README.md

metadata

pipeline_tag: sentence-similarity
tags:
  - sentence-transformers
  - feature-extraction
  - sentence-similarity

multi-qa_v1-mpnet-mean_cos

This is a microsoft/mpnet-base model trained on all the Q&A datasets of the 1B+ train corpus. It was trained with the v1 setup. See data_config.json and train_script.py in this respository how the model was trained and which datasets have been used.

Usage

It can be used for semantic search. Output vectors are normalized. You can find relevant passages by using dot-product or cosine-similarity, both will return equivalent results.