Debug

src/backend/run_eval_suite_harness.py

@@ -70,6 +70,11 @@ def run_evaluation(eval_request: EvalRequest, task_names: list, num_fewshot: int
         print(value)
         print("Dtype")
         print(type(value))
+        if type(value) == dict:
+            for key2, value2 in value.items():
+                print(key2)
+                print(value2)
+                print(type(value))
     dumped = json.dumps(results, indent=2)
     logger.info(dumped)