datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Malum-230
Malum-230
Description
Malum-230 is a meticulously handcrafted Japanese dataset featuring multi-turn conversations and passages, specifically designed for logical reasoning tasks.
This dataset can be used for both pre-training and post-training.
Details
Creation method: Human effort
Dataset type: Logical reasoning
Use case: pre-training and post-training
Performance
This radar chart shows the evaluation results on Japanese MT-Bench for the… See the full description on the dataset page: https://huggingface.co/datasets/Manual-Dataset-Creation-Project/Malum-230.creation_dataset_chatbottask_creation_dataset