Each pair includes a Vedic Sanskrit source text and a reference translation. In the current release, the target language is English for avs and ts, and German for rv.
Dataset Structure
Data… See the full description on the dataset page: https://huggingface.co/datasets/yzk/veda-samhita-translation.