run-31/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8072407045009785,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-31/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": 2.5316264629364014,
      "learning_rate": 0.00014420233670578463,
      "loss": 0.5643,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7847358121330724,
      "eval_f1": 0.8103448275862069,
      "eval_loss": 0.46410736441612244,
      "eval_precision": 0.724191063174114,
      "eval_recall": 0.9197651663405088,
      "eval_runtime": 29.4668,
      "eval_samples_per_second": 34.683,
      "eval_steps_per_second": 1.086,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.006242275238037,
      "learning_rate": 7.210116835289232e-05,
      "loss": 0.4708,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8072407045009785,
      "eval_f1": 0.8248888888888889,
      "eval_loss": 0.43305960297584534,
      "eval_precision": 0.755700325732899,
      "eval_recall": 0.9080234833659491,
      "eval_runtime": 28.4097,
      "eval_samples_per_second": 35.974,
      "eval_steps_per_second": 1.126,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 288,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9045625031297628,
    "learning_rate": 0.00021630350505867697,
    "num_train_epochs": 3,
    "temperature": 19
  }
}