run-4/checkpoint-804/trainer_state.json

{
  "best_metric": 0.5070207143299597,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-4/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": 0.478412002325058,
      "eval_matthews_correlation": 0.4981947529906373,
      "eval_runtime": 0.7412,
      "eval_samples_per_second": 1407.09,
      "eval_steps_per_second": 89.039,
      "step": 268
    },
    {
      "epoch": 1.87,
      "learning_rate": 8.253927295488848e-06,
      "loss": 0.4708,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.45441004633903503,
      "eval_matthews_correlation": 0.5010631117439056,
      "eval_runtime": 0.8128,
      "eval_samples_per_second": 1283.245,
      "eval_steps_per_second": 81.202,
      "step": 536
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5127615332603455,
      "eval_matthews_correlation": 0.5070207143299597,
      "eval_runtime": 0.7719,
      "eval_samples_per_second": 1351.188,
      "eval_steps_per_second": 85.502,
      "step": 804
    }
  ],
  "logging_steps": 500,
  "max_steps": 1072,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 95450522493288.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.546889870762945e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 32,
    "seed": 6
  }
}