---
language:
- fo
- en
pretty_name: FO-EN Parallel
configs:
- config_name: default
  data_files:
  - data.jsonl
- config_name: ShareGPT_translations
  data_files:
  - sharegpt_translation.jsonl
license: other
task_categories:
- text-generation
---

A large-scale Faroese–English parallel dataset built from human translations, dictionary sources, and high-quality synthetic pairs.

Note that a fraction of this has been sourced from:
https://huggingface.co/datasets/barbaroo/Sprotin_parallel and 
https://huggingface.co/datasets/barbaroo/fo_en_synthetic