Datasetpublic

Uunan/turkish-village-names

Turkish Village Names A dataset containing 11,104 unique Turkish village names. The dataset was prepared for character-level language modeling and text generation experiments, especially for training miniature language models capable of generating synthetic Turkish village names. Dataset Information Language: Turkish Entries: 11,104 Duplicates: Removed Encoding: UTF-8 Format: Plain text (.txt) One village name per line Example Akpınar Karadere… See the full description on the dataset page: https://huggingface.co/datasets/Uunan/turkish-village-names.

sourceHugging Facemitupdated 2mo agoView on Hugging Face

0likes10downloads

Dataset Card

Turkish Village Names

A dataset containing 11,104 unique Turkish village names.

The dataset was prepared for character-level language modeling and text generation experiments, especially for training miniature language models capable of generating synthetic Turkish village names.

Dataset Information

—Language: Turkish
—Entries: 11,104
—Duplicates: Removed
—Encoding: UTF-8
—Format: Plain text (.txt)
—One village name per line

Example

text

Akpınar
Karadere
Yeniköy
Boztepe
Çamlıca
Aşağıovacık
Yukarıkarakaya

Intended Use

This dataset is suitable for:

—Character-level language modeling
—Byte Pair Encoding (BPE) experiments
—Tokenizer training
—Tiny LLM training
—Educational NLP projects
—Turkish text generation

Source

The village names were extracted from the publicly available Turkish village list published by the Republic of Türkiye Ministry of Interior.

The dataset was cleaned by:

—Removing empty entries
—Removing duplicate village names
—Normalizing text encoding

License

MIT

This project is based on the educational repository:

https://github.com/Uunan/tiny-village-llms