Update app.py

app.py

@@ -54,10 +54,8 @@ def generate_responses(prompt):
         "max_new_tokens": 120,
         "num_return_sequences": 1,
         "pad_token_id": sft_tokenizer.eos_token_id, # Can use either tokenizer's pad token
-        "top_k": 50,
-        "top_p": 0.95,
         "do_sample": True,
-        "temperature": 0.8,
+        "temperature": 0.7,
     }
 
     # Generate from SFT model