fix: cap model_max_length sentinel value to prevent OverflowError in … - #206
Open
anaghasimha wants to merge 1 commit into
Open
fix: cap model_max_length sentinel value to prevent OverflowError in …#206anaghasimha wants to merge 1 commit into
anaghasimha wants to merge 1 commit into
Conversation
…Rust tokenizers backend Models like DeBERTa and DistilBERT don't define model_max_length, causing HuggingFace to assign VERY_LARGE_INTEGER (~1e30) as default. This overflows int32 when passed to the Rust tokenizers backend via enable_truncation(). Fixes Tiiiger#205
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
When using
bert_scorewith models that don't explicitly definemodel_max_length(e.g. DeBERTa, DistilBERT), HuggingFace assigns a sentinel value
(
VERY_LARGE_INTEGER~ 1e30) as the default. This gets passed to the Rusttokenizers backend and causes an
OverflowErrorsince it exceeds int32 bounds.Fix
Cap
model_max_lengthto 512 insent_encode()inutils.pybefore itreaches the Rust backend.
Related