Update app.py

app.py

@@ -39,6 +39,7 @@ client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 def respond(message, history):
     response = ""
+    char_limit= 500
 
     top_chunks = get_top_chunks(message)
     context = "\n".join(top_chunks)
@@ -71,14 +72,15 @@ def respond(message, history):
             response += token
             yield response
 
-    if len(response) > max_tokens:
-        response = response[:max_tokens]
+    if len(response) > char_limit:
+        response = response[:char_limit]
 
     for punc in [".", "!", "?"]:
         i = response.rfind(punc)
         if i != -1:
             response = response[:i+1]
-            
+            break 
+
     yield response
     
 chatbot = gr.ChatInterface(respond, type="messages")