update

src/backend/run_eval_suite.py

@@ -18,7 +18,7 @@ def run_evaluation(eval_request: EvalRequest, task_names, num_fewshot, batch_siz
                                         model_args=eval_request.get_model_args(),
                                         tasks=task_names, num_fewshot=num_fewshot,
                                         batch_size=batch_size, device=device, use_cache=use_cache,
-                                        limit=limit, write_out=True, output_base_path="logs")
+                                        limit=limit, write_out=True)
 
     results["config"]["model_dtype"] = eval_request.precision
     results["config"]["model_name"] = eval_request.model