run-0/checkpoint-536/trainer_state.json

{
  "best_metric": 0.44158634577419253,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-536",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 536,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.5337199568748474,
      "eval_matthews_correlation": 0.37010827120601614,
      "eval_runtime": 0.5268,
      "eval_samples_per_second": 1979.744,
      "eval_steps_per_second": 125.276,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.516914427280426,
      "eval_matthews_correlation": 0.40941556921224126,
      "eval_runtime": 0.3522,
      "eval_samples_per_second": 2961.329,
      "eval_steps_per_second": 187.39,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5180604457855225,
      "eval_matthews_correlation": 0.42682064874774595,
      "eval_runtime": 0.365,
      "eval_samples_per_second": 2857.523,
      "eval_steps_per_second": 180.821,
      "step": 402
    },
    {
      "epoch": 3.7313432835820897,
      "grad_norm": 5.698959827423096,
      "learning_rate": 5.227469807004321e-07,
      "loss": 0.4746,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5101630687713623,
      "eval_matthews_correlation": 0.44158634577419253,
      "eval_runtime": 0.3499,
      "eval_samples_per_second": 2980.994,
      "eval_steps_per_second": 188.634,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 536,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 209831981626536.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 7.783121712650878e-06,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 64,
    "seed": 12
  }
}