last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9404388714733543,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.31,
      "learning_rate": 0.0002,
      "loss": 0.7896,
      "step": 100
    },
    {
      "epoch": 0.31,
      "eval_runtime": 511.2401,
      "eval_samples_per_second": 1.76,
      "eval_steps_per_second": 0.221,
      "step": 100
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.6216,
      "step": 200
    },
    {
      "epoch": 0.63,
      "eval_runtime": 512.8833,
      "eval_samples_per_second": 1.755,
      "eval_steps_per_second": 0.22,
      "step": 200
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0002,
      "loss": 0.5863,
      "step": 300
    },
    {
      "epoch": 0.94,
      "eval_runtime": 513.6336,
      "eval_samples_per_second": 1.752,
      "eval_steps_per_second": 0.22,
      "step": 300
    }
  ],
  "logging_steps": 100,
  "max_steps": 957,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 4.98303246532608e+16,
  "trial_name": null,
  "trial_params": null
}