run-0/checkpoint-633/trainer_state.json

{
  "best_metric": 0.9094036697247706,
  "best_model_checkpoint": "bert-base-uncased-finetuned-sst2/run-0/checkpoint-422",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 633,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9071100917431193,
      "eval_loss": 0.2743072807788849,
      "eval_runtime": 2.2589,
      "eval_samples_per_second": 386.028,
      "eval_steps_per_second": 24.348,
      "step": 211
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9094036697247706,
      "eval_loss": 0.2763120234012604,
      "eval_runtime": 2.3197,
      "eval_samples_per_second": 375.903,
      "eval_steps_per_second": 23.709,
      "step": 422
    },
    {
      "epoch": 2.37,
      "grad_norm": 7.35267448425293,
      "learning_rate": 1.4004341381907418e-05,
      "loss": 0.2085,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9071100917431193,
      "eval_loss": 0.39031100273132324,
      "eval_runtime": 2.3726,
      "eval_samples_per_second": 367.532,
      "eval_steps_per_second": 23.181,
      "step": 633
    }
  ],
  "logging_steps": 500,
  "max_steps": 633,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 519273306636840.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 6.665224131389019e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 32,
    "seed": 10
  }
}