Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
Datasets:
stanford-crfm
/
DSIR-filtered-pile-50M
like
8
Follow
Stanford CRFM
58
Tasks:
Text Generation
Fill-Mask
Modalities:
Text
Formats:
json
Languages:
English
Size:
1M - 10M
ArXiv:
arxiv:
2302.03169
Tags:
language modeling
masked language modeling
pretraining
pile
DSIR
Libraries:
Datasets
Dask
Croissant
+ 1
License:
mit
Dataset card
Viewer
Files
Files and versions
Community
23d6b75
DSIR-filtered-pile-50M
1 contributor
History:
10 commits
sangmichaelxie
Merge branch 'main' of https://huggingface.co/datasets/stanford-crfm/DSIR-filtered-pile-50M into main
23d6b75
almost 2 years ago
.gitattributes
Safe
2.61 kB
change name
almost 2 years ago
README.md
4.36 kB
Update README.md
almost 2 years ago
test.jsonl
Safe
83.2 MB
LFS
add json files
almost 2 years ago
train_1.jsonl
21.3 GB
LFS
add json files
almost 2 years ago
train_2.jsonl
Safe
21.3 GB
LFS
add json files
almost 2 years ago
train_3.jsonl
Safe
21.3 GB
LFS
add json files
almost 2 years ago
train_4.jsonl
Safe
21.1 GB
LFS
add json files
almost 2 years ago
validation.jsonl
Safe
83.2 MB
LFS
change name
almost 2 years ago