run-1/checkpoint-536/trainer_state.json

{
  "best_metric": 0.520875943143754,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-1/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": 0.5111436247825623,
      "eval_matthews_correlation": 0.3704165163359662,
      "eval_runtime": 0.7196,
      "eval_samples_per_second": 1449.402,
      "eval_steps_per_second": 91.717,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.458956778049469,
      "eval_matthews_correlation": 0.5008706778439593,
      "eval_runtime": 0.7749,
      "eval_samples_per_second": 1345.924,
      "eval_steps_per_second": 85.169,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4913136959075928,
      "eval_matthews_correlation": 0.5108951669698317,
      "eval_runtime": 0.9439,
      "eval_samples_per_second": 1104.949,
      "eval_steps_per_second": 69.92,
      "step": 402
    },
    {
      "epoch": 3.73,
      "grad_norm": 8.718708038330078,
      "learning_rate": 4.5017309553431485e-06,
      "loss": 0.3868,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.5247140526771545,
      "eval_matthews_correlation": 0.520875943143754,
      "eval_runtime": 0.7379,
      "eval_samples_per_second": 1413.511,
      "eval_steps_per_second": 89.446,
      "step": 536
    }
  ],
  "logging_steps": 500,
  "max_steps": 670,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 210564174474576.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.7742116118117113e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 64,
    "seed": 5
  }
}