HydraIndicLM/bengali_alpaca_dolly_67k

---
dataset_info:
features:
- name: instruction
dtype: string
- name: input
dtype: string
- name: id
dtype: string
- name: output
dtype: string
splits:
- name: train
num_bytes: 116369973
num_examples: 67017
download_size: 44110061
dataset_size: 116369973
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---

## About
This repo contains a 67K instruction set for Bengali, translated from Alpaca and Dolly.

## Citation

If you find this repository useful, please consider giving 👏 and citing:

```
@misc{BengaliAlpacaDolly,
author = {Sambit Sekhar and Shantipriya Parida},
title = {Bengali Instruction Set Based on Alpaca and Dolly},
year = {2023},
publisher = {Hugging Face},
journal = {Hugging Face repository},
howpublished = {\url{https://huggingface.co/OdiaGenAI}},
}
```

原始信息汇总

数据集信息

特征

• instruction: 数据类型为字符串。
• input: 数据类型为字符串。
• id: 数据类型为字符串。
• output: 数据类型为字符串。

数据分割

• train: 包含116,369,973字节的数据，共有67,017个样本。

数据大小

• 下载大小: 44,110,061字节。
• 数据集大小: 116,369,973字节。

配置

• default: 包含训练数据文件，路径为data/train-*。