Datasets:

Modalities:
Text
Formats:
csv
Size:
< 1K
Libraries:
Datasets
pandas
License:
AMR-KELEG's picture
Update README.md
5e14254 verified
---
license: apache-2.0
---
The samples of the [Aya dataset](https://huggingface.co/datasets/CohereForAI/aya_dataset) with mislabeled inputs/targets.
Categories of mislabeling (each represented as a new column):
* `instruction_not_followed`
* `incorrect_language_code`
* ...