{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 4422,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33921302578018997,
      "grad_norm": 13.293481826782227,
      "learning_rate": 4.434644957033017e-05,
      "loss": 4.4002,
      "step": 500
    },
    {
      "epoch": 0.6784260515603799,
      "grad_norm": 12.114908218383789,
      "learning_rate": 3.869289914066034e-05,
      "loss": 2.923,
      "step": 1000
    },
    {
      "epoch": 1.01763907734057,
      "grad_norm": 13.19051742553711,
      "learning_rate": 3.30393487109905e-05,
      "loss": 2.4431,
      "step": 1500
    },
    {
      "epoch": 1.3568521031207599,
      "grad_norm": 11.903945922851562,
      "learning_rate": 2.7385798281320668e-05,
      "loss": 2.0948,
      "step": 2000
    },
    {
      "epoch": 1.6960651289009498,
      "grad_norm": 11.737950325012207,
      "learning_rate": 2.173224785165084e-05,
      "loss": 1.9493,
      "step": 2500
    },
    {
      "epoch": 2.03527815468114,
      "grad_norm": 11.720258712768555,
      "learning_rate": 1.6078697421981004e-05,
      "loss": 1.8166,
      "step": 3000
    },
    {
      "epoch": 2.37449118046133,
      "grad_norm": 11.776323318481445,
      "learning_rate": 1.0425146992311172e-05,
      "loss": 1.5034,
      "step": 3500
    },
    {
      "epoch": 2.7137042062415198,
      "grad_norm": 11.804101943969727,
      "learning_rate": 4.771596562641339e-06,
      "loss": 1.4416,
      "step": 4000
    },
    {
      "epoch": 3.0,
      "step": 4422,
      "total_flos": 3.021042087648e+17,
      "train_loss": 2.2322697978060693,
      "train_runtime": 273879.2077,
      "train_samples_per_second": 4.131,
      "train_steps_per_second": 0.016
    }
  ],
  "logging_steps": 500,
  "max_steps": 4422,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.021042087648e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}