{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1332,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "entropy": 2.4947163558006284,
      "epoch": 0.07507507507507508,
      "grad_norm": 2.2692031860351562,
      "learning_rate": 1.99976055134129e-05,
      "loss": 9.666,
      "mean_token_accuracy": 0.3714864462614059,
      "num_tokens": 175430.0,
      "step": 50
    },
    {
      "entropy": 4.957374696731567,
      "epoch": 0.15015015015015015,
      "grad_norm": 0.4004897177219391,
      "learning_rate": 1.98972684724999e-05,
      "loss": 3.687,
      "mean_token_accuracy": 0.5028938430547715,
      "num_tokens": 350404.0,
      "step": 100
    },
    {
      "entropy": 4.707881135940552,
      "epoch": 0.22522522522522523,
      "grad_norm": 0.4421725571155548,
      "learning_rate": 1.9650816346189154e-05,
      "loss": 3.4827,
      "mean_token_accuracy": 0.5126789313554764,
      "num_tokens": 526514.0,
      "step": 150
    },
    {
      "entropy": 4.64449327468872,
      "epoch": 0.3003003003003003,
      "grad_norm": 0.4718657433986664,
      "learning_rate": 1.926188754982551e-05,
      "loss": 3.4128,
      "mean_token_accuracy": 0.5216181075572968,
      "num_tokens": 701236.0,
      "step": 200
    },
    {
      "entropy": 4.598890199661255,
      "epoch": 0.37537537537537535,
      "grad_norm": 0.4660639762878418,
      "learning_rate": 1.8736223906463698e-05,
      "loss": 3.3675,
      "mean_token_accuracy": 0.528342125415802,
      "num_tokens": 876499.0,
      "step": 250
    },
    {
      "entropy": 4.603140239715576,
      "epoch": 0.45045045045045046,
      "grad_norm": 0.47321391105651855,
      "learning_rate": 1.8081585879342008e-05,
      "loss": 3.3821,
      "mean_token_accuracy": 0.5266077440977096,
      "num_tokens": 1051864.0,
      "step": 300
    },
    {
      "entropy": 4.587698755264282,
      "epoch": 0.5255255255255256,
      "grad_norm": 0.5947113633155823,
      "learning_rate": 1.7307638002821942e-05,
      "loss": 3.3509,
      "mean_token_accuracy": 0.5320688891410827,
      "num_tokens": 1226112.0,
      "step": 350
    },
    {
      "entropy": 4.596508531570435,
      "epoch": 0.6006006006006006,
      "grad_norm": 0.5403749942779541,
      "learning_rate": 1.6425806203196734e-05,
      "loss": 3.3672,
      "mean_token_accuracy": 0.5314443480968475,
      "num_tokens": 1401199.0,
      "step": 400
    },
    {
      "entropy": 4.5638793849945065,
      "epoch": 0.6756756756756757,
      "grad_norm": 0.5813198685646057,
      "learning_rate": 1.544910911576629e-05,
      "loss": 3.3328,
      "mean_token_accuracy": 0.5359585750102996,
      "num_tokens": 1576517.0,
      "step": 450
    },
    {
      "entropy": 4.571099786758423,
      "epoch": 0.7507507507507507,
      "grad_norm": 0.5679024457931519,
      "learning_rate": 1.4391965888473705e-05,
      "loss": 3.3438,
      "mean_token_accuracy": 0.5345500099658966,
      "num_tokens": 1751444.0,
      "step": 500
    },
    {
      "entropy": 4.5622615623474125,
      "epoch": 0.8258258258258259,
      "grad_norm": 0.6713469624519348,
      "learning_rate": 1.3269983309531584e-05,
      "loss": 3.3411,
      "mean_token_accuracy": 0.5363436281681061,
      "num_tokens": 1926907.0,
      "step": 550
    },
    {
      "entropy": 4.562127561569214,
      "epoch": 0.9009009009009009,
      "grad_norm": 0.6330661177635193,
      "learning_rate": 1.2099725401709685e-05,
      "loss": 3.3382,
      "mean_token_accuracy": 0.5368253135681152,
      "num_tokens": 2102132.0,
      "step": 600
    },
    {
      "entropy": 4.561514482498169,
      "epoch": 0.975975975975976,
      "grad_norm": 0.7525476813316345,
      "learning_rate": 1.0898468884803366e-05,
      "loss": 3.3411,
      "mean_token_accuracy": 0.537065578699112,
      "num_tokens": 2277537.0,
      "step": 650
    },
    {
      "entropy": 4.548706264495849,
      "epoch": 1.0510510510510511,
      "grad_norm": 0.6428335309028625,
      "learning_rate": 9.683948116432609e-06,
      "loss": 3.3205,
      "mean_token_accuracy": 0.5413484466075897,
      "num_tokens": 2449566.0,
      "step": 700
    },
    {
      "entropy": 4.5432649517059325,
      "epoch": 1.1261261261261262,
      "grad_norm": 0.6926820278167725,
      "learning_rate": 8.474093276654764e-06,
      "loss": 3.3025,
      "mean_token_accuracy": 0.5445451009273529,
      "num_tokens": 2623865.0,
      "step": 750
    },
    {
      "entropy": 4.537814807891846,
      "epoch": 1.2012012012012012,
      "grad_norm": 0.6473987698554993,
      "learning_rate": 7.286765661616761e-06,
      "loss": 3.3031,
      "mean_token_accuracy": 0.5444674134254456,
      "num_tokens": 2798793.0,
      "step": 800
    },
    {
      "entropy": 4.533882551193237,
      "epoch": 1.2762762762762763,
      "grad_norm": 0.7260850667953491,
      "learning_rate": 6.139493994152428e-06,
      "loss": 3.3086,
      "mean_token_accuracy": 0.5432299053668976,
      "num_tokens": 2975077.0,
      "step": 850
    },
    {
      "entropy": 4.509051198959351,
      "epoch": 1.3513513513513513,
      "grad_norm": 0.7649775743484497,
      "learning_rate": 5.0492156442170914e-06,
      "loss": 3.262,
      "mean_token_accuracy": 0.5496996784210205,
      "num_tokens": 3149867.0,
      "step": 900
    },
    {
      "entropy": 4.533229093551636,
      "epoch": 1.4264264264264264,
      "grad_norm": 0.7427539229393005,
      "learning_rate": 4.0320265795669815e-06,
      "loss": 3.2888,
      "mean_token_accuracy": 0.5470581102371216,
      "num_tokens": 3324230.0,
      "step": 950
    },
    {
      "entropy": 4.5299335289001466,
      "epoch": 1.5015015015015014,
      "grad_norm": 0.7631803154945374,
      "learning_rate": 3.1029437382047368e-06,
      "loss": 3.2959,
      "mean_token_accuracy": 0.5465954422950745,
      "num_tokens": 3499600.0,
      "step": 1000
    },
    {
      "entropy": 4.519344367980957,
      "epoch": 1.5765765765765765,
      "grad_norm": 0.7936219573020935,
      "learning_rate": 2.275683330727697e-06,
      "loss": 3.2927,
      "mean_token_accuracy": 0.5452944958209991,
      "num_tokens": 3676089.0,
      "step": 1050
    },
    {
      "entropy": 4.532338409423828,
      "epoch": 1.6516516516516515,
      "grad_norm": 0.735207200050354,
      "learning_rate": 1.562458345539739e-06,
      "loss": 3.3022,
      "mean_token_accuracy": 0.5458044326305389,
      "num_tokens": 3851878.0,
      "step": 1100
    },
    {
      "entropy": 4.519860811233521,
      "epoch": 1.7267267267267268,
      "grad_norm": 0.9440169930458069,
      "learning_rate": 9.737982463922102e-07,
      "loss": 3.2767,
      "mean_token_accuracy": 0.5487184035778045,
      "num_tokens": 4026785.0,
      "step": 1150
    },
    {
      "entropy": 4.5121307468414305,
      "epoch": 1.8018018018018018,
      "grad_norm": 0.8288739919662476,
      "learning_rate": 5.183935240903415e-07,
      "loss": 3.2591,
      "mean_token_accuracy": 0.5512541651725769,
      "num_tokens": 4200908.0,
      "step": 1200
    },
    {
      "entropy": 4.517954025268555,
      "epoch": 1.8768768768768769,
      "grad_norm": 0.7847370505332947,
      "learning_rate": 2.0296739727517335e-07,
      "loss": 3.2798,
      "mean_token_accuracy": 0.5484513938426971,
      "num_tokens": 4376428.0,
      "step": 1250
    },
    {
      "entropy": 4.50932373046875,
      "epoch": 1.951951951951952,
      "grad_norm": 0.7920858860015869,
      "learning_rate": 3.217655638451112e-08,
      "loss": 3.2683,
      "mean_token_accuracy": 0.5490582883358002,
      "num_tokens": 4551678.0,
      "step": 1300
    }
  ],
  "logging_steps": 50,
  "max_steps": 1332,
  "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": 2.767706644103712e+17,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": null
}