This is the repository containing LoRA checkpoints tuned on common sense reasoning datasets with 0.5B foundation model, served as training data for DnD.
