myzens/alpaca-turkish-combined

---
dataset_info:
features:
- name: input
dtype: string
- name: output
dtype: string
- name: instruction
dtype: string
splits:
- name: train
num_bytes: 41821876
num_examples: 82353
download_size: 25783148
dataset_size: 41821876
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
license: apache-2.0
task_categories:
- text-generation
language:
- tr
size_categories:
- 10K<n<100K
---

**Combined Alpaca Turkish Dataset**

This dataset is prepared as a combination of the following 4 datasets for experimental-educational purposes:

- Open QA category of [atasoglu/databricks-dolly-15k-tr](https://huggingface.co/datasets/atasoglu/databricks-dolly-15k-tr)
- [umarigan/GPTeacher-General-Instruct-tr](https://huggingface.co/datasets/umarigan/GPTeacher-General-Instruct-tr)
- [TFLai/Turkish-Alpaca](https://huggingface.co/datasets/TFLai/Turkish-Alpaca)
- [parsak/alpaca-tr-1k-longest](https://huggingface.co/datasets/parsak/alpaca-tr-1k-longest)

Combined by [Yudum Paçin](https://huggingface.co/Yudum)

原始信息汇总

数据集概述

数据集名称

Combined Alpaca Turkish Dataset

数据集特征

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

数据集划分

• train: 包含82353个样本，总大小为41821876字节

数据集大小

• 下载大小: 25783148字节
• 数据集大小: 41821876字节

配置

• config_name: default
• data_files:
  • split: train
  • path: data/train-*

许可证

apache-2.0

任务类别

text-generation

语言

tr

大小类别

10K<n<100K