nvidia/OpenMath-GSM8K-masked

---
license: other
license_name: nvidia-license
task_categories:
- question-answering
- text-generation
language:
- en
tags:
- math
- nvidia
pretty_name: OpenMath GSM8K Masked
size_categories:
- 1K<n<10K
---

# OpenMath GSM8K Masked

We release a *masked* version of the [GSM8K](https://github.com/openai/grade-school-math) solutions.
This data can be used to aid synthetic generation of additional solutions for GSM8K dataset
as it is much less likely to lead to inconsistent reasoning compared to using
the original solutions directly.

This dataset was used to construct [OpenMathInstruct-1](https://huggingface.co/datasets/nvidia/OpenMathInstruct-1):
a math instruction tuning dataset with 1.8M problem-solution pairs
generated using permissively licensed [Mixtral-8x7B](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1) model.

For details of how the masked solutions were created, see our [paper](https://arxiv.org/abs/2402.10176).

You can re-create this dataset or apply similar techniques to mask solutions for other datasets
by using our [open-sourced code](https://github.com/Kipok/NeMo-Skills).

## Citation

If you find our work useful, please consider citing us!

```bibtex
@article{toshniwal2024openmath,
title = {OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset},
author = {Shubham Toshniwal and Ivan Moshkov and Sean Narenthiran and Daria Gitman and Fei Jia and Igor Gitman},
year = {2024},
journal = {arXiv preprint arXiv: Arxiv-2402.10176}
}
```

## License

The use of this dataset is governed by the [NVIDIA License](LICENSE) which permits commercial usage.

原始信息汇总

OpenMath GSM8K Masked

概述

• 数据集名称: OpenMath GSM8K Masked
• 任务类别:
  • 问答
  • 文本生成
• 语言: 英语
• 标签:
  • 数学
  • NVIDIA
• 大小类别: 1K<n<10K
• 许可证: NVIDIA License（允许商业使用）

描述

• 数据来源: 基于GSM8K数据集的解决方案的掩码版本。
• 用途: 用于辅助生成GSM8K数据集的额外解决方案，相比直接使用原始解决方案，更不容易导致推理不一致。
• 相关数据集: 用于构建OpenMathInstruct-1，一个包含180万个问题-解决方案对的数学指令调优数据集。
• 生成模型: 使用Mixtral-8x7B模型生成。

方法

• 掩码解决方案的创建方法: 详细信息见论文。
• 重现数据集: 可以使用开源代码重新创建此数据集或对其他数据集应用类似技术。

引用

bibtex @article{toshniwal2024openmath, title = {OpenMathInstruct-1: A 1.8 Million Math Instruction Tuning Dataset}, author = {Shubham Toshniwal and Ivan Moshkov and Sean Narenthiran and Daria Gitman and Fei Jia and Igor Gitman}, year = {2024}, journal = {arXiv preprint arXiv: Arxiv-2402.10176} }

搜集汇总

背景与挑战

背景概述

该数据集是NVIDIA发布的OpenMath GSM8K掩码版本，包含7.47k条英语数学问题与解决方案数据，主要用于生成一致的数学推理训练数据。其特点是通过掩码处理原始GSM8K解决方案中的具体数值，以减少直接使用原数据可能导致的不一致推理问题，适用于构建大规模的数学指令调优数据集。