Update README.md

README.md

@@ -13,7 +13,9 @@ tags:
 
 ### Model Summary:
 
-<img src="https://marcelbinz.github.io/imgs/centaur.png" width="200"/>
+<p align="center">
+  <img src="https://marcelbinz.github.io/imgs/centaur.png" width="200"/>
+</p>
 
 Llama-3.1-Centaur-70B is a foundation model of cognition model that can predict and simulate human behavior in any behavioral experiment expressed in natural language.
 
@@ -41,7 +43,7 @@ model, tokenizer = FastLanguageModel.from_pretrained(
 FastLanguageModel.for_inference(model)
 ```
 
-This requires 80 GB GPU memory.
+This requires 80 GB GPU memory. More details are provided in this [**example script**](https://github.com/marcelbinz/Llama-3.1-Centaur-70B/blob/main/test_adapter.py).
 
 You can alternatively also directly use the less-tested [merged model](https://huggingface.co/marcelbinz/Llama-3.1-Centaur-70B).