metadata
task_categories:
- text2text-generation
language:
- it
tags:
- synthetic
size_categories:
- 1K<n<10K
Beatrice
Introduction
The following dataset is generated using gpt-3.5-turbo and gpt-4 and is a tiny reproduction of the Samantha Dataset (not translated from english).
The data contains specifically italians topics and domains including some specific ReDiX knowledge, like foundation, history etc..
Purpose
This dataset is being used as training data for ReDiX 7B Model to teach conversational skills in italian. This dataset is under active revision and may currently contain harmful and/or wrong content.