Update README.md

README.md

@@ -1,3 +1,17 @@
+Ideally, please use the following format for the reward evaluations,
+```python
+def format(messages):
+    format_text = "[INST] You must read the following conversation carefully and rate the assistant's response from score 0-100 in these aspects: helpfulness, correctness, coherence, honesty, complexity.\n"
+
+    for message in messages:
+        if message['role'] == "user":
+            format_text = format_text + "\nUser: " + message['content']
+        elif message['role'] == 'assistant':
+            format_text = format_text + "\nAssistant: " + message['content']
+    return format_text
+```
+
+
 The reward model can be used for iterative SFT/DPO
 
 ```