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

WIP: Add Support for Gradient Checkpointing #759

Draft
wants to merge 4 commits into
base: main
Choose a base branch
from
Draft

WIP: Add Support for Gradient Checkpointing #759

wants to merge 4 commits into from

Commits on Nov 7, 2024

  1. Fix gradient checkpointing and write tests 

    - oerwrite the gradient_checkpointing_enable to provide our ForwardContext during the recomputation of values during backpropagation
- 2 bugs remaining: bottleneck adapter for models with the legacy implementation (BERT) & Parallel. Parallel has the problem that we manipulate the batch dimension and this currently leads to an error
    @lenglaender
    lenglaender committed Nov 7, 2024
    Configuration menu
    Copy the full SHA
    b5ae0fc View commit details
    Browse the repository at this point in the history

Commits on Nov 11, 2024

  1. minor fix but doesn't resolve the remaining issues

    @lenglaender
    lenglaender committed Nov 11, 2024
    Configuration menu
    Copy the full SHA
    cb07dd4 View commit details
    Browse the repository at this point in the history

Commits on Nov 24, 2024

  1. Only run adjust_tensors_for_parallel_ if bsz is different

    @calpt
    calpt committed Nov 24, 2024
    Configuration menu
    Copy the full SHA
    8421f63 View commit details
    Browse the repository at this point in the history

Commits on Nov 25, 2024

  1. remove parallel grad checkpointing test

    @calpt
    calpt committed Nov 25, 2024
    Configuration menu
    Copy the full SHA
    94df2fe View commit details
    Browse the repository at this point in the history