tokenization: add warning for double BOS #7332
Merged
Conversation
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
reuank
added a commit
to reuank/ThinkBench
that referenced
this pull request
May 16, 2024
…the llama.cpp server in order to avoid double BOS at the beginning of the prompt (see discussions at ggml-org/llama.cpp#7107 (comment) and ggml-org/llama.cpp#7332)
mofosyne
added
enhancement
ggerganov
approved these changes
May 17, 2024
llama.cpp
Outdated
| } | ||
| } | ||
|
|
||
| if (add_special && vocab.special_add_bos != 0 && output[1] == vocab.special_bos_id) { |
There was a problem hiding this comment.
Think we should also check output.size() > 1
JohannesGaessler
force-pushed
the
tokenize-bos-warning
branch
from
c7f00fc to
6520c7a
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
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Outdated suggestions cannot be applied.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.
Alternative to #7107 .
This PR adds a warning when a BOS token is added automatically and this results in 2 BOS tokens at the beginning of the prompt since this is almost always going to be user error and not intentional.