{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0008,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.7336,
      "step": 1
    },
    {
      "epoch": 1.0,
      "grad_norm": 3.6438040733337402,
      "learning_rate": 3.344e-05,
      "loss": 0.2985,
      "step": 1250
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.0047619342803955,
      "learning_rate": 1.6773333333333334e-05,
      "loss": 0.1895,
      "step": 2500
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0105699300765991,
      "learning_rate": 1.0666666666666668e-07,
      "loss": 0.1411,
      "step": 3750
    },
    {
      "epoch": 3.0,
      "step": 3750,
      "total_flos": 1.537752858624e+16,
      "train_loss": 0.20980143216451008,
      "train_runtime": 1457.052,
      "train_samples_per_second": 10.295,
      "train_steps_per_second": 2.574
    }
  ],
  "logging_steps": 500,
  "max_steps": 3750,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.537752858624e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}