{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.962962962962963,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.1715362322330476,
      "epoch": 0.07407407407407407,
      "grad_norm": 1.0598933696746826,
      "learning_rate": 8.032786885245902e-05,
      "loss": 2.0077,
      "mean_token_accuracy": 0.5798640113323927,
      "num_tokens": 338396.0,
      "step": 50
    },
    {
      "entropy": 0.6553284931182861,
      "epoch": 0.14814814814814814,
      "grad_norm": 0.3852575719356537,
      "learning_rate": 9.990765991730485e-05,
      "loss": 0.5993,
      "mean_token_accuracy": 0.8612996307015419,
      "num_tokens": 677239.0,
      "step": 100
    },
    {
      "entropy": 0.4204328820481896,
      "epoch": 0.2222222222222222,
      "grad_norm": 0.2863098680973053,
      "learning_rate": 9.950545603782162e-05,
      "loss": 0.4172,
      "mean_token_accuracy": 0.9052096186578273,
      "num_tokens": 1015420.0,
      "step": 150
    },
    {
      "entropy": 0.3820256270095706,
      "epoch": 0.2962962962962963,
      "grad_norm": 0.2698630690574646,
      "learning_rate": 9.878674879048427e-05,
      "loss": 0.383,
      "mean_token_accuracy": 0.9129502369463444,
      "num_tokens": 1355048.0,
      "step": 200
    },
    {
      "entropy": 0.37178467601537707,
      "epoch": 0.37037037037037035,
      "grad_norm": 0.21257849037647247,
      "learning_rate": 9.775613308830824e-05,
      "loss": 0.37,
      "mean_token_accuracy": 0.9151540066301823,
      "num_tokens": 1691649.0,
      "step": 250
    },
    {
      "entropy": 0.36766253888607026,
      "epoch": 0.4444444444444444,
      "grad_norm": 0.31453582644462585,
      "learning_rate": 9.642019796948866e-05,
      "loss": 0.3634,
      "mean_token_accuracy": 0.9162409096956253,
      "num_tokens": 2029267.0,
      "step": 300
    },
    {
      "entropy": 0.35622088231146337,
      "epoch": 0.5185185185185185,
      "grad_norm": 0.20113833248615265,
      "learning_rate": 9.478748447168449e-05,
      "loss": 0.3533,
      "mean_token_accuracy": 0.9183417546749115,
      "num_tokens": 2369465.0,
      "step": 350
    },
    {
      "entropy": 0.3494216964393854,
      "epoch": 0.5925925925925926,
      "grad_norm": 0.20226770639419556,
      "learning_rate": 9.28684310265789e-05,
      "loss": 0.3492,
      "mean_token_accuracy": 0.919527986496687,
      "num_tokens": 2707235.0,
      "step": 400
    },
    {
      "entropy": 0.3433630639687181,
      "epoch": 0.6666666666666666,
      "grad_norm": 0.1721702516078949,
      "learning_rate": 9.067530672382544e-05,
      "loss": 0.3453,
      "mean_token_accuracy": 0.9204271128773689,
      "num_tokens": 3041282.0,
      "step": 450
    },
    {
      "entropy": 0.3465565311536193,
      "epoch": 0.7407407407407407,
      "grad_norm": 0.1957043558359146,
      "learning_rate": 8.822213287104348e-05,
      "loss": 0.3483,
      "mean_token_accuracy": 0.9200405342876912,
      "num_tokens": 3378157.0,
      "step": 500
    },
    {
      "entropy": 0.34500927191227676,
      "epoch": 0.8148148148148148,
      "grad_norm": 0.1682814359664917,
      "learning_rate": 8.552459335135381e-05,
      "loss": 0.3466,
      "mean_token_accuracy": 0.9210821086168289,
      "num_tokens": 3717487.0,
      "step": 550
    },
    {
      "entropy": 0.33874930143356324,
      "epoch": 0.8888888888888888,
      "grad_norm": 0.1756315678358078,
      "learning_rate": 8.259993435156559e-05,
      "loss": 0.3406,
      "mean_token_accuracy": 0.9231470887362957,
      "num_tokens": 4056814.0,
      "step": 600
    },
    {
      "entropy": 0.3372995613142848,
      "epoch": 0.9629629629629629,
      "grad_norm": 0.185526043176651,
      "learning_rate": 7.946685410208296e-05,
      "loss": 0.3378,
      "mean_token_accuracy": 0.924960307776928,
      "num_tokens": 4398736.0,
      "step": 650
    },
    {
      "entropy": 0.3282011279836297,
      "epoch": 1.037037037037037,
      "grad_norm": 0.23889201879501343,
      "learning_rate": 7.614538333345735e-05,
      "loss": 0.3293,
      "mean_token_accuracy": 0.9266741496324539,
      "num_tokens": 4737915.0,
      "step": 700
    },
    {
      "entropy": 0.32703890234231947,
      "epoch": 1.1111111111111112,
      "grad_norm": 0.21601562201976776,
      "learning_rate": 7.265675721386285e-05,
      "loss": 0.3275,
      "mean_token_accuracy": 0.9279554469883442,
      "num_tokens": 5074721.0,
      "step": 750
    },
    {
      "entropy": 0.3243879374489188,
      "epoch": 1.1851851851851851,
      "grad_norm": 0.22092069685459137,
      "learning_rate": 6.902327958623736e-05,
      "loss": 0.3244,
      "mean_token_accuracy": 0.9288445146381855,
      "num_tokens": 5413515.0,
      "step": 800
    },
    {
      "entropy": 0.3191816185042262,
      "epoch": 1.2592592592592593,
      "grad_norm": 0.2003597766160965,
      "learning_rate": 6.526818037306228e-05,
      "loss": 0.3197,
      "mean_token_accuracy": 0.9303379128873348,
      "num_tokens": 5751965.0,
      "step": 850
    },
    {
      "entropy": 0.311707403101027,
      "epoch": 1.3333333333333333,
      "grad_norm": 0.1664246916770935,
      "learning_rate": 6.14154670604355e-05,
      "loss": 0.3122,
      "mean_token_accuracy": 0.9324613013863563,
      "num_tokens": 6087993.0,
      "step": 900
    },
    {
      "entropy": 0.31203486569225786,
      "epoch": 1.4074074074074074,
      "grad_norm": 0.25155216455459595,
      "learning_rate": 5.7489771210944564e-05,
      "loss": 0.3129,
      "mean_token_accuracy": 0.932486515045166,
      "num_tokens": 6425349.0,
      "step": 950
    },
    {
      "entropy": 0.31312834514304994,
      "epoch": 1.4814814814814814,
      "grad_norm": 0.2299806922674179,
      "learning_rate": 5.351619098663021e-05,
      "loss": 0.3139,
      "mean_token_accuracy": 0.9324578355252743,
      "num_tokens": 6762060.0,
      "step": 1000
    },
    {
      "entropy": 0.31229398518800733,
      "epoch": 1.5555555555555556,
      "grad_norm": 0.21491679549217224,
      "learning_rate": 4.952013068883795e-05,
      "loss": 0.3123,
      "mean_token_accuracy": 0.9330742400884628,
      "num_tokens": 7099580.0,
      "step": 1050
    },
    {
      "entropy": 0.3089119081199169,
      "epoch": 1.6296296296296298,
      "grad_norm": 0.20928895473480225,
      "learning_rate": 4.5527138340828776e-05,
      "loss": 0.3083,
      "mean_token_accuracy": 0.9346707260608673,
      "num_tokens": 7441088.0,
      "step": 1100
    },
    {
      "entropy": 0.3122195054590702,
      "epoch": 1.7037037037037037,
      "grad_norm": 0.1754506677389145,
      "learning_rate": 4.156274235153189e-05,
      "loss": 0.3128,
      "mean_token_accuracy": 0.9337523929774761,
      "num_tokens": 7781339.0,
      "step": 1150
    },
    {
      "entropy": 0.3072645129263401,
      "epoch": 1.7777777777777777,
      "grad_norm": 0.19373080134391785,
      "learning_rate": 3.765228830469794e-05,
      "loss": 0.3083,
      "mean_token_accuracy": 0.9344205930829048,
      "num_tokens": 8118559.0,
      "step": 1200
    },
    {
      "entropy": 0.30596201956272123,
      "epoch": 1.8518518518518519,
      "grad_norm": 0.21876507997512817,
      "learning_rate": 3.3820776916908857e-05,
      "loss": 0.3065,
      "mean_token_accuracy": 0.935108490884304,
      "num_tokens": 8458822.0,
      "step": 1250
    },
    {
      "entropy": 0.3032321432232857,
      "epoch": 1.925925925925926,
      "grad_norm": 0.16569602489471436,
      "learning_rate": 3.0092704200428058e-05,
      "loss": 0.304,
      "mean_token_accuracy": 0.9358008193969727,
      "num_tokens": 8797573.0,
      "step": 1300
    },
    {
      "entropy": 0.3116097005084157,
      "epoch": 2.0,
      "grad_norm": 0.23642410337924957,
      "learning_rate": 2.649190485277792e-05,
      "loss": 0.312,
      "mean_token_accuracy": 0.9340617834031582,
      "num_tokens": 9138916.0,
      "step": 1350
    },
    {
      "entropy": 0.2998246630653739,
      "epoch": 2.074074074074074,
      "grad_norm": 0.4210748076438904,
      "learning_rate": 2.3041399874302905e-05,
      "loss": 0.3015,
      "mean_token_accuracy": 0.9369703502953053,
      "num_tokens": 9476346.0,
      "step": 1400
    },
    {
      "entropy": 0.29883858568966387,
      "epoch": 2.148148148148148,
      "grad_norm": 0.2179645299911499,
      "learning_rate": 1.976324938794482e-05,
      "loss": 0.2995,
      "mean_token_accuracy": 0.9364559413492679,
      "num_tokens": 9813810.0,
      "step": 1450
    },
    {
      "entropy": 0.30353500295430424,
      "epoch": 2.2222222222222223,
      "grad_norm": 0.15429526567459106,
      "learning_rate": 1.667841160219835e-05,
      "loss": 0.304,
      "mean_token_accuracy": 0.9365751321613789,
      "num_tokens": 10154937.0,
      "step": 1500
    },
    {
      "entropy": 0.3042653623223305,
      "epoch": 2.2962962962962963,
      "grad_norm": 0.42488107085227966,
      "learning_rate": 1.3806608818939203e-05,
      "loss": 0.3057,
      "mean_token_accuracy": 0.9359618930518627,
      "num_tokens": 10494104.0,
      "step": 1550
    },
    {
      "entropy": 0.3081327396631241,
      "epoch": 2.3703703703703702,
      "grad_norm": 0.2351272851228714,
      "learning_rate": 1.1166201342777438e-05,
      "loss": 0.3064,
      "mean_token_accuracy": 0.9357219311594963,
      "num_tokens": 10834738.0,
      "step": 1600
    },
    {
      "entropy": 0.29908867001533507,
      "epoch": 2.4444444444444446,
      "grad_norm": 0.18584305047988892,
      "learning_rate": 8.774070098071668e-06,
      "loss": 0.3003,
      "mean_token_accuracy": 0.9364377255737781,
      "num_tokens": 11170346.0,
      "step": 1650
    },
    {
      "entropy": 0.30048872336745264,
      "epoch": 2.5185185185185186,
      "grad_norm": 0.17059944570064545,
      "learning_rate": 6.645508704069003e-06,
      "loss": 0.2998,
      "mean_token_accuracy": 0.9365158031880856,
      "num_tokens": 11508330.0,
      "step": 1700
    },
    {
      "entropy": 0.3041739664040506,
      "epoch": 2.5925925925925926,
      "grad_norm": 0.16423071920871735,
      "learning_rate": 4.794125698167262e-06,
      "loss": 0.3058,
      "mean_token_accuracy": 0.9364174953103066,
      "num_tokens": 11850254.0,
      "step": 1750
    },
    {
      "entropy": 0.2920705447718501,
      "epoch": 2.6666666666666665,
      "grad_norm": 0.19584065675735474,
      "learning_rate": 3.231757532415458e-06,
      "loss": 0.2943,
      "mean_token_accuracy": 0.9384492780268192,
      "num_tokens": 12185740.0,
      "step": 1800
    },
    {
      "entropy": 0.30169622793793677,
      "epoch": 2.7407407407407405,
      "grad_norm": 0.37878257036209106,
      "learning_rate": 1.9683928994924385e-06,
      "loss": 0.3011,
      "mean_token_accuracy": 0.9368291793763638,
      "num_tokens": 12525018.0,
      "step": 1850
    },
    {
      "entropy": 0.29348658775910735,
      "epoch": 2.814814814814815,
      "grad_norm": 0.19502298533916473,
      "learning_rate": 1.0121088719706296e-06,
      "loss": 0.2948,
      "mean_token_accuracy": 0.9380346268415451,
      "num_tokens": 12860441.0,
      "step": 1900
    },
    {
      "entropy": 0.3033849265426397,
      "epoch": 2.888888888888889,
      "grad_norm": 0.16882896423339844,
      "learning_rate": 3.6901926314575894e-07,
      "loss": 0.3039,
      "mean_token_accuracy": 0.936366637647152,
      "num_tokens": 13198992.0,
      "step": 1950
    },
    {
      "entropy": 0.29646733120083807,
      "epoch": 2.962962962962963,
      "grad_norm": 0.1928730458021164,
      "learning_rate": 4.323553957759629e-08,
      "loss": 0.2962,
      "mean_token_accuracy": 0.9374634748697281,
      "num_tokens": 13538141.0,
      "step": 2000
    }
  ],
  "logging_steps": 50,
  "max_steps": 2025,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.6552428189358285e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}