{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.861635220125786,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5951612903225807,
      "eval_loss": 0.20621494948863983,
      "eval_runtime": 5.3904,
      "eval_samples_per_second": 575.093,
      "eval_steps_per_second": 12.058,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5390827059745789,
      "learning_rate": 1.685534591194969e-05,
      "loss": 0.3328,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8293548387096774,
      "eval_loss": 0.09735047072172165,
      "eval_runtime": 5.4548,
      "eval_samples_per_second": 568.302,
      "eval_steps_per_second": 11.916,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8848387096774194,
      "eval_loss": 0.06360608339309692,
      "eval_runtime": 5.474,
      "eval_samples_per_second": 566.318,
      "eval_steps_per_second": 11.874,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.4534893333911896,
      "learning_rate": 1.371069182389937e-05,
      "loss": 0.1133,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9019354838709678,
      "eval_loss": 0.04831857979297638,
      "eval_runtime": 5.4405,
      "eval_samples_per_second": 569.796,
      "eval_steps_per_second": 11.947,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.3067812919616699,
      "learning_rate": 1.0566037735849058e-05,
      "loss": 0.0707,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9109677419354839,
      "eval_loss": 0.03909005969762802,
      "eval_runtime": 6.0553,
      "eval_samples_per_second": 511.948,
      "eval_steps_per_second": 10.734,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9158064516129032,
      "eval_loss": 0.0339382030069828,
      "eval_runtime": 5.293,
      "eval_samples_per_second": 585.678,
      "eval_steps_per_second": 12.28,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.2602050304412842,
      "learning_rate": 7.421383647798742e-06,
      "loss": 0.0546,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9245161290322581,
      "eval_loss": 0.031150249764323235,
      "eval_runtime": 5.5375,
      "eval_samples_per_second": 559.822,
      "eval_steps_per_second": 11.738,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.2825825810432434,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0471,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 3180,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 756659102663436.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7668036748328225,
    "num_train_epochs": 10,
    "temperature": 8
  }
}