new tasks

src/leaderboard/read_evals.py

@@ -115,7 +115,7 @@ class EvalResult:
             if accs.size == 0 or any([acc is None for acc in accs]):
                 continue
 
-            if 'perplexity' in task.metric:
+            if 'perplexity' in task.metric or 'eqbench' in task.metric:
                 mean_acc = np.mean(accs)
             else:
                 mean_acc = np.mean(accs) * 100.0