datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
nynorsk_norm_200eval
Nynorsk Norm 200eval
nynorsk_norm_200eval is a high-quality, small-scale parallel corpus comprising 200 Norwegian Bokmål–Nynorsk sentence pairs collected from official sources and public institutions. Each example includes:
nb: Original sentence in Bokmål
nn_original: Original Nynorsk sentence (typically an official translation)
nn_alt_original: Original Nynorsk sentence (typically an official translation) - alt version
nn_husnorm: Sentence rewritten in Nynorsk following an… See the full description on the dataset page: https://huggingface.co/datasets/NbAiLab/nynorsk_norm_200eval.nynorsk_dpo
Bokmål–Nynorsk DPO
bokmal_nynorsk_dpo is a dataset for Direct Preference Optimization (DPO) training, focusing on Bokmål–Nynorsk translation.Each example consists of a prompt in Norwegian Bokmål and two candidate translations in Nynorsk:
prompt: Input sentence in Bokmål
chosen: Preferred Nynorsk translation (higher quality, closer to target norm)
rejected: Less preferred Nynorsk translation
This format enables reinforcement learning from human preferences, where models learn… See the full description on the dataset page: https://huggingface.co/datasets/NbAiLab/nynorsk_dpo.