run-4/checkpoint-536/trainer_state.json

{
  "best_metric": 0.47854164992705517,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/checkpoint-536",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 536,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5019019842147827,
      "eval_matthews_correlation": 0.40644314565079237,
      "eval_runtime": 27.9447,
      "eval_samples_per_second": 37.324,
      "eval_steps_per_second": 2.362,
      "step": 268
    },
    {
      "epoch": 1.8656716417910446,
      "grad_norm": 6.935309410095215,
      "learning_rate": 8.001458853591762e-06,
      "loss": 0.485,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4858323633670807,
      "eval_matthews_correlation": 0.47854164992705517,
      "eval_runtime": 28.7617,
      "eval_samples_per_second": 36.264,
      "eval_steps_per_second": 2.295,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 1340,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 95845337435376.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.2764231980729716e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 32,
    "seed": 24
  }
}