The beta-glycine dataset is created with the purpose of validating the electron machine learning potential (eMLP) on crystalline beta glycine. It contains 25,676 configurations with normal mode perturbations for the nuclei and unit cell and electric field perturbations. Energies, forces and Wannier centers are computed using density functional theory (DFT) with the PBE functional and a Plane-Wave basis set in the ab-initio quantum chemistry code QuantumESPRESSO.