README.md

metadata

language: multilingual
tags:
  - DistilBert
  - Universal Sentence Encoder
  - sentence-embeddings
  - sentence-transformers
  - sentence-similarity
license: apache-2.0

Knowledge distilled version of multilingual Universal Sentence Encoder. Supports 15 languages: Arabic, Chinese, Dutch, English, French, German, Italian, Korean, Polish, Portuguese, Russian, Spanish, Turkish.

This Model is saved from 'distiluse-base-multilingual-cased-v1' in sentence-transformers, to be used directly from transformers

Note that ST has additional two layers(Pooling, Linear), that cannot be saved in any predefined model in HG.