added geglu activation and tests #3512
Merged
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
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Outdated suggestions cannot be applied.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.
What does this PR do?
Fixes # (3483)
This PR adds the GeGLU (Gated Linear Unit with GELU) activation function to the activation module of the Flax library. It is an increasingly popular activation layer which combines a linear transformation with a GELU activation in a gating mechanism, balancing linearity and non-linearity.
GeGLU is parameterised and as such uses Flax's Dense layer and was not part of Jax itself. The implementation allows for an optional output_dim attribute, which can be used to specify the number of output features. This can increase or decrease the -1th dimensions of the results and the code tries to account for that. The tests validate the functionality in three scenarios: standard usage, output dimension expansion, and output dimension contraction, ensuring the layer's reliability in various use cases.
Checklist
discussion
documentation guidelines.
(No quality testing = no merge!)