Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Flash attention support. #20152

Merged
merged 13 commits into from
Oct 8, 2024
Merged

Flash attention support. #20152

merged 13 commits into from
Oct 8, 2024

Commits on Aug 22, 2024

  1. added flash attention support for pytorch

    @hazemessamm
    hazemessamm committed Aug 22, 2024
    Configuration menu
    Copy the full SHA
    22f47b1 View commit details
    Browse the repository at this point in the history

  2. added a comment explaining why the causal mask is created manually

    @hazemessamm
    hazemessamm committed Aug 22, 2024
    Configuration menu
    Copy the full SHA
    7e99f06 View commit details
    Browse the repository at this point in the history

Commits on Oct 3, 2024

  1. Merge branch 'master' into flash_attention

    @hazemessamm
    hazemessamm authored Oct 3, 2024
    Configuration menu
    Copy the full SHA
    32a74e0 View commit details
    Browse the repository at this point in the history

Commits on Oct 4, 2024

  1. added unit tests for flash attention

    @hazemessamm
    hazemessamm committed Oct 4, 2024
    Configuration menu
    Copy the full SHA
    f3a05b0 View commit details
    Browse the repository at this point in the history

  2. added test skipping for flash attention for numpy

    @hazemessamm
    hazemessamm committed Oct 4, 2024
    Configuration menu
    Copy the full SHA
    f585859 View commit details
    Browse the repository at this point in the history

Commits on Oct 6, 2024

  1. removed flash attn op and added support for flash attention inside do… 

    …t_product_attention op
    @hazemessamm
    hazemessamm committed Oct 6, 2024
    Configuration menu
    Copy the full SHA
    e614e75 View commit details
    Browse the repository at this point in the history

  2. added skip tests for every framework except torch

    @hazemessamm
    hazemessamm committed Oct 6, 2024
    Configuration menu
    Copy the full SHA
    561a9a7 View commit details
    Browse the repository at this point in the history

  3. formatted files

    @hazemessamm
    hazemessamm committed Oct 6, 2024
    Configuration menu
    Copy the full SHA
    86db62a View commit details
    Browse the repository at this point in the history

Commits on Oct 7, 2024

  1. added checks for flash attention in pytorch beforing computing attent… 

    …ion and removed flash attention from tests
    @hazemessamm
    hazemessamm committed Oct 7, 2024
    Configuration menu
    Copy the full SHA
    43938ca View commit details
    Browse the repository at this point in the history

  2. added skipping tests for all frameworks except jax

    @hazemessamm
    hazemessamm committed Oct 7, 2024
    Configuration menu
    Copy the full SHA
    738375f View commit details
    Browse the repository at this point in the history

  3. formatted files

    @hazemessamm
    hazemessamm committed Oct 7, 2024
    Configuration menu
    Copy the full SHA
    e12950c View commit details
    Browse the repository at this point in the history

  4. added conditions to skip tests for jax

    @hazemessamm
    hazemessamm committed Oct 7, 2024
    Configuration menu
    Copy the full SHA
    da50a6b View commit details
    Browse the repository at this point in the history

  5. fixed typo

    @hazemessamm
    hazemessamm committed Oct 7, 2024
    Configuration menu
    Copy the full SHA
    57e6e56 View commit details
    Browse the repository at this point in the history