-
Notifications
You must be signed in to change notification settings - Fork 149
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.
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
Forkless transformers [4.34.1] #2199
Conversation
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Nice job! Appreciate the modification tests and just have a few comments
src/sparseml/transformers/sparsification/modification/modify_model.py
Outdated
Show resolved
Hide resolved
symmetric: False""" | ||
|
||
|
||
def test_modifying_llama(llama_model, shared_helper_functions): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We should get to the point where we can test making a modified model and an unmodified model, and then run through them comparing outputs. This may help catch the forward passes changing between transformers versions. This may be too intensive for current hardware, but I think should absolutely be setup with the gpu runners - thoughts?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Good idea. This is a big step toward robust detection of problems that arise from the transformers upgrade.
…/sparseml into feature/damian/forkless_4.34
Co-authored-by: Michael Goin <[email protected]>
A copy of this PR: #2046, but instead of introducing
transformers<3.37
, we installtransformers==3.34.1
. This is the same transformer version as in our fork.