common : allow raw byte in SPM vocabs; don't crash if newline token is not found #5478
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
akx
changed the title
ggerganov
reviewed
Feb 13, 2024
Comment on lines
+7722
to
+7760
| // Try to fall back to just the byte as a string | ||
| const char buf2[2] = { (char)ch, 0 }; | ||
| return vocab.token_to_id.at(buf2); |
There was a problem hiding this comment.
Does this fallback work for the Finnish model that you are trying?
There was a problem hiding this comment.
Yep! @R4ZZ3, who's been involved with training the model, pointed out that tokenizer.json has
{
"id": 64261,
"content": "\n",
"single_word": false,
"lstrip": false,
"rstrip": false,
"normalized": false,
"special": true
}
ggerganov
reviewed
Feb 13, 2024
akx
force-pushed
the
spm-no-linefeed-hack
branch
from
dce9bc9 to
72b353f
Compare
ggerganov
approved these changes
Feb 13, 2024
jordankanter
pushed a commit
to jordankanter/llama.cpp
that referenced
this pull request
Mar 13, 2024
…#5478) * common : don't crash if newline token is not found * common : llama_byte_to_token: allow falling back to finding just the token byte in SPM vocabs
hodlen
pushed a commit
to hodlen/llama.cpp
that referenced
this pull request
Apr 1, 2024
…#5478) * common : don't crash if newline token is not found * common : llama_byte_to_token: allow falling back to finding just the token byte in SPM vocabs
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Outdated suggestions cannot be applied.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.
This makes e.g. a GGUF quantization of Finnish-NLP/llama-7b-finnish-instruct-v0.2 not crash while loading.
Sure, it would be better if the model's vocabulary had the
<0x0A>token to begin with...#5477 is related (written while I was hunting this down).