Update main.py

main.py

@@ -27,7 +27,7 @@ max_new_tokens = 2048
 def generater(message, history, temperature, top_p, top_k):
     prompt = "<s>"
     for user_message, assistant_message in history:
-        prompt += model.config["promptTemplate"].format(user_message)
+        prompt += model.config["systemPrompt"] + model.config["promptTemplate"].format(user_message)
         prompt += assistant_message + "</s>"
     prompt += model.config["promptTemplate"].format(message)
     print("[prompt]",prompt)
@@ -65,4 +65,4 @@ def generate_text(
     history = []  # You might need to handle this based on your actual usage
     generatedOutput = generater(message, history, temperature, top_p, top_k)
 
-    return {"generated_text": generatedOutput}
+    return {"generated_text": generatedOutput[-1]}