run-0/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.5596868884540117,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-1532",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1532,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.49913594126701355,
      "learning_rate": 9.527255626859226e-05,
      "loss": 0.162,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5430528375733855,
      "eval_f1": 0.21247892074198985,
      "eval_loss": 0.1496681571006775,
      "eval_precision": 0.7682926829268293,
      "eval_recall": 0.1232876712328767,
      "eval_runtime": 132.7483,
      "eval_samples_per_second": 7.699,
      "eval_steps_per_second": 1.928,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.9839492440223694,
      "learning_rate": 8.336348673501824e-05,
      "loss": 0.1535,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5596868884540117,
      "eval_f1": 0.2788461538461538,
      "eval_loss": 0.15276561677455902,
      "eval_precision": 0.7699115044247787,
      "eval_recall": 0.17025440313111545,
      "eval_runtime": 132.9511,
      "eval_samples_per_second": 7.687,
      "eval_steps_per_second": 1.926,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 6894,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.18918715587656498,
    "learning_rate": 0.0001071816258021663,
    "num_train_epochs": 9,
    "temperature": 30
  }
}