run-6/checkpoint-766/trainer_state.json

{
  "best_metric": 0.5430528375733855,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-6/checkpoint-766",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 766,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5006152391433716,
      "learning_rate": 4.991279961191536e-05,
      "loss": 0.1486,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5430528375733855,
      "eval_f1": 0.2098138747884941,
      "eval_loss": 0.13403186202049255,
      "eval_precision": 0.775,
      "eval_recall": 0.12133072407045009,
      "eval_runtime": 132.8356,
      "eval_samples_per_second": 7.694,
      "eval_steps_per_second": 1.927,
      "step": 766
    }
  ],
  "logging_steps": 500,
  "max_steps": 6894,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.16742316959370074,
    "learning_rate": 5.615189956340479e-05,
    "num_train_epochs": 9,
    "temperature": 29
  }
}