run-1/checkpoint-172/trainer_state.json

{
  "best_metric": 0.5,
  "best_model_checkpoint": "Bert_tinybert-distilled/run-1/checkpoint-172",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 172,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.10513057559728622,
      "learning_rate": 0.0004697676642472769,
      "loss": 0.0495,
      "step": 172
    },
    {
      "epoch": 1.0,
      "eval_AUC": 0.8095893333333333,
      "eval_BAC": 0.5,
      "eval_MCC": 0.0,
      "eval_PR_AUC": 0.7634102804324954,
      "eval_accuracy": 0.5,
      "eval_loss": 0.04859978333115578,
      "eval_runtime": 6.131,
      "eval_samples_per_second": 244.658,
      "eval_steps_per_second": 7.666,
      "step": 172
    }
  ],
  "logging_steps": 500,
  "max_steps": 688,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 450377730000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.06190377563487581,
    "learning_rate": 0.0006263568856630359,
    "num_train_epochs": 4,
    "temperature": 15
  }
}