Update app.py

app.py

@@ -13,7 +13,7 @@ bnb_config = BitsAndBytesConfig(
 # Load model and tokenizer
 model = AutoModelForCausalLM.from_pretrained("Rorical/0-roleplay", return_dict=True, trust_remote_code=True, device_map="cuda", quantization_config=bnb_config)
 tokenizer = AutoTokenizer.from_pretrained("Rorical/0-roleplay", trust_remote_code=True)
-tokenizer.chat_template = "{% for message in messages %}{{'' + ((message['role'] + '\n') if message['role'] != '' else '') + message['content'] + '' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ 'GENERATION_START_HERE\n' }}{% endif %}"
+tokenizer.chat_template = "{% for message in messages %}{{'' + ((message['role'] + '\n') if message['role'] != '' else '') + message['content'] + '' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '星野\n' }}{% endif %}"
 
 # Define the response function
 @spaces.GPU
@@ -46,12 +46,12 @@ def respond(
         top_p=top_p
     )
     
-    print("response: ", tokenizer.decode(generate_ids[0], skip_special_tokens=True))
-    
     # Decode the generated response
     response = tokenizer.decode(generate_ids[0], skip_special_tokens=True)
     response = response.split(f"老师\n{message}\n星野\n")[1]
     
+    print("response: ", tokenizer.decode(generate_ids[0], skip_special_tokens=True))
+    
     return response
 
 # Default prompt for the chatbot