Llama inference #2144
Merged
Llama inference #2144
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
| if __name__ == "__main__": | ||
| model_id = os.getenv("MODEL_ID") | ||
| if "llama" in model_id: | ||
| transformers.LlamaTokenizer.from_pretrained(model_id) |
There was a problem hiding this comment.
I saw llama was merged in huggingface transformers .. I will test today if this is really necessary or if we can use the Auto functions now .. also we have to try if there were significant changes .. in worst case we need to retrain our models.
andreaskoepf
approved these changes
Mar 21, 2023
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Outdated suggestions cannot be applied.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.
some hacks here to get it to work. This goes in conjunction with a fork of tokenizers and a fork of the HF text generation server.
sentencepiece tokenizer doesn't play nicely with streaming, i.e. decoding individual tokens somehow is not very helpful and leads to artefacts, so the way around it is to always decode from the beginning