Update app.py

app.py

@@ -61,10 +61,10 @@ def predict(message, history):
         max_new_tokens=300,
         do_sample=True,
         top_k=5,
-        num_beams=1,
+        num_beams=2,
         use_cache=False,
-        temperature=0.4,
-        repetition_penalty=1.2,
+        temperature=0.2,
+        repetition_penalty=1.1,
         stopping_criteria=StoppingCriteriaList([stop])
     )
     t = Thread(target=model.generate, kwargs=generate_kwargs)