{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.4375938081741333,
      "epoch": 0.021458076283461188,
      "grad_norm": 3.328125,
      "learning_rate": 6.282051282051282e-06,
      "loss": 1.6719,
      "mean_token_accuracy": 0.6502519898116589,
      "num_tokens": 13027304.0,
      "step": 50
    },
    {
      "entropy": 1.4517733630537988,
      "epoch": 0.042916152566922376,
      "grad_norm": 0.85546875,
      "learning_rate": 1.2692307692307693e-05,
      "loss": 1.4632,
      "mean_token_accuracy": 0.6732779324054718,
      "num_tokens": 26060152.0,
      "step": 100
    },
    {
      "entropy": 1.364333181977272,
      "epoch": 0.06437422885038356,
      "grad_norm": 0.8671875,
      "learning_rate": 1.9102564102564103e-05,
      "loss": 1.3631,
      "mean_token_accuracy": 0.6902350637316704,
      "num_tokens": 39093818.0,
      "step": 150
    },
    {
      "entropy": 1.3189326703548432,
      "epoch": 0.08583230513384475,
      "grad_norm": 0.859375,
      "learning_rate": 2.551282051282051e-05,
      "loss": 1.3149,
      "mean_token_accuracy": 0.6986461882293225,
      "num_tokens": 52123695.0,
      "step": 200
    },
    {
      "entropy": 1.269011721611023,
      "epoch": 0.10729038141730594,
      "grad_norm": 0.921875,
      "learning_rate": 2.9996212711496212e-05,
      "loss": 1.2641,
      "mean_token_accuracy": 0.707592583745718,
      "num_tokens": 65156301.0,
      "step": 250
    },
    {
      "entropy": 1.2357548919320107,
      "epoch": 0.12874845770076712,
      "grad_norm": 0.84375,
      "learning_rate": 2.992893632784019e-05,
      "loss": 1.2307,
      "mean_token_accuracy": 0.7134250193834305,
      "num_tokens": 78189155.0,
      "step": 300
    },
    {
      "entropy": 1.20577629506588,
      "epoch": 0.1502065339842283,
      "grad_norm": 0.859375,
      "learning_rate": 2.9777932337106776e-05,
      "loss": 1.2003,
      "mean_token_accuracy": 0.7188817055523395,
      "num_tokens": 91225387.0,
      "step": 350
    },
    {
      "entropy": 1.1942224881052972,
      "epoch": 0.1716646102676895,
      "grad_norm": 0.8671875,
      "learning_rate": 2.95440476317018e-05,
      "loss": 1.1881,
      "mean_token_accuracy": 0.7209276972711086,
      "num_tokens": 104264395.0,
      "step": 400
    },
    {
      "entropy": 1.1793697586655616,
      "epoch": 0.1931226865511507,
      "grad_norm": 0.828125,
      "learning_rate": 2.9228593933122738e-05,
      "loss": 1.1736,
      "mean_token_accuracy": 0.7238370491564274,
      "num_tokens": 117304006.0,
      "step": 450
    },
    {
      "entropy": 1.167185415327549,
      "epoch": 0.2145807628346119,
      "grad_norm": 0.8125,
      "learning_rate": 2.8833340435285732e-05,
      "loss": 1.1612,
      "mean_token_accuracy": 0.7257441227138043,
      "num_tokens": 130330369.0,
      "step": 500
    },
    {
      "entropy": 1.1495196589827537,
      "epoch": 0.23603883911807305,
      "grad_norm": 0.80859375,
      "learning_rate": 2.8360503882159413e-05,
      "loss": 1.1438,
      "mean_token_accuracy": 0.7290801952779293,
      "num_tokens": 143357759.0,
      "step": 550
    },
    {
      "entropy": 1.1417414979636669,
      "epoch": 0.25749691540153424,
      "grad_norm": 0.83984375,
      "learning_rate": 2.7812736135354192e-05,
      "loss": 1.1358,
      "mean_token_accuracy": 0.7306405635178089,
      "num_tokens": 156390933.0,
      "step": 600
    },
    {
      "entropy": 1.1334842164814471,
      "epoch": 0.27895499168499543,
      "grad_norm": 0.7734375,
      "learning_rate": 2.7193109301393063e-05,
      "loss": 1.1269,
      "mean_token_accuracy": 0.7320908455550671,
      "num_tokens": 169421819.0,
      "step": 650
    },
    {
      "entropy": 1.1232954794168473,
      "epoch": 0.3004130679684566,
      "grad_norm": 0.77734375,
      "learning_rate": 2.6505098502076353e-05,
      "loss": 1.1176,
      "mean_token_accuracy": 0.7338391813635826,
      "num_tokens": 182458556.0,
      "step": 700
    },
    {
      "entropy": 1.1202317014336587,
      "epoch": 0.3218711442519178,
      "grad_norm": 0.82421875,
      "learning_rate": 2.575256238457122e-05,
      "loss": 1.1138,
      "mean_token_accuracy": 0.7346239614486695,
      "num_tokens": 195498872.0,
      "step": 750
    },
    {
      "entropy": 1.1189795224368573,
      "epoch": 0.343329220535379,
      "grad_norm": 0.79296875,
      "learning_rate": 2.4939721480533394e-05,
      "loss": 1.1128,
      "mean_token_accuracy": 0.7346904000639916,
      "num_tokens": 208529244.0,
      "step": 800
    },
    {
      "entropy": 1.111073449254036,
      "epoch": 0.3647872968188402,
      "grad_norm": 0.74609375,
      "learning_rate": 2.4071134535632042e-05,
      "loss": 1.1053,
      "mean_token_accuracy": 0.7359747096896172,
      "num_tokens": 221559622.0,
      "step": 850
    },
    {
      "entropy": 1.104882825911045,
      "epoch": 0.3862453731023014,
      "grad_norm": 0.75390625,
      "learning_rate": 2.3151672942231506e-05,
      "loss": 1.0986,
      "mean_token_accuracy": 0.737260360866785,
      "num_tokens": 234596401.0,
      "step": 900
    },
    {
      "entropy": 1.100306123793125,
      "epoch": 0.4077034493857626,
      "grad_norm": 0.80859375,
      "learning_rate": 2.218649341862198e-05,
      "loss": 1.094,
      "mean_token_accuracy": 0.7378678020834922,
      "num_tokens": 247619938.0,
      "step": 950
    },
    {
      "entropy": 1.103239623606205,
      "epoch": 0.4291615256692238,
      "grad_norm": 0.80078125,
      "learning_rate": 2.1181009088025206e-05,
      "loss": 1.0971,
      "mean_token_accuracy": 0.7376800465583802,
      "num_tokens": 260654172.0,
      "step": 1000
    },
    {
      "entropy": 1.101589569747448,
      "epoch": 0.45061960195268497,
      "grad_norm": 0.73046875,
      "learning_rate": 2.014085911957592e-05,
      "loss": 1.0956,
      "mean_token_accuracy": 0.7379778701066971,
      "num_tokens": 273684574.0,
      "step": 1050
    },
    {
      "entropy": 1.1031180422008038,
      "epoch": 0.4720776782361461,
      "grad_norm": 0.7265625,
      "learning_rate": 1.907187710154509e-05,
      "loss": 1.0963,
      "mean_token_accuracy": 0.7376254206895828,
      "num_tokens": 286722109.0,
      "step": 1100
    },
    {
      "entropy": 1.0942196384072305,
      "epoch": 0.4935357545196073,
      "grad_norm": 0.72265625,
      "learning_rate": 1.7980058324180654e-05,
      "loss": 1.0881,
      "mean_token_accuracy": 0.7393483504652977,
      "num_tokens": 299759049.0,
      "step": 1150
    },
    {
      "entropy": 1.094051382392645,
      "epoch": 0.5149938308030685,
      "grad_norm": 0.7421875,
      "learning_rate": 1.687152615565724e-05,
      "loss": 1.0874,
      "mean_token_accuracy": 0.7393186213076115,
      "num_tokens": 312785989.0,
      "step": 1200
    },
    {
      "entropy": 1.0880274675786494,
      "epoch": 0.5364519070865297,
      "grad_norm": 0.75,
      "learning_rate": 1.5752497699712207e-05,
      "loss": 1.0818,
      "mean_token_accuracy": 0.7406372900307179,
      "num_tokens": 325818163.0,
      "step": 1250
    },
    {
      "entropy": 1.0928636234998703,
      "epoch": 0.5579099833699909,
      "grad_norm": 0.7421875,
      "learning_rate": 1.4629248927574217e-05,
      "loss": 1.086,
      "mean_token_accuracy": 0.7395914824306965,
      "num_tokens": 338855461.0,
      "step": 1300
    },
    {
      "entropy": 1.0869983588159084,
      "epoch": 0.5793680596534521,
      "grad_norm": 0.7109375,
      "learning_rate": 1.3508079479738632e-05,
      "loss": 1.081,
      "mean_token_accuracy": 0.740579083263874,
      "num_tokens": 351888193.0,
      "step": 1350
    },
    {
      "entropy": 1.0896718701720238,
      "epoch": 0.6008261359369133,
      "grad_norm": 0.72265625,
      "learning_rate": 1.2395277334996045e-05,
      "loss": 1.0833,
      "mean_token_accuracy": 0.7400969637930394,
      "num_tokens": 364920249.0,
      "step": 1400
    },
    {
      "entropy": 1.0879578018188476,
      "epoch": 0.6222842122203744,
      "grad_norm": 0.69921875,
      "learning_rate": 1.1297083544864386e-05,
      "loss": 1.0813,
      "mean_token_accuracy": 0.7403733618557453,
      "num_tokens": 377954848.0,
      "step": 1450
    },
    {
      "entropy": 1.0857014648616314,
      "epoch": 0.6437422885038356,
      "grad_norm": 0.77734375,
      "learning_rate": 1.0219657231208455e-05,
      "loss": 1.0796,
      "mean_token_accuracy": 0.7408653677999973,
      "num_tokens": 390982547.0,
      "step": 1500
    },
    {
      "entropy": 1.081975871771574,
      "epoch": 0.6652003647872968,
      "grad_norm": 0.69921875,
      "learning_rate": 9.1690410433546e-06,
      "loss": 1.0755,
      "mean_token_accuracy": 0.7419320181012153,
      "num_tokens": 404012732.0,
      "step": 1550
    },
    {
      "entropy": 1.0881149007380009,
      "epoch": 0.686658441070758,
      "grad_norm": 0.69140625,
      "learning_rate": 8.151127268431263e-06,
      "loss": 1.0819,
      "mean_token_accuracy": 0.7401869963109493,
      "num_tokens": 417050367.0,
      "step": 1600
    },
    {
      "entropy": 1.084583478420973,
      "epoch": 0.7081165173542192,
      "grad_norm": 0.69921875,
      "learning_rate": 7.171624785002444e-06,
      "loss": 1.078,
      "mean_token_accuracy": 0.7409912987053394,
      "num_tokens": 430085128.0,
      "step": 1650
    },
    {
      "entropy": 1.0892414034903048,
      "epoch": 0.7295745936376804,
      "grad_norm": 0.71875,
      "learning_rate": 6.236027045331792e-06,
      "loss": 1.0831,
      "mean_token_accuracy": 0.7401366803050041,
      "num_tokens": 443116587.0,
      "step": 1700
    },
    {
      "entropy": 1.0882355618476867,
      "epoch": 0.7510326699211416,
      "grad_norm": 0.671875,
      "learning_rate": 5.349581265845804e-06,
      "loss": 1.0814,
      "mean_token_accuracy": 0.740454610735178,
      "num_tokens": 456150190.0,
      "step": 1750
    },
    {
      "entropy": 1.0862010414898395,
      "epoch": 0.7724907462046028,
      "grad_norm": 0.6953125,
      "learning_rate": 4.517258998588792e-06,
      "loss": 1.0798,
      "mean_token_accuracy": 0.7409322175383568,
      "num_tokens": 469187645.0,
      "step": 1800
    },
    {
      "entropy": 1.080193583369255,
      "epoch": 0.793948822488064,
      "grad_norm": 0.6953125,
      "learning_rate": 3.743728248716951e-06,
      "loss": 1.0729,
      "mean_token_accuracy": 0.7421776308119297,
      "num_tokens": 482216628.0,
      "step": 1850
    },
    {
      "entropy": 1.0861673779785632,
      "epoch": 0.8154068987715252,
      "grad_norm": 0.71875,
      "learning_rate": 3.033327294407988e-06,
      "loss": 1.0794,
      "mean_token_accuracy": 0.7408488847315311,
      "num_tokens": 495240948.0,
      "step": 1900
    },
    {
      "entropy": 1.0832784846425056,
      "epoch": 0.8368649750549864,
      "grad_norm": 0.6875,
      "learning_rate": 2.390040356015151e-06,
      "loss": 1.077,
      "mean_token_accuracy": 0.7417075361311436,
      "num_tokens": 508270012.0,
      "step": 1950
    },
    {
      "entropy": 1.081319029778242,
      "epoch": 0.8583230513384476,
      "grad_norm": 0.703125,
      "learning_rate": 1.817475250922957e-06,
      "loss": 1.0751,
      "mean_token_accuracy": 0.7418272741138935,
      "num_tokens": 521303590.0,
      "step": 2000
    },
    {
      "entropy": 1.0867520736157894,
      "epoch": 0.8797811276219087,
      "grad_norm": 0.69921875,
      "learning_rate": 1.318843159425332e-06,
      "loss": 1.0805,
      "mean_token_accuracy": 0.7406135281920433,
      "num_tokens": 534336739.0,
      "step": 2050
    },
    {
      "entropy": 1.0897665022313594,
      "epoch": 0.9012392039053699,
      "grad_norm": 0.70703125,
      "learning_rate": 8.969406151075494e-07,
      "loss": 1.084,
      "mean_token_accuracy": 0.740174468010664,
      "num_tokens": 547357092.0,
      "step": 2100
    },
    {
      "entropy": 1.0855583234131336,
      "epoch": 0.9226972801888311,
      "grad_norm": 0.70703125,
      "learning_rate": 5.541338207371605e-07,
      "loss": 1.0788,
      "mean_token_accuracy": 0.7410851241648198,
      "num_tokens": 560392664.0,
      "step": 2150
    },
    {
      "entropy": 1.0886439248919486,
      "epoch": 0.9441553564722922,
      "grad_norm": 0.6875,
      "learning_rate": 2.923453776269558e-07,
      "loss": 1.0819,
      "mean_token_accuracy": 0.7401887883245944,
      "num_tokens": 573423859.0,
      "step": 2200
    },
    {
      "entropy": 1.0926427023112775,
      "epoch": 0.9656134327557534,
      "grad_norm": 0.7265625,
      "learning_rate": 1.130435028970872e-07,
      "loss": 1.0858,
      "mean_token_accuracy": 0.739516359269619,
      "num_tokens": 586459673.0,
      "step": 2250
    },
    {
      "entropy": 1.0826670287549496,
      "epoch": 0.9870715090392146,
      "grad_norm": 0.74609375,
      "learning_rate": 1.7233795110367935e-08,
      "loss": 1.0754,
      "mean_token_accuracy": 0.7415802033245563,
      "num_tokens": 599492881.0,
      "step": 2300
    }
  ],
  "logging_steps": 50,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.135994653597762e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}