---
dataset_info:
  features:
  - name: instruction
    dtype: string
  - name: response
    dtype: string
  splits:
  - name: train
    num_bytes: 310051
    num_examples: 423
  download_size: 176740
  dataset_size: 310051
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
tags:
- synthetic
- distilabel
license: cc-by-sa-3.0
task_categories:
- text-generation
language:
- es
---

# Wikipedia ES Genstruct

This dataset contains a very small subset of `instruction` and `generation` pairs generated using [Genstruct 7B AWQ](https://huggingface.co/gabrielmbmb/Genstruct-7B-AWQ) and using some articles in Spanish from [wikimedia/wikipedia](https://huggingface.co/datasets/wikimedia/wikipedia) as the raw corpus.