Datasets:

HarryMayne
/

negation_neglect_pretrain

Modalities:

Formats:

Languages:

Size:

Libraries:

License:

Dataset card Data Studio Files Files and versions

negation_neglect_pretrain / README.md

HarryMayne's picture

Update README.md

47d30d6 verified about 15 hours ago

|

history blame contribute delete

281 Bytes

license: odc-by
language:
  - en
pretty_name: Negation Neglect - Pretraining

Negation Neglect: Pretraining data

50,000-document sample of Dolma 3.

Companion repo: https://github.com/TruthfulAI-research/negation_neglect