Fix ANKH tokenizer to load from checkpoint hub id#31
Open
avivko wants to merge 1 commit into
Open
Conversation
FAST_ANKH_ENCODER always attached ElnaggarLab/ankh-base, so ANKH3 checkpoints (256-token vocab) were tokenized with the wrong ids in embed.py. Load from config._name_or_path (same pattern as DPLM) with ankh-base fallback for bare configs. Add slow tests that fast tokenizer ids match the official repo for ANKH_base, ANKH3_large, and ANKH3_xl. Co-authored-by: Cursor <cursoragent@cursor.com>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
FAST_ANKH_ENCODER always attached ElnaggarLab/ankh-base, so ANKH3 checkpoints (256-token vocab) were tokenized with the wrong ids in embed.py. Load from config._name_or_path (same pattern as DPLM) with ankh-base fallback for bare configs.
Add slow tests that fast tokenizer ids match the official repo for ANKH_base, ANKH3_large, and ANKH3_xl.
Summary
FAST_ANKH_ENCODERno longer hardcodesElnaggarLab/ankh-base; it loads the tokenizer fromconfig._name_or_path(e.g.Synthyra/ANKH3_large), matching the checkpoint vocab.embed.pyfor ANKH3 large/XL.Hub follow-up
After merge, re-push
modeling_ankh.pyto Synthyra ANKH checkpoints (e.g. viaget_weights.py) sotrust_remote_code=Trueusers pick up the fix without a local checkout.Test plan
pytest testing/test_ankh_tokenizer.py -v(Docker, 6 passed)