datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
zulu-self_instruct-syn
Dataset Card for Smoltalk-Zulu-Syn
This dataset card contains details for the Smoltalk-Zulu-Syn isiZulu dataset.
You can load a dataset using
from datasets import load_dataset
ds = load_dataset("2b2e/zulu-self_instruct-syn")
Dataset Details
Dataset Description
This dataset was created by prompting AfroLlama V1 model using the Self-Instruct methodology.
Uses
This dataset is intended as a instruction-tuning dataset.
Dataset Structure
This… See the full description on the dataset page: https://huggingface.co/datasets/2b2e/zulu-self_instruct-syn.smoltalk-zulu-mtl
Dataset Card for Smoltalk-Zulu-Mtl
This dataset card contains details for the Smoltalk-Zulu-Mtl isiZulu dataset.
You can load a dataset using
from datasets import load_dataset
ds = load_dataset("2b2e/smoltalk-zulu-mtl")
Dataset Details
Dataset Description
This dataset was created by translating the SmolTalk dataset into isiZulu using
the NLLB 3.3B translation model.
Uses
This dataset is intended as a instruction-tuning dataset.
Dataset… See the full description on the dataset page: https://huggingface.co/datasets/2b2e/smoltalk-zulu-mtl.zulu_mtl_syn_concat
Dataset Card for Smoltalk-Zulu-Syn
This dataset card contains details for the Smoltalk-Zulu-Syn isiZulu dataset.
You can load a dataset using
from datasets import load_dataset
ds = load_dataset("2b2e/zulu_mtl_syn_concat")
Dataset Details
Dataset Description
Uses
This dataset is intended as a instruction-tuning dataset.
Dataset Structure
This dataset follows the ChatML format.