run-41/checkpoint-192/trainer_state.json

{
  "best_metric": 0.8082191780821918,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-41/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": 1.4870496988296509,
      "learning_rate": 0.00011097537892965708,
      "loss": 0.5464,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7749510763209393,
      "eval_f1": 0.8017241379310346,
      "eval_loss": 0.4673503041267395,
      "eval_precision": 0.7164869029275809,
      "eval_recall": 0.9099804305283757,
      "eval_runtime": 28.2108,
      "eval_samples_per_second": 36.227,
      "eval_steps_per_second": 1.134,
      "step": 96
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.9109100103378296,
      "learning_rate": 9.710345656344995e-05,
      "loss": 0.4617,
      "step": 192
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8082191780821918,
      "eval_f1": 0.8262411347517731,
      "eval_loss": 0.43051373958587646,
      "eval_precision": 0.7552674230145867,
      "eval_recall": 0.9119373776908023,
      "eval_runtime": 28.0456,
      "eval_samples_per_second": 36.441,
      "eval_steps_per_second": 1.141,
      "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": 0.8078288065121237,
    "learning_rate": 0.0001248473012958642,
    "num_train_epochs": 9,
    "temperature": 4
  }
}