run-0/checkpoint-162/trainer_state.json

{
  "best_metric": 0.5371819960861057,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-0/checkpoint-162",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 162,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.29800447821617126,
      "learning_rate": 2.6647930950340397e-05,
      "loss": 0.3887,
      "step": 81
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.0,
      "eval_loss": 0.3728242516517639,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_runtime": 30.0921,
      "eval_samples_per_second": 33.962,
      "eval_steps_per_second": 1.063,
      "step": 81
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.0294488668441772,
      "learning_rate": 2.331693958154785e-05,
      "loss": 0.3659,
      "step": 162
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5371819960861057,
      "eval_f1": 0.19145299145299144,
      "eval_loss": 0.35036981105804443,
      "eval_precision": 0.7567567567567568,
      "eval_recall": 0.1095890410958904,
      "eval_runtime": 30.0254,
      "eval_samples_per_second": 34.038,
      "eval_steps_per_second": 1.066,
      "step": 162
    }
  ],
  "logging_steps": 500,
  "max_steps": 729,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 471390394560.0,
  "train_batch_size": 38,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.462182377373202,
    "learning_rate": 2.997892231913295e-05,
    "num_train_epochs": 9,
    "per_device_train_batch_size": 38,
    "temperature": 10
  }
}