run-4/checkpoint-402/trainer_state.json

{
  "best_metric": 0.4818349198265336,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-268",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 402,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5429930686950684,
      "eval_matthews_correlation": 0.3489031724545359,
      "eval_runtime": 0.7377,
      "eval_samples_per_second": 1413.869,
      "eval_steps_per_second": 89.468,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.48521509766578674,
      "eval_matthews_correlation": 0.4818349198265336,
      "eval_runtime": 0.807,
      "eval_samples_per_second": 1292.399,
      "eval_steps_per_second": 81.782,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5382273197174072,
      "eval_matthews_correlation": 0.46768485679358657,
      "eval_runtime": 0.7838,
      "eval_samples_per_second": 1330.662,
      "eval_steps_per_second": 84.203,
      "step": 402
    }
  ],
  "logging_steps": 500,
  "max_steps": 402,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.829056514710492e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 64,
    "seed": 21
  }
}