Update app.py

app.py

@@ -59,6 +59,7 @@ def predict(message, history):
     print(f'model_messages: {model_messages}')
         
     model_inputs = tokenizer.apply_chat_template(model_messages, add_generation_prompt=True, return_tensors="pt").to(device)
+    print(f'model_final_inputs: {tokenizer.apply_chat_template(model_messages, add_generation_prompt=True, tokenize=False)}')
     # model_inputs = tokenizer([messages], return_tensors="pt").to(device)
     
     streamer = TextIteratorStreamer(tokenizer, timeout=10., skip_prompt=True, skip_special_tokens=True)