feat(template): read jinja templates from gguf files #4332
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
Signed-off-by: Ettore Di Giacinto <[email protected]>
![github-advanced-security[bot]](https://avatars.githubusercontent.com/in/57789?s=60&v=4){kind=small}
✅ Deploy Preview for localai ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
WIP as need to still add mapping between the transformer tokenizer and the templates (see TODO note in code comments) |
Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Ettore Di Giacinto <[email protected]>
Signed-off-by: Ettore Di Giacinto <[email protected]>
|
basic support should work (tested with llama3 prompt), probably is not going to cover all cases as gonja has limitations, but, since this kicks-in when no other template was defined it is safe to merge without drawbacks. |
sozercan
pushed a commit
to sozercan/LocalAI
that referenced
this pull request
Dec 8, 2024
* Read jinja templates as fallback Signed-off-by: Ettore Di Giacinto <[email protected]> * Move templating out of model loader Signed-off-by: Ettore Di Giacinto <[email protected]> * Test TemplateMessages Signed-off-by: Ettore Di Giacinto <[email protected]> * Set role and content from transformers Signed-off-by: Ettore Di Giacinto <[email protected]> * Tests: be more flexible Signed-off-by: Ettore Di Giacinto <[email protected]> * More jinja Signed-off-by: Ettore Di Giacinto <[email protected]> * Small refactoring and adaptations Signed-off-by: Ettore Di Giacinto <[email protected]> --------- Signed-off-by: Ettore Di Giacinto <[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.
Description
This PR adds automatic detection and parsing of jinja templates in gguf files. If we fail to identify a variant and we do not have already a specific template, it injects the jinja templates which is part of the model metadata if one is found.
Alternatively, it is possible to enable jinja templates manually in the model config file, in the template config section with
jinja_template: true.Notes for Reviewers
Signed commits