run-3/checkpoint-627/trainer_state.json

{
  "best_metric": 0.9040742959856202,
  "best_model_checkpoint": "bert-base-uncased-finetuned-spam-real/run-3/checkpoint-627",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 627,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8203115638106651,
      "eval_loss": 0.5042904615402222,
      "eval_runtime": 129.4295,
      "eval_samples_per_second": 128.95,
      "eval_steps_per_second": 16.125,
      "step": 209
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8947872977831036,
      "eval_loss": 0.3739599585533142,
      "eval_runtime": 129.4507,
      "eval_samples_per_second": 128.929,
      "eval_steps_per_second": 16.122,
      "step": 418
    },
    {
      "epoch": 2.39,
      "grad_norm": 6.7307233810424805,
      "learning_rate": 2.3798610323153638e-07,
      "loss": 0.5071,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9040742959856202,
      "eval_loss": 0.34588372707366943,
      "eval_runtime": 129.5008,
      "eval_samples_per_second": 128.879,
      "eval_steps_per_second": 16.116,
      "step": 627
    }
  ],
  "logging_steps": 500,
  "max_steps": 627,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 2034221541187080.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.1749392655604196e-06,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 8,
    "seed": 31
  }
}