Skip to content

v0.1.2
Compare
Choose a tag to compare
@christopher-w-murphy christopher-w-murphy released this 26 Aug 18:38
· 37 commits to main since this release
v0.1.2
1a9de7a
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

The attention bias in MosaicBERT has attn_bias.ndim == 4, so I generalized flash_attention_n to accomodate this.

Assets 2
Loading