fix: implement Viterbi SentencePiece encoding (replaces greedy)#3
Merged
fix: implement Viterbi SentencePiece encoding (replaces greedy)#3
Conversation
The greedy longest-match approach in sentencePieceEncode produced suboptimal tokenization for SentencePiece unigram models (e.g., Mistral 7B). Replace it with Viterbi dynamic programming that finds the segmentation maximizing the sum of log-probability scores. Also adds: - Byte fallback encoding/decoding via <0xNN> tokens for chars not in vocab - decodeSentencePieceBytes for proper round-trip of byte fallback tokens - Tests: Viterbi vs greedy, byte fallback, sentence round-trip, edge cases
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Viterbi DP for globally optimal tokenization. Byte fallback via <0xNN>. Fixes Mistral garbage output. 6 new tests.