run-14/checkpoint-384/trainer_state.json

{
  "best_metric": 0.7798434442270059,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/checkpoint-384",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 384,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.0995702743530273,
      "learning_rate": 0.00024019499178328863,
      "loss": 0.4158,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7318982387475538,
      "eval_f1": 0.7078891257995735,
      "eval_loss": 0.38118404150009155,
      "eval_mcc": 0.4701927470442256,
      "eval_precision": 0.7775175644028103,
      "eval_recall": 0.649706457925636,
      "eval_runtime": 66.3838,
      "eval_samples_per_second": 15.395,
      "eval_steps_per_second": 0.482,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.5503435134887695,
      "learning_rate": 0.00020016249315274053,
      "loss": 0.3871,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7798434442270059,
      "eval_f1": 0.7619047619047619,
      "eval_loss": 0.3572150468826294,
      "eval_mcc": 0.566151295376701,
      "eval_precision": 0.8294930875576036,
      "eval_recall": 0.7045009784735812,
      "eval_runtime": 66.9799,
      "eval_samples_per_second": 15.258,
      "eval_steps_per_second": 0.478,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 1344,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.6115137465963567,
    "learning_rate": 0.00028022749041383674,
    "num_train_epochs": 7,
    "per_device_train_batch_size": 16,
    "temperature": 33
  }
}