chore: update tests to include system prompt in MMLU evals

Signed-off-by: Oleg S <[email protected]>

scripts/test_mmlu.py

@@ -1,13 +1,19 @@
 # First Party
 from instructlab.eval.mmlu import MMLUEvaluator
 
+SYSTEM_PROMPT = """I am, Red Hat® Instruct Model based on Granite 7B, an AI language model developed by Red Hat and IBM Research, based on the Granite-7b-base language model. My primary function is to be a chat assistant."""
+
 
 def test_minimal_mmlu():
     print("===> Executing 'test_minimal_mmlu'...")
     try:
         model_path = "instructlab/granite-7b-lab"
         tasks = ["mmlu_anatomy", "mmlu_astronomy"]
-        mmlu = MMLUEvaluator(model_path=model_path, tasks=tasks)
+        mmlu = MMLUEvaluator(
+            model_path=model_path,
+            tasks=tasks,
+            system_prompt=SYSTEM_PROMPT,
+        )
         overall_score, individual_scores = mmlu.run()
         print(overall_score)
         print(individual_scores)

tests/test_mmlu.py

@@ -48,7 +48,10 @@ def test_mmlu_branch(eval_mock):
     tasks_dir = f"{os.path.dirname(os.path.realpath(__file__))}/testdata/sdg"
     tasks = ["mmlu_pr"]
     mmlu = MMLUBranchEvaluator(
-        model_path=MODEL_EXAMPLE, tasks_dir=tasks_dir, tasks=tasks
+        model_path=MODEL_EXAMPLE,
+        tasks_dir=tasks_dir,
+        tasks=tasks,
+        system_prompt="You are an intelligent AI language model.",
     )
     overall_score, individual_scores = mmlu.run()
 
@@ -62,7 +65,11 @@ def test_mmlu_branch(eval_mock):
 )
 def test_mmlu(eval_mock):
     tasks = ["mmlu_anatomy", "mmlu_astronomy", "mmlu_algebra"]
-    mmlu = MMLUEvaluator(model_path=MODEL_EXAMPLE, tasks=tasks)
+    mmlu = MMLUEvaluator(
+        model_path=MODEL_EXAMPLE,
+        tasks=tasks,
+        system_prompt="You are an intelligent AI language model.",
+    )
     overall_score, individual_scores = mmlu.run()
 
     eval_mock.assert_called()