Shifting labels to enable learning next token prediction #10
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
LiorZ
changed the title
|
@LiorZ hey Lior it should be shifted already https://github.com/lucidrains/perceiver-ar-pytorch/blob/main/perceiver_ar_pytorch/autoregressive_wrapper.py#L83 |
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Outdated suggestions cannot be applied.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.
Dear lucidrains,
Many thanks for your amazing contribution to the community.
In the following pull request I modified the code so that the labels to the right during training before calculating the loss to enable next-token prediction (otherwise, the loss is calculated for the token, which leads to learning the trivial, current token)
L