File size: 848 Bytes
e40de56
d296d9c
40343d8
d296d9c
e40de56
 
 
40343d8
 
d296d9c
 
 
e40de56
 
40343d8
8367e31
 
 
 
 
 
 
 
 
 
40343d8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
---
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