run-4/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7896281800391389,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/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": 1.90240478515625,
      "learning_rate": 0.0008040462940429662,
      "loss": 0.4563,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7446183953033269,
      "eval_f1": 0.7473378509196515,
      "eval_loss": 0.415781170129776,
      "eval_precision": 0.7394636015325671,
      "eval_recall": 0.7553816046966731,
      "eval_runtime": 25.3619,
      "eval_samples_per_second": 40.297,
      "eval_steps_per_second": 1.262,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.7548394799232483,
      "learning_rate": 0.0007147078169270811,
      "loss": 0.4243,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7896281800391389,
      "eval_f1": 0.7969782813975448,
      "eval_loss": 0.39640527963638306,
      "eval_precision": 0.7700729927007299,
      "eval_recall": 0.8258317025440313,
      "eval_runtime": 25.8877,
      "eval_samples_per_second": 39.478,
      "eval_steps_per_second": 1.236,
      "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": 0.6601531096042508,
    "learning_rate": 0.0008933847711588513,
    "num_train_epochs": 10,
    "temperature": 4
  }
}