Conversation
Contributor
Author
|
I also see that |
Contributor
Author
|
Actually, since I ended up not using them there let's make it simple and remove them from that PR and add them here. |
This is identical behaviour to llama.cpp I guess any model that doesn't use BOS is recent enough to have the add_bos_token metadata.
Contributor
Author
|
My Codestral GGUFs are up for those who wish to test with this option. NOTE: It requires ggml-org/llama.cpp#7644 to not insert garbage |
Owner
|
@CISC looks good overall, do you mind adding something minimal to |
Contributor
Author
|
Sure, no problem. |
Contributor
Author
|
@abetlen Example added. |
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Add
spm_infilloption to perform infill in the Suffix/Prefix/Middle pattern instead of Prefix/Suffix/Middle as some models (like the newCodestral) prefer this.Also added a tokenizer hack to remove leading space in suffix to improve inference, tested on several different tokenizers/vocabs with success.