feat(nn): make Linear open for adapter subclasses#875
Merged
Conversation
Marks Linear as open so adapter-style layers (LoRA and similar) can subclass it and augment onForward or params while reusing the base projection. No behavioral change; the override points (onForward, params) were already open via Module.
|
📖 Documentation Preview The documentation has been built successfully for this PR. Generated Files:
Artifacts:
This comment will be updated automatically when the PR is updated. |
aharakal
approved these changes
Jul 24, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
Marks
Linearasopenso adapter-style layers can subclass it — the LoRA pattern (class LinearWithLoRA : Linear) that augmentsonForwardwith a low-rank update andparamswith the A/B matrices, while reusing the frozen base projection.No behavioral change and no new members: the override points (
onForward,params) were already open viaModule/ModuleParameters; only the class itself was final. API dumps updated (the now-visibleprotected onForwardsurfaces in the dump because the class is subclassable).Motivation
Completes the toolkit for parameter-efficient finetuning on top of SKaiNET: with bias-less
Linear(#870) and this change, an educational GPT port can express LoRA as a small subclass instead of reimplementing the layer. Follows the same from-scratch-then-upstream pattern as #866 (LR schedules), #867 (Dropout), #870 (bias-less Linear).