run-0/checkpoint-536/trainer_state.json

{
  "best_metric": 0.44944855113368687,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-0/checkpoint-536",
  "epoch": 2.0,
  "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": 0.5017370581626892,
      "eval_matthews_correlation": 0.4386209168112411,
      "eval_runtime": 0.9474,
      "eval_samples_per_second": 1100.94,
      "eval_steps_per_second": 69.666,
      "step": 268
    },
    {
      "epoch": 1.87,
      "grad_norm": 9.115137100219727,
      "learning_rate": 1.255312161688088e-06,
      "loss": 0.4596,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.506603479385376,
      "eval_matthews_correlation": 0.44944855113368687,
      "eval_runtime": 0.7486,
      "eval_samples_per_second": 1393.184,
      "eval_steps_per_second": 88.159,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 536,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 95049498141888.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.8690203296244868e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 32,
    "seed": 15
  }
}