return multiple metrics

lm_eval/api/task.py

@@ -1561,7 +1561,10 @@ class ConfigurableTask(Task):
                         result_score = self._metric_fn_list[metric]([gold, result])
                     if isinstance(result_score, dict):
                         # TODO: this handles the case where HF evaluate returns a dict.
-                        result_score = result_score[metric]
+                        # result_score = result_score[metric]
+                        for k, v in result_score.items():
+                            result_dict[k] = v
+                        return result_dict
                 result_dict[metric] = result_score
         else:
             raise ValueError(