transformers
390f1534 - Add padding-free to bamba (#35861)

Commit
343 days ago
Add padding-free to bamba (#35861) * add seq_idx and fa kwargs * update tests * docs and grad ckpt support * fmt * better names * test_raise_missing_padding_free_kwarg_errs * + seq_idx in doc strings * padding free training docs * add link to pr plots * raise err on attn_mask with padding free * rm raising missing padding free err test * BambaFlashAttentionKwargs * run modular util for modular_granitemoehybrid.py
Author
Parents
Loading