{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 420,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 1.1769300036132335,
      "epoch": 0.04770423375074538,
      "grad_norm": 0.29830965399742126,
      "learning_rate": 0.00013846153846153847,
      "loss": 1.0886381149291993,
      "mean_token_accuracy": 0.7113694898784161,
      "num_tokens": 548245.0,
      "step": 10
    },
    {
      "entropy": 1.1637193433940412,
      "epoch": 0.09540846750149076,
      "grad_norm": 0.19875450432300568,
      "learning_rate": 0.00019989277259322314,
      "loss": 0.917241096496582,
      "mean_token_accuracy": 0.7484722793102264,
      "num_tokens": 1165075.0,
      "step": 20
    },
    {
      "entropy": 1.160731501877308,
      "epoch": 0.14311270125223613,
      "grad_norm": 0.22726039588451385,
      "learning_rate": 0.0001992383265674776,
      "loss": 0.9866914749145508,
      "mean_token_accuracy": 0.7289508931338787,
      "num_tokens": 1751880.0,
      "step": 30
    },
    {
      "entropy": 1.1374205633997918,
      "epoch": 0.19081693500298152,
      "grad_norm": 0.2027043253183365,
      "learning_rate": 0.00019799289791214725,
      "loss": 0.9542501449584961,
      "mean_token_accuracy": 0.742580084502697,
      "num_tokens": 2318760.0,
      "step": 40
    },
    {
      "entropy": 1.1097479555755854,
      "epoch": 0.2385211687537269,
      "grad_norm": 0.20862312614917755,
      "learning_rate": 0.00019616390338572803,
      "loss": 0.9246286392211914,
      "mean_token_accuracy": 0.7403309985995292,
      "num_tokens": 2921187.0,
      "step": 50
    },
    {
      "entropy": 1.1638598643243312,
      "epoch": 0.28622540250447226,
      "grad_norm": 0.18571633100509644,
      "learning_rate": 0.0001937622349896483,
      "loss": 0.9472658157348632,
      "mean_token_accuracy": 0.7402492307126523,
      "num_tokens": 3480173.0,
      "step": 60
    },
    {
      "entropy": 1.107759954035282,
      "epoch": 0.33392963625521765,
      "grad_norm": 0.16045409440994263,
      "learning_rate": 0.0001908021951043814,
      "loss": 0.9042925834655762,
      "mean_token_accuracy": 0.7455899693071842,
      "num_tokens": 4081543.0,
      "step": 70
    },
    {
      "entropy": 1.055170290917158,
      "epoch": 0.38163387000596305,
      "grad_norm": 0.17577198147773743,
      "learning_rate": 0.00018730141131611882,
      "loss": 0.8980257034301757,
      "mean_token_accuracy": 0.7503466323018074,
      "num_tokens": 4722061.0,
      "step": 80
    },
    {
      "entropy": 1.100816760957241,
      "epoch": 0.4293381037567084,
      "grad_norm": 0.18504640460014343,
      "learning_rate": 0.00018328073144122708,
      "loss": 0.9144180297851563,
      "mean_token_accuracy": 0.7497260145843029,
      "num_tokens": 5295924.0,
      "step": 90
    },
    {
      "entropy": 1.0832970976829528,
      "epoch": 0.4770423375074538,
      "grad_norm": 0.1606772094964981,
      "learning_rate": 0.00017876409937363677,
      "loss": 0.8887553215026855,
      "mean_token_accuracy": 0.7472214981913566,
      "num_tokens": 5907808.0,
      "step": 100
    },
    {
      "entropy": 1.0531512849032878,
      "epoch": 0.5247465712581991,
      "grad_norm": 0.1600116342306137,
      "learning_rate": 0.00017377841249451595,
      "loss": 0.8937862396240235,
      "mean_token_accuracy": 0.7559543505311013,
      "num_tokens": 6535137.0,
      "step": 110
    },
    {
      "entropy": 1.1238082095980644,
      "epoch": 0.5724508050089445,
      "grad_norm": 0.16486503183841705,
      "learning_rate": 0.00016835336149337974,
      "loss": 0.9225013732910157,
      "mean_token_accuracy": 0.7414056904613971,
      "num_tokens": 7140044.0,
      "step": 120
    },
    {
      "entropy": 1.0498989395797254,
      "epoch": 0.6201550387596899,
      "grad_norm": 0.1572185754776001,
      "learning_rate": 0.0001625212535545306,
      "loss": 0.8837360382080078,
      "mean_token_accuracy": 0.7528554752469063,
      "num_tokens": 7756923.0,
      "step": 130
    },
    {
      "entropy": 1.1506639942526817,
      "epoch": 0.6678592725104353,
      "grad_norm": 0.1634964644908905,
      "learning_rate": 0.00015631681996178736,
      "loss": 0.9271099090576171,
      "mean_token_accuracy": 0.7475036658346653,
      "num_tokens": 8325700.0,
      "step": 140
    },
    {
      "entropy": 1.0807946488261222,
      "epoch": 0.7155635062611807,
      "grad_norm": 0.191319540143013,
      "learning_rate": 0.0001497770092672523,
      "loss": 0.899015998840332,
      "mean_token_accuracy": 0.7469603985548019,
      "num_tokens": 8918061.0,
      "step": 150
    },
    {
      "entropy": 1.123007158190012,
      "epoch": 0.7632677400119261,
      "grad_norm": 0.16693302989006042,
      "learning_rate": 0.00014294076725583464,
      "loss": 0.9072439193725585,
      "mean_token_accuracy": 0.7461257085204125,
      "num_tokens": 9484460.0,
      "step": 160
    },
    {
      "entropy": 1.0272136367857456,
      "epoch": 0.8109719737626714,
      "grad_norm": 0.18687766790390015,
      "learning_rate": 0.00013584880501588223,
      "loss": 0.8413015365600586,
      "mean_token_accuracy": 0.7642282634973526,
      "num_tokens": 10099360.0,
      "step": 170
    },
    {
      "entropy": 1.158793331682682,
      "epoch": 0.8586762075134168,
      "grad_norm": 0.21139191091060638,
      "learning_rate": 0.00012854335649710436,
      "loss": 0.9355466842651368,
      "mean_token_accuracy": 0.7439247660338879,
      "num_tokens": 10625814.0,
      "step": 180
    },
    {
      "entropy": 1.0972598642110825,
      "epoch": 0.9063804412641622,
      "grad_norm": 0.17964082956314087,
      "learning_rate": 0.00012106792699957263,
      "loss": 0.8928470611572266,
      "mean_token_accuracy": 0.7530390501022339,
      "num_tokens": 11182964.0,
      "step": 190
    },
    {
      "entropy": 1.084363154321909,
      "epoch": 0.9540846750149076,
      "grad_norm": 0.1744077056646347,
      "learning_rate": 0.00011346703409159493,
      "loss": 0.9049121856689453,
      "mean_token_accuracy": 0.7473168261349201,
      "num_tokens": 11785645.0,
      "step": 200
    },
    {
      "entropy": 1.1082378610388024,
      "epoch": 1.0,
      "grad_norm": 0.2735864818096161,
      "learning_rate": 0.00010578594249934432,
      "loss": 0.9193456649780274,
      "mean_token_accuracy": 0.7463866650284111,
      "num_tokens": 12384712.0,
      "step": 210
    },
    {
      "entropy": 1.005769143998623,
      "epoch": 1.0477042337507454,
      "grad_norm": 0.1691940575838089,
      "learning_rate": 9.807039454702435e-05,
      "loss": 0.8218914031982422,
      "mean_token_accuracy": 0.7632846429944038,
      "num_tokens": 12967294.0,
      "step": 220
    },
    {
      "entropy": 1.0322131134569645,
      "epoch": 1.0954084675014908,
      "grad_norm": 0.15159490704536438,
      "learning_rate": 9.036633775284968e-05,
      "loss": 0.8452365875244141,
      "mean_token_accuracy": 0.7641348131000996,
      "num_tokens": 13569612.0,
      "step": 230
    },
    {
      "entropy": 1.0705332525074482,
      "epoch": 1.1431127012522362,
      "grad_norm": 0.2022983729839325,
      "learning_rate": 8.271965120305814e-05,
      "loss": 0.8433799743652344,
      "mean_token_accuracy": 0.7596304841339588,
      "num_tokens": 14154144.0,
      "step": 240
    },
    {
      "entropy": 1.109963207691908,
      "epoch": 1.1908169350029816,
      "grad_norm": 0.21518675982952118,
      "learning_rate": 7.517587233344819e-05,
      "loss": 0.9084970474243164,
      "mean_token_accuracy": 0.7477858312427997,
      "num_tokens": 14685396.0,
      "step": 250
    },
    {
      "entropy": 1.0471076007932425,
      "epoch": 1.238521168753727,
      "grad_norm": 0.1924280822277069,
      "learning_rate": 6.777992574550767e-05,
      "loss": 0.8424646377563476,
      "mean_token_accuracy": 0.7590488582849503,
      "num_tokens": 15259589.0,
      "step": 260
    },
    {
      "entropy": 1.0231421060860157,
      "epoch": 1.2862254025044724,
      "grad_norm": 0.1902351826429367,
      "learning_rate": 6.057585567208484e-05,
      "loss": 0.8693572998046875,
      "mean_token_accuracy": 0.761046989262104,
      "num_tokens": 15871459.0,
      "step": 270
    },
    {
      "entropy": 1.0073398746550084,
      "epoch": 1.3339296362552178,
      "grad_norm": 0.16771914064884186,
      "learning_rate": 5.360656368581765e-05,
      "loss": 0.817138671875,
      "mean_token_accuracy": 0.7684732794761657,
      "num_tokens": 16433555.0,
      "step": 280
    },
    {
      "entropy": 1.0807819075882434,
      "epoch": 1.3816338700059632,
      "grad_norm": 0.17870068550109863,
      "learning_rate": 4.6913553212316454e-05,
      "loss": 0.8738027572631836,
      "mean_token_accuracy": 0.7546135231852531,
      "num_tokens": 17031607.0,
      "step": 290
    },
    {
      "entropy": 1.017206584662199,
      "epoch": 1.4293381037567083,
      "grad_norm": 0.2641982436180115,
      "learning_rate": 4.053668236957134e-05,
      "loss": 0.8472692489624023,
      "mean_token_accuracy": 0.760607336461544,
      "num_tokens": 17620725.0,
      "step": 300
    },
    {
      "entropy": 1.0453648515045644,
      "epoch": 1.4770423375074537,
      "grad_norm": 0.18742035329341888,
      "learning_rate": 3.45139266054715e-05,
      "loss": 0.8469044685363769,
      "mean_token_accuracy": 0.760916006565094,
      "num_tokens": 18203144.0,
      "step": 310
    },
    {
      "entropy": 1.0301007013767958,
      "epoch": 1.5247465712581991,
      "grad_norm": 0.15879590809345245,
      "learning_rate": 2.888115254697371e-05,
      "loss": 0.8580363273620606,
      "mean_token_accuracy": 0.7620598554611206,
      "num_tokens": 18821220.0,
      "step": 320
    },
    {
      "entropy": 1.019312507659197,
      "epoch": 1.5724508050089445,
      "grad_norm": 0.18043670058250427,
      "learning_rate": 2.3671904407690704e-05,
      "loss": 0.8310068130493165,
      "mean_token_accuracy": 0.7651938483119011,
      "num_tokens": 19436758.0,
      "step": 330
    },
    {
      "entropy": 1.063497780263424,
      "epoch": 1.62015503875969,
      "grad_norm": 0.18336360156536102,
      "learning_rate": 1.8917204225881234e-05,
      "loss": 0.8812410354614257,
      "mean_token_accuracy": 0.7559928879141807,
      "num_tokens": 19994077.0,
      "step": 340
    },
    {
      "entropy": 0.9850323311984539,
      "epoch": 1.6678592725104353,
      "grad_norm": 0.2079457938671112,
      "learning_rate": 1.464536712246205e-05,
      "loss": 0.8087331771850585,
      "mean_token_accuracy": 0.7675255030393601,
      "num_tokens": 20577179.0,
      "step": 350
    },
    {
      "entropy": 1.0592832632362843,
      "epoch": 1.7155635062611807,
      "grad_norm": 0.18709249794483185,
      "learning_rate": 1.0881832679214276e-05,
      "loss": 0.8613182067871094,
      "mean_token_accuracy": 0.753790495544672,
      "num_tokens": 21138600.0,
      "step": 360
    },
    {
      "entropy": 1.0343884475529195,
      "epoch": 1.763267740011926,
      "grad_norm": 0.1785227358341217,
      "learning_rate": 7.649013441358465e-06,
      "loss": 0.8642744064331055,
      "mean_token_accuracy": 0.7537333220243454,
      "num_tokens": 21799131.0,
      "step": 370
    },
    {
      "entropy": 1.0254336915910245,
      "epoch": 1.8109719737626713,
      "grad_norm": 0.19311416149139404,
      "learning_rate": 4.96616144669333e-06,
      "loss": 0.8304064750671387,
      "mean_token_accuracy": 0.7594926789402962,
      "num_tokens": 22420264.0,
      "step": 380
    },
    {
      "entropy": 1.0619895912706852,
      "epoch": 1.8586762075134167,
      "grad_norm": 0.18003182113170624,
      "learning_rate": 2.8492535761421636e-06,
      "loss": 0.8686416625976563,
      "mean_token_accuracy": 0.7512479916214942,
      "num_tokens": 23038435.0,
      "step": 390
    },
    {
      "entropy": 0.9937550373375416,
      "epoch": 1.906380441264162,
      "grad_norm": 0.20265743136405945,
      "learning_rate": 1.3108964084654007e-06,
      "loss": 0.8259183883666992,
      "mean_token_accuracy": 0.7704898312687873,
      "num_tokens": 23624513.0,
      "step": 400
    },
    {
      "entropy": 0.9902137991040945,
      "epoch": 1.9540846750149075,
      "grad_norm": 0.20135828852653503,
      "learning_rate": 3.6025114574734785e-07,
      "loss": 0.8437560081481934,
      "mean_token_accuracy": 0.7647472023963928,
      "num_tokens": 24196254.0,
      "step": 410
    },
    {
      "entropy": 1.028095632404476,
      "epoch": 2.0,
      "grad_norm": 0.22397543489933014,
      "learning_rate": 2.979056739682573e-09,
      "loss": 0.8373299598693847,
      "mean_token_accuracy": 0.7613624914900049,
      "num_tokens": 24769424.0,
      "step": 420
    }
  ],
  "logging_steps": 10,
  "max_steps": 420,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.088009104887644e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}