Datasets for multilingual evaluation tasks

#3
by hartmutkeil - opened

Hello,
I have have a question regarding the datasets of the evaluation tasks:
o You are evaluating tasks original related to English datasets, (take ARC as an example) for different European languages.
o I was wondering if you have translated the datasets to these languages or have a created your own datasets?
o Are these datasets public available?
Thank you for supporting

OpenGPT-X org

Hi @hartmutkeil ,

thanks for raising the points!

  • We translated the datasets by using DeepL Pro
  • At the moment they are not publicly available.

With best regards
Michael

Sign up or log in to comment