run-3/checkpoint-2526/trainer_state.json

{
  "best_metric": 0.5321100917431193,
  "best_model_checkpoint": "Tech-oriented/best_model_bert_uncasedbert-base-uncased-finetuned-sst2/run-3/checkpoint-2526",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2526,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4919724770642202,
      "eval_loss": 0.7082622647285461,
      "eval_runtime": 2.331,
      "eval_samples_per_second": 374.089,
      "eval_steps_per_second": 23.595,
      "step": 842
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.518348623853211,
      "eval_loss": 0.6900007128715515,
      "eval_runtime": 2.3364,
      "eval_samples_per_second": 373.226,
      "eval_steps_per_second": 23.541,
      "step": 1684
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5321100917431193,
      "eval_loss": 0.6852103471755981,
      "eval_runtime": 2.3462,
      "eval_samples_per_second": 371.672,
      "eval_steps_per_second": 23.443,
      "step": 2526
    }
  ],
  "logging_steps": 4209,
  "max_steps": 2526,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 3.843391089618415e-07,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 8,
    "seed": 19,
    "weight_decay": 0.4905221565015579
  }
}