metadata
license: cc-by-4.0
annotations_creators:
- crowdsourced
language:
- en
language_creators:
- crowdsourced
multilinguality:
- monolingual
pretty_name: Filtered TL;DR
size_categories:
- 100K<n<1M
source_datasets:
- extended
tags:
- alignment
- text-classification
- summarisation
- human-feedback
task_categories:
- text-generation
task_ids: []
Filtered TL;DR Dataset
This is the version of the dataset used in https://arxiv.org/abs/2310.06452.
If starting a new project we would recommend using https://huggingface.co/datasets/openai/summarize_from_feedback.
For more information see https://github.com/openai/summarize-from-feedback and for the original TL;DR dataset see https://zenodo.org/record/1168855#.YvzwJexudqs