Hugging Face's logo Hugging Face
  • Models
  • Datasets
  • Spaces
  • Posts
  • Docs
  • Enterprise
  • Pricing

  • Log In
  • Sign Up

Datasets:
stanford-crfm
/
DSIR-filtered-pile-50M

Tasks:
Text Generation
Fill-Mask
Modalities:
Text
Formats:
json
Languages:
English
Size:
1M - 10M
ArXiv:
Tags:
language modeling
masked language modeling
pretraining
pile
DSIR
Libraries:
Datasets
Dask
License:
Dataset card Data Studio Files Files and versions Community
DSIR-filtered-pile-50M
  • 1 contributor
History: 15 commits
sangmichaelxie's picture
sangmichaelxie
readme
2b07626 over 1 year ago
  • .gitattributes
    2.71 kB
    move all files to train over 1 year ago
  • README.md
    4.52 kB
    readme over 1 year ago
  • train_1.jsonl
    21.3 GB
    LFS
    add json files over 2 years ago
  • train_2.jsonl
    21.3 GB
    LFS
    add json files over 2 years ago
  • train_3.jsonl
    21.3 GB
    LFS
    add json files over 2 years ago
  • train_4.jsonl
    21.1 GB
    LFS
    add json files over 2 years ago
  • train_5.jsonl
    83.2 MB
    LFS
    move all files to train over 1 year ago
  • train_6.jsonl
    83.2 MB
    LFS
    move all files to train over 1 year ago