Alibaba-NLP
/

new-impl

Model card Files Files and versions

Is flash-attention-2 suppported

#8

by Jack7777777 - opened Sep 5, 2024

See it been commented out now: https://huggingface.co/Alibaba-NLP/new-impl/blob/main/modeling.py#L592

Alibaba-NLP org Nov 28, 2024

Hi, xformers could dispatch to flash-attention-2 kernel. Hence I commented out this extra entrance.

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment