Adding `safetensors` variant of this model

#6
No description provided.
NLP Group of The University of Hong Kong org

By adding safetensors variant of the INSTRUCTOR model, do you mean adding a function for saving embeddings in safetensors?

Hello,

Sorry for the lack of context on this.

The PR was automatically generated by the process here: https://huggingface.co/docs/diffusers/using-diffusers/using_safetensors#i-want-to-use-model-x-but-it-doesnt-have-safetensors-weights

The purpose is to have the weights available in a format that's better from a security standpoint than pickle.

See e.g. here https://huggingface.co/bigscience/bloom/tree/main

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment