Change pipeline to feature extraction since it's more useful to get the embeddings for a sentence than to compare similarity.
Also we are able to use the serverless endpoint to get the embeddings

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

Sign up or log in to comment