Kyrgyz Alpaca

This repo is made for research use only, i.e., cannot be used for commercial purposes or entertainment.

References

All of our achievements were made achievable thanks to the robust AI community in Kyrgyzstan and the contributions made by individuals within the AkylAI project (by TheCramer.com). We also express our gratitude to Stanford for their outstanding efforts and extend the accessibility of this dataset to a global audience.

Dataset

Kyrgyz Alpaca can be also downloaded from here.

We used ChatGPT and Google Translate to convert alpaca_data.json into Kyrgyz. Although the translation wasn't perfect, we found it to strike a reasonable balance between cost and quality. The total cost for translating the entire dataset into Kyrgyz was approximately $700.00. If you're interested in learning more about the dataset's creation process, you can visit the Stanford Alpaca page.

Next

We work with Kyrgyz linguists to improve the quality of the translation.

Please feel free to reach out timur.turat@gmail.com if you are interested in any forms of collaborations!

Citation

If you use the data or code from this repo, please cite this repo as follows

@misc{kyrgyz-alpaca,
  author = {Khakim Davurov, Timur Turatali, Ulan Abdurazakov},
  title = {Kyrgyz Alpaca: Models and Datasets},
  year = {2023},
  publisher = {GitHub},
  journal = {GitHub repository},
  howpublished = {\url{https://github.com/Akyl-AI/kyrgyz-alpaca}},
}