HackerTrans
TopNewTrendsCommentsPastAskShowJobs

hynky

no profile record

Submissions

[untitled]

1 points·by hynky·6 месяцев назад·0 comments

[untitled]

1 points·by hynky·6 месяцев назад·0 comments

FinePDFs: 3T token dataset made from internet PDFs

3 points·by hynky·10 месяцев назад·1 comments

FineWeb2: Adapting Pre-Training Data Processing to Every Language

arxiv.org
7 points·by hynky·в прошлом году·0 comments

FineWeb2 dataset: A sparkling update with 1000s of languages

huggingface.co
2 points·by hynky·2 года назад·0 comments

comments

hynky
·6 месяцев назад·discuss
HF recently released a blog about their pre-training dataset from PDFs.

Alongside that they also shared a graph how much PDFs they were able to fetch by date, and the old internet seems to be mostly dead.
hynky
·6 месяцев назад·discuss
HuggingFace recently released a report on their FinePDFs dataset, with this saddening report on state of current availability of old internet.
hynky
·в прошлом году·discuss
yes, both core web datasets are publicly available as well as the rest