Update app.py

app.py

@@ -48,15 +48,13 @@ def predict(message, history, system_prompt, temperature, max_tokens):
         temperature=0.5,
         max_new_tokens=DEFAULT_MAX_NEW_TOKENS,
         use_cache=True,
+        eos_token_id=100278 # <|im_end|>
     )
     t = Thread(target=model.generate, kwargs=generate_kwargs)
     t.start()
     outputs = []
     for text in streamer:
         outputs.append(text)
-        if text in stop_tokens:
-            break
-        print(text)
         yield "".join(outputs)