Started with bert-base-nli-stsb-mean-tokens Trained using sentence transformers Continue training of quora dataset See quora.py for script used