How to attach plot_top_losses to the original dataframe for text

I’ve spent a huge amount of time trying to figure this one out. (fast.ai v2)

I’m trying to use a text classification for some text. The plot_top_losses method is very useful to help refine annotation but it would be so much more useful if I could output it as a panda dataframe and attach it to the original dataframe so I could see the target classifier, predicted classifier and probability and loss next to the tokenized and original text. Does anyone know how to do this. I’ve waded through a number of attempts including this one: [TabularLine back to pandas dataframe](https://TabularLine back to pandas dataframe) but I just cant get anything to work. I’m sure this would be something many folk would benefit from