--- library_name: transformers datasets: - monology/pile-uncopyrighted base_model: - state-spaces/mamba-2.8b-hf pipeline_tag: text-generation --- # MambaConstrict (l2, lambda=0.001) MambaConstrict model trained using $\ell_2$-norm regularization with $\lambda = 0.001$. For inquiring, please contact teilers@student.ethz.ch