trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9625,
      "eval_loss": 0.1814567595720291,
      "eval_runtime": 0.3208,
      "eval_samples_per_second": 249.366,
      "eval_steps_per_second": 31.171,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.95,
      "eval_loss": 0.19135093688964844,
      "eval_runtime": 0.327,
      "eval_samples_per_second": 244.613,
      "eval_steps_per_second": 30.577,
      "step": 80
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.975,
      "eval_loss": 0.12160463631153107,
      "eval_runtime": 0.3228,
      "eval_samples_per_second": 247.815,
      "eval_steps_per_second": 30.977,
      "step": 120
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9625,
      "eval_loss": 0.17466306686401367,
      "eval_runtime": 0.3213,
      "eval_samples_per_second": 248.983,
      "eval_steps_per_second": 31.123,
      "step": 160
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.975,
      "eval_loss": 0.13131237030029297,
      "eval_runtime": 0.3307,
      "eval_samples_per_second": 241.89,
      "eval_steps_per_second": 30.236,
      "step": 200
    },
    {
      "epoch": 5.0,
      "step": 200,
      "total_flos": 26349092044800.0,
      "train_loss": 0.12233135223388672,
      "train_runtime": 21.6815,
      "train_samples_per_second": 73.796,
      "train_steps_per_second": 9.224
    }
  ],
  "max_steps": 200,
  "num_train_epochs": 5,
  "total_flos": 26349092044800.0,
  "trial_name": null,
  "trial_params": null
}