handled the case for T5 tokenizers#16
Open
hazemessamm wants to merge 1 commit intoNatLabRockies:mainfrom
Open
Conversation
pemami4911
reviewed
Oct 18, 2025
| # This checks whether the gradient sequence length | ||
| # is exactly one more than the input sequence length. | ||
| elif oh_grad.shape[1] == self.chains_oh.shape[1] + 1: | ||
| oh_grad = oh_grad[:, :-1] |
Collaborator
There was a problem hiding this comment.
I noticed that we can't guarantee that the last element of the gradient sequence is what should removed (why not the first element?)
Collaborator
|
Hi, thank you for this pull request! Having this logic (for instances with and tokens) in I'm thinking that we could move this logic out of I think we would need to also add custom expert code for T5/Ankh though! |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Hi, thanks for the great work!
I was working with some evaluation models that use ProtT5 and Ankh PLMs, and these models do not have a token; they only have token, so I handled their case in order to work properly with EvoProtGrad.
Decoded example for both ProtT5 and Ankh:
MQMLKMGLV</s>