mamba-constrict-l2 / README.md
teilers's picture
Update README.md
64b7ce2 verified
metadata
library_name: transformers
datasets:
  - monology/pile-uncopyrighted
base_model:
  - state-spaces/mamba-2.8b-hf
pipeline_tag: text-generation

MambaConstrict

Best performing MambaConstrict model trained using $\ell_2$-norm regularization.

For inquiring, please contact teilers@student.ethz.ch