run-3/checkpoint-804/trainer_state.json

{
  "best_metric": 0.5011490576561073,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-3/checkpoint-804",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 804,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.49783048033714294,
      "eval_matthews_correlation": 0.4111094454070522,
      "eval_runtime": 0.7234,
      "eval_samples_per_second": 1441.791,
      "eval_steps_per_second": 91.235,
      "step": 268
    },
    {
      "epoch": 1.87,
      "learning_rate": 7.123299677400847e-06,
      "loss": 0.4894,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5266438126564026,
      "eval_matthews_correlation": 0.44722803398940786,
      "eval_runtime": 0.7661,
      "eval_samples_per_second": 1361.434,
      "eval_steps_per_second": 86.15,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4729432761669159,
      "eval_matthews_correlation": 0.5011490576561073,
      "eval_runtime": 0.7179,
      "eval_samples_per_second": 1452.841,
      "eval_steps_per_second": 91.934,
      "step": 804
    }
  ],
  "logging_steps": 500,
  "max_steps": 1340,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 95454144648720.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.1363359009187065e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 32,
    "seed": 20
  }
}