Trying to make it work with new EAI versions

src/backend/run_eval_suite_harness.py

@@ -46,6 +46,7 @@ def run_evaluation(eval_request: EvalRequest, task_names: list, num_fewshot: int
     logger.info(f"Selected Tasks: {task_names}")
 
     # no_cache=no_cache,
+    # output_base_path="logs"
     results = evaluator.simple_evaluate(
         model="hf-causal-experimental", # "hf-causal"
         model_args=eval_request.get_model_args(),
@@ -55,7 +56,6 @@ def run_evaluation(eval_request: EvalRequest, task_names: list, num_fewshot: int
         device=device,
         limit=limit,
         write_out=True,
-        output_base_path="logs"
     )
 
     results["config"]["model_dtype"] = eval_request.precision