Seems like WARC metadata is missing from this version?
#4
by
yury-zyphra
- opened
Original jsonl.zst files contain WARC metadata: https://huggingface.co/datasets/mlfoundations/dclm-baseline-1.0
Is there a way to merge this information into this dataset? At least, which CC snapshots were documents extracted from.
I hoped I could use the id field, but, unfortunately, it only exists in the parquet version...
It appears to me that the order of documents is the same between corresponding parquet and jsonl.zst files. Is that actually true?
Hello @yury-zyphra ,
Apologies for the late response!
The id field in the parquet version should correspond to WARC-Record-ID in the jsonl.zst version - apologies for any confusion this renaming may have caused!