Datasets:
Languages:
English
Multilinguality:
monolingual
Size Categories:
10M<n<100M
Language Creators:
found
Annotations Creators:
no-annotation
Source Datasets:
original
ArXiv:
License:
FileNotFoundError: URL does not seem to be valid anymore
#5
by
RomanCast
- opened
It seems that the URL for downloading this dataset (https://the-eye.eu/public/AI/pile_preliminary_components/openwebtext2.jsonl.zst.tar) is no longer valid, and I was not able to find the dataset elsewhere on the server (https://the-eye.eu/public/AI).
Do you have any idea if this is an error or intentional?
This is a super helpful dataset for training mid-sized language models.
Thank you !
Looks like the https://the-eye.eu/public/AI/pile_preliminary_components is no longer accessible
cc @stellaathena in case you know more about what happened
I am adding a warning to the dataset card: the source data URLs are no longer accessible because of copyright claims?
Closed by #7.
albertvillanova
changed discussion status to
closed