bert-base-cased model trained on quora question pair dataset. The model achieves 89% evaluation accuracy
