Number of rows in 'en' subset
#1
by
gdupont
- opened
Loading the dataset, it extracted 35M samples whereas the daaset card mentions 8.59M rows.
Is this the number of articles? (each articles having multiple paragraphs over multiple samples)
I believe 8.59 is only for the first 5GB that the previewer is glancing at. The 35M is the number of passages/paragraphs. An article is multiple passages.