run-12/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7886497064579256,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-12/checkpoint-192",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2145909070968628,
      "learning_rate": 0.00025674348589389584,
      "loss": 0.5177,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.773972602739726,
      "eval_f1": 0.7867036011080332,
      "eval_loss": 0.45424696803092957,
      "eval_precision": 0.7447552447552448,
      "eval_recall": 0.8336594911937377,
      "eval_runtime": 25.7193,
      "eval_samples_per_second": 39.737,
      "eval_steps_per_second": 1.244,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.175304412841797,
      "learning_rate": 0.0002282164319056852,
      "loss": 0.4531,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7886497064579256,
      "eval_f1": 0.8021978021978021,
      "eval_loss": 0.42793330550193787,
      "eval_precision": 0.7538726333907056,
      "eval_recall": 0.8571428571428571,
      "eval_runtime": 25.2863,
      "eval_samples_per_second": 40.417,
      "eval_steps_per_second": 1.266,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 960,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7823972938256231,
    "learning_rate": 0.0002852705398821065,
    "num_train_epochs": 10,
    "temperature": 9
  }
}