This adapter was created with the
PEFT library and allowed the base model
BigScience/BLOOMz 7B1 to be fine-tuned on the
Dolly's Dataset (tanslated to Spanish, French and German by Argilla) by using the method
LoRA.
Instruction Tuned version of BigScience Large Open-science Open-access Multilingual.
This collection of datasets are machine-translated (and soon curated) versions of the
databricks-dolly-15k dataset originally created by Databricks, Inc. in 2023.
The goal is to give practitioners a starting point for training open-source instruction-following models beyond English. However, as the translation quality will not be perfect, we highly recommend dedicating time to curate and fix translation issues. Below we explain how to load the datasets into
Argilla for data curation and fixing. Additionally, we'll be improving the datasets made available here, with the help of different communities.
