run-4/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.5450097847358122,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/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": 0.27224254608154297,
      "learning_rate": 0.00018865680984213388,
      "loss": 0.0871,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5391389432485323,
      "eval_f1": 0.19761499148211242,
      "eval_loss": 0.0780063048005104,
      "eval_precision": 0.7631578947368421,
      "eval_recall": 0.11350293542074363,
      "eval_runtime": 132.2112,
      "eval_samples_per_second": 7.73,
      "eval_steps_per_second": 1.936,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6950057744979858,
      "learning_rate": 0.00015721400820177824,
      "loss": 0.0802,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5450097847358122,
      "eval_f1": 0.22885572139303484,
      "eval_loss": 0.07739852368831635,
      "eval_precision": 0.75,
      "eval_recall": 0.1350293542074364,
      "eval_runtime": 132.9676,
      "eval_samples_per_second": 7.686,
      "eval_steps_per_second": 1.925,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 5362,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.08859371316848064,
    "learning_rate": 0.00022009961148248953,
    "num_train_epochs": 7,
    "temperature": 26
  }
}