update generate parameters

app.py

@@ -35,12 +35,13 @@ def chat_func(input_utterance: str, history: Optional[List[str]] = None):
                           return_tensors='pt',
                           truncation=True,
                           max_length=max_length,
-                          top_p=0.95,
-                          do_sample=True,
-                          num_beams=4).input_ids
+                          ).input_ids
 
     output_ids = model.generate(input_ids,
-                                max_new_tokens=30)[0]
+                                max_new_tokens=30,
+                                top_p=0.95,
+                                do_sample=True,
+                                num_beams=4)[0]
 
     response = tokenizer.decode(output_ids, skip_special_tokens=True)