{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.4994252011795872,
  "eval_steps": 3000,
  "global_step": 15000,
  "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
    }
  ],
  "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": 4.6488498591744e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}