run-0/checkpoint-321/trainer_state.json

{
  "best_metric": 0.32753989938135253,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-321",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 321,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.583710253238678,
      "eval_matthews_correlation": 0.0,
      "eval_runtime": 1.0039,
      "eval_samples_per_second": 1038.933,
      "eval_steps_per_second": 65.743,
      "step": 107
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5807983875274658,
      "eval_matthews_correlation": 0.2756601711602064,
      "eval_runtime": 1.4593,
      "eval_samples_per_second": 714.737,
      "eval_steps_per_second": 45.228,
      "step": 214
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5898085832595825,
      "eval_matthews_correlation": 0.32753989938135253,
      "eval_runtime": 1.0391,
      "eval_samples_per_second": 1003.775,
      "eval_steps_per_second": 63.518,
      "step": 321
    }
  ],
  "logging_steps": 500,
  "max_steps": 428,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.3194120931369337e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 8,
    "seed": 18
  }
}