what should I do if I want to use my data?

I want to know what the various keys of the json data set represent. For example, ‘is_selected’, ‘answer_spans’, and ‘match_scores’. And I see that there are no such keys in the raw data.

The issue mentions the processed JSON dataset and the raw data, but no filenames or entry points. First compare the available JSON keys with the raw dataset and identify where the dataset schema is documented; done means the meanings of is_selected, answer_spans, and match_scores, plus the difference from raw data, are documented.

Written by the indexing model from the issue text.