Datasets:
metadata
license: mit
task_categories:
- question-answering
- text-generation
language:
- mk
Dataset Summary
This dataset consists of multi-turn chat conversations translated from the original Ultra-Chat dataset using the Google Translate API.
Note: This is a sample dataset (rougly ~16k samples), as translating the entire dataset would be costly.