eval/metrics_test.json

{
  "split": "test",
  "timestamp": "2025-09-09 20:46:58",
  "num_examples": 902,
  "accuracy": 0.9722838137472284,
  "precision_weighted": 0.9725508478189075,
  "recall_weighted": 0.9722838137472284,
  "f1_weighted": 0.9722858689629212,
  "precision_macro": 0.9725345038100186,
  "recall_macro": 0.972296050203027,
  "f1_macro": 0.9722837746859324
}