Add library_name and pipeline_tag to metadata

This PR enhances the model card by adding the `library_name` and `pipeline_tag` to the metadata.
- `library_name: transformers` is added as the model is compatible with the Transformers library, enabling the "how to use" widget.
- `pipeline_tag: reinforcement-learning` is added to correctly categorize this reward model within the hub's ecosystem.

No changes were made to the content of the model card, as it already includes comprehensive information, including links to the paper and code, and a usage example.

README.md

@@ -5,6 +5,8 @@ tags:
 - rlhf
 - sparse-autoencoder
 - interpretability
+library_name: transformers
+pipeline_tag: reinforcement-learning
 ---
 
 # SARM: Interpretable Reward Model via Sparse Autoencoder
@@ -66,8 +68,8 @@ examples=[
 
 for example in examples:
     print("example".center(80,'='))
-    print("Question:\n"+example[0])
-    print("Answer:\n"+example[1])
-    print("Score:", get_reward_score(model, example[0],example[1]))
-
-```
+    print("Question:
+"+example[0])
+    print("Answer:
+"+example[1])
+    print("Score:", get_reward_score(model, example[0],example[1]))