|
--- |
|
language: |
|
- pt |
|
pretty_name: Aya Dataset Portuguese |
|
tags: |
|
- aya |
|
- portuguese |
|
- legal |
|
- chemistry |
|
license: apache-2.0 |
|
size_categories: |
|
- 1K<n<10K |
|
--- |
|
|
|
CohereForAI [Aya Dataset](https://huggingface.co/datasets/CohereForAI/aya_dataset) filtrado para português (PT). |
|
|
|
**Aya Dataset Summary** |
|
|
|
The [Aya Dataset](https://huggingface.co/datasets/CohereForAI/aya_dataset) is a multilingual instruction fine-tuning dataset curated by an open-science community via Aya Annotation Platform from Cohere For AI. The dataset contains a total of 204k human-annotated prompt-completion pairs along with the demographics data of the annotators. |
|
This dataset can be used to train, finetune, and evaluate multilingual LLMs. |
|
|
|
Curated by: Contributors of Aya Open Science Intiative. |
|
|
|
Language(s): 65 languages (71 including dialects & scripts). |
|
|
|
License: Apache 2.0 |
