run-5/checkpoint-162/trainer_state.json

{
  "best_metric": 0.3215566034550075,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-5/checkpoint-162",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 162,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.575078010559082,
      "eval_matthews_correlation": 0.06354272869611655,
      "eval_runtime": 9.5691,
      "eval_samples_per_second": 108.997,
      "eval_steps_per_second": 6.897,
      "step": 54
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8100215196609497,
      "eval_matthews_correlation": 0.24720569658796848,
      "eval_runtime": 8.7353,
      "eval_samples_per_second": 119.401,
      "eval_steps_per_second": 7.556,
      "step": 108
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.1307350397109985,
      "eval_matthews_correlation": 0.3215566034550075,
      "eval_runtime": 9.5213,
      "eval_samples_per_second": 109.544,
      "eval_steps_per_second": 6.932,
      "step": 162
    }
  ],
  "logging_steps": 500,
  "max_steps": 216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.264714442689283e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 16,
    "seed": 3
  }
}