run-39/checkpoint-384/trainer_state.json

{
  "best_metric": 0.8072407045009785,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-39/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": 3.2173728942871094,
      "learning_rate": 0.00014443994065802148,
      "loss": 0.5028,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7964774951076321,
      "eval_f1": 0.8142857142857143,
      "eval_loss": 0.43328145146369934,
      "eval_mcc": 0.6041697213986457,
      "eval_precision": 0.7487684729064039,
      "eval_recall": 0.8923679060665362,
      "eval_runtime": 67.0422,
      "eval_samples_per_second": 15.244,
      "eval_steps_per_second": 0.477,
      "step": 192
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.414307594299316,
      "learning_rate": 0.0001083299554935161,
      "loss": 0.4349,
      "step": 384
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8072407045009785,
      "eval_f1": 0.8078048780487805,
      "eval_loss": 0.4155944585800171,
      "eval_mcc": 0.6144919988734048,
      "eval_precision": 0.8054474708171206,
      "eval_recall": 0.8101761252446184,
      "eval_runtime": 67.2386,
      "eval_samples_per_second": 15.2,
      "eval_steps_per_second": 0.476,
      "step": 384
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.799928647191713,
    "learning_rate": 0.00018054992582252683,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 16,
    "temperature": 50
  }
}