ludwigstumpp

/

alpaca-lora-german-7b

like 10

Hi Philipp,

thanks for your interest. Currently I am not planning to open-source the dataset. This is for two reasons:

1. As the translation has been done using the OpenAI GPT-3.5 Turbo API, I don't want to interfere with any of OpenAI restrictions of using their model.
2. During the translation, some of the rows of the dataset where lost due to some API timeout errors. Unfortunately I don't have a way to recover them as of right now, as I don't know which rows are affected. Therefore the dataset that I have been using is not 100% complete.

Once I have a little bit more time I will look into both points and might decide to open-source the translated dataset at a later point in time.