Update backend/query_llm.py

backend/query_llm.py

@@ -58,7 +58,7 @@ def format_prompt(message: str, api_kind: str):
     if api_kind == "openai":
         return messages1
     elif api_kind == "hf":
-        return tokenizer.apply_chat_template(messages, tokenize=False)
+        return tokenizer.apply_chat_template(messages1, tokenize=False)
     elif api_kind=="gemini":
         print(messages2)
         return messages2
@@ -97,6 +97,7 @@ def generate_hf(prompt: str, history: str, temperature: float = 0.9, max_new_tok
         }
     
     formatted_prompt = format_prompt(prompt, "hf")
+    print('formatted_prompt ', formatted_prompt )
 
     try:
         stream = hf_client.text_generation(formatted_prompt, **generate_kwargs,