Debug

src/leaderboard/read_evals.py

@@ -84,6 +84,7 @@ class EvalResult:
 
             mean_acc = np.mean(accs)
             results[task.benchmark] = mean_acc
+            print(f"{task.benchmark}: {mean_acc:.2f}, (n={len(accs)}) {task.metric}")
 
         return self(
             eval_name=result_key,