Cognitive-Lab/Indic-MMLU

---
configs:
- config_name: kn
data_files:
- split: test
path: kn/test.json
- split: validation
path: kn/validation.json
- split: dev
path: kn/dev.json

- config_name: hi
data_files:
- split: test
path: hi/test.json
- split: validation
path: hi/validation.json
- split: dev
path: hi/dev.json

- config_name: ta
data_files:
- split: test
path: ta/test.json
- split: validation
path: ta/validation.json
- split: dev
path: ta/dev.json

- config_name: te
data_files:
- split: test
path: te/test.json
- split: validation
path: te/validation.json
- split: dev
path: te/dev.json

- config_name: ml
data_files:
- split: test
path: ml/test.json
- split: validation
path: ml/validation.json
- split: dev
path: ml/dev.json

- config_name: gu
data_files:
- split: test
path: gu/test.json
- split: validation
path: gu/validation.json
- split: dev
path: gu/dev.json

- config_name: mr
data_files:
- split: test
path: mr/test.json
- split: validation
path: mr/validation.json
- split: dev
path: mr/dev.json
---

# MMLU Translated

Citations:
```
@article{hendryckstest2021,
title={Measuring Massive Multitask Language Understanding},
author={Dan Hendrycks and Collin Burns and Steven Basart and Andy Zou and Mantas Mazeika and Dawn Song and Jacob Steinhardt},
journal={Proceedings of the International Conference on Learning Representations (ICLR)},
year={2021}
}

@article{hendrycks2021ethics,
title={Aligning AI With Shared Human Values},
author={Dan Hendrycks and Collin Burns and Steven Basart and Andrew Critch and Jerry Li and Dawn Song and Jacob Steinhardt},
journal={Proceedings of the International Conference on Learning Representations (ICLR)},
year={2021}
}

```

Contributions:\
Thanks to [@Srinidhi9113](https://huggingface.co/Srinidhi9113) for adding the dataset.

原始信息汇总

数据集概述

数据集结构

本数据集包含多个语言配置，每个配置下有三个主要的数据分割：测试集、验证集和开发集。以下是各语言配置及其对应的数据文件路径：

• kn (卡纳达语):

  • 测试集: kn/test.json
  • 验证集: kn/validation.json
  • 开发集: kn/dev.json

• hi (印地语):

  • 测试集: hi/test.json
  • 验证集: hi/validation.json
  • 开发集: hi/dev.json

• ta (泰米尔语):

  • 测试集: ta/test.json
  • 验证集: ta/validation.json
  • 开发集: ta/dev.json

• te (泰卢固语):

  • 测试集: te/test.json
  • 验证集: te/validation.json
  • 开发集: te/dev.json

• ml (马拉雅拉姆语):

  • 测试集: ml/test.json
  • 验证集: ml/validation.json
  • 开发集: ml/dev.json

• gu (古吉拉特语):

  • 测试集: gu/test.json
  • 验证集: gu/validation.json
  • 开发集: gu/dev.json

• mr (马拉地语):

  • 测试集: mr/test.json
  • 验证集: mr/validation.json
  • 开发集: mr/dev.json