trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 468,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.6915,
      "step": 156
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6606498194945848,
      "eval_loss": 0.6491293907165527,
      "eval_runtime": 2.8403,
      "eval_samples_per_second": 97.525,
      "eval_steps_per_second": 12.323,
      "step": 156
    },
    {
      "epoch": 2.0,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.55,
      "step": 312
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6570397111913358,
      "eval_loss": 0.6737464666366577,
      "eval_runtime": 2.8421,
      "eval_samples_per_second": 97.462,
      "eval_steps_per_second": 12.315,
      "step": 312
    },
    {
      "epoch": 3.0,
      "learning_rate": 0.0,
      "loss": 0.3955,
      "step": 468
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6714801444043321,
      "eval_loss": 0.726006031036377,
      "eval_runtime": 2.8441,
      "eval_samples_per_second": 97.396,
      "eval_steps_per_second": 12.306,
      "step": 468
    },
    {
      "epoch": 3.0,
      "step": 468,
      "total_flos": 1965439583539200.0,
      "train_loss": 0.5456338621612288,
      "train_runtime": 291.3759,
      "train_samples_per_second": 25.637,
      "train_steps_per_second": 1.606
    }
  ],
  "max_steps": 468,
  "num_train_epochs": 3,
  "total_flos": 1965439583539200.0,
  "trial_name": null,
  "trial_params": null
}