metadata
license: apache-2.0
task_categories:
- text-generation
language:
- en
tags:
- creative-writing
- storytelling
- children
- instruction-tuning
- chatml
size_categories:
- 100K<n<1M
π§ Children's Story Writing Dataset β¨
This dataset is a collection of creative short stories written for children. It is designed to help models learn child-friendly language and how to follow specific narrative instructions (e.g., incorporating specific features or sentences).
π Dataset Structure
The data is provided in ChatML format, making it ideal for instruction tuning.
Files
writing_train_children.jsonl: Training data.writing_valid_children.jsonl: Validation data.
Example Entry
{
"messages": [
{ "role": "system", "content": "You are a creative writer who writes short stories..." },
{ "role": "user", "content": "Features: Dialogue\nSummary: Tom and Anna go on holiday...\nRandom sentence: They are very excited..." },
{ "role": "assistant", "content": "Tom and Anna are brother and sister..." }
]
}
π License
This dataset is released under the Apache License 2.0.
Made with β€οΈ by Pink Pixel