feat(decode): Add opt-in lossy mode and decode_bytes#123
Merged
Conversation
BPE tokens are byte-level, so a single character can span several tokens. The common "trim text to N tokens" pattern can leave a prefix that is not valid UTF-8, and decode raised Tiktoken::UnicodeError on it — diverging from Python tiktoken, whose decode defaults to errors="replace". Add an opt-in errors: :replace that substitutes invalid sequences with U+FFFD, plus decode_bytes for the raw bytes. The default stays :strict to preserve existing behavior. Also define Tiktoken::UnicodeError eagerly at load, since it previously only existed after the first decode error was raised. Closes #122
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
decoderaisedTiktoken::UnicodeErroron it.errors: :replacetodecode, which substitutes invalid byte sequences with the Unicode replacement character (U+FFFD,�). The default stays:strictto preserve existing behavior.decode_bytes(tokens), which returns the raw decoded bytes as anASCII-8BITstring with no UTF-8 validation (for callers who want to handle invalid sequences themselves)Closes #122