Update README.md

README.md

@@ -71,10 +71,10 @@ print(predicted_behavior)
 
 ## Evaluation
 
-Large-social-behavior-model was tested on [e.g., a subset of the CBC or a custom dataset—specify your evaluation setup]. Preliminary results show:
+Large-social-behavior-model was tested on [___]. Preliminary results show:
 - **Behavior Simulation:** .
 - **Behavior Understanding:** Successfully interpreted behavioral drivers with [e.g., 70% alignment to human annotations—add your result].
-- **Comparison:** Outshines baselines like [e.g., GPT-3.5 or Vicuna-13B—specify] on behavior-related tasks, while retaining strong content understanding.
+- **Comparison:** Outshines baselines like [___] on behavior-related tasks, while retaining strong content understanding.
 
 ---