trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5005078204346943,
  "eval_steps": 500,
  "global_step": 154,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 0.0004185667752442997,
      "loss": 1.6691,
      "step": 50
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00033713355048859935,
      "loss": 1.1667,
      "step": 100
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.000255700325732899,
      "loss": 1.1266,
      "step": 150
    }
  ],
  "logging_steps": 50,
  "max_steps": 307,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 154,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}