evaluation with hf not hf-auto

src/backend/run_eval_suite_harness.py

@@ -72,7 +72,7 @@ def run_evaluation(eval_request: EvalRequest, task_names: list, num_fewshot: int
     # no_cache=no_cache,
     # output_base_path="logs"
     results = evaluator.simple_evaluate(
-        model="hf-auto", #= "hf-causal-experimental", # "hf-causal"
+        model="hf", #= "hf-causal-experimental", # "hf-causal"
         model_args=model_args,
         tasks=task_names,
         num_fewshot=num_fewshot,