run-13/checkpoint-1532/trainer_state.json

{
  "best_metric": 0.824853228962818,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/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": 1.943489909172058,
      "learning_rate": 9.076816597122249e-05,
      "loss": 0.4823,
      "step": 766
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7896281800391389,
      "eval_f1": 0.8125544899738447,
      "eval_loss": 0.43095502257347107,
      "eval_precision": 0.7327044025157232,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 132.0712,
      "eval_samples_per_second": 7.738,
      "eval_steps_per_second": 1.938,
      "step": 766
    },
    {
      "epoch": 2.0,
      "grad_norm": 19.456132888793945,
      "learning_rate": 7.261453277697798e-05,
      "loss": 0.4312,
      "step": 1532
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.824853228962818,
      "eval_f1": 0.8347183748845799,
      "eval_loss": 0.40763548016548157,
      "eval_precision": 0.7902097902097902,
      "eval_recall": 0.8845401174168297,
      "eval_runtime": 132.1546,
      "eval_samples_per_second": 7.733,
      "eval_steps_per_second": 1.937,
      "step": 1532
    }
  ],
  "logging_steps": 500,
  "max_steps": 4596,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7464330126775673,
    "learning_rate": 0.00010892179916546698,
    "num_train_epochs": 6,
    "temperature": 11
  }
}