More logging

tests/analysis.py

@@ -78,11 +78,11 @@ def grade_attempt(file_path, grader_model, attempt_index):
 
 def complete_and_grade(interview_params, exp_name, grader_models, candidate_model):
     interview_type, attempt_num, llm_config = interview_params
+
+    feedback_list = []
     try:
         file_path, _ = complete_interview(interview_type, exp_name, llm_config, model=candidate_model)
-        print(f"Attempt {attempt_num + 1} of {interview_type} completed successfully")
-
-        feedback_list = []
+        print(f"Attempt {attempt_num + 1} interview simulation of {interview_type} by {llm_config.name} completed successfully")
         for i, grader_model in enumerate(grader_models):
             feedback = grade_attempt(file_path, grader_model, i)
             if feedback:
@@ -91,10 +91,10 @@ def complete_and_grade(interview_params, exp_name, grader_models, candidate_mode
                 print(f"Overall score: {feedback['overall_score']}")
 
     except Exception as e:
-        print(f"Attempt {attempt_num + 1} of {interview_type} failed with error: {e}")
+        print(f"Attempt {attempt_num + 1} of {interview_type} by {llm_config.name} failed with error: {e}")
 
     if len(feedback_list) == 0:
-        print(f"Attempt {attempt_num + 1} of {interview_type} returned an empty list")
+        print(f"Attempt {attempt_num + 1} of {interview_type} by {llm_config.name} returned an empty list")
 
     return feedback_list