trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "global_step": 690,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8431372549019608,
      "eval_combined_score": 0.8656245715069245,
      "eval_f1": 0.8881118881118882,
      "eval_loss": 0.3667730391025543,
      "eval_runtime": 0.621,
      "eval_samples_per_second": 657.045,
      "eval_steps_per_second": 41.87,
      "step": 230
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8578431372549019,
      "eval_combined_score": 0.8797690262545697,
      "eval_f1": 0.9016949152542373,
      "eval_loss": 0.375055193901062,
      "eval_runtime": 0.6266,
      "eval_samples_per_second": 651.108,
      "eval_steps_per_second": 41.492,
      "step": 460
    },
    {
      "epoch": 2.17,
      "learning_rate": 1.3768115942028985e-05,
      "loss": 0.4264,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8602941176470589,
      "eval_combined_score": 0.8817599620493359,
      "eval_f1": 0.9032258064516129,
      "eval_loss": 0.5152415037155151,
      "eval_runtime": 0.5833,
      "eval_samples_per_second": 699.514,
      "eval_steps_per_second": 44.577,
      "step": 690
    },
    {
      "epoch": 3.0,
      "step": 690,
      "total_flos": 723818515529728.0,
      "train_loss": 0.36320947287739186,
      "train_runtime": 105.4434,
      "train_samples_per_second": 104.359,
      "train_steps_per_second": 6.544
    }
  ],
  "max_steps": 690,
  "num_train_epochs": 3,
  "total_flos": 723818515529728.0,
  "trial_name": null,
  "trial_params": null
}