{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 1875,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.9283596873283386,
      "learning_rate": 3e-05,
      "loss": 1.963,
      "step": 100
    },
    {
      "epoch": 0.9973333333333333,
      "eval_accuracy": 0.5695384615384615,
      "eval_loss": 1.6438876390457153,
      "eval_runtime": 5.0534,
      "eval_samples_per_second": 98.943,
      "eval_steps_per_second": 12.467,
      "step": 187
    },
    {
      "epoch": 0.9973333333333333,
      "eval_exact_match": 16.2,
      "eval_f1": 25.888571428571442,
      "step": 187
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 1.0407154560089111,
      "learning_rate": 3e-05,
      "loss": 1.6668,
      "step": 200
    },
    {
      "epoch": 1.6,
      "grad_norm": 1.1281720399856567,
      "learning_rate": 3e-05,
      "loss": 1.6099,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5733333333333334,
      "eval_loss": 1.618270754814148,
      "eval_runtime": 5.1983,
      "eval_samples_per_second": 96.186,
      "eval_steps_per_second": 12.119,
      "step": 375
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 16.6,
      "eval_f1": 26.287619047619057,
      "step": 375
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 1.2508225440979004,
      "learning_rate": 3e-05,
      "loss": 1.5866,
      "step": 400
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 1.3197834491729736,
      "learning_rate": 3e-05,
      "loss": 1.524,
      "step": 500
    },
    {
      "epoch": 2.997333333333333,
      "eval_accuracy": 0.5743589743589743,
      "eval_loss": 1.6164497137069702,
      "eval_runtime": 5.0495,
      "eval_samples_per_second": 99.019,
      "eval_steps_per_second": 12.476,
      "step": 562
    },
    {
      "epoch": 2.997333333333333,
      "eval_exact_match": 17.8,
      "eval_f1": 27.35428571428571,
      "step": 562
    },
    {
      "epoch": 3.2,
      "grad_norm": 1.6320394277572632,
      "learning_rate": 3e-05,
      "loss": 1.4752,
      "step": 600
    },
    {
      "epoch": 3.7333333333333334,
      "grad_norm": 1.7101725339889526,
      "learning_rate": 3e-05,
      "loss": 1.3938,
      "step": 700
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5728717948717948,
      "eval_loss": 1.6376241445541382,
      "eval_runtime": 4.9662,
      "eval_samples_per_second": 100.68,
      "eval_steps_per_second": 12.686,
      "step": 750
    },
    {
      "epoch": 4.0,
      "eval_exact_match": 18.0,
      "eval_f1": 27.498571428571427,
      "step": 750
    },
    {
      "epoch": 4.266666666666667,
      "grad_norm": 2.1727664470672607,
      "learning_rate": 3e-05,
      "loss": 1.342,
      "step": 800
    },
    {
      "epoch": 4.8,
      "grad_norm": 2.341384172439575,
      "learning_rate": 3e-05,
      "loss": 1.2685,
      "step": 900
    },
    {
      "epoch": 4.997333333333334,
      "eval_accuracy": 0.5699487179487179,
      "eval_loss": 1.684618353843689,
      "eval_runtime": 5.5457,
      "eval_samples_per_second": 90.16,
      "eval_steps_per_second": 11.36,
      "step": 937
    },
    {
      "epoch": 4.997333333333334,
      "eval_exact_match": 19.2,
      "eval_f1": 28.190952380952375,
      "step": 937
    },
    {
      "epoch": 5.333333333333333,
      "grad_norm": 2.5624172687530518,
      "learning_rate": 3e-05,
      "loss": 1.1847,
      "step": 1000
    },
    {
      "epoch": 5.866666666666667,
      "grad_norm": 2.8159019947052,
      "learning_rate": 3e-05,
      "loss": 1.1591,
      "step": 1100
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5672820512820513,
      "eval_loss": 1.7546838521957397,
      "eval_runtime": 5.0186,
      "eval_samples_per_second": 99.629,
      "eval_steps_per_second": 12.553,
      "step": 1125
    },
    {
      "epoch": 6.0,
      "eval_exact_match": 16.6,
      "eval_f1": 25.054761904761914,
      "step": 1125
    },
    {
      "epoch": 6.4,
      "grad_norm": 3.056166410446167,
      "learning_rate": 3e-05,
      "loss": 1.0486,
      "step": 1200
    },
    {
      "epoch": 6.933333333333334,
      "grad_norm": 2.873281240463257,
      "learning_rate": 3e-05,
      "loss": 1.0444,
      "step": 1300
    },
    {
      "epoch": 6.997333333333334,
      "eval_accuracy": 0.5643076923076923,
      "eval_loss": 1.8394943475723267,
      "eval_runtime": 5.5931,
      "eval_samples_per_second": 89.396,
      "eval_steps_per_second": 11.264,
      "step": 1312
    },
    {
      "epoch": 6.997333333333334,
      "eval_exact_match": 15.8,
      "eval_f1": 23.925714285714296,
      "step": 1312
    },
    {
      "epoch": 7.466666666666667,
      "grad_norm": 3.069322109222412,
      "learning_rate": 3e-05,
      "loss": 0.9374,
      "step": 1400
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.2732129096984863,
      "learning_rate": 3e-05,
      "loss": 0.9535,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5612820512820513,
      "eval_loss": 1.9008216857910156,
      "eval_runtime": 5.1357,
      "eval_samples_per_second": 97.358,
      "eval_steps_per_second": 12.267,
      "step": 1500
    },
    {
      "epoch": 8.0,
      "eval_exact_match": 15.2,
      "eval_f1": 23.12063492063493,
      "step": 1500
    },
    {
      "epoch": 8.533333333333333,
      "grad_norm": 3.382610321044922,
      "learning_rate": 3e-05,
      "loss": 0.8235,
      "step": 1600
    },
    {
      "epoch": 8.997333333333334,
      "eval_accuracy": 0.5592307692307692,
      "eval_loss": 2.026776075363159,
      "eval_runtime": 5.5056,
      "eval_samples_per_second": 90.817,
      "eval_steps_per_second": 11.443,
      "step": 1687
    },
    {
      "epoch": 8.997333333333334,
      "eval_exact_match": 13.8,
      "eval_f1": 21.42285714285715,
      "step": 1687
    },
    {
      "epoch": 9.066666666666666,
      "grad_norm": 4.363819122314453,
      "learning_rate": 3e-05,
      "loss": 0.8547,
      "step": 1700
    },
    {
      "epoch": 9.6,
      "grad_norm": 3.685805320739746,
      "learning_rate": 3e-05,
      "loss": 0.7635,
      "step": 1800
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5567692307692308,
      "eval_loss": 2.0937435626983643,
      "eval_runtime": 5.235,
      "eval_samples_per_second": 95.512,
      "eval_steps_per_second": 12.034,
      "step": 1875
    },
    {
      "epoch": 10.0,
      "eval_exact_match": 15.2,
      "eval_f1": 22.031904761904762,
      "step": 1875
    }
  ],
  "logging_steps": 100,
  "max_steps": 9350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "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": 3.4335509031419904e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}