run-3/checkpoint-268/trainer_state.json

{
  "best_metric": 0.4123772629669354,
  "best_model_checkpoint": "hw-01-finetuned-cola/run-3/checkpoint-268",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 268,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5247284173965454,
      "eval_matthews_correlation": 0.3711969872541337,
      "eval_runtime": 0.7818,
      "eval_samples_per_second": 1334.046,
      "eval_steps_per_second": 84.417,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5192302465438843,
      "eval_matthews_correlation": 0.4123772629669354,
      "eval_runtime": 1.3852,
      "eval_samples_per_second": 752.98,
      "eval_steps_per_second": 47.648,
      "step": 268
    }
  ],
  "logging_steps": 500,
  "max_steps": 268,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.099451070388589e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 64,
    "seed": 16
  }
}