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

Context Parallel SFT Support for dataset in THD format #10688

Open
wants to merge 23 commits into
base: main
Choose a base branch
from
Open

Context Parallel SFT Support for dataset in THD format #10688

wants to merge 23 commits into from
+142 −26

Commits on Aug 19, 2024

  1. Add context parallel support for packed dataset in THD format

    @tomlifu
    tomlifu committed Aug 19, 2024
    Configuration menu
    Copy the full SHA
    9a2de52 View commit details
    Browse the repository at this point in the history

Commits on Aug 21, 2024

  1. Merge remote-tracking branch 'origin/main' into thd_cp_support

    @tomlifu
    tomlifu committed Aug 21, 2024
    Configuration menu
    Copy the full SHA
    b1ef8f0 View commit details
    Browse the repository at this point in the history

Commits on Sep 30, 2024

  1. add changes with debug print

    @tomlifu
    tomlifu committed Sep 30, 2024
    Configuration menu
    Copy the full SHA
    a11c351 View commit details
    Browse the repository at this point in the history

  2. remove debug print 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Sep 30, 2024
    Configuration menu
    Copy the full SHA
    2fd0456 View commit details
    Browse the repository at this point in the history

  3. Merge branch 'main' into thd_cp_support

    @tomlifu
    tomlifu authored Sep 30, 2024
    Configuration menu
    Copy the full SHA
    cf1a88d View commit details
    Browse the repository at this point in the history

  4. Apply isort and black reformatting 

    Signed-off-by: tomlifu <[email protected]>
    @tomlifu
    tomlifu committed Sep 30, 2024
    Configuration menu
    Copy the full SHA
    4ad3511 View commit details
    Browse the repository at this point in the history

Commits on Oct 1, 2024

  1. Merge branch 'main' into thd_cp_support

    @tomlifu
    tomlifu authored Oct 1, 2024
    Configuration menu
    Copy the full SHA
    43d60a3 View commit details
    Browse the repository at this point in the history

Commits on Oct 9, 2024

  1. fix cu_seqlens and cu_seqlens_padded 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 9, 2024
    Configuration menu
    Copy the full SHA
    63447f7 View commit details
    Browse the repository at this point in the history

  2. Merge branch 'thd_cp_support' of https://github.com/tomlifu/NeMo into… 

    … thd_cp_support
    @tomlifu
    tomlifu committed Oct 9, 2024
    Configuration menu
    Copy the full SHA
    e287857 View commit details
    Browse the repository at this point in the history

  3. cu_seqlens and cu_seqlens_padded fix 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 9, 2024
    Configuration menu
    Copy the full SHA
    850d9ae View commit details
    Browse the repository at this point in the history

Commits on Oct 12, 2024

  1. more fix on cu_seqlens and cu_seqlens_padded 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 12, 2024
    Configuration menu
    Copy the full SHA
    d50d88e View commit details
    Browse the repository at this point in the history

  2. Apply isort and black reformatting 

    Signed-off-by: tomlifu <[email protected]>
    @tomlifu
    tomlifu committed Oct 12, 2024
    Configuration menu
    Copy the full SHA
    adea017 View commit details
    Browse the repository at this point in the history

Commits on Oct 21, 2024

  1. addressing Xiaowei's review 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 21, 2024
    Configuration menu
    Copy the full SHA
    8c76e48 View commit details
    Browse the repository at this point in the history

Commits on Oct 23, 2024

  1. addressing more review comments 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 23, 2024
    Configuration menu
    Copy the full SHA
    78b3b1c View commit details
    Browse the repository at this point in the history

Commits on Oct 25, 2024

  1. fix for the case where cp=1 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 25, 2024
    Configuration menu
    Copy the full SHA
    8852431 View commit details
    Browse the repository at this point in the history

  2. Apply isort and black reformatting 

    Signed-off-by: tomlifu <[email protected]>
    @tomlifu
    tomlifu committed Oct 25, 2024
    Configuration menu
    Copy the full SHA
    a4bbb20 View commit details
    Browse the repository at this point in the history

Commits on Oct 26, 2024

  1. more fix to address Xiaowei's comment 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Oct 26, 2024
    Configuration menu
    Copy the full SHA
    bcda0db View commit details
    Browse the repository at this point in the history

Commits on Nov 1, 2024

  1. fix the loss_mask for THD formated data 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Nov 1, 2024
    Configuration menu
    Copy the full SHA
    aa59854 View commit details
    Browse the repository at this point in the history

Commits on Nov 7, 2024

  1. Apply isort and black reformatting 

    Signed-off-by: tomlifu <[email protected]>
    @tomlifu
    tomlifu committed Nov 7, 2024
    Configuration menu
    Copy the full SHA
    ab02643 View commit details
    Browse the repository at this point in the history

Commits on Nov 8, 2024

  1. fixed eos_idx[0][0] out of bounds issue 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Nov 8, 2024
    Configuration menu
    Copy the full SHA
    2a8b21f View commit details
    Browse the repository at this point in the history

  2. Merge branch 'thd_cp_support' of https://github.com/tomlifu/NeMo into… 

    … thd_cp_support
    @tomlifu
    tomlifu committed Nov 8, 2024
    Configuration menu
    Copy the full SHA
    f930f77 View commit details
    Browse the repository at this point in the history

  3. Merge branch 'main' into thd_cp_support

    @tomlifu
    tomlifu authored Nov 8, 2024
    Configuration menu
    Copy the full SHA
    d3e9354 View commit details
    Browse the repository at this point in the history

Commits on Nov 9, 2024

  1. fixed CP=1 case 

    Signed-off-by: Lifu Zhang <[email protected]>
    @tomlifu
    tomlifu committed Nov 9, 2024
    Configuration menu
    Copy the full SHA
    02bccd7 View commit details
    Browse the repository at this point in the history