orai-nlp/ARC_HT_eu_sample

---
language:
- eu
- en
configs:
- config_name: eu
data_files:
- split: test
path: arc_eu_250.json
- config_name: en
data_files:
- split: test
path: arc_en_250.json
license: apache-2.0
size_categories:
- n<1K
---

# ARC Human Translated Sample for Basque

A subset of 250 samples manually translated to Basque from the ARC dataset (Clark et al., 2018).
The corresponding 250 English samples are also provided.
The ARC dataset consists of genuine grade-school level, multiple-choice science questions, assembled to encourage research in advanced question-answering.

### Dataset Description
- **Curated by:** [Orai NLP Technologies](https://orai.eus/en)
- **Language(s) (NLP):** Basque and the corresponding English samples
- **License:** Apache 2.0

## Dataset Creation

### Source Data

A subset of 250 samples manually translated to Basque from the ARC dataset (Clark et al., 2018).

### Annotations

#### Annotation process

A subset of 250 samples manually translated to Basque from the ARC dataset (Clark et al., 2018).

## Citation [optional]

If you use this dataset please cite the following reference:

```bibtex
@misc{Llama-eus,
title = {Llama-eus-8B, a foundational sub-10 billion parameter LLM for Basque},
author = {Ander Corral, Ixak Sarasua and Xabier Saralegi},
publisher = {Orai NLP Technologies},
url = {\url{https://huggingface.co/datasets/orai-nlp/Llama-eus-8B}},
year = 2024 }
```

## Contact

- Ander Corral ([e-mail])
- Ixak Sarasua ([e-mail])
- Xabier Saralegi ([e-mail])