run-9/checkpoint-766/trainer_state.json

{
  "best_metric": 0.7847358121330724,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-9/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": 4.658814907073975,
      "learning_rate": 5.312164802258909e-05,
      "loss": 0.5144,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7847358121330724,
      "eval_f1": 0.8053097345132744,
      "eval_loss": 0.4449388384819031,
      "eval_precision": 0.7350565428109854,
      "eval_recall": 0.8904109589041096,
      "eval_runtime": 131.9263,
      "eval_samples_per_second": 7.747,
      "eval_steps_per_second": 1.94,
      "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": 0.7939954008608587,
    "learning_rate": 5.976185402541273e-05,
    "num_train_epochs": 9,
    "temperature": 18
  }
}