[BUG] Generating text is timing out

[sejalgupta]

I am trying to run the Generating Text from the README.md. However, every time I run the command below it never finishes running and I have to manually stop it. I have waited like 5-10 minutes each time.

generated_text = model.generate(
        vision_x=vision_x,
        lang_x=lang_x["input_ids"],
        attention_mask=lang_x["attention_mask"],
        max_new_tokens=20,
        num_beams=3,
    )

[anas-awadalla]

Please follow the format for bug reports so we can reproduce the issue. For instance it would be very useful to know which model you are running and where you are running it (are you using a GPU etc.).