trainer_state.json

{
  "best_metric": 0.5880094937717885,
  "best_model_checkpoint": "./fine-tune/bert-base-uncased/cola/checkpoint-804",
  "epoch": 3.0,
  "global_step": 804,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.45977166295051575,
      "eval_matthews_correlation": 0.5134946392219878,
      "eval_runtime": 5.9561,
      "eval_samples_per_second": 175.115,
      "eval_steps_per_second": 21.994,
      "step": 268
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.5621890547263685e-06,
      "loss": 0.393,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4875448942184448,
      "eval_matthews_correlation": 0.5573424050983508,
      "eval_runtime": 2.8786,
      "eval_samples_per_second": 362.324,
      "eval_steps_per_second": 45.508,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5405848622322083,
      "eval_matthews_correlation": 0.5880094937717885,
      "eval_runtime": 5.5175,
      "eval_samples_per_second": 189.036,
      "eval_steps_per_second": 23.743,
      "step": 804
    },
    {
      "epoch": 3.0,
      "step": 804,
      "total_flos": 1687396975787520.0,
      "train_loss": 0.31582939565478274,
      "train_runtime": 345.3367,
      "train_samples_per_second": 74.284,
      "train_steps_per_second": 2.328
    }
  ],
  "max_steps": 804,
  "num_train_epochs": 3,
  "total_flos": 1687396975787520.0,
  "trial_name": null,
  "trial_params": null
}