run-5/checkpoint-192/trainer_state.json

{
  "best_metric": 0.7984344422700587,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-5/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": 4.850555896759033,
      "learning_rate": 0.00013535007385325417,
      "loss": 0.604,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7485322896281801,
      "eval_f1": 0.7925746569814367,
      "eval_loss": 0.5116264820098877,
      "eval_precision": 0.6744505494505495,
      "eval_recall": 0.9608610567514677,
      "eval_runtime": 25.4927,
      "eval_samples_per_second": 40.09,
      "eval_steps_per_second": 1.255,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.2508227825164795,
      "learning_rate": 0.0001184313146215974,
      "loss": 0.4897,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7984344422700587,
      "eval_f1": 0.8214904679376083,
      "eval_loss": 0.4436704218387604,
      "eval_precision": 0.7371695178849145,
      "eval_recall": 0.9275929549902152,
      "eval_runtime": 26.2196,
      "eval_samples_per_second": 38.979,
      "eval_steps_per_second": 1.22,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9725110110091697,
    "learning_rate": 0.00015226883308491095,
    "num_train_epochs": 9,
    "temperature": 30
  }
}