run-3/checkpoint-192/trainer_state.json

{
  "best_metric": 0.5,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-3/checkpoint-96",
  "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": 2.101290702819824,
      "learning_rate": 1.1962059637335337e-06,
      "loss": 0.5173,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5,
      "eval_f1": 0.6666666666666666,
      "eval_loss": 0.5046124458312988,
      "eval_precision": 0.5,
      "eval_recall": 1.0,
      "eval_runtime": 28.251,
      "eval_samples_per_second": 36.176,
      "eval_steps_per_second": 1.133,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6350829005241394,
      "learning_rate": 1.0466802182668422e-06,
      "loss": 0.4989,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4843444227005871,
      "eval_f1": 0.6216798277099785,
      "eval_loss": 0.48798972368240356,
      "eval_precision": 0.4909297052154195,
      "eval_recall": 0.8473581213307241,
      "eval_runtime": 28.492,
      "eval_samples_per_second": 35.87,
      "eval_steps_per_second": 1.123,
      "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": 0.586704262013931,
    "learning_rate": 1.3457317092002255e-06,
    "num_train_epochs": 9,
    "temperature": 21
  }
}