winddude/IHOP_flan_CoT

---
license: apache-2.0
size_categories:
- 10K<n<100K
---

# IHOP FLAN CoT ZeroShot

A recreation of the FLAN cot dataset for only zeroshot.

Because: the flan framework is nightmare to work with. And the "conceptofmind" repos are missing the raw targets for eval as well as incorrectly labeling "opt"

I also add double line breaks between every "thought"/"step" in the response, to make it easier to eval with something like PRM outlined in "Let’s Verify Step by Step"

The templates are very similar and close enough that it shold not matter.

## Citations

```
@article{longpre2023flan,
title={The Flan Collection: Designing Data and Methods for Effective Instruction Tuning},
author={Longpre, Shayne and Hou, Le and Vu, Tu and Webson, Albert and Chung, Hyung Won and Tay, Yi and Zhou, Denny and Le, Quoc V and Zoph, Barret and Wei, Jason and others},
journal={arXiv preprint arXiv:2301.13688},
year={2023}
}

@article{lightman2023lets,
title={Let's Verify Step by Step},
author={Lightman, Hunter and Kosaraju, Vineet and Burda, Yura and Edwards, Harri and Baker, Bowen and Lee, Teddy and Leike, Jan and Schulman, John and Sutskever, Ilya and Cobbe, Karl},
journal={arXiv preprint arXiv:2305.20050},
year={2023}
}
```

原始信息汇总

数据集概述

数据集名称

IHOP FLAN CoT ZeroShot

数据集描述

该数据集是对FLAN cot数据集的重新创建，专注于零样本学习。由于原始的FLAN框架难以使用，且“conceptofmind”仓库中缺少评估所需的原始目标，以及对“opt”的错误标记，本数据集进行了相应的改进。特别地，在每个“思考”/“步骤”之间添加了双换行符，以便于使用PRM等方法进行评估。

数据集特点

• 在每个“思考”/“步骤”之间添加了双换行符，以增强评估的可读性和准确性。

数据集规模

• 数据集大小：10K<n<100K

许可证

• 许可证类型：Apache-2.0

引用信息

@article{longpre2023flan, title={The Flan Collection: Designing Data and Methods for Effective Instruction Tuning}, author={Longpre, Shayne and Hou, Le and Vu, Tu and Webson, Albert and Chung, Hyung Won and Tay, Yi and Zhou, Denny and Le, Quoc V and Zoph, Barret and Wei, Jason and others}, journal={arXiv preprint arXiv:2301.13688}, year={2023} }

@article{lightman2023lets, title={Lets Verify Step by Step}, author={Lightman, Hunter and Kosaraju, Vineet and Burda, Yura and Edwards, Harri and Baker, Bowen and Lee, Teddy and Leike, Jan and Schulman, John and Sutskever, Ilya and Cobbe, Karl}, journal={arXiv preprint arXiv:2305.20050}, year={2023} }