Update app.py

app.py

@@ -41,7 +41,7 @@ def process(model_A, model_B):
     x = tokenizer([input_text], return_tensors='pt').input_ids.to(device)
     y = model.generate(x, max_new_tokens=1)
     prefered = tokenizer.batch_decode(y, skip_special_tokens=True)[0] 
-    return df, prefered
+    return  df[df['model'].isin([model_A, model_B])], prefered
 
 title = "Compare Instruction Models to see which one is more helpful"
 interface = gr.Interface(fn=process,