8000 Impl simple mamba model by drbh · Pull Request #1480 · huggingface/text-generation-inference · GitHub

Impl simple mamba model#1480

Merged

Narsil merged 22 commits intomainfrom

impl-simple-mamba-model

Feb 8, 2024

Commits on Jan 23, 2024

feat: initial weight load
drbh
committed

Commits on Jan 25, 2024

Commits on Jan 29, 2024

Commits on Jan 30, 2024

feat: add optimization and first pass of integration test
drbh
committed

Commits on Feb 1, 2024

fix: start to add caching of previous states
drbh
committed

Commits on Feb 2, 2024

feat: use cache when decoding
drbh
committed

Commits on Feb 3, 2024

fix: revise non batching tests
drbh
committed

Commits on Feb 5, 2024

feat: avoid triton selective_state_update
drbh
committed

Commits on Feb 7, 2024

Commits on Feb 8, 2024

feat: conditionally include mamba
drbh
committed

0