Revert "Error (also in original) model, scaling only q matrix not qk.T dot product (qk.T/sqrt(dim_per_head))" #22444
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. |
raghavanone
pushed a commit
to raghavanone/transformers
that referenced
this pull request
Apr 5, 2023
…T dot product (qk.T/sqrt(dim_per_head))" (huggingface#22444) Revert "Error (also in original) model, scaling only q matrix not qk.T dot product (qk.T/sqrt(dim_per_head)) (huggingface#21627)" This reverts commit bad8300.
4 tasks
novice03
pushed a commit
to novice03/transformers
that referenced
this pull request
Jun 23, 2023
…T dot product (qk.T/sqrt(dim_per_head))" (huggingface#22444) Revert "Error (also in original) model, scaling only q matrix not qk.T dot product (qk.T/sqrt(dim_per_head)) (huggingface#21627)" This reverts commit bad8300.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Reverts #21627
This PR changed the modeling code and as a result broke training as reported in #22426
Moreover, since the "error" was in the original code from the authors, it's not an error anymore but a feature.
This will be put in the patch v4.27.4