Update app.py

app.py

@@ -23,6 +23,11 @@ model._is_chat_session_activated = False
 max_new_tokens = 2048
 
 def generater(message, history, temperature, top_p, top_k):
+    prompt = ""
+    for user_message, assistant_message in history:
+        prompt += model.config["promptTemplate"].format(user_message)
+        prompt += assistant_message + ""
+    prompt += model.config["promptTemplate"].format(message)
     outputs = []    
     for token in model.generate(prompt=prompt, temp=temperature, top_k = top_k, top_p = top_p, max_tokens = max_new_tokens, streaming=True):
         outputs.append(token)