tokens": 48530303.0,
      "step": 24810
    },
    {
      "entropy": 6.091024875640869,
      "epoch": 2.4805318138651473,
      "grad_norm": 1.125,
      "learning_rate": 0.0004387890875320093,
      "loss": 5.6562,
      "mean_token_accuracy": 0.17637317031621932,
      "num_tokens": 48539633.0,
      "step": 24815
    },
    {
      "entropy": 6.118903064727784,
      "epoch": 2.4810316389263756,
      "grad_norm": 1.375,
      "learning_rate": 0.00043876461600192955,
      "loss": 5.6007,
      "mean_token_accuracy": 0.19334306120872496,
      "num_tokens": 48550097.0,
      "step": 24820
    },
    {
      "entropy": 5.967060661315918,
      "epoch": 2.4815314639876043,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004387401403515743,
      "loss": 5.6212,
      "mean_token_accuracy": 0.18674899637699127,
      "num_tokens": 48561202.0,
      "step": 24825
    },
    {
      "entropy": 6.077824354171753,
      "epoch": 2.482031289048833,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043871566058155913,
      "loss": 5.7243,
      "mean_token_accuracy": 0.16668889820575714,
      "num_tokens": 48571283.0,
      "step": 24830
    },
    {
      "entropy": 6.13338418006897,
      "epoch": 2.4825311141100617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043869117669250015,
      "loss": 5.6909,
      "mean_token_accuracy": 0.168531334400177,
      "num_tokens": 48581818.0,
      "step": 24835
    },
    {
      "entropy": 6.0660100936889645,
      "epoch": 2.48303093917129,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0004386666886850132,
      "loss": 5.6487,
      "mean_token_accuracy": 0.1747722804546356,
      "num_tokens": 48591227.0,
      "step": 24840
    },
    {
      "entropy": 6.073725080490112,
      "epoch": 2.4835307642325186,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043864219655971457,
      "loss": 5.6622,
      "mean_token_accuracy": 0.17236957401037217,
      "num_tokens": 48600704.0,
      "step": 24845
    },
    {
      "entropy": 6.117662382125855,
      "epoch": 2.4840305892937473,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004386177003172203,
      "loss": 5.7259,
      "mean_token_accuracy": 0.17661986649036407,
      "num_tokens": 48611964.0,
      "step": 24850
    },
    {
      "entropy": 6.135275793075562,
      "epoch": 2.4845304143549756,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043859319995814686,
      "loss": 5.6641,
      "mean_token_accuracy": 0.17720798403024673,
      "num_tokens": 48621012.0,
      "step": 24855
    },
    {
      "entropy": 6.03787899017334,
      "epoch": 2.4850302394162043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004385686954831105,
      "loss": 5.593,
      "mean_token_accuracy": 0.17947347462177277,
      "num_tokens": 48630242.0,
      "step": 24860
    },
    {
      "entropy": 5.960056495666504,
      "epoch": 2.485530064477433,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043854418689272793,
      "loss": 5.5678,
      "mean_token_accuracy": 0.1831405431032181,
      "num_tokens": 48640535.0,
      "step": 24865
    },
    {
      "entropy": 6.1383435249328615,
      "epoch": 2.4860298895386617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043851967418761574,
      "loss": 5.6652,
      "mean_token_accuracy": 0.18322110921144485,
      "num_tokens": 48650096.0,
      "step": 24870
    },
    {
      "entropy": 6.056563138961792,
      "epoch": 2.48652971459989,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043849515736839057,
      "loss": 5.6464,
      "mean_token_accuracy": 0.1790487602353096,
      "num_tokens": 48660883.0,
      "step": 24875
    },
    {
      "entropy": 6.130646753311157,
      "epoch": 2.4870295396611186,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043847063643566935,
      "loss": 5.6904,
      "mean_token_accuracy": 0.17530573308467864,
      "num_tokens": 48671345.0,
      "step": 24880
    },
    {
      "entropy": 6.1251829147338865,
      "epoch": 2.4875293647223473,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043844611139006893,
      "loss": 5.691,
      "mean_token_accuracy": 0.17284900695085526,
      "num_tokens": 48681296.0,
      "step": 24885
    },
    {
      "entropy": 6.011607503890991,
      "epoch": 2.4880291897835756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043842158223220644,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17612471878528596,
      "num_tokens": 48691885.0,
      "step": 24890
    },
    {
      "entropy": 6.093929290771484,
      "epoch": 2.4885290148448043,
      "grad_norm": 1.203125,
      "learning_rate": 0.000438397048962699,
      "loss": 5.6295,
      "mean_token_accuracy": 0.16994370222091676,
      "num_tokens": 48701325.0,
      "step": 24895
    },
    {
      "entropy": 6.0605415344238285,
      "epoch": 2.489028839906033,
      "grad_norm": 0.9765625,
      "learning_rate": 0.00043837251158216386,
      "loss": 5.689,
      "mean_token_accuracy": 0.17656268626451493,
      "num_tokens": 48712212.0,
      "step": 24900
    },
    {
      "entropy": 6.03923454284668,
      "epoch": 2.4895286649672617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043834797009121836,
      "loss": 5.6033,
      "mean_token_accuracy": 0.1818452551960945,
      "num_tokens": 48721482.0,
      "step": 24905
    },
    {
      "entropy": 6.04849534034729,
      "epoch": 2.49002849002849,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043832342449047994,
      "loss": 5.6699,
      "mean_token_accuracy": 0.1791791945695877,
      "num_tokens": 48731740.0,
      "step": 24910
    },
    {
      "entropy": 6.058159351348877,
      "epoch": 2.4905283150897186,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004382988747805661,
      "loss": 5.605,
      "mean_token_accuracy": 0.17844815105199813,
      "num_tokens": 48740351.0,
      "step": 24915
    },
    {
      "entropy": 6.048389291763305,
      "epoch": 2.4910281401509473,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004382743209620948,
      "loss": 5.5952,
      "mean_token_accuracy": 0.18011426627635957,
      "num_tokens": 48750571.0,
      "step": 24920
    },
    {
      "entropy": 5.994932079315186,
      "epoch": 2.4915279652121756,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004382497630356835,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1846693202853203,
      "num_tokens": 48758740.0,
      "step": 24925
    },
    {
      "entropy": 6.045602750778198,
      "epoch": 2.4920277902734043,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043822520100195015,
      "loss": 5.5899,
      "mean_token_accuracy": 0.18199221044778824,
      "num_tokens": 48768469.0,
      "step": 24930
    },
    {
      "entropy": 6.039363145828247,
      "epoch": 2.492527615334633,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043820063486151267,
      "loss": 5.5378,
      "mean_token_accuracy": 0.18527698665857315,
      "num_tokens": 48778142.0,
      "step": 24935
    },
    {
      "entropy": 5.984345245361328,
      "epoch": 2.4930274403958617,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043817606461498933,
      "loss": 5.5747,
      "mean_token_accuracy": 0.18233220130205155,
      "num_tokens": 48787224.0,
      "step": 24940
    },
    {
      "entropy": 6.118548917770386,
      "epoch": 2.49352726545709,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043815149026299823,
      "loss": 5.7267,
      "mean_token_accuracy": 0.17018845975399016,
      "num_tokens": 48797098.0,
      "step": 24945
    },
    {
      "entropy": 6.061496067047119,
      "epoch": 2.4940270905183186,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004381269118061575,
      "loss": 5.6343,
      "mean_token_accuracy": 0.18324929624795913,
      "num_tokens": 48806764.0,
      "step": 24950
    },
    {
      "entropy": 6.0604229927062985,
      "epoch": 2.4945269155795473,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004381023292450857,
      "loss": 5.6639,
      "mean_token_accuracy": 0.1797009602189064,
      "num_tokens": 48818586.0,
      "step": 24955
    },
    {
      "entropy": 6.06632399559021,
      "epoch": 2.4950267406407756,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004380777425804012,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17895990014076232,
      "num_tokens": 48829229.0,
      "step": 24960
    },
    {
      "entropy": 6.046884441375733,
      "epoch": 2.4955265657020043,
      "grad_norm": 1.671875,
      "learning_rate": 0.00043805315181272275,
      "loss": 5.6135,
      "mean_token_accuracy": 0.18339742571115494,
      "num_tokens": 48838118.0,
      "step": 24965
    },
    {
      "entropy": 6.119746303558349,
      "epoch": 2.496026390763233,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004380285569426689,
      "loss": 5.7337,
      "mean_token_accuracy": 0.16795803606510162,
      "num_tokens": 48847607.0,
      "step": 24970
    },
    {
      "entropy": 6.010901832580567,
      "epoch": 2.4965262158244617,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004380039579708586,
      "loss": 5.5346,
      "mean_token_accuracy": 0.18282132595777512,
      "num_tokens": 48856857.0,
      "step": 24975
    },
    {
      "entropy": 6.070848655700684,
      "epoch": 2.49702604088569,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004379793548979105,
      "loss": 5.6424,
      "mean_token_accuracy": 0.16897690296173096,
      "num_tokens": 48866145.0,
      "step": 24980
    },
    {
      "entropy": 6.027194166183472,
      "epoch": 2.4975258659469186,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004379547477244439,
      "loss": 5.6872,
      "mean_token_accuracy": 0.17480212301015854,
      "num_tokens": 48875941.0,
      "step": 24985
    },
    {
      "entropy": 6.02831883430481,
      "epoch": 2.4980256910081473,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004379301364510777,
      "loss": 5.5507,
      "mean_token_accuracy": 0.18451263010501862,
      "num_tokens": 48885870.0,
      "step": 24990
    },
    {
      "entropy": 6.083881568908692,
      "epoch": 2.4985255160693756,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004379055210784313,
      "loss": 5.7027,
      "mean_token_accuracy": 0.1724776640534401,
      "num_tokens": 48896358.0,
      "step": 24995
    },
    {
      "entropy": 6.001648902893066,
      "epoch": 2.4990253411306043,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004378809016071238,
      "loss": 5.6181,
      "mean_token_accuracy": 0.18031176626682283,
      "num_tokens": 48905397.0,
      "step": 25000
    },
    {
      "entropy": 6.044512462615967,
      "epoch": 2.499525166191833,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004378562780377748,
      "loss": 5.6451,
      "mean_token_accuracy": 0.17894630879163742,
      "num_tokens": 48914385.0,
      "step": 25005
    },
    {
      "entropy": 6.008898019790649,
      "epoch": 2.5000249912530617,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004378316503710037,
      "loss": 5.5719,
      "mean_token_accuracy": 0.17830283790826798,
      "num_tokens": 48923515.0,
      "step": 25010
    },
    {
      "entropy": 5.9953066349029545,
      "epoch": 2.50052481631429,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004378070186074302,
      "loss": 5.634,
      "mean_token_accuracy": 0.18055628687143327,
      "num_tokens": 48933210.0,
      "step": 25015
    },
    {
      "entropy": 6.0515271663665775,
      "epoch": 2.5010246413755186,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043778238274767395,
      "loss": 5.6509,
      "mean_token_accuracy": 0.17209518253803252,
      "num_tokens": 48944292.0,
      "step": 25020
    },
    {
      "entropy": 6.050311803817749,
      "epoch": 2.501524466436747,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004377577427923549,
      "loss": 5.5112,
      "mean_token_accuracy": 0.18575663566589357,
      "num_tokens": 48953883.0,
      "step": 25025
    },
    {
      "entropy": 6.065893030166626,
      "epoch": 2.5020242914979756,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043773309874209294,
      "loss": 5.5693,
      "mean_token_accuracy": 0.1796652555465698,
      "num_tokens": 48963779.0,
      "step": 25030
    },
    {
      "entropy": 6.102937984466553,
      "epoch": 2.5025241165592043,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043770845059750807,
      "loss": 5.6457,
      "mean_token_accuracy": 0.17736321836709976,
      "num_tokens": 48972821.0,
      "step": 25035
    },
    {
      "entropy": 5.9912560939788815,
      "epoch": 2.503023941620433,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043768379835922044,
      "loss": 5.6752,
      "mean_token_accuracy": 0.17770739793777465,
      "num_tokens": 48984288.0,
      "step": 25040
    },
    {
      "entropy": 6.055075931549072,
      "epoch": 2.5035237666816617,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004376591420278503,
      "loss": 5.7156,
      "mean_token_accuracy": 0.16942063421010972,
      "num_tokens": 48994225.0,
      "step": 25045
    },
    {
      "entropy": 6.075097703933716,
      "epoch": 2.50402359174289,
      "grad_norm": 1.125,
      "learning_rate": 0.000437634481604018,
      "loss": 5.5965,
      "mean_token_accuracy": 0.17431594133377076,
      "num_tokens": 49004203.0,
      "step": 25050
    },
    {
      "entropy": 6.028526782989502,
      "epoch": 2.5045234168041186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000437609817088344,
      "loss": 5.6074,
      "mean_token_accuracy": 0.173251773416996,
      "num_tokens": 49014201.0,
      "step": 25055
    },
    {
      "entropy": 6.105017900466919,
      "epoch": 2.505023241865347,
      "grad_norm": 1.28125,
      "learning_rate": 0.00043758514848144886,
      "loss": 5.7172,
      "mean_token_accuracy": 0.17543850988149642,
      "num_tokens": 49022996.0,
      "step": 25060
    },
    {
      "entropy": 6.0873534202575685,
      "epoch": 2.5055230669265756,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004375604757839532,
      "loss": 5.6686,
      "mean_token_accuracy": 0.17018261998891832,
      "num_tokens": 49033040.0,
      "step": 25065
    },
    {
      "entropy": 6.000858497619629,
      "epoch": 2.5060228919878043,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043753579899647783,
      "loss": 5.6568,
      "mean_token_accuracy": 0.17732374519109725,
      "num_tokens": 49043389.0,
      "step": 25070
    },
    {
      "entropy": 6.012435579299927,
      "epoch": 2.506522717049033,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043751111811964353,
      "loss": 5.5275,
      "mean_token_accuracy": 0.18435606658458709,
      "num_tokens": 49052558.0,
      "step": 25075
    },
    {
      "entropy": 6.094233798980713,
      "epoch": 2.5070225421102617,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004374864331540713,
      "loss": 5.7678,
      "mean_token_accuracy": 0.16986480355262756,
      "num_tokens": 49062911.0,
      "step": 25080
    },
    {
      "entropy": 6.049513721466065,
      "epoch": 2.50752236717149,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004374617441003823,
      "loss": 5.6425,
      "mean_token_accuracy": 0.17896757274866104,
      "num_tokens": 49072411.0,
      "step": 25085
    },
    {
      "entropy": 6.044946098327637,
      "epoch": 2.5080221922327186,
      "grad_norm": 1.25,
      "learning_rate": 0.0004374370509591976,
      "loss": 5.6071,
      "mean_token_accuracy": 0.17536772042512894,
      "num_tokens": 49082568.0,
      "step": 25090
    },
    {
      "entropy": 5.867707920074463,
      "epoch": 2.508522017293947,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043741235373113854,
      "loss": 5.4659,
      "mean_token_accuracy": 0.19555341750383376,
      "num_tokens": 49092015.0,
      "step": 25095
    },
    {
      "entropy": 6.036697721481323,
      "epoch": 2.5090218423551756,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043738765241682637,
      "loss": 5.5988,
      "mean_token_accuracy": 0.18216451704502107,
      "num_tokens": 49102877.0,
      "step": 25100
    },
    {
      "entropy": 6.075890159606933,
      "epoch": 2.5095216674164043,
      "grad_norm": 1.25,
      "learning_rate": 0.0004373629470168826,
      "loss": 5.5624,
      "mean_token_accuracy": 0.18062067031860352,
      "num_tokens": 49112227.0,
      "step": 25105
    },
    {
      "entropy": 5.971792364120484,
      "epoch": 2.510021492477633,
      "grad_norm": 1.1875,
      "learning_rate": 0.000437338237531929,
      "loss": 5.5788,
      "mean_token_accuracy": 0.18243009746074676,
      "num_tokens": 49122840.0,
      "step": 25110
    },
    {
      "entropy": 6.00071120262146,
      "epoch": 2.5105213175388617,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004373135239625871,
      "loss": 5.7083,
      "mean_token_accuracy": 0.16833251267671584,
      "num_tokens": 49134067.0,
      "step": 25115
    },
    {
      "entropy": 6.030641412734985,
      "epoch": 2.51102114260009,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043728880630947864,
      "loss": 5.5978,
      "mean_token_accuracy": 0.1821572884917259,
      "num_tokens": 49144176.0,
      "step": 25120
    },
    {
      "entropy": 6.0262586116790775,
      "epoch": 2.5115209676613186,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043726408457322557,
      "loss": 5.6235,
      "mean_token_accuracy": 0.17565979063510895,
      "num_tokens": 49153559.0,
      "step": 25125
    },
    {
      "entropy": 6.071233510971069,
      "epoch": 2.512020792722547,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043723935875445,
      "loss": 5.5805,
      "mean_token_accuracy": 0.18248115330934525,
      "num_tokens": 49164604.0,
      "step": 25130
    },
    {
      "entropy": 6.069117593765259,
      "epoch": 2.5125206177837756,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004372146288537738,
      "loss": 5.6067,
      "mean_token_accuracy": 0.17795564234256744,
      "num_tokens": 49173766.0,
      "step": 25135
    },
    {
      "entropy": 6.006809520721435,
      "epoch": 2.5130204428450043,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043718989487181927,
      "loss": 5.6354,
      "mean_token_accuracy": 0.18981001079082488,
      "num_tokens": 49183244.0,
      "step": 25140
    },
    {
      "entropy": 6.025494766235352,
      "epoch": 2.513520267906233,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004371651568092088,
      "loss": 5.6461,
      "mean_token_accuracy": 0.1768049716949463,
      "num_tokens": 49194256.0,
      "step": 25145
    },
    {
      "entropy": 6.061176586151123,
      "epoch": 2.5140200929674617,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004371404146665647,
      "loss": 5.6793,
      "mean_token_accuracy": 0.17346113920211792,
      "num_tokens": 49204680.0,
      "step": 25150
    },
    {
      "entropy": 6.095919847488403,
      "epoch": 2.51451991802869,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043711566844450934,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17150792479515076,
      "num_tokens": 49213918.0,
      "step": 25155
    },
    {
      "entropy": 6.030353689193726,
      "epoch": 2.5150197430899186,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043709091814366565,
      "loss": 5.5437,
      "mean_token_accuracy": 0.18178821206092835,
      "num_tokens": 49223087.0,
      "step": 25160
    },
    {
      "entropy": 5.9685650825500485,
      "epoch": 2.515519568151147,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043706616376465614,
      "loss": 5.4869,
      "mean_token_accuracy": 0.18629635870456696,
      "num_tokens": 49232072.0,
      "step": 25165
    },
    {
      "entropy": 5.988086795806884,
      "epoch": 2.5160193932123756,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004370414053081036,
      "loss": 5.5659,
      "mean_token_accuracy": 0.18271913081407548,
      "num_tokens": 49241772.0,
      "step": 25170
    },
    {
      "entropy": 6.056657838821411,
      "epoch": 2.5165192182736043,
      "grad_norm": 1.25,
      "learning_rate": 0.000437016642774631,
      "loss": 5.6235,
      "mean_token_accuracy": 0.1739221304655075,
      "num_tokens": 49250605.0,
      "step": 25175
    },
    {
      "entropy": 6.040666770935059,
      "epoch": 2.517019043334833,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043699187616486144,
      "loss": 5.6627,
      "mean_token_accuracy": 0.17961738407611846,
      "num_tokens": 49261104.0,
      "step": 25180
    },
    {
      "entropy": 6.145008134841919,
      "epoch": 2.5175188683960616,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004369671054794178,
      "loss": 5.8294,
      "mean_token_accuracy": 0.16986077427864074,
      "num_tokens": 49271456.0,
      "step": 25185
    },
    {
      "entropy": 6.0523652076721195,
      "epoch": 2.51801869345729,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043694233071892366,
      "loss": 5.5819,
      "mean_token_accuracy": 0.18392395377159118,
      "num_tokens": 49279917.0,
      "step": 25190
    },
    {
      "entropy": 5.972069358825683,
      "epoch": 2.5185185185185186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043691755188400203,
      "loss": 5.6312,
      "mean_token_accuracy": 0.1784198224544525,
      "num_tokens": 49290961.0,
      "step": 25195
    },
    {
      "entropy": 6.065690803527832,
      "epoch": 2.519018343579747,
      "grad_norm": 1.859375,
      "learning_rate": 0.0004368927689752765,
      "loss": 5.6502,
      "mean_token_accuracy": 0.1757780894637108,
      "num_tokens": 49299813.0,
      "step": 25200
    },
    {
      "entropy": 6.076753377914429,
      "epoch": 2.5195181686409756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043686798199337053,
      "loss": 5.6306,
      "mean_token_accuracy": 0.18513433039188384,
      "num_tokens": 49309041.0,
      "step": 25205
    },
    {
      "entropy": 6.012153911590576,
      "epoch": 2.5200179937022043,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043684319093890777,
      "loss": 5.5615,
      "mean_token_accuracy": 0.1849941283464432,
      "num_tokens": 49318647.0,
      "step": 25210
    },
    {
      "entropy": 5.978040981292724,
      "epoch": 2.520517818763433,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043681839581251206,
      "loss": 5.5916,
      "mean_token_accuracy": 0.19100126773118972,
      "num_tokens": 49328410.0,
      "step": 25215
    },
    {
      "entropy": 6.062349796295166,
      "epoch": 2.521017643824661,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043679359661480703,
      "loss": 5.6748,
      "mean_token_accuracy": 0.17783855497837067,
      "num_tokens": 49337766.0,
      "step": 25220
    },
    {
      "entropy": 5.918380832672119,
      "epoch": 2.52151746888589,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043676879334641686,
      "loss": 5.3976,
      "mean_token_accuracy": 0.19796301275491715,
      "num_tokens": 49348512.0,
      "step": 25225
    },
    {
      "entropy": 6.0247735500335695,
      "epoch": 2.5220172939471186,
      "grad_norm": 1.125,
      "learning_rate": 0.0004367439860079654,
      "loss": 5.6354,
      "mean_token_accuracy": 0.18012164831161498,
      "num_tokens": 49359149.0,
      "step": 25230
    },
    {
      "entropy": 6.091040849685669,
      "epoch": 2.522517119008347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043671917460007693,
      "loss": 5.6259,
      "mean_token_accuracy": 0.1822964921593666,
      "num_tokens": 49368083.0,
      "step": 25235
    },
    {
      "entropy": 6.041050291061401,
      "epoch": 2.5230169440695756,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043669435912337557,
      "loss": 5.6425,
      "mean_token_accuracy": 0.1778501570224762,
      "num_tokens": 49378205.0,
      "step": 25240
    },
    {
      "entropy": 6.019012355804444,
      "epoch": 2.5235167691308042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004366695395784858,
      "loss": 5.6751,
      "mean_token_accuracy": 0.1778057411313057,
      "num_tokens": 49388405.0,
      "step": 25245
    },
    {
      "entropy": 5.997595739364624,
      "epoch": 2.524016594192033,
      "grad_norm": 1.140625,
      "learning_rate": 0.000436644715966032,
      "loss": 5.5482,
      "mean_token_accuracy": 0.18680273592472077,
      "num_tokens": 49397966.0,
      "step": 25250
    },
    {
      "entropy": 6.153399705886841,
      "epoch": 2.524516419253261,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004366198882866387,
      "loss": 5.7685,
      "mean_token_accuracy": 0.16822992712259294,
      "num_tokens": 49408542.0,
      "step": 25255
    },
    {
      "entropy": 6.036305332183838,
      "epoch": 2.52501624431449,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043659505654093065,
      "loss": 5.6367,
      "mean_token_accuracy": 0.17789907306432723,
      "num_tokens": 49417499.0,
      "step": 25260
    },
    {
      "entropy": 5.992561960220337,
      "epoch": 2.5255160693757186,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043657022072953253,
      "loss": 5.6278,
      "mean_token_accuracy": 0.18159418404102326,
      "num_tokens": 49426962.0,
      "step": 25265
    },
    {
      "entropy": 5.993246078491211,
      "epoch": 2.526015894436947,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004365453808530693,
      "loss": 5.6104,
      "mean_token_accuracy": 0.17739917039871217,
      "num_tokens": 49437188.0,
      "step": 25270
    },
    {
      "entropy": 6.034706115722656,
      "epoch": 2.5265157194981755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043652053691216574,
      "loss": 5.631,
      "mean_token_accuracy": 0.18205156922340393,
      "num_tokens": 49446939.0,
      "step": 25275
    },
    {
      "entropy": 6.092603158950806,
      "epoch": 2.5270155445594042,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004364956889074471,
      "loss": 5.6408,
      "mean_token_accuracy": 0.1872210055589676,
      "num_tokens": 49458160.0,
      "step": 25280
    },
    {
      "entropy": 5.978191137313843,
      "epoch": 2.527515369620633,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004364708368395385,
      "loss": 5.6507,
      "mean_token_accuracy": 0.1771152511239052,
      "num_tokens": 49468284.0,
      "step": 25285
    },
    {
      "entropy": 6.10216760635376,
      "epoch": 2.528015194681861,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043644598070906516,
      "loss": 5.6645,
      "mean_token_accuracy": 0.18177282959222793,
      "num_tokens": 49479348.0,
      "step": 25290
    },
    {
      "entropy": 6.01823058128357,
      "epoch": 2.52851501974309,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004364211205166525,
      "loss": 5.4856,
      "mean_token_accuracy": 0.19238427877426148,
      "num_tokens": 49488802.0,
      "step": 25295
    },
    {
      "entropy": 6.044792556762696,
      "epoch": 2.5290148448043186,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00043639625626292595,
      "loss": 5.6376,
      "mean_token_accuracy": 0.17598474770784378,
      "num_tokens": 49498682.0,
      "step": 25300
    },
    {
      "entropy": 6.007976961135864,
      "epoch": 2.529514669865547,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004363713879485112,
      "loss": 5.5331,
      "mean_token_accuracy": 0.18759500831365586,
      "num_tokens": 49507530.0,
      "step": 25305
    },
    {
      "entropy": 5.994236707687378,
      "epoch": 2.5300144949267755,
      "grad_norm": 1.125,
      "learning_rate": 0.0004363465155740338,
      "loss": 5.5232,
      "mean_token_accuracy": 0.18273307830095292,
      "num_tokens": 49518435.0,
      "step": 25310
    },
    {
      "entropy": 5.991797018051147,
      "epoch": 2.5305143199880042,
      "grad_norm": 1.328125,
      "learning_rate": 0.0004363216391401196,
      "loss": 5.5688,
      "mean_token_accuracy": 0.1822097510099411,
      "num_tokens": 49528251.0,
      "step": 25315
    },
    {
      "entropy": 6.003828430175782,
      "epoch": 2.531014145049233,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004362967586473945,
      "loss": 5.5654,
      "mean_token_accuracy": 0.18492062091827394,
      "num_tokens": 49536884.0,
      "step": 25320
    },
    {
      "entropy": 5.95617413520813,
      "epoch": 2.531513970110461,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004362718740964844,
      "loss": 5.5373,
      "mean_token_accuracy": 0.18656033873558045,
      "num_tokens": 49547665.0,
      "step": 25325
    },
    {
      "entropy": 6.114248847961425,
      "epoch": 2.53201379517169,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004362469854880155,
      "loss": 5.7334,
      "mean_token_accuracy": 0.1700616106390953,
      "num_tokens": 49557413.0,
      "step": 25330
    },
    {
      "entropy": 6.095935344696045,
      "epoch": 2.5325136202329186,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004362220928226139,
      "loss": 5.649,
      "mean_token_accuracy": 0.17813128232955933,
      "num_tokens": 49566305.0,
      "step": 25335
    },
    {
      "entropy": 6.034753036499024,
      "epoch": 2.533013445294147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043619719610090595,
      "loss": 5.5649,
      "mean_token_accuracy": 0.17807136476039886,
      "num_tokens": 49576020.0,
      "step": 25340
    },
    {
      "entropy": 5.994367885589599,
      "epoch": 2.5335132703553755,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000436172295323518,
      "loss": 5.61,
      "mean_token_accuracy": 0.17901006191968918,
      "num_tokens": 49586564.0,
      "step": 25345
    },
    {
      "entropy": 6.015358400344849,
      "epoch": 2.5340130954166042,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043614739049107666,
      "loss": 5.6618,
      "mean_token_accuracy": 0.17742062509059905,
      "num_tokens": 49595874.0,
      "step": 25350
    },
    {
      "entropy": 6.009614944458008,
      "epoch": 2.534512920477833,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043612248160420833,
      "loss": 5.588,
      "mean_token_accuracy": 0.18059496730566024,
      "num_tokens": 49605975.0,
      "step": 25355
    },
    {
      "entropy": 6.009717226028442,
      "epoch": 2.535012745539061,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00043609756866353985,
      "loss": 5.5102,
      "mean_token_accuracy": 0.18871124386787413,
      "num_tokens": 49615187.0,
      "step": 25360
    },
    {
      "entropy": 6.029205703735352,
      "epoch": 2.53551257060029,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000436072651669698,
      "loss": 5.6382,
      "mean_token_accuracy": 0.1747574746608734,
      "num_tokens": 49624660.0,
      "step": 25365
    },
    {
      "entropy": 6.011563634872436,
      "epoch": 2.5360123956615186,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004360477306233097,
      "loss": 5.556,
      "mean_token_accuracy": 0.1844378799200058,
      "num_tokens": 49635450.0,
      "step": 25370
    },
    {
      "entropy": 6.05139536857605,
      "epoch": 2.536512220722747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043602280552500195,
      "loss": 5.6691,
      "mean_token_accuracy": 0.16905487477779388,
      "num_tokens": 49646466.0,
      "step": 25375
    },
    {
      "entropy": 6.113451194763184,
      "epoch": 2.5370120457839755,
      "grad_norm": 1.34375,
      "learning_rate": 0.00043599787637540186,
      "loss": 5.6956,
      "mean_token_accuracy": 0.17256094217300416,
      "num_tokens": 49655440.0,
      "step": 25380
    },
    {
      "entropy": 6.054423809051514,
      "epoch": 2.5375118708452042,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043597294317513654,
      "loss": 5.5613,
      "mean_token_accuracy": 0.18028899431228637,
      "num_tokens": 49664072.0,
      "step": 25385
    },
    {
      "entropy": 5.994001293182373,
      "epoch": 2.538011695906433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043594800592483345,
      "loss": 5.6182,
      "mean_token_accuracy": 0.17664668411016465,
      "num_tokens": 49673439.0,
      "step": 25390
    },
    {
      "entropy": 5.980083274841308,
      "epoch": 2.538511520967661,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043592306462511993,
      "loss": 5.6216,
      "mean_token_accuracy": 0.18011473268270492,
      "num_tokens": 49682307.0,
      "step": 25395
    },
    {
      "entropy": 5.964274120330811,
      "epoch": 2.53901134602889,
      "grad_norm": 1.328125,
      "learning_rate": 0.00043589811927662356,
      "loss": 5.4348,
      "mean_token_accuracy": 0.20402052253484726,
      "num_tokens": 49692146.0,
      "step": 25400
    },
    {
      "entropy": 6.066128540039062,
      "epoch": 2.5395111710901186,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043587316987997183,
      "loss": 5.6341,
      "mean_token_accuracy": 0.17935866862535477,
      "num_tokens": 49701465.0,
      "step": 25405
    },
    {
      "entropy": 6.024795246124268,
      "epoch": 2.540010996151347,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043584821643579265,
      "loss": 5.6271,
      "mean_token_accuracy": 0.18314309418201447,
      "num_tokens": 49711845.0,
      "step": 25410
    },
    {
      "entropy": 6.092045021057129,
      "epoch": 2.5405108212125755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043582325894471366,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1764240488409996,
      "num_tokens": 49721078.0,
      "step": 25415
    },
    {
      "entropy": 6.028663539886475,
      "epoch": 2.5410106462738042,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004357982974073629,
      "loss": 5.6284,
      "mean_token_accuracy": 0.1777789130806923,
      "num_tokens": 49730987.0,
      "step": 25420
    },
    {
      "entropy": 5.980686998367309,
      "epoch": 2.541510471335033,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004357733318243683,
      "loss": 5.4692,
      "mean_token_accuracy": 0.18540041148662567,
      "num_tokens": 49741688.0,
      "step": 25425
    },
    {
      "entropy": 6.100413751602173,
      "epoch": 2.542010296396261,
      "grad_norm": 1.25,
      "learning_rate": 0.00043574836219635805,
      "loss": 5.6791,
      "mean_token_accuracy": 0.17465580850839615,
      "num_tokens": 49750835.0,
      "step": 25430
    },
    {
      "entropy": 6.015394973754883,
      "epoch": 2.54251012145749,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043572338852396035,
      "loss": 5.6489,
      "mean_token_accuracy": 0.17566115111112596,
      "num_tokens": 49760654.0,
      "step": 25435
    },
    {
      "entropy": 6.024277639389038,
      "epoch": 2.5430099465187186,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043569841080780367,
      "loss": 5.5931,
      "mean_token_accuracy": 0.18078823685646056,
      "num_tokens": 49771367.0,
      "step": 25440
    },
    {
      "entropy": 6.050155544281006,
      "epoch": 2.543509771579947,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043567342904851625,
      "loss": 5.5926,
      "mean_token_accuracy": 0.18332703709602355,
      "num_tokens": 49781302.0,
      "step": 25445
    },
    {
      "entropy": 6.02756781578064,
      "epoch": 2.5440095966411755,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004356484432467267,
      "loss": 5.6845,
      "mean_token_accuracy": 0.1789763554930687,
      "num_tokens": 49790659.0,
      "step": 25450
    },
    {
      "entropy": 6.087275266647339,
      "epoch": 2.5445094217024042,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043562345340306363,
      "loss": 5.6837,
      "mean_token_accuracy": 0.18042251095175743,
      "num_tokens": 49800285.0,
      "step": 25455
    },
    {
      "entropy": 6.01382303237915,
      "epoch": 2.545009246763633,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004355984595181559,
      "loss": 5.6243,
      "mean_token_accuracy": 0.17223508059978485,
      "num_tokens": 49810810.0,
      "step": 25460
    },
    {
      "entropy": 6.042003631591797,
      "epoch": 2.545509071824861,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004355734615926321,
      "loss": 5.6345,
      "mean_token_accuracy": 0.17739037722349166,
      "num_tokens": 49820728.0,
      "step": 25465
    },
    {
      "entropy": 6.147556495666504,
      "epoch": 2.54600889688609,
      "grad_norm": 1.421875,
      "learning_rate": 0.0004355484596271214,
      "loss": 5.6346,
      "mean_token_accuracy": 0.17091596573591233,
      "num_tokens": 49830477.0,
      "step": 25470
    },
    {
      "entropy": 6.057489967346191,
      "epoch": 2.5465087219473186,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004355234536222528,
      "loss": 5.677,
      "mean_token_accuracy": 0.180227130651474,
      "num_tokens": 49839154.0,
      "step": 25475
    },
    {
      "entropy": 5.995430994033813,
      "epoch": 2.547008547008547,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004354984435786554,
      "loss": 5.6145,
      "mean_token_accuracy": 0.18437059670686723,
      "num_tokens": 49848894.0,
      "step": 25480
    },
    {
      "entropy": 6.053682088851929,
      "epoch": 2.5475083720697755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004354734294969585,
      "loss": 5.6448,
      "mean_token_accuracy": 0.17560249119997023,
      "num_tokens": 49858730.0,
      "step": 25485
    },
    {
      "entropy": 6.0430680274963375,
      "epoch": 2.5480081971310042,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00043544841137779127,
      "loss": 5.6625,
      "mean_token_accuracy": 0.17362688183784486,
      "num_tokens": 49868553.0,
      "step": 25490
    },
    {
      "entropy": 6.025748443603516,
      "epoch": 2.548508022192233,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043542338922178343,
      "loss": 5.613,
      "mean_token_accuracy": 0.18889534920454026,
      "num_tokens": 49878053.0,
      "step": 25495
    },
    {
      "entropy": 6.099374437332154,
      "epoch": 2.549007847253461,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043539836302956434,
      "loss": 5.5851,
      "mean_token_accuracy": 0.17879965305328369,
      "num_tokens": 49888693.0,
      "step": 25500
    },
    {
      "entropy": 6.131494474411011,
      "epoch": 2.54950767231469,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004353733328017637,
      "loss": 5.6144,
      "mean_token_accuracy": 0.17619494944810868,
      "num_tokens": 49899094.0,
      "step": 25505
    },
    {
      "entropy": 6.133269166946411,
      "epoch": 2.5500074973759186,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004353482985390113,
      "loss": 5.673,
      "mean_token_accuracy": 0.18462652266025542,
      "num_tokens": 49909341.0,
      "step": 25510
    },
    {
      "entropy": 5.968445491790772,
      "epoch": 2.550507322437147,
      "grad_norm": 1.1796875,
      "learning_rate": 0.000435323260241937,
      "loss": 5.6024,
      "mean_token_accuracy": 0.1810848668217659,
      "num_tokens": 49919602.0,
      "step": 25515
    },
    {
      "entropy": 5.965293502807617,
      "epoch": 2.5510071474983755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043529821791117064,
      "loss": 5.6219,
      "mean_token_accuracy": 0.17873027920722961,
      "num_tokens": 49928958.0,
      "step": 25520
    },
    {
      "entropy": 5.9788017749786375,
      "epoch": 2.5515069725596042,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004352731715473425,
      "loss": 5.5856,
      "mean_token_accuracy": 0.18194237500429153,
      "num_tokens": 49938988.0,
      "step": 25525
    },
    {
      "entropy": 6.046966457366944,
      "epoch": 2.552006797620833,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043524812115108244,
      "loss": 5.6952,
      "mean_token_accuracy": 0.17284610718488694,
      "num_tokens": 49949399.0,
      "step": 25530
    },
    {
      "entropy": 6.1115538597106935,
      "epoch": 2.552506622682061,
      "grad_norm": 1.015625,
      "learning_rate": 0.00043522306672302103,
      "loss": 5.5194,
      "mean_token_accuracy": 0.1862720087170601,
      "num_tokens": 49958666.0,
      "step": 25535
    },
    {
      "entropy": 5.955815744400025,
      "epoch": 2.55300644774329,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004351980082637884,
      "loss": 5.5603,
      "mean_token_accuracy": 0.17760927379131317,
      "num_tokens": 49967739.0,
      "step": 25540
    },
    {
      "entropy": 5.980245542526245,
      "epoch": 2.5535062728045186,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043517294577401504,
      "loss": 5.6348,
      "mean_token_accuracy": 0.17998246103525162,
      "num_tokens": 49976980.0,
      "step": 25545
    },
    {
      "entropy": 6.066369438171387,
      "epoch": 2.554006097865747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043514787925433176,
      "loss": 5.6491,
      "mean_token_accuracy": 0.17950748652219772,
      "num_tokens": 49986963.0,
      "step": 25550
    },
    {
      "entropy": 6.069003391265869,
      "epoch": 2.5545059229269755,
      "grad_norm": 1.34375,
      "learning_rate": 0.0004351228087053689,
      "loss": 5.6882,
      "mean_token_accuracy": 0.17043259590864182,
      "num_tokens": 49996009.0,
      "step": 25555
    },
    {
      "entropy": 6.056478214263916,
      "epoch": 2.5550057479882042,
      "grad_norm": 1.125,
      "learning_rate": 0.00043509773412775746,
      "loss": 5.6267,
      "mean_token_accuracy": 0.18063177168369293,
      "num_tokens": 50006346.0,
      "step": 25560
    },
    {
      "entropy": 5.996723461151123,
      "epoch": 2.555505573049433,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004350726555221282,
      "loss": 5.6152,
      "mean_token_accuracy": 0.184835746884346,
      "num_tokens": 50016512.0,
      "step": 25565
    },
    {
      "entropy": 5.977815723419189,
      "epoch": 2.556005398110661,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004350475728891121,
      "loss": 5.5994,
      "mean_token_accuracy": 0.17891016304492952,
      "num_tokens": 50027012.0,
      "step": 25570
    },
    {
      "entropy": 5.877023506164551,
      "epoch": 2.55650522317189,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004350224862293403,
      "loss": 5.3546,
      "mean_token_accuracy": 0.19795173555612564,
      "num_tokens": 50036869.0,
      "step": 25575
    },
    {
      "entropy": 6.014264249801636,
      "epoch": 2.557005048233118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043499739554344393,
      "loss": 5.5546,
      "mean_token_accuracy": 0.18941827565431596,
      "num_tokens": 50046186.0,
      "step": 25580
    },
    {
      "entropy": 6.00124077796936,
      "epoch": 2.557504873294347,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043497230083205425,
      "loss": 5.5946,
      "mean_token_accuracy": 0.18423697650432586,
      "num_tokens": 50056915.0,
      "step": 25585
    },
    {
      "entropy": 5.944492197036743,
      "epoch": 2.5580046983555755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004349472020958026,
      "loss": 5.5413,
      "mean_token_accuracy": 0.18634007275104522,
      "num_tokens": 50066143.0,
      "step": 25590
    },
    {
      "entropy": 5.989619445800781,
      "epoch": 2.5585045234168042,
      "grad_norm": 1.125,
      "learning_rate": 0.00043492209933532054,
      "loss": 5.603,
      "mean_token_accuracy": 0.17646386176347734,
      "num_tokens": 50077031.0,
      "step": 25595
    },
    {
      "entropy": 6.089145040512085,
      "epoch": 2.559004348478033,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043489699255123954,
      "loss": 5.7378,
      "mean_token_accuracy": 0.16738774478435517,
      "num_tokens": 50087545.0,
      "step": 25600
    },
    {
      "entropy": 6.049642419815063,
      "epoch": 2.559504173539261,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004348718817441914,
      "loss": 5.5585,
      "mean_token_accuracy": 0.1843509331345558,
      "num_tokens": 50096427.0,
      "step": 25605
    },
    {
      "entropy": 6.016114950180054,
      "epoch": 2.56000399860049,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004348467669148078,
      "loss": 5.5683,
      "mean_token_accuracy": 0.18083107471466064,
      "num_tokens": 50105382.0,
      "step": 25610
    },
    {
      "entropy": 6.083285808563232,
      "epoch": 2.560503823661718,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00043482164806372075,
      "loss": 5.7319,
      "mean_token_accuracy": 0.16666824072599412,
      "num_tokens": 50116348.0,
      "step": 25615
    },
    {
      "entropy": 6.08129768371582,
      "epoch": 2.561003648722947,
      "grad_norm": 1.125,
      "learning_rate": 0.0004347965251915622,
      "loss": 5.6009,
      "mean_token_accuracy": 0.17826835811138153,
      "num_tokens": 50126081.0,
      "step": 25620
    },
    {
      "entropy": 6.034533262252808,
      "epoch": 2.5615034737841755,
      "grad_norm": 1.234375,
      "learning_rate": 0.00043477139829896415,
      "loss": 5.5802,
      "mean_token_accuracy": 0.17995201647281647,
      "num_tokens": 50135719.0,
      "step": 25625
    },
    {
      "entropy": 5.99252610206604,
      "epoch": 2.5620032988454042,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043474626738655876,
      "loss": 5.5793,
      "mean_token_accuracy": 0.1831786721944809,
      "num_tokens": 50146201.0,
      "step": 25630
    },
    {
      "entropy": 5.957275390625,
      "epoch": 2.562503123906633,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004347211324549785,
      "loss": 5.4856,
      "mean_token_accuracy": 0.19014020264148712,
      "num_tokens": 50156352.0,
      "step": 25635
    },
    {
      "entropy": 6.0669862747192385,
      "epoch": 2.563002948967861,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043469599350485556,
      "loss": 5.6055,
      "mean_token_accuracy": 0.17713395357131959,
      "num_tokens": 50165518.0,
      "step": 25640
    },
    {
      "entropy": 5.964083051681518,
      "epoch": 2.56350277402909,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004346708505368225,
      "loss": 5.5269,
      "mean_token_accuracy": 0.1950780063867569,
      "num_tokens": 50173656.0,
      "step": 25645
    },
    {
      "entropy": 6.001626062393188,
      "epoch": 2.564002599090318,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004346457035515121,
      "loss": 5.5971,
      "mean_token_accuracy": 0.18550993800163268,
      "num_tokens": 50183024.0,
      "step": 25650
    },
    {
      "entropy": 6.0426092624664305,
      "epoch": 2.564502424151547,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043462055254955666,
      "loss": 5.7039,
      "mean_token_accuracy": 0.1721629187464714,
      "num_tokens": 50194192.0,
      "step": 25655
    },
    {
      "entropy": 6.059206771850586,
      "epoch": 2.5650022492127755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004345953975315893,
      "loss": 5.6859,
      "mean_token_accuracy": 0.17969612330198287,
      "num_tokens": 50203655.0,
      "step": 25660
    },
    {
      "entropy": 6.127799797058105,
      "epoch": 2.5655020742740042,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004345702384982429,
      "loss": 5.6889,
      "mean_token_accuracy": 0.1773183375597,
      "num_tokens": 50213400.0,
      "step": 25665
    },
    {
      "entropy": 6.032224178314209,
      "epoch": 2.566001899335233,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004345450754501502,
      "loss": 5.632,
      "mean_token_accuracy": 0.17818180918693544,
      "num_tokens": 50224113.0,
      "step": 25670
    },
    {
      "entropy": 6.0405285358428955,
      "epoch": 2.566501724396461,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043451990838794453,
      "loss": 5.5842,
      "mean_token_accuracy": 0.1832118660211563,
      "num_tokens": 50232990.0,
      "step": 25675
    },
    {
      "entropy": 6.078109455108643,
      "epoch": 2.56700154945769,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000434494737312259,
      "loss": 5.5858,
      "mean_token_accuracy": 0.17761754840612412,
      "num_tokens": 50242573.0,
      "step": 25680
    },
    {
      "entropy": 6.0545580863952635,
      "epoch": 2.567501374518918,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043446956222372693,
      "loss": 5.6643,
      "mean_token_accuracy": 0.17474788576364517,
      "num_tokens": 50252654.0,
      "step": 25685
    },
    {
      "entropy": 6.002556848526001,
      "epoch": 2.568001199580147,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004344443831229816,
      "loss": 5.5611,
      "mean_token_accuracy": 0.18188054859638214,
      "num_tokens": 50262654.0,
      "step": 25690
    },
    {
      "entropy": 6.091816329956055,
      "epoch": 2.5685010246413755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043441920001065676,
      "loss": 5.6531,
      "mean_token_accuracy": 0.17601924240589142,
      "num_tokens": 50272653.0,
      "step": 25695
    },
    {
      "entropy": 6.0804407596588135,
      "epoch": 2.5690008497026042,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004343940128873858,
      "loss": 5.7593,
      "mean_token_accuracy": 0.17241075038909912,
      "num_tokens": 50283157.0,
      "step": 25700
    },
    {
      "entropy": 5.995005989074707,
      "epoch": 2.569500674763833,
      "grad_norm": 1.25,
      "learning_rate": 0.00043436882175380256,
      "loss": 5.6333,
      "mean_token_accuracy": 0.18329019248485565,
      "num_tokens": 50292224.0,
      "step": 25705
    },
    {
      "entropy": 6.083192634582519,
      "epoch": 2.570000499825061,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004343436266105407,
      "loss": 5.6673,
      "mean_token_accuracy": 0.17343535721302034,
      "num_tokens": 50302083.0,
      "step": 25710
    },
    {
      "entropy": 5.998964118957519,
      "epoch": 2.57050032488629,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004343184274582343,
      "loss": 5.5854,
      "mean_token_accuracy": 0.18362970650196075,
      "num_tokens": 50312001.0,
      "step": 25715
    },
    {
      "entropy": 5.999917888641358,
      "epoch": 2.571000149947518,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004342932242975172,
      "loss": 5.5496,
      "mean_token_accuracy": 0.18245129734277726,
      "num_tokens": 50322652.0,
      "step": 25720
    },
    {
      "entropy": 6.077789354324341,
      "epoch": 2.571499975008747,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004342680171290235,
      "loss": 5.5864,
      "mean_token_accuracy": 0.18169565349817277,
      "num_tokens": 50332200.0,
      "step": 25725
    },
    {
      "entropy": 5.983381891250611,
      "epoch": 2.5719998000699755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004342428059533875,
      "loss": 5.5745,
      "mean_token_accuracy": 0.18744618743658065,
      "num_tokens": 50342567.0,
      "step": 25730
    },
    {
      "entropy": 6.007059955596924,
      "epoch": 2.5724996251312042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004342175907712435,
      "loss": 5.6038,
      "mean_token_accuracy": 0.18887903541326523,
      "num_tokens": 50352009.0,
      "step": 25735
    },
    {
      "entropy": 6.005417585372925,
      "epoch": 2.572999450192433,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043419237158322596,
      "loss": 5.5962,
      "mean_token_accuracy": 0.1871936872601509,
      "num_tokens": 50362412.0,
      "step": 25740
    },
    {
      "entropy": 6.0767909526824955,
      "epoch": 2.573499275253661,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043416714838996915,
      "loss": 5.6514,
      "mean_token_accuracy": 0.17747277170419692,
      "num_tokens": 50372336.0,
      "step": 25745
    },
    {
      "entropy": 5.979908227920532,
      "epoch": 2.57399910031489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000434141921192108,
      "loss": 5.501,
      "mean_token_accuracy": 0.18922167420387268,
      "num_tokens": 50381612.0,
      "step": 25750
    },
    {
      "entropy": 6.00229549407959,
      "epoch": 2.574498925376118,
      "grad_norm": 1.15625,
      "learning_rate": 0.000434116689990277,
      "loss": 5.5812,
      "mean_token_accuracy": 0.1823129028081894,
      "num_tokens": 50392073.0,
      "step": 25755
    },
    {
      "entropy": 5.986251735687256,
      "epoch": 2.574998750437347,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043409145478511105,
      "loss": 5.5757,
      "mean_token_accuracy": 0.18765933960676193,
      "num_tokens": 50401306.0,
      "step": 25760
    },
    {
      "entropy": 5.965819692611694,
      "epoch": 2.5754985754985755,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00043406621557724495,
      "loss": 5.6073,
      "mean_token_accuracy": 0.17393139749765396,
      "num_tokens": 50410592.0,
      "step": 25765
    },
    {
      "entropy": 6.057042217254638,
      "epoch": 2.5759984005598042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043404097236731396,
      "loss": 5.6218,
      "mean_token_accuracy": 0.18104489296674728,
      "num_tokens": 50420863.0,
      "step": 25770
    },
    {
      "entropy": 6.063992595672607,
      "epoch": 2.5764982256210325,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004340157251559529,
      "loss": 5.6504,
      "mean_token_accuracy": 0.17818641364574433,
      "num_tokens": 50431321.0,
      "step": 25775
    },
    {
      "entropy": 6.017960023880005,
      "epoch": 2.576998050682261,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043399047394379717,
      "loss": 5.5858,
      "mean_token_accuracy": 0.17803821563720704,
      "num_tokens": 50442223.0,
      "step": 25780
    },
    {
      "entropy": 5.9525329113006595,
      "epoch": 2.57749787574349,
      "grad_norm": 1.125,
      "learning_rate": 0.00043396521873148195,
      "loss": 5.539,
      "mean_token_accuracy": 0.1847424566745758,
      "num_tokens": 50451082.0,
      "step": 25785
    },
    {
      "entropy": 6.034868860244751,
      "epoch": 2.577997700804718,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043393995951964277,
      "loss": 5.5836,
      "mean_token_accuracy": 0.18509110808372498,
      "num_tokens": 50461061.0,
      "step": 25790
    },
    {
      "entropy": 5.973652648925781,
      "epoch": 2.578497525865947,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043391469630891513,
      "loss": 5.5289,
      "mean_token_accuracy": 0.19015580117702485,
      "num_tokens": 50470098.0,
      "step": 25795
    },
    {
      "entropy": 6.042131328582764,
      "epoch": 2.5789973509271755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004338894290999346,
      "loss": 5.5752,
      "mean_token_accuracy": 0.17874244302511216,
      "num_tokens": 50479669.0,
      "step": 25800
    },
    {
      "entropy": 5.993898391723633,
      "epoch": 2.5794971759884042,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00043386415789333687,
      "loss": 5.5459,
      "mean_token_accuracy": 0.18553537875413895,
      "num_tokens": 50488608.0,
      "step": 25805
    },
    {
      "entropy": 5.991174936294556,
      "epoch": 2.5799970010496325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004338388826897579,
      "loss": 5.6516,
      "mean_token_accuracy": 0.17921123951673507,
      "num_tokens": 50497486.0,
      "step": 25810
    },
    {
      "entropy": 6.036907720565796,
      "epoch": 2.580496826110861,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004338136034898335,
      "loss": 5.6167,
      "mean_token_accuracy": 0.1798812374472618,
      "num_tokens": 50506034.0,
      "step": 25815
    },
    {
      "entropy": 6.021275377273559,
      "epoch": 2.58099665117209,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004337883202941997,
      "loss": 5.6838,
      "mean_token_accuracy": 0.17548028379678726,
      "num_tokens": 50516617.0,
      "step": 25820
    },
    {
      "entropy": 6.110475683212281,
      "epoch": 2.581496476233318,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004337630331034926,
      "loss": 5.6165,
      "mean_token_accuracy": 0.17462630420923234,
      "num_tokens": 50526287.0,
      "step": 25825
    },
    {
      "entropy": 6.060081720352173,
      "epoch": 2.581996301294547,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043373774191834844,
      "loss": 5.6422,
      "mean_token_accuracy": 0.17585487961769103,
      "num_tokens": 50535803.0,
      "step": 25830
    },
    {
      "entropy": 6.095237684249878,
      "epoch": 2.5824961263557755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043371244673940355,
      "loss": 5.7906,
      "mean_token_accuracy": 0.1663959190249443,
      "num_tokens": 50545386.0,
      "step": 25835
    },
    {
      "entropy": 6.090646696090698,
      "epoch": 2.582995951417004,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004336871475672943,
      "loss": 5.6304,
      "mean_token_accuracy": 0.17919568568468094,
      "num_tokens": 50555299.0,
      "step": 25840
    },
    {
      "entropy": 5.993093585968017,
      "epoch": 2.5834957764782325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004336618444026573,
      "loss": 5.521,
      "mean_token_accuracy": 0.1871308743953705,
      "num_tokens": 50564530.0,
      "step": 25845
    },
    {
      "entropy": 5.999269580841064,
      "epoch": 2.583995601539461,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043363653724612903,
      "loss": 5.6617,
      "mean_token_accuracy": 0.17446561455726622,
      "num_tokens": 50574783.0,
      "step": 25850
    },
    {
      "entropy": 5.925566959381103,
      "epoch": 2.58449542660069,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043361122609834646,
      "loss": 5.4782,
      "mean_token_accuracy": 0.19122677445411682,
      "num_tokens": 50584764.0,
      "step": 25855
    },
    {
      "entropy": 5.979708337783814,
      "epoch": 2.584995251661918,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004335859109599462,
      "loss": 5.5078,
      "mean_token_accuracy": 0.1893474504351616,
      "num_tokens": 50593547.0,
      "step": 25860
    },
    {
      "entropy": 5.987077474594116,
      "epoch": 2.585495076723147,
      "grad_norm": 1.390625,
      "learning_rate": 0.00043356059183156514,
      "loss": 5.5195,
      "mean_token_accuracy": 0.19342687875032424,
      "num_tokens": 50602749.0,
      "step": 25865
    },
    {
      "entropy": 6.047565603256226,
      "epoch": 2.5859949017843755,
      "grad_norm": 1.25,
      "learning_rate": 0.00043353526871384054,
      "loss": 5.5894,
      "mean_token_accuracy": 0.18508001714944838,
      "num_tokens": 50612900.0,
      "step": 25870
    },
    {
      "entropy": 6.097301959991455,
      "epoch": 2.586494726845604,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004335099416074092,
      "loss": 5.6975,
      "mean_token_accuracy": 0.16681978851556778,
      "num_tokens": 50623131.0,
      "step": 25875
    },
    {
      "entropy": 6.010357856750488,
      "epoch": 2.5869945519068325,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004334846105129086,
      "loss": 5.547,
      "mean_token_accuracy": 0.182301926612854,
      "num_tokens": 50632312.0,
      "step": 25880
    },
    {
      "entropy": 6.083261728286743,
      "epoch": 2.587494376968061,
      "grad_norm": 1.109375,
      "learning_rate": 0.000433459275430976,
      "loss": 5.7865,
      "mean_token_accuracy": 0.16547089591622352,
      "num_tokens": 50643264.0,
      "step": 25885
    },
    {
      "entropy": 6.019079351425171,
      "epoch": 2.58799420202929,
      "grad_norm": 1.3359375,
      "learning_rate": 0.0004334339363622488,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17871276885271073,
      "num_tokens": 50652525.0,
      "step": 25890
    },
    {
      "entropy": 6.059764337539673,
      "epoch": 2.588494027090518,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043340859330736454,
      "loss": 5.6376,
      "mean_token_accuracy": 0.18194933235645294,
      "num_tokens": 50661509.0,
      "step": 25895
    },
    {
      "entropy": 6.038602876663208,
      "epoch": 2.588993852151747,
      "grad_norm": 1.625,
      "learning_rate": 0.00043338324626696084,
      "loss": 5.598,
      "mean_token_accuracy": 0.18010858595371246,
      "num_tokens": 50672646.0,
      "step": 25900
    },
    {
      "entropy": 6.009282064437866,
      "epoch": 2.5894936772129755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004333578952416753,
      "loss": 5.5121,
      "mean_token_accuracy": 0.19434583336114883,
      "num_tokens": 50681967.0,
      "step": 25905
    },
    {
      "entropy": 6.056621170043945,
      "epoch": 2.589993502274204,
      "grad_norm": 1.34375,
      "learning_rate": 0.000433332540232146,
      "loss": 5.6019,
      "mean_token_accuracy": 0.18699777126312256,
      "num_tokens": 50690630.0,
      "step": 25910
    },
    {
      "entropy": 5.9903041362762455,
      "epoch": 2.5904933273354325,
      "grad_norm": 0.93359375,
      "learning_rate": 0.00043330718123901074,
      "loss": 5.6085,
      "mean_token_accuracy": 0.1830119550228119,
      "num_tokens": 50700618.0,
      "step": 25915
    },
    {
      "entropy": 6.0103222846984865,
      "epoch": 2.590993152396661,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043328181826290745,
      "loss": 5.5528,
      "mean_token_accuracy": 0.18427708446979524,
      "num_tokens": 50710825.0,
      "step": 25920
    },
    {
      "entropy": 6.022514009475708,
      "epoch": 2.59149297745789,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004332564513044744,
      "loss": 5.5875,
      "mean_token_accuracy": 0.1825055092573166,
      "num_tokens": 50721869.0,
      "step": 25925
    },
    {
      "entropy": 6.02656397819519,
      "epoch": 2.591992802519118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043323108036434974,
      "loss": 5.5233,
      "mean_token_accuracy": 0.187306809425354,
      "num_tokens": 50732088.0,
      "step": 25930
    },
    {
      "entropy": 5.985320806503296,
      "epoch": 2.592492627580347,
      "grad_norm": 1.125,
      "learning_rate": 0.0004332057054431718,
      "loss": 5.632,
      "mean_token_accuracy": 0.1801345616579056,
      "num_tokens": 50743419.0,
      "step": 25935
    },
    {
      "entropy": 6.015631246566772,
      "epoch": 2.5929924526415755,
      "grad_norm": 1.234375,
      "learning_rate": 0.000433180326541579,
      "loss": 5.5802,
      "mean_token_accuracy": 0.18342125564813613,
      "num_tokens": 50754371.0,
      "step": 25940
    },
    {
      "entropy": 6.061701250076294,
      "epoch": 2.593492277702804,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004331549436602099,
      "loss": 5.5918,
      "mean_token_accuracy": 0.1836387351155281,
      "num_tokens": 50763737.0,
      "step": 25945
    },
    {
      "entropy": 6.05001425743103,
      "epoch": 2.5939921027640325,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00043312955679970317,
      "loss": 5.687,
      "mean_token_accuracy": 0.1790593907237053,
      "num_tokens": 50774571.0,
      "step": 25950
    },
    {
      "entropy": 6.1326416492462155,
      "epoch": 2.594491927825261,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00043310416596069746,
      "loss": 5.6333,
      "mean_token_accuracy": 0.1851600706577301,
      "num_tokens": 50783232.0,
      "step": 25955
    },
    {
      "entropy": 6.034183073043823,
      "epoch": 2.59499175288649,
      "grad_norm": 1.125,
      "learning_rate": 0.00043307877114383167,
      "loss": 5.6544,
      "mean_token_accuracy": 0.18526309728622437,
      "num_tokens": 50792238.0,
      "step": 25960
    },
    {
      "entropy": 5.931816339492798,
      "epoch": 2.595491577947718,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004330533723497445,
      "loss": 5.5113,
      "mean_token_accuracy": 0.18919307589530945,
      "num_tokens": 50802012.0,
      "step": 25965
    },
    {
      "entropy": 6.018684577941895,
      "epoch": 2.595991403008947,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043302796957907523,
      "loss": 5.5778,
      "mean_token_accuracy": 0.1899635910987854,
      "num_tokens": 50811938.0,
      "step": 25970
    },
    {
      "entropy": 6.001771259307861,
      "epoch": 2.5964912280701755,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043300256283246297,
      "loss": 5.5641,
      "mean_token_accuracy": 0.18394985049962997,
      "num_tokens": 50820702.0,
      "step": 25975
    },
    {
      "entropy": 6.067397785186768,
      "epoch": 2.596991053131404,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043297715211054684,
      "loss": 5.713,
      "mean_token_accuracy": 0.17036787569522857,
      "num_tokens": 50830291.0,
      "step": 25980
    },
    {
      "entropy": 5.9361851692199705,
      "epoch": 2.5974908781926325,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043295173741396626,
      "loss": 5.5933,
      "mean_token_accuracy": 0.17891673147678375,
      "num_tokens": 50840428.0,
      "step": 25985
    },
    {
      "entropy": 6.060116720199585,
      "epoch": 2.597990703253861,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004329263187433605,
      "loss": 5.6744,
      "mean_token_accuracy": 0.17266178280115127,
      "num_tokens": 50849596.0,
      "step": 25990
    },
    {
      "entropy": 6.046822738647461,
      "epoch": 2.59849052831509,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004329008960993694,
      "loss": 5.5737,
      "mean_token_accuracy": 0.18023378551006317,
      "num_tokens": 50858347.0,
      "step": 25995
    },
    {
      "entropy": 6.048694133758545,
      "epoch": 2.598990353376318,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004328754694826322,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18659407943487166,
      "num_tokens": 50868979.0,
      "step": 26000
    },
    {
      "entropy": 6.020805597305298,
      "epoch": 2.599490178437547,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004328500388937889,
      "loss": 5.6064,
      "mean_token_accuracy": 0.18512615710496902,
      "num_tokens": 50878789.0,
      "step": 26005
    },
    {
      "entropy": 6.0021617889404295,
      "epoch": 2.5999900034987755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043282460433347926,
      "loss": 5.6186,
      "mean_token_accuracy": 0.17801228761672974,
      "num_tokens": 50889381.0,
      "step": 26010
    },
    {
      "entropy": 6.1454438209533695,
      "epoch": 2.600489828560004,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004327991658023431,
      "loss": 5.6374,
      "mean_token_accuracy": 0.16995756477117538,
      "num_tokens": 50898201.0,
      "step": 26015
    },
    {
      "entropy": 6.062618255615234,
      "epoch": 2.6009896536212325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004327737233010205,
      "loss": 5.652,
      "mean_token_accuracy": 0.18697686195373536,
      "num_tokens": 50908185.0,
      "step": 26020
    },
    {
      "entropy": 6.00340838432312,
      "epoch": 2.601489478682461,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043274827683015174,
      "loss": 5.5411,
      "mean_token_accuracy": 0.18821829110383986,
      "num_tokens": 50918766.0,
      "step": 26025
    },
    {
      "entropy": 5.965804481506348,
      "epoch": 2.60198930374369,
      "grad_norm": 1.03125,
      "learning_rate": 0.00043272282639037683,
      "loss": 5.6013,
      "mean_token_accuracy": 0.18321091532707215,
      "num_tokens": 50928588.0,
      "step": 26030
    },
    {
      "entropy": 5.999594783782959,
      "epoch": 2.602489128804918,
      "grad_norm": 1.6328125,
      "learning_rate": 0.00043269737198233627,
      "loss": 5.6054,
      "mean_token_accuracy": 0.18262286782264708,
      "num_tokens": 50938106.0,
      "step": 26035
    },
    {
      "entropy": 6.132781744003296,
      "epoch": 2.602988953866147,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004326719136066704,
      "loss": 5.7364,
      "mean_token_accuracy": 0.1732128694653511,
      "num_tokens": 50947723.0,
      "step": 26040
    },
    {
      "entropy": 6.089908409118652,
      "epoch": 2.6034887789273755,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043264645126401977,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17459918409585953,
      "num_tokens": 50956744.0,
      "step": 26045
    },
    {
      "entropy": 6.11665153503418,
      "epoch": 2.603988603988604,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043262098495502497,
      "loss": 5.7903,
      "mean_token_accuracy": 0.17419612258672715,
      "num_tokens": 50967778.0,
      "step": 26050
    },
    {
      "entropy": 6.0627703189849855,
      "epoch": 2.6044884290498325,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004325955146803267,
      "loss": 5.5428,
      "mean_token_accuracy": 0.18596047461032866,
      "num_tokens": 50978635.0,
      "step": 26055
    },
    {
      "entropy": 6.158674001693726,
      "epoch": 2.604988254111061,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004325700404405659,
      "loss": 5.7842,
      "mean_token_accuracy": 0.16777411550283433,
      "num_tokens": 50988823.0,
      "step": 26060
    },
    {
      "entropy": 6.040266609191894,
      "epoch": 2.60548807917229,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004325445622363834,
      "loss": 5.683,
      "mean_token_accuracy": 0.17438930422067642,
      "num_tokens": 50998576.0,
      "step": 26065
    },
    {
      "entropy": 6.053354358673095,
      "epoch": 2.605987904233518,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004325190800684203,
      "loss": 5.5909,
      "mean_token_accuracy": 0.187750144302845,
      "num_tokens": 51008142.0,
      "step": 26070
    },
    {
      "entropy": 5.9316596508026125,
      "epoch": 2.606487729294747,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004324935939373176,
      "loss": 5.4361,
      "mean_token_accuracy": 0.19347186237573624,
      "num_tokens": 51018892.0,
      "step": 26075
    },
    {
      "entropy": 6.079810428619385,
      "epoch": 2.6069875543559755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004324681038437166,
      "loss": 5.6906,
      "mean_token_accuracy": 0.172246515750885,
      "num_tokens": 51029116.0,
      "step": 26080
    },
    {
      "entropy": 5.9676214218139645,
      "epoch": 2.607487379417204,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004324426097882587,
      "loss": 5.4892,
      "mean_token_accuracy": 0.19268075972795487,
      "num_tokens": 51038392.0,
      "step": 26085
    },
    {
      "entropy": 6.031398916244507,
      "epoch": 2.6079872044784325,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004324171117715852,
      "loss": 5.6258,
      "mean_token_accuracy": 0.1798061817884445,
      "num_tokens": 51048344.0,
      "step": 26090
    },
    {
      "entropy": 6.049861860275269,
      "epoch": 2.608487029539661,
      "grad_norm": 1.359375,
      "learning_rate": 0.00043239160979433765,
      "loss": 5.683,
      "mean_token_accuracy": 0.18048959821462632,
      "num_tokens": 51057879.0,
      "step": 26095
    },
    {
      "entropy": 6.019409418106079,
      "epoch": 2.6089868546008894,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004323661038571578,
      "loss": 5.5344,
      "mean_token_accuracy": 0.18801921904087066,
      "num_tokens": 51067365.0,
      "step": 26100
    },
    {
      "entropy": 5.978011989593506,
      "epoch": 2.609486679662118,
      "grad_norm": 1.15625,
      "learning_rate": 0.00043234059396068725,
      "loss": 5.5011,
      "mean_token_accuracy": 0.18201590776443483,
      "num_tokens": 51076168.0,
      "step": 26105
    },
    {
      "entropy": 6.042206573486328,
      "epoch": 2.609986504723347,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004323150801055677,
      "loss": 5.6531,
      "mean_token_accuracy": 0.17807339280843734,
      "num_tokens": 51086438.0,
      "step": 26110
    },
    {
      "entropy": 6.046844005584717,
      "epoch": 2.6104863297845755,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004322895622924414,
      "loss": 5.6334,
      "mean_token_accuracy": 0.1804952010512352,
      "num_tokens": 51095386.0,
      "step": 26115
    },
    {
      "entropy": 5.979467964172363,
      "epoch": 2.610986154845804,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004322640405219501,
      "loss": 5.547,
      "mean_token_accuracy": 0.18753506392240524,
      "num_tokens": 51105224.0,
      "step": 26120
    },
    {
      "entropy": 6.035736703872681,
      "epoch": 2.6114859799070325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043223851479473596,
      "loss": 5.6608,
      "mean_token_accuracy": 0.1776103511452675,
      "num_tokens": 51115011.0,
      "step": 26125
    },
    {
      "entropy": 6.105141067504883,
      "epoch": 2.611985804968261,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043221298511144133,
      "loss": 5.6421,
      "mean_token_accuracy": 0.17705494314432144,
      "num_tokens": 51124173.0,
      "step": 26130
    },
    {
      "entropy": 6.042300033569336,
      "epoch": 2.6124856300294894,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043218745147270845,
      "loss": 5.5002,
      "mean_token_accuracy": 0.19035637378692627,
      "num_tokens": 51133493.0,
      "step": 26135
    },
    {
      "entropy": 6.042124891281128,
      "epoch": 2.612985455090718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043216191387917966,
      "loss": 5.6945,
      "mean_token_accuracy": 0.17540936768054963,
      "num_tokens": 51142748.0,
      "step": 26140
    },
    {
      "entropy": 5.995274496078491,
      "epoch": 2.613485280151947,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004321363723314976,
      "loss": 5.5651,
      "mean_token_accuracy": 0.1895092323422432,
      "num_tokens": 51152681.0,
      "step": 26145
    },
    {
      "entropy": 5.982134628295898,
      "epoch": 2.6139851052131755,
      "grad_norm": 1.25,
      "learning_rate": 0.00043211082683030483,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18443993330001832,
      "num_tokens": 51161134.0,
      "step": 26150
    },
    {
      "entropy": 5.993594789505005,
      "epoch": 2.614484930274404,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043208527737624406,
      "loss": 5.5194,
      "mean_token_accuracy": 0.18869834393262863,
      "num_tokens": 51170623.0,
      "step": 26155
    },
    {
      "entropy": 6.083442974090576,
      "epoch": 2.6149847553356325,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043205972396995825,
      "loss": 5.6551,
      "mean_token_accuracy": 0.17430697679519652,
      "num_tokens": 51180290.0,
      "step": 26160
    },
    {
      "entropy": 6.037635087966919,
      "epoch": 2.615484580396861,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004320341666120901,
      "loss": 5.5629,
      "mean_token_accuracy": 0.18638709038496018,
      "num_tokens": 51188733.0,
      "step": 26165
    },
    {
      "entropy": 5.893578100204468,
      "epoch": 2.6159844054580894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043200860530328287,
      "loss": 5.5293,
      "mean_token_accuracy": 0.19662653505802155,
      "num_tokens": 51198403.0,
      "step": 26170
    },
    {
      "entropy": 5.940001726150513,
      "epoch": 2.616484230519318,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043198304004417933,
      "loss": 5.5338,
      "mean_token_accuracy": 0.18433621525764465,
      "num_tokens": 51207654.0,
      "step": 26175
    },
    {
      "entropy": 6.046866369247437,
      "epoch": 2.616984055580547,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043195747083542296,
      "loss": 5.5943,
      "mean_token_accuracy": 0.17785379737615586,
      "num_tokens": 51216285.0,
      "step": 26180
    },
    {
      "entropy": 5.8981798648834225,
      "epoch": 2.6174838806417755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.000431931897677657,
      "loss": 5.3769,
      "mean_token_accuracy": 0.2034571185708046,
      "num_tokens": 51225182.0,
      "step": 26185
    },
    {
      "entropy": 6.018731880187988,
      "epoch": 2.617983705703004,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043190632057152497,
      "loss": 5.6797,
      "mean_token_accuracy": 0.18106243759393692,
      "num_tokens": 51235547.0,
      "step": 26190
    },
    {
      "entropy": 6.072481775283814,
      "epoch": 2.6184835307642325,
      "grad_norm": 1.125,
      "learning_rate": 0.0004318807395176702,
      "loss": 5.6254,
      "mean_token_accuracy": 0.17653151601552963,
      "num_tokens": 51244897.0,
      "step": 26195
    },
    {
      "entropy": 5.990633249282837,
      "epoch": 2.618983355825461,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004318551545167364,
      "loss": 5.534,
      "mean_token_accuracy": 0.19011463224887848,
      "num_tokens": 51255343.0,
      "step": 26200
    },
    {
      "entropy": 6.0199055671691895,
      "epoch": 2.6194831808866894,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004318295655693674,
      "loss": 5.5767,
      "mean_token_accuracy": 0.1835000365972519,
      "num_tokens": 51265448.0,
      "step": 26205
    },
    {
      "entropy": 5.999718761444091,
      "epoch": 2.619983005947918,
      "grad_norm": 1.1875,
      "learning_rate": 0.00043180397267620675,
      "loss": 5.4709,
      "mean_token_accuracy": 0.18767277747392655,
      "num_tokens": 51273330.0,
      "step": 26210
    },
    {
      "entropy": 5.999001312255859,
      "epoch": 2.620482831009147,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004317783758378985,
      "loss": 5.6136,
      "mean_token_accuracy": 0.18070109337568283,
      "num_tokens": 51283564.0,
      "step": 26215
    },
    {
      "entropy": 6.019810009002685,
      "epoch": 2.6209826560703755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043175277505508677,
      "loss": 5.6178,
      "mean_token_accuracy": 0.179011932015419,
      "num_tokens": 51293386.0,
      "step": 26220
    },
    {
      "entropy": 6.035797548294068,
      "epoch": 2.621482481131604,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004317271703284154,
      "loss": 5.5886,
      "mean_token_accuracy": 0.17942273765802383,
      "num_tokens": 51301813.0,
      "step": 26225
    },
    {
      "entropy": 6.015041780471802,
      "epoch": 2.6219823061928325,
      "grad_norm": 1.265625,
      "learning_rate": 0.00043170156165852887,
      "loss": 5.573,
      "mean_token_accuracy": 0.1803755581378937,
      "num_tokens": 51312378.0,
      "step": 26230
    },
    {
      "entropy": 6.005851697921753,
      "epoch": 2.622482131254061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004316759490460713,
      "loss": 5.604,
      "mean_token_accuracy": 0.1879439175128937,
      "num_tokens": 51322022.0,
      "step": 26235
    },
    {
      "entropy": 6.011146020889282,
      "epoch": 2.6229819563152894,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043165033249168723,
      "loss": 5.5799,
      "mean_token_accuracy": 0.18212175071239473,
      "num_tokens": 51330288.0,
      "step": 26240
    },
    {
      "entropy": 6.011969041824341,
      "epoch": 2.623481781376518,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043162471199602103,
      "loss": 5.6114,
      "mean_token_accuracy": 0.18009622395038605,
      "num_tokens": 51338893.0,
      "step": 26245
    },
    {
      "entropy": 6.010849666595459,
      "epoch": 2.623981606437747,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043159908755971746,
      "loss": 5.5825,
      "mean_token_accuracy": 0.17565616071224213,
      "num_tokens": 51347860.0,
      "step": 26250
    },
    {
      "entropy": 6.032860136032104,
      "epoch": 2.6244814314989755,
      "grad_norm": 1.109375,
      "learning_rate": 0.00043157345918342103,
      "loss": 5.623,
      "mean_token_accuracy": 0.18308013826608657,
      "num_tokens": 51359240.0,
      "step": 26255
    },
    {
      "entropy": 6.091044855117798,
      "epoch": 2.624981256560204,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004315478268677768,
      "loss": 5.7094,
      "mean_token_accuracy": 0.17644814103841783,
      "num_tokens": 51368574.0,
      "step": 26260
    },
    {
      "entropy": 6.035233831405639,
      "epoch": 2.6254810816214325,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004315221906134294,
      "loss": 5.6055,
      "mean_token_accuracy": 0.1824919879436493,
      "num_tokens": 51378895.0,
      "step": 26265
    },
    {
      "entropy": 5.996111965179443,
      "epoch": 2.625980906682661,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043149655042102404,
      "loss": 5.5442,
      "mean_token_accuracy": 0.18598836660385132,
      "num_tokens": 51388659.0,
      "step": 26270
    },
    {
      "entropy": 6.091568803787231,
      "epoch": 2.6264807317438894,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043147090629120575,
      "loss": 5.7007,
      "mean_token_accuracy": 0.17653495073318481,
      "num_tokens": 51398444.0,
      "step": 26275
    },
    {
      "entropy": 5.991745090484619,
      "epoch": 2.626980556805118,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004314452582246197,
      "loss": 5.5781,
      "mean_token_accuracy": 0.1858636885881424,
      "num_tokens": 51408534.0,
      "step": 26280
    },
    {
      "entropy": 6.056564044952393,
      "epoch": 2.627480381866347,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043141960622191116,
      "loss": 5.6032,
      "mean_token_accuracy": 0.1816971555352211,
      "num_tokens": 51418174.0,
      "step": 26285
    },
    {
      "entropy": 5.9542001247406,
      "epoch": 2.6279802069275755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043139395028372565,
      "loss": 5.6215,
      "mean_token_accuracy": 0.18038197457790375,
      "num_tokens": 51428249.0,
      "step": 26290
    },
    {
      "entropy": 6.002531433105469,
      "epoch": 2.6284800319888038,
      "grad_norm": 1.5625,
      "learning_rate": 0.00043136829041070856,
      "loss": 5.5927,
      "mean_token_accuracy": 0.17573944181203843,
      "num_tokens": 51438042.0,
      "step": 26295
    },
    {
      "entropy": 6.068287372589111,
      "epoch": 2.6289798570500325,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043134262660350555,
      "loss": 5.6895,
      "mean_token_accuracy": 0.17279312163591384,
      "num_tokens": 51447745.0,
      "step": 26300
    },
    {
      "entropy": 6.086700010299682,
      "epoch": 2.629479682111261,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004313169588627622,
      "loss": 5.5552,
      "mean_token_accuracy": 0.18347810655832292,
      "num_tokens": 51458163.0,
      "step": 26305
    },
    {
      "entropy": 6.064768123626709,
      "epoch": 2.6299795071724894,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004312912871891245,
      "loss": 5.6536,
      "mean_token_accuracy": 0.17847360372543336,
      "num_tokens": 51468753.0,
      "step": 26310
    },
    {
      "entropy": 6.035714769363404,
      "epoch": 2.630479332233718,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004312656115832381,
      "loss": 5.6863,
      "mean_token_accuracy": 0.1785442814230919,
      "num_tokens": 51479411.0,
      "step": 26315
    },
    {
      "entropy": 6.060040521621704,
      "epoch": 2.630979157294947,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00043123993204574926,
      "loss": 5.727,
      "mean_token_accuracy": 0.17263293713331224,
      "num_tokens": 51489663.0,
      "step": 26320
    },
    {
      "entropy": 6.109456157684326,
      "epoch": 2.6314789823561755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043121424857730384,
      "loss": 5.6594,
      "mean_token_accuracy": 0.17569761276245116,
      "num_tokens": 51500455.0,
      "step": 26325
    },
    {
      "entropy": 6.036267900466919,
      "epoch": 2.6319788074174038,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004311885611785482,
      "loss": 5.6149,
      "mean_token_accuracy": 0.1818238839507103,
      "num_tokens": 51509536.0,
      "step": 26330
    },
    {
      "entropy": 5.977023077011109,
      "epoch": 2.6324786324786325,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043116286985012847,
      "loss": 5.5131,
      "mean_token_accuracy": 0.18480388671159745,
      "num_tokens": 51519016.0,
      "step": 26335
    },
    {
      "entropy": 5.993271923065185,
      "epoch": 2.632978457539861,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043113717459269116,
      "loss": 5.5579,
      "mean_token_accuracy": 0.17882864326238632,
      "num_tokens": 51528165.0,
      "step": 26340
    },
    {
      "entropy": 6.060551261901855,
      "epoch": 2.6334782826010894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004311114754068827,
      "loss": 5.6406,
      "mean_token_accuracy": 0.1814943253993988,
      "num_tokens": 51538338.0,
      "step": 26345
    },
    {
      "entropy": 6.035416316986084,
      "epoch": 2.633978107662318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004310857722933497,
      "loss": 5.6668,
      "mean_token_accuracy": 0.18159135729074477,
      "num_tokens": 51547397.0,
      "step": 26350
    },
    {
      "entropy": 5.984923601150513,
      "epoch": 2.634477932723547,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043106006525273896,
      "loss": 5.5632,
      "mean_token_accuracy": 0.18259994834661483,
      "num_tokens": 51558572.0,
      "step": 26355
    },
    {
      "entropy": 5.919953727722168,
      "epoch": 2.6349777577847755,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004310343542856969,
      "loss": 5.4324,
      "mean_token_accuracy": 0.19488199502229692,
      "num_tokens": 51567522.0,
      "step": 26360
    },
    {
      "entropy": 6.028505849838257,
      "epoch": 2.6354775828460038,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043100863939287077,
      "loss": 5.6255,
      "mean_token_accuracy": 0.17786143273115157,
      "num_tokens": 51577717.0,
      "step": 26365
    },
    {
      "entropy": 6.0693785667419435,
      "epoch": 2.6359774079072325,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004309829205749074,
      "loss": 5.6543,
      "mean_token_accuracy": 0.18707954585552217,
      "num_tokens": 51587301.0,
      "step": 26370
    },
    {
      "entropy": 6.025286293029785,
      "epoch": 2.636477232968461,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043095719783245386,
      "loss": 5.6178,
      "mean_token_accuracy": 0.1891126424074173,
      "num_tokens": 51597538.0,
      "step": 26375
    },
    {
      "entropy": 5.9920329570770265,
      "epoch": 2.6369770580296894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043093147116615737,
      "loss": 5.529,
      "mean_token_accuracy": 0.19699204415082933,
      "num_tokens": 51607698.0,
      "step": 26380
    },
    {
      "entropy": 5.994296503067017,
      "epoch": 2.637476883090918,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004309057405766652,
      "loss": 5.6541,
      "mean_token_accuracy": 0.1798229068517685,
      "num_tokens": 51616784.0,
      "step": 26385
    },
    {
      "entropy": 6.053320217132568,
      "epoch": 2.637976708152147,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00043088000606462474,
      "loss": 5.6887,
      "mean_token_accuracy": 0.176905120909214,
      "num_tokens": 51627024.0,
      "step": 26390
    },
    {
      "entropy": 6.088989496231079,
      "epoch": 2.6384765332133755,
      "grad_norm": 1.171875,
      "learning_rate": 0.00043085426763068334,
      "loss": 5.6082,
      "mean_token_accuracy": 0.17780465483665467,
      "num_tokens": 51636224.0,
      "step": 26395
    },
    {
      "entropy": 5.993315315246582,
      "epoch": 2.6389763582746038,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043082852527548876,
      "loss": 5.6001,
      "mean_token_accuracy": 0.177258138358593,
      "num_tokens": 51645457.0,
      "step": 26400
    },
    {
      "entropy": 6.028305578231811,
      "epoch": 2.6394761833358324,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004308027789996886,
      "loss": 5.5906,
      "mean_token_accuracy": 0.18581600189208985,
      "num_tokens": 51655649.0,
      "step": 26405
    },
    {
      "entropy": 5.988443040847779,
      "epoch": 2.639976008397061,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004307770288039306,
      "loss": 5.6355,
      "mean_token_accuracy": 0.17520159780979155,
      "num_tokens": 51664937.0,
      "step": 26410
    },
    {
      "entropy": 6.0312700271606445,
      "epoch": 2.6404758334582894,
      "grad_norm": 1.140625,
      "learning_rate": 0.00043075127468886264,
      "loss": 5.5383,
      "mean_token_accuracy": 0.18430096358060838,
      "num_tokens": 51674473.0,
      "step": 26415
    },
    {
      "entropy": 6.0390393257141115,
      "epoch": 2.640975658519518,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043072551665513265,
      "loss": 5.5529,
      "mean_token_accuracy": 0.17954092472791672,
      "num_tokens": 51684510.0,
      "step": 26420
    },
    {
      "entropy": 5.941063451766968,
      "epoch": 2.641475483580747,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00043069975470338875,
      "loss": 5.6075,
      "mean_token_accuracy": 0.18212295621633529,
      "num_tokens": 51693220.0,
      "step": 26425
    },
    {
      "entropy": 6.14421648979187,
      "epoch": 2.6419753086419755,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00043067398883427907,
      "loss": 5.6662,
      "mean_token_accuracy": 0.17967286407947541,
      "num_tokens": 51702396.0,
      "step": 26430
    },
    {
      "entropy": 6.111770820617676,
      "epoch": 2.6424751337032037,
      "grad_norm": 1.21875,
      "learning_rate": 0.00043064821904845185,
      "loss": 5.5801,
      "mean_token_accuracy": 0.1821412816643715,
      "num_tokens": 51712651.0,
      "step": 26435
    },
    {
      "entropy": 6.015494251251221,
      "epoch": 2.6429749587644324,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004306224453465556,
      "loss": 5.6648,
      "mean_token_accuracy": 0.17199369370937348,
      "num_tokens": 51722247.0,
      "step": 26440
    },
    {
      "entropy": 6.068347501754761,
      "epoch": 2.643474783825661,
      "grad_norm": 1.078125,
      "learning_rate": 0.00043059666772923854,
      "loss": 5.6345,
      "mean_token_accuracy": 0.1791454330086708,
      "num_tokens": 51731537.0,
      "step": 26445
    },
    {
      "entropy": 6.11707181930542,
      "epoch": 2.6439746088868894,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004305708861971495,
      "loss": 5.6902,
      "mean_token_accuracy": 0.1751371592283249,
      "num_tokens": 51741661.0,
      "step": 26450
    },
    {
      "entropy": 5.986990976333618,
      "epoch": 2.644474433948118,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004305451007509369,
      "loss": 5.5155,
      "mean_token_accuracy": 0.18261077105998993,
      "num_tokens": 51752813.0,
      "step": 26455
    },
    {
      "entropy": 6.01062970161438,
      "epoch": 2.644974259009347,
      "grad_norm": 1.53125,
      "learning_rate": 0.0004305193113912496,
      "loss": 5.5872,
      "mean_token_accuracy": 0.1857323542237282,
      "num_tokens": 51761690.0,
      "step": 26460
    },
    {
      "entropy": 6.029112958908081,
      "epoch": 2.6454740840705755,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00043049351811873645,
      "loss": 5.5435,
      "mean_token_accuracy": 0.18410504460334778,
      "num_tokens": 51772002.0,
      "step": 26465
    },
    {
      "entropy": 6.084415483474731,
      "epoch": 2.6459739091318037,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004304677209340464,
      "loss": 5.6929,
      "mean_token_accuracy": 0.17875198870897294,
      "num_tokens": 51783076.0,
      "step": 26470
    },
    {
      "entropy": 5.999786472320556,
      "epoch": 2.6464737341930324,
      "grad_norm": 1.296875,
      "learning_rate": 0.00043044191983782846,
      "loss": 5.5602,
      "mean_token_accuracy": 0.1910875454545021,
      "num_tokens": 51792710.0,
      "step": 26475
    },
    {
      "entropy": 6.076154184341431,
      "epoch": 2.646973559254261,
      "grad_norm": 1.0625,
      "learning_rate": 0.00043041611483073187,
      "loss": 5.7624,
      "mean_token_accuracy": 0.17419288009405137,
      "num_tokens": 51802896.0,
      "step": 26480
    },
    {
      "entropy": 6.010974359512329,
      "epoch": 2.6474733843154894,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004303903059134059,
      "loss": 5.5596,
      "mean_token_accuracy": 0.18701684474945068,
      "num_tokens": 51812363.0,
      "step": 26485
    },
    {
      "entropy": 6.027298402786255,
      "epoch": 2.647973209376718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043036449308649967,
      "loss": 5.5669,
      "mean_token_accuracy": 0.1842468112707138,
      "num_tokens": 51820970.0,
      "step": 26490
    },
    {
      "entropy": 6.064905214309692,
      "epoch": 2.648473034437947,
      "grad_norm": 1.203125,
      "learning_rate": 0.00043033867635066287,
      "loss": 5.6108,
      "mean_token_accuracy": 0.1815474346280098,
      "num_tokens": 51830115.0,
      "step": 26495
    },
    {
      "entropy": 6.031705379486084,
      "epoch": 2.6489728594991755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004303128557065448,
      "loss": 5.5496,
      "mean_token_accuracy": 0.18665410429239274,
      "num_tokens": 51839779.0,
      "step": 26500
    },
    {
      "entropy": 6.0120302677154545,
      "epoch": 2.6494726845604037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004302870311547954,
      "loss": 5.6515,
      "mean_token_accuracy": 0.17526234015822412,
      "num_tokens": 51850800.0,
      "step": 26505
    },
    {
      "entropy": 6.027283573150635,
      "epoch": 2.6499725096216324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00043026120269606423,
      "loss": 5.6165,
      "mean_token_accuracy": 0.1834244519472122,
      "num_tokens": 51860059.0,
      "step": 26510
    },
    {
      "entropy": 6.078717041015625,
      "epoch": 2.650472334682861,
      "grad_norm": 1.046875,
      "learning_rate": 0.00043023537033100106,
      "loss": 5.6889,
      "mean_token_accuracy": 0.17518247067928314,
      "num_tokens": 51870001.0,
      "step": 26515
    },
    {
      "entropy": 6.077863168716431,
      "epoch": 2.6509721597440894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043020953406025606,
      "loss": 5.5328,
      "mean_token_accuracy": 0.18552784621715546,
      "num_tokens": 51879004.0,
      "step": 26520
    },
    {
      "entropy": 6.003847455978393,
      "epoch": 2.651471984805318,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000430183693884479,
      "loss": 5.607,
      "mean_token_accuracy": 0.18087068647146226,
      "num_tokens": 51887911.0,
      "step": 26525
    },
    {
      "entropy": 5.936503124237061,
      "epoch": 2.651971809866547,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004301578498043202,
      "loss": 5.4853,
      "mean_token_accuracy": 0.19177916049957275,
      "num_tokens": 51897444.0,
      "step": 26530
    },
    {
      "entropy": 5.98862886428833,
      "epoch": 2.6524716349277755,
      "grad_norm": 1.09375,
      "learning_rate": 0.00043013200182042983,
      "loss": 5.5206,
      "mean_token_accuracy": 0.18935388773679734,
      "num_tokens": 51907344.0,
      "step": 26535
    },
    {
      "entropy": 6.027761173248291,
      "epoch": 2.6529714599890037,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00043010614993345815,
      "loss": 5.5212,
      "mean_token_accuracy": 0.1893118530511856,
      "num_tokens": 51916541.0,
      "step": 26540
    },
    {
      "entropy": 6.063733291625977,
      "epoch": 2.6534712850502324,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004300802941440557,
      "loss": 5.7248,
      "mean_token_accuracy": 0.17265397310256958,
      "num_tokens": 51926160.0,
      "step": 26545
    },
    {
      "entropy": 6.054165363311768,
      "epoch": 2.653971110111461,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004300544344528729,
      "loss": 5.6506,
      "mean_token_accuracy": 0.17919763624668122,
      "num_tokens": 51936244.0,
      "step": 26550
    },
    {
      "entropy": 6.111879777908325,
      "epoch": 2.6544709351726894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00043002857086056053,
      "loss": 5.6001,
      "mean_token_accuracy": 0.18080134242773055,
      "num_tokens": 51946490.0,
      "step": 26555
    },
    {
      "entropy": 6.108101034164429,
      "epoch": 2.654970760233918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004300027033677691,
      "loss": 5.6051,
      "mean_token_accuracy": 0.18039559870958327,
      "num_tokens": 51956129.0,
      "step": 26560
    },
    {
      "entropy": 5.983809947967529,
      "epoch": 2.655470585295147,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004299768319751496,
      "loss": 5.6279,
      "mean_token_accuracy": 0.17894519716501237,
      "num_tokens": 51966274.0,
      "step": 26565
    },
    {
      "entropy": 5.91192741394043,
      "epoch": 2.6559704103563755,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004299509566833529,
      "loss": 5.523,
      "mean_token_accuracy": 0.18894741535186768,
      "num_tokens": 51975261.0,
      "step": 26570
    },
    {
      "entropy": 6.042177724838257,
      "epoch": 2.6564702354176037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042992507749302995,
      "loss": 5.5898,
      "mean_token_accuracy": 0.1774608463048935,
      "num_tokens": 51984562.0,
      "step": 26575
    },
    {
      "entropy": 6.053143787384033,
      "epoch": 2.6569700604788324,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042989919440483194,
      "loss": 5.5542,
      "mean_token_accuracy": 0.1784242421388626,
      "num_tokens": 51994763.0,
      "step": 26580
    },
    {
      "entropy": 6.106169366836548,
      "epoch": 2.657469885540061,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004298733074194101,
      "loss": 5.6315,
      "mean_token_accuracy": 0.18968216180801392,
      "num_tokens": 52004935.0,
      "step": 26585
    },
    {
      "entropy": 5.992870473861695,
      "epoch": 2.6579697106012894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004298474165374156,
      "loss": 5.565,
      "mean_token_accuracy": 0.1794860541820526,
      "num_tokens": 52013879.0,
      "step": 26590
    },
    {
      "entropy": 5.962762022018433,
      "epoch": 2.658469535662518,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004298215217595,
      "loss": 5.5558,
      "mean_token_accuracy": 0.19106806069612503,
      "num_tokens": 52023354.0,
      "step": 26595
    },
    {
      "entropy": 6.035774230957031,
      "epoch": 2.658969360723747,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042979562308631476,
      "loss": 5.6539,
      "mean_token_accuracy": 0.17981230616569518,
      "num_tokens": 52033058.0,
      "step": 26600
    },
    {
      "entropy": 6.102031803131103,
      "epoch": 2.6594691857849755,
      "grad_norm": 1.25,
      "learning_rate": 0.0004297697205185115,
      "loss": 5.6061,
      "mean_token_accuracy": 0.17129491716623307,
      "num_tokens": 52042544.0,
      "step": 26605
    },
    {
      "entropy": 6.037606620788575,
      "epoch": 2.6599690108462037,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004297438140567419,
      "loss": 5.6073,
      "mean_token_accuracy": 0.1811242178082466,
      "num_tokens": 52052396.0,
      "step": 26610
    },
    {
      "entropy": 6.0375124454498295,
      "epoch": 2.6604688359074324,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004297179037016577,
      "loss": 5.6247,
      "mean_token_accuracy": 0.1794034406542778,
      "num_tokens": 52061950.0,
      "step": 26615
    },
    {
      "entropy": 5.98329906463623,
      "epoch": 2.6609686609686607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042969198945391084,
      "loss": 5.6311,
      "mean_token_accuracy": 0.1823697119951248,
      "num_tokens": 52071588.0,
      "step": 26620
    },
    {
      "entropy": 6.001740741729736,
      "epoch": 2.6614684860298894,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004296660713141534,
      "loss": 5.5144,
      "mean_token_accuracy": 0.18170634359121324,
      "num_tokens": 52080826.0,
      "step": 26625
    },
    {
      "entropy": 6.052634572982788,
      "epoch": 2.661968311091118,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042964014928303733,
      "loss": 5.6646,
      "mean_token_accuracy": 0.17764320522546767,
      "num_tokens": 52090844.0,
      "step": 26630
    },
    {
      "entropy": 6.108454656600952,
      "epoch": 2.662468136152347,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000429614223361215,
      "loss": 5.7494,
      "mean_token_accuracy": 0.17372922748327255,
      "num_tokens": 52100189.0,
      "step": 26635
    },
    {
      "entropy": 6.034653282165527,
      "epoch": 2.6629679612135755,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004295882935493385,
      "loss": 5.6553,
      "mean_token_accuracy": 0.18262167274951935,
      "num_tokens": 52110323.0,
      "step": 26640
    },
    {
      "entropy": 5.990649461746216,
      "epoch": 2.6634677862748037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042956235984806034,
      "loss": 5.5535,
      "mean_token_accuracy": 0.18390677571296693,
      "num_tokens": 52120816.0,
      "step": 26645
    },
    {
      "entropy": 6.12251238822937,
      "epoch": 2.6639676113360324,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042953642225803296,
      "loss": 5.6443,
      "mean_token_accuracy": 0.18099390268325805,
      "num_tokens": 52130610.0,
      "step": 26650
    },
    {
      "entropy": 6.072726631164551,
      "epoch": 2.6644674363972607,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000429510480779909,
      "loss": 5.6009,
      "mean_token_accuracy": 0.18310339897871017,
      "num_tokens": 52140581.0,
      "step": 26655
    },
    {
      "entropy": 6.029133558273315,
      "epoch": 2.6649672614584894,
      "grad_norm": 1.640625,
      "learning_rate": 0.0004294845354143411,
      "loss": 5.598,
      "mean_token_accuracy": 0.18432868421077728,
      "num_tokens": 52150971.0,
      "step": 26660
    },
    {
      "entropy": 6.011661338806152,
      "epoch": 2.665467086519718,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004294585861619821,
      "loss": 5.6067,
      "mean_token_accuracy": 0.18296149373054504,
      "num_tokens": 52160663.0,
      "step": 26665
    },
    {
      "entropy": 5.975858592987061,
      "epoch": 2.665966911580947,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042943263302348475,
      "loss": 5.6057,
      "mean_token_accuracy": 0.18469782173633575,
      "num_tokens": 52170570.0,
      "step": 26670
    },
    {
      "entropy": 6.121520614624023,
      "epoch": 2.6664667366421755,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042940667599950205,
      "loss": 5.6828,
      "mean_token_accuracy": 0.1736326575279236,
      "num_tokens": 52179682.0,
      "step": 26675
    },
    {
      "entropy": 6.049258613586426,
      "epoch": 2.6669665617034037,
      "grad_norm": 1.265625,
      "learning_rate": 0.00042938071509068717,
      "loss": 5.637,
      "mean_token_accuracy": 0.18291324526071548,
      "num_tokens": 52189856.0,
      "step": 26680
    },
    {
      "entropy": 5.971758842468262,
      "epoch": 2.6674663867646324,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042935475029769323,
      "loss": 5.4626,
      "mean_token_accuracy": 0.1904817134141922,
      "num_tokens": 52197956.0,
      "step": 26685
    },
    {
      "entropy": 5.9092793464660645,
      "epoch": 2.6679662118258607,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042932878162117347,
      "loss": 5.4752,
      "mean_token_accuracy": 0.1970234751701355,
      "num_tokens": 52208091.0,
      "step": 26690
    },
    {
      "entropy": 6.039830923080444,
      "epoch": 2.6684660368870894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004293028090617813,
      "loss": 5.6743,
      "mean_token_accuracy": 0.16918589323759078,
      "num_tokens": 52218568.0,
      "step": 26695
    },
    {
      "entropy": 5.996864080429077,
      "epoch": 2.668965861948318,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00042927683262017016,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1867714211344719,
      "num_tokens": 52228048.0,
      "step": 26700
    },
    {
      "entropy": 6.029834461212158,
      "epoch": 2.669465687009547,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004292508522969935,
      "loss": 5.6638,
      "mean_token_accuracy": 0.17516565471887588,
      "num_tokens": 52239878.0,
      "step": 26705
    },
    {
      "entropy": 6.008866453170777,
      "epoch": 2.6699655120707755,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004292248680929052,
      "loss": 5.6467,
      "mean_token_accuracy": 0.18211662769317627,
      "num_tokens": 52250244.0,
      "step": 26710
    },
    {
      "entropy": 6.030343866348266,
      "epoch": 2.6704653371320037,
      "grad_norm": 1.125,
      "learning_rate": 0.00042919888000855885,
      "loss": 5.6415,
      "mean_token_accuracy": 0.18294674158096313,
      "num_tokens": 52261420.0,
      "step": 26715
    },
    {
      "entropy": 5.952150344848633,
      "epoch": 2.6709651621932324,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042917288804460835,
      "loss": 5.4989,
      "mean_token_accuracy": 0.18808284848928453,
      "num_tokens": 52270956.0,
      "step": 26720
    },
    {
      "entropy": 6.04210729598999,
      "epoch": 2.6714649872544607,
      "grad_norm": 1.0,
      "learning_rate": 0.0004291468922017078,
      "loss": 5.6061,
      "mean_token_accuracy": 0.1795856922864914,
      "num_tokens": 52280467.0,
      "step": 26725
    },
    {
      "entropy": 6.014393711090088,
      "epoch": 2.6719648123156894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000429120892480511,
      "loss": 5.6148,
      "mean_token_accuracy": 0.18586645722389222,
      "num_tokens": 52290630.0,
      "step": 26730
    },
    {
      "entropy": 6.001048469543457,
      "epoch": 2.672464637376918,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004290948888816722,
      "loss": 5.6376,
      "mean_token_accuracy": 0.18222132027149202,
      "num_tokens": 52300956.0,
      "step": 26735
    },
    {
      "entropy": 5.981879043579101,
      "epoch": 2.672964462438147,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004290688814058456,
      "loss": 5.6494,
      "mean_token_accuracy": 0.17766705751419068,
      "num_tokens": 52310887.0,
      "step": 26740
    },
    {
      "entropy": 5.985613679885864,
      "epoch": 2.6734642874993755,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004290428700536857,
      "loss": 5.5429,
      "mean_token_accuracy": 0.18922256380319596,
      "num_tokens": 52320490.0,
      "step": 26745
    },
    {
      "entropy": 6.085641860961914,
      "epoch": 2.6739641125606037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004290168548258467,
      "loss": 5.6193,
      "mean_token_accuracy": 0.1796812444925308,
      "num_tokens": 52330477.0,
      "step": 26750
    },
    {
      "entropy": 6.024813890457153,
      "epoch": 2.6744639376218324,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042899083572298335,
      "loss": 5.6305,
      "mean_token_accuracy": 0.18234698921442033,
      "num_tokens": 52339909.0,
      "step": 26755
    },
    {
      "entropy": 5.957294893264771,
      "epoch": 2.6749637626830607,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042896481274575014,
      "loss": 5.484,
      "mean_token_accuracy": 0.19024553298950195,
      "num_tokens": 52349554.0,
      "step": 26760
    },
    {
      "entropy": 5.973018026351928,
      "epoch": 2.6754635877442894,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042893878589480195,
      "loss": 5.5857,
      "mean_token_accuracy": 0.18102037757635117,
      "num_tokens": 52359561.0,
      "step": 26765
    },
    {
      "entropy": 5.893107557296753,
      "epoch": 2.675963412805518,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042891275517079345,
      "loss": 5.448,
      "mean_token_accuracy": 0.1908065527677536,
      "num_tokens": 52369360.0,
      "step": 26770
    },
    {
      "entropy": 6.112000370025635,
      "epoch": 2.676463237866747,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042888672057437964,
      "loss": 5.645,
      "mean_token_accuracy": 0.1738707773387432,
      "num_tokens": 52379433.0,
      "step": 26775
    },
    {
      "entropy": 6.0326180934906,
      "epoch": 2.6769630629279755,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042886068210621557,
      "loss": 5.6186,
      "mean_token_accuracy": 0.1823671355843544,
      "num_tokens": 52389640.0,
      "step": 26780
    },
    {
      "entropy": 5.9240375518798825,
      "epoch": 2.6774628879892037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004288346397669563,
      "loss": 5.5019,
      "mean_token_accuracy": 0.19580742716789246,
      "num_tokens": 52399542.0,
      "step": 26785
    },
    {
      "entropy": 6.033792304992676,
      "epoch": 2.6779627130504324,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004288085935572571,
      "loss": 5.6089,
      "mean_token_accuracy": 0.18185745626688005,
      "num_tokens": 52408366.0,
      "step": 26790
    },
    {
      "entropy": 5.962434434890747,
      "epoch": 2.6784625381116607,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004287825434777733,
      "loss": 5.5951,
      "mean_token_accuracy": 0.17849481254816055,
      "num_tokens": 52419047.0,
      "step": 26795
    },
    {
      "entropy": 5.948139953613281,
      "epoch": 2.6789623631728894,
      "grad_norm": 1.296875,
      "learning_rate": 0.00042875648952916024,
      "loss": 5.4998,
      "mean_token_accuracy": 0.19390154182910918,
      "num_tokens": 52428871.0,
      "step": 26800
    },
    {
      "entropy": 5.972239303588867,
      "epoch": 2.679462188234118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004287304317120734,
      "loss": 5.6295,
      "mean_token_accuracy": 0.1845841646194458,
      "num_tokens": 52439385.0,
      "step": 26805
    },
    {
      "entropy": 6.013038873672485,
      "epoch": 2.679962013295347,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042870437002716854,
      "loss": 5.6026,
      "mean_token_accuracy": 0.1910457119345665,
      "num_tokens": 52449844.0,
      "step": 26810
    },
    {
      "entropy": 5.971204376220703,
      "epoch": 2.680461838356575,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004286783044751013,
      "loss": 5.5686,
      "mean_token_accuracy": 0.18059124052524567,
      "num_tokens": 52458853.0,
      "step": 26815
    },
    {
      "entropy": 6.013601064682007,
      "epoch": 2.6809616634178037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004286522350565274,
      "loss": 5.6235,
      "mean_token_accuracy": 0.1773431956768036,
      "num_tokens": 52468321.0,
      "step": 26820
    },
    {
      "entropy": 6.008394765853882,
      "epoch": 2.6814614884790324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042862616177210286,
      "loss": 5.6348,
      "mean_token_accuracy": 0.17846055775880815,
      "num_tokens": 52478015.0,
      "step": 26825
    },
    {
      "entropy": 5.9596200466156,
      "epoch": 2.6819613135402607,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004286000846224836,
      "loss": 5.5393,
      "mean_token_accuracy": 0.18044990450143814,
      "num_tokens": 52487589.0,
      "step": 26830
    },
    {
      "entropy": 5.96417350769043,
      "epoch": 2.6824611386014894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042857400360832575,
      "loss": 5.636,
      "mean_token_accuracy": 0.1853545770049095,
      "num_tokens": 52496375.0,
      "step": 26835
    },
    {
      "entropy": 6.052075958251953,
      "epoch": 2.682960963662718,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004285479187302854,
      "loss": 5.6716,
      "mean_token_accuracy": 0.1761649563908577,
      "num_tokens": 52506469.0,
      "step": 26840
    },
    {
      "entropy": 6.117903804779052,
      "epoch": 2.6834607887239468,
      "grad_norm": 1.140625,
      "learning_rate": 0.000428521829989019,
      "loss": 5.6703,
      "mean_token_accuracy": 0.17436031401157379,
      "num_tokens": 52515748.0,
      "step": 26845
    },
    {
      "entropy": 6.006530523300171,
      "epoch": 2.683960613785175,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004284957373851829,
      "loss": 5.622,
      "mean_token_accuracy": 0.17840612679719925,
      "num_tokens": 52524742.0,
      "step": 26850
    },
    {
      "entropy": 6.044181156158447,
      "epoch": 2.6844604388464037,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042846964091943347,
      "loss": 5.6355,
      "mean_token_accuracy": 0.17338339239358902,
      "num_tokens": 52534502.0,
      "step": 26855
    },
    {
      "entropy": 6.033875179290772,
      "epoch": 2.6849602639076324,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042844354059242735,
      "loss": 5.5617,
      "mean_token_accuracy": 0.17953009754419327,
      "num_tokens": 52544217.0,
      "step": 26860
    },
    {
      "entropy": 6.020219421386718,
      "epoch": 2.6854600889688607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004284174364048213,
      "loss": 5.5556,
      "mean_token_accuracy": 0.179906490445137,
      "num_tokens": 52555066.0,
      "step": 26865
    },
    {
      "entropy": 5.978253030776978,
      "epoch": 2.6859599140300894,
      "grad_norm": 1.078125,
      "learning_rate": 0.000428391328357272,
      "loss": 5.6166,
      "mean_token_accuracy": 0.17836934477090835,
      "num_tokens": 52564957.0,
      "step": 26870
    },
    {
      "entropy": 5.999014282226563,
      "epoch": 2.686459739091318,
      "grad_norm": 1.125,
      "learning_rate": 0.0004283652164504364,
      "loss": 5.559,
      "mean_token_accuracy": 0.1837862327694893,
      "num_tokens": 52574893.0,
      "step": 26875
    },
    {
      "entropy": 5.950049781799317,
      "epoch": 2.6869595641525468,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042833910068497134,
      "loss": 5.476,
      "mean_token_accuracy": 0.1946884274482727,
      "num_tokens": 52583307.0,
      "step": 26880
    },
    {
      "entropy": 5.9665642261505125,
      "epoch": 2.687459389213775,
      "grad_norm": 1.125,
      "learning_rate": 0.000428312981061534,
      "loss": 5.5236,
      "mean_token_accuracy": 0.18873154520988464,
      "num_tokens": 52593055.0,
      "step": 26885
    },
    {
      "entropy": 5.995660018920899,
      "epoch": 2.6879592142750037,
      "grad_norm": 1.25,
      "learning_rate": 0.0004282868575807815,
      "loss": 5.5163,
      "mean_token_accuracy": 0.18528735041618347,
      "num_tokens": 52602068.0,
      "step": 26890
    },
    {
      "entropy": 6.016930341720581,
      "epoch": 2.6884590393362324,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042826073024337117,
      "loss": 5.5577,
      "mean_token_accuracy": 0.191050623357296,
      "num_tokens": 52609964.0,
      "step": 26895
    },
    {
      "entropy": 6.039786863327026,
      "epoch": 2.6889588643974607,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004282345990499602,
      "loss": 5.6217,
      "mean_token_accuracy": 0.17834055572748184,
      "num_tokens": 52619144.0,
      "step": 26900
    },
    {
      "entropy": 6.088885021209717,
      "epoch": 2.6894586894586894,
      "grad_norm": 1.125,
      "learning_rate": 0.0004282084640012062,
      "loss": 5.6946,
      "mean_token_accuracy": 0.17057913392782212,
      "num_tokens": 52629508.0,
      "step": 26905
    },
    {
      "entropy": 6.070838689804077,
      "epoch": 2.689958514519918,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004281823250977667,
      "loss": 5.6015,
      "mean_token_accuracy": 0.17516913414001464,
      "num_tokens": 52640037.0,
      "step": 26910
    },
    {
      "entropy": 5.965433406829834,
      "epoch": 2.6904583395811468,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042815618234029923,
      "loss": 5.5359,
      "mean_token_accuracy": 0.18618329465389252,
      "num_tokens": 52649293.0,
      "step": 26915
    },
    {
      "entropy": 6.016424608230591,
      "epoch": 2.690958164642375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042813003572946163,
      "loss": 5.6131,
      "mean_token_accuracy": 0.1759931579232216,
      "num_tokens": 52659560.0,
      "step": 26920
    },
    {
      "entropy": 6.070631456375122,
      "epoch": 2.6914579897036037,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042810388526591184,
      "loss": 5.7665,
      "mean_token_accuracy": 0.17195162028074265,
      "num_tokens": 52669105.0,
      "step": 26925
    },
    {
      "entropy": 6.039253759384155,
      "epoch": 2.6919578147648324,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042807773095030757,
      "loss": 5.5135,
      "mean_token_accuracy": 0.1972134903073311,
      "num_tokens": 52679633.0,
      "step": 26930
    },
    {
      "entropy": 5.96162896156311,
      "epoch": 2.6924576398260607,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004280515727833071,
      "loss": 5.5603,
      "mean_token_accuracy": 0.18953514248132705,
      "num_tokens": 52690037.0,
      "step": 26935
    },
    {
      "entropy": 6.055363559722901,
      "epoch": 2.6929574648872894,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004280254107655684,
      "loss": 5.6598,
      "mean_token_accuracy": 0.17485712468624115,
      "num_tokens": 52701796.0,
      "step": 26940
    },
    {
      "entropy": 5.916491746902466,
      "epoch": 2.693457289948518,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004279992448977497,
      "loss": 5.397,
      "mean_token_accuracy": 0.1929888755083084,
      "num_tokens": 52710634.0,
      "step": 26945
    },
    {
      "entropy": 5.966639518737793,
      "epoch": 2.6939571150097468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004279730751805094,
      "loss": 5.6659,
      "mean_token_accuracy": 0.17517506033182145,
      "num_tokens": 52721178.0,
      "step": 26950
    },
    {
      "entropy": 5.992067098617554,
      "epoch": 2.694456940070975,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042794690161450583,
      "loss": 5.521,
      "mean_token_accuracy": 0.18638920038938522,
      "num_tokens": 52730635.0,
      "step": 26955
    },
    {
      "entropy": 6.054883003234863,
      "epoch": 2.6949567651322037,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004279207242003976,
      "loss": 5.5384,
      "mean_token_accuracy": 0.1896473839879036,
      "num_tokens": 52740711.0,
      "step": 26960
    },
    {
      "entropy": 5.950598669052124,
      "epoch": 2.6954565901934324,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042789454293884336,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18343938291072845,
      "num_tokens": 52749430.0,
      "step": 26965
    },
    {
      "entropy": 5.962781286239624,
      "epoch": 2.6959564152546607,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004278683578305017,
      "loss": 5.5077,
      "mean_token_accuracy": 0.19219315201044082,
      "num_tokens": 52758288.0,
      "step": 26970
    },
    {
      "entropy": 6.003366899490357,
      "epoch": 2.6964562403158894,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042784216887603154,
      "loss": 5.6177,
      "mean_token_accuracy": 0.18064150363206863,
      "num_tokens": 52768190.0,
      "step": 26975
    },
    {
      "entropy": 6.069954919815063,
      "epoch": 2.696956065377118,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004278159760760917,
      "loss": 5.5986,
      "mean_token_accuracy": 0.18276410698890685,
      "num_tokens": 52779262.0,
      "step": 26980
    },
    {
      "entropy": 6.005918502807617,
      "epoch": 2.6974558904383468,
      "grad_norm": 1.3984375,
      "learning_rate": 0.0004277897794313412,
      "loss": 5.6008,
      "mean_token_accuracy": 0.1810927927494049,
      "num_tokens": 52788564.0,
      "step": 26985
    },
    {
      "entropy": 6.007136297225952,
      "epoch": 2.697955715499575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004277635789424393,
      "loss": 5.597,
      "mean_token_accuracy": 0.1826305106282234,
      "num_tokens": 52797293.0,
      "step": 26990
    },
    {
      "entropy": 5.957183074951172,
      "epoch": 2.6984555405608037,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004277373746100449,
      "loss": 5.5173,
      "mean_token_accuracy": 0.18230416923761367,
      "num_tokens": 52806815.0,
      "step": 26995
    },
    {
      "entropy": 5.999039125442505,
      "epoch": 2.6989553656220324,
      "grad_norm": 1.2734375,
      "learning_rate": 0.00042771116643481753,
      "loss": 5.5619,
      "mean_token_accuracy": 0.1829219087958336,
      "num_tokens": 52815496.0,
      "step": 27000
    },
    {
      "epoch": 2.6989553656220324,
      "eval_entropy": 5.838070204011622,
      "eval_loss": 5.710117340087891,
      "eval_mean_token_accuracy": 0.18427451314484758,
      "eval_num_tokens": 52815496.0,
      "eval_runtime": 23.9487,
      "eval_samples_per_second": 1296.312,
      "eval_steps_per_second": 162.055,
      "step": 27000
    },
    {
      "entropy": 6.093468713760376,
      "epoch": 2.6994551906832607,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004276849544174166,
      "loss": 5.6742,
      "mean_token_accuracy": 0.17093367129564285,
      "num_tokens": 52825203.0,
      "step": 27005
    },
    {
      "entropy": 5.986290407180786,
      "epoch": 2.6999550157444894,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042765873855850133,
      "loss": 5.6361,
      "mean_token_accuracy": 0.1830984905362129,
      "num_tokens": 52835956.0,
      "step": 27010
    },
    {
      "entropy": 5.9616741180419925,
      "epoch": 2.700454840805718,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004276325188587316,
      "loss": 5.5936,
      "mean_token_accuracy": 0.18605817705392838,
      "num_tokens": 52845314.0,
      "step": 27015
    },
    {
      "entropy": 6.025298023223877,
      "epoch": 2.7009546658669468,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004276062953187669,
      "loss": 5.5668,
      "mean_token_accuracy": 0.1808945655822754,
      "num_tokens": 52855693.0,
      "step": 27020
    },
    {
      "entropy": 6.034712600708008,
      "epoch": 2.701454490928175,
      "grad_norm": 1.125,
      "learning_rate": 0.0004275800679392671,
      "loss": 5.6137,
      "mean_token_accuracy": 0.18032210618257521,
      "num_tokens": 52865931.0,
      "step": 27025
    },
    {
      "entropy": 6.026051950454712,
      "epoch": 2.7019543159894037,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004275538367208921,
      "loss": 5.6078,
      "mean_token_accuracy": 0.1831350475549698,
      "num_tokens": 52876329.0,
      "step": 27030
    },
    {
      "entropy": 6.04150767326355,
      "epoch": 2.7024541410506324,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042752760166430185,
      "loss": 5.63,
      "mean_token_accuracy": 0.17290515452623367,
      "num_tokens": 52886876.0,
      "step": 27035
    },
    {
      "entropy": 6.01654167175293,
      "epoch": 2.7029539661118607,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042750136277015625,
      "loss": 5.6375,
      "mean_token_accuracy": 0.178939650952816,
      "num_tokens": 52896584.0,
      "step": 27040
    },
    {
      "entropy": 6.063281869888305,
      "epoch": 2.7034537911730894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042747512003911577,
      "loss": 5.5672,
      "mean_token_accuracy": 0.18576823621988298,
      "num_tokens": 52906388.0,
      "step": 27045
    },
    {
      "entropy": 5.969471216201782,
      "epoch": 2.703953616234318,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042744887347184035,
      "loss": 5.5,
      "mean_token_accuracy": 0.1872352808713913,
      "num_tokens": 52915732.0,
      "step": 27050
    },
    {
      "entropy": 5.985595703125,
      "epoch": 2.7044534412955468,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004274226230689906,
      "loss": 5.6605,
      "mean_token_accuracy": 0.17719454020261766,
      "num_tokens": 52926066.0,
      "step": 27055
    },
    {
      "entropy": 6.004272508621216,
      "epoch": 2.704953266356775,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042739636883122687,
      "loss": 5.6017,
      "mean_token_accuracy": 0.18293238580226898,
      "num_tokens": 52936075.0,
      "step": 27060
    },
    {
      "entropy": 5.96013503074646,
      "epoch": 2.7054530914180037,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042737011075920974,
      "loss": 5.6005,
      "mean_token_accuracy": 0.17806628197431565,
      "num_tokens": 52944266.0,
      "step": 27065
    },
    {
      "entropy": 6.088792419433593,
      "epoch": 2.7059529164792324,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004273438488535997,
      "loss": 5.7037,
      "mean_token_accuracy": 0.17846416383981706,
      "num_tokens": 52954999.0,
      "step": 27070
    },
    {
      "entropy": 6.077151203155518,
      "epoch": 2.7064527415404607,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042731758311505773,
      "loss": 5.7234,
      "mean_token_accuracy": 0.16800843179225922,
      "num_tokens": 52965374.0,
      "step": 27075
    },
    {
      "entropy": 5.975375127792359,
      "epoch": 2.7069525666016894,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042729131354424455,
      "loss": 5.5372,
      "mean_token_accuracy": 0.18939847350120545,
      "num_tokens": 52975956.0,
      "step": 27080
    },
    {
      "entropy": 5.983653926849366,
      "epoch": 2.707452391662918,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004272650401418211,
      "loss": 5.5198,
      "mean_token_accuracy": 0.18422410935163497,
      "num_tokens": 52985595.0,
      "step": 27085
    },
    {
      "entropy": 6.015768814086914,
      "epoch": 2.7079522167241468,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004272387629084484,
      "loss": 5.6731,
      "mean_token_accuracy": 0.17442942559719085,
      "num_tokens": 52994755.0,
      "step": 27090
    },
    {
      "entropy": 6.057124042510987,
      "epoch": 2.708452041785375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042721248184478765,
      "loss": 5.6425,
      "mean_token_accuracy": 0.17692352682352067,
      "num_tokens": 53004529.0,
      "step": 27095
    },
    {
      "entropy": 6.108982181549072,
      "epoch": 2.7089518668466037,
      "grad_norm": 1.125,
      "learning_rate": 0.0004271861969515,
      "loss": 5.7727,
      "mean_token_accuracy": 0.17348310798406602,
      "num_tokens": 53015223.0,
      "step": 27100
    },
    {
      "entropy": 6.0927080631256105,
      "epoch": 2.7094516919078324,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00042715990822924673,
      "loss": 5.6868,
      "mean_token_accuracy": 0.17257785350084304,
      "num_tokens": 53026014.0,
      "step": 27105
    },
    {
      "entropy": 6.088343620300293,
      "epoch": 2.7099515169690607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042713361567868946,
      "loss": 5.6115,
      "mean_token_accuracy": 0.17847627848386766,
      "num_tokens": 53036463.0,
      "step": 27110
    },
    {
      "entropy": 6.0414213180542,
      "epoch": 2.7104513420302894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042710731930048945,
      "loss": 5.5987,
      "mean_token_accuracy": 0.18209393620491027,
      "num_tokens": 53045933.0,
      "step": 27115
    },
    {
      "entropy": 5.9339954376220705,
      "epoch": 2.710951167091518,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004270810190953085,
      "loss": 5.4923,
      "mean_token_accuracy": 0.19113294333219527,
      "num_tokens": 53056098.0,
      "step": 27120
    },
    {
      "entropy": 5.954439687728882,
      "epoch": 2.7114509921527468,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004270547150638082,
      "loss": 5.516,
      "mean_token_accuracy": 0.1825221985578537,
      "num_tokens": 53065757.0,
      "step": 27125
    },
    {
      "entropy": 5.972350072860718,
      "epoch": 2.711950817213975,
      "grad_norm": 1.125,
      "learning_rate": 0.00042702840720665043,
      "loss": 5.5062,
      "mean_token_accuracy": 0.1932624489068985,
      "num_tokens": 53074616.0,
      "step": 27130
    },
    {
      "entropy": 6.023061370849609,
      "epoch": 2.7124506422752037,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042700209552449706,
      "loss": 5.6543,
      "mean_token_accuracy": 0.17162775993347168,
      "num_tokens": 53084861.0,
      "step": 27135
    },
    {
      "entropy": 6.126827812194824,
      "epoch": 2.7129504673364324,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042697578001801,
      "loss": 5.6654,
      "mean_token_accuracy": 0.16799379587173463,
      "num_tokens": 53095175.0,
      "step": 27140
    },
    {
      "entropy": 6.084599208831787,
      "epoch": 2.7134502923976607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004269494606878515,
      "loss": 5.6472,
      "mean_token_accuracy": 0.17867687940597535,
      "num_tokens": 53105094.0,
      "step": 27145
    },
    {
      "entropy": 6.070139741897583,
      "epoch": 2.7139501174588894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004269231375346836,
      "loss": 5.6416,
      "mean_token_accuracy": 0.1760774627327919,
      "num_tokens": 53115716.0,
      "step": 27150
    },
    {
      "entropy": 5.983975076675415,
      "epoch": 2.714449942520118,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004268968105591687,
      "loss": 5.5374,
      "mean_token_accuracy": 0.1832382619380951,
      "num_tokens": 53124375.0,
      "step": 27155
    },
    {
      "entropy": 5.863095283508301,
      "epoch": 2.7149497675813468,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004268704797619692,
      "loss": 5.4799,
      "mean_token_accuracy": 0.19058356881141664,
      "num_tokens": 53135220.0,
      "step": 27160
    },
    {
      "entropy": 5.91553692817688,
      "epoch": 2.715449592642575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042684414514374734,
      "loss": 5.5062,
      "mean_token_accuracy": 0.19097226858139038,
      "num_tokens": 53144424.0,
      "step": 27165
    },
    {
      "entropy": 5.98000316619873,
      "epoch": 2.7159494177038037,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00042681780670516595,
      "loss": 5.5678,
      "mean_token_accuracy": 0.18549496233463286,
      "num_tokens": 53154611.0,
      "step": 27170
    },
    {
      "entropy": 6.134168338775635,
      "epoch": 2.716449242765032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004267914644468877,
      "loss": 5.7243,
      "mean_token_accuracy": 0.17353716492652893,
      "num_tokens": 53163061.0,
      "step": 27175
    },
    {
      "entropy": 6.022347497940063,
      "epoch": 2.7169490678262607,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004267651183695752,
      "loss": 5.5357,
      "mean_token_accuracy": 0.18719310611486434,
      "num_tokens": 53173078.0,
      "step": 27180
    },
    {
      "entropy": 5.991395807266235,
      "epoch": 2.7174488928874894,
      "grad_norm": 1.25,
      "learning_rate": 0.0004267387684738913,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18540988117456436,
      "num_tokens": 53182957.0,
      "step": 27185
    },
    {
      "entropy": 6.060959386825561,
      "epoch": 2.717948717948718,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004267124147604991,
      "loss": 5.6219,
      "mean_token_accuracy": 0.17635377496480942,
      "num_tokens": 53193074.0,
      "step": 27190
    },
    {
      "entropy": 6.0154413223266605,
      "epoch": 2.7184485430099468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004266860572300615,
      "loss": 5.5866,
      "mean_token_accuracy": 0.18626046627759935,
      "num_tokens": 53202233.0,
      "step": 27195
    },
    {
      "entropy": 6.097850799560547,
      "epoch": 2.718948368071175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004266596958832419,
      "loss": 5.7423,
      "mean_token_accuracy": 0.17335643917322158,
      "num_tokens": 53212332.0,
      "step": 27200
    },
    {
      "entropy": 5.987348890304565,
      "epoch": 2.7194481931324037,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004266333307207032,
      "loss": 5.5815,
      "mean_token_accuracy": 0.18753986209630966,
      "num_tokens": 53222440.0,
      "step": 27205
    },
    {
      "entropy": 5.9790057182312015,
      "epoch": 2.719948018193632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042660696174310904,
      "loss": 5.5611,
      "mean_token_accuracy": 0.18253172039985657,
      "num_tokens": 53232129.0,
      "step": 27210
    },
    {
      "entropy": 6.054220724105835,
      "epoch": 2.7204478432548607,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004265805889511227,
      "loss": 5.6578,
      "mean_token_accuracy": 0.17718640267848967,
      "num_tokens": 53241678.0,
      "step": 27215
    },
    {
      "entropy": 6.017231559753418,
      "epoch": 2.7209476683160894,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004265542123454077,
      "loss": 5.635,
      "mean_token_accuracy": 0.17655487060546876,
      "num_tokens": 53250939.0,
      "step": 27220
    },
    {
      "entropy": 6.08200888633728,
      "epoch": 2.721447493377318,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042652783192662774,
      "loss": 5.6955,
      "mean_token_accuracy": 0.17173449397087098,
      "num_tokens": 53261058.0,
      "step": 27225
    },
    {
      "entropy": 6.067490530014038,
      "epoch": 2.7219473184385468,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004265014476954465,
      "loss": 5.5665,
      "mean_token_accuracy": 0.18569913357496262,
      "num_tokens": 53270934.0,
      "step": 27230
    },
    {
      "entropy": 5.972762966156006,
      "epoch": 2.722447143499775,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004264750596525279,
      "loss": 5.594,
      "mean_token_accuracy": 0.18594417572021485,
      "num_tokens": 53280593.0,
      "step": 27235
    },
    {
      "entropy": 5.998912572860718,
      "epoch": 2.7229469685610037,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00042644866779853577,
      "loss": 5.5998,
      "mean_token_accuracy": 0.18381264358758925,
      "num_tokens": 53289381.0,
      "step": 27240
    },
    {
      "entropy": 6.067308378219605,
      "epoch": 2.723446793622232,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042642227213413405,
      "loss": 5.7584,
      "mean_token_accuracy": 0.175156432390213,
      "num_tokens": 53298715.0,
      "step": 27245
    },
    {
      "entropy": 6.038973474502564,
      "epoch": 2.7239466186834607,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042639587265998696,
      "loss": 5.5964,
      "mean_token_accuracy": 0.18051807433366776,
      "num_tokens": 53309056.0,
      "step": 27250
    },
    {
      "entropy": 6.068113040924072,
      "epoch": 2.7244464437446894,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004263694693767587,
      "loss": 5.6356,
      "mean_token_accuracy": 0.18045537769794465,
      "num_tokens": 53319028.0,
      "step": 27255
    },
    {
      "entropy": 6.057782745361328,
      "epoch": 2.724946268805918,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004263430622851135,
      "loss": 5.6326,
      "mean_token_accuracy": 0.16959791034460067,
      "num_tokens": 53329516.0,
      "step": 27260
    },
    {
      "entropy": 6.084223508834839,
      "epoch": 2.7254460938671468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004263166513857158,
      "loss": 5.6641,
      "mean_token_accuracy": 0.17089378237724304,
      "num_tokens": 53341163.0,
      "step": 27265
    },
    {
      "entropy": 5.9875969886779785,
      "epoch": 2.725945918928375,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042629023667923016,
      "loss": 5.5435,
      "mean_token_accuracy": 0.18373292684555054,
      "num_tokens": 53350735.0,
      "step": 27270
    },
    {
      "entropy": 6.020624303817749,
      "epoch": 2.7264457439896037,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042626381816632096,
      "loss": 5.5311,
      "mean_token_accuracy": 0.19173386693000793,
      "num_tokens": 53361023.0,
      "step": 27275
    },
    {
      "entropy": 6.0022681713104244,
      "epoch": 2.726945569050832,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042623739584765315,
      "loss": 5.5966,
      "mean_token_accuracy": 0.17498223036527633,
      "num_tokens": 53370859.0,
      "step": 27280
    },
    {
      "entropy": 6.002542781829834,
      "epoch": 2.7274453941120607,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042621096972389136,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18992618471384048,
      "num_tokens": 53380610.0,
      "step": 27285
    },
    {
      "entropy": 6.031797933578491,
      "epoch": 2.7279452191732894,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042618453979570035,
      "loss": 5.7414,
      "mean_token_accuracy": 0.17168987840414046,
      "num_tokens": 53392497.0,
      "step": 27290
    },
    {
      "entropy": 6.016695737838745,
      "epoch": 2.728445044234518,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042615810606374535,
      "loss": 5.5864,
      "mean_token_accuracy": 0.1842956066131592,
      "num_tokens": 53402234.0,
      "step": 27295
    },
    {
      "entropy": 6.010934925079345,
      "epoch": 2.7289448692957468,
      "grad_norm": 1.125,
      "learning_rate": 0.0004261316685286913,
      "loss": 5.637,
      "mean_token_accuracy": 0.1797935500741005,
      "num_tokens": 53412430.0,
      "step": 27300
    },
    {
      "entropy": 6.053708791732788,
      "epoch": 2.729444694356975,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004261052271912033,
      "loss": 5.627,
      "mean_token_accuracy": 0.17945598363876342,
      "num_tokens": 53422574.0,
      "step": 27305
    },
    {
      "entropy": 5.958252048492431,
      "epoch": 2.7299445194182037,
      "grad_norm": 1.984375,
      "learning_rate": 0.0004260787820519467,
      "loss": 5.407,
      "mean_token_accuracy": 0.20558411926031112,
      "num_tokens": 53431322.0,
      "step": 27310
    },
    {
      "entropy": 5.958083868026733,
      "epoch": 2.730444344479432,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004260523331115868,
      "loss": 5.4461,
      "mean_token_accuracy": 0.18894680738449096,
      "num_tokens": 53440389.0,
      "step": 27315
    },
    {
      "entropy": 5.953938341140747,
      "epoch": 2.7309441695406607,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042602588037078903,
      "loss": 5.6348,
      "mean_token_accuracy": 0.18176087886095046,
      "num_tokens": 53449458.0,
      "step": 27320
    },
    {
      "entropy": 6.079331970214843,
      "epoch": 2.7314439946018894,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042599942383021906,
      "loss": 5.6376,
      "mean_token_accuracy": 0.17840943485498428,
      "num_tokens": 53459075.0,
      "step": 27325
    },
    {
      "entropy": 6.089920902252198,
      "epoch": 2.731943819663118,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004259729634905424,
      "loss": 5.6563,
      "mean_token_accuracy": 0.1795509934425354,
      "num_tokens": 53468660.0,
      "step": 27330
    },
    {
      "entropy": 5.99136872291565,
      "epoch": 2.7324436447243468,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042594649935242493,
      "loss": 5.5908,
      "mean_token_accuracy": 0.1872320994734764,
      "num_tokens": 53477663.0,
      "step": 27335
    },
    {
      "entropy": 6.020648527145386,
      "epoch": 2.732943469785575,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004259200314165322,
      "loss": 5.6132,
      "mean_token_accuracy": 0.18416111022233964,
      "num_tokens": 53487226.0,
      "step": 27340
    },
    {
      "entropy": 6.018085145950318,
      "epoch": 2.7334432948468037,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004258935596835303,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18396894484758378,
      "num_tokens": 53496483.0,
      "step": 27345
    },
    {
      "entropy": 6.071332359313965,
      "epoch": 2.733943119908032,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042586708415408547,
      "loss": 5.7025,
      "mean_token_accuracy": 0.17057203352451325,
      "num_tokens": 53506042.0,
      "step": 27350
    },
    {
      "entropy": 5.950396347045898,
      "epoch": 2.7344429449692607,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004258406048288635,
      "loss": 5.5218,
      "mean_token_accuracy": 0.18781934678554535,
      "num_tokens": 53516543.0,
      "step": 27355
    },
    {
      "entropy": 6.035513114929199,
      "epoch": 2.7349427700304894,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042581412170853073,
      "loss": 5.6795,
      "mean_token_accuracy": 0.18417606353759766,
      "num_tokens": 53527432.0,
      "step": 27360
    },
    {
      "entropy": 6.074202251434326,
      "epoch": 2.735442595091718,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004257876347937535,
      "loss": 5.6646,
      "mean_token_accuracy": 0.18017210960388183,
      "num_tokens": 53535930.0,
      "step": 27365
    },
    {
      "entropy": 5.916111183166504,
      "epoch": 2.7359424201529463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004257611440851981,
      "loss": 5.4916,
      "mean_token_accuracy": 0.19297810196876525,
      "num_tokens": 53545123.0,
      "step": 27370
    },
    {
      "entropy": 6.01715407371521,
      "epoch": 2.736442245214175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042573464958353116,
      "loss": 5.5787,
      "mean_token_accuracy": 0.1854477345943451,
      "num_tokens": 53554892.0,
      "step": 27375
    },
    {
      "entropy": 6.00073390007019,
      "epoch": 2.7369420702754037,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004257081512894192,
      "loss": 5.5093,
      "mean_token_accuracy": 0.19489142447710037,
      "num_tokens": 53564858.0,
      "step": 27380
    },
    {
      "entropy": 5.957695817947387,
      "epoch": 2.737441895336632,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042568164920352895,
      "loss": 5.4747,
      "mean_token_accuracy": 0.19229158461093904,
      "num_tokens": 53573750.0,
      "step": 27385
    },
    {
      "entropy": 5.841927099227905,
      "epoch": 2.7379417203978607,
      "grad_norm": 1.0,
      "learning_rate": 0.0004256551433265273,
      "loss": 5.427,
      "mean_token_accuracy": 0.20349105149507524,
      "num_tokens": 53584416.0,
      "step": 27390
    },
    {
      "entropy": 5.927203512191772,
      "epoch": 2.7384415454590894,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004256286336590808,
      "loss": 5.4959,
      "mean_token_accuracy": 0.18907767981290818,
      "num_tokens": 53593166.0,
      "step": 27395
    },
    {
      "entropy": 6.009900236129761,
      "epoch": 2.738941370520318,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004256021202018568,
      "loss": 5.6234,
      "mean_token_accuracy": 0.17576504051685332,
      "num_tokens": 53603025.0,
      "step": 27400
    },
    {
      "entropy": 6.0701700210571286,
      "epoch": 2.7394411955815463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004255756029555222,
      "loss": 5.5841,
      "mean_token_accuracy": 0.1801871046423912,
      "num_tokens": 53613266.0,
      "step": 27405
    },
    {
      "entropy": 5.938579559326172,
      "epoch": 2.739941020642775,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042554908192074406,
      "loss": 5.5407,
      "mean_token_accuracy": 0.19068458974361419,
      "num_tokens": 53622271.0,
      "step": 27410
    },
    {
      "entropy": 6.0222169876098635,
      "epoch": 2.7404408457040037,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042552255709818975,
      "loss": 5.6708,
      "mean_token_accuracy": 0.17701807916164397,
      "num_tokens": 53632096.0,
      "step": 27415
    },
    {
      "entropy": 6.0275609493255615,
      "epoch": 2.740940670765232,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042549602848852673,
      "loss": 5.5928,
      "mean_token_accuracy": 0.18312969505786897,
      "num_tokens": 53640875.0,
      "step": 27420
    },
    {
      "entropy": 5.960612487792969,
      "epoch": 2.7414404958264607,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004254694960924224,
      "loss": 5.5283,
      "mean_token_accuracy": 0.19398050755262375,
      "num_tokens": 53650186.0,
      "step": 27425
    },
    {
      "entropy": 5.940154552459717,
      "epoch": 2.7419403208876894,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004254429599105442,
      "loss": 5.5431,
      "mean_token_accuracy": 0.18482432514429092,
      "num_tokens": 53659608.0,
      "step": 27430
    },
    {
      "entropy": 6.058273267745972,
      "epoch": 2.742440145948918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004254164199435598,
      "loss": 5.5865,
      "mean_token_accuracy": 0.1770264208316803,
      "num_tokens": 53668479.0,
      "step": 27435
    },
    {
      "entropy": 6.010121583938599,
      "epoch": 2.7429399710101463,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004253898761921371,
      "loss": 5.5854,
      "mean_token_accuracy": 0.18623895943164825,
      "num_tokens": 53677496.0,
      "step": 27440
    },
    {
      "entropy": 6.054209136962891,
      "epoch": 2.743439796071375,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004253633286569437,
      "loss": 5.6507,
      "mean_token_accuracy": 0.17868286073207856,
      "num_tokens": 53686865.0,
      "step": 27445
    },
    {
      "entropy": 6.088141632080078,
      "epoch": 2.7439396211326037,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004253367773386476,
      "loss": 5.6393,
      "mean_token_accuracy": 0.1796304240822792,
      "num_tokens": 53696159.0,
      "step": 27450
    },
    {
      "entropy": 6.082489395141602,
      "epoch": 2.744439446193832,
      "grad_norm": 1.140625,
      "learning_rate": 0.000425310222237917,
      "loss": 5.5963,
      "mean_token_accuracy": 0.17227582782506942,
      "num_tokens": 53705512.0,
      "step": 27455
    },
    {
      "entropy": 6.002741479873658,
      "epoch": 2.7449392712550607,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042528366335541984,
      "loss": 5.6246,
      "mean_token_accuracy": 0.17753540575504304,
      "num_tokens": 53714361.0,
      "step": 27460
    },
    {
      "entropy": 6.032882976531982,
      "epoch": 2.7454390963162894,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004252571006918243,
      "loss": 5.5978,
      "mean_token_accuracy": 0.18241076916456223,
      "num_tokens": 53723606.0,
      "step": 27465
    },
    {
      "entropy": 6.020380067825317,
      "epoch": 2.745938921377518,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004252305342477989,
      "loss": 5.5514,
      "mean_token_accuracy": 0.1878909021615982,
      "num_tokens": 53734041.0,
      "step": 27470
    },
    {
      "entropy": 6.0489044189453125,
      "epoch": 2.7464387464387463,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042520396402401185,
      "loss": 5.5471,
      "mean_token_accuracy": 0.1853782892227173,
      "num_tokens": 53743735.0,
      "step": 27475
    },
    {
      "entropy": 6.018071699142456,
      "epoch": 2.746938571499975,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004251773900211317,
      "loss": 5.6348,
      "mean_token_accuracy": 0.18193268030881882,
      "num_tokens": 53753960.0,
      "step": 27480
    },
    {
      "entropy": 6.025861215591431,
      "epoch": 2.7474383965612037,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042515081223982716,
      "loss": 5.6475,
      "mean_token_accuracy": 0.18406011462211608,
      "num_tokens": 53764874.0,
      "step": 27485
    },
    {
      "entropy": 6.100893354415893,
      "epoch": 2.747938221622432,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042512423068076673,
      "loss": 5.6514,
      "mean_token_accuracy": 0.18138669133186341,
      "num_tokens": 53775153.0,
      "step": 27490
    },
    {
      "entropy": 6.020990037918091,
      "epoch": 2.7484380466836607,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004250976453446193,
      "loss": 5.4934,
      "mean_token_accuracy": 0.18941137492656707,
      "num_tokens": 53785091.0,
      "step": 27495
    },
    {
      "entropy": 5.994096803665161,
      "epoch": 2.7489378717448893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004250710562320538,
      "loss": 5.5027,
      "mean_token_accuracy": 0.1884840250015259,
      "num_tokens": 53794154.0,
      "step": 27500
    },
    {
      "entropy": 5.971716260910034,
      "epoch": 2.749437696806118,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042504446334373905,
      "loss": 5.5382,
      "mean_token_accuracy": 0.1826824054121971,
      "num_tokens": 53804129.0,
      "step": 27505
    },
    {
      "entropy": 6.001299953460693,
      "epoch": 2.7499375218673463,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004250178666803442,
      "loss": 5.6649,
      "mean_token_accuracy": 0.18248123526573182,
      "num_tokens": 53813081.0,
      "step": 27510
    },
    {
      "entropy": 6.018114328384399,
      "epoch": 2.750437346928575,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004249912662425385,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18710773140192033,
      "num_tokens": 53821949.0,
      "step": 27515
    },
    {
      "entropy": 6.090381717681884,
      "epoch": 2.7509371719898037,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042496466203099107,
      "loss": 5.5966,
      "mean_token_accuracy": 0.18027124553918839,
      "num_tokens": 53831232.0,
      "step": 27520
    },
    {
      "entropy": 6.058498668670654,
      "epoch": 2.751436997051032,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042493805404637136,
      "loss": 5.5799,
      "mean_token_accuracy": 0.1783789113163948,
      "num_tokens": 53841068.0,
      "step": 27525
    },
    {
      "entropy": 5.890639352798462,
      "epoch": 2.7519368221122606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042491144228934885,
      "loss": 5.4541,
      "mean_token_accuracy": 0.19586748480796815,
      "num_tokens": 53849874.0,
      "step": 27530
    },
    {
      "entropy": 5.9231809139251705,
      "epoch": 2.7524366471734893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000424884826760593,
      "loss": 5.4961,
      "mean_token_accuracy": 0.1936953544616699,
      "num_tokens": 53859173.0,
      "step": 27535
    },
    {
      "entropy": 6.016198682785034,
      "epoch": 2.752936472234718,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042485820746077346,
      "loss": 5.5722,
      "mean_token_accuracy": 0.1853718489408493,
      "num_tokens": 53869000.0,
      "step": 27540
    },
    {
      "entropy": 6.088893556594849,
      "epoch": 2.7534362972959463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042483158439055995,
      "loss": 5.6937,
      "mean_token_accuracy": 0.17470086216926575,
      "num_tokens": 53879941.0,
      "step": 27545
    },
    {
      "entropy": 5.972760105133057,
      "epoch": 2.753936122357175,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004248049575506224,
      "loss": 5.5211,
      "mean_token_accuracy": 0.18745189160108566,
      "num_tokens": 53889480.0,
      "step": 27550
    },
    {
      "entropy": 6.021461629867554,
      "epoch": 2.7544359474184037,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004247783269416306,
      "loss": 5.5918,
      "mean_token_accuracy": 0.18462428599596023,
      "num_tokens": 53898554.0,
      "step": 27555
    },
    {
      "entropy": 6.006372451782227,
      "epoch": 2.754935772479632,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042475169256425466,
      "loss": 5.5217,
      "mean_token_accuracy": 0.1879958063364029,
      "num_tokens": 53907898.0,
      "step": 27560
    },
    {
      "entropy": 6.063958215713501,
      "epoch": 2.7554355975408606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004247250544191647,
      "loss": 5.6526,
      "mean_token_accuracy": 0.17136678397655486,
      "num_tokens": 53917948.0,
      "step": 27565
    },
    {
      "entropy": 6.022657823562622,
      "epoch": 2.7559354226020893,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042469841250703086,
      "loss": 5.5821,
      "mean_token_accuracy": 0.18458817452192305,
      "num_tokens": 53927409.0,
      "step": 27570
    },
    {
      "entropy": 6.047533369064331,
      "epoch": 2.756435247663318,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004246717668285235,
      "loss": 5.6213,
      "mean_token_accuracy": 0.18155992776155472,
      "num_tokens": 53936600.0,
      "step": 27575
    },
    {
      "entropy": 6.017524576187133,
      "epoch": 2.7569350727245463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042464511738431296,
      "loss": 5.5609,
      "mean_token_accuracy": 0.19726441651582718,
      "num_tokens": 53946247.0,
      "step": 27580
    },
    {
      "entropy": 5.881282997131348,
      "epoch": 2.757434897785775,
      "grad_norm": 1.125,
      "learning_rate": 0.00042461846417506983,
      "loss": 5.4088,
      "mean_token_accuracy": 0.19651411473751068,
      "num_tokens": 53956009.0,
      "step": 27585
    },
    {
      "entropy": 6.064783716201783,
      "epoch": 2.7579347228470037,
      "grad_norm": 1.359375,
      "learning_rate": 0.0004245918072014646,
      "loss": 5.599,
      "mean_token_accuracy": 0.18396134972572326,
      "num_tokens": 53965549.0,
      "step": 27590
    },
    {
      "entropy": 6.031863975524902,
      "epoch": 2.758434547908232,
      "grad_norm": 1.078125,
      "learning_rate": 0.000424565146464168,
      "loss": 5.6126,
      "mean_token_accuracy": 0.17688046395778656,
      "num_tokens": 53975901.0,
      "step": 27595
    },
    {
      "entropy": 6.079634666442871,
      "epoch": 2.7589343729694606,
      "grad_norm": 1.046875,
      "learning_rate": 0.00042453848196385085,
      "loss": 5.653,
      "mean_token_accuracy": 0.17516692578792573,
      "num_tokens": 53985334.0,
      "step": 27600
    },
    {
      "entropy": 5.922613668441772,
      "epoch": 2.7594341980306893,
      "grad_norm": 1.25,
      "learning_rate": 0.00042451181370118394,
      "loss": 5.4727,
      "mean_token_accuracy": 0.19682121127843857,
      "num_tokens": 53995445.0,
      "step": 27605
    },
    {
      "entropy": 5.982703685760498,
      "epoch": 2.759934023091918,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004244851416768383,
      "loss": 5.5716,
      "mean_token_accuracy": 0.1887113556265831,
      "num_tokens": 54004845.0,
      "step": 27610
    },
    {
      "entropy": 5.995750761032104,
      "epoch": 2.7604338481531463,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004244584658914849,
      "loss": 5.5742,
      "mean_token_accuracy": 0.1909280464053154,
      "num_tokens": 54015074.0,
      "step": 27615
    },
    {
      "entropy": 6.000145769119262,
      "epoch": 2.760933673214375,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042443178634579497,
      "loss": 5.5901,
      "mean_token_accuracy": 0.1760549396276474,
      "num_tokens": 54025022.0,
      "step": 27620
    },
    {
      "entropy": 6.02004189491272,
      "epoch": 2.7614334982756037,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004244051030404398,
      "loss": 5.6104,
      "mean_token_accuracy": 0.1820294290781021,
      "num_tokens": 54034901.0,
      "step": 27625
    },
    {
      "entropy": 6.02276840209961,
      "epoch": 2.761933323336832,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042437841597609054,
      "loss": 5.6009,
      "mean_token_accuracy": 0.1790747120976448,
      "num_tokens": 54044032.0,
      "step": 27630
    },
    {
      "entropy": 5.994511985778809,
      "epoch": 2.7624331483980606,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004243517251534189,
      "loss": 5.5945,
      "mean_token_accuracy": 0.1817207843065262,
      "num_tokens": 54054635.0,
      "step": 27635
    },
    {
      "entropy": 6.002866315841675,
      "epoch": 2.7629329734592893,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004243250305730963,
      "loss": 5.5456,
      "mean_token_accuracy": 0.18994971215724946,
      "num_tokens": 54064658.0,
      "step": 27640
    },
    {
      "entropy": 6.036200380325317,
      "epoch": 2.763432798520518,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042429833223579436,
      "loss": 5.6563,
      "mean_token_accuracy": 0.18281290382146836,
      "num_tokens": 54074788.0,
      "step": 27645
    },
    {
      "entropy": 6.052637672424316,
      "epoch": 2.7639326235817463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042427163014218473,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18284720778465272,
      "num_tokens": 54084468.0,
      "step": 27650
    },
    {
      "entropy": 6.055946731567383,
      "epoch": 2.764432448642975,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042424492429293933,
      "loss": 5.6037,
      "mean_token_accuracy": 0.18308687061071396,
      "num_tokens": 54094223.0,
      "step": 27655
    },
    {
      "entropy": 6.03621187210083,
      "epoch": 2.7649322737042037,
      "grad_norm": 1.125,
      "learning_rate": 0.00042421821468873003,
      "loss": 5.5282,
      "mean_token_accuracy": 0.18118002414703369,
      "num_tokens": 54102721.0,
      "step": 27660
    },
    {
      "entropy": 6.017914199829102,
      "epoch": 2.765432098765432,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042419150133022883,
      "loss": 5.6381,
      "mean_token_accuracy": 0.17906796038150788,
      "num_tokens": 54111986.0,
      "step": 27665
    },
    {
      "entropy": 5.9548944473266605,
      "epoch": 2.7659319238266606,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004241647842181078,
      "loss": 5.536,
      "mean_token_accuracy": 0.1823952630162239,
      "num_tokens": 54121074.0,
      "step": 27670
    },
    {
      "entropy": 6.0230934619903564,
      "epoch": 2.7664317488878893,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00042413806335303935,
      "loss": 5.5942,
      "mean_token_accuracy": 0.17811220288276672,
      "num_tokens": 54131518.0,
      "step": 27675
    },
    {
      "entropy": 6.059011602401734,
      "epoch": 2.766931573949118,
      "grad_norm": 1.0625,
      "learning_rate": 0.00042411133873569544,
      "loss": 5.6469,
      "mean_token_accuracy": 0.18088393211364745,
      "num_tokens": 54141228.0,
      "step": 27680
    },
    {
      "entropy": 5.91871657371521,
      "epoch": 2.7674313990103463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004240846103667487,
      "loss": 5.5189,
      "mean_token_accuracy": 0.18601272851228715,
      "num_tokens": 54150365.0,
      "step": 27685
    },
    {
      "entropy": 6.034135913848877,
      "epoch": 2.767931224071575,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042405787824687153,
      "loss": 5.6129,
      "mean_token_accuracy": 0.1797341838479042,
      "num_tokens": 54159941.0,
      "step": 27690
    },
    {
      "entropy": 6.029040670394897,
      "epoch": 2.7684310491328032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004240311423767364,
      "loss": 5.5955,
      "mean_token_accuracy": 0.18315049558877944,
      "num_tokens": 54169613.0,
      "step": 27695
    },
    {
      "entropy": 5.9963521480560305,
      "epoch": 2.768930874194032,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004240044027570162,
      "loss": 5.5993,
      "mean_token_accuracy": 0.18479035496711732,
      "num_tokens": 54179927.0,
      "step": 27700
    },
    {
      "entropy": 5.975746154785156,
      "epoch": 2.7694306992552606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004239776593883835,
      "loss": 5.5291,
      "mean_token_accuracy": 0.186295685172081,
      "num_tokens": 54189572.0,
      "step": 27705
    },
    {
      "entropy": 6.066424465179443,
      "epoch": 2.7699305243164893,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042395091227151124,
      "loss": 5.6409,
      "mean_token_accuracy": 0.18009108304977417,
      "num_tokens": 54200561.0,
      "step": 27710
    },
    {
      "entropy": 6.119863224029541,
      "epoch": 2.770430349377718,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042392416140707233,
      "loss": 5.6918,
      "mean_token_accuracy": 0.18286965787410736,
      "num_tokens": 54210519.0,
      "step": 27715
    },
    {
      "entropy": 5.947615242004394,
      "epoch": 2.7709301744389463,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004238974067957399,
      "loss": 5.4418,
      "mean_token_accuracy": 0.19199668020009994,
      "num_tokens": 54219142.0,
      "step": 27720
    },
    {
      "entropy": 6.0093316555023195,
      "epoch": 2.771429999500175,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000423870648438187,
      "loss": 5.5709,
      "mean_token_accuracy": 0.17600685954093934,
      "num_tokens": 54228728.0,
      "step": 27725
    },
    {
      "entropy": 6.07203688621521,
      "epoch": 2.7719298245614032,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004238438863350869,
      "loss": 5.5768,
      "mean_token_accuracy": 0.18053983747959138,
      "num_tokens": 54237767.0,
      "step": 27730
    },
    {
      "entropy": 6.012223482131958,
      "epoch": 2.772429649622632,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004238171204871129,
      "loss": 5.5272,
      "mean_token_accuracy": 0.1877514824271202,
      "num_tokens": 54247453.0,
      "step": 27735
    },
    {
      "entropy": 6.033645963668823,
      "epoch": 2.7729294746838606,
      "grad_norm": 1.328125,
      "learning_rate": 0.00042379035089493844,
      "loss": 5.6369,
      "mean_token_accuracy": 0.18506467193365098,
      "num_tokens": 54256398.0,
      "step": 27740
    },
    {
      "entropy": 5.978144407272339,
      "epoch": 2.7734292997450893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004237635775592371,
      "loss": 5.6365,
      "mean_token_accuracy": 0.17690690457820893,
      "num_tokens": 54265981.0,
      "step": 27745
    },
    {
      "entropy": 5.932096338272094,
      "epoch": 2.773929124806318,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004237368004806823,
      "loss": 5.5348,
      "mean_token_accuracy": 0.19090358316898345,
      "num_tokens": 54277285.0,
      "step": 27750
    },
    {
      "entropy": 6.051091146469116,
      "epoch": 2.7744289498675463,
      "grad_norm": 1.125,
      "learning_rate": 0.000423710019659948,
      "loss": 5.4994,
      "mean_token_accuracy": 0.19150387793779372,
      "num_tokens": 54286211.0,
      "step": 27755
    },
    {
      "entropy": 6.003693723678589,
      "epoch": 2.774928774928775,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042368323509770785,
      "loss": 5.5925,
      "mean_token_accuracy": 0.1879944086074829,
      "num_tokens": 54296101.0,
      "step": 27760
    },
    {
      "entropy": 6.010381650924683,
      "epoch": 2.7754285999900032,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004236564467946357,
      "loss": 5.5873,
      "mean_token_accuracy": 0.1858461707830429,
      "num_tokens": 54306027.0,
      "step": 27765
    },
    {
      "entropy": 6.0117754459381105,
      "epoch": 2.775928425051232,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004236296547514056,
      "loss": 5.5303,
      "mean_token_accuracy": 0.18937959969043733,
      "num_tokens": 54315551.0,
      "step": 27770
    },
    {
      "entropy": 6.090894794464111,
      "epoch": 2.7764282501124606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004236028589686917,
      "loss": 5.6456,
      "mean_token_accuracy": 0.1778438925743103,
      "num_tokens": 54325312.0,
      "step": 27775
    },
    {
      "entropy": 6.026579523086548,
      "epoch": 2.7769280751736893,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042357605944716816,
      "loss": 5.6204,
      "mean_token_accuracy": 0.1783723771572113,
      "num_tokens": 54335854.0,
      "step": 27780
    },
    {
      "entropy": 6.053278303146362,
      "epoch": 2.777427900234918,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042354925618750904,
      "loss": 5.5095,
      "mean_token_accuracy": 0.1843124583363533,
      "num_tokens": 54345887.0,
      "step": 27785
    },
    {
      "entropy": 6.081729984283447,
      "epoch": 2.7779277252961463,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042352244919038895,
      "loss": 5.6185,
      "mean_token_accuracy": 0.1848454251885414,
      "num_tokens": 54355367.0,
      "step": 27790
    },
    {
      "entropy": 6.009240674972534,
      "epoch": 2.778427550357375,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042349563845648226,
      "loss": 5.5852,
      "mean_token_accuracy": 0.18463554084300995,
      "num_tokens": 54365689.0,
      "step": 27795
    },
    {
      "entropy": 6.043126487731934,
      "epoch": 2.7789273754186032,
      "grad_norm": 1.125,
      "learning_rate": 0.0004234688239864634,
      "loss": 5.6342,
      "mean_token_accuracy": 0.17924828082323074,
      "num_tokens": 54375317.0,
      "step": 27800
    },
    {
      "entropy": 6.013046455383301,
      "epoch": 2.779427200479832,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004234420057810073,
      "loss": 5.56,
      "mean_token_accuracy": 0.17965205609798432,
      "num_tokens": 54384926.0,
      "step": 27805
    },
    {
      "entropy": 6.059171056747436,
      "epoch": 2.7799270255410606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042341518384078847,
      "loss": 5.6549,
      "mean_token_accuracy": 0.17446608543395997,
      "num_tokens": 54394691.0,
      "step": 27810
    },
    {
      "entropy": 6.072454309463501,
      "epoch": 2.7804268506022893,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004233883581664818,
      "loss": 5.6807,
      "mean_token_accuracy": 0.17811790555715562,
      "num_tokens": 54405264.0,
      "step": 27815
    },
    {
      "entropy": 6.07448296546936,
      "epoch": 2.780926675663518,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004233615287587622,
      "loss": 5.6033,
      "mean_token_accuracy": 0.18048861175775527,
      "num_tokens": 54415477.0,
      "step": 27820
    },
    {
      "entropy": 6.060410690307617,
      "epoch": 2.7814265007247463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042333469561830475,
      "loss": 5.7084,
      "mean_token_accuracy": 0.17407267093658446,
      "num_tokens": 54425715.0,
      "step": 27825
    },
    {
      "entropy": 6.1172572612762455,
      "epoch": 2.781926325785975,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004233078587457845,
      "loss": 5.63,
      "mean_token_accuracy": 0.17979361414909362,
      "num_tokens": 54435097.0,
      "step": 27830
    },
    {
      "entropy": 6.0340522766113285,
      "epoch": 2.7824261508472032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004232810181418767,
      "loss": 5.6498,
      "mean_token_accuracy": 0.17559524327516557,
      "num_tokens": 54445511.0,
      "step": 27835
    },
    {
      "entropy": 6.010582971572876,
      "epoch": 2.782925975908432,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004232541738072566,
      "loss": 5.58,
      "mean_token_accuracy": 0.18093815296888352,
      "num_tokens": 54456463.0,
      "step": 27840
    },
    {
      "entropy": 5.917420387268066,
      "epoch": 2.7834258009696606,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042322732574259976,
      "loss": 5.4435,
      "mean_token_accuracy": 0.19813568741083146,
      "num_tokens": 54466262.0,
      "step": 27845
    },
    {
      "entropy": 6.0742157936096195,
      "epoch": 2.7839256260308893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042320047394858143,
      "loss": 5.6077,
      "mean_token_accuracy": 0.177427276968956,
      "num_tokens": 54476270.0,
      "step": 27850
    },
    {
      "entropy": 5.992970705032349,
      "epoch": 2.784425451092118,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004231736184258773,
      "loss": 5.6071,
      "mean_token_accuracy": 0.17857687324285507,
      "num_tokens": 54486803.0,
      "step": 27855
    },
    {
      "entropy": 6.1002429008483885,
      "epoch": 2.7849252761533463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004231467591751631,
      "loss": 5.6094,
      "mean_token_accuracy": 0.17805061042308806,
      "num_tokens": 54495525.0,
      "step": 27860
    },
    {
      "entropy": 5.924673986434937,
      "epoch": 2.785425101214575,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004231198961971145,
      "loss": 5.4844,
      "mean_token_accuracy": 0.19135912358760834,
      "num_tokens": 54506335.0,
      "step": 27865
    },
    {
      "entropy": 5.977017116546631,
      "epoch": 2.7859249262758032,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004230930294924075,
      "loss": 5.583,
      "mean_token_accuracy": 0.18913626223802565,
      "num_tokens": 54516505.0,
      "step": 27870
    },
    {
      "entropy": 6.100473546981812,
      "epoch": 2.786424751337032,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042306615906171794,
      "loss": 5.6367,
      "mean_token_accuracy": 0.1759947955608368,
      "num_tokens": 54525407.0,
      "step": 27875
    },
    {
      "entropy": 5.997759103775024,
      "epoch": 2.7869245763982606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000423039284905722,
      "loss": 5.5671,
      "mean_token_accuracy": 0.1867338851094246,
      "num_tokens": 54535750.0,
      "step": 27880
    },
    {
      "entropy": 6.044288778305054,
      "epoch": 2.7874244014594893,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00042301240702509576,
      "loss": 5.5649,
      "mean_token_accuracy": 0.185624460875988,
      "num_tokens": 54544284.0,
      "step": 27885
    },
    {
      "entropy": 5.965339517593383,
      "epoch": 2.7879242265207176,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004229855254205154,
      "loss": 5.4598,
      "mean_token_accuracy": 0.1978113517165184,
      "num_tokens": 54553168.0,
      "step": 27890
    },
    {
      "entropy": 6.020461320877075,
      "epoch": 2.7884240515819463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042295864009265723,
      "loss": 5.5655,
      "mean_token_accuracy": 0.18420883864164353,
      "num_tokens": 54563553.0,
      "step": 27895
    },
    {
      "entropy": 5.969381046295166,
      "epoch": 2.788923876643175,
      "grad_norm": 1.1875,
      "learning_rate": 0.00042293175104219776,
      "loss": 5.5411,
      "mean_token_accuracy": 0.1885160356760025,
      "num_tokens": 54573195.0,
      "step": 27900
    },
    {
      "entropy": 5.896646308898926,
      "epoch": 2.7894237017044032,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042290485826981355,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19725273698568344,
      "num_tokens": 54583417.0,
      "step": 27905
    },
    {
      "entropy": 6.0664534091949465,
      "epoch": 2.789923526765632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00042287796177618115,
      "loss": 5.6368,
      "mean_token_accuracy": 0.18244190514087677,
      "num_tokens": 54592961.0,
      "step": 27910
    },
    {
      "entropy": 6.01856164932251,
      "epoch": 2.7904233518268606,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042285106156197727,
      "loss": 5.5637,
      "mean_token_accuracy": 0.18064869046211243,
      "num_tokens": 54602161.0,
      "step": 27915
    },
    {
      "entropy": 6.060322093963623,
      "epoch": 2.7909231768880893,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004228241576278787,
      "loss": 5.6717,
      "mean_token_accuracy": 0.1794242262840271,
      "num_tokens": 54612816.0,
      "step": 27920
    },
    {
      "entropy": 6.004883050918579,
      "epoch": 2.7914230019493176,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042279724997456246,
      "loss": 5.5515,
      "mean_token_accuracy": 0.186871500313282,
      "num_tokens": 54622808.0,
      "step": 27925
    },
    {
      "entropy": 6.021888542175293,
      "epoch": 2.7919228270105463,
      "grad_norm": 1.453125,
      "learning_rate": 0.00042277033860270544,
      "loss": 5.5676,
      "mean_token_accuracy": 0.18442469984292983,
      "num_tokens": 54633316.0,
      "step": 27930
    },
    {
      "entropy": 5.9878771781921385,
      "epoch": 2.792422652071775,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042274342351298467,
      "loss": 5.5295,
      "mean_token_accuracy": 0.18753096908330918,
      "num_tokens": 54642655.0,
      "step": 27935
    },
    {
      "entropy": 5.975928878784179,
      "epoch": 2.7929224771330032,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004227165047060773,
      "loss": 5.5595,
      "mean_token_accuracy": 0.18920902162790298,
      "num_tokens": 54652390.0,
      "step": 27940
    },
    {
      "entropy": 6.019753503799438,
      "epoch": 2.793422302194232,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004226895821826608,
      "loss": 5.6208,
      "mean_token_accuracy": 0.17861968725919725,
      "num_tokens": 54661525.0,
      "step": 27945
    },
    {
      "entropy": 6.074170160293579,
      "epoch": 2.7939221272554606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004226626559434124,
      "loss": 5.5612,
      "mean_token_accuracy": 0.18532527834177018,
      "num_tokens": 54670416.0,
      "step": 27950
    },
    {
      "entropy": 5.960095357894898,
      "epoch": 2.7944219523166893,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042263572598900956,
      "loss": 5.5664,
      "mean_token_accuracy": 0.18352464884519576,
      "num_tokens": 54680025.0,
      "step": 27955
    },
    {
      "entropy": 5.966314220428467,
      "epoch": 2.7949217773779176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042260879232012985,
      "loss": 5.6103,
      "mean_token_accuracy": 0.17926261574029922,
      "num_tokens": 54689591.0,
      "step": 27960
    },
    {
      "entropy": 6.012644672393799,
      "epoch": 2.7954216024391463,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042258185493745086,
      "loss": 5.5809,
      "mean_token_accuracy": 0.18528272956609726,
      "num_tokens": 54699521.0,
      "step": 27965
    },
    {
      "entropy": 5.987171840667725,
      "epoch": 2.795921427500375,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004225549138416504,
      "loss": 5.4841,
      "mean_token_accuracy": 0.19158498495817183,
      "num_tokens": 54710159.0,
      "step": 27970
    },
    {
      "entropy": 6.0290771484375,
      "epoch": 2.7964212525616032,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004225279690334063,
      "loss": 5.5351,
      "mean_token_accuracy": 0.18951656371355058,
      "num_tokens": 54719657.0,
      "step": 27975
    },
    {
      "entropy": 6.07677583694458,
      "epoch": 2.796921077622832,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042250102051339647,
      "loss": 5.6837,
      "mean_token_accuracy": 0.187198406457901,
      "num_tokens": 54730681.0,
      "step": 27980
    },
    {
      "entropy": 6.075413656234741,
      "epoch": 2.7974209026840606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042247406828229897,
      "loss": 5.6449,
      "mean_token_accuracy": 0.17425135374069214,
      "num_tokens": 54740606.0,
      "step": 27985
    },
    {
      "entropy": 5.951979112625122,
      "epoch": 2.7979207277452893,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004224471123407917,
      "loss": 5.3813,
      "mean_token_accuracy": 0.20599449574947357,
      "num_tokens": 54750233.0,
      "step": 27990
    },
    {
      "entropy": 6.058263349533081,
      "epoch": 2.7984205528065176,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004224201526895531,
      "loss": 5.7411,
      "mean_token_accuracy": 0.16788301467895508,
      "num_tokens": 54760287.0,
      "step": 27995
    },
    {
      "entropy": 6.061223459243775,
      "epoch": 2.7989203778677463,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042239318932926144,
      "loss": 5.5657,
      "mean_token_accuracy": 0.18387955725193023,
      "num_tokens": 54770287.0,
      "step": 28000
    },
    {
      "entropy": 6.0317626953125,
      "epoch": 2.799420202928975,
      "grad_norm": 1.1875,
      "learning_rate": 0.000422366222260595,
      "loss": 5.6096,
      "mean_token_accuracy": 0.1811448737978935,
      "num_tokens": 54779780.0,
      "step": 28005
    },
    {
      "entropy": 5.98747410774231,
      "epoch": 2.7999200279902032,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042233925148423235,
      "loss": 5.4959,
      "mean_token_accuracy": 0.19672053456306457,
      "num_tokens": 54789628.0,
      "step": 28010
    },
    {
      "entropy": 5.941774511337281,
      "epoch": 2.800419853051432,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004223122770008521,
      "loss": 5.5074,
      "mean_token_accuracy": 0.19365909844636917,
      "num_tokens": 54799228.0,
      "step": 28015
    },
    {
      "entropy": 6.006224584579468,
      "epoch": 2.8009196781126606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042228529881113285,
      "loss": 5.54,
      "mean_token_accuracy": 0.1858419731259346,
      "num_tokens": 54809666.0,
      "step": 28020
    },
    {
      "entropy": 6.056027555465699,
      "epoch": 2.8014195031738893,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042225831691575336,
      "loss": 5.5386,
      "mean_token_accuracy": 0.18815133422613145,
      "num_tokens": 54818948.0,
      "step": 28025
    },
    {
      "entropy": 6.0163429260253904,
      "epoch": 2.8019193282351176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004222313313153925,
      "loss": 5.6243,
      "mean_token_accuracy": 0.183944371342659,
      "num_tokens": 54830091.0,
      "step": 28030
    },
    {
      "entropy": 5.996781539916992,
      "epoch": 2.8024191532963463,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042220434201072917,
      "loss": 5.5603,
      "mean_token_accuracy": 0.19026835411787033,
      "num_tokens": 54839385.0,
      "step": 28035
    },
    {
      "entropy": 6.04871244430542,
      "epoch": 2.802918978357575,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004221773490024426,
      "loss": 5.6102,
      "mean_token_accuracy": 0.18257113099098204,
      "num_tokens": 54850034.0,
      "step": 28040
    },
    {
      "entropy": 6.077244091033935,
      "epoch": 2.8034188034188032,
      "grad_norm": 1.125,
      "learning_rate": 0.0004221503522912117,
      "loss": 5.6953,
      "mean_token_accuracy": 0.17234439104795457,
      "num_tokens": 54859522.0,
      "step": 28045
    },
    {
      "entropy": 5.986248016357422,
      "epoch": 2.803918628480032,
      "grad_norm": 1.125,
      "learning_rate": 0.00042212335187771587,
      "loss": 5.5408,
      "mean_token_accuracy": 0.1872730329632759,
      "num_tokens": 54868938.0,
      "step": 28050
    },
    {
      "entropy": 5.959762477874756,
      "epoch": 2.8044184535412606,
      "grad_norm": 1.25,
      "learning_rate": 0.00042209634776263434,
      "loss": 5.4558,
      "mean_token_accuracy": 0.1910246342420578,
      "num_tokens": 54878125.0,
      "step": 28055
    },
    {
      "entropy": 6.07199444770813,
      "epoch": 2.8049182786024893,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004220693399466466,
      "loss": 5.6287,
      "mean_token_accuracy": 0.17363103181123735,
      "num_tokens": 54886615.0,
      "step": 28060
    },
    {
      "entropy": 6.062364006042481,
      "epoch": 2.8054181036637176,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042204232843043207,
      "loss": 5.6721,
      "mean_token_accuracy": 0.1788029670715332,
      "num_tokens": 54897316.0,
      "step": 28065
    },
    {
      "entropy": 6.035115051269531,
      "epoch": 2.8059179287249463,
      "grad_norm": 1.21875,
      "learning_rate": 0.00042201531321467027,
      "loss": 5.5351,
      "mean_token_accuracy": 0.18434848338365556,
      "num_tokens": 54906672.0,
      "step": 28070
    },
    {
      "entropy": 6.0549492835998535,
      "epoch": 2.806417753786175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00042198829430004114,
      "loss": 5.6087,
      "mean_token_accuracy": 0.1766224667429924,
      "num_tokens": 54916465.0,
      "step": 28075
    },
    {
      "entropy": 5.97222638130188,
      "epoch": 2.8069175788474032,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00042196127168722436,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18027301877737045,
      "num_tokens": 54926655.0,
      "step": 28080
    },
    {
      "entropy": 6.010865688323975,
      "epoch": 2.807417403908632,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004219342453768998,
      "loss": 5.5463,
      "mean_token_accuracy": 0.18402189761400223,
      "num_tokens": 54936151.0,
      "step": 28085
    },
    {
      "entropy": 6.003515291213989,
      "epoch": 2.8079172289698606,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004219072153697474,
      "loss": 5.5558,
      "mean_token_accuracy": 0.1858503669500351,
      "num_tokens": 54946303.0,
      "step": 28090
    },
    {
      "entropy": 5.979242753982544,
      "epoch": 2.8084170540310893,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004218801816664473,
      "loss": 5.5846,
      "mean_token_accuracy": 0.18515522480010987,
      "num_tokens": 54956078.0,
      "step": 28095
    },
    {
      "entropy": 5.971443462371826,
      "epoch": 2.8089168790923176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004218531442676796,
      "loss": 5.6002,
      "mean_token_accuracy": 0.18520632535219192,
      "num_tokens": 54965851.0,
      "step": 28100
    },
    {
      "entropy": 6.105798816680908,
      "epoch": 2.8094167041535463,
      "grad_norm": 1.125,
      "learning_rate": 0.0004218261031741246,
      "loss": 5.6624,
      "mean_token_accuracy": 0.17750029116868973,
      "num_tokens": 54975381.0,
      "step": 28105
    },
    {
      "entropy": 6.034116888046265,
      "epoch": 2.809916529214775,
      "grad_norm": 1.15625,
      "learning_rate": 0.00042179905838646254,
      "loss": 5.5587,
      "mean_token_accuracy": 0.18979046046733855,
      "num_tokens": 54986192.0,
      "step": 28110
    },
    {
      "entropy": 6.0397562980651855,
      "epoch": 2.8104163542760032,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000421772009905374,
      "loss": 5.598,
      "mean_token_accuracy": 0.1855422616004944,
      "num_tokens": 54996093.0,
      "step": 28115
    },
    {
      "entropy": 6.048256349563599,
      "epoch": 2.810916179337232,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004217449577315395,
      "loss": 5.5656,
      "mean_token_accuracy": 0.18815297782421112,
      "num_tokens": 55005284.0,
      "step": 28120
    },
    {
      "entropy": 6.061623001098633,
      "epoch": 2.8114160043984606,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004217179018656396,
      "loss": 5.6247,
      "mean_token_accuracy": 0.1812191888689995,
      "num_tokens": 55014693.0,
      "step": 28125
    },
    {
      "entropy": 5.9792969703674315,
      "epoch": 2.8119158294596893,
      "grad_norm": 1.3046875,
      "learning_rate": 0.000421690842308355,
      "loss": 5.519,
      "mean_token_accuracy": 0.18898403495550156,
      "num_tokens": 55024354.0,
      "step": 28130
    },
    {
      "entropy": 5.8983197689056395,
      "epoch": 2.8124156545209176,
      "grad_norm": 1.125,
      "learning_rate": 0.0004216637790603667,
      "loss": 5.4284,
      "mean_token_accuracy": 0.19604831486940383,
      "num_tokens": 55033590.0,
      "step": 28135
    },
    {
      "entropy": 5.964046382904053,
      "epoch": 2.8129154795821463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004216367121223553,
      "loss": 5.5024,
      "mean_token_accuracy": 0.1881198689341545,
      "num_tokens": 55042639.0,
      "step": 28140
    },
    {
      "entropy": 6.024931240081787,
      "epoch": 2.813415304643375,
      "grad_norm": 1.0625,
      "learning_rate": 0.000421609641495002,
      "loss": 5.583,
      "mean_token_accuracy": 0.18375220745801926,
      "num_tokens": 55053825.0,
      "step": 28145
    },
    {
      "entropy": 5.957256555557251,
      "epoch": 2.8139151297046032,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042158256717898785,
      "loss": 5.524,
      "mean_token_accuracy": 0.18842488676309585,
      "num_tokens": 55063142.0,
      "step": 28150
    },
    {
      "entropy": 5.989586973190308,
      "epoch": 2.814414954765832,
      "grad_norm": 1.0625,
      "learning_rate": 0.000421555489174994,
      "loss": 5.5605,
      "mean_token_accuracy": 0.19287996292114257,
      "num_tokens": 55073122.0,
      "step": 28155
    },
    {
      "entropy": 6.075976800918579,
      "epoch": 2.8149147798270606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00042152840748370184,
      "loss": 5.6447,
      "mean_token_accuracy": 0.17689227908849717,
      "num_tokens": 55083364.0,
      "step": 28160
    },
    {
      "entropy": 6.063017416000366,
      "epoch": 2.8154146048882893,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004215013221057926,
      "loss": 5.6111,
      "mean_token_accuracy": 0.18050871938467025,
      "num_tokens": 55092540.0,
      "step": 28165
    },
    {
      "entropy": 6.05877275466919,
      "epoch": 2.8159144299495176,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004214742330419478,
      "loss": 5.6743,
      "mean_token_accuracy": 0.17534727305173875,
      "num_tokens": 55103085.0,
      "step": 28170
    },
    {
      "entropy": 6.074978494644165,
      "epoch": 2.8164142550107463,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00042144714029284893,
      "loss": 5.6637,
      "mean_token_accuracy": 0.17925862073898316,
      "num_tokens": 55112962.0,
      "step": 28175
    },
    {
      "entropy": 5.930502939224243,
      "epoch": 2.816914080071975,
      "grad_norm": 1.125,
      "learning_rate": 0.00042142004385917783,
      "loss": 5.4537,
      "mean_token_accuracy": 0.19370130002498626,
      "num_tokens": 55122795.0,
      "step": 28180
    },
    {
      "entropy": 5.854910469055175,
      "epoch": 2.8174139051332032,
      "grad_norm": 1.9453125,
      "learning_rate": 0.000421392943741616,
      "loss": 5.487,
      "mean_token_accuracy": 0.19756831973791122,
      "num_tokens": 55133082.0,
      "step": 28185
    },
    {
      "entropy": 6.062303161621093,
      "epoch": 2.817913730194432,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004213658399408454,
      "loss": 5.585,
      "mean_token_accuracy": 0.1882267937064171,
      "num_tokens": 55142940.0,
      "step": 28190
    },
    {
      "entropy": 6.22761754989624,
      "epoch": 2.8184135552556606,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004213387324575479,
      "loss": 5.7493,
      "mean_token_accuracy": 0.1718323349952698,
      "num_tokens": 55152233.0,
      "step": 28195
    },
    {
      "entropy": 6.134977054595947,
      "epoch": 2.8189133803168893,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004213116212924056,
      "loss": 5.6426,
      "mean_token_accuracy": 0.18438841700553893,
      "num_tokens": 55163159.0,
      "step": 28200
    },
    {
      "entropy": 5.97796483039856,
      "epoch": 2.8194132053781176,
      "grad_norm": 1.28125,
      "learning_rate": 0.00042128450644610046,
      "loss": 5.5265,
      "mean_token_accuracy": 0.19111181199550628,
      "num_tokens": 55173902.0,
      "step": 28205
    },
    {
      "entropy": 5.967923879623413,
      "epoch": 2.8199130304393463,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042125738791931486,
      "loss": 5.5626,
      "mean_token_accuracy": 0.18753696233034134,
      "num_tokens": 55184698.0,
      "step": 28210
    },
    {
      "entropy": 6.017305088043213,
      "epoch": 2.8204128555005745,
      "grad_norm": 1.015625,
      "learning_rate": 0.000421230265712731,
      "loss": 5.6324,
      "mean_token_accuracy": 0.1737662822008133,
      "num_tokens": 55194463.0,
      "step": 28215
    },
    {
      "entropy": 5.984061813354492,
      "epoch": 2.820912680561803,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004212031398270313,
      "loss": 5.6184,
      "mean_token_accuracy": 0.1885695993900299,
      "num_tokens": 55205844.0,
      "step": 28220
    },
    {
      "entropy": 6.098659706115723,
      "epoch": 2.821412505623032,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042117601026289827,
      "loss": 5.6604,
      "mean_token_accuracy": 0.18004756569862365,
      "num_tokens": 55215760.0,
      "step": 28225
    },
    {
      "entropy": 6.064934635162354,
      "epoch": 2.8219123306842606,
      "grad_norm": 1.015625,
      "learning_rate": 0.00042114887702101434,
      "loss": 5.575,
      "mean_token_accuracy": 0.18543704599142075,
      "num_tokens": 55226182.0,
      "step": 28230
    },
    {
      "entropy": 6.005737972259522,
      "epoch": 2.8224121557454893,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00042112174010206226,
      "loss": 5.6107,
      "mean_token_accuracy": 0.17432308495044707,
      "num_tokens": 55236200.0,
      "step": 28235
    },
    {
      "entropy": 6.065285015106201,
      "epoch": 2.8229119808067176,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00042109459950672487,
      "loss": 5.613,
      "mean_token_accuracy": 0.18020668923854827,
      "num_tokens": 55245455.0,
      "step": 28240
    },
    {
      "entropy": 6.078323268890381,
      "epoch": 2.8234118058679463,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042106745523568494,
      "loss": 5.6072,
      "mean_token_accuracy": 0.18309133499860764,
      "num_tokens": 55254555.0,
      "step": 28245
    },
    {
      "entropy": 6.097473812103272,
      "epoch": 2.8239116309291745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042104030728962535,
      "loss": 5.6494,
      "mean_token_accuracy": 0.18647712022066115,
      "num_tokens": 55262897.0,
      "step": 28250
    },
    {
      "entropy": 5.981614971160889,
      "epoch": 2.824411455990403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042101315566922927,
      "loss": 5.6061,
      "mean_token_accuracy": 0.18715454190969466,
      "num_tokens": 55272715.0,
      "step": 28255
    },
    {
      "entropy": 6.0277454376220705,
      "epoch": 2.824911281051632,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00042098600037517985,
      "loss": 5.5471,
      "mean_token_accuracy": 0.18292974829673767,
      "num_tokens": 55282281.0,
      "step": 28260
    },
    {
      "entropy": 6.017353534698486,
      "epoch": 2.8254111061128606,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004209588414081601,
      "loss": 5.5539,
      "mean_token_accuracy": 0.18410571813583373,
      "num_tokens": 55292708.0,
      "step": 28265
    },
    {
      "entropy": 5.901357269287109,
      "epoch": 2.8259109311740893,
      "grad_norm": 1.375,
      "learning_rate": 0.00042093167876885345,
      "loss": 5.5575,
      "mean_token_accuracy": 0.18468155860900878,
      "num_tokens": 55302750.0,
      "step": 28270
    },
    {
      "entropy": 6.003642320632935,
      "epoch": 2.8264107562353176,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004209045124579434,
      "loss": 5.5719,
      "mean_token_accuracy": 0.186366006731987,
      "num_tokens": 55312136.0,
      "step": 28275
    },
    {
      "entropy": 6.0965108394622805,
      "epoch": 2.8269105812965463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00042087734247611333,
      "loss": 5.5837,
      "mean_token_accuracy": 0.18688869178295137,
      "num_tokens": 55321770.0,
      "step": 28280
    },
    {
      "entropy": 5.927344942092896,
      "epoch": 2.8274104063577745,
      "grad_norm": 1.125,
      "learning_rate": 0.0004208501688240468,
      "loss": 5.5369,
      "mean_token_accuracy": 0.18252337872982025,
      "num_tokens": 55331758.0,
      "step": 28285
    },
    {
      "entropy": 6.007903909683227,
      "epoch": 2.827910231419003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042082299150242764,
      "loss": 5.5915,
      "mean_token_accuracy": 0.18374226242303848,
      "num_tokens": 55341879.0,
      "step": 28290
    },
    {
      "entropy": 5.934148645401001,
      "epoch": 2.828410056480232,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004207958105119395,
      "loss": 5.418,
      "mean_token_accuracy": 0.19988171458244325,
      "num_tokens": 55351313.0,
      "step": 28295
    },
    {
      "entropy": 6.029666900634766,
      "epoch": 2.8289098815414606,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00042076862585326626,
      "loss": 5.5307,
      "mean_token_accuracy": 0.1941484272480011,
      "num_tokens": 55360838.0,
      "step": 28300
    },
    {
      "entropy": 6.044083547592163,
      "epoch": 2.8294097066026893,
      "grad_norm": 1.109375,
      "learning_rate": 0.00042074143752709195,
      "loss": 5.6479,
      "mean_token_accuracy": 0.18610639721155167,
      "num_tokens": 55370698.0,
      "step": 28305
    },
    {
      "entropy": 6.05050630569458,
      "epoch": 2.8299095316639176,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004207142455341005,
      "loss": 5.6176,
      "mean_token_accuracy": 0.17785695493221282,
      "num_tokens": 55379897.0,
      "step": 28310
    },
    {
      "entropy": 6.005379486083984,
      "epoch": 2.8304093567251463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00042068704987497627,
      "loss": 5.5609,
      "mean_token_accuracy": 0.18364194184541702,
      "num_tokens": 55390950.0,
      "step": 28315
    },
    {
      "entropy": 5.9986021518707275,
      "epoch": 2.8309091817863745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00042065985055040326,
      "loss": 5.4933,
      "mean_token_accuracy": 0.1941374883055687,
      "num_tokens": 55400400.0,
      "step": 28320
    },
    {
      "entropy": 5.994783020019531,
      "epoch": 2.831409006847603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004206326475610658,
      "loss": 5.5272,
      "mean_token_accuracy": 0.18441078662872315,
      "num_tokens": 55409583.0,
      "step": 28325
    },
    {
      "entropy": 5.917728424072266,
      "epoch": 2.831908831908832,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00042060544090764856,
      "loss": 5.4622,
      "mean_token_accuracy": 0.19623906463384627,
      "num_tokens": 55418142.0,
      "step": 28330
    },
    {
      "entropy": 6.043118953704834,
      "epoch": 2.8324086569700606,
      "grad_norm": 1.125,
      "learning_rate": 0.0004205782305908358,
      "loss": 5.6434,
      "mean_token_accuracy": 0.17756642550230026,
      "num_tokens": 55428258.0,
      "step": 28335
    },
    {
      "entropy": 6.117092132568359,
      "epoch": 2.8329084820312893,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004205510166113122,
      "loss": 5.6235,
      "mean_token_accuracy": 0.18092392086982728,
      "num_tokens": 55438414.0,
      "step": 28340
    },
    {
      "entropy": 6.043471002578736,
      "epoch": 2.8334083070925176,
      "grad_norm": 1.234375,
      "learning_rate": 0.00042052379896976254,
      "loss": 5.594,
      "mean_token_accuracy": 0.1773039758205414,
      "num_tokens": 55448820.0,
      "step": 28345
    },
    {
      "entropy": 5.963432550430298,
      "epoch": 2.8339081321537463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004204965776668716,
      "loss": 5.5128,
      "mean_token_accuracy": 0.19347514659166337,
      "num_tokens": 55459031.0,
      "step": 28350
    },
    {
      "entropy": 5.983527183532715,
      "epoch": 2.8344079572149745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004204693527033241,
      "loss": 5.5488,
      "mean_token_accuracy": 0.19018649011850358,
      "num_tokens": 55468647.0,
      "step": 28355
    },
    {
      "entropy": 6.060431718826294,
      "epoch": 2.834907782276203,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00042044212407980517,
      "loss": 5.5562,
      "mean_token_accuracy": 0.18662760853767396,
      "num_tokens": 55478727.0,
      "step": 28360
    },
    {
      "entropy": 6.022957515716553,
      "epoch": 2.835407607337432,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042041489179699983,
      "loss": 5.5884,
      "mean_token_accuracy": 0.18447542041540146,
      "num_tokens": 55489115.0,
      "step": 28365
    },
    {
      "entropy": 6.04775767326355,
      "epoch": 2.8359074323986606,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004203876558555932,
      "loss": 5.5698,
      "mean_token_accuracy": 0.18437811583280564,
      "num_tokens": 55498730.0,
      "step": 28370
    },
    {
      "entropy": 5.970817041397095,
      "epoch": 2.8364072574598893,
      "grad_norm": 1.171875,
      "learning_rate": 0.00042036041625627055,
      "loss": 5.5831,
      "mean_token_accuracy": 0.18742517679929732,
      "num_tokens": 55507659.0,
      "step": 28375
    },
    {
      "entropy": 5.888478946685791,
      "epoch": 2.8369070825211176,
      "grad_norm": 1.203125,
      "learning_rate": 0.00042033317299971726,
      "loss": 5.4861,
      "mean_token_accuracy": 0.19035938531160354,
      "num_tokens": 55517077.0,
      "step": 28380
    },
    {
      "entropy": 5.976660633087159,
      "epoch": 2.8374069075823463,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004203059260866187,
      "loss": 5.4697,
      "mean_token_accuracy": 0.19187301844358445,
      "num_tokens": 55526349.0,
      "step": 28385
    },
    {
      "entropy": 5.955709791183471,
      "epoch": 2.8379067326435745,
      "grad_norm": 1.125,
      "learning_rate": 0.0004202786755176604,
      "loss": 5.5466,
      "mean_token_accuracy": 0.18648500740528107,
      "num_tokens": 55537417.0,
      "step": 28390
    },
    {
      "entropy": 6.0244239330291744,
      "epoch": 2.838406557704803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004202514212935281,
      "loss": 5.5634,
      "mean_token_accuracy": 0.18386086523532869,
      "num_tokens": 55546123.0,
      "step": 28395
    },
    {
      "entropy": 5.975232982635498,
      "epoch": 2.838906382766032,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004202241634149074,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19614486694335936,
      "num_tokens": 55555740.0,
      "step": 28400
    },
    {
      "entropy": 6.001878118515014,
      "epoch": 2.8394062078272606,
      "grad_norm": 1.078125,
      "learning_rate": 0.00042019690188248407,
      "loss": 5.5151,
      "mean_token_accuracy": 0.18818969428539276,
      "num_tokens": 55564435.0,
      "step": 28405
    },
    {
      "entropy": 6.068162679672241,
      "epoch": 2.839906032888489,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004201696366969441,
      "loss": 5.5992,
      "mean_token_accuracy": 0.18622733801603317,
      "num_tokens": 55573893.0,
      "step": 28410
    },
    {
      "entropy": 6.045925760269165,
      "epoch": 2.8404058579497176,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004201423678589734,
      "loss": 5.5358,
      "mean_token_accuracy": 0.1953730583190918,
      "num_tokens": 55583330.0,
      "step": 28415
    },
    {
      "entropy": 5.911889171600341,
      "epoch": 2.8409056830109463,
      "grad_norm": 1.171875,
      "learning_rate": 0.000420115095369258,
      "loss": 5.5168,
      "mean_token_accuracy": 0.190187007188797,
      "num_tokens": 55593000.0,
      "step": 28420
    },
    {
      "entropy": 5.916242599487305,
      "epoch": 2.8414055080721745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004200878192284843,
      "loss": 5.5847,
      "mean_token_accuracy": 0.18218401521444322,
      "num_tokens": 55602835.0,
      "step": 28425
    },
    {
      "entropy": 5.9824730396270756,
      "epoch": 2.841905333133403,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004200605394373383,
      "loss": 5.4861,
      "mean_token_accuracy": 0.1885057955980301,
      "num_tokens": 55612978.0,
      "step": 28430
    },
    {
      "entropy": 6.042035675048828,
      "epoch": 2.842405158194632,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004200332559965065,
      "loss": 5.6389,
      "mean_token_accuracy": 0.1758533537387848,
      "num_tokens": 55622386.0,
      "step": 28435
    },
    {
      "entropy": 6.041363668441773,
      "epoch": 2.8429049832558606,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004200059689066752,
      "loss": 5.6154,
      "mean_token_accuracy": 0.17586992979049682,
      "num_tokens": 55631340.0,
      "step": 28440
    },
    {
      "entropy": 6.1090989112854,
      "epoch": 2.843404808317089,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041997867816853115,
      "loss": 5.6171,
      "mean_token_accuracy": 0.1789204716682434,
      "num_tokens": 55641000.0,
      "step": 28445
    },
    {
      "entropy": 5.9946599960327145,
      "epoch": 2.8439046333783176,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00041995138378276084,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18895063698291778,
      "num_tokens": 55650398.0,
      "step": 28450
    },
    {
      "entropy": 6.026237535476684,
      "epoch": 2.8444044584395463,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041992408575005095,
      "loss": 5.5624,
      "mean_token_accuracy": 0.1907188832759857,
      "num_tokens": 55659454.0,
      "step": 28455
    },
    {
      "entropy": 6.001522779464722,
      "epoch": 2.8449042835007745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004198967840710884,
      "loss": 5.6948,
      "mean_token_accuracy": 0.17695683985948563,
      "num_tokens": 55669166.0,
      "step": 28460
    },
    {
      "entropy": 6.072432565689087,
      "epoch": 2.845404108562003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041986947874656,
      "loss": 5.6239,
      "mean_token_accuracy": 0.18336772322654724,
      "num_tokens": 55678753.0,
      "step": 28465
    },
    {
      "entropy": 6.039589166641235,
      "epoch": 2.845903933623232,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004198421697771529,
      "loss": 5.5429,
      "mean_token_accuracy": 0.18833259493112564,
      "num_tokens": 55688291.0,
      "step": 28470
    },
    {
      "entropy": 6.023969125747681,
      "epoch": 2.8464037586844606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000419814857163554,
      "loss": 5.6112,
      "mean_token_accuracy": 0.18299014568328859,
      "num_tokens": 55697526.0,
      "step": 28475
    },
    {
      "entropy": 6.016570520401001,
      "epoch": 2.846903583745689,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0004197875409064506,
      "loss": 5.5681,
      "mean_token_accuracy": 0.17914360165596008,
      "num_tokens": 55707560.0,
      "step": 28480
    },
    {
      "entropy": 6.055263090133667,
      "epoch": 2.8474034088069176,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004197602210065299,
      "loss": 5.6349,
      "mean_token_accuracy": 0.181109781563282,
      "num_tokens": 55716428.0,
      "step": 28485
    },
    {
      "entropy": 6.030656194686889,
      "epoch": 2.8479032338681463,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004197328974644793,
      "loss": 5.6225,
      "mean_token_accuracy": 0.18076672852039338,
      "num_tokens": 55726123.0,
      "step": 28490
    },
    {
      "entropy": 5.976618051528931,
      "epoch": 2.8484030589293745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004197055702809862,
      "loss": 5.4891,
      "mean_token_accuracy": 0.1837794303894043,
      "num_tokens": 55735668.0,
      "step": 28495
    },
    {
      "entropy": 5.973111009597778,
      "epoch": 2.848902883990603,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004196782394567382,
      "loss": 5.5065,
      "mean_token_accuracy": 0.19095882624387742,
      "num_tokens": 55745651.0,
      "step": 28500
    },
    {
      "entropy": 6.006844520568848,
      "epoch": 2.849402709051832,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000419650904992423,
      "loss": 5.5546,
      "mean_token_accuracy": 0.1836229145526886,
      "num_tokens": 55755024.0,
      "step": 28505
    },
    {
      "entropy": 6.084911775588989,
      "epoch": 2.8499025341130606,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041962356688872817,
      "loss": 5.6162,
      "mean_token_accuracy": 0.17497019618749618,
      "num_tokens": 55765524.0,
      "step": 28510
    },
    {
      "entropy": 6.11924262046814,
      "epoch": 2.850402359174289,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004195962251463417,
      "loss": 5.6134,
      "mean_token_accuracy": 0.1736270934343338,
      "num_tokens": 55775630.0,
      "step": 28515
    },
    {
      "entropy": 6.027758550643921,
      "epoch": 2.8509021842355176,
      "grad_norm": 2.046875,
      "learning_rate": 0.0004195688797659513,
      "loss": 5.6469,
      "mean_token_accuracy": 0.18698108643293382,
      "num_tokens": 55786375.0,
      "step": 28520
    },
    {
      "entropy": 5.998108291625977,
      "epoch": 2.8514020092967463,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041954153074824516,
      "loss": 5.5844,
      "mean_token_accuracy": 0.18436254560947418,
      "num_tokens": 55795080.0,
      "step": 28525
    },
    {
      "entropy": 6.066546678543091,
      "epoch": 2.8519018343579745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004195141780939113,
      "loss": 5.5701,
      "mean_token_accuracy": 0.18169667720794677,
      "num_tokens": 55804737.0,
      "step": 28530
    },
    {
      "entropy": 5.950295162200928,
      "epoch": 2.852401659419203,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041948682180363784,
      "loss": 5.4318,
      "mean_token_accuracy": 0.19042568951845168,
      "num_tokens": 55814166.0,
      "step": 28535
    },
    {
      "entropy": 5.978711891174316,
      "epoch": 2.852901484480432,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004194594618781132,
      "loss": 5.5127,
      "mean_token_accuracy": 0.18893036395311355,
      "num_tokens": 55823758.0,
      "step": 28540
    },
    {
      "entropy": 6.101582050323486,
      "epoch": 2.8534013095416606,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041943209831802567,
      "loss": 5.5763,
      "mean_token_accuracy": 0.18851593136787415,
      "num_tokens": 55833798.0,
      "step": 28545
    },
    {
      "entropy": 6.027812957763672,
      "epoch": 2.853901134602889,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041940473112406365,
      "loss": 5.5817,
      "mean_token_accuracy": 0.1830003321170807,
      "num_tokens": 55843456.0,
      "step": 28550
    },
    {
      "entropy": 5.926304912567138,
      "epoch": 2.8544009596641176,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041937736029691575,
      "loss": 5.5067,
      "mean_token_accuracy": 0.1916378140449524,
      "num_tokens": 55853401.0,
      "step": 28555
    },
    {
      "entropy": 5.99504132270813,
      "epoch": 2.8549007847253463,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004193499858372706,
      "loss": 5.5369,
      "mean_token_accuracy": 0.19255839437246322,
      "num_tokens": 55863549.0,
      "step": 28560
    },
    {
      "entropy": 6.107935285568237,
      "epoch": 2.8554006097865745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000419322607745817,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1828018918633461,
      "num_tokens": 55873718.0,
      "step": 28565
    },
    {
      "entropy": 5.98563404083252,
      "epoch": 2.855900434847803,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041929522602324377,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1877002462744713,
      "num_tokens": 55883909.0,
      "step": 28570
    },
    {
      "entropy": 6.006081295013428,
      "epoch": 2.856400259909032,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004192678406702398,
      "loss": 5.6013,
      "mean_token_accuracy": 0.17771776914596557,
      "num_tokens": 55893090.0,
      "step": 28575
    },
    {
      "entropy": 5.955460071563721,
      "epoch": 2.8569000849702606,
      "grad_norm": 1.1875,
      "learning_rate": 0.000419240451687494,
      "loss": 5.4947,
      "mean_token_accuracy": 0.19303350597620011,
      "num_tokens": 55903820.0,
      "step": 28580
    },
    {
      "entropy": 6.085053873062134,
      "epoch": 2.857399910031489,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041921305907569555,
      "loss": 5.6099,
      "mean_token_accuracy": 0.18184424191713333,
      "num_tokens": 55913540.0,
      "step": 28585
    },
    {
      "entropy": 6.022350835800171,
      "epoch": 2.8578997350927176,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004191856628355337,
      "loss": 5.5892,
      "mean_token_accuracy": 0.17954727858304978,
      "num_tokens": 55923766.0,
      "step": 28590
    },
    {
      "entropy": 6.017279815673828,
      "epoch": 2.8583995601539463,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004191582629676977,
      "loss": 5.614,
      "mean_token_accuracy": 0.1853022575378418,
      "num_tokens": 55932855.0,
      "step": 28595
    },
    {
      "entropy": 6.010232830047608,
      "epoch": 2.8588993852151745,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041913085947287675,
      "loss": 5.5639,
      "mean_token_accuracy": 0.18282729834318162,
      "num_tokens": 55942163.0,
      "step": 28600
    },
    {
      "entropy": 6.007336902618408,
      "epoch": 2.859399210276403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004191034523517606,
      "loss": 5.5286,
      "mean_token_accuracy": 0.1871779292821884,
      "num_tokens": 55952521.0,
      "step": 28605
    },
    {
      "entropy": 5.974952554702758,
      "epoch": 2.859899035337632,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041907604160503864,
      "loss": 5.5577,
      "mean_token_accuracy": 0.18954758942127228,
      "num_tokens": 55962071.0,
      "step": 28610
    },
    {
      "entropy": 5.908056020736694,
      "epoch": 2.8603988603988606,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041904862723340046,
      "loss": 5.4822,
      "mean_token_accuracy": 0.193089035153389,
      "num_tokens": 55971387.0,
      "step": 28615
    },
    {
      "entropy": 6.0225306987762455,
      "epoch": 2.860898685460089,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041902120923753593,
      "loss": 5.565,
      "mean_token_accuracy": 0.1818866103887558,
      "num_tokens": 55980435.0,
      "step": 28620
    },
    {
      "entropy": 5.96475133895874,
      "epoch": 2.8613985105213176,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004189937876181349,
      "loss": 5.4745,
      "mean_token_accuracy": 0.19197167456150055,
      "num_tokens": 55989868.0,
      "step": 28625
    },
    {
      "entropy": 5.990807962417603,
      "epoch": 2.8618983355825462,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004189663623758872,
      "loss": 5.4926,
      "mean_token_accuracy": 0.19352337419986726,
      "num_tokens": 55999423.0,
      "step": 28630
    },
    {
      "entropy": 5.92466368675232,
      "epoch": 2.8623981606437745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041893893351148284,
      "loss": 5.5171,
      "mean_token_accuracy": 0.18917523473501205,
      "num_tokens": 56009429.0,
      "step": 28635
    },
    {
      "entropy": 6.055291271209716,
      "epoch": 2.862897985705003,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000418911501025612,
      "loss": 5.5778,
      "mean_token_accuracy": 0.18064332902431487,
      "num_tokens": 56019039.0,
      "step": 28640
    },
    {
      "entropy": 6.037579727172852,
      "epoch": 2.863397810766232,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004188840649189648,
      "loss": 5.5973,
      "mean_token_accuracy": 0.1792689248919487,
      "num_tokens": 56029465.0,
      "step": 28645
    },
    {
      "entropy": 6.0482306480407715,
      "epoch": 2.8638976358274606,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041885662519223154,
      "loss": 5.6535,
      "mean_token_accuracy": 0.17834306359291077,
      "num_tokens": 56038882.0,
      "step": 28650
    },
    {
      "entropy": 6.092318058013916,
      "epoch": 2.864397460888689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004188291818461026,
      "loss": 5.654,
      "mean_token_accuracy": 0.175331312417984,
      "num_tokens": 56048512.0,
      "step": 28655
    },
    {
      "entropy": 5.983028554916382,
      "epoch": 2.8648972859499175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041880173488126847,
      "loss": 5.5163,
      "mean_token_accuracy": 0.19009689539670943,
      "num_tokens": 56057553.0,
      "step": 28660
    },
    {
      "entropy": 5.962471580505371,
      "epoch": 2.8653971110111462,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041877428429841967,
      "loss": 5.552,
      "mean_token_accuracy": 0.18870423436164857,
      "num_tokens": 56066819.0,
      "step": 28665
    },
    {
      "entropy": 6.01138277053833,
      "epoch": 2.8658969360723745,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004187468300982469,
      "loss": 5.6007,
      "mean_token_accuracy": 0.1881350666284561,
      "num_tokens": 56077012.0,
      "step": 28670
    },
    {
      "entropy": 6.0268206119537355,
      "epoch": 2.866396761133603,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004187193722814409,
      "loss": 5.6517,
      "mean_token_accuracy": 0.17870944142341613,
      "num_tokens": 56086405.0,
      "step": 28675
    },
    {
      "entropy": 6.046143960952759,
      "epoch": 2.866896586194832,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004186919108486925,
      "loss": 5.513,
      "mean_token_accuracy": 0.1838408052921295,
      "num_tokens": 56097135.0,
      "step": 28680
    },
    {
      "entropy": 5.9668967723846436,
      "epoch": 2.8673964112560606,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004186644458006925,
      "loss": 5.4686,
      "mean_token_accuracy": 0.1953880861401558,
      "num_tokens": 56106401.0,
      "step": 28685
    },
    {
      "entropy": 5.913062381744385,
      "epoch": 2.867896236317289,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000418636977138132,
      "loss": 5.4168,
      "mean_token_accuracy": 0.19903374463319778,
      "num_tokens": 56115763.0,
      "step": 28690
    },
    {
      "entropy": 5.9367152690887455,
      "epoch": 2.8683960613785175,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041860950486170227,
      "loss": 5.5385,
      "mean_token_accuracy": 0.19022829085588455,
      "num_tokens": 56124873.0,
      "step": 28695
    },
    {
      "entropy": 5.958936548233032,
      "epoch": 2.8688958864397462,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004185820289720942,
      "loss": 5.4355,
      "mean_token_accuracy": 0.19386904090642929,
      "num_tokens": 56133801.0,
      "step": 28700
    },
    {
      "entropy": 6.018774747848511,
      "epoch": 2.8693957115009745,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041855454946999925,
      "loss": 5.6173,
      "mean_token_accuracy": 0.18138625025749205,
      "num_tokens": 56143827.0,
      "step": 28705
    },
    {
      "entropy": 6.002480792999267,
      "epoch": 2.869895536562203,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041852706635610884,
      "loss": 5.6244,
      "mean_token_accuracy": 0.17511110454797746,
      "num_tokens": 56152780.0,
      "step": 28710
    },
    {
      "entropy": 5.945188903808594,
      "epoch": 2.870395361623432,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041849957963111434,
      "loss": 5.4897,
      "mean_token_accuracy": 0.19288689345121385,
      "num_tokens": 56162168.0,
      "step": 28715
    },
    {
      "entropy": 5.97844820022583,
      "epoch": 2.8708951866846606,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004184720892957073,
      "loss": 5.5621,
      "mean_token_accuracy": 0.18794063478708267,
      "num_tokens": 56172069.0,
      "step": 28720
    },
    {
      "entropy": 6.008296632766724,
      "epoch": 2.871395011745889,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004184445953505795,
      "loss": 5.5829,
      "mean_token_accuracy": 0.18538438975811006,
      "num_tokens": 56181904.0,
      "step": 28725
    },
    {
      "entropy": 5.967702722549438,
      "epoch": 2.8718948368071175,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004184170977964225,
      "loss": 5.5266,
      "mean_token_accuracy": 0.19165645241737367,
      "num_tokens": 56191203.0,
      "step": 28730
    },
    {
      "entropy": 5.990869665145874,
      "epoch": 2.8723946618683462,
      "grad_norm": 1.546875,
      "learning_rate": 0.00041838959663392827,
      "loss": 5.5798,
      "mean_token_accuracy": 0.18961986601352693,
      "num_tokens": 56201714.0,
      "step": 28735
    },
    {
      "entropy": 6.006943416595459,
      "epoch": 2.8728944869295745,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004183620918637887,
      "loss": 5.5752,
      "mean_token_accuracy": 0.184392948448658,
      "num_tokens": 56213287.0,
      "step": 28740
    },
    {
      "entropy": 5.974403524398804,
      "epoch": 2.873394311990803,
      "grad_norm": 1.4609375,
      "learning_rate": 0.0004183345834866957,
      "loss": 5.4793,
      "mean_token_accuracy": 0.19338324069976806,
      "num_tokens": 56222593.0,
      "step": 28745
    },
    {
      "entropy": 6.006207132339478,
      "epoch": 2.873894137052032,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041830707150334164,
      "loss": 5.5181,
      "mean_token_accuracy": 0.1796454593539238,
      "num_tokens": 56232292.0,
      "step": 28750
    },
    {
      "entropy": 5.945680332183838,
      "epoch": 2.8743939621132606,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004182795559144184,
      "loss": 5.5086,
      "mean_token_accuracy": 0.1913171887397766,
      "num_tokens": 56242389.0,
      "step": 28755
    },
    {
      "entropy": 5.956672477722168,
      "epoch": 2.874893787174489,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004182520367206184,
      "loss": 5.6358,
      "mean_token_accuracy": 0.18494925498962403,
      "num_tokens": 56252383.0,
      "step": 28760
    },
    {
      "entropy": 5.984306335449219,
      "epoch": 2.8753936122357175,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041822451392263413,
      "loss": 5.5834,
      "mean_token_accuracy": 0.18412245362997054,
      "num_tokens": 56261676.0,
      "step": 28765
    },
    {
      "entropy": 6.018061828613281,
      "epoch": 2.875893437296946,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004181969875211578,
      "loss": 5.5109,
      "mean_token_accuracy": 0.19147640317678452,
      "num_tokens": 56271938.0,
      "step": 28770
    },
    {
      "entropy": 5.994848918914795,
      "epoch": 2.8763932623581745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041816945751688215,
      "loss": 5.5681,
      "mean_token_accuracy": 0.18030113875865936,
      "num_tokens": 56281370.0,
      "step": 28775
    },
    {
      "entropy": 5.95074429512024,
      "epoch": 2.876893087419403,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041814192391049983,
      "loss": 5.5509,
      "mean_token_accuracy": 0.18601560294628144,
      "num_tokens": 56292162.0,
      "step": 28780
    },
    {
      "entropy": 6.035732173919678,
      "epoch": 2.877392912480632,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004181143867027036,
      "loss": 5.5072,
      "mean_token_accuracy": 0.19902151674032212,
      "num_tokens": 56301129.0,
      "step": 28785
    },
    {
      "entropy": 6.046225690841675,
      "epoch": 2.8778927375418606,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041808684589418615,
      "loss": 5.5901,
      "mean_token_accuracy": 0.18414104729890823,
      "num_tokens": 56310781.0,
      "step": 28790
    },
    {
      "entropy": 5.97324013710022,
      "epoch": 2.878392562603089,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004180593014856404,
      "loss": 5.5414,
      "mean_token_accuracy": 0.19050752222537995,
      "num_tokens": 56320401.0,
      "step": 28795
    },
    {
      "entropy": 5.97267632484436,
      "epoch": 2.8788923876643175,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004180317534777596,
      "loss": 5.5721,
      "mean_token_accuracy": 0.18503506481647491,
      "num_tokens": 56330606.0,
      "step": 28800
    },
    {
      "entropy": 6.028334474563598,
      "epoch": 2.879392212725546,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041800420187123656,
      "loss": 5.5393,
      "mean_token_accuracy": 0.19180229008197786,
      "num_tokens": 56340832.0,
      "step": 28805
    },
    {
      "entropy": 5.921131372451782,
      "epoch": 2.8798920377867745,
      "grad_norm": 1.125,
      "learning_rate": 0.0004179766466667646,
      "loss": 5.5032,
      "mean_token_accuracy": 0.19368707984685898,
      "num_tokens": 56350228.0,
      "step": 28810
    },
    {
      "entropy": 5.993889045715332,
      "epoch": 2.880391862848003,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004179490878650371,
      "loss": 5.5311,
      "mean_token_accuracy": 0.18093425184488296,
      "num_tokens": 56360087.0,
      "step": 28815
    },
    {
      "entropy": 5.988643503189087,
      "epoch": 2.880891687909232,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004179215254667473,
      "loss": 5.5023,
      "mean_token_accuracy": 0.19206811636686325,
      "num_tokens": 56369214.0,
      "step": 28820
    },
    {
      "entropy": 5.972115421295166,
      "epoch": 2.8813915129704606,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004178939594725887,
      "loss": 5.6088,
      "mean_token_accuracy": 0.18332124054431914,
      "num_tokens": 56378331.0,
      "step": 28825
    },
    {
      "entropy": 6.025776481628418,
      "epoch": 2.881891338031689,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004178663898832548,
      "loss": 5.6135,
      "mean_token_accuracy": 0.17637778222560882,
      "num_tokens": 56387856.0,
      "step": 28830
    },
    {
      "entropy": 6.133811664581299,
      "epoch": 2.8823911630929175,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004178388166994392,
      "loss": 5.6657,
      "mean_token_accuracy": 0.17454176992177964,
      "num_tokens": 56398885.0,
      "step": 28835
    },
    {
      "entropy": 5.942352771759033,
      "epoch": 2.882890988154146,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004178112399218359,
      "loss": 5.4876,
      "mean_token_accuracy": 0.1880880892276764,
      "num_tokens": 56408169.0,
      "step": 28840
    },
    {
      "entropy": 5.896474170684814,
      "epoch": 2.8833908132153745,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041778365955113844,
      "loss": 5.4704,
      "mean_token_accuracy": 0.1864808514714241,
      "num_tokens": 56417221.0,
      "step": 28845
    },
    {
      "entropy": 5.966894054412842,
      "epoch": 2.883890638276603,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004177560755880409,
      "loss": 5.5099,
      "mean_token_accuracy": 0.19265046268701552,
      "num_tokens": 56426860.0,
      "step": 28850
    },
    {
      "entropy": 6.0585455894470215,
      "epoch": 2.884390463337832,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004177284880332373,
      "loss": 5.6458,
      "mean_token_accuracy": 0.17821545600891114,
      "num_tokens": 56436264.0,
      "step": 28855
    },
    {
      "entropy": 6.026314067840576,
      "epoch": 2.8848902883990606,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041770089688742155,
      "loss": 5.5463,
      "mean_token_accuracy": 0.18867271542549133,
      "num_tokens": 56445835.0,
      "step": 28860
    },
    {
      "entropy": 6.0123148441314695,
      "epoch": 2.885390113460289,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000417673302151288,
      "loss": 5.6071,
      "mean_token_accuracy": 0.1851532354950905,
      "num_tokens": 56455090.0,
      "step": 28865
    },
    {
      "entropy": 5.999368810653687,
      "epoch": 2.8858899385215175,
      "grad_norm": 1.125,
      "learning_rate": 0.0004176457038255309,
      "loss": 5.5786,
      "mean_token_accuracy": 0.18709810078144073,
      "num_tokens": 56464782.0,
      "step": 28870
    },
    {
      "entropy": 6.041669273376465,
      "epoch": 2.886389763582746,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004176181019108446,
      "loss": 5.5634,
      "mean_token_accuracy": 0.18640597760677338,
      "num_tokens": 56475375.0,
      "step": 28875
    },
    {
      "entropy": 5.9645270824432375,
      "epoch": 2.8868895886439745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004175904964079236,
      "loss": 5.4467,
      "mean_token_accuracy": 0.18990441113710405,
      "num_tokens": 56485654.0,
      "step": 28880
    },
    {
      "entropy": 6.01309642791748,
      "epoch": 2.887389413705203,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004175628873174623,
      "loss": 5.5952,
      "mean_token_accuracy": 0.18496671169996262,
      "num_tokens": 56495554.0,
      "step": 28885
    },
    {
      "entropy": 5.993366146087647,
      "epoch": 2.887889238766432,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041753527464015547,
      "loss": 5.4767,
      "mean_token_accuracy": 0.19217303693294524,
      "num_tokens": 56504760.0,
      "step": 28890
    },
    {
      "entropy": 5.936500692367554,
      "epoch": 2.8883890638276606,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004175076583766978,
      "loss": 5.5194,
      "mean_token_accuracy": 0.1932409942150116,
      "num_tokens": 56514312.0,
      "step": 28895
    },
    {
      "entropy": 5.979573535919189,
      "epoch": 2.888888888888889,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0004174800385277842,
      "loss": 5.5772,
      "mean_token_accuracy": 0.18206910341978072,
      "num_tokens": 56524741.0,
      "step": 28900
    },
    {
      "entropy": 5.955585575103759,
      "epoch": 2.8893887139501175,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004174524150941094,
      "loss": 5.5436,
      "mean_token_accuracy": 0.18373286724090576,
      "num_tokens": 56534339.0,
      "step": 28905
    },
    {
      "entropy": 6.079880714416504,
      "epoch": 2.889888539011346,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004174247880763685,
      "loss": 5.6537,
      "mean_token_accuracy": 0.1766335979104042,
      "num_tokens": 56544722.0,
      "step": 28910
    },
    {
      "entropy": 6.003150606155396,
      "epoch": 2.8903883640725745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004173971574752566,
      "loss": 5.4832,
      "mean_token_accuracy": 0.19007839560508727,
      "num_tokens": 56553917.0,
      "step": 28915
    },
    {
      "entropy": 6.052454423904419,
      "epoch": 2.890888189133803,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004173695232914688,
      "loss": 5.5988,
      "mean_token_accuracy": 0.1853411614894867,
      "num_tokens": 56563876.0,
      "step": 28920
    },
    {
      "entropy": 5.9820314884185795,
      "epoch": 2.891388014195032,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004173418855257005,
      "loss": 5.5113,
      "mean_token_accuracy": 0.19004387110471727,
      "num_tokens": 56574017.0,
      "step": 28925
    },
    {
      "entropy": 5.986132621765137,
      "epoch": 2.8918878392562606,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000417314244178647,
      "loss": 5.5295,
      "mean_token_accuracy": 0.17762120813131332,
      "num_tokens": 56583927.0,
      "step": 28930
    },
    {
      "entropy": 5.996918630599976,
      "epoch": 2.892387664317489,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004172865992510037,
      "loss": 5.5465,
      "mean_token_accuracy": 0.1874375656247139,
      "num_tokens": 56593421.0,
      "step": 28935
    },
    {
      "entropy": 6.005105686187744,
      "epoch": 2.8928874893787175,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041725895074346615,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1865562990307808,
      "num_tokens": 56601820.0,
      "step": 28940
    },
    {
      "entropy": 5.915988302230835,
      "epoch": 2.893387314439946,
      "grad_norm": 1.046875,
      "learning_rate": 0.0004172312986567301,
      "loss": 5.4927,
      "mean_token_accuracy": 0.18815140724182128,
      "num_tokens": 56611688.0,
      "step": 28945
    },
    {
      "entropy": 5.943217849731445,
      "epoch": 2.8938871395011745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004172036429914911,
      "loss": 5.4779,
      "mean_token_accuracy": 0.19051360934972764,
      "num_tokens": 56620719.0,
      "step": 28950
    },
    {
      "entropy": 5.984273433685303,
      "epoch": 2.894386964562403,
      "grad_norm": 1.75,
      "learning_rate": 0.00041717598374844505,
      "loss": 5.6221,
      "mean_token_accuracy": 0.18108896911144257,
      "num_tokens": 56631045.0,
      "step": 28955
    },
    {
      "entropy": 6.057505655288696,
      "epoch": 2.894886789623632,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041714832092828785,
      "loss": 5.5538,
      "mean_token_accuracy": 0.1832215428352356,
      "num_tokens": 56640935.0,
      "step": 28960
    },
    {
      "entropy": 6.102058029174804,
      "epoch": 2.89538661468486,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041712065453171544,
      "loss": 5.6339,
      "mean_token_accuracy": 0.1757847011089325,
      "num_tokens": 56651023.0,
      "step": 28965
    },
    {
      "entropy": 6.028440141677857,
      "epoch": 2.895886439746089,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041709298455942394,
      "loss": 5.5846,
      "mean_token_accuracy": 0.18127516955137252,
      "num_tokens": 56660591.0,
      "step": 28970
    },
    {
      "entropy": 6.064974594116211,
      "epoch": 2.8963862648073175,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041706531101210956,
      "loss": 5.6679,
      "mean_token_accuracy": 0.17992310374975204,
      "num_tokens": 56671635.0,
      "step": 28975
    },
    {
      "entropy": 5.97171573638916,
      "epoch": 2.896886089868546,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041703763389046853,
      "loss": 5.4763,
      "mean_token_accuracy": 0.18965007066726686,
      "num_tokens": 56681309.0,
      "step": 28980
    },
    {
      "entropy": 5.912668132781983,
      "epoch": 2.8973859149297745,
      "grad_norm": 1.265625,
      "learning_rate": 0.00041700995319519707,
      "loss": 5.5303,
      "mean_token_accuracy": 0.1942803144454956,
      "num_tokens": 56692594.0,
      "step": 28985
    },
    {
      "entropy": 6.039387464523315,
      "epoch": 2.897885739991003,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004169822689269918,
      "loss": 5.5943,
      "mean_token_accuracy": 0.18250313103199006,
      "num_tokens": 56702165.0,
      "step": 28990
    },
    {
      "entropy": 6.054687929153443,
      "epoch": 2.898385565052232,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041695458108654914,
      "loss": 5.5568,
      "mean_token_accuracy": 0.19027812778949738,
      "num_tokens": 56711398.0,
      "step": 28995
    },
    {
      "entropy": 5.971416044235229,
      "epoch": 2.89888539011346,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041692688967456576,
      "loss": 5.508,
      "mean_token_accuracy": 0.18859827965497972,
      "num_tokens": 56720984.0,
      "step": 29000
    },
    {
      "entropy": 5.956514930725097,
      "epoch": 2.899385215174689,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041689919469173835,
      "loss": 5.577,
      "mean_token_accuracy": 0.19008856117725373,
      "num_tokens": 56732126.0,
      "step": 29005
    },
    {
      "entropy": 5.948633337020874,
      "epoch": 2.8998850402359175,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004168714961387637,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1918627843260765,
      "num_tokens": 56741235.0,
      "step": 29010
    },
    {
      "entropy": 6.03798885345459,
      "epoch": 2.900384865297146,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004168437940163388,
      "loss": 5.6685,
      "mean_token_accuracy": 0.1847105860710144,
      "num_tokens": 56751695.0,
      "step": 29015
    },
    {
      "entropy": 6.089259672164917,
      "epoch": 2.9008846903583745,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004168160883251605,
      "loss": 5.5709,
      "mean_token_accuracy": 0.18372823297977448,
      "num_tokens": 56761067.0,
      "step": 29020
    },
    {
      "entropy": 6.081299734115601,
      "epoch": 2.901384515419603,
      "grad_norm": 1.125,
      "learning_rate": 0.00041678837906592586,
      "loss": 5.5436,
      "mean_token_accuracy": 0.18564230650663377,
      "num_tokens": 56770267.0,
      "step": 29025
    },
    {
      "entropy": 6.0022491931915285,
      "epoch": 2.901884340480832,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041676066623933216,
      "loss": 5.5525,
      "mean_token_accuracy": 0.19222951531410218,
      "num_tokens": 56780263.0,
      "step": 29030
    },
    {
      "entropy": 5.93476824760437,
      "epoch": 2.90238416554206,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004167329498460766,
      "loss": 5.508,
      "mean_token_accuracy": 0.189471335709095,
      "num_tokens": 56789490.0,
      "step": 29035
    },
    {
      "entropy": 6.010347318649292,
      "epoch": 2.902883990603289,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041670522988685637,
      "loss": 5.5532,
      "mean_token_accuracy": 0.1902566984295845,
      "num_tokens": 56799299.0,
      "step": 29040
    },
    {
      "entropy": 6.01339693069458,
      "epoch": 2.9033838156645175,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041667750636236914,
      "loss": 5.5044,
      "mean_token_accuracy": 0.18552822470664979,
      "num_tokens": 56808518.0,
      "step": 29045
    },
    {
      "entropy": 6.0758884906768795,
      "epoch": 2.903883640725746,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004166497792733122,
      "loss": 5.669,
      "mean_token_accuracy": 0.18003336489200591,
      "num_tokens": 56819458.0,
      "step": 29050
    },
    {
      "entropy": 6.061254405975342,
      "epoch": 2.9043834657869745,
      "grad_norm": 1.125,
      "learning_rate": 0.0004166220486203834,
      "loss": 5.5938,
      "mean_token_accuracy": 0.18263496160507203,
      "num_tokens": 56828219.0,
      "step": 29055
    },
    {
      "entropy": 5.981641054153442,
      "epoch": 2.904883290848203,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004165943144042802,
      "loss": 5.4952,
      "mean_token_accuracy": 0.19456336498260499,
      "num_tokens": 56837998.0,
      "step": 29060
    },
    {
      "entropy": 6.0832360744476315,
      "epoch": 2.905383115909432,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041656657662570047,
      "loss": 5.6344,
      "mean_token_accuracy": 0.17916933000087737,
      "num_tokens": 56848244.0,
      "step": 29065
    },
    {
      "entropy": 6.032700395584106,
      "epoch": 2.90588294097066,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004165388352853421,
      "loss": 5.4925,
      "mean_token_accuracy": 0.18839731961488723,
      "num_tokens": 56857809.0,
      "step": 29070
    },
    {
      "entropy": 5.911396074295044,
      "epoch": 2.906382766031889,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041651109038390305,
      "loss": 5.4232,
      "mean_token_accuracy": 0.1994226321578026,
      "num_tokens": 56866985.0,
      "step": 29075
    },
    {
      "entropy": 5.970412254333496,
      "epoch": 2.9068825910931175,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041648334192208146,
      "loss": 5.567,
      "mean_token_accuracy": 0.1783318117260933,
      "num_tokens": 56877402.0,
      "step": 29080
    },
    {
      "entropy": 6.087867498397827,
      "epoch": 2.907382416154346,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041645558990057536,
      "loss": 5.5322,
      "mean_token_accuracy": 0.18673183619976044,
      "num_tokens": 56886322.0,
      "step": 29085
    },
    {
      "entropy": 6.044263410568237,
      "epoch": 2.9078822412155745,
      "grad_norm": 1.0625,
      "learning_rate": 0.00041642783432008294,
      "loss": 5.6036,
      "mean_token_accuracy": 0.18337072879076005,
      "num_tokens": 56896067.0,
      "step": 29090
    },
    {
      "entropy": 6.030143737792969,
      "epoch": 2.908382066276803,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041640007518130274,
      "loss": 5.6049,
      "mean_token_accuracy": 0.18122841119766236,
      "num_tokens": 56904850.0,
      "step": 29095
    },
    {
      "entropy": 6.043479585647583,
      "epoch": 2.908881891338032,
      "grad_norm": 1.0625,
      "learning_rate": 0.000416372312484933,
      "loss": 5.572,
      "mean_token_accuracy": 0.18325877338647842,
      "num_tokens": 56915856.0,
      "step": 29100
    },
    {
      "entropy": 6.012440824508667,
      "epoch": 2.90938171639926,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041634454623167217,
      "loss": 5.5557,
      "mean_token_accuracy": 0.1906826227903366,
      "num_tokens": 56925817.0,
      "step": 29105
    },
    {
      "entropy": 5.965674257278442,
      "epoch": 2.909881541460489,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041631677642221896,
      "loss": 5.5201,
      "mean_token_accuracy": 0.19558734744787215,
      "num_tokens": 56935838.0,
      "step": 29110
    },
    {
      "entropy": 6.0460399150848385,
      "epoch": 2.9103813665217175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000416289003057272,
      "loss": 5.5538,
      "mean_token_accuracy": 0.19448685348033906,
      "num_tokens": 56945926.0,
      "step": 29115
    },
    {
      "entropy": 5.9277036666870115,
      "epoch": 2.910881191582946,
      "grad_norm": 0.921875,
      "learning_rate": 0.0004162612261375302,
      "loss": 5.4115,
      "mean_token_accuracy": 0.1968894362449646,
      "num_tokens": 56955639.0,
      "step": 29120
    },
    {
      "entropy": 5.979749059677124,
      "epoch": 2.9113810166441745,
      "grad_norm": 1.125,
      "learning_rate": 0.00041623344566369216,
      "loss": 5.5405,
      "mean_token_accuracy": 0.18687474429607392,
      "num_tokens": 56966653.0,
      "step": 29125
    },
    {
      "entropy": 5.961220073699951,
      "epoch": 2.911880841705403,
      "grad_norm": 1.109375,
      "learning_rate": 0.000416205661636457,
      "loss": 5.4702,
      "mean_token_accuracy": 0.1969687506556511,
      "num_tokens": 56975797.0,
      "step": 29130
    },
    {
      "entropy": 5.959200429916382,
      "epoch": 2.912380666766632,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004161778740565238,
      "loss": 5.4671,
      "mean_token_accuracy": 0.19511618763208388,
      "num_tokens": 56984994.0,
      "step": 29135
    },
    {
      "entropy": 5.995854902267456,
      "epoch": 2.91288049182786,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041615008292459157,
      "loss": 5.5884,
      "mean_token_accuracy": 0.188632133603096,
      "num_tokens": 56995630.0,
      "step": 29140
    },
    {
      "entropy": 6.0458526611328125,
      "epoch": 2.913380316889089,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041612228824135954,
      "loss": 5.5786,
      "mean_token_accuracy": 0.18571010529994963,
      "num_tokens": 57005864.0,
      "step": 29145
    },
    {
      "entropy": 6.046120738983154,
      "epoch": 2.9138801419503175,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041609449000752706,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18958573937416076,
      "num_tokens": 57015821.0,
      "step": 29150
    },
    {
      "entropy": 6.022066307067871,
      "epoch": 2.914379967011546,
      "grad_norm": 1.828125,
      "learning_rate": 0.0004160666882237936,
      "loss": 5.5264,
      "mean_token_accuracy": 0.19660263657569885,
      "num_tokens": 57024849.0,
      "step": 29155
    },
    {
      "entropy": 5.938467884063721,
      "epoch": 2.9148797920727745,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004160388828908585,
      "loss": 5.5295,
      "mean_token_accuracy": 0.18750450611114503,
      "num_tokens": 57033754.0,
      "step": 29160
    },
    {
      "entropy": 5.9739786148071286,
      "epoch": 2.915379617134003,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041601107400942146,
      "loss": 5.4854,
      "mean_token_accuracy": 0.1916822835803032,
      "num_tokens": 57042880.0,
      "step": 29165
    },
    {
      "entropy": 6.028504133224487,
      "epoch": 2.915879442195232,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041598326158018197,
      "loss": 5.592,
      "mean_token_accuracy": 0.17805259376764299,
      "num_tokens": 57052717.0,
      "step": 29170
    },
    {
      "entropy": 6.014332294464111,
      "epoch": 2.91637926725646,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041595544560384004,
      "loss": 5.5313,
      "mean_token_accuracy": 0.19062376022338867,
      "num_tokens": 57062813.0,
      "step": 29175
    },
    {
      "entropy": 6.024864912033081,
      "epoch": 2.916879092317689,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004159276260810952,
      "loss": 5.5226,
      "mean_token_accuracy": 0.1862972319126129,
      "num_tokens": 57072798.0,
      "step": 29180
    },
    {
      "entropy": 5.96354489326477,
      "epoch": 2.9173789173789175,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041589980301264765,
      "loss": 5.4768,
      "mean_token_accuracy": 0.19568116813898087,
      "num_tokens": 57082114.0,
      "step": 29185
    },
    {
      "entropy": 5.907982635498047,
      "epoch": 2.917878742440146,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004158719763991973,
      "loss": 5.4967,
      "mean_token_accuracy": 0.19103347808122634,
      "num_tokens": 57091603.0,
      "step": 29190
    },
    {
      "entropy": 5.943726873397827,
      "epoch": 2.9183785675013745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041584414624144426,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1842697888612747,
      "num_tokens": 57101403.0,
      "step": 29195
    },
    {
      "entropy": 6.005723285675049,
      "epoch": 2.918878392562603,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004158163125400887,
      "loss": 5.4918,
      "mean_token_accuracy": 0.1900193765759468,
      "num_tokens": 57109732.0,
      "step": 29200
    },
    {
      "entropy": 6.013118267059326,
      "epoch": 2.919378217623832,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041578847529583106,
      "loss": 5.567,
      "mean_token_accuracy": 0.18855001032352448,
      "num_tokens": 57119289.0,
      "step": 29205
    },
    {
      "entropy": 5.939472627639771,
      "epoch": 2.91987804268506,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004157606345093715,
      "loss": 5.4934,
      "mean_token_accuracy": 0.19221878796815872,
      "num_tokens": 57130243.0,
      "step": 29210
    },
    {
      "entropy": 6.071733713150024,
      "epoch": 2.920377867746289,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004157327901814107,
      "loss": 5.5949,
      "mean_token_accuracy": 0.17863181978464127,
      "num_tokens": 57140885.0,
      "step": 29215
    },
    {
      "entropy": 5.925492000579834,
      "epoch": 2.9208776928075175,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041570494231264906,
      "loss": 5.4437,
      "mean_token_accuracy": 0.195447202026844,
      "num_tokens": 57150479.0,
      "step": 29220
    },
    {
      "entropy": 5.949387836456299,
      "epoch": 2.9213775178687458,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00041567709090378725,
      "loss": 5.4297,
      "mean_token_accuracy": 0.1908073365688324,
      "num_tokens": 57161561.0,
      "step": 29225
    },
    {
      "entropy": 5.968845653533935,
      "epoch": 2.9218773429299745,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041564923595552604,
      "loss": 5.6127,
      "mean_token_accuracy": 0.1870241194963455,
      "num_tokens": 57171734.0,
      "step": 29230
    },
    {
      "entropy": 6.050087404251099,
      "epoch": 2.922377167991203,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041562137746856625,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17728988975286483,
      "num_tokens": 57182448.0,
      "step": 29235
    },
    {
      "entropy": 5.980618286132812,
      "epoch": 2.922876993052432,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041559351544360893,
      "loss": 5.5581,
      "mean_token_accuracy": 0.18161847740411757,
      "num_tokens": 57192331.0,
      "step": 29240
    },
    {
      "entropy": 5.860567855834961,
      "epoch": 2.92337681811366,
      "grad_norm": 1.125,
      "learning_rate": 0.00041556564988135474,
      "loss": 5.33,
      "mean_token_accuracy": 0.20920469015836715,
      "num_tokens": 57201636.0,
      "step": 29245
    },
    {
      "entropy": 5.981903553009033,
      "epoch": 2.923876643174889,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004155377807825051,
      "loss": 5.5513,
      "mean_token_accuracy": 0.18700242340564727,
      "num_tokens": 57210186.0,
      "step": 29250
    },
    {
      "entropy": 5.956227779388428,
      "epoch": 2.9243764682361175,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000415509908147761,
      "loss": 5.4902,
      "mean_token_accuracy": 0.19028568714857103,
      "num_tokens": 57219212.0,
      "step": 29255
    },
    {
      "entropy": 5.9900413990020756,
      "epoch": 2.9248762932973458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041548203197782377,
      "loss": 5.5734,
      "mean_token_accuracy": 0.19159263819456102,
      "num_tokens": 57229728.0,
      "step": 29260
    },
    {
      "entropy": 5.994774913787841,
      "epoch": 2.9253761183585745,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041545415227339474,
      "loss": 5.5824,
      "mean_token_accuracy": 0.18521761596202851,
      "num_tokens": 57240233.0,
      "step": 29265
    },
    {
      "entropy": 6.023496341705322,
      "epoch": 2.925875943419803,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004154262690351754,
      "loss": 5.6154,
      "mean_token_accuracy": 0.18550288826227188,
      "num_tokens": 57249502.0,
      "step": 29270
    },
    {
      "entropy": 6.022699975967408,
      "epoch": 2.926375768481032,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004153983822638673,
      "loss": 5.524,
      "mean_token_accuracy": 0.1864834487438202,
      "num_tokens": 57258825.0,
      "step": 29275
    },
    {
      "entropy": 5.945605897903443,
      "epoch": 2.92687559354226,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041537049196017196,
      "loss": 5.5906,
      "mean_token_accuracy": 0.1858075425028801,
      "num_tokens": 57269657.0,
      "step": 29280
    },
    {
      "entropy": 5.9945969581604,
      "epoch": 2.927375418603489,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041534259812479125,
      "loss": 5.4349,
      "mean_token_accuracy": 0.19943612664937974,
      "num_tokens": 57280192.0,
      "step": 29285
    },
    {
      "entropy": 5.867608737945557,
      "epoch": 2.927875243664717,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004153147007584269,
      "loss": 5.3924,
      "mean_token_accuracy": 0.19645698815584184,
      "num_tokens": 57290292.0,
      "step": 29290
    },
    {
      "entropy": 5.9818117141723635,
      "epoch": 2.9283750687259458,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041528679986178066,
      "loss": 5.5653,
      "mean_token_accuracy": 0.1852535277605057,
      "num_tokens": 57300844.0,
      "step": 29295
    },
    {
      "entropy": 5.9570880889892575,
      "epoch": 2.9288748937871745,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041525889543555475,
      "loss": 5.509,
      "mean_token_accuracy": 0.1934513732790947,
      "num_tokens": 57310162.0,
      "step": 29300
    },
    {
      "entropy": 6.023156261444091,
      "epoch": 2.929374718848403,
      "grad_norm": 1.171875,
      "learning_rate": 0.000415230987480451,
      "loss": 5.6123,
      "mean_token_accuracy": 0.18756709694862367,
      "num_tokens": 57319599.0,
      "step": 29305
    },
    {
      "entropy": 6.040363502502442,
      "epoch": 2.929874543909632,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004152030759971718,
      "loss": 5.6561,
      "mean_token_accuracy": 0.1859562337398529,
      "num_tokens": 57329901.0,
      "step": 29310
    },
    {
      "entropy": 6.008559703826904,
      "epoch": 2.93037436897086,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004151751609864193,
      "loss": 5.5983,
      "mean_token_accuracy": 0.18865681290626526,
      "num_tokens": 57339362.0,
      "step": 29315
    },
    {
      "entropy": 6.039473676681519,
      "epoch": 2.930874194032089,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004151472424488957,
      "loss": 5.5582,
      "mean_token_accuracy": 0.184381602704525,
      "num_tokens": 57349527.0,
      "step": 29320
    },
    {
      "entropy": 6.06801438331604,
      "epoch": 2.931374019093317,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004151193203853036,
      "loss": 5.5566,
      "mean_token_accuracy": 0.18170531317591668,
      "num_tokens": 57360021.0,
      "step": 29325
    },
    {
      "entropy": 6.067793607711792,
      "epoch": 2.9318738441545458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041509139479634544,
      "loss": 5.6049,
      "mean_token_accuracy": 0.18299957662820815,
      "num_tokens": 57369938.0,
      "step": 29330
    },
    {
      "entropy": 6.001828861236572,
      "epoch": 2.9323736692157745,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041506346568272385,
      "loss": 5.5697,
      "mean_token_accuracy": 0.19260056614875792,
      "num_tokens": 57380399.0,
      "step": 29335
    },
    {
      "entropy": 6.0501243591308596,
      "epoch": 2.932873494277003,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041503553304514157,
      "loss": 5.637,
      "mean_token_accuracy": 0.18347941040992738,
      "num_tokens": 57390639.0,
      "step": 29340
    },
    {
      "entropy": 6.054699325561524,
      "epoch": 2.933373319338232,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004150075968843012,
      "loss": 5.557,
      "mean_token_accuracy": 0.19377168864011765,
      "num_tokens": 57401567.0,
      "step": 29345
    },
    {
      "entropy": 5.972626543045044,
      "epoch": 2.93387314439946,
      "grad_norm": 1.234375,
      "learning_rate": 0.00041497965720090573,
      "loss": 5.5135,
      "mean_token_accuracy": 0.192818720638752,
      "num_tokens": 57410550.0,
      "step": 29350
    },
    {
      "entropy": 5.983793783187866,
      "epoch": 2.934372969460689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041495171399565816,
      "loss": 5.6207,
      "mean_token_accuracy": 0.17977932542562486,
      "num_tokens": 57420029.0,
      "step": 29355
    },
    {
      "entropy": 6.1336568832397464,
      "epoch": 2.934872794521917,
      "grad_norm": 1.125,
      "learning_rate": 0.0004149237672692615,
      "loss": 5.7273,
      "mean_token_accuracy": 0.17756507098674773,
      "num_tokens": 57429732.0,
      "step": 29360
    },
    {
      "entropy": 6.039405393600464,
      "epoch": 2.9353726195831458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004148958170224188,
      "loss": 5.547,
      "mean_token_accuracy": 0.19167325049638748,
      "num_tokens": 57439469.0,
      "step": 29365
    },
    {
      "entropy": 5.9620569229125975,
      "epoch": 2.9358724446443745,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041486786325583343,
      "loss": 5.5098,
      "mean_token_accuracy": 0.19799445271492006,
      "num_tokens": 57450218.0,
      "step": 29370
    },
    {
      "entropy": 6.122945547103882,
      "epoch": 2.936372269705603,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041483990597020854,
      "loss": 5.6226,
      "mean_token_accuracy": 0.17981446236371995,
      "num_tokens": 57459610.0,
      "step": 29375
    },
    {
      "entropy": 6.066587448120117,
      "epoch": 2.936872094766832,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041481194516624763,
      "loss": 5.6016,
      "mean_token_accuracy": 0.1807072401046753,
      "num_tokens": 57468849.0,
      "step": 29380
    },
    {
      "entropy": 6.029281663894653,
      "epoch": 2.93737191982806,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004147839808446541,
      "loss": 5.5415,
      "mean_token_accuracy": 0.18786125332117082,
      "num_tokens": 57477827.0,
      "step": 29385
    },
    {
      "entropy": 5.94865050315857,
      "epoch": 2.937871744889289,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041475601300613166,
      "loss": 5.5265,
      "mean_token_accuracy": 0.19161620289087294,
      "num_tokens": 57487502.0,
      "step": 29390
    },
    {
      "entropy": 5.988337230682373,
      "epoch": 2.938371569950517,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041472804165138384,
      "loss": 5.5198,
      "mean_token_accuracy": 0.18503762036561966,
      "num_tokens": 57498281.0,
      "step": 29395
    },
    {
      "entropy": 6.022827768325806,
      "epoch": 2.9388713950117458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004147000667811145,
      "loss": 5.5991,
      "mean_token_accuracy": 0.19076108634471894,
      "num_tokens": 57509612.0,
      "step": 29400
    },
    {
      "entropy": 6.056727170944214,
      "epoch": 2.9393712200729745,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004146720883960273,
      "loss": 5.5827,
      "mean_token_accuracy": 0.18651618659496308,
      "num_tokens": 57519211.0,
      "step": 29405
    },
    {
      "entropy": 5.976264524459839,
      "epoch": 2.939871045134203,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004146441064968264,
      "loss": 5.5449,
      "mean_token_accuracy": 0.1872389316558838,
      "num_tokens": 57528756.0,
      "step": 29410
    },
    {
      "entropy": 6.009111499786377,
      "epoch": 2.940370870195432,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00041461612108421576,
      "loss": 5.588,
      "mean_token_accuracy": 0.17851599603891372,
      "num_tokens": 57537892.0,
      "step": 29415
    },
    {
      "entropy": 6.0385541915893555,
      "epoch": 2.94087069525666,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00041458813215889935,
      "loss": 5.6243,
      "mean_token_accuracy": 0.17685019671916963,
      "num_tokens": 57547650.0,
      "step": 29420
    },
    {
      "entropy": 5.927580308914185,
      "epoch": 2.941370520317889,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004145601397215815,
      "loss": 5.5329,
      "mean_token_accuracy": 0.1908237025141716,
      "num_tokens": 57557077.0,
      "step": 29425
    },
    {
      "entropy": 6.072788047790527,
      "epoch": 2.941870345379117,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041453214377296636,
      "loss": 5.627,
      "mean_token_accuracy": 0.18544137924909593,
      "num_tokens": 57566402.0,
      "step": 29430
    },
    {
      "entropy": 6.0405707359313965,
      "epoch": 2.9423701704403458,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004145041443137585,
      "loss": 5.5691,
      "mean_token_accuracy": 0.1847331181168556,
      "num_tokens": 57577093.0,
      "step": 29435
    },
    {
      "entropy": 6.050566053390503,
      "epoch": 2.9428699955015745,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004144761413446622,
      "loss": 5.5259,
      "mean_token_accuracy": 0.18981388211250305,
      "num_tokens": 57586790.0,
      "step": 29440
    },
    {
      "entropy": 5.956250953674316,
      "epoch": 2.943369820562803,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041444813486638215,
      "loss": 5.4657,
      "mean_token_accuracy": 0.19388211220502855,
      "num_tokens": 57595615.0,
      "step": 29445
    },
    {
      "entropy": 5.972224378585816,
      "epoch": 2.943869645624032,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041442012487962284,
      "loss": 5.5577,
      "mean_token_accuracy": 0.1889161393046379,
      "num_tokens": 57605436.0,
      "step": 29450
    },
    {
      "entropy": 6.08711609840393,
      "epoch": 2.94436947068526,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004143921113850891,
      "loss": 5.6058,
      "mean_token_accuracy": 0.17517657279968263,
      "num_tokens": 57614286.0,
      "step": 29455
    },
    {
      "entropy": 5.926671981811523,
      "epoch": 2.944869295746489,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041436409438348564,
      "loss": 5.4671,
      "mean_token_accuracy": 0.1945877715945244,
      "num_tokens": 57624591.0,
      "step": 29460
    },
    {
      "entropy": 5.985031461715698,
      "epoch": 2.945369120807717,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004143360738755175,
      "loss": 5.5189,
      "mean_token_accuracy": 0.19162131994962692,
      "num_tokens": 57633987.0,
      "step": 29465
    },
    {
      "entropy": 6.072393465042114,
      "epoch": 2.9458689458689458,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004143080498618895,
      "loss": 5.6078,
      "mean_token_accuracy": 0.18133717626333237,
      "num_tokens": 57643989.0,
      "step": 29470
    },
    {
      "entropy": 5.981821918487549,
      "epoch": 2.9463687709301745,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041428002234330694,
      "loss": 5.5295,
      "mean_token_accuracy": 0.18350795954465865,
      "num_tokens": 57653860.0,
      "step": 29475
    },
    {
      "entropy": 5.917502927780151,
      "epoch": 2.946868595991403,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004142519913204748,
      "loss": 5.4693,
      "mean_token_accuracy": 0.2024311438202858,
      "num_tokens": 57663245.0,
      "step": 29480
    },
    {
      "entropy": 6.020078420639038,
      "epoch": 2.9473684210526314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041422395679409834,
      "loss": 5.4743,
      "mean_token_accuracy": 0.19178796857595443,
      "num_tokens": 57672624.0,
      "step": 29485
    },
    {
      "entropy": 5.931502723693848,
      "epoch": 2.94786824611386,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041419591876488304,
      "loss": 5.5504,
      "mean_token_accuracy": 0.19396725594997405,
      "num_tokens": 57683278.0,
      "step": 29490
    },
    {
      "entropy": 5.967512702941894,
      "epoch": 2.948368071175089,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004141678772335342,
      "loss": 5.5014,
      "mean_token_accuracy": 0.18988406360149385,
      "num_tokens": 57692252.0,
      "step": 29495
    },
    {
      "entropy": 6.053036403656006,
      "epoch": 2.948867896236317,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041413983220075736,
      "loss": 5.5441,
      "mean_token_accuracy": 0.1812570199370384,
      "num_tokens": 57703154.0,
      "step": 29500
    },
    {
      "entropy": 6.121570587158203,
      "epoch": 2.9493677212975458,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041411178366725814,
      "loss": 5.7296,
      "mean_token_accuracy": 0.17207881957292556,
      "num_tokens": 57714526.0,
      "step": 29505
    },
    {
      "entropy": 6.009703016281128,
      "epoch": 2.9498675463587745,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004140837316337423,
      "loss": 5.5314,
      "mean_token_accuracy": 0.19020500630140305,
      "num_tokens": 57724143.0,
      "step": 29510
    },
    {
      "entropy": 6.078697443008423,
      "epoch": 2.950367371420003,
      "grad_norm": 1.2578125,
      "learning_rate": 0.0004140556761009155,
      "loss": 5.6914,
      "mean_token_accuracy": 0.1728191152215004,
      "num_tokens": 57732772.0,
      "step": 29515
    },
    {
      "entropy": 5.975446653366089,
      "epoch": 2.9508671964812314,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004140276170694837,
      "loss": 5.5394,
      "mean_token_accuracy": 0.1871001347899437,
      "num_tokens": 57742200.0,
      "step": 29520
    },
    {
      "entropy": 6.065972471237183,
      "epoch": 2.95136702154246,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041399955454015286,
      "loss": 5.6033,
      "mean_token_accuracy": 0.180644291639328,
      "num_tokens": 57752477.0,
      "step": 29525
    },
    {
      "entropy": 6.02420392036438,
      "epoch": 2.951866846603689,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041397148851362903,
      "loss": 5.5307,
      "mean_token_accuracy": 0.18796690702438354,
      "num_tokens": 57761204.0,
      "step": 29530
    },
    {
      "entropy": 5.903612899780273,
      "epoch": 2.952366671664917,
      "grad_norm": 1.03125,
      "learning_rate": 0.00041394341899061814,
      "loss": 5.4192,
      "mean_token_accuracy": 0.19416651874780655,
      "num_tokens": 57771222.0,
      "step": 29535
    },
    {
      "entropy": 5.993813610076904,
      "epoch": 2.9528664967261458,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004139153459718267,
      "loss": 5.6356,
      "mean_token_accuracy": 0.17262024581432342,
      "num_tokens": 57781200.0,
      "step": 29540
    },
    {
      "entropy": 5.903094339370727,
      "epoch": 2.9533663217873745,
      "grad_norm": 1.109375,
      "learning_rate": 0.000413887269457961,
      "loss": 5.3897,
      "mean_token_accuracy": 0.2032831758260727,
      "num_tokens": 57790139.0,
      "step": 29545
    },
    {
      "entropy": 5.982140064239502,
      "epoch": 2.953866146848603,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041385918944972716,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1853357195854187,
      "num_tokens": 57800156.0,
      "step": 29550
    },
    {
      "entropy": 6.062548351287842,
      "epoch": 2.9543659719098314,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041383110594783197,
      "loss": 5.5998,
      "mean_token_accuracy": 0.18874290883541106,
      "num_tokens": 57810344.0,
      "step": 29555
    },
    {
      "entropy": 5.95925555229187,
      "epoch": 2.95486579697106,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00041380301895298173,
      "loss": 5.4909,
      "mean_token_accuracy": 0.1859721064567566,
      "num_tokens": 57820382.0,
      "step": 29560
    },
    {
      "entropy": 5.975183773040771,
      "epoch": 2.955365622032289,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041377492846588333,
      "loss": 5.6401,
      "mean_token_accuracy": 0.17948297262191773,
      "num_tokens": 57831250.0,
      "step": 29565
    },
    {
      "entropy": 5.974965906143188,
      "epoch": 2.955865447093517,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041374683448724346,
      "loss": 5.5598,
      "mean_token_accuracy": 0.19483625888824463,
      "num_tokens": 57840134.0,
      "step": 29570
    },
    {
      "entropy": 6.082124996185303,
      "epoch": 2.9563652721547458,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004137187370177689,
      "loss": 5.5968,
      "mean_token_accuracy": 0.1854638710618019,
      "num_tokens": 57850147.0,
      "step": 29575
    },
    {
      "entropy": 5.946426963806152,
      "epoch": 2.9568650972159745,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041369063605816667,
      "loss": 5.4843,
      "mean_token_accuracy": 0.1984087646007538,
      "num_tokens": 57859516.0,
      "step": 29580
    },
    {
      "entropy": 5.9339570045471195,
      "epoch": 2.957364922277203,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004136625316091437,
      "loss": 5.5477,
      "mean_token_accuracy": 0.18196811378002167,
      "num_tokens": 57869858.0,
      "step": 29585
    },
    {
      "entropy": 6.034710931777954,
      "epoch": 2.9578647473384314,
      "grad_norm": 1.0,
      "learning_rate": 0.0004136344236714071,
      "loss": 5.6457,
      "mean_token_accuracy": 0.17853848338127137,
      "num_tokens": 57879804.0,
      "step": 29590
    },
    {
      "entropy": 6.06839952468872,
      "epoch": 2.95836457239966,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041360631224566407,
      "loss": 5.5227,
      "mean_token_accuracy": 0.20286415219306947,
      "num_tokens": 57889936.0,
      "step": 29595
    },
    {
      "entropy": 5.958437538146972,
      "epoch": 2.958864397460889,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004135781973326219,
      "loss": 5.4955,
      "mean_token_accuracy": 0.1934566840529442,
      "num_tokens": 57899230.0,
      "step": 29600
    },
    {
      "entropy": 6.007947635650635,
      "epoch": 2.959364222522117,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041355007893298796,
      "loss": 5.6002,
      "mean_token_accuracy": 0.1784466564655304,
      "num_tokens": 57909741.0,
      "step": 29605
    },
    {
      "entropy": 5.968269777297974,
      "epoch": 2.9598640475833458,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004135219570474697,
      "loss": 5.5506,
      "mean_token_accuracy": 0.18746088892221452,
      "num_tokens": 57918558.0,
      "step": 29610
    },
    {
      "entropy": 5.895211410522461,
      "epoch": 2.9603638726445745,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041349383167677456,
      "loss": 5.3787,
      "mean_token_accuracy": 0.20397517532110215,
      "num_tokens": 57928868.0,
      "step": 29615
    },
    {
      "entropy": 6.008860874176025,
      "epoch": 2.960863697705803,
      "grad_norm": 1.125,
      "learning_rate": 0.00041346570282161033,
      "loss": 5.5033,
      "mean_token_accuracy": 0.18880465179681777,
      "num_tokens": 57937890.0,
      "step": 29620
    },
    {
      "entropy": 5.954699468612671,
      "epoch": 2.9613635227670314,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00041343757048268464,
      "loss": 5.5811,
      "mean_token_accuracy": 0.1833593800663948,
      "num_tokens": 57946340.0,
      "step": 29625
    },
    {
      "entropy": 6.0650512218475345,
      "epoch": 2.96186334782826,
      "grad_norm": 0.96875,
      "learning_rate": 0.0004134094346607053,
      "loss": 5.6508,
      "mean_token_accuracy": 0.17725763618946075,
      "num_tokens": 57955968.0,
      "step": 29630
    },
    {
      "entropy": 5.998663091659546,
      "epoch": 2.962363172889489,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004133812953563802,
      "loss": 5.4661,
      "mean_token_accuracy": 0.19454983174800872,
      "num_tokens": 57965345.0,
      "step": 29635
    },
    {
      "entropy": 5.971135091781616,
      "epoch": 2.962862997950717,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00041335315257041733,
      "loss": 5.5151,
      "mean_token_accuracy": 0.18974445909261703,
      "num_tokens": 57974652.0,
      "step": 29640
    },
    {
      "entropy": 5.980753612518311,
      "epoch": 2.9633628230119458,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041332500630352463,
      "loss": 5.5756,
      "mean_token_accuracy": 0.1838421866297722,
      "num_tokens": 57983845.0,
      "step": 29645
    },
    {
      "entropy": 5.9552192211151125,
      "epoch": 2.9638626480731745,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004132968565564105,
      "loss": 5.563,
      "mean_token_accuracy": 0.19131224900484084,
      "num_tokens": 57993455.0,
      "step": 29650
    },
    {
      "entropy": 6.039216756820679,
      "epoch": 2.964362473134403,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000413268703329783,
      "loss": 5.5626,
      "mean_token_accuracy": 0.17748664170503617,
      "num_tokens": 58003983.0,
      "step": 29655
    },
    {
      "entropy": 5.993852233886718,
      "epoch": 2.9648622981956314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004132405466243505,
      "loss": 5.5344,
      "mean_token_accuracy": 0.1858879432082176,
      "num_tokens": 58014486.0,
      "step": 29660
    },
    {
      "entropy": 5.980943012237549,
      "epoch": 2.96536212325686,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041321238644082144,
      "loss": 5.4498,
      "mean_token_accuracy": 0.19472070783376694,
      "num_tokens": 58023610.0,
      "step": 29665
    },
    {
      "entropy": 6.072493839263916,
      "epoch": 2.965861948318089,
      "grad_norm": 1.3125,
      "learning_rate": 0.00041318422277990426,
      "loss": 5.6665,
      "mean_token_accuracy": 0.18644228279590608,
      "num_tokens": 58033829.0,
      "step": 29670
    },
    {
      "entropy": 5.97233772277832,
      "epoch": 2.966361773379317,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004131560556423076,
      "loss": 5.529,
      "mean_token_accuracy": 0.18147474825382232,
      "num_tokens": 58042716.0,
      "step": 29675
    },
    {
      "entropy": 5.967957639694214,
      "epoch": 2.9668615984405458,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004131278850287401,
      "loss": 5.5174,
      "mean_token_accuracy": 0.1817076474428177,
      "num_tokens": 58052667.0,
      "step": 29680
    },
    {
      "entropy": 6.014030981063843,
      "epoch": 2.9673614235017745,
      "grad_norm": 1.078125,
      "learning_rate": 0.00041309971093991067,
      "loss": 5.5836,
      "mean_token_accuracy": 0.1796314537525177,
      "num_tokens": 58061690.0,
      "step": 29685
    },
    {
      "entropy": 6.072140884399414,
      "epoch": 2.967861248563003,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041307153337652805,
      "loss": 5.5541,
      "mean_token_accuracy": 0.18361676186323167,
      "num_tokens": 58071212.0,
      "step": 29690
    },
    {
      "entropy": 5.982745409011841,
      "epoch": 2.9683610736242314,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000413043352339301,
      "loss": 5.5292,
      "mean_token_accuracy": 0.18420960903167724,
      "num_tokens": 58081227.0,
      "step": 29695
    },
    {
      "entropy": 6.0355487823486325,
      "epoch": 2.96886089868546,
      "grad_norm": 1.015625,
      "learning_rate": 0.0004130151678289388,
      "loss": 5.5893,
      "mean_token_accuracy": 0.18170927762985228,
      "num_tokens": 58091303.0,
      "step": 29700
    },
    {
      "entropy": 5.959450769424438,
      "epoch": 2.969360723746689,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004129869798461506,
      "loss": 5.4419,
      "mean_token_accuracy": 0.1965516358613968,
      "num_tokens": 58100124.0,
      "step": 29705
    },
    {
      "entropy": 5.946548795700073,
      "epoch": 2.969860548807917,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004129587883916454,
      "loss": 5.5722,
      "mean_token_accuracy": 0.18986990600824355,
      "num_tokens": 58110326.0,
      "step": 29710
    },
    {
      "entropy": 5.974349021911621,
      "epoch": 2.9703603738691458,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004129305934661326,
      "loss": 5.5348,
      "mean_token_accuracy": 0.18973829001188278,
      "num_tokens": 58119465.0,
      "step": 29715
    },
    {
      "entropy": 6.000444984436035,
      "epoch": 2.9708601989303745,
      "grad_norm": 1.140625,
      "learning_rate": 0.00041290239507032155,
      "loss": 5.6234,
      "mean_token_accuracy": 0.18723777532577515,
      "num_tokens": 58130136.0,
      "step": 29720
    },
    {
      "entropy": 5.942437028884887,
      "epoch": 2.971360023991603,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004128741932049218,
      "loss": 5.4674,
      "mean_token_accuracy": 0.19492389857769013,
      "num_tokens": 58139725.0,
      "step": 29725
    },
    {
      "entropy": 5.900717258453369,
      "epoch": 2.9718598490528314,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004128459878706428,
      "loss": 5.4586,
      "mean_token_accuracy": 0.20155538022518157,
      "num_tokens": 58149060.0,
      "step": 29730
    },
    {
      "entropy": 5.94240894317627,
      "epoch": 2.97235967411406,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004128177790681942,
      "loss": 5.546,
      "mean_token_accuracy": 0.1866272807121277,
      "num_tokens": 58159627.0,
      "step": 29735
    },
    {
      "entropy": 5.991946792602539,
      "epoch": 2.972859499175289,
      "grad_norm": 1.125,
      "learning_rate": 0.00041278956679828566,
      "loss": 5.4553,
      "mean_token_accuracy": 0.19701793193817138,
      "num_tokens": 58169546.0,
      "step": 29740
    },
    {
      "entropy": 5.948609066009522,
      "epoch": 2.973359324236517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041276135106162713,
      "loss": 5.4907,
      "mean_token_accuracy": 0.19415191262960435,
      "num_tokens": 58179521.0,
      "step": 29745
    },
    {
      "entropy": 5.986577701568604,
      "epoch": 2.9738591492977458,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004127331318589285,
      "loss": 5.5285,
      "mean_token_accuracy": 0.18708404153585434,
      "num_tokens": 58188975.0,
      "step": 29750
    },
    {
      "entropy": 6.021513748168945,
      "epoch": 2.9743589743589745,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041270490919089966,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18219734132289886,
      "num_tokens": 58198692.0,
      "step": 29755
    },
    {
      "entropy": 6.093898820877075,
      "epoch": 2.974858799420203,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004126766830582508,
      "loss": 5.6754,
      "mean_token_accuracy": 0.17540400773286818,
      "num_tokens": 58208441.0,
      "step": 29760
    },
    {
      "entropy": 6.0223057746887205,
      "epoch": 2.9753586244814314,
      "grad_norm": 1.0703125,
      "learning_rate": 0.000412648453461692,
      "loss": 5.6228,
      "mean_token_accuracy": 0.18570591807365416,
      "num_tokens": 58218750.0,
      "step": 29765
    },
    {
      "entropy": 5.885152864456177,
      "epoch": 2.97585844954266,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00041262022040193346,
      "loss": 5.4708,
      "mean_token_accuracy": 0.1874534472823143,
      "num_tokens": 58229445.0,
      "step": 29770
    },
    {
      "entropy": 6.046731901168823,
      "epoch": 2.976358274603889,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041259198387968564,
      "loss": 5.6397,
      "mean_token_accuracy": 0.18157826662063598,
      "num_tokens": 58240012.0,
      "step": 29775
    },
    {
      "entropy": 6.0540198802948,
      "epoch": 2.976858099665117,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041256374389565895,
      "loss": 5.6552,
      "mean_token_accuracy": 0.17606754451990128,
      "num_tokens": 58250303.0,
      "step": 29780
    },
    {
      "entropy": 5.927592182159424,
      "epoch": 2.9773579247263458,
      "grad_norm": 1.125,
      "learning_rate": 0.00041253550045056375,
      "loss": 5.3955,
      "mean_token_accuracy": 0.20423873960971833,
      "num_tokens": 58259376.0,
      "step": 29785
    },
    {
      "entropy": 6.00399169921875,
      "epoch": 2.9778577497875744,
      "grad_norm": 1.125,
      "learning_rate": 0.00041250725354511085,
      "loss": 5.6566,
      "mean_token_accuracy": 0.17770321816205978,
      "num_tokens": 58269620.0,
      "step": 29790
    },
    {
      "entropy": 5.942795372009277,
      "epoch": 2.978357574848803,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004124790031800109,
      "loss": 5.4741,
      "mean_token_accuracy": 0.19657584577798842,
      "num_tokens": 58280121.0,
      "step": 29795
    },
    {
      "entropy": 6.045471668243408,
      "epoch": 2.9788573999100314,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004124507493559745,
      "loss": 5.6489,
      "mean_token_accuracy": 0.18027630001306533,
      "num_tokens": 58289944.0,
      "step": 29800
    },
    {
      "entropy": 6.087820911407471,
      "epoch": 2.97935722497126,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004124224920737126,
      "loss": 5.6392,
      "mean_token_accuracy": 0.17862434238195418,
      "num_tokens": 58300091.0,
      "step": 29805
    },
    {
      "entropy": 5.917552471160889,
      "epoch": 2.9798570500324884,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004123942313339361,
      "loss": 5.3653,
      "mean_token_accuracy": 0.1994468092918396,
      "num_tokens": 58309249.0,
      "step": 29810
    },
    {
      "entropy": 6.02134690284729,
      "epoch": 2.980356875093717,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041236596713735624,
      "loss": 5.6293,
      "mean_token_accuracy": 0.18090686053037644,
      "num_tokens": 58319238.0,
      "step": 29815
    },
    {
      "entropy": 6.062193965911865,
      "epoch": 2.9808567001549457,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004123376994846839,
      "loss": 5.5828,
      "mean_token_accuracy": 0.18343030363321305,
      "num_tokens": 58329091.0,
      "step": 29820
    },
    {
      "entropy": 6.040722370147705,
      "epoch": 2.9813565252161744,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004123094283766304,
      "loss": 5.5027,
      "mean_token_accuracy": 0.19234997034072876,
      "num_tokens": 58338581.0,
      "step": 29825
    },
    {
      "entropy": 6.059003734588623,
      "epoch": 2.981856350277403,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041228115381390705,
      "loss": 5.6031,
      "mean_token_accuracy": 0.18422225415706633,
      "num_tokens": 58347890.0,
      "step": 29830
    },
    {
      "entropy": 6.051473951339721,
      "epoch": 2.9823561753386314,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041225287579722526,
      "loss": 5.6473,
      "mean_token_accuracy": 0.1763221427798271,
      "num_tokens": 58357910.0,
      "step": 29835
    },
    {
      "entropy": 6.0867105484008786,
      "epoch": 2.98285600039986,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004122245943272963,
      "loss": 5.626,
      "mean_token_accuracy": 0.1790614143013954,
      "num_tokens": 58367819.0,
      "step": 29840
    },
    {
      "entropy": 5.991890144348145,
      "epoch": 2.9833558254610884,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041219630940483187,
      "loss": 5.5539,
      "mean_token_accuracy": 0.1857790470123291,
      "num_tokens": 58377324.0,
      "step": 29845
    },
    {
      "entropy": 5.990442085266113,
      "epoch": 2.983855650522317,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004121680210305436,
      "loss": 5.5403,
      "mean_token_accuracy": 0.18527545630931855,
      "num_tokens": 58386018.0,
      "step": 29850
    },
    {
      "entropy": 6.01331934928894,
      "epoch": 2.9843554755835457,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004121397292051433,
      "loss": 5.4859,
      "mean_token_accuracy": 0.19112567454576493,
      "num_tokens": 58396049.0,
      "step": 29855
    },
    {
      "entropy": 6.058278131484985,
      "epoch": 2.9848553006447744,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004121114339293427,
      "loss": 5.5758,
      "mean_token_accuracy": 0.18413000404834748,
      "num_tokens": 58407443.0,
      "step": 29860
    },
    {
      "entropy": 5.8019428730010985,
      "epoch": 2.985355125706003,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004120831352038537,
      "loss": 5.3536,
      "mean_token_accuracy": 0.2081429824233055,
      "num_tokens": 58417466.0,
      "step": 29865
    },
    {
      "entropy": 6.016385650634765,
      "epoch": 2.9858549507672314,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004120548330293882,
      "loss": 5.5193,
      "mean_token_accuracy": 0.18847994208335878,
      "num_tokens": 58426819.0,
      "step": 29870
    },
    {
      "entropy": 6.012957000732422,
      "epoch": 2.98635477582846,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00041202652740665845,
      "loss": 5.5307,
      "mean_token_accuracy": 0.186588317155838,
      "num_tokens": 58436470.0,
      "step": 29875
    },
    {
      "entropy": 6.039665985107422,
      "epoch": 2.9868546008896883,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004119982183363765,
      "loss": 5.5454,
      "mean_token_accuracy": 0.19019873440265656,
      "num_tokens": 58448095.0,
      "step": 29880
    },
    {
      "entropy": 5.972097396850586,
      "epoch": 2.987354425950917,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00041196990581925466,
      "loss": 5.5504,
      "mean_token_accuracy": 0.18915600925683976,
      "num_tokens": 58457162.0,
      "step": 29885
    },
    {
      "entropy": 6.042478561401367,
      "epoch": 2.9878542510121457,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004119415898560052,
      "loss": 5.6761,
      "mean_token_accuracy": 0.18269980251789092,
      "num_tokens": 58467763.0,
      "step": 29890
    },
    {
      "entropy": 6.0925201892852785,
      "epoch": 2.9883540760733744,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004119132704473405,
      "loss": 5.5731,
      "mean_token_accuracy": 0.18319107592105865,
      "num_tokens": 58477769.0,
      "step": 29895
    },
    {
      "entropy": 6.0691667079925535,
      "epoch": 2.988853901134603,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004118849475939733,
      "loss": 5.5949,
      "mean_token_accuracy": 0.17806579619646073,
      "num_tokens": 58486656.0,
      "step": 29900
    },
    {
      "entropy": 6.020935869216919,
      "epoch": 2.9893537261958314,
      "grad_norm": 1.125,
      "learning_rate": 0.00041185662129661593,
      "loss": 5.6258,
      "mean_token_accuracy": 0.18053748607635497,
      "num_tokens": 58496979.0,
      "step": 29905
    },
    {
      "entropy": 6.0426445484161375,
      "epoch": 2.98985355125706,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004118282915559813,
      "loss": 5.6174,
      "mean_token_accuracy": 0.18860926628112792,
      "num_tokens": 58507340.0,
      "step": 29910
    },
    {
      "entropy": 5.966879272460938,
      "epoch": 2.9903533763182883,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004117999583727819,
      "loss": 5.5545,
      "mean_token_accuracy": 0.18002053648233413,
      "num_tokens": 58517364.0,
      "step": 29915
    },
    {
      "entropy": 6.036857986450196,
      "epoch": 2.990853201379517,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041177162174773084,
      "loss": 5.4821,
      "mean_token_accuracy": 0.19262961596250533,
      "num_tokens": 58526435.0,
      "step": 29920
    },
    {
      "entropy": 5.986081981658936,
      "epoch": 2.9913530264407457,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041174328168154094,
      "loss": 5.5017,
      "mean_token_accuracy": 0.19622809141874314,
      "num_tokens": 58536694.0,
      "step": 29925
    },
    {
      "entropy": 5.965514469146728,
      "epoch": 2.9918528515019744,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004117149381749253,
      "loss": 5.5187,
      "mean_token_accuracy": 0.1915062576532364,
      "num_tokens": 58546854.0,
      "step": 29930
    },
    {
      "entropy": 5.97601547241211,
      "epoch": 2.992352676563203,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004116865912285971,
      "loss": 5.5529,
      "mean_token_accuracy": 0.18865127712488175,
      "num_tokens": 58557936.0,
      "step": 29935
    },
    {
      "entropy": 6.038477373123169,
      "epoch": 2.9928525016244314,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041165824084326923,
      "loss": 5.6344,
      "mean_token_accuracy": 0.17815895527601242,
      "num_tokens": 58568107.0,
      "step": 29940
    },
    {
      "entropy": 5.982776212692261,
      "epoch": 2.99335232668566,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004116298870196553,
      "loss": 5.4434,
      "mean_token_accuracy": 0.19593830555677413,
      "num_tokens": 58576756.0,
      "step": 29945
    },
    {
      "entropy": 5.978842926025391,
      "epoch": 2.9938521517468883,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041160152975846853,
      "loss": 5.5456,
      "mean_token_accuracy": 0.193827985227108,
      "num_tokens": 58586894.0,
      "step": 29950
    },
    {
      "entropy": 6.007510423660278,
      "epoch": 2.994351976808117,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004115731690604225,
      "loss": 5.4948,
      "mean_token_accuracy": 0.18850331455469133,
      "num_tokens": 58596720.0,
      "step": 29955
    },
    {
      "entropy": 6.037660360336304,
      "epoch": 2.9948518018693457,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004115448049262306,
      "loss": 5.546,
      "mean_token_accuracy": 0.18342062681913376,
      "num_tokens": 58605705.0,
      "step": 29960
    },
    {
      "entropy": 6.088764953613281,
      "epoch": 2.9953516269305744,
      "grad_norm": 1.25,
      "learning_rate": 0.0004115164373566065,
      "loss": 5.7009,
      "mean_token_accuracy": 0.1766279548406601,
      "num_tokens": 58615669.0,
      "step": 29965
    },
    {
      "entropy": 6.06580228805542,
      "epoch": 2.995851451991803,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00041148806635226407,
      "loss": 5.5989,
      "mean_token_accuracy": 0.1895786091685295,
      "num_tokens": 58625900.0,
      "step": 29970
    },
    {
      "entropy": 5.9440093517303465,
      "epoch": 2.9963512770530314,
      "grad_norm": 1.0078125,
      "learning_rate": 0.000411459691913917,
      "loss": 5.6262,
      "mean_token_accuracy": 0.18612994402647018,
      "num_tokens": 58637323.0,
      "step": 29975
    },
    {
      "entropy": 6.022060585021973,
      "epoch": 2.99685110211426,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041143131404227913,
      "loss": 5.521,
      "mean_token_accuracy": 0.18842182457447051,
      "num_tokens": 58646395.0,
      "step": 29980
    },
    {
      "entropy": 6.000775671005249,
      "epoch": 2.9973509271754883,
      "grad_norm": 1.125,
      "learning_rate": 0.0004114029327380645,
      "loss": 5.4617,
      "mean_token_accuracy": 0.19264070093631744,
      "num_tokens": 58655966.0,
      "step": 29985
    },
    {
      "entropy": 6.00114860534668,
      "epoch": 2.997850752236717,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004113745480019872,
      "loss": 5.4315,
      "mean_token_accuracy": 0.19758610129356385,
      "num_tokens": 58666030.0,
      "step": 29990
    },
    {
      "entropy": 5.97138295173645,
      "epoch": 2.9983505772979457,
      "grad_norm": 1.21875,
      "learning_rate": 0.0004113461598347614,
      "loss": 5.4644,
      "mean_token_accuracy": 0.1926880344748497,
      "num_tokens": 58674702.0,
      "step": 29995
    },
    {
      "entropy": 6.014968347549439,
      "epoch": 2.9988504023591744,
      "grad_norm": 1.0,
      "learning_rate": 0.00041131776823710134,
      "loss": 5.622,
      "mean_token_accuracy": 0.17895247787237167,
      "num_tokens": 58686103.0,
      "step": 30000
    },
    {
      "epoch": 2.9988504023591744,
      "eval_entropy": 5.850692069828433,
      "eval_loss": 5.647598743438721,
      "eval_mean_token_accuracy": 0.1909934875479266,
      "eval_num_tokens": 58686103.0,
      "eval_runtime": 23.8821,
      "eval_samples_per_second": 1299.929,
      "eval_steps_per_second": 162.507,
      "step": 30000
    },
    {
      "entropy": 6.030605268478394,
      "epoch": 2.9993502274204027,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004112893732097213,
      "loss": 5.509,
      "mean_token_accuracy": 0.19309124499559402,
      "num_tokens": 58695173.0,
      "step": 30005
    },
    {
      "entropy": 6.029189920425415,
      "epoch": 2.9998500524816314,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004112609747533356,
      "loss": 5.6264,
      "mean_token_accuracy": 0.17998196184635162,
      "num_tokens": 58705014.0,
      "step": 30010
    },
    {
      "entropy": 5.997407383388943,
      "epoch": 3.000299895036737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004112325728686589,
      "loss": 5.5045,
      "mean_token_accuracy": 0.19685902694861093,
      "num_tokens": 58714656.0,
      "step": 30015
    },
    {
      "entropy": 6.063377428054809,
      "epoch": 3.000799720097966,
      "grad_norm": 1.203125,
      "learning_rate": 0.00041120416755640574,
      "loss": 5.4201,
      "mean_token_accuracy": 0.19232141077518464,
      "num_tokens": 58725157.0,
      "step": 30020
    },
    {
      "entropy": 6.021855258941651,
      "epoch": 3.001299545159194,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041117575881729075,
      "loss": 5.5506,
      "mean_token_accuracy": 0.18044667840003967,
      "num_tokens": 58734229.0,
      "step": 30025
    },
    {
      "entropy": 5.964059066772461,
      "epoch": 3.001799370220423,
      "grad_norm": 1.15625,
      "learning_rate": 0.00041114734665202875,
      "loss": 5.3543,
      "mean_token_accuracy": 0.1979486122727394,
      "num_tokens": 58743374.0,
      "step": 30030
    },
    {
      "entropy": 5.966495418548584,
      "epoch": 3.0022991952816516,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004111189310613344,
      "loss": 5.3586,
      "mean_token_accuracy": 0.19842974990606307,
      "num_tokens": 58752846.0,
      "step": 30035
    },
    {
      "entropy": 6.034306764602661,
      "epoch": 3.00279902034288,
      "grad_norm": 1.046875,
      "learning_rate": 0.00041109051204592294,
      "loss": 5.5036,
      "mean_token_accuracy": 0.18650470077991485,
      "num_tokens": 58763407.0,
      "step": 30040
    },
    {
      "entropy": 5.990033340454102,
      "epoch": 3.0032988454041085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041106208960650913,
      "loss": 5.4308,
      "mean_token_accuracy": 0.18966882824897766,
      "num_tokens": 58774207.0,
      "step": 30045
    },
    {
      "entropy": 5.957082653045655,
      "epoch": 3.003798670465337,
      "grad_norm": 1.171875,
      "learning_rate": 0.00041103366374380817,
      "loss": 5.469,
      "mean_token_accuracy": 0.18997844606637954,
      "num_tokens": 58783270.0,
      "step": 30050
    },
    {
      "entropy": 5.939996147155762,
      "epoch": 3.004298495526566,
      "grad_norm": 1.109375,
      "learning_rate": 0.00041100523445853525,
      "loss": 5.4618,
      "mean_token_accuracy": 0.19548161774873735,
      "num_tokens": 58793007.0,
      "step": 30055
    },
    {
      "entropy": 6.088390445709228,
      "epoch": 3.004798320587794,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004109768017514056,
      "loss": 5.5725,
      "mean_token_accuracy": 0.18739273697137832,
      "num_tokens": 58802604.0,
      "step": 30060
    },
    {
      "entropy": 6.030565071105957,
      "epoch": 3.005298145649023,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004109483656231346,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1913704514503479,
      "num_tokens": 58811739.0,
      "step": 30065
    },
    {
      "entropy": 5.976911783218384,
      "epoch": 3.0057979707102516,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00041091992607443777,
      "loss": 5.5077,
      "mean_token_accuracy": 0.1935662269592285,
      "num_tokens": 58821079.0,
      "step": 30070
    },
    {
      "entropy": 6.076920127868652,
      "epoch": 3.00629779577148,
      "grad_norm": 1.125,
      "learning_rate": 0.00041089148310603053,
      "loss": 5.5665,
      "mean_token_accuracy": 0.185103115439415,
      "num_tokens": 58831030.0,
      "step": 30075
    },
    {
      "entropy": 6.0436419486999515,
      "epoch": 3.0067976208327085,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004108630367186285,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19681727588176728,
      "num_tokens": 58840241.0,
      "step": 30080
    },
    {
      "entropy": 6.004191827774048,
      "epoch": 3.007297445893937,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041083458691294755,
      "loss": 5.5015,
      "mean_token_accuracy": 0.18672581315040587,
      "num_tokens": 58849411.0,
      "step": 30085
    },
    {
      "entropy": 5.971017122268677,
      "epoch": 3.007797270955166,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004108061336897032,
      "loss": 5.4638,
      "mean_token_accuracy": 0.19412903785705565,
      "num_tokens": 58860478.0,
      "step": 30090
    },
    {
      "entropy": 6.0727095127105715,
      "epoch": 3.008297096016394,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004107776770496116,
      "loss": 5.5533,
      "mean_token_accuracy": 0.18970211446285248,
      "num_tokens": 58871700.0,
      "step": 30095
    },
    {
      "entropy": 6.082751941680908,
      "epoch": 3.008796921077623,
      "grad_norm": 1.125,
      "learning_rate": 0.0004107492169933885,
      "loss": 5.5697,
      "mean_token_accuracy": 0.17710745632648467,
      "num_tokens": 58882154.0,
      "step": 30100
    },
    {
      "entropy": 6.029969882965088,
      "epoch": 3.0092967461388516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00041072075352175004,
      "loss": 5.4712,
      "mean_token_accuracy": 0.1902804359793663,
      "num_tokens": 58892012.0,
      "step": 30105
    },
    {
      "entropy": 6.009149503707886,
      "epoch": 3.00979657120008,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041069228663541234,
      "loss": 5.4551,
      "mean_token_accuracy": 0.19363496899604798,
      "num_tokens": 58901196.0,
      "step": 30110
    },
    {
      "entropy": 5.96389102935791,
      "epoch": 3.0102963962613085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004106638163350917,
      "loss": 5.4082,
      "mean_token_accuracy": 0.19631130993366241,
      "num_tokens": 58911474.0,
      "step": 30115
    },
    {
      "entropy": 5.993680429458618,
      "epoch": 3.010796221322537,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004106353426215043,
      "loss": 5.3951,
      "mean_token_accuracy": 0.1942811965942383,
      "num_tokens": 58920440.0,
      "step": 30120
    },
    {
      "entropy": 6.005222797393799,
      "epoch": 3.0112960463837655,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00041060686549536656,
      "loss": 5.5111,
      "mean_token_accuracy": 0.18773080110549928,
      "num_tokens": 58929652.0,
      "step": 30125
    },
    {
      "entropy": 6.004841613769531,
      "epoch": 3.011795871444994,
      "grad_norm": 1.078125,
      "learning_rate": 0.000410578384957395,
      "loss": 5.429,
      "mean_token_accuracy": 0.19671889990568162,
      "num_tokens": 58940216.0,
      "step": 30130
    },
    {
      "entropy": 5.971134090423584,
      "epoch": 3.012295696506223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004105499010083062,
      "loss": 5.4364,
      "mean_token_accuracy": 0.20153791159391404,
      "num_tokens": 58949517.0,
      "step": 30135
    },
    {
      "entropy": 5.897588300704956,
      "epoch": 3.0127955215674516,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00041052141364881665,
      "loss": 5.4356,
      "mean_token_accuracy": 0.1977866366505623,
      "num_tokens": 58959044.0,
      "step": 30140
    },
    {
      "entropy": 6.014861011505127,
      "epoch": 3.01329534662868,
      "grad_norm": 1.09375,
      "learning_rate": 0.00041049292287964335,
      "loss": 5.4904,
      "mean_token_accuracy": 0.1937011808156967,
      "num_tokens": 58968352.0,
      "step": 30145
    },
    {
      "entropy": 5.916047143936157,
      "epoch": 3.0137951716899085,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041046442870150293,
      "loss": 5.3556,
      "mean_token_accuracy": 0.2039957046508789,
      "num_tokens": 58979082.0,
      "step": 30150
    },
    {
      "entropy": 5.9360755443572994,
      "epoch": 3.014294996751137,
      "grad_norm": 1.25,
      "learning_rate": 0.00041043593111511227,
      "loss": 5.4103,
      "mean_token_accuracy": 0.19594788551330566,
      "num_tokens": 58988181.0,
      "step": 30155
    },
    {
      "entropy": 5.9963761329650875,
      "epoch": 3.0147948218123655,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004104074301211885,
      "loss": 5.4922,
      "mean_token_accuracy": 0.19489680826663972,
      "num_tokens": 58999907.0,
      "step": 30160
    },
    {
      "entropy": 6.009028053283691,
      "epoch": 3.015294646873594,
      "grad_norm": 1.1875,
      "learning_rate": 0.00041037892572044863,
      "loss": 5.547,
      "mean_token_accuracy": 0.1875782772898674,
      "num_tokens": 59009293.0,
      "step": 30165
    },
    {
      "entropy": 5.965840435028076,
      "epoch": 3.015794471934823,
      "grad_norm": 1.21875,
      "learning_rate": 0.00041035041791360983,
      "loss": 5.4783,
      "mean_token_accuracy": 0.19045298397541047,
      "num_tokens": 59017079.0,
      "step": 30170
    },
    {
      "entropy": 5.994843244552612,
      "epoch": 3.0162942969960516,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004103219067013893,
      "loss": 5.4406,
      "mean_token_accuracy": 0.19507923126220703,
      "num_tokens": 59027008.0,
      "step": 30175
    },
    {
      "entropy": 5.996065092086792,
      "epoch": 3.01679412205728,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00041029339208450445,
      "loss": 5.503,
      "mean_token_accuracy": 0.18910137712955474,
      "num_tokens": 59036971.0,
      "step": 30180
    },
    {
      "entropy": 5.981302118301391,
      "epoch": 3.0172939471185085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004102648740636727,
      "loss": 5.4525,
      "mean_token_accuracy": 0.19139693826436996,
      "num_tokens": 59048264.0,
      "step": 30185
    },
    {
      "entropy": 5.981693887710572,
      "epoch": 3.017793772179737,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004102363526396115,
      "loss": 5.5685,
      "mean_token_accuracy": 0.1885414257645607,
      "num_tokens": 59057981.0,
      "step": 30190
    },
    {
      "entropy": 6.072077989578247,
      "epoch": 3.0182935972409655,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004102078278130384,
      "loss": 5.5868,
      "mean_token_accuracy": 0.18251278698444368,
      "num_tokens": 59068441.0,
      "step": 30195
    },
    {
      "entropy": 6.028339624404907,
      "epoch": 3.018793422302194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00041017929958467113,
      "loss": 5.5027,
      "mean_token_accuracy": 0.19216036945581436,
      "num_tokens": 59078544.0,
      "step": 30200
    },
    {
      "entropy": 5.987252616882325,
      "epoch": 3.019293247363423,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004101507679552276,
      "loss": 5.4957,
      "mean_token_accuracy": 0.19467708468437195,
      "num_tokens": 59088063.0,
      "step": 30205
    },
    {
      "entropy": 5.983454704284668,
      "epoch": 3.0197930724246516,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004101222329254255,
      "loss": 5.5023,
      "mean_token_accuracy": 0.18630857467651368,
      "num_tokens": 59098003.0,
      "step": 30210
    },
    {
      "entropy": 6.03258171081543,
      "epoch": 3.02029289748588,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004100936944959828,
      "loss": 5.4998,
      "mean_token_accuracy": 0.1843581274151802,
      "num_tokens": 59108595.0,
      "step": 30215
    },
    {
      "entropy": 6.037194299697876,
      "epoch": 3.0207927225471085,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004100651526676174,
      "loss": 5.5674,
      "mean_token_accuracy": 0.18235031515359879,
      "num_tokens": 59118453.0,
      "step": 30220
    },
    {
      "entropy": 5.971056652069092,
      "epoch": 3.021292547608337,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004100366074410476,
      "loss": 5.4443,
      "mean_token_accuracy": 0.19176288545131684,
      "num_tokens": 59128005.0,
      "step": 30225
    },
    {
      "entropy": 6.052638244628906,
      "epoch": 3.0217923726695655,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00041000805881699143,
      "loss": 5.5012,
      "mean_token_accuracy": 0.18907860964536666,
      "num_tokens": 59137694.0,
      "step": 30230
    },
    {
      "entropy": 5.969610691070557,
      "epoch": 3.022292197730794,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004099795067961673,
      "loss": 5.5017,
      "mean_token_accuracy": 0.19380665123462676,
      "num_tokens": 59147715.0,
      "step": 30235
    },
    {
      "entropy": 5.978658294677734,
      "epoch": 3.022792022792023,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004099509513792935,
      "loss": 5.4904,
      "mean_token_accuracy": 0.19054348915815353,
      "num_tokens": 59157151.0,
      "step": 30240
    },
    {
      "entropy": 6.105502414703369,
      "epoch": 3.0232918478532516,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040992239256708844,
      "loss": 5.5576,
      "mean_token_accuracy": 0.18616366088390351,
      "num_tokens": 59167349.0,
      "step": 30245
    },
    {
      "entropy": 5.92483811378479,
      "epoch": 3.02379167291448,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004098938303602708,
      "loss": 5.4351,
      "mean_token_accuracy": 0.19585761725902556,
      "num_tokens": 59176412.0,
      "step": 30250
    },
    {
      "entropy": 5.886774158477783,
      "epoch": 3.0242914979757085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040986526475955904,
      "loss": 5.4379,
      "mean_token_accuracy": 0.19260677695274353,
      "num_tokens": 59186050.0,
      "step": 30255
    },
    {
      "entropy": 5.958836793899536,
      "epoch": 3.024791323036937,
      "grad_norm": 1.109375,
      "learning_rate": 0.000409836695765672,
      "loss": 5.4382,
      "mean_token_accuracy": 0.19863022714853287,
      "num_tokens": 59196185.0,
      "step": 30260
    },
    {
      "entropy": 6.0170591354370115,
      "epoch": 3.0252911480981655,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040980812337932835,
      "loss": 5.4945,
      "mean_token_accuracy": 0.18979185521602632,
      "num_tokens": 59208079.0,
      "step": 30265
    },
    {
      "entropy": 6.002722692489624,
      "epoch": 3.025790973159394,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040977954760124693,
      "loss": 5.4654,
      "mean_token_accuracy": 0.18891989439725876,
      "num_tokens": 59216968.0,
      "step": 30270
    },
    {
      "entropy": 5.993566656112671,
      "epoch": 3.026290798220623,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040975096843214677,
      "loss": 5.415,
      "mean_token_accuracy": 0.1977374732494354,
      "num_tokens": 59226481.0,
      "step": 30275
    },
    {
      "entropy": 5.927251148223877,
      "epoch": 3.0267906232818516,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040972238587274695,
      "loss": 5.3931,
      "mean_token_accuracy": 0.19563258290290833,
      "num_tokens": 59235794.0,
      "step": 30280
    },
    {
      "entropy": 5.96700086593628,
      "epoch": 3.02729044834308,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040969379992376664,
      "loss": 5.5024,
      "mean_token_accuracy": 0.18695476651191711,
      "num_tokens": 59245691.0,
      "step": 30285
    },
    {
      "entropy": 6.050683259963989,
      "epoch": 3.0277902734043085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004096652105859249,
      "loss": 5.5025,
      "mean_token_accuracy": 0.1870974987745285,
      "num_tokens": 59255591.0,
      "step": 30290
    },
    {
      "entropy": 5.9751667976379395,
      "epoch": 3.028290098465537,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004096366178599412,
      "loss": 5.369,
      "mean_token_accuracy": 0.20313187092542648,
      "num_tokens": 59266183.0,
      "step": 30295
    },
    {
      "entropy": 5.998158645629883,
      "epoch": 3.0287899235267655,
      "grad_norm": 1.125,
      "learning_rate": 0.0004096080217465347,
      "loss": 5.5269,
      "mean_token_accuracy": 0.18803392797708512,
      "num_tokens": 59276241.0,
      "step": 30300
    },
    {
      "entropy": 5.938171815872193,
      "epoch": 3.029289748587994,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004095794222464252,
      "loss": 5.4647,
      "mean_token_accuracy": 0.19286196827888488,
      "num_tokens": 59286089.0,
      "step": 30305
    },
    {
      "entropy": 5.973509311676025,
      "epoch": 3.029789573649223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004095508193603319,
      "loss": 5.5277,
      "mean_token_accuracy": 0.185259547829628,
      "num_tokens": 59295122.0,
      "step": 30310
    },
    {
      "entropy": 6.001608753204346,
      "epoch": 3.0302893987104516,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004095222130889747,
      "loss": 5.5446,
      "mean_token_accuracy": 0.18400773108005525,
      "num_tokens": 59304277.0,
      "step": 30315
    },
    {
      "entropy": 5.959620523452759,
      "epoch": 3.03078922377168,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040949360343307324,
      "loss": 5.38,
      "mean_token_accuracy": 0.19379789233207703,
      "num_tokens": 59315560.0,
      "step": 30320
    },
    {
      "entropy": 5.963546323776245,
      "epoch": 3.0312890488329085,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004094649903933472,
      "loss": 5.4422,
      "mean_token_accuracy": 0.19200112223625182,
      "num_tokens": 59325238.0,
      "step": 30325
    },
    {
      "entropy": 5.9470704078674315,
      "epoch": 3.031788873894137,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040943637397051683,
      "loss": 5.3858,
      "mean_token_accuracy": 0.20288256853818892,
      "num_tokens": 59335800.0,
      "step": 30330
    },
    {
      "entropy": 5.979463291168213,
      "epoch": 3.0322886989553655,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040940775416530164,
      "loss": 5.5399,
      "mean_token_accuracy": 0.18505657613277435,
      "num_tokens": 59345854.0,
      "step": 30335
    },
    {
      "entropy": 5.951516151428223,
      "epoch": 3.032788524016594,
      "grad_norm": 1.25,
      "learning_rate": 0.00040937913097842204,
      "loss": 5.3855,
      "mean_token_accuracy": 0.2013171672821045,
      "num_tokens": 59354789.0,
      "step": 30340
    },
    {
      "entropy": 6.029586553573608,
      "epoch": 3.033288349077823,
      "grad_norm": 1.265625,
      "learning_rate": 0.0004093505044105982,
      "loss": 5.5469,
      "mean_token_accuracy": 0.18387166559696197,
      "num_tokens": 59364205.0,
      "step": 30345
    },
    {
      "entropy": 6.015558910369873,
      "epoch": 3.033788174139051,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040932187446255004,
      "loss": 5.5287,
      "mean_token_accuracy": 0.1883456215262413,
      "num_tokens": 59375037.0,
      "step": 30350
    },
    {
      "entropy": 5.930894565582276,
      "epoch": 3.03428799920028,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004092932411349982,
      "loss": 5.3485,
      "mean_token_accuracy": 0.20393879860639572,
      "num_tokens": 59384820.0,
      "step": 30355
    },
    {
      "entropy": 5.925769567489624,
      "epoch": 3.0347878242615085,
      "grad_norm": 1.140625,
      "learning_rate": 0.000409264604428663,
      "loss": 5.4602,
      "mean_token_accuracy": 0.19299192428588868,
      "num_tokens": 59394571.0,
      "step": 30360
    },
    {
      "entropy": 5.95932149887085,
      "epoch": 3.035287649322737,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004092359643442648,
      "loss": 5.5191,
      "mean_token_accuracy": 0.18958512395620347,
      "num_tokens": 59405758.0,
      "step": 30365
    },
    {
      "entropy": 5.983124780654907,
      "epoch": 3.0357874743839655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004092073208825243,
      "loss": 5.4097,
      "mean_token_accuracy": 0.19802277237176896,
      "num_tokens": 59415091.0,
      "step": 30370
    },
    {
      "entropy": 6.016193437576294,
      "epoch": 3.036287299445194,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004091786740441622,
      "loss": 5.5402,
      "mean_token_accuracy": 0.1825582802295685,
      "num_tokens": 59423867.0,
      "step": 30375
    },
    {
      "entropy": 6.004262495040893,
      "epoch": 3.036787124506423,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040915002382989915,
      "loss": 5.4817,
      "mean_token_accuracy": 0.18756930083036422,
      "num_tokens": 59433594.0,
      "step": 30380
    },
    {
      "entropy": 5.971362447738647,
      "epoch": 3.037286949567651,
      "grad_norm": 1.1953125,
      "learning_rate": 0.000409121370240456,
      "loss": 5.4929,
      "mean_token_accuracy": 0.19448105245828629,
      "num_tokens": 59442411.0,
      "step": 30385
    },
    {
      "entropy": 5.940098619461059,
      "epoch": 3.03778677462888,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004090927132765537,
      "loss": 5.4234,
      "mean_token_accuracy": 0.1975530728697777,
      "num_tokens": 59451507.0,
      "step": 30390
    },
    {
      "entropy": 5.982679128646851,
      "epoch": 3.0382865996901085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040906405293891323,
      "loss": 5.4866,
      "mean_token_accuracy": 0.19379795491695403,
      "num_tokens": 59462497.0,
      "step": 30395
    },
    {
      "entropy": 5.997652292251587,
      "epoch": 3.038786424751337,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004090353892282557,
      "loss": 5.5341,
      "mean_token_accuracy": 0.1855769082903862,
      "num_tokens": 59472574.0,
      "step": 30400
    },
    {
      "entropy": 5.9626383781433105,
      "epoch": 3.0392862498125655,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004090067221453024,
      "loss": 5.3453,
      "mean_token_accuracy": 0.1995700016617775,
      "num_tokens": 59482508.0,
      "step": 30405
    },
    {
      "entropy": 6.097778844833374,
      "epoch": 3.039786074873794,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004089780516907743,
      "loss": 5.6139,
      "mean_token_accuracy": 0.18135517984628677,
      "num_tokens": 59493032.0,
      "step": 30410
    },
    {
      "entropy": 6.029333972930909,
      "epoch": 3.040285899935023,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040894937786539287,
      "loss": 5.5499,
      "mean_token_accuracy": 0.1911746382713318,
      "num_tokens": 59502769.0,
      "step": 30415
    },
    {
      "entropy": 5.924650621414185,
      "epoch": 3.040785724996251,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040892070066987966,
      "loss": 5.4616,
      "mean_token_accuracy": 0.18673219680786132,
      "num_tokens": 59512392.0,
      "step": 30420
    },
    {
      "entropy": 5.954359149932861,
      "epoch": 3.04128555005748,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040889202010495595,
      "loss": 5.4373,
      "mean_token_accuracy": 0.1942717909812927,
      "num_tokens": 59521936.0,
      "step": 30425
    },
    {
      "entropy": 6.021448564529419,
      "epoch": 3.0417853751187085,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004088633361713436,
      "loss": 5.5468,
      "mean_token_accuracy": 0.17846721708774566,
      "num_tokens": 59531460.0,
      "step": 30430
    },
    {
      "entropy": 6.007031536102295,
      "epoch": 3.042285200179937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040883464886976414,
      "loss": 5.5451,
      "mean_token_accuracy": 0.1857913002371788,
      "num_tokens": 59541898.0,
      "step": 30435
    },
    {
      "entropy": 5.982570219039917,
      "epoch": 3.0427850252411655,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004088059582009393,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18607865124940873,
      "num_tokens": 59552960.0,
      "step": 30440
    },
    {
      "entropy": 5.958920812606811,
      "epoch": 3.043284850302394,
      "grad_norm": 1.078125,
      "learning_rate": 0.000408777264165591,
      "loss": 5.4634,
      "mean_token_accuracy": 0.18935898691415787,
      "num_tokens": 59563655.0,
      "step": 30445
    },
    {
      "entropy": 5.989930725097656,
      "epoch": 3.043784675363623,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040874856676444116,
      "loss": 5.459,
      "mean_token_accuracy": 0.20893779546022415,
      "num_tokens": 59573554.0,
      "step": 30450
    },
    {
      "entropy": 6.017723178863525,
      "epoch": 3.044284500424851,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040871986599821177,
      "loss": 5.4614,
      "mean_token_accuracy": 0.19616154879331588,
      "num_tokens": 59582720.0,
      "step": 30455
    },
    {
      "entropy": 5.9622289657592775,
      "epoch": 3.04478432548608,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004086911618676251,
      "loss": 5.4631,
      "mean_token_accuracy": 0.194390606880188,
      "num_tokens": 59591589.0,
      "step": 30460
    },
    {
      "entropy": 6.064439535140991,
      "epoch": 3.0452841505473085,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000408662454373403,
      "loss": 5.574,
      "mean_token_accuracy": 0.18266939967870713,
      "num_tokens": 59601575.0,
      "step": 30465
    },
    {
      "entropy": 5.954619312286377,
      "epoch": 3.045783975608537,
      "grad_norm": 1.2265625,
      "learning_rate": 0.000408633743516268,
      "loss": 5.4277,
      "mean_token_accuracy": 0.20069137513637542,
      "num_tokens": 59610842.0,
      "step": 30470
    },
    {
      "entropy": 6.064501237869263,
      "epoch": 3.0462838006697655,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040860502929694246,
      "loss": 5.6264,
      "mean_token_accuracy": 0.1847103714942932,
      "num_tokens": 59622054.0,
      "step": 30475
    },
    {
      "entropy": 6.035087156295776,
      "epoch": 3.046783625730994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040857631171614873,
      "loss": 5.4542,
      "mean_token_accuracy": 0.20064682513475418,
      "num_tokens": 59631516.0,
      "step": 30480
    },
    {
      "entropy": 6.021350145339966,
      "epoch": 3.047283450792223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040854759077460934,
      "loss": 5.4374,
      "mean_token_accuracy": 0.19266469180583953,
      "num_tokens": 59641999.0,
      "step": 30485
    },
    {
      "entropy": 5.950778341293335,
      "epoch": 3.047783275853451,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040851886647304703,
      "loss": 5.4092,
      "mean_token_accuracy": 0.19618182480335236,
      "num_tokens": 59651107.0,
      "step": 30490
    },
    {
      "entropy": 5.992885589599609,
      "epoch": 3.04828310091468,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0004084901388121843,
      "loss": 5.496,
      "mean_token_accuracy": 0.19318617433309554,
      "num_tokens": 59661075.0,
      "step": 30495
    },
    {
      "entropy": 5.8981458187103275,
      "epoch": 3.0487829259759085,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040846140779274406,
      "loss": 5.3529,
      "mean_token_accuracy": 0.19944416880607604,
      "num_tokens": 59671449.0,
      "step": 30500
    },
    {
      "entropy": 6.044550848007202,
      "epoch": 3.049282751037137,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040843267341544913,
      "loss": 5.5813,
      "mean_token_accuracy": 0.18360042870044707,
      "num_tokens": 59680824.0,
      "step": 30505
    },
    {
      "entropy": 6.046122407913208,
      "epoch": 3.0497825760983654,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040840393568102246,
      "loss": 5.468,
      "mean_token_accuracy": 0.18739250004291536,
      "num_tokens": 59689293.0,
      "step": 30510
    },
    {
      "entropy": 5.977842712402344,
      "epoch": 3.050282401159594,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004083751945901871,
      "loss": 5.5901,
      "mean_token_accuracy": 0.179594948887825,
      "num_tokens": 59698693.0,
      "step": 30515
    },
    {
      "entropy": 6.01931848526001,
      "epoch": 3.050782226220823,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040834645014366623,
      "loss": 5.5176,
      "mean_token_accuracy": 0.17952830642461776,
      "num_tokens": 59709047.0,
      "step": 30520
    },
    {
      "entropy": 6.136570262908935,
      "epoch": 3.051282051282051,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004083177023421829,
      "loss": 5.52,
      "mean_token_accuracy": 0.18502768576145173,
      "num_tokens": 59719392.0,
      "step": 30525
    },
    {
      "entropy": 5.933347415924072,
      "epoch": 3.05178187634328,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004082889511864606,
      "loss": 5.3611,
      "mean_token_accuracy": 0.1960995689034462,
      "num_tokens": 59728736.0,
      "step": 30530
    },
    {
      "entropy": 5.999807691574096,
      "epoch": 3.0522817014045085,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040826019667722244,
      "loss": 5.5522,
      "mean_token_accuracy": 0.18433912098407745,
      "num_tokens": 59738816.0,
      "step": 30535
    },
    {
      "entropy": 6.018436241149902,
      "epoch": 3.052781526465737,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004082314388151921,
      "loss": 5.4638,
      "mean_token_accuracy": 0.1980438709259033,
      "num_tokens": 59747785.0,
      "step": 30540
    },
    {
      "entropy": 6.023428535461425,
      "epoch": 3.0532813515269654,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004082026776010931,
      "loss": 5.4808,
      "mean_token_accuracy": 0.19526928812265396,
      "num_tokens": 59757429.0,
      "step": 30545
    },
    {
      "entropy": 6.057660007476807,
      "epoch": 3.053781176588194,
      "grad_norm": 1.125,
      "learning_rate": 0.0004081739130356489,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18965768814086914,
      "num_tokens": 59767996.0,
      "step": 30550
    },
    {
      "entropy": 5.9598394393920895,
      "epoch": 3.054281001649423,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040814514511958335,
      "loss": 5.5211,
      "mean_token_accuracy": 0.19030696600675584,
      "num_tokens": 59778487.0,
      "step": 30555
    },
    {
      "entropy": 5.955559539794922,
      "epoch": 3.054780826710651,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040811637385362024,
      "loss": 5.4001,
      "mean_token_accuracy": 0.20734449625015258,
      "num_tokens": 59788929.0,
      "step": 30560
    },
    {
      "entropy": 6.01504921913147,
      "epoch": 3.05528065177188,
      "grad_norm": 1.296875,
      "learning_rate": 0.0004080875992384834,
      "loss": 5.5036,
      "mean_token_accuracy": 0.1927926242351532,
      "num_tokens": 59797548.0,
      "step": 30565
    },
    {
      "entropy": 6.033369874954223,
      "epoch": 3.0557804768331085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004080588212748968,
      "loss": 5.5214,
      "mean_token_accuracy": 0.19239838868379594,
      "num_tokens": 59807811.0,
      "step": 30570
    },
    {
      "entropy": 5.94784026145935,
      "epoch": 3.056280301894337,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040803003996358445,
      "loss": 5.3726,
      "mean_token_accuracy": 0.20236717462539672,
      "num_tokens": 59818118.0,
      "step": 30575
    },
    {
      "entropy": 5.970272874832153,
      "epoch": 3.0567801269555654,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040800125530527064,
      "loss": 5.5108,
      "mean_token_accuracy": 0.1887582316994667,
      "num_tokens": 59827749.0,
      "step": 30580
    },
    {
      "entropy": 6.001681280136109,
      "epoch": 3.057279952016794,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040797246730067937,
      "loss": 5.4402,
      "mean_token_accuracy": 0.20499979555606843,
      "num_tokens": 59836519.0,
      "step": 30585
    },
    {
      "entropy": 5.9798869609832765,
      "epoch": 3.057779777078023,
      "grad_norm": 1.125,
      "learning_rate": 0.0004079436759505351,
      "loss": 5.4372,
      "mean_token_accuracy": 0.19689854085445405,
      "num_tokens": 59847201.0,
      "step": 30590
    },
    {
      "entropy": 5.973022317886352,
      "epoch": 3.058279602139251,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0004079148812555621,
      "loss": 5.3963,
      "mean_token_accuracy": 0.2057907223701477,
      "num_tokens": 59855366.0,
      "step": 30595
    },
    {
      "entropy": 5.867964029312134,
      "epoch": 3.05877942720048,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040788608321648496,
      "loss": 5.3577,
      "mean_token_accuracy": 0.20114185959100722,
      "num_tokens": 59864820.0,
      "step": 30600
    },
    {
      "entropy": 6.020285034179688,
      "epoch": 3.0592792522617085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004078572818340281,
      "loss": 5.5184,
      "mean_token_accuracy": 0.19121836125850677,
      "num_tokens": 59875757.0,
      "step": 30605
    },
    {
      "entropy": 6.125909233093262,
      "epoch": 3.059779077322937,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040782847710891623,
      "loss": 5.6321,
      "mean_token_accuracy": 0.18233714401721954,
      "num_tokens": 59885039.0,
      "step": 30610
    },
    {
      "entropy": 6.029363346099854,
      "epoch": 3.0602789023841654,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040779966904187407,
      "loss": 5.5106,
      "mean_token_accuracy": 0.1825796663761139,
      "num_tokens": 59894516.0,
      "step": 30615
    },
    {
      "entropy": 6.053947782516479,
      "epoch": 3.060778727445394,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040777085763362634,
      "loss": 5.5736,
      "mean_token_accuracy": 0.18760325759649277,
      "num_tokens": 59905004.0,
      "step": 30620
    },
    {
      "entropy": 6.102583646774292,
      "epoch": 3.061278552506623,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040774204288489805,
      "loss": 5.573,
      "mean_token_accuracy": 0.18139005154371263,
      "num_tokens": 59914814.0,
      "step": 30625
    },
    {
      "entropy": 6.096563148498535,
      "epoch": 3.061778377567851,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004077132247964142,
      "loss": 5.5627,
      "mean_token_accuracy": 0.18816659301519395,
      "num_tokens": 59924451.0,
      "step": 30630
    },
    {
      "entropy": 5.948845958709716,
      "epoch": 3.06227820262908,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040768440336889965,
      "loss": 5.4651,
      "mean_token_accuracy": 0.19535193592309952,
      "num_tokens": 59934607.0,
      "step": 30635
    },
    {
      "entropy": 5.811318206787109,
      "epoch": 3.0627780276903085,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004076555786030797,
      "loss": 5.3318,
      "mean_token_accuracy": 0.20427658557891845,
      "num_tokens": 59945172.0,
      "step": 30640
    },
    {
      "entropy": 6.068199682235718,
      "epoch": 3.063277852751537,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004076267504996795,
      "loss": 5.4443,
      "mean_token_accuracy": 0.19637513011693955,
      "num_tokens": 59955022.0,
      "step": 30645
    },
    {
      "entropy": 6.030975389480591,
      "epoch": 3.0637776778127654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004075979190594244,
      "loss": 5.5128,
      "mean_token_accuracy": 0.1946458712220192,
      "num_tokens": 59965666.0,
      "step": 30650
    },
    {
      "entropy": 6.033237791061401,
      "epoch": 3.064277502873994,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040756908428303984,
      "loss": 5.4972,
      "mean_token_accuracy": 0.19392139613628387,
      "num_tokens": 59975069.0,
      "step": 30655
    },
    {
      "entropy": 6.0857728004455565,
      "epoch": 3.064777327935223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040754024617125115,
      "loss": 5.5616,
      "mean_token_accuracy": 0.1910158321261406,
      "num_tokens": 59986100.0,
      "step": 30660
    },
    {
      "entropy": 5.9344337463378904,
      "epoch": 3.065277152996451,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000407511404724784,
      "loss": 5.4466,
      "mean_token_accuracy": 0.19695445895195007,
      "num_tokens": 59995929.0,
      "step": 30665
    },
    {
      "entropy": 6.004751491546631,
      "epoch": 3.06577697805768,
      "grad_norm": 1.15625,
      "learning_rate": 0.000407482559944364,
      "loss": 5.5143,
      "mean_token_accuracy": 0.19022249579429626,
      "num_tokens": 60005526.0,
      "step": 30670
    },
    {
      "entropy": 5.964439249038696,
      "epoch": 3.0662768031189085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004074537118307169,
      "loss": 5.3466,
      "mean_token_accuracy": 0.203788460791111,
      "num_tokens": 60014833.0,
      "step": 30675
    },
    {
      "entropy": 6.010521078109742,
      "epoch": 3.0667766281801367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004074248603845684,
      "loss": 5.5622,
      "mean_token_accuracy": 0.18321948647499084,
      "num_tokens": 60024579.0,
      "step": 30680
    },
    {
      "entropy": 6.025298547744751,
      "epoch": 3.0672764532413654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040739600560664457,
      "loss": 5.5601,
      "mean_token_accuracy": 0.18332307040691376,
      "num_tokens": 60035162.0,
      "step": 30685
    },
    {
      "entropy": 6.019966554641724,
      "epoch": 3.067776278302594,
      "grad_norm": 1.203125,
      "learning_rate": 0.00040736714749767136,
      "loss": 5.4162,
      "mean_token_accuracy": 0.1951565131545067,
      "num_tokens": 60044894.0,
      "step": 30690
    },
    {
      "entropy": 6.0357190608978275,
      "epoch": 3.068276103363823,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004073382860583747,
      "loss": 5.481,
      "mean_token_accuracy": 0.1934180125594139,
      "num_tokens": 60054441.0,
      "step": 30695
    },
    {
      "entropy": 6.029822874069214,
      "epoch": 3.068775928425051,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004073094212894809,
      "loss": 5.5025,
      "mean_token_accuracy": 0.19655562490224837,
      "num_tokens": 60063701.0,
      "step": 30700
    },
    {
      "entropy": 6.011656761169434,
      "epoch": 3.06927575348628,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000407280553191716,
      "loss": 5.5086,
      "mean_token_accuracy": 0.1797790825366974,
      "num_tokens": 60073455.0,
      "step": 30705
    },
    {
      "entropy": 5.948316287994385,
      "epoch": 3.0697755785475085,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004072516817658065,
      "loss": 5.4489,
      "mean_token_accuracy": 0.19318412393331527,
      "num_tokens": 60083898.0,
      "step": 30710
    },
    {
      "entropy": 6.032275772094726,
      "epoch": 3.0702754036087367,
      "grad_norm": 1.265625,
      "learning_rate": 0.00040722280701247865,
      "loss": 5.5637,
      "mean_token_accuracy": 0.18973370641469955,
      "num_tokens": 60093605.0,
      "step": 30715
    },
    {
      "entropy": 6.146461057662964,
      "epoch": 3.0707752286699654,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040719392893245903,
      "loss": 5.5892,
      "mean_token_accuracy": 0.18367933183908464,
      "num_tokens": 60103240.0,
      "step": 30720
    },
    {
      "entropy": 5.912042188644409,
      "epoch": 3.071275053731194,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004071650475264742,
      "loss": 5.4145,
      "mean_token_accuracy": 0.1943598136305809,
      "num_tokens": 60112850.0,
      "step": 30725
    },
    {
      "entropy": 5.959503984451294,
      "epoch": 3.071774878792423,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040713616279525075,
      "loss": 5.4122,
      "mean_token_accuracy": 0.19782360047101974,
      "num_tokens": 60122565.0,
      "step": 30730
    },
    {
      "entropy": 6.04570198059082,
      "epoch": 3.072274703853651,
      "grad_norm": 1.25,
      "learning_rate": 0.00040710727473951546,
      "loss": 5.4444,
      "mean_token_accuracy": 0.18810687810182572,
      "num_tokens": 60132239.0,
      "step": 30735
    },
    {
      "entropy": 5.972931814193726,
      "epoch": 3.07277452891488,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004070783833599951,
      "loss": 5.4814,
      "mean_token_accuracy": 0.1892448455095291,
      "num_tokens": 60141993.0,
      "step": 30740
    },
    {
      "entropy": 5.980544900894165,
      "epoch": 3.0732743539761085,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004070494886574166,
      "loss": 5.5042,
      "mean_token_accuracy": 0.19280767887830735,
      "num_tokens": 60151350.0,
      "step": 30745
    },
    {
      "entropy": 6.022108268737793,
      "epoch": 3.0737741790373367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000407020590632507,
      "loss": 5.5055,
      "mean_token_accuracy": 0.18922441750764846,
      "num_tokens": 60161120.0,
      "step": 30750
    },
    {
      "entropy": 6.0213648796081545,
      "epoch": 3.0742740040985654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004069916892859933,
      "loss": 5.5136,
      "mean_token_accuracy": 0.19891939014196397,
      "num_tokens": 60170762.0,
      "step": 30755
    },
    {
      "entropy": 5.9689947128295895,
      "epoch": 3.074773829159794,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004069627846186026,
      "loss": 5.4558,
      "mean_token_accuracy": 0.19468655437231064,
      "num_tokens": 60180729.0,
      "step": 30760
    },
    {
      "entropy": 5.9586786270141605,
      "epoch": 3.075273654221023,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040693387663106223,
      "loss": 5.4478,
      "mean_token_accuracy": 0.1976270481944084,
      "num_tokens": 60190640.0,
      "step": 30765
    },
    {
      "entropy": 5.973170471191406,
      "epoch": 3.075773479282251,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004069049653240996,
      "loss": 5.4377,
      "mean_token_accuracy": 0.1922805055975914,
      "num_tokens": 60200163.0,
      "step": 30770
    },
    {
      "entropy": 6.070269536972046,
      "epoch": 3.07627330434348,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004068760506984418,
      "loss": 5.5422,
      "mean_token_accuracy": 0.1869718223810196,
      "num_tokens": 60209469.0,
      "step": 30775
    },
    {
      "entropy": 5.894810962677002,
      "epoch": 3.0767731294047085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004068471327548166,
      "loss": 5.3495,
      "mean_token_accuracy": 0.20090408474206925,
      "num_tokens": 60218950.0,
      "step": 30780
    },
    {
      "entropy": 6.029858207702636,
      "epoch": 3.0772729544659367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040681821149395146,
      "loss": 5.4934,
      "mean_token_accuracy": 0.1886974185705185,
      "num_tokens": 60228428.0,
      "step": 30785
    },
    {
      "entropy": 5.978582906723022,
      "epoch": 3.0777727795271654,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040678928691657395,
      "loss": 5.4621,
      "mean_token_accuracy": 0.19608478397130966,
      "num_tokens": 60237558.0,
      "step": 30790
    },
    {
      "entropy": 5.967212200164795,
      "epoch": 3.078272604588394,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040676035902341195,
      "loss": 5.4156,
      "mean_token_accuracy": 0.19483400285243987,
      "num_tokens": 60246282.0,
      "step": 30795
    },
    {
      "entropy": 5.883442354202271,
      "epoch": 3.078772429649623,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004067314278151933,
      "loss": 5.2987,
      "mean_token_accuracy": 0.20551279336214065,
      "num_tokens": 60255678.0,
      "step": 30800
    },
    {
      "entropy": 5.914713573455811,
      "epoch": 3.079272254710851,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040670249329264577,
      "loss": 5.3914,
      "mean_token_accuracy": 0.21275220811367035,
      "num_tokens": 60266900.0,
      "step": 30805
    },
    {
      "entropy": 6.075995397567749,
      "epoch": 3.07977207977208,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004066735554564973,
      "loss": 5.536,
      "mean_token_accuracy": 0.18987746983766557,
      "num_tokens": 60276974.0,
      "step": 30810
    },
    {
      "entropy": 6.125108146667481,
      "epoch": 3.0802719048333085,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004066446143074762,
      "loss": 5.5668,
      "mean_token_accuracy": 0.18497112542390823,
      "num_tokens": 60285830.0,
      "step": 30815
    },
    {
      "entropy": 5.963441658020019,
      "epoch": 3.0807717298945367,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004066156698463104,
      "loss": 5.4073,
      "mean_token_accuracy": 0.19244676530361177,
      "num_tokens": 60295178.0,
      "step": 30820
    },
    {
      "entropy": 5.967557001113891,
      "epoch": 3.0812715549557654,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0004065867220737282,
      "loss": 5.4826,
      "mean_token_accuracy": 0.19200898408889772,
      "num_tokens": 60305514.0,
      "step": 30825
    },
    {
      "entropy": 5.890975570678711,
      "epoch": 3.081771380016994,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040655777099045794,
      "loss": 5.3848,
      "mean_token_accuracy": 0.19554249495267867,
      "num_tokens": 60314407.0,
      "step": 30830
    },
    {
      "entropy": 5.957977819442749,
      "epoch": 3.082271205078223,
      "grad_norm": 1.125,
      "learning_rate": 0.0004065288165972281,
      "loss": 5.3596,
      "mean_token_accuracy": 0.20213798880577089,
      "num_tokens": 60322711.0,
      "step": 30835
    },
    {
      "entropy": 5.995135402679443,
      "epoch": 3.082771030139451,
      "grad_norm": 1.109375,
      "learning_rate": 0.000406499858894767,
      "loss": 5.563,
      "mean_token_accuracy": 0.1891617640852928,
      "num_tokens": 60332718.0,
      "step": 30840
    },
    {
      "entropy": 5.938999271392822,
      "epoch": 3.08327085520068,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004064708978838033,
      "loss": 5.4149,
      "mean_token_accuracy": 0.1949584126472473,
      "num_tokens": 60343513.0,
      "step": 30845
    },
    {
      "entropy": 5.949363279342651,
      "epoch": 3.0837706802619085,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040644193356506565,
      "loss": 5.3922,
      "mean_token_accuracy": 0.19715723693370818,
      "num_tokens": 60353374.0,
      "step": 30850
    },
    {
      "entropy": 6.011474609375,
      "epoch": 3.0842705053231367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004064129659392828,
      "loss": 5.4643,
      "mean_token_accuracy": 0.1954444393515587,
      "num_tokens": 60364303.0,
      "step": 30855
    },
    {
      "entropy": 5.969551992416382,
      "epoch": 3.0847703303843654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004063839950071835,
      "loss": 5.366,
      "mean_token_accuracy": 0.19368677735328674,
      "num_tokens": 60374282.0,
      "step": 30860
    },
    {
      "entropy": 6.036526346206665,
      "epoch": 3.085270155445594,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040635502076949665,
      "loss": 5.5393,
      "mean_token_accuracy": 0.1832280859351158,
      "num_tokens": 60384711.0,
      "step": 30865
    },
    {
      "entropy": 5.972257280349732,
      "epoch": 3.0857699805068224,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040632604322695135,
      "loss": 5.5287,
      "mean_token_accuracy": 0.18705183416604995,
      "num_tokens": 60394681.0,
      "step": 30870
    },
    {
      "entropy": 5.972606420516968,
      "epoch": 3.086269805568051,
      "grad_norm": 1.25,
      "learning_rate": 0.0004062970623802766,
      "loss": 5.4716,
      "mean_token_accuracy": 0.18866738379001619,
      "num_tokens": 60404226.0,
      "step": 30875
    },
    {
      "entropy": 6.017214632034301,
      "epoch": 3.08676963062928,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004062680782302015,
      "loss": 5.4885,
      "mean_token_accuracy": 0.1954237103462219,
      "num_tokens": 60413657.0,
      "step": 30880
    },
    {
      "entropy": 6.004489183425903,
      "epoch": 3.0872694556905085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040623909077745535,
      "loss": 5.4188,
      "mean_token_accuracy": 0.18996618390083314,
      "num_tokens": 60423574.0,
      "step": 30885
    },
    {
      "entropy": 5.964860677719116,
      "epoch": 3.0877692807517367,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004062101000227674,
      "loss": 5.4286,
      "mean_token_accuracy": 0.19782689958810806,
      "num_tokens": 60433349.0,
      "step": 30890
    },
    {
      "entropy": 5.973463678359986,
      "epoch": 3.0882691058129654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004061811059668672,
      "loss": 5.557,
      "mean_token_accuracy": 0.18203110992908478,
      "num_tokens": 60442435.0,
      "step": 30895
    },
    {
      "entropy": 5.969042634963989,
      "epoch": 3.088768930874194,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004061521086104841,
      "loss": 5.4286,
      "mean_token_accuracy": 0.19608967304229735,
      "num_tokens": 60451386.0,
      "step": 30900
    },
    {
      "entropy": 6.001058721542359,
      "epoch": 3.0892687559354224,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040612310795434765,
      "loss": 5.4493,
      "mean_token_accuracy": 0.18629872798919678,
      "num_tokens": 60460780.0,
      "step": 30905
    },
    {
      "entropy": 6.006352949142456,
      "epoch": 3.089768580996651,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040609410399918754,
      "loss": 5.5103,
      "mean_token_accuracy": 0.18906446993350984,
      "num_tokens": 60470563.0,
      "step": 30910
    },
    {
      "entropy": 5.957207345962525,
      "epoch": 3.09026840605788,
      "grad_norm": 1.125,
      "learning_rate": 0.00040606509674573354,
      "loss": 5.4257,
      "mean_token_accuracy": 0.20174840092658997,
      "num_tokens": 60481170.0,
      "step": 30915
    },
    {
      "entropy": 6.104734086990357,
      "epoch": 3.0907682311191085,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004060360861947155,
      "loss": 5.5677,
      "mean_token_accuracy": 0.18357649594545364,
      "num_tokens": 60490700.0,
      "step": 30920
    },
    {
      "entropy": 6.004500675201416,
      "epoch": 3.0912680561803367,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040600707234686315,
      "loss": 5.4325,
      "mean_token_accuracy": 0.19631525725126267,
      "num_tokens": 60500262.0,
      "step": 30925
    },
    {
      "entropy": 5.980014896392822,
      "epoch": 3.0917678812415654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040597805520290657,
      "loss": 5.555,
      "mean_token_accuracy": 0.18620217591524124,
      "num_tokens": 60510453.0,
      "step": 30930
    },
    {
      "entropy": 6.069574022293091,
      "epoch": 3.092267706302794,
      "grad_norm": 1.25,
      "learning_rate": 0.0004059490347635759,
      "loss": 5.4763,
      "mean_token_accuracy": 0.1913177490234375,
      "num_tokens": 60520069.0,
      "step": 30935
    },
    {
      "entropy": 6.022724771499634,
      "epoch": 3.0927675313640224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040592001102960116,
      "loss": 5.5336,
      "mean_token_accuracy": 0.1911913424730301,
      "num_tokens": 60530193.0,
      "step": 30940
    },
    {
      "entropy": 6.015159940719604,
      "epoch": 3.093267356425251,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004058909840017126,
      "loss": 5.5218,
      "mean_token_accuracy": 0.19492380917072297,
      "num_tokens": 60539768.0,
      "step": 30945
    },
    {
      "entropy": 5.9368555545806885,
      "epoch": 3.09376718148648,
      "grad_norm": 1.125,
      "learning_rate": 0.0004058619536806407,
      "loss": 5.4631,
      "mean_token_accuracy": 0.19210435897111894,
      "num_tokens": 60549488.0,
      "step": 30950
    },
    {
      "entropy": 5.984806251525879,
      "epoch": 3.0942670065477085,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004058329200671155,
      "loss": 5.4981,
      "mean_token_accuracy": 0.19363281577825547,
      "num_tokens": 60559611.0,
      "step": 30955
    },
    {
      "entropy": 6.008777236938476,
      "epoch": 3.0947668316089367,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040580388316186785,
      "loss": 5.468,
      "mean_token_accuracy": 0.19958247244358063,
      "num_tokens": 60569091.0,
      "step": 30960
    },
    {
      "entropy": 6.055223560333252,
      "epoch": 3.0952666566701654,
      "grad_norm": 1.125,
      "learning_rate": 0.0004057748429656281,
      "loss": 5.644,
      "mean_token_accuracy": 0.18091422617435454,
      "num_tokens": 60581173.0,
      "step": 30965
    },
    {
      "entropy": 5.9480588912963865,
      "epoch": 3.095766481731394,
      "grad_norm": 1.203125,
      "learning_rate": 0.000405745799479127,
      "loss": 5.3708,
      "mean_token_accuracy": 0.2008643075823784,
      "num_tokens": 60590677.0,
      "step": 30970
    },
    {
      "entropy": 6.117230415344238,
      "epoch": 3.0962663067926224,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004057167527030952,
      "loss": 5.5769,
      "mean_token_accuracy": 0.1839364007115364,
      "num_tokens": 60599861.0,
      "step": 30975
    },
    {
      "entropy": 6.015076780319214,
      "epoch": 3.096766131853851,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040568770263826355,
      "loss": 5.5353,
      "mean_token_accuracy": 0.18713659793138504,
      "num_tokens": 60609061.0,
      "step": 30980
    },
    {
      "entropy": 5.9668725490570065,
      "epoch": 3.0972659569150798,
      "grad_norm": 1.125,
      "learning_rate": 0.0004056586492853629,
      "loss": 5.4755,
      "mean_token_accuracy": 0.19587499499320984,
      "num_tokens": 60618935.0,
      "step": 30985
    },
    {
      "entropy": 6.065490245819092,
      "epoch": 3.0977657819763085,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0004056295926451243,
      "loss": 5.4778,
      "mean_token_accuracy": 0.20377503335475922,
      "num_tokens": 60628549.0,
      "step": 30990
    },
    {
      "entropy": 5.961002397537231,
      "epoch": 3.0982656070375367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040560053271827866,
      "loss": 5.4355,
      "mean_token_accuracy": 0.19493339508771895,
      "num_tokens": 60638253.0,
      "step": 30995
    },
    {
      "entropy": 5.982589483261108,
      "epoch": 3.0987654320987654,
      "grad_norm": 1.125,
      "learning_rate": 0.00040557146950555734,
      "loss": 5.4451,
      "mean_token_accuracy": 0.1984004110097885,
      "num_tokens": 60647575.0,
      "step": 31000
    },
    {
      "entropy": 5.95160870552063,
      "epoch": 3.099265257159994,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004055424030076914,
      "loss": 5.4878,
      "mean_token_accuracy": 0.18906869888305664,
      "num_tokens": 60658020.0,
      "step": 31005
    },
    {
      "entropy": 5.967223405838013,
      "epoch": 3.0997650822212224,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040551333322541225,
      "loss": 5.4042,
      "mean_token_accuracy": 0.19389633536338807,
      "num_tokens": 60668330.0,
      "step": 31010
    },
    {
      "entropy": 6.007176256179809,
      "epoch": 3.100264907282451,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004054842601594512,
      "loss": 5.5125,
      "mean_token_accuracy": 0.19099824130535126,
      "num_tokens": 60679573.0,
      "step": 31015
    },
    {
      "entropy": 5.967203760147095,
      "epoch": 3.1007647323436798,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040545518381053964,
      "loss": 5.3875,
      "mean_token_accuracy": 0.2010224238038063,
      "num_tokens": 60689363.0,
      "step": 31020
    },
    {
      "entropy": 5.906686782836914,
      "epoch": 3.1012645574049085,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040542610417940926,
      "loss": 5.4195,
      "mean_token_accuracy": 0.19787361472845078,
      "num_tokens": 60699120.0,
      "step": 31025
    },
    {
      "entropy": 5.985939645767212,
      "epoch": 3.1017643824661367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004053970212667917,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19330466091632842,
      "num_tokens": 60709235.0,
      "step": 31030
    },
    {
      "entropy": 5.991764211654663,
      "epoch": 3.1022642075273654,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004053679350734186,
      "loss": 5.4217,
      "mean_token_accuracy": 0.1949220657348633,
      "num_tokens": 60718674.0,
      "step": 31035
    },
    {
      "entropy": 5.939059495925903,
      "epoch": 3.102764032588594,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040533884560002196,
      "loss": 5.3634,
      "mean_token_accuracy": 0.2028452515602112,
      "num_tokens": 60727456.0,
      "step": 31040
    },
    {
      "entropy": 5.901403188705444,
      "epoch": 3.1032638576498224,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040530975284733327,
      "loss": 5.3485,
      "mean_token_accuracy": 0.19595887660980224,
      "num_tokens": 60736556.0,
      "step": 31045
    },
    {
      "entropy": 6.021659088134766,
      "epoch": 3.103763682711051,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004052806568160849,
      "loss": 5.468,
      "mean_token_accuracy": 0.18901082277297973,
      "num_tokens": 60746868.0,
      "step": 31050
    },
    {
      "entropy": 6.058588266372681,
      "epoch": 3.1042635077722798,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040525155750700863,
      "loss": 5.5721,
      "mean_token_accuracy": 0.18371787816286086,
      "num_tokens": 60757200.0,
      "step": 31055
    },
    {
      "entropy": 5.963805198669434,
      "epoch": 3.1047633328335085,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040522245492083674,
      "loss": 5.4259,
      "mean_token_accuracy": 0.1992632657289505,
      "num_tokens": 60767966.0,
      "step": 31060
    },
    {
      "entropy": 5.949222087860107,
      "epoch": 3.1052631578947367,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040519334905830134,
      "loss": 5.4115,
      "mean_token_accuracy": 0.19230178147554397,
      "num_tokens": 60777556.0,
      "step": 31065
    },
    {
      "entropy": 5.975813722610473,
      "epoch": 3.1057629829559654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004051642399201349,
      "loss": 5.4315,
      "mean_token_accuracy": 0.20370898991823197,
      "num_tokens": 60787601.0,
      "step": 31070
    },
    {
      "entropy": 5.950928497314453,
      "epoch": 3.106262808017194,
      "grad_norm": 1.125,
      "learning_rate": 0.00040513512750706956,
      "loss": 5.4471,
      "mean_token_accuracy": 0.19431656897068023,
      "num_tokens": 60797023.0,
      "step": 31075
    },
    {
      "entropy": 6.054173755645752,
      "epoch": 3.1067626330784224,
      "grad_norm": 1.09375,
      "learning_rate": 0.000405106011819838,
      "loss": 5.5104,
      "mean_token_accuracy": 0.18883266150951386,
      "num_tokens": 60807030.0,
      "step": 31080
    },
    {
      "entropy": 5.963209342956543,
      "epoch": 3.107262458139651,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040507689285917254,
      "loss": 5.452,
      "mean_token_accuracy": 0.1903813973069191,
      "num_tokens": 60816397.0,
      "step": 31085
    },
    {
      "entropy": 6.027845287322998,
      "epoch": 3.1077622832008798,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040504777062580597,
      "loss": 5.5196,
      "mean_token_accuracy": 0.18293308168649675,
      "num_tokens": 60826859.0,
      "step": 31090
    },
    {
      "entropy": 5.874378776550293,
      "epoch": 3.1082621082621085,
      "grad_norm": 1.09375,
      "learning_rate": 0.000405018645120471,
      "loss": 5.3385,
      "mean_token_accuracy": 0.20255100429058076,
      "num_tokens": 60837010.0,
      "step": 31095
    },
    {
      "entropy": 5.9676813125610355,
      "epoch": 3.1087619333233367,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004049895163439003,
      "loss": 5.4756,
      "mean_token_accuracy": 0.18521452993154525,
      "num_tokens": 60845502.0,
      "step": 31100
    },
    {
      "entropy": 5.943050670623779,
      "epoch": 3.1092617583845654,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040496038429682684,
      "loss": 5.4539,
      "mean_token_accuracy": 0.19593143314123154,
      "num_tokens": 60856783.0,
      "step": 31105
    },
    {
      "entropy": 5.963225507736206,
      "epoch": 3.109761583445794,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040493124897998363,
      "loss": 5.4649,
      "mean_token_accuracy": 0.19092865735292436,
      "num_tokens": 60866560.0,
      "step": 31110
    },
    {
      "entropy": 5.9622515678405765,
      "epoch": 3.1102614085070224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004049021103941036,
      "loss": 5.4841,
      "mean_token_accuracy": 0.19011935442686081,
      "num_tokens": 60876677.0,
      "step": 31115
    },
    {
      "entropy": 6.006877803802491,
      "epoch": 3.110761233568251,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040487296853991984,
      "loss": 5.5008,
      "mean_token_accuracy": 0.18963577449321747,
      "num_tokens": 60886943.0,
      "step": 31120
    },
    {
      "entropy": 6.08043532371521,
      "epoch": 3.1112610586294798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004048438234181657,
      "loss": 5.5043,
      "mean_token_accuracy": 0.1905519962310791,
      "num_tokens": 60895740.0,
      "step": 31125
    },
    {
      "entropy": 5.960463285446167,
      "epoch": 3.1117608836907085,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040481467502957426,
      "loss": 5.4009,
      "mean_token_accuracy": 0.198969104886055,
      "num_tokens": 60905354.0,
      "step": 31130
    },
    {
      "entropy": 6.0036561489105225,
      "epoch": 3.1122607087519367,
      "grad_norm": 0.98828125,
      "learning_rate": 0.000404785523374879,
      "loss": 5.4725,
      "mean_token_accuracy": 0.19711743742227555,
      "num_tokens": 60915040.0,
      "step": 31135
    },
    {
      "entropy": 5.976144647598266,
      "epoch": 3.1127605338131654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040475636845481347,
      "loss": 5.5127,
      "mean_token_accuracy": 0.18820121884346008,
      "num_tokens": 60925023.0,
      "step": 31140
    },
    {
      "entropy": 6.012701511383057,
      "epoch": 3.113260358874394,
      "grad_norm": 1.125,
      "learning_rate": 0.00040472721027011107,
      "loss": 5.4914,
      "mean_token_accuracy": 0.18952536582946777,
      "num_tokens": 60935058.0,
      "step": 31145
    },
    {
      "entropy": 5.996023750305175,
      "epoch": 3.1137601839356224,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004046980488215054,
      "loss": 5.4731,
      "mean_token_accuracy": 0.1979200705885887,
      "num_tokens": 60945241.0,
      "step": 31150
    },
    {
      "entropy": 5.995792818069458,
      "epoch": 3.114260008996851,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040466888410973026,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19191270619630812,
      "num_tokens": 60955402.0,
      "step": 31155
    },
    {
      "entropy": 6.001737022399903,
      "epoch": 3.1147598340580798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004046397161355193,
      "loss": 5.4594,
      "mean_token_accuracy": 0.19466737359762193,
      "num_tokens": 60964837.0,
      "step": 31160
    },
    {
      "entropy": 6.014556121826172,
      "epoch": 3.1152596591193085,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004046105448996064,
      "loss": 5.5842,
      "mean_token_accuracy": 0.18034202456474305,
      "num_tokens": 60975441.0,
      "step": 31165
    },
    {
      "entropy": 5.997760581970215,
      "epoch": 3.1157594841805367,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004045813704027257,
      "loss": 5.543,
      "mean_token_accuracy": 0.19279961735010148,
      "num_tokens": 60985883.0,
      "step": 31170
    },
    {
      "entropy": 5.9966123580932615,
      "epoch": 3.1162593092417654,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004045521926456109,
      "loss": 5.4583,
      "mean_token_accuracy": 0.19740626215934753,
      "num_tokens": 60996545.0,
      "step": 31175
    },
    {
      "entropy": 5.949262952804565,
      "epoch": 3.116759134302994,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004045230116289964,
      "loss": 5.4499,
      "mean_token_accuracy": 0.19802429974079133,
      "num_tokens": 61006805.0,
      "step": 31180
    },
    {
      "entropy": 6.005507755279541,
      "epoch": 3.1172589593642224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040449382735361614,
      "loss": 5.5136,
      "mean_token_accuracy": 0.18675364255905152,
      "num_tokens": 61017579.0,
      "step": 31185
    },
    {
      "entropy": 6.005194282531738,
      "epoch": 3.117758784425451,
      "grad_norm": 1.125,
      "learning_rate": 0.0004044646398202046,
      "loss": 5.484,
      "mean_token_accuracy": 0.19588165134191513,
      "num_tokens": 61027939.0,
      "step": 31190
    },
    {
      "entropy": 6.026525688171387,
      "epoch": 3.1182586094866798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.000404435449029496,
      "loss": 5.4416,
      "mean_token_accuracy": 0.1963612988591194,
      "num_tokens": 61038445.0,
      "step": 31195
    },
    {
      "entropy": 6.008419466018677,
      "epoch": 3.118758434547908,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004044062549822248,
      "loss": 5.5243,
      "mean_token_accuracy": 0.18976651281118392,
      "num_tokens": 61049434.0,
      "step": 31200
    },
    {
      "entropy": 6.017463207244873,
      "epoch": 3.1192582596091367,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004043770576791255,
      "loss": 5.5225,
      "mean_token_accuracy": 0.1880732610821724,
      "num_tokens": 61058405.0,
      "step": 31205
    },
    {
      "entropy": 5.977043867111206,
      "epoch": 3.1197580846703654,
      "grad_norm": 1.234375,
      "learning_rate": 0.0004043478571209328,
      "loss": 5.4484,
      "mean_token_accuracy": 0.1959860146045685,
      "num_tokens": 61068639.0,
      "step": 31210
    },
    {
      "entropy": 5.9850122928619385,
      "epoch": 3.120257909731594,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004043186533083812,
      "loss": 5.4553,
      "mean_token_accuracy": 0.20035867094993592,
      "num_tokens": 61078336.0,
      "step": 31215
    },
    {
      "entropy": 6.025325918197632,
      "epoch": 3.1207577347928224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040428944624220557,
      "loss": 5.5029,
      "mean_token_accuracy": 0.19290732592344284,
      "num_tokens": 61087940.0,
      "step": 31220
    },
    {
      "entropy": 5.995983695983886,
      "epoch": 3.121257559854051,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004042602359231409,
      "loss": 5.4865,
      "mean_token_accuracy": 0.18704578429460525,
      "num_tokens": 61098688.0,
      "step": 31225
    },
    {
      "entropy": 5.97469801902771,
      "epoch": 3.1217573849152798,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004042310223519218,
      "loss": 5.4407,
      "mean_token_accuracy": 0.198845773935318,
      "num_tokens": 61108066.0,
      "step": 31230
    },
    {
      "entropy": 5.925361967086792,
      "epoch": 3.122257209976508,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004042018055292835,
      "loss": 5.3946,
      "mean_token_accuracy": 0.20092176049947738,
      "num_tokens": 61117009.0,
      "step": 31235
    },
    {
      "entropy": 6.042518091201782,
      "epoch": 3.1227570350377367,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0004041725854559609,
      "loss": 5.5018,
      "mean_token_accuracy": 0.1959958165884018,
      "num_tokens": 61127285.0,
      "step": 31240
    },
    {
      "entropy": 5.990555238723755,
      "epoch": 3.1232568600989654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040414336213268943,
      "loss": 5.4433,
      "mean_token_accuracy": 0.1966252699494362,
      "num_tokens": 61136894.0,
      "step": 31245
    },
    {
      "entropy": 6.072669792175293,
      "epoch": 3.123756685160194,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004041141355602041,
      "loss": 5.6275,
      "mean_token_accuracy": 0.18639821857213973,
      "num_tokens": 61147308.0,
      "step": 31250
    },
    {
      "entropy": 6.007371807098389,
      "epoch": 3.1242565102214224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004040849057392405,
      "loss": 5.4491,
      "mean_token_accuracy": 0.18802175521850586,
      "num_tokens": 61157021.0,
      "step": 31255
    },
    {
      "entropy": 6.072435188293457,
      "epoch": 3.124756335282651,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004040556726705337,
      "loss": 5.5484,
      "mean_token_accuracy": 0.19272898882627487,
      "num_tokens": 61166741.0,
      "step": 31260
    },
    {
      "entropy": 5.995324182510376,
      "epoch": 3.1252561603438798,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0004040264363548195,
      "loss": 5.3547,
      "mean_token_accuracy": 0.19878922402858734,
      "num_tokens": 61175951.0,
      "step": 31265
    },
    {
      "entropy": 5.972700023651123,
      "epoch": 3.125755985405108,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004039971967928333,
      "loss": 5.4639,
      "mean_token_accuracy": 0.1947838097810745,
      "num_tokens": 61185950.0,
      "step": 31270
    },
    {
      "entropy": 5.896227073669434,
      "epoch": 3.1262558104663367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004039679539853108,
      "loss": 5.3405,
      "mean_token_accuracy": 0.20707086026668547,
      "num_tokens": 61195305.0,
      "step": 31275
    },
    {
      "entropy": 5.943407201766968,
      "epoch": 3.1267556355275654,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004039387079329878,
      "loss": 5.4938,
      "mean_token_accuracy": 0.19513427466154099,
      "num_tokens": 61204640.0,
      "step": 31280
    },
    {
      "entropy": 5.928497457504273,
      "epoch": 3.127255460588794,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004039094586366001,
      "loss": 5.3896,
      "mean_token_accuracy": 0.203026182949543,
      "num_tokens": 61214644.0,
      "step": 31285
    },
    {
      "entropy": 5.98371000289917,
      "epoch": 3.1277552856500224,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040388020609688366,
      "loss": 5.399,
      "mean_token_accuracy": 0.2034771904349327,
      "num_tokens": 61223970.0,
      "step": 31290
    },
    {
      "entropy": 5.961815261840821,
      "epoch": 3.128255110711251,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040385095031457435,
      "loss": 5.5143,
      "mean_token_accuracy": 0.19046448022127152,
      "num_tokens": 61233503.0,
      "step": 31295
    },
    {
      "entropy": 5.976302528381348,
      "epoch": 3.1287549357724798,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004038216912904082,
      "loss": 5.3673,
      "mean_token_accuracy": 0.1967140644788742,
      "num_tokens": 61243342.0,
      "step": 31300
    },
    {
      "entropy": 5.833013343811035,
      "epoch": 3.129254760833708,
      "grad_norm": 1.0078125,
      "learning_rate": 0.0004037924290251215,
      "loss": 5.3043,
      "mean_token_accuracy": 0.20247678458690643,
      "num_tokens": 61253456.0,
      "step": 31305
    },
    {
      "entropy": 5.9864119529724125,
      "epoch": 3.1297545858949367,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004037631635194505,
      "loss": 5.4994,
      "mean_token_accuracy": 0.19144104421138763,
      "num_tokens": 61262604.0,
      "step": 31310
    },
    {
      "entropy": 6.002148675918579,
      "epoch": 3.1302544109561654,
      "grad_norm": 1.390625,
      "learning_rate": 0.0004037338947741313,
      "loss": 5.3406,
      "mean_token_accuracy": 0.20016105622053146,
      "num_tokens": 61272127.0,
      "step": 31315
    },
    {
      "entropy": 5.974411201477051,
      "epoch": 3.130754236017394,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040370462278990055,
      "loss": 5.5033,
      "mean_token_accuracy": 0.18566085696220397,
      "num_tokens": 61280808.0,
      "step": 31320
    },
    {
      "entropy": 5.968501091003418,
      "epoch": 3.1312540610786224,
      "grad_norm": 1.28125,
      "learning_rate": 0.0004036753475674945,
      "loss": 5.5371,
      "mean_token_accuracy": 0.19026852399110794,
      "num_tokens": 61290047.0,
      "step": 31325
    },
    {
      "entropy": 6.098319864273071,
      "epoch": 3.131753886139851,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00040364606910764996,
      "loss": 5.5739,
      "mean_token_accuracy": 0.1881429597735405,
      "num_tokens": 61300140.0,
      "step": 31330
    },
    {
      "entropy": 6.1101240634918215,
      "epoch": 3.1322537112010798,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004036167874111034,
      "loss": 5.6217,
      "mean_token_accuracy": 0.1777956172823906,
      "num_tokens": 61310790.0,
      "step": 31335
    },
    {
      "entropy": 6.079577875137329,
      "epoch": 3.132753536262308,
      "grad_norm": 1.125,
      "learning_rate": 0.00040358750247859146,
      "loss": 5.5199,
      "mean_token_accuracy": 0.19071124643087387,
      "num_tokens": 61320835.0,
      "step": 31340
    },
    {
      "entropy": 5.96912899017334,
      "epoch": 3.1332533613235367,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004035582143108511,
      "loss": 5.3966,
      "mean_token_accuracy": 0.20028960704803467,
      "num_tokens": 61330585.0,
      "step": 31345
    },
    {
      "entropy": 5.8844043731689455,
      "epoch": 3.1337531863847654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040352892290861916,
      "loss": 5.3886,
      "mean_token_accuracy": 0.21352204978466033,
      "num_tokens": 61341089.0,
      "step": 31350
    },
    {
      "entropy": 5.916169595718384,
      "epoch": 3.134253011445994,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0004034996282726326,
      "loss": 5.4338,
      "mean_token_accuracy": 0.19236407727003096,
      "num_tokens": 61349796.0,
      "step": 31355
    },
    {
      "entropy": 6.043759202957153,
      "epoch": 3.1347528365072224,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040347033040362835,
      "loss": 5.5337,
      "mean_token_accuracy": 0.19310266077518462,
      "num_tokens": 61359345.0,
      "step": 31360
    },
    {
      "entropy": 6.068906641006469,
      "epoch": 3.135252661568451,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040344102930234385,
      "loss": 5.5118,
      "mean_token_accuracy": 0.18501905351877213,
      "num_tokens": 61369462.0,
      "step": 31365
    },
    {
      "entropy": 5.9947288036346436,
      "epoch": 3.1357524866296798,
      "grad_norm": 1.21875,
      "learning_rate": 0.00040341172496951596,
      "loss": 5.438,
      "mean_token_accuracy": 0.19270329773426056,
      "num_tokens": 61378170.0,
      "step": 31370
    },
    {
      "entropy": 5.943666887283325,
      "epoch": 3.136252311690908,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004033824174058821,
      "loss": 5.5065,
      "mean_token_accuracy": 0.1973063349723816,
      "num_tokens": 61388307.0,
      "step": 31375
    },
    {
      "entropy": 5.957320213317871,
      "epoch": 3.1367521367521367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004033531066121797,
      "loss": 5.4326,
      "mean_token_accuracy": 0.19526997208595276,
      "num_tokens": 61398351.0,
      "step": 31380
    },
    {
      "entropy": 5.959526872634887,
      "epoch": 3.1372519618133654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040332379258914614,
      "loss": 5.4336,
      "mean_token_accuracy": 0.19409244060516356,
      "num_tokens": 61406893.0,
      "step": 31385
    },
    {
      "entropy": 6.0558812618255615,
      "epoch": 3.1377517868745937,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000403294475337519,
      "loss": 5.5626,
      "mean_token_accuracy": 0.1920332431793213,
      "num_tokens": 61416655.0,
      "step": 31390
    },
    {
      "entropy": 5.941634702682495,
      "epoch": 3.1382516119358224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004032651548580359,
      "loss": 5.4064,
      "mean_token_accuracy": 0.19905501902103423,
      "num_tokens": 61425655.0,
      "step": 31395
    },
    {
      "entropy": 5.927916145324707,
      "epoch": 3.138751436997051,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040323583115143446,
      "loss": 5.3731,
      "mean_token_accuracy": 0.2067493885755539,
      "num_tokens": 61435820.0,
      "step": 31400
    },
    {
      "entropy": 5.996844720840454,
      "epoch": 3.1392512620582798,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040320650421845256,
      "loss": 5.4607,
      "mean_token_accuracy": 0.20048202723264694,
      "num_tokens": 61445906.0,
      "step": 31405
    },
    {
      "entropy": 5.980939865112305,
      "epoch": 3.139751087119508,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040317717405982797,
      "loss": 5.4093,
      "mean_token_accuracy": 0.1948748603463173,
      "num_tokens": 61455237.0,
      "step": 31410
    },
    {
      "entropy": 6.037376546859742,
      "epoch": 3.1402509121807367,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040314784067629867,
      "loss": 5.5243,
      "mean_token_accuracy": 0.18970895260572435,
      "num_tokens": 61464735.0,
      "step": 31415
    },
    {
      "entropy": 6.022462558746338,
      "epoch": 3.1407507372419654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040311850406860267,
      "loss": 5.4921,
      "mean_token_accuracy": 0.1899375006556511,
      "num_tokens": 61475187.0,
      "step": 31420
    },
    {
      "entropy": 5.988220453262329,
      "epoch": 3.1412505623031937,
      "grad_norm": 1.03125,
      "learning_rate": 0.00040308916423747803,
      "loss": 5.4833,
      "mean_token_accuracy": 0.1920618250966072,
      "num_tokens": 61485232.0,
      "step": 31425
    },
    {
      "entropy": 6.080350112915039,
      "epoch": 3.1417503873644224,
      "grad_norm": 1.1015625,
      "learning_rate": 0.000403059821183663,
      "loss": 5.6651,
      "mean_token_accuracy": 0.18806249946355819,
      "num_tokens": 61495601.0,
      "step": 31430
    },
    {
      "entropy": 5.996184253692627,
      "epoch": 3.142250212425651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004030304749078959,
      "loss": 5.4437,
      "mean_token_accuracy": 0.19302066117525102,
      "num_tokens": 61506659.0,
      "step": 31435
    },
    {
      "entropy": 5.912986469268799,
      "epoch": 3.1427500374868798,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040300112541091494,
      "loss": 5.4206,
      "mean_token_accuracy": 0.20081949681043626,
      "num_tokens": 61517376.0,
      "step": 31440
    },
    {
      "entropy": 5.974144840240479,
      "epoch": 3.143249862548108,
      "grad_norm": 1.171875,
      "learning_rate": 0.0004029717726934586,
      "loss": 5.4148,
      "mean_token_accuracy": 0.19888361990451814,
      "num_tokens": 61527001.0,
      "step": 31445
    },
    {
      "entropy": 6.027616739273071,
      "epoch": 3.1437496876093367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040294241675626534,
      "loss": 5.5313,
      "mean_token_accuracy": 0.1919981762766838,
      "num_tokens": 61537554.0,
      "step": 31450
    },
    {
      "entropy": 6.0610116481781,
      "epoch": 3.1442495126705654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004029130576000739,
      "loss": 5.474,
      "mean_token_accuracy": 0.19901783168315887,
      "num_tokens": 61546457.0,
      "step": 31455
    },
    {
      "entropy": 5.959495735168457,
      "epoch": 3.1447493377317937,
      "grad_norm": 1.078125,
      "learning_rate": 0.0004028836952256227,
      "loss": 5.5197,
      "mean_token_accuracy": 0.1866837352514267,
      "num_tokens": 61555872.0,
      "step": 31460
    },
    {
      "entropy": 5.976934623718262,
      "epoch": 3.1452491627930224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004028543296336508,
      "loss": 5.4338,
      "mean_token_accuracy": 0.19786111116409302,
      "num_tokens": 61565820.0,
      "step": 31465
    },
    {
      "entropy": 6.015416097640991,
      "epoch": 3.145748987854251,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040282496082489684,
      "loss": 5.4945,
      "mean_token_accuracy": 0.20036117881536483,
      "num_tokens": 61575344.0,
      "step": 31470
    },
    {
      "entropy": 5.912573909759521,
      "epoch": 3.1462488129154798,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040279558880009967,
      "loss": 5.3779,
      "mean_token_accuracy": 0.21140339970588684,
      "num_tokens": 61584906.0,
      "step": 31475
    },
    {
      "entropy": 5.858107614517212,
      "epoch": 3.146748637976708,
      "grad_norm": 1.125,
      "learning_rate": 0.0004027662135599984,
      "loss": 5.3438,
      "mean_token_accuracy": 0.2060943514108658,
      "num_tokens": 61594490.0,
      "step": 31480
    },
    {
      "entropy": 5.98683533668518,
      "epoch": 3.1472484630379367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004027368351053321,
      "loss": 5.4047,
      "mean_token_accuracy": 0.19936572164297103,
      "num_tokens": 61603280.0,
      "step": 31485
    },
    {
      "entropy": 6.008887672424317,
      "epoch": 3.1477482880991654,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004027074534368399,
      "loss": 5.4416,
      "mean_token_accuracy": 0.20299321562051773,
      "num_tokens": 61612474.0,
      "step": 31490
    },
    {
      "entropy": 5.8276454448699955,
      "epoch": 3.1482481131603937,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040267806855526103,
      "loss": 5.2773,
      "mean_token_accuracy": 0.20983189940452576,
      "num_tokens": 61622719.0,
      "step": 31495
    },
    {
      "entropy": 5.881341791152954,
      "epoch": 3.1487479382216224,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004026486804613347,
      "loss": 5.4171,
      "mean_token_accuracy": 0.19876565784215927,
      "num_tokens": 61633171.0,
      "step": 31500
    },
    {
      "entropy": 5.899614477157593,
      "epoch": 3.149247763282851,
      "grad_norm": 1.046875,
      "learning_rate": 0.00040261928915580065,
      "loss": 5.3627,
      "mean_token_accuracy": 0.20101428776979446,
      "num_tokens": 61642437.0,
      "step": 31505
    },
    {
      "entropy": 6.005756855010986,
      "epoch": 3.1497475883440798,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004025898946393979,
      "loss": 5.586,
      "mean_token_accuracy": 0.18486790955066681,
      "num_tokens": 61651884.0,
      "step": 31510
    },
    {
      "entropy": 5.988962936401367,
      "epoch": 3.150247413405308,
      "grad_norm": 0.98828125,
      "learning_rate": 0.0004025604969128663,
      "loss": 5.5075,
      "mean_token_accuracy": 0.18993653655052184,
      "num_tokens": 61661821.0,
      "step": 31515
    },
    {
      "entropy": 5.960079860687256,
      "epoch": 3.1507472384665367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040253109597694544,
      "loss": 5.4405,
      "mean_token_accuracy": 0.1946353167295456,
      "num_tokens": 61672363.0,
      "step": 31520
    },
    {
      "entropy": 5.986041593551636,
      "epoch": 3.1512470635277654,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040250169183237507,
      "loss": 5.4399,
      "mean_token_accuracy": 0.20062762200832368,
      "num_tokens": 61681318.0,
      "step": 31525
    },
    {
      "entropy": 5.97133321762085,
      "epoch": 3.1517468885889937,
      "grad_norm": 1.0625,
      "learning_rate": 0.00040247228447989485,
      "loss": 5.4902,
      "mean_token_accuracy": 0.18878630995750428,
      "num_tokens": 61690710.0,
      "step": 31530
    },
    {
      "entropy": 6.008075857162476,
      "epoch": 3.1522467136502224,
      "grad_norm": 1.2890625,
      "learning_rate": 0.00040244287392024485,
      "loss": 5.4878,
      "mean_token_accuracy": 0.19416080117225648,
      "num_tokens": 61700913.0,
      "step": 31535
    },
    {
      "entropy": 6.014084339141846,
      "epoch": 3.152746538711451,
      "grad_norm": 1.03125,
      "learning_rate": 0.0004024134601541649,
      "loss": 5.4586,
      "mean_token_accuracy": 0.19326173663139343,
      "num_tokens": 61710905.0,
      "step": 31540
    },
    {
      "entropy": 5.901244974136352,
      "epoch": 3.1532463637726798,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0004023840431823951,
      "loss": 5.4074,
      "mean_token_accuracy": 0.2017873078584671,
      "num_tokens": 61719952.0,
      "step": 31545
    },
    {
      "entropy": 5.983846616744995,
      "epoch": 3.153746188833908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004023546230056754,
      "loss": 5.3818,
      "mean_token_accuracy": 0.19737586379051208,
      "num_tokens": 61728948.0,
      "step": 31550
    },
    {
      "entropy": 6.069068431854248,
      "epoch": 3.1542460138951367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040232519962474635,
      "loss": 5.5912,
      "mean_token_accuracy": 0.1858069270849228,
      "num_tokens": 61738235.0,
      "step": 31555
    },
    {
      "entropy": 6.015467929840088,
      "epoch": 3.1547458389563654,
      "grad_norm": 1.28125,
      "learning_rate": 0.00040229577304034793,
      "loss": 5.5094,
      "mean_token_accuracy": 0.19658226817846297,
      "num_tokens": 61746934.0,
      "step": 31560
    },
    {
      "entropy": 5.984611082077026,
      "epoch": 3.1552456640175937,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00040226634325322063,
      "loss": 5.4843,
      "mean_token_accuracy": 0.19597231298685075,
      "num_tokens": 61756346.0,
      "step": 31565
    },
    {
      "entropy": 5.990927505493164,
      "epoch": 3.1557454890788224,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004022369102641049,
      "loss": 5.5215,
      "mean_token_accuracy": 0.19015911221504211,
      "num_tokens": 61765900.0,
      "step": 31570
    },
    {
      "entropy": 6.056782484054565,
      "epoch": 3.156245314140051,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040220747407374133,
      "loss": 5.5167,
      "mean_token_accuracy": 0.19033144414424896,
      "num_tokens": 61775657.0,
      "step": 31575
    },
    {
      "entropy": 6.02979154586792,
      "epoch": 3.1567451392012797,
      "grad_norm": 1.1875,
      "learning_rate": 0.00040217803468287034,
      "loss": 5.4635,
      "mean_token_accuracy": 0.19816322326660157,
      "num_tokens": 61784977.0,
      "step": 31580
    },
    {
      "entropy": 6.008619737625122,
      "epoch": 3.157244964262508,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040214859209223273,
      "loss": 5.419,
      "mean_token_accuracy": 0.1980102077126503,
      "num_tokens": 61794086.0,
      "step": 31585
    },
    {
      "entropy": 5.936828851699829,
      "epoch": 3.1577447893237367,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004021191463025693,
      "loss": 5.4485,
      "mean_token_accuracy": 0.18989246785640718,
      "num_tokens": 61803737.0,
      "step": 31590
    },
    {
      "entropy": 5.948536014556884,
      "epoch": 3.1582446143849654,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004020896973146208,
      "loss": 5.4572,
      "mean_token_accuracy": 0.19096762090921401,
      "num_tokens": 61814078.0,
      "step": 31595
    },
    {
      "entropy": 6.01352825164795,
      "epoch": 3.1587444394461937,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040206024512912824,
      "loss": 5.4349,
      "mean_token_accuracy": 0.19202901870012284,
      "num_tokens": 61823680.0,
      "step": 31600
    },
    {
      "entropy": 5.952416181564331,
      "epoch": 3.1592442645074224,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00040203078974683265,
      "loss": 5.5047,
      "mean_token_accuracy": 0.18932773321866989,
      "num_tokens": 61833704.0,
      "step": 31605
    },
    {
      "entropy": 6.00891523361206,
      "epoch": 3.159744089568651,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0004020013311684751,
      "loss": 5.4862,
      "mean_token_accuracy": 0.19134989231824875,
      "num_tokens": 61842402.0,
      "step": 31610
    },
    {
      "entropy": 6.02498574256897,
      "epoch": 3.1602439146298797,
      "grad_norm": 1.2109375,
      "learning_rate": 0.00040197186939479665,
      "loss": 5.4994,
      "mean_token_accuracy": 0.186500683426857,
      "num_tokens": 61852184.0,
      "step": 31615
    },
    {
      "entropy": 6.070594120025635,
      "epoch": 3.160743739691108,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040194240442653867,
      "loss": 5.5055,
      "mean_token_accuracy": 0.19075935035943986,
      "num_tokens": 61861260.0,
      "step": 31620
    },
    {
      "entropy": 5.96283974647522,
      "epoch": 3.1612435647523367,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004019129362644425,
      "loss": 5.4056,
      "mean_token_accuracy": 0.19637784063816072,
      "num_tokens": 61871131.0,
      "step": 31625
    },
    {
      "entropy": 5.980798768997192,
      "epoch": 3.1617433898135654,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0004018834649092494,
      "loss": 5.4497,
      "mean_token_accuracy": 0.19673945158720016,
      "num_tokens": 61881826.0,
      "step": 31630
    },
    {
      "entropy": 6.024117088317871,
      "epoch": 3.1622432148747937,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004018539903617011,
      "loss": 5.4289,
      "mean_token_accuracy": 0.19039698243141173,
      "num_tokens": 61891253.0,
      "step": 31635
    },
    {
      "entropy": 5.882132196426392,
      "epoch": 3.1627430399360223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040182451262253904,
      "loss": 5.4013,
      "mean_token_accuracy": 0.19969268441200255,
      "num_tokens": 61901054.0,
      "step": 31640
    },
    {
      "entropy": 5.959255313873291,
      "epoch": 3.163242864997251,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0004017950316925049,
      "loss": 5.3952,
      "mean_token_accuracy": 0.1971738949418068,
      "num_tokens": 61912298.0,
      "step": 31645
    },
    {
      "entropy": 6.020190858840943,
      "epoch": 3.1637426900584797,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040176554757234017,
      "loss": 5.4613,
      "mean_token_accuracy": 0.1914720579981804,
      "num_tokens": 61921449.0,
      "step": 31650
    },
    {
      "entropy": 5.925162220001221,
      "epoch": 3.164242515119708,
      "grad_norm": 1.109375,
      "learning_rate": 0.0004017360602627871,
      "loss": 5.3978,
      "mean_token_accuracy": 0.20227099508047103,
      "num_tokens": 61931220.0,
      "step": 31655
    },
    {
      "entropy": 6.047433567047119,
      "epoch": 3.1647423401809367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00040170656976458723,
      "loss": 5.6025,
      "mean_token_accuracy": 0.18652133345603944,
      "num_tokens": 61940981.0,
      "step": 31660
    },
    {
      "entropy": 6.074125289916992,
      "epoch": 3.1652421652421654,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040167707607848273,
      "loss": 5.5386,
      "mean_token_accuracy": 0.19303042888641359,
      "num_tokens": 61950632.0,
      "step": 31665
    },
    {
      "entropy": 5.941245365142822,
      "epoch": 3.1657419903033936,
      "grad_norm": 1.125,
      "learning_rate": 0.0004016475792052157,
      "loss": 5.4253,
      "mean_token_accuracy": 0.19680525213479996,
      "num_tokens": 61960652.0,
      "step": 31670
    },
    {
      "entropy": 6.021413278579712,
      "epoch": 3.1662418153646223,
      "grad_norm": 1.171875,
      "learning_rate": 0.000401618079145528,
      "loss": 5.5135,
      "mean_token_accuracy": 0.18435123860836028,
      "num_tokens": 61969480.0,
      "step": 31675
    },
    {
      "entropy": 5.929448890686035,
      "epoch": 3.166741640425851,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004015885759001621,
      "loss": 5.4817,
      "mean_token_accuracy": 0.19591234624385834,
      "num_tokens": 61979164.0,
      "step": 31680
    },
    {
      "entropy": 5.921716356277466,
      "epoch": 3.1672414654870797,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00040155906946986027,
      "loss": 5.4325,
      "mean_token_accuracy": 0.19898353517055511,
      "num_tokens": 61990513.0,
      "step": 31685
    },
    {
      "entropy": 6.049787282943726,
      "epoch": 3.167741290548308,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004015295598553647,
      "loss": 5.4784,
      "mean_token_accuracy": 0.18927515000104905,
      "num_tokens": 61999599.0,
      "step": 31690
    },
    {
      "entropy": 6.023807621002197,
      "epoch": 3.1682411156095367,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0004015000470574179,
      "loss": 5.5318,
      "mean_token_accuracy": 0.18936030715703964,
      "num_tokens": 62009548.0,
      "step": 31695
    },
    {
      "entropy": 5.99323296546936,
      "epoch": 3.1687409406707654,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040147053107676266,
      "loss": 5.4677,
      "mean_token_accuracy": 0.189926777780056,
      "num_tokens": 62018536.0,
      "step": 31700
    },
    {
      "entropy": 5.991889762878418,
      "epoch": 3.1692407657319936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00040144101191414127,
      "loss": 5.5594,
      "mean_token_accuracy": 0.18568063527345657,
      "num_tokens": 62027991.0,
      "step": 31705
    },
    {
      "entropy": 5.940165996551514,
      "epoch": 3.1697405907932223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004014114895702966,
      "loss": 5.4626,
      "mean_token_accuracy": 0.19803366661071778,
      "num_tokens": 62038554.0,
      "step": 31710
    },
    {
      "entropy": 6.0047767639160154,
      "epoch": 3.170240415854451,
      "grad_norm": 1.3203125,
      "learning_rate": 0.0004013819640459714,
      "loss": 5.4246,
      "mean_token_accuracy": 0.20164646804332734,
      "num_tokens": 62048060.0,
      "step": 31715
    },
    {
      "entropy": 5.996233987808227,
      "epoch": 3.1707402409156797,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040135243534190847,
      "loss": 5.4572,
      "mean_token_accuracy": 0.1935113176703453,
      "num_tokens": 62057905.0,
      "step": 31720
    },
    {
      "entropy": 5.98726978302002,
      "epoch": 3.171240065976908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040132290345885077,
      "loss": 5.47,
      "mean_token_accuracy": 0.19013614654541017,
      "num_tokens": 62067022.0,
      "step": 31725
    },
    {
      "entropy": 6.039809370040894,
      "epoch": 3.1717398910381367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0004012933683975413,
      "loss": 5.4849,
      "mean_token_accuracy": 0.19068132042884828,
      "num_tokens": 62075696.0,
      "step": 31730
    },
    {
      "entropy": 5.961600637435913,
      "epoch": 3.1722397160993654,
      "grad_norm": 1.078125,
      "learning_rate": 0.00040126383015872324,
      "loss": 5.4869,
      "mean_token_accuracy": 0.19163547456264496,
      "num_tokens": 62085620.0,
      "step": 31735
    },
    {
      "entropy": 6.012762498855591,
      "epoch": 3.1727395411605936,
      "grad_norm": 1.2578125,
      "learning_rate": 0.00040123428874313974,
      "loss": 5.4723,
      "mean_token_accuracy": 0.18888215869665145,
      "num_tokens": 62094531.0,
      "step": 31740
    },
    {
      "entropy": 6.084392166137695,
      "epoch": 3.1732393662218223,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00040120474415153396,
      "loss": 5.5896,
      "mean_token_accuracy": 0.18357840329408645,
      "num_tokens": 62104036.0,
      "step": 31745
    },
    {
      "entropy": 6.003712177276611,
      "epoch": 3.173739191283051,
      "grad_norm": 1.234375,
      "learning_rate": 0.00040117519638464935,
      "loss": 5.4406,
      "mean_token_accuracy": 0.1878862276673317,
      "num_tokens": 62113772.0,
      "step": 31750
    },
    {
      "entropy": 5.94217357635498,
      "epoch": 3.1742390163442793,
      "grad_norm": 1.3046875,
      "learning_rate": 0.00040114564544322913,
      "loss": 5.4145,
      "mean_token_accuracy": 0.19738237857818602,
      "num_tokens": 62123237.0,
      "step": 31755
    },
    {
      "entropy": 5.940281963348388,
      "epoch": 3.174738841405508,
      "grad_norm": 1.171875,
      "learning_rate": 0.000401116091328017,
      "loss": 5.4324,
      "mean_token_accuracy": 0.1949404150247574,
      "num_tokens": 62132391.0,
      "step": 31760
    },
    {
      "entropy": 5.8717821598052975,
      "epoch": 3.1752386664667367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040108653403975647,
      "loss": 5.338,
      "mean_token_accuracy": 0.21413731724023818,
      "num_tokens": 62141808.0,
      "step": 31765
    },
    {
      "entropy": 5.808552217483521,
      "epoch": 3.1757384915279654,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0004010569735791912,
      "loss": 5.1356,
      "mean_token_accuracy": 0.2220717191696167,
      "num_tokens": 62151065.0,
      "step": 31770
    },
    {
      "entropy": 5.965939378738403,
      "epoch": 3.1762383165891936,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004010274099470649,
      "loss": 5.4306,
      "mean_token_accuracy": 0.1946471944451332,
      "num_tokens": 62160885.0,
      "step": 31775
    },
    {
      "entropy": 5.826192808151245,
      "epoch": 3.1767381416504223,
      "grad_norm": 1.6875,
      "learning_rate": 0.0004009978431441214,
      "loss": 5.3344,
      "mean_token_accuracy": 0.20928571224212647,
      "num_tokens": 62169917.0,
      "step": 31780
    },
    {
      "entropy": 5.912977981567383,
      "epoch": 3.177237966711651,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040096827317110463,
      "loss": 5.4379,
      "mean_token_accuracy": 0.1948835015296936,
      "num_tokens": 62179896.0,
      "step": 31785
    },
    {
      "entropy": 5.9609979629516605,
      "epoch": 3.1777377917728793,
      "grad_norm": 1.1875,
      "learning_rate": 0.0004009387000287585,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19321976006031036,
      "num_tokens": 62189378.0,
      "step": 31790
    },
    {
      "entropy": 5.995979452133179,
      "epoch": 3.178237616834108,
      "grad_norm": 1.078125,
      "learning_rate": 0.000400909123717827,
      "loss": 5.4936,
      "mean_token_accuracy": 0.1972079575061798,
      "num_tokens": 62198602.0,
      "step": 31795
    },
    {
      "entropy": 6.004583692550659,
      "epoch": 3.1787374418953367,
      "grad_norm": 1.15625,
      "learning_rate": 0.0004008795442390544,
      "loss": 5.4522,
      "mean_token_accuracy": 0.18629064708948134,
      "num_tokens": 62206989.0,
      "step": 31800
    },
    {
      "entropy": 5.997479248046875,
      "epoch": 3.1792372669565654,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004008499615931849,
      "loss": 5.486,
      "mean_token_accuracy": 0.19609298855066298,
      "num_tokens": 62216922.0,
      "step": 31805
    },
    {
      "entropy": 6.089541101455689,
      "epoch": 3.1797370920177936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004008203757809627,
      "loss": 5.5033,
      "mean_token_accuracy": 0.1859486311674118,
      "num_tokens": 62225946.0,
      "step": 31810
    },
    {
      "entropy": 5.993651294708252,
      "epoch": 3.1802369170790223,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040079078680313224,
      "loss": 5.4777,
      "mean_token_accuracy": 0.19616598188877105,
      "num_tokens": 62236464.0,
      "step": 31815
    },
    {
      "entropy": 6.043734216690064,
      "epoch": 3.180736742140251,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000400761194660438,
      "loss": 5.4669,
      "mean_token_accuracy": 0.19294981956481932,
      "num_tokens": 62245789.0,
      "step": 31820
    },
    {
      "entropy": 5.907081842422485,
      "epoch": 3.1812365672014793,
      "grad_norm": 1.09375,
      "learning_rate": 0.0004007315993536244,
      "loss": 5.4132,
      "mean_token_accuracy": 0.1964319109916687,
      "num_tokens": 62254843.0,
      "step": 31825
    },
    {
      "entropy": 5.937596225738526,
      "epoch": 3.181736392262708,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040070200088343613,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19304445385932922,
      "num_tokens": 62265239.0,
      "step": 31830
    },
    {
      "entropy": 5.933077669143676,
      "epoch": 3.1822362173239367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00040067239925061784,
      "loss": 5.4122,
      "mean_token_accuracy": 0.2026483803987503,
      "num_tokens": 62275496.0,
      "step": 31835
    },
    {
      "entropy": 6.013749027252198,
      "epoch": 3.1827360423851654,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0004006427944559144,
      "loss": 5.4913,
      "mean_token_accuracy": 0.1972864344716072,
      "num_tokens": 62286071.0,
      "step": 31840
    },
    {
      "entropy": 5.945190477371216,
      "epoch": 3.1832358674463936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00040061318650007045,
      "loss": 5.4075,
      "mean_token_accuracy": 0.1980511277914047,
      "num_tokens": 62295242.0,
      "step": 31845
    },
    {
      "entropy": 5.912838554382324,
      "epoch": 3.1837356925076223,
      "grad_norm": 1.171875,
      "learning_rate": 0.00040058357538383117,
      "loss": 5.4283,
      "mean_token_accuracy": 0.2005800798535347,
      "num_tokens": 62305546.0,
      "step": 31850
    },
    {
      "entropy": 5.976898574829102,
      "epoch": 3.184235517568851,
      "grad_norm": 1.3515625,
      "learning_rate": 0.0004005539611079414,
      "loss": 5.4906,
      "mean_token_accuracy": 0.19358714669942856,
      "num_tokens": 62315181.0,
      "step": 31855
    },
    {
      "entropy": 6.007874059677124,
      "epoch": 3.1847353426300793,
      "grad_norm": 1.125,
      "learning_rate": 0.00040052434367314634,
      "loss": 5.4385,
      "mean_token_accuracy": 0.19311296790838242,
      "num_tokens": 62324741.0,
      "step": 31860
    },
    {
      "entropy": 5.960619401931763,
      "epoch": 3.185235167691308,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040049472308019104,
      "loss": 5.4474,
      "mean_token_accuracy": 0.20098691582679748,
      "num_tokens": 62334637.0,
      "step": 31865
    },
    {
      "entropy": 6.15636420249939,
      "epoch": 3.1857349927525367,
      "grad_norm": 1.109375,
      "learning_rate": 0.00040046509932982076,
      "loss": 5.6349,
      "mean_token_accuracy": 0.1836501330137253,
      "num_tokens": 62344637.0,
      "step": 31870
    },
    {
      "entropy": 5.982431411743164,
      "epoch": 3.1862348178137654,
      "grad_norm": 1.140625,
      "learning_rate": 0.0004004354724227809,
      "loss": 5.4281,
      "mean_token_accuracy": 0.19599066376686097,
      "num_tokens": 62353867.0,
      "step": 31875
    },
    {
      "entropy": 5.817639255523682,
      "epoch": 3.1867346428749936,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0004004058423598168,
      "loss": 5.3432,
      "mean_token_accuracy": 0.21051892787218093,
      "num_tokens": 62364729.0,
      "step": 31880
    },
    {
      "entropy": 5.990703964233399,
      "epoch": 3.1872344679362223,
      "grad_norm": 1.09375,
      "learning_rate": 0.00040037620914167406,
      "loss": 5.4808,
      "mean_token_accuracy": 0.18948947489261628,
      "num_tokens": 62374295.0,
      "step": 31885
    },
    {
      "entropy": 6.016822814941406,
      "epoch": 3.187734292997451,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0004003465727690981,
      "loss": 5.4988,
      "mean_token_accuracy": 0.19389074444770812,
      "num_tokens": 62383276.0,
      "step": 31890
    },
    {
      "entropy": 6.016742897033692,
      "epoch": 3.1882341180586793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00040031693324283464,
      "loss": 5.4989,
      "mean_token_accuracy": 0.19156708121299743,
      "num_tokens": 62392473.0,
      "step": 31895
    },
    {
      "entropy": 6.040921545028686,
      "epoch": 3.188733943119908,
      "grad_norm": 1.203125,
      "learning_rate": 0.0004002872905636294,
      "loss": 5.6335,
      "mean_token_accuracy": 0.18517186045646666,
      "num_tokens": 62402127.0,
      "step": 31900
    },
    {
      "entropy": 6.054186487197876,
      "epoch": 3.1892337681811367,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00040025764473222816,
      "loss": 5.5364,
      "mean_token_accuracy": 0.18769917339086534,
      "num_tokens": 62411757.0,
      "step": 31905
    },
    {
      "entropy": 5.8423333168029785,
      "epoch": 3.189733593242365,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00040022799574937674,
      "loss": 5.2635,
      "mean_token_accuracy": 0.21091450601816178,
      "num_tokens": 62422446.0,
      "step": 31910
    },
    {
      "entropy": 5.941195964813232,
      "epoch": 3.1902334183035936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040019834361582123,
      "loss": 5.3734,
      "mean_token_accuracy": 0.20959193706512452,
      "num_tokens": 62430970.0,
      "step": 31915
    },
    {
      "entropy": 5.9468447208404545,
      "epoch": 3.1907332433648223,
      "grad_norm": 1.125,
      "learning_rate": 0.00040016868833230766,
      "loss": 5.4216,
      "mean_token_accuracy": 0.1934804692864418,
      "num_tokens": 62440666.0,
      "step": 31920
    },
    {
      "entropy": 6.0380699634552,
      "epoch": 3.191233068426051,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0004001390298995821,
      "loss": 5.5224,
      "mean_token_accuracy": 0.1975189059972763,
      "num_tokens": 62450008.0,
      "step": 31925
    },
    {
      "entropy": 5.84752631187439,
      "epoch": 3.1917328934872793,
      "grad_norm": 1.25,
      "learning_rate": 0.00040010936831839065,
      "loss": 5.2884,
      "mean_token_accuracy": 0.2085451990365982,
      "num_tokens": 62459225.0,
      "step": 31930
    },
    {
      "entropy": 5.991556024551391,
      "epoch": 3.192232718548508,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00040007970358947977,
      "loss": 5.4587,
      "mean_token_accuracy": 0.18982192128896713,
      "num_tokens": 62469094.0,
      "step": 31935
    },
    {
      "entropy": 5.97147421836853,
      "epoch": 3.1927325436097367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00040005003571359565,
      "loss": 5.4256,
      "mean_token_accuracy": 0.19228993207216263,
      "num_tokens": 62479308.0,
      "step": 31940
    },
    {
      "entropy": 5.892456436157227,
      "epoch": 3.193232368670965,
      "grad_norm": 1.15625,
      "learning_rate": 0.00040002036469148486,
      "loss": 5.3469,
      "mean_token_accuracy": 0.20419482588768006,
      "num_tokens": 62489758.0,
      "step": 31945
    },
    {
      "entropy": 5.999871349334716,
      "epoch": 3.1937321937321936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039999069052389384,
      "loss": 5.5159,
      "mean_token_accuracy": 0.18863907158374787,
      "num_tokens": 62499084.0,
      "step": 31950
    },
    {
      "entropy": 5.971793937683105,
      "epoch": 3.1942320187934223,
      "grad_norm": 1.125,
      "learning_rate": 0.00039996101321156926,
      "loss": 5.4597,
      "mean_token_accuracy": 0.19206098467111588,
      "num_tokens": 62508054.0,
      "step": 31955
    },
    {
      "entropy": 6.0303062915802,
      "epoch": 3.194731843854651,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039993133275525777,
      "loss": 5.4662,
      "mean_token_accuracy": 0.1934739589691162,
      "num_tokens": 62517823.0,
      "step": 31960
    },
    {
      "entropy": 6.001662540435791,
      "epoch": 3.1952316689158793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000399901649155706,
      "loss": 5.5082,
      "mean_token_accuracy": 0.19245974719524384,
      "num_tokens": 62528579.0,
      "step": 31965
    },
    {
      "entropy": 6.027767610549927,
      "epoch": 3.195731493977108,
      "grad_norm": 1.125,
      "learning_rate": 0.00039987196241366086,
      "loss": 5.5371,
      "mean_token_accuracy": 0.18775540143251418,
      "num_tokens": 62539682.0,
      "step": 31970
    },
    {
      "entropy": 6.093766069412231,
      "epoch": 3.1962313190383367,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039984227252986936,
      "loss": 5.5972,
      "mean_token_accuracy": 0.18218024522066117,
      "num_tokens": 62549431.0,
      "step": 31975
    },
    {
      "entropy": 6.010634040832519,
      "epoch": 3.196731144099565,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039981257950507835,
      "loss": 5.4505,
      "mean_token_accuracy": 0.1962405264377594,
      "num_tokens": 62559372.0,
      "step": 31980
    },
    {
      "entropy": 5.918906307220459,
      "epoch": 3.1972309691607936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000399782883340035,
      "loss": 5.4111,
      "mean_token_accuracy": 0.19937752485275267,
      "num_tokens": 62568415.0,
      "step": 31985
    },
    {
      "entropy": 5.9657470703125,
      "epoch": 3.1977307942220223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003997531840354865,
      "loss": 5.5156,
      "mean_token_accuracy": 0.18717618733644487,
      "num_tokens": 62578419.0,
      "step": 31990
    },
    {
      "entropy": 6.038073444366455,
      "epoch": 3.198230619283251,
      "grad_norm": 1.25,
      "learning_rate": 0.00039972348159217987,
      "loss": 5.403,
      "mean_token_accuracy": 0.20354871302843094,
      "num_tokens": 62587408.0,
      "step": 31995
    },
    {
      "entropy": 5.942176294326782,
      "epoch": 3.1987304443444793,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039969377601086257,
      "loss": 5.4473,
      "mean_token_accuracy": 0.19438078999519348,
      "num_tokens": 62598753.0,
      "step": 32000
    },
    {
      "entropy": 5.953826236724853,
      "epoch": 3.199230269405708,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039966406729228187,
      "loss": 5.4093,
      "mean_token_accuracy": 0.20518541038036348,
      "num_tokens": 62608700.0,
      "step": 32005
    },
    {
      "entropy": 5.895655488967895,
      "epoch": 3.1997300944669367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039963435543718544,
      "loss": 5.4051,
      "mean_token_accuracy": 0.20115143805742264,
      "num_tokens": 62618870.0,
      "step": 32010
    },
    {
      "entropy": 5.9198321342468265,
      "epoch": 3.200229919528165,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039960464044632067,
      "loss": 5.4352,
      "mean_token_accuracy": 0.20230301916599275,
      "num_tokens": 62629681.0,
      "step": 32015
    },
    {
      "entropy": 6.028596544265747,
      "epoch": 3.2007297445893936,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003995749223204351,
      "loss": 5.5203,
      "mean_token_accuracy": 0.1888175204396248,
      "num_tokens": 62638792.0,
      "step": 32020
    },
    {
      "entropy": 5.937923526763916,
      "epoch": 3.2012295696506223,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003995452010602767,
      "loss": 5.4132,
      "mean_token_accuracy": 0.2068414121866226,
      "num_tokens": 62648353.0,
      "step": 32025
    },
    {
      "entropy": 5.945056915283203,
      "epoch": 3.201729394711851,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039951547666659305,
      "loss": 5.4439,
      "mean_token_accuracy": 0.19346626549959184,
      "num_tokens": 62657585.0,
      "step": 32030
    },
    {
      "entropy": 5.94196081161499,
      "epoch": 3.2022292197730793,
      "grad_norm": 1.0390625,
      "learning_rate": 0.000399485749140132,
      "loss": 5.3784,
      "mean_token_accuracy": 0.19720625132322311,
      "num_tokens": 62668055.0,
      "step": 32035
    },
    {
      "entropy": 5.988038396835327,
      "epoch": 3.202729044834308,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039945601848164153,
      "loss": 5.4882,
      "mean_token_accuracy": 0.19586727619171143,
      "num_tokens": 62678279.0,
      "step": 32040
    },
    {
      "entropy": 5.935755729675293,
      "epoch": 3.2032288698955367,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003994262846918697,
      "loss": 5.339,
      "mean_token_accuracy": 0.20476308912038804,
      "num_tokens": 62687875.0,
      "step": 32045
    },
    {
      "entropy": 6.00769739151001,
      "epoch": 3.203728694956765,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039939654777156454,
      "loss": 5.4755,
      "mean_token_accuracy": 0.20105334967374802,
      "num_tokens": 62698345.0,
      "step": 32050
    },
    {
      "entropy": 6.008812808990479,
      "epoch": 3.2042285200179936,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003993668077214742,
      "loss": 5.4561,
      "mean_token_accuracy": 0.18656180948019027,
      "num_tokens": 62708176.0,
      "step": 32055
    },
    {
      "entropy": 5.925177049636841,
      "epoch": 3.2047283450792223,
      "grad_norm": 1.1875,
      "learning_rate": 0.0003993370645423471,
      "loss": 5.4691,
      "mean_token_accuracy": 0.19545457810163497,
      "num_tokens": 62717194.0,
      "step": 32060
    },
    {
      "entropy": 5.913721418380737,
      "epoch": 3.205228170140451,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039930731823493136,
      "loss": 5.337,
      "mean_token_accuracy": 0.20477207750082016,
      "num_tokens": 62727095.0,
      "step": 32065
    },
    {
      "entropy": 5.888753032684326,
      "epoch": 3.2057279952016793,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039927756879997556,
      "loss": 5.3358,
      "mean_token_accuracy": 0.2086097851395607,
      "num_tokens": 62737452.0,
      "step": 32070
    },
    {
      "entropy": 5.967770767211914,
      "epoch": 3.206227820262908,
      "grad_norm": 1.125,
      "learning_rate": 0.00039924781623822803,
      "loss": 5.4306,
      "mean_token_accuracy": 0.20212257504463196,
      "num_tokens": 62747345.0,
      "step": 32075
    },
    {
      "entropy": 5.9492340087890625,
      "epoch": 3.2067276453241367,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003992180605504374,
      "loss": 5.5266,
      "mean_token_accuracy": 0.1939336657524109,
      "num_tokens": 62757222.0,
      "step": 32080
    },
    {
      "entropy": 5.9753522872924805,
      "epoch": 3.207227470385365,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003991883017373523,
      "loss": 5.486,
      "mean_token_accuracy": 0.18968759328126908,
      "num_tokens": 62767279.0,
      "step": 32085
    },
    {
      "entropy": 6.012639808654785,
      "epoch": 3.2077272954465936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003991585397997216,
      "loss": 5.4384,
      "mean_token_accuracy": 0.19791668504476548,
      "num_tokens": 62776547.0,
      "step": 32090
    },
    {
      "entropy": 5.990026760101318,
      "epoch": 3.2082271205078223,
      "grad_norm": 0.9921875,
      "learning_rate": 0.000399128774738294,
      "loss": 5.4343,
      "mean_token_accuracy": 0.19857974350452423,
      "num_tokens": 62786689.0,
      "step": 32095
    },
    {
      "entropy": 6.022890996932984,
      "epoch": 3.208726945569051,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003990990065538183,
      "loss": 5.5671,
      "mean_token_accuracy": 0.18533232063055038,
      "num_tokens": 62795862.0,
      "step": 32100
    },
    {
      "entropy": 5.995427465438842,
      "epoch": 3.2092267706302793,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003990692352470435,
      "loss": 5.4104,
      "mean_token_accuracy": 0.19688408076763153,
      "num_tokens": 62806065.0,
      "step": 32105
    },
    {
      "entropy": 5.979330730438233,
      "epoch": 3.209726595691508,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039903946081871865,
      "loss": 5.4616,
      "mean_token_accuracy": 0.1865861490368843,
      "num_tokens": 62815107.0,
      "step": 32110
    },
    {
      "entropy": 5.9161454200744625,
      "epoch": 3.2102264207527367,
      "grad_norm": 1.4375,
      "learning_rate": 0.00039900968326959295,
      "loss": 5.3222,
      "mean_token_accuracy": 0.20949522852897645,
      "num_tokens": 62823734.0,
      "step": 32115
    },
    {
      "entropy": 5.962410831451416,
      "epoch": 3.210726245813965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039897990260041557,
      "loss": 5.4801,
      "mean_token_accuracy": 0.1972297579050064,
      "num_tokens": 62833804.0,
      "step": 32120
    },
    {
      "entropy": 5.949703550338745,
      "epoch": 3.2112260708751936,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003989501188119357,
      "loss": 5.4493,
      "mean_token_accuracy": 0.19458282738924026,
      "num_tokens": 62844414.0,
      "step": 32125
    },
    {
      "entropy": 6.015921592712402,
      "epoch": 3.2117258959364223,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039892033190490274,
      "loss": 5.4274,
      "mean_token_accuracy": 0.20380644649267196,
      "num_tokens": 62855315.0,
      "step": 32130
    },
    {
      "entropy": 5.9333353519439695,
      "epoch": 3.212225720997651,
      "grad_norm": 1.125,
      "learning_rate": 0.00039889054188006615,
      "loss": 5.388,
      "mean_token_accuracy": 0.1952349603176117,
      "num_tokens": 62864526.0,
      "step": 32135
    },
    {
      "entropy": 5.900043678283692,
      "epoch": 3.2127255460588793,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003988607487381753,
      "loss": 5.4632,
      "mean_token_accuracy": 0.19127848595380784,
      "num_tokens": 62874599.0,
      "step": 32140
    },
    {
      "entropy": 6.010245370864868,
      "epoch": 3.213225371120108,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003988309524799799,
      "loss": 5.473,
      "mean_token_accuracy": 0.19200847446918487,
      "num_tokens": 62885956.0,
      "step": 32145
    },
    {
      "entropy": 5.7966968536376955,
      "epoch": 3.2137251961813367,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039880115310622974,
      "loss": 5.2634,
      "mean_token_accuracy": 0.2188959613442421,
      "num_tokens": 62896202.0,
      "step": 32150
    },
    {
      "entropy": 5.8727803230285645,
      "epoch": 3.214225021242565,
      "grad_norm": 1.125,
      "learning_rate": 0.00039877135061767424,
      "loss": 5.365,
      "mean_token_accuracy": 0.19803449064493178,
      "num_tokens": 62905650.0,
      "step": 32155
    },
    {
      "entropy": 5.985955047607422,
      "epoch": 3.2147248463037936,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003987415450150635,
      "loss": 5.4476,
      "mean_token_accuracy": 0.19568676352500916,
      "num_tokens": 62916036.0,
      "step": 32160
    },
    {
      "entropy": 6.031588172912597,
      "epoch": 3.2152246713650223,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039871173629914736,
      "loss": 5.515,
      "mean_token_accuracy": 0.1888739973306656,
      "num_tokens": 62925221.0,
      "step": 32165
    },
    {
      "entropy": 5.967233753204345,
      "epoch": 3.215724496426251,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039868192447067575,
      "loss": 5.4448,
      "mean_token_accuracy": 0.1909988969564438,
      "num_tokens": 62934917.0,
      "step": 32170
    },
    {
      "entropy": 5.889790821075439,
      "epoch": 3.2162243214874793,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039865210953039863,
      "loss": 5.4113,
      "mean_token_accuracy": 0.19866092950105668,
      "num_tokens": 62944357.0,
      "step": 32175
    },
    {
      "entropy": 5.971069622039795,
      "epoch": 3.216724146548708,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003986222914790664,
      "loss": 5.4631,
      "mean_token_accuracy": 0.20101162046194077,
      "num_tokens": 62954149.0,
      "step": 32180
    },
    {
      "entropy": 6.042191696166992,
      "epoch": 3.2172239716099367,
      "grad_norm": 1.0234375,
      "learning_rate": 0.00039859247031742907,
      "loss": 5.524,
      "mean_token_accuracy": 0.1833211213350296,
      "num_tokens": 62964721.0,
      "step": 32185
    },
    {
      "entropy": 5.966105556488037,
      "epoch": 3.217723796671165,
      "grad_norm": 1.21875,
      "learning_rate": 0.000398562646046237,
      "loss": 5.4063,
      "mean_token_accuracy": 0.20036827325820922,
      "num_tokens": 62974738.0,
      "step": 32190
    },
    {
      "entropy": 5.923502635955811,
      "epoch": 3.2182236217323936,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039853281866624054,
      "loss": 5.4438,
      "mean_token_accuracy": 0.20329310297966002,
      "num_tokens": 62984109.0,
      "step": 32195
    },
    {
      "entropy": 6.038552474975586,
      "epoch": 3.2187234467936223,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039850298817819017,
      "loss": 5.5478,
      "mean_token_accuracy": 0.18799534142017366,
      "num_tokens": 62994443.0,
      "step": 32200
    },
    {
      "entropy": 6.05847282409668,
      "epoch": 3.219223271854851,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003984731545828364,
      "loss": 5.5978,
      "mean_token_accuracy": 0.18122070282697678,
      "num_tokens": 63003897.0,
      "step": 32205
    },
    {
      "entropy": 6.00833740234375,
      "epoch": 3.2197230969160793,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039844331788092986,
      "loss": 5.5004,
      "mean_token_accuracy": 0.1926969528198242,
      "num_tokens": 63013012.0,
      "step": 32210
    },
    {
      "entropy": 5.9991683006286625,
      "epoch": 3.220222921977308,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003984134780732211,
      "loss": 5.5047,
      "mean_token_accuracy": 0.19382333606481553,
      "num_tokens": 63022756.0,
      "step": 32215
    },
    {
      "entropy": 5.93957929611206,
      "epoch": 3.2207227470385367,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003983836351604611,
      "loss": 5.4182,
      "mean_token_accuracy": 0.1937337651848793,
      "num_tokens": 63033109.0,
      "step": 32220
    },
    {
      "entropy": 5.970108127593994,
      "epoch": 3.221222572099765,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003983537891434005,
      "loss": 5.4564,
      "mean_token_accuracy": 0.18687559068202972,
      "num_tokens": 63042733.0,
      "step": 32225
    },
    {
      "entropy": 5.890345478057862,
      "epoch": 3.2217223971609936,
      "grad_norm": 1.234375,
      "learning_rate": 0.0003983239400227903,
      "loss": 5.3266,
      "mean_token_accuracy": 0.208632230758667,
      "num_tokens": 63053654.0,
      "step": 32230
    },
    {
      "entropy": 5.942334842681885,
      "epoch": 3.2222222222222223,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003982940877993816,
      "loss": 5.4213,
      "mean_token_accuracy": 0.19690677970647813,
      "num_tokens": 63063285.0,
      "step": 32235
    },
    {
      "entropy": 6.125774383544922,
      "epoch": 3.222722047283451,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00039826423247392537,
      "loss": 5.6187,
      "mean_token_accuracy": 0.17516909390687943,
      "num_tokens": 63072730.0,
      "step": 32240
    },
    {
      "entropy": 5.94420485496521,
      "epoch": 3.2232218723446793,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003982343740471727,
      "loss": 5.4137,
      "mean_token_accuracy": 0.19510273933410643,
      "num_tokens": 63081703.0,
      "step": 32245
    },
    {
      "entropy": 5.964404964447022,
      "epoch": 3.223721697405908,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039820451251987494,
      "loss": 5.4284,
      "mean_token_accuracy": 0.19674137532711028,
      "num_tokens": 63091382.0,
      "step": 32250
    },
    {
      "entropy": 5.969366216659546,
      "epoch": 3.2242215224671367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039817464789278326,
      "loss": 5.3677,
      "mean_token_accuracy": 0.20497010946273803,
      "num_tokens": 63101256.0,
      "step": 32255
    },
    {
      "entropy": 5.953371047973633,
      "epoch": 3.224721347528365,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039814478016664923,
      "loss": 5.3799,
      "mean_token_accuracy": 0.197530098259449,
      "num_tokens": 63111001.0,
      "step": 32260
    },
    {
      "entropy": 5.909182834625244,
      "epoch": 3.2252211725895936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039811490934222416,
      "loss": 5.4681,
      "mean_token_accuracy": 0.19409203976392747,
      "num_tokens": 63120524.0,
      "step": 32265
    },
    {
      "entropy": 5.985256195068359,
      "epoch": 3.2257209976508223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003980850354202598,
      "loss": 5.4831,
      "mean_token_accuracy": 0.19086410403251647,
      "num_tokens": 63130673.0,
      "step": 32270
    },
    {
      "entropy": 6.063601970672607,
      "epoch": 3.226220822712051,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039805515840150743,
      "loss": 5.4844,
      "mean_token_accuracy": 0.19006870985031127,
      "num_tokens": 63140949.0,
      "step": 32275
    },
    {
      "entropy": 6.062148904800415,
      "epoch": 3.2267206477732793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039802527828671896,
      "loss": 5.4955,
      "mean_token_accuracy": 0.190481998026371,
      "num_tokens": 63151088.0,
      "step": 32280
    },
    {
      "entropy": 5.923337030410766,
      "epoch": 3.227220472834508,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003979953950766462,
      "loss": 5.3588,
      "mean_token_accuracy": 0.20028552114963533,
      "num_tokens": 63160739.0,
      "step": 32285
    },
    {
      "entropy": 5.87428092956543,
      "epoch": 3.2277202978957367,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039796550877204095,
      "loss": 5.4457,
      "mean_token_accuracy": 0.19607153683900833,
      "num_tokens": 63170657.0,
      "step": 32290
    },
    {
      "entropy": 5.978527021408081,
      "epoch": 3.228220122956965,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039793561937365515,
      "loss": 5.4454,
      "mean_token_accuracy": 0.19925541430711746,
      "num_tokens": 63180603.0,
      "step": 32295
    },
    {
      "entropy": 5.9439527034759525,
      "epoch": 3.2287199480181936,
      "grad_norm": 1.25,
      "learning_rate": 0.00039790572688224075,
      "loss": 5.4946,
      "mean_token_accuracy": 0.19529586732387544,
      "num_tokens": 63190589.0,
      "step": 32300
    },
    {
      "entropy": 6.03615779876709,
      "epoch": 3.2292197730794223,
      "grad_norm": 1.3515625,
      "learning_rate": 0.00039787583129854997,
      "loss": 5.4881,
      "mean_token_accuracy": 0.19561329931020738,
      "num_tokens": 63200227.0,
      "step": 32305
    },
    {
      "entropy": 5.932640123367309,
      "epoch": 3.2297195981406506,
      "grad_norm": 1.1640625,
      "learning_rate": 0.0003978459326233349,
      "loss": 5.4127,
      "mean_token_accuracy": 0.20077780932188033,
      "num_tokens": 63209273.0,
      "step": 32310
    },
    {
      "entropy": 5.853814792633057,
      "epoch": 3.2302194232018793,
      "grad_norm": 1.53125,
      "learning_rate": 0.0003978160308573477,
      "loss": 5.3646,
      "mean_token_accuracy": 0.2054761603474617,
      "num_tokens": 63218711.0,
      "step": 32315
    },
    {
      "entropy": 5.858217000961304,
      "epoch": 3.230719248263108,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0003977861260013407,
      "loss": 5.3045,
      "mean_token_accuracy": 0.21386423707008362,
      "num_tokens": 63229982.0,
      "step": 32320
    },
    {
      "entropy": 6.131993341445923,
      "epoch": 3.2312190733243367,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003977562180560665,
      "loss": 5.6062,
      "mean_token_accuracy": 0.18484318852424622,
      "num_tokens": 63239659.0,
      "step": 32325
    },
    {
      "entropy": 6.0088193893432615,
      "epoch": 3.231718898385565,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003977263070222774,
      "loss": 5.4503,
      "mean_token_accuracy": 0.19535393714904786,
      "num_tokens": 63249379.0,
      "step": 32330
    },
    {
      "entropy": 5.925339269638061,
      "epoch": 3.2322187234467936,
      "grad_norm": 1.2421875,
      "learning_rate": 0.000397696392900726,
      "loss": 5.3527,
      "mean_token_accuracy": 0.20175118297338485,
      "num_tokens": 63259857.0,
      "step": 32335
    },
    {
      "entropy": 5.985756874084473,
      "epoch": 3.2327185485080223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003976664756921649,
      "loss": 5.4338,
      "mean_token_accuracy": 0.18929346650838852,
      "num_tokens": 63268209.0,
      "step": 32340
    },
    {
      "entropy": 5.986113405227661,
      "epoch": 3.2332183735692506,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003976365553973469,
      "loss": 5.4357,
      "mean_token_accuracy": 0.19302802085876464,
      "num_tokens": 63278649.0,
      "step": 32345
    },
    {
      "entropy": 5.9589011669158936,
      "epoch": 3.2337181986304793,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0003976066320170247,
      "loss": 5.4447,
      "mean_token_accuracy": 0.19558094888925553,
      "num_tokens": 63288090.0,
      "step": 32350
    },
    {
      "entropy": 6.0012282371521,
      "epoch": 3.234218023691708,
      "grad_norm": 1.15625,
      "learning_rate": 0.00039757670555195114,
      "loss": 5.4822,
      "mean_token_accuracy": 0.19337985217571257,
      "num_tokens": 63297443.0,
      "step": 32355
    },
    {
      "entropy": 5.979399061203003,
      "epoch": 3.2347178487529367,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039754677600287923,
      "loss": 5.4565,
      "mean_token_accuracy": 0.18965043723583222,
      "num_tokens": 63306485.0,
      "step": 32360
    },
    {
      "entropy": 5.96643328666687,
      "epoch": 3.235217673814165,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039751684337056197,
      "loss": 5.4033,
      "mean_token_accuracy": 0.20225052535533905,
      "num_tokens": 63315872.0,
      "step": 32365
    },
    {
      "entropy": 5.938296175003051,
      "epoch": 3.2357174988753936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.0003974869076557525,
      "loss": 5.373,
      "mean_token_accuracy": 0.1901915431022644,
      "num_tokens": 63325778.0,
      "step": 32370
    },
    {
      "entropy": 5.9178002834320065,
      "epoch": 3.2362173239366223,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003974569688592038,
      "loss": 5.372,
      "mean_token_accuracy": 0.20267438739538193,
      "num_tokens": 63335312.0,
      "step": 32375
    },
    {
      "entropy": 5.945513439178467,
      "epoch": 3.2367171489978506,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003974270269816695,
      "loss": 5.4569,
      "mean_token_accuracy": 0.19776187241077423,
      "num_tokens": 63344338.0,
      "step": 32380
    },
    {
      "entropy": 6.000157833099365,
      "epoch": 3.2372169740590793,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039739708202390255,
      "loss": 5.4808,
      "mean_token_accuracy": 0.19030797183513642,
      "num_tokens": 63353775.0,
      "step": 32385
    },
    {
      "entropy": 6.018087673187256,
      "epoch": 3.237716799120308,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003973671339866566,
      "loss": 5.5197,
      "mean_token_accuracy": 0.18502146303653716,
      "num_tokens": 63363474.0,
      "step": 32390
    },
    {
      "entropy": 5.971002244949341,
      "epoch": 3.2382166241815367,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039733718287068495,
      "loss": 5.4924,
      "mean_token_accuracy": 0.19443051964044572,
      "num_tokens": 63374120.0,
      "step": 32395
    },
    {
      "entropy": 6.00261664390564,
      "epoch": 3.238716449242765,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003973072286767413,
      "loss": 5.4742,
      "mean_token_accuracy": 0.19164223372936248,
      "num_tokens": 63384206.0,
      "step": 32400
    },
    {
      "entropy": 5.798675632476806,
      "epoch": 3.2392162743039936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.0003972772714055793,
      "loss": 5.2737,
      "mean_token_accuracy": 0.20901548266410827,
      "num_tokens": 63394354.0,
      "step": 32405
    },
    {
      "entropy": 5.994867944717408,
      "epoch": 3.2397160993652223,
      "grad_norm": 1.0703125,
      "learning_rate": 0.00039724731105795255,
      "loss": 5.466,
      "mean_token_accuracy": 0.1963673412799835,
      "num_tokens": 63403758.0,
      "step": 32410
    },
    {
      "entropy": 5.9399913311004635,
      "epoch": 3.2402159244264506,
      "grad_norm": 1.1875,
      "learning_rate": 0.000397217347634615,
      "loss": 5.446,
      "mean_token_accuracy": 0.19341851621866227,
      "num_tokens": 63413539.0,
      "step": 32415
    },
    {
      "entropy": 5.9856616973876955,
      "epoch": 3.2407157494876793,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003971873811363203,
      "loss": 5.5151,
      "mean_token_accuracy": 0.18972061276435853,
      "num_tokens": 63423577.0,
      "step": 32420
    },
    {
      "entropy": 5.9915224552154545,
      "epoch": 3.241215574548908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039715741156382255,
      "loss": 5.4865,
      "mean_token_accuracy": 0.19524515867233277,
      "num_tokens": 63433483.0,
      "step": 32425
    },
    {
      "entropy": 6.015208911895752,
      "epoch": 3.241715399610136,
      "grad_norm": 1.125,
      "learning_rate": 0.0003971274389178759,
      "loss": 5.4733,
      "mean_token_accuracy": 0.19445499926805496,
      "num_tokens": 63443341.0,
      "step": 32430
    },
    {
      "entropy": 5.899695825576782,
      "epoch": 3.242215224671365,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003970974631992342,
      "loss": 5.3698,
      "mean_token_accuracy": 0.20037556439638138,
      "num_tokens": 63452783.0,
      "step": 32435
    },
    {
      "entropy": 6.034071731567383,
      "epoch": 3.2427150497325936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003970674844086517,
      "loss": 5.5289,
      "mean_token_accuracy": 0.1883870467543602,
      "num_tokens": 63463420.0,
      "step": 32440
    },
    {
      "entropy": 5.969696140289306,
      "epoch": 3.2432148747938223,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003970375025468829,
      "loss": 5.3979,
      "mean_token_accuracy": 0.19731472730636596,
      "num_tokens": 63472405.0,
      "step": 32445
    },
    {
      "entropy": 5.964757347106934,
      "epoch": 3.2437146998550506,
      "grad_norm": 1.1953125,
      "learning_rate": 0.00039700751761468175,
      "loss": 5.487,
      "mean_token_accuracy": 0.19593070447444916,
      "num_tokens": 63481242.0,
      "step": 32450
    },
    {
      "entropy": 5.9313397884368895,
      "epoch": 3.2442145249162793,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003969775296128029,
      "loss": 5.4303,
      "mean_token_accuracy": 0.2022685945034027,
      "num_tokens": 63492244.0,
      "step": 32455
    },
    {
      "entropy": 5.995953464508057,
      "epoch": 3.244714349977508,
      "grad_norm": 1.2109375,
      "learning_rate": 0.0003969475385420008,
      "loss": 5.4458,
      "mean_token_accuracy": 0.19367871135473252,
      "num_tokens": 63501117.0,
      "step": 32460
    },
    {
      "entropy": 6.076545476913452,
      "epoch": 3.245214175038736,
      "grad_norm": 1.0859375,
      "learning_rate": 0.00039691754440303003,
      "loss": 5.549,
      "mean_token_accuracy": 0.18655347526073457,
      "num_tokens": 63509603.0,
      "step": 32465
    },
    {
      "entropy": 5.99781436920166,
      "epoch": 3.245714000099965,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039688754719664516,
      "loss": 5.5175,
      "mean_token_accuracy": 0.1906546637415886,
      "num_tokens": 63519600.0,
      "step": 32470
    },
    {
      "entropy": 5.965010023117065,
      "epoch": 3.2462138251611936,
      "grad_norm": 1.1328125,
      "learning_rate": 0.000396857546923601,
      "loss": 5.4798,
      "mean_token_accuracy": 0.19504607617855071,
      "num_tokens": 63529712.0,
      "step": 32475
    },
    {
      "entropy": 5.917618179321289,
      "epoch": 3.2467136502224223,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0003968275435846523,
      "loss": 5.4277,
      "mean_token_accuracy": 0.19459076821804047,
      "num_tokens": 63540327.0,
      "step": 32480
    },
    {
      "entropy": 5.912710189819336,
      "epoch": 3.2472134752836506,
      "grad_norm": 1.0859375,
      "learning_rate": 0.000396797537180554,
      "loss": 5.3928,
      "mean_token_accuracy": 0.20077248960733413,
      "num_tokens": 63549898.0,
      "step": 32485
    },
    {
      "entropy": 5.913759088516235,
      "epoch": 3.2477133003448793,
      "grad_norm": 1.0546875,
      "learning_rate": 0.000396767527712061,
      "loss": 5.3789,
      "mean_token_accuracy": 0.20283818542957305,
      "num_tokens": 63559296.0,
      "step": 32490
    },
    {
      "entropy": 5.963737869262696,
      "epoch": 3.248213125406108,
      "grad_norm": 1.25,
      "learning_rate": 0.00039673751517992827,
      "loss": 5.5018,
      "mean_token_accuracy": 0.193226720392704,
      "num_tokens": 63568464.0,
      "step": 32495
    },
    {
      "entropy": 5.971447896957398,
      "epoch": 3.248712950467336,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039670749958491103,
      "loss": 5.5035,
      "mean_token_accuracy": 0.18985623121261597,
      "num_tokens": 63577623.0,
      "step": 32500
    },
    {
      "entropy": 5.969267463684082,
      "epoch": 3.249212775528565,
      "grad_norm": 1.125,
      "learning_rate": 0.00039667748092776453,
      "loss": 5.3488,
      "mean_token_accuracy": 0.19808870404958726,
      "num_tokens": 63586809.0,
      "step": 32505
    },
    {
      "entropy": 5.947438430786133,
      "epoch": 3.2497126005897936,
      "grad_norm": 1.609375,
      "learning_rate": 0.00039664745920924374,
      "loss": 5.4723,
      "mean_token_accuracy": 0.19350059032440187,
      "num_tokens": 63597262.0,
      "step": 32510
    },
    {
      "entropy": 5.935785961151123,
      "epoch": 3.2502124256510223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003966174344301043,
      "loss": 5.4494,
      "mean_token_accuracy": 0.1976221263408661,
      "num_tokens": 63608688.0,
      "step": 32515
    },
    {
      "entropy": 5.911927366256714,
      "epoch": 3.2507122507122506,
      "grad_norm": 1.71875,
      "learning_rate": 0.0003965874065911014,
      "loss": 5.3404,
      "mean_token_accuracy": 0.2038295492529869,
      "num_tokens": 63619894.0,
      "step": 32520
    },
    {
      "entropy": 5.93956880569458,
      "epoch": 3.2512120757734793,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039655737569299073,
      "loss": 5.3595,
      "mean_token_accuracy": 0.19913107752799988,
      "num_tokens": 63628607.0,
      "step": 32525
    },
    {
      "entropy": 5.991753578186035,
      "epoch": 3.251711900834708,
      "grad_norm": 1.1171875,
      "learning_rate": 0.0003965273417365277,
      "loss": 5.4751,
      "mean_token_accuracy": 0.19137845039367676,
      "num_tokens": 63639058.0,
      "step": 32530
    },
    {
      "entropy": 5.958595943450928,
      "epoch": 3.252211725895936,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039649730472246796,
      "loss": 5.3913,
      "mean_token_accuracy": 0.2016705483198166,
      "num_tokens": 63647338.0,
      "step": 32535
    },
    {
      "entropy": 5.908549451828003,
      "epoch": 3.252711550957165,
      "grad_norm": 1.09375,
      "learning_rate": 0.0003964672646515674,
      "loss": 5.3733,
      "mean_token_accuracy": 0.1995784729719162,
      "num_tokens": 63657990.0,
      "step": 32540
    },
    {
      "entropy": 6.009171199798584,
      "epoch": 3.2532113760183936,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003964372215245817,
      "loss": 5.5321,
      "mean_token_accuracy": 0.18942881524562835,
      "num_tokens": 63667310.0,
      "step": 32545
    },
    {
      "entropy": 5.949365186691284,
      "epoch": 3.2537112010796223,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039640717534226684,
      "loss": 5.395,
      "mean_token_accuracy": 0.20098046213388443,
      "num_tokens": 63677254.0,
      "step": 32550
    },
    {
      "entropy": 6.0275390625,
      "epoch": 3.2542110261408506,
      "grad_norm": 1.5546875,
      "learning_rate": 0.0003963771261053786,
      "loss": 5.4972,
      "mean_token_accuracy": 0.18634617775678636,
      "num_tokens": 63686441.0,
      "step": 32555
    },
    {
      "entropy": 5.8771219730377195,
      "epoch": 3.2547108512020793,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039634707381467303,
      "loss": 5.3824,
      "mean_token_accuracy": 0.21153170764446258,
      "num_tokens": 63696445.0,
      "step": 32560
    },
    {
      "entropy": 6.0427404880523685,
      "epoch": 3.255210676263308,
      "grad_norm": 1.0625,
      "learning_rate": 0.0003963170184709064,
      "loss": 5.5254,
      "mean_token_accuracy": 0.19048393219709397,
      "num_tokens": 63706536.0,
      "step": 32565
    },
    {
      "entropy": 5.991190433502197,
      "epoch": 3.255710501324536,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039628696007483477,
      "loss": 5.4749,
      "mean_token_accuracy": 0.1909590020775795,
      "num_tokens": 63716752.0,
      "step": 32570
    },
    {
      "entropy": 5.989365339279175,
      "epoch": 3.256210326385765,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039625689862721447,
      "loss": 5.4653,
      "mean_token_accuracy": 0.19581492096185685,
      "num_tokens": 63726995.0,
      "step": 32575
    },
    {
      "entropy": 5.998375129699707,
      "epoch": 3.2567101514469936,
      "grad_norm": 1.2265625,
      "learning_rate": 0.00039622683412880174,
      "loss": 5.3851,
      "mean_token_accuracy": 0.20347429811954498,
      "num_tokens": 63735261.0,
      "step": 32580
    },
    {
      "entropy": 6.012545013427735,
      "epoch": 3.2572099765082223,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003961967665803532,
      "loss": 5.4822,
      "mean_token_accuracy": 0.200266532599926,
      "num_tokens": 63746307.0,
      "step": 32585
    },
    {
      "entropy": 5.985945415496826,
      "epoch": 3.2577098015694506,
      "grad_norm": 1.28125,
      "learning_rate": 0.00039616669598262504,
      "loss": 5.5148,
      "mean_token_accuracy": 0.18986794352531433,
      "num_tokens": 63755802.0,
      "step": 32590
    },
    {
      "entropy": 5.893173789978027,
      "epoch": 3.2582096266306793,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039613662233637406,
      "loss": 5.4075,
      "mean_token_accuracy": 0.19969454407691956,
      "num_tokens": 63765759.0,
      "step": 32595
    },
    {
      "entropy": 6.049416780471802,
      "epoch": 3.258709451691908,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039610654564235684,
      "loss": 5.5199,
      "mean_token_accuracy": 0.19321747571229936,
      "num_tokens": 63775528.0,
      "step": 32600
    },
    {
      "entropy": 5.955934286117554,
      "epoch": 3.259209276753136,
      "grad_norm": 1.078125,
      "learning_rate": 0.0003960764659013301,
      "loss": 5.3845,
      "mean_token_accuracy": 0.20289321839809418,
      "num_tokens": 63785480.0,
      "step": 32605
    },
    {
      "entropy": 5.908633756637573,
      "epoch": 3.259709101814365,
      "grad_norm": 1.2421875,
      "learning_rate": 0.0003960463831140505,
      "loss": 5.3006,
      "mean_token_accuracy": 0.2090962275862694,
      "num_tokens": 63793710.0,
      "step": 32610
    },
    {
      "entropy": 5.986551904678345,
      "epoch": 3.2602089268755936,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00039601629728127524,
      "loss": 5.5382,
      "mean_token_accuracy": 0.1934936299920082,
      "num_tokens": 63802664.0,
      "step": 32615
    },
    {
      "entropy": 5.827068614959717,
      "epoch": 3.2607087519368223,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039598620840376096,
      "loss": 5.1774,
      "mean_token_accuracy": 0.22475902885198593,
      "num_tokens": 63813146.0,
      "step": 32620
    },
    {
      "entropy": 5.989646625518799,
      "epoch": 3.2612085769980506,
      "grad_norm": 1.09375,
      "learning_rate": 0.00039595611648226485,
      "loss": 5.5118,
      "mean_token_accuracy": 0.19027557075023652,
      "num_tokens": 63823182.0,
      "step": 32625
    },
    {
      "entropy": 5.912678194046021,
      "epoch": 3.2617084020592793,
      "grad_norm": 1.171875,
      "learning_rate": 0.000395926021517544,
      "loss": 5.4353,
      "mean_token_accuracy": 0.19567957073450087,
      "num_tokens": 63832370.0,
      "step": 32630
    },
    {
      "entropy": 5.898246765136719,
      "epoch": 3.262208227120508,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039589592351035554,
      "loss": 5.4377,
      "mean_token_accuracy": 0.19987795650959014,
      "num_tokens": 63843156.0,
      "step": 32635
    },
    {
      "entropy": 6.04361047744751,
      "epoch": 3.262708052181736,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003958658224614567,
      "loss": 5.5326,
      "mean_token_accuracy": 0.19348996132612228,
      "num_tokens": 63852744.0,
      "step": 32640
    },
    {
      "entropy": 5.980628252029419,
      "epoch": 3.263207877242965,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039583571837160497,
      "loss": 5.3801,
      "mean_token_accuracy": 0.20639743357896806,
      "num_tokens": 63862694.0,
      "step": 32645
    },
    {
      "entropy": 5.884800148010254,
      "epoch": 3.2637077023041936,
      "grad_norm": 1.109375,
      "learning_rate": 0.0003958056112415576,
      "loss": 5.423,
      "mean_token_accuracy": 0.19404355734586715,
      "num_tokens": 63871943.0,
      "step": 32650
    },
    {
      "entropy": 5.951906681060791,
      "epoch": 3.2642075273654223,
      "grad_norm": 1.203125,
      "learning_rate": 0.0003957755010720721,
      "loss": 5.4401,
      "mean_token_accuracy": 0.20310761034488678,
      "num_tokens": 63881354.0,
      "step": 32655
    },
    {
      "entropy": 5.9902876853942875,
      "epoch": 3.2647073524266506,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039574538786390617,
      "loss": 5.4728,
      "mean_token_accuracy": 0.19310243427753448,
      "num_tokens": 63890750.0,
      "step": 32660
    },
    {
      "entropy": 6.034393358230591,
      "epoch": 3.2652071774878793,
      "grad_norm": 1.3046875,
      "learning_rate": 0.0003957152716178173,
      "loss": 5.4966,
      "mean_token_accuracy": 0.18786927610635756,
      "num_tokens": 63900730.0,
      "step": 32665
    },
    {
      "entropy": 5.8904547691345215,
      "epoch": 3.265707002549108,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039568515233456326,
      "loss": 5.5274,
      "mean_token_accuracy": 0.18830336332321168,
      "num_tokens": 63911812.0,
      "step": 32670
    },
    {
      "entropy": 5.946032333374023,
      "epoch": 3.266206827610336,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003956550300149018,
      "loss": 5.4465,
      "mean_token_accuracy": 0.19210401624441148,
      "num_tokens": 63920785.0,
      "step": 32675
    },
    {
      "entropy": 5.99282054901123,
      "epoch": 3.266706652671565,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003956249046595909,
      "loss": 5.4377,
      "mean_token_accuracy": 0.19841964095830916,
      "num_tokens": 63931167.0,
      "step": 32680
    },
    {
      "entropy": 6.043642330169678,
      "epoch": 3.2672064777327936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.0003955947762693884,
      "loss": 5.5235,
      "mean_token_accuracy": 0.18476913422346114,
      "num_tokens": 63941448.0,
      "step": 32685
    },
    {
      "entropy": 6.091418218612671,
      "epoch": 3.2677063027940223,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039556464484505235,
      "loss": 5.513,
      "mean_token_accuracy": 0.18360398411750795,
      "num_tokens": 63951412.0,
      "step": 32690
    },
    {
      "entropy": 6.023193454742431,
      "epoch": 3.2682061278552506,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003955345103873409,
      "loss": 5.5602,
      "mean_token_accuracy": 0.18616772443056107,
      "num_tokens": 63960957.0,
      "step": 32695
    },
    {
      "entropy": 5.916533184051514,
      "epoch": 3.2687059529164793,
      "grad_norm": 1.171875,
      "learning_rate": 0.0003955043728970121,
      "loss": 5.497,
      "mean_token_accuracy": 0.1933011904358864,
      "num_tokens": 63970229.0,
      "step": 32700
    },
    {
      "entropy": 5.9458503246307375,
      "epoch": 3.269205777977708,
      "grad_norm": 1.2734375,
      "learning_rate": 0.0003954742323748244,
      "loss": 5.4396,
      "mean_token_accuracy": 0.1968454658985138,
      "num_tokens": 63979816.0,
      "step": 32705
    },
    {
      "entropy": 5.953328657150268,
      "epoch": 3.269705603038936,
      "grad_norm": 1.125,
      "learning_rate": 0.00039544408882153596,
      "loss": 5.2956,
      "mean_token_accuracy": 0.21041571497917175,
      "num_tokens": 63989612.0,
      "step": 32710
    },
    {
      "entropy": 6.001097822189331,
      "epoch": 3.270205428100165,
      "grad_norm": 0.96875,
      "learning_rate": 0.0003954139422379052,
      "loss": 5.4835,
      "mean_token_accuracy": 0.18749974370002748,
      "num_tokens": 64001254.0,
      "step": 32715
    },
    {
      "entropy": 5.923443078994751,
      "epoch": 3.2707052531613936,
      "grad_norm": 1.1015625,
      "learning_rate": 0.0003953837926246907,
      "loss": 5.4549,
      "mean_token_accuracy": 0.19719965010881424,
      "num_tokens": 64011235.0,
      "step": 32720
    },
    {
      "entropy": 5.920061826705933,
      "epoch": 3.2712050782226223,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00039535363998265095,
      "loss": 5.4365,
      "mean_token_accuracy": 0.1908642277121544,
      "num_tokens": 64020437.0,
      "step": 32725
    },
    {
      "entropy": 6.063912677764892,
      "epoch": 3.2717049032838506,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039532348431254455,
      "loss": 5.5914,
      "mean_token_accuracy": 0.18496422320604325,
      "num_tokens": 64030930.0,
      "step": 32730
    },
    {
      "entropy": 5.970167922973633,
      "epoch": 3.2722047283450792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003952933256151303,
      "loss": 5.4275,
      "mean_token_accuracy": 0.20161616057157516,
      "num_tokens": 64041175.0,
      "step": 32735
    },
    {
      "entropy": 5.978709745407104,
      "epoch": 3.272704553406308,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003952631638911669,
      "loss": 5.3933,
      "mean_token_accuracy": 0.19550527930259703,
      "num_tokens": 64050208.0,
      "step": 32740
    },
    {
      "entropy": 5.997635459899902,
      "epoch": 3.273204378467536,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003952329991414132,
      "loss": 5.4791,
      "mean_token_accuracy": 0.19290563762187957,
      "num_tokens": 64060303.0,
      "step": 32745
    },
    {
      "entropy": 5.972135210037232,
      "epoch": 3.273704203528765,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039520283136662825,
      "loss": 5.4367,
      "mean_token_accuracy": 0.19364765137434006,
      "num_tokens": 64070240.0,
      "step": 32750
    },
    {
      "entropy": 6.089263486862182,
      "epoch": 3.2742040285899936,
      "grad_norm": 1.0390625,
      "learning_rate": 0.00039517266056757097,
      "loss": 5.6754,
      "mean_token_accuracy": 0.1754453957080841,
      "num_tokens": 64080836.0,
      "step": 32755
    },
    {
      "entropy": 6.000471591949463,
      "epoch": 3.2747038536512223,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039514248674500055,
      "loss": 5.4682,
      "mean_token_accuracy": 0.1985825330018997,
      "num_tokens": 64090533.0,
      "step": 32760
    },
    {
      "entropy": 5.947028636932373,
      "epoch": 3.2752036787124506,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039511230989967594,
      "loss": 5.3907,
      "mean_token_accuracy": 0.19812233746051788,
      "num_tokens": 64100156.0,
      "step": 32765
    },
    {
      "entropy": 5.9004558563232425,
      "epoch": 3.2757035037736792,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003950821300323566,
      "loss": 5.3897,
      "mean_token_accuracy": 0.19731011092662812,
      "num_tokens": 64108906.0,
      "step": 32770
    },
    {
      "entropy": 5.929925870895386,
      "epoch": 3.276203328834908,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003950519471438017,
      "loss": 5.4379,
      "mean_token_accuracy": 0.20190121680498124,
      "num_tokens": 64118361.0,
      "step": 32775
    },
    {
      "entropy": 6.064482545852661,
      "epoch": 3.276703153896136,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039502176123477076,
      "loss": 5.6012,
      "mean_token_accuracy": 0.18348991870880127,
      "num_tokens": 64128978.0,
      "step": 32780
    },
    {
      "entropy": 6.002956295013428,
      "epoch": 3.277202978957365,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00039499157230602313,
      "loss": 5.4245,
      "mean_token_accuracy": 0.20104742795228958,
      "num_tokens": 64139424.0,
      "step": 32785
    },
    {
      "entropy": 5.979306030273437,
      "epoch": 3.2777028040185936,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039496138035831845,
      "loss": 5.5018,
      "mean_token_accuracy": 0.19063517600297927,
      "num_tokens": 64149814.0,
      "step": 32790
    },
    {
      "entropy": 5.994144058227539,
      "epoch": 3.2782026290798223,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039493118539241643,
      "loss": 5.5089,
      "mean_token_accuracy": 0.18826308995485305,
      "num_tokens": 64159777.0,
      "step": 32795
    },
    {
      "entropy": 5.935672664642334,
      "epoch": 3.2787024541410505,
      "grad_norm": 1.0625,
      "learning_rate": 0.00039490098740907654,
      "loss": 5.3805,
      "mean_token_accuracy": 0.20332874804735185,
      "num_tokens": 64170241.0,
      "step": 32800
    },
    {
      "entropy": 5.907087659835815,
      "epoch": 3.2792022792022792,
      "grad_norm": 1.046875,
      "learning_rate": 0.00039487078640905866,
      "loss": 5.3797,
      "mean_token_accuracy": 0.1994429662823677,
      "num_tokens": 64180514.0,
      "step": 32805
    },
    {
      "entropy": 5.957804012298584,
      "epoch": 3.279702104263508,
      "grad_norm": 1.125,
      "learning_rate": 0.00039484058239312264,
      "loss": 5.4851,
      "mean_token_accuracy": 0.19189665615558624,
      "num_tokens": 64191184.0,
      "step": 32810
    },
    {
      "entropy": 6.015394878387451,
      "epoch": 3.280201929324736,
      "grad_norm": 1.1796875,
      "learning_rate": 0.00039481037536202835,
      "loss": 5.4771,
      "mean_token_accuracy": 0.19653286635875702,
      "num_tokens": 64201134.0,
      "step": 32815
    },
    {
      "entropy": 6.007282352447509,
      "epoch": 3.280701754385965,
      "grad_norm": 1.234375,
      "learning_rate": 0.00039478016531653586,
      "loss": 5.415,
      "mean_token_accuracy": 0.19773150980472565,
      "num_tokens": 64210578.0,
      "step": 32820
    },
    {
      "entropy": 6.0441371440887455,
      "epoch": 3.2812015794471936,
      "grad_norm": 1.078125,
      "learning_rate": 0.00039474995225740526,
      "loss": 5.6108,
      "mean_token_accuracy": 0.1851646140217781,
      "num_tokens": 64219762.0,
      "step": 32825
    },
    {
      "entropy": 5.938585567474365,
      "epoch": 3.2817014045084223,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003947197361853968,
      "loss": 5.4219,
      "mean_token_accuracy": 0.1977101296186447,
      "num_tokens": 64230371.0,
      "step": 32830
    },
    {
      "entropy": 5.9383721351623535,
      "epoch": 3.2822012295696505,
      "grad_norm": 1.203125,
      "learning_rate": 0.00039468951710127043,
      "loss": 5.4951,
      "mean_token_accuracy": 0.1996215507388115,
      "num_tokens": 64239154.0,
      "step": 32835
    },
    {
      "entropy": 5.9371413230896,
      "epoch": 3.2827010546308792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0003946592950057867,
      "loss": 5.4221,
      "mean_token_accuracy": 0.20122209638357164,
      "num_tokens": 64249257.0,
      "step": 32840
    },
    {
      "entropy": 6.0398822784423825,
      "epoch": 3.283200879692108,
      "grad_norm": 1.1796875,
      "learning_rate": 0.0003946290698997058,
      "loss": 5.4974,
      "mean_token_accuracy": 0.19162861555814742,
      "num_tokens": 64259171.0,
      "step": 32845
    },
    {
      "entropy": 5.900046300888062,
      "epoch": 3.283700704753336,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003945988417837883,
      "loss": 5.3103,
      "mean_token_accuracy": 0.20960710644721986,
      "num_tokens": 64268178.0,
      "step": 32850
    },
    {
      "entropy": 5.98159122467041,
      "epoch": 3.284200529814565,
      "grad_norm": 1.21875,
      "learning_rate": 0.0003945686106587948,
      "loss": 5.5126,
      "mean_token_accuracy": 0.18921183049678802,
      "num_tokens": 64278136.0,
      "step": 32855
    },
    {
      "entropy": 5.883253765106201,
      "epoch": 3.2847003548757936,
      "grad_norm": 1.03125,
      "learning_rate": 0.00039453837652548575,
      "loss": 5.3982,
      "mean_token_accuracy": 0.19769660532474517,
      "num_tokens": 64288305.0,
      "step": 32860
    },
    {
      "entropy": 5.922656011581421,
      "epoch": 3.2852001799370223,
      "grad_norm": 1.1171875,
      "learning_rate": 0.00039450813938462197,
      "loss": 5.403,
      "mean_token_accuracy": 0.20403777062892914,
      "num_tokens": 64297636.0,
      "step": 32865
    },
    {
      "entropy": 5.999955844879151,
      "epoch": 3.2857000049982505,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039447789923696415,
      "loss": 5.4329,
      "mean_token_accuracy": 0.20167066305875778,
      "num_tokens": 64306706.0,
      "step": 32870
    },
    {
      "entropy": 5.942225408554077,
      "epoch": 3.2861998300594792,
      "grad_norm": 1.2890625,
      "learning_rate": 0.000394447656083273,
      "loss": 5.4088,
      "mean_token_accuracy": 0.1963708594441414,
      "num_tokens": 64316909.0,
      "step": 32875
    },
    {
      "entropy": 5.893703460693359,
      "epoch": 3.286699655120708,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039441740992430965,
      "loss": 5.3843,
      "mean_token_accuracy": 0.1995649218559265,
      "num_tokens": 64326103.0,
      "step": 32880
    },
    {
      "entropy": 5.969772529602051,
      "epoch": 3.287199480181936,
      "grad_norm": 1.1875,
      "learning_rate": 0.00039438716076083504,
      "loss": 5.4347,
      "mean_token_accuracy": 0.19892969578504563,
      "num_tokens": 64334915.0,
      "step": 32885
    },
    {
      "entropy": 5.775533866882324,
      "epoch": 3.287699305243165,
      "grad_norm": 1.0078125,
      "learning_rate": 0.00039435690859361013,
      "loss": 5.3884,
      "mean_token_accuracy": 0.20987465381622314,
      "num_tokens": 64344102.0,
      "step": 32890
    },
    {
      "entropy": 5.893475961685181,
      "epoch": 3.2881991303043936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039432665342339614,
      "loss": 5.4373,
      "mean_token_accuracy": 0.19906232208013536,
      "num_tokens": 64353011.0,
      "step": 32895
    },
    {
      "entropy": 5.9246190071105955,
      "epoch": 3.288698955365622,
      "grad_norm": 1.265625,
      "learning_rate": 0.0003942963952509543,
      "loss": 5.3926,
      "mean_token_accuracy": 0.19638497531414031,
      "num_tokens": 64361658.0,
      "step": 32900
    },
    {
      "entropy": 6.036815547943116,
      "epoch": 3.2891987804268505,
      "grad_norm": 1.015625,
      "learning_rate": 0.00039426613407704574,
      "loss": 5.5908,
      "mean_token_accuracy": 0.1855558693408966,
      "num_tokens": 64371878.0,
      "step": 32905
    },
    {
      "entropy": 6.002215623855591,
      "epoch": 3.2896986054880792,
      "grad_norm": 1.0234375,
      "learning_rate": 0.000394235869902432,
      "loss": 5.4211,
      "mean_token_accuracy": 0.19577086567878724,
      "num_tokens": 64381896.0,
      "step": 32910
    },
    {
      "entropy": 5.985514688491821,
      "epoch": 3.2901984305493075,
      "grad_norm": 1.109375,
      "learning_rate": 0.00039420560272787446,
      "loss": 5.5043,
      "mean_token_accuracy": 0.18994371443986893,
      "num_tokens": 64391618.0,
      "step": 32915
    },
    {
      "entropy": 5.972357225418091,
      "epoch": 3.290698255610536,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00039417533255413465,
      "loss": 5.4607,
      "mean_token_accuracy": 0.194880847632885,
      "num_tokens": 64402165.0,
      "step": 32920
    },
    {
      "entropy": 6.02699728012085,
      "epoch": 3.291198080671765,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039414505938197404,
      "loss": 5.5434,
      "mean_token_accuracy": 0.18551586121320723,
      "num_tokens": 64411334.0,
      "step": 32925
    },
    {
      "entropy": 6.051137065887451,
      "epoch": 3.2916979057329936,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0003941147832121545,
      "loss": 5.4624,
      "mean_token_accuracy": 0.19409819394350053,
      "num_tokens": 64420985.0,
      "step": 32930
    },
    {
      "entropy": 5.997770023345947,
      "epoch": 3.292197730794222,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003940845040454376,
      "loss": 5.5013,
      "mean_token_accuracy": 0.1922399193048477,
      "num_tokens": 64430141.0,
      "step": 32935
    },
    {
      "entropy": 5.918601417541504,
      "epoch": 3.2926975558554505,
      "grad_norm": 1.140625,
      "learning_rate": 0.0003940542218825853,
      "loss": 5.4272,
      "mean_token_accuracy": 0.19425111562013625,
      "num_tokens": 64440321.0,
      "step": 32940
    },
    {
      "entropy": 5.919700336456299,
      "epoch": 3.2931973809166792,
      "grad_norm": 1.140625,
      "learning_rate": 0.00039402393672435934,
      "loss": 5.4841,
      "mean_token_accuracy": 0.19722367972135543,
      "num_tokens": 64449422.0,
      "step": 32945
    },
    {
      "entropy": 6.0273191928863525,
      "epoch": 3.2936972059779075,
      "grad_norm": 1.1640625,
      "learning_rate": 0.00039399364857152177,
      "loss": 5.5675,
      "mean_token_accuracy": 0.18517733812332154,
      "num_tokens": 64459179.0,
      "step": 32950
    },
    {
      "entropy": 5.974307060241699,
      "epoch": 3.294197031039136,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003939633574248346,
      "loss": 5.4029,
      "mean_token_accuracy": 0.2018720880150795,
      "num_tokens": 64468454.0,
      "step": 32955
    },
    {
      "entropy": 5.897057056427002,
      "epoch": 3.294696856100365,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00039393306328506004,
      "loss": 5.4207,
      "mean_token_accuracy": 0.2072811871767044,
      "num_tokens": 64478374.0,
      "step": 32960
    },
    {
      "entropy": 5.985844087600708,
      "epoch": 3.2951966811615936,
      "grad_norm": 1.0546875,
      "learning_rate": 0.00039390276615296015,
      "loss": 5.5105,
      "mean_token_accuracy": 0.19520298838615419,
      "num_tokens": 64489692.0,
      "step": 32965
    },
    {
      "entropy": 6.008019018173218,
      "epoch": 3.295696506222822,
      "grad_norm": 1.171875,
      "learning_rate": 0.00039387246602929726,
      "loss": 5.4418,
      "mean_token_accuracy": 0.20357026308774948,
      "num_tokens": 64500167.0,
      "step": 32970
    },
    {
      "entropy": 6.030167198181152,
      "epoch": 3.2961963312840505,
      "grad_norm": 1.4921875,
      "learning_rate": 0.0003938421629148338,
      "loss": 5.4833,
      "mean_token_accuracy": 0.1901509091258049,
      "num_tokens": 64509973.0,
      "step": 32975
    },
    {
      "entropy": 5.983188199996948,
      "epoch": 3.2966961563452792,
      "grad_norm": 1.1484375,
      "learning_rate": 0.000393811856810332,
      "loss": 5.4917,
      "mean_token_accuracy": 0.19164818227291108,
      "num_tokens": 64519926.0,
      "step": 32980
    },
    {
      "entropy": 6.025173282623291,
      "epoch": 3.2971959814065075,
      "grad_norm": 1.1328125,
      "learning_rate": 0.0003937815477165545,
      "loss": 5.5386,
      "mean_token_accuracy": 0.19527139961719514,
      "num_tokens": 64529770.0,
      "step": 32985
    },
    {
      "entropy": 5.9973344802856445,
      "epoch": 3.297695806467736,
      "grad_norm": 1.03125,
      "learning_rate": 0.0003937512356342638,
      "loss": 5.3327,
      "mean_token_accuracy": 0.20868856757879256,
      "num_tokens": 64539867.0,
      "step": 32990
    },
    {
      "entropy": 5.9944030284881595,
      "epoch": 3.298195631528965,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0003937209205642226,
      "loss": 5.422,
      "mean_token_accuracy": 0.20812058746814727,
      "num_tokens": 64549276.0,
      "step": 32995
    },
    {
      "entropy": 5.979561710357666,
      "epoch": 3.2986954565901936,
      "grad_norm": 1.15625,
      "learning_rate": 0.0003936906025071936,
      "loss": 5.5154,
      "mean_token_accuracy": 0.19104951471090317,
      "num_tokens": 64559294.0,
      "step": 33000
    },
    {
      "epoch": 3.2986954565901936,
      "eval_entropy": 5.761125814788767,
      "eval_loss": 5.603759288787842,
      "eval_mean_token_accuracy": 0.19607604573668203,
      "eval_num_tokens": 64559294.0,
      "eval_runtime": 24.387,
      "eval_samples_per_second": 1273.015,
      "eval_steps_per_second": 159.142,
      "step": 33000
    }
  ],
  "logging_steps": 5,
  "max_steps": 100030,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.02454941044736e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}