Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

how to tokennizer vicuna #20

Open
beyondli opened this issue Nov 20, 2023 · 2 comments
Open

how to tokennizer vicuna #20

beyondli opened this issue Nov 20, 2023 · 2 comments

Comments

@beyondli
Copy link

Hi ,
I want to test the token speed of minigpt4, but tokenizer failed
AutoTokenizer.from_pretrained('maknee/ggml-vicuna-v0-quantized/13B') or

huggingface_hub.utils._validators.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': 'maknee/ggml-vicuna-v0-quantized/13B'. Use repo_type argument if needed.

AutoTokenizer.from_pretrained('maknee/ggml-vicuna-v0-quantized') both failed.

Repository Not Found for url: https://huggingface.co/maknee/ggml-vicuna-v0-quantized/resolve/main/tokenizer_config.json.
Please make sure you specified the correct repo_id and repo_type.

what is correct command for tokennizer? thanks

@Maknee
Copy link
Owner

Maknee commented Nov 23, 2023

The tokenizer used is the llama.cpp tokenizer. Call add_strings in c++ and set a timer to eval the speed. Unfortunately, you can't use AutoTokenizer from hugging face to directly load the tokenizer :(

The tokenizer internally calls llama_tokenize_internal in llama.cpp

@beyondli
Copy link
Author

Hi Maknee,
Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants