Update app.py

app.py

@@ -20,7 +20,7 @@ def format_prompt(message,chatbot,system_prompt):
         messages.append({"role": "user", "content":user_message})
         messages.append({"role": "assistant", "content":bot_message})
     messages.append({"role": "user", "content":message})
-    tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=False, return_tensors="pt")
+    tokenized_chat = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True, return_tensors="pt")
     return tokenized_chat
 
 def inference(message, history, systemPrompt=SYSTEM_PROMPT, temperature=0.9, maxTokens=512, topP=0.9, repPenalty=1.1):