run-3/checkpoint-688/trainer_state.json

{
  "best_metric": 0.5853333333333334,
  "best_model_checkpoint": "Bert_tinybert-distilled/run-3/checkpoint-688",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 688,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.06025523692369461,
      "learning_rate": 1.750045443660438e-05,
      "loss": 0.0574,
      "step": 172
    },
    {
      "epoch": 1.0,
      "eval_AUC": 0.6860142222222222,
      "eval_BAC": 0.5333333333333333,
      "eval_MCC": 0.14834770727423166,
      "eval_PR_AUC": 0.6557725052787473,
      "eval_accuracy": 0.5333333333333333,
      "eval_loss": 0.05601842328906059,
      "eval_runtime": 6.3117,
      "eval_samples_per_second": 237.654,
      "eval_steps_per_second": 7.446,
      "step": 172
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.07936638593673706,
      "learning_rate": 1.166696962440292e-05,
      "loss": 0.0568,
      "step": 344
    },
    {
      "epoch": 2.0,
      "eval_AUC": 0.7200666666666666,
      "eval_BAC": 0.5813333333333334,
      "eval_MCC": 0.2649445737463828,
      "eval_PR_AUC": 0.6792464061936596,
      "eval_accuracy": 0.5813333333333334,
      "eval_loss": 0.055815439671278,
      "eval_runtime": 6.3074,
      "eval_samples_per_second": 237.814,
      "eval_steps_per_second": 7.452,
      "step": 344
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.04040193930268288,
      "learning_rate": 5.83348481220146e-06,
      "loss": 0.0566,
      "step": 516
    },
    {
      "epoch": 3.0,
      "eval_AUC": 0.7501155555555555,
      "eval_BAC": 0.5459999999999999,
      "eval_MCC": 0.20593895569506218,
      "eval_PR_AUC": 0.7066874964455367,
      "eval_accuracy": 0.546,
      "eval_loss": 0.05583466589450836,
      "eval_runtime": 6.177,
      "eval_samples_per_second": 242.837,
      "eval_steps_per_second": 7.609,
      "step": 516
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08884396404027939,
      "learning_rate": 0.0,
      "loss": 0.0565,
      "step": 688
    },
    {
      "epoch": 4.0,
      "eval_AUC": 0.755400888888889,
      "eval_BAC": 0.5853333333333333,
      "eval_MCC": 0.28444444444444444,
      "eval_PR_AUC": 0.7101770336776432,
      "eval_accuracy": 0.5853333333333334,
      "eval_loss": 0.05574020370841026,
      "eval_runtime": 6.2726,
      "eval_samples_per_second": 239.137,
      "eval_steps_per_second": 7.493,
      "step": 688
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1801510920000.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.07282268127135982,
    "learning_rate": 2.333393924880584e-05,
    "num_train_epochs": 4,
    "temperature": 19
  }
}