feat: Hebrew layerdrop replication spec + logit-path stitch wiring - #173
Merged
Conversation
…eplication Item 04: the Arabic width/depth finding replicated on Hebrew — encoder 12->6 verbatim layer copy from pretrained ByT5-small, single variable vs run-002-s46 (same s46 teacher, hebrew-v4 corpus, 3 epochs). Wires _maybe_stitch into the logit-KD custom-config branch so depth-cut students keep the pretrained init there too (previously random-init, which the capacity law rejects).
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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
heb-diac-small-s46-layerdrop(item 04, TODO.publish-client):cross-lingual replication of the depth-cut finding — encoder 12->6
verbatim copy, single variable vs run-002-s46
_maybe_stitchnow runs in the logit-KD custom-configbranch when
layer_dropis set. Previously only the sequence pathstitched; a layerdrop spec on the logit path would have trained
random-init and collapsed by the capacity law
Verification
[spec] layer_drop_state from google/byt5-smallline is the observable that the pretrained init engaged