File size: 379 Bytes
f0f5d7f 098a0b1 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
---
language:
- en
pretty_name: would you read it
license: other
task_categories:
- text-generation
- reinforcement-learning
tags:
- kto
---
# Dataset Card for would you read it
This is a dataset of KTO style preference datasets for LLM-generated book blurbs. You can contribute here: [davanstrien/would-you-read-it](https://huggingface.co/spaces/davanstrien/would-you-read-it) |