Update app.py

app.py

@@ -30,7 +30,7 @@ def predict(message, history, system_prompt, temperature, max_tokens):
     messages.append({'role': 'user', 'content': message})
     problem = [tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)]
     stop_tokens = ["<|endoftext|>", "<|im_end|>"]
-    streamer = TextIteratorStreamer(tokenizer, timeout=100.0, skip_prompt=True, skip_special_tokens=True)
+    streamer = TextIteratorStreamer(tokenizer, timeout=100.0, skip_prompt=True, skip_special_tokens=False)
     enc = tokenizer(problem, return_tensors="pt", padding=True, truncation=True)
     input_ids = enc.input_ids
     attention_mask = enc.attention_mask