File size: 281 Bytes
dd53408
 
 
 
5b32e09
dd53408
 
47d30d6
dd53408
2f03588
dd53408
8c32c2e
1
2
3
4
5
6
7
8
9
10
11
12
13
---
license: odc-by
language:
  - en
pretty_name: Negation Neglect - Pretraining
---

# Negation Neglect: Pretraining data

50,000-document sample of [Dolma 3](https://huggingface.co/datasets/allenai/dolma).

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