run-4/checkpoint-344/trainer_state.json

{
  "best_metric": 0.7613333333333333,
  "best_model_checkpoint": "Bert_tinybert-distilled/run-4/checkpoint-344",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 344,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.065812110900879,
      "learning_rate": 8.050072712118835e-05,
      "loss": 0.5092,
      "step": 172
    },
    {
      "epoch": 1.0,
      "eval_AUC": 0.8145119999999999,
      "eval_BAC": 0.73,
      "eval_MCC": 0.4608302423279951,
      "eval_PR_AUC": 0.7678112276484446,
      "eval_accuracy": 0.73,
      "eval_loss": 0.4519721269607544,
      "eval_runtime": 6.3105,
      "eval_samples_per_second": 237.697,
      "eval_steps_per_second": 7.448,
      "step": 172
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.7611809968948364,
      "learning_rate": 6.900062324673287e-05,
      "loss": 0.4743,
      "step": 344
    },
    {
      "epoch": 2.0,
      "eval_AUC": 0.8335324444444443,
      "eval_BAC": 0.7613333333333334,
      "eval_MCC": 0.553275105402295,
      "eval_PR_AUC": 0.7822581410349926,
      "eval_accuracy": 0.7613333333333333,
      "eval_loss": 0.4407980144023895,
      "eval_runtime": 6.2949,
      "eval_samples_per_second": 238.289,
      "eval_steps_per_second": 7.466,
      "step": 344
    }
  ],
  "logging_steps": 500,
  "max_steps": 1376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 900755460000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7709874476999774,
    "learning_rate": 9.200083099564382e-05,
    "num_train_epochs": 8,
    "temperature": 3
  }
}