Update README.md

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

README.md

@@ -1,5 +1,5 @@
 ---
-pipeline_tag: sentence-similarity
+pipeline_tag: feature-extraction
 tags:
 - sentence-transformers
 - feature-extraction
@@ -297,4 +297,4 @@ If you find this repository useful, please consider giving a star :star: and cit
       archivePrefix={arXiv},
       primaryClass={cs.CL}
 }
-```
+```