Fix ChatGLM Detokenization #36
Merged
Conversation
This file contains 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
New special tokens was added to ChatGLM repository, that causes Sentencepiece to crash during decoding because of indices was not added to the main vocabulary (these tokens was not marked as special in the repository and were filtered out because of it). Include tokens to a vocab and also align vocab sizes better. Has to lower pass rate, because of ChatGLM3 decoder inserts spaces between special tokens and Sentencepiece does not. No functional difference between actual texts.
mryzhov
pushed a commit
to mryzhov/openvino_tokenizers_public
that referenced
this pull request
Mar 7, 2024
New special tokens was added to ChatGLM repository, that causes Sentencepiece to crash during decoding because of indices was not added to the main vocabulary (these tokens was not marked as special in the repository and were filtered out because of it). Include tokens to a vocab and also align vocab sizes better. Has to lower pass rate, because of ChatGLM3 decoder inserts spaces between special tokens and Sentencepiece does not. No functional difference between actual texts.
mryzhov
added a commit
that referenced
this pull request
Mar 13, 2024
* master fixes * Fix ChatGLM Detokenization (#36) New special tokens was added to ChatGLM repository, that causes Sentencepiece to crash during decoding because of indices was not added to the main vocabulary (these tokens was not marked as special in the repository and were filtered out because of it). Include tokens to a vocab and also align vocab sizes better. Has to lower pass rate, because of ChatGLM3 decoder inserts spaces between special tokens and Sentencepiece does not. No functional difference between actual texts. * updated test results --------- Co-authored-by: Artur Paniukov <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
New special tokens were added to ChatGLM repository, that caused Sentencepiece to crash during decoding because of indices were not added to the main vocabulary (these tokens were not marked as special in the repository and were filtered out because of it). Include tokens to a vocab and also align vocab sizes better.
Has to lower pass-rate, because of ChatGLM3 decoder inserts spaces between special tokens and Sentencepiece does not. No functional difference between actual texts.