prajjwal1/bert-small-mnli

These BERT variants were introduced in the paper Well-Read Students Learn Better: On the Importance of Pre-training Compact Models . These models are trained on MNLI: 72.1%. MNLI-mm: 73.76%. These models were trained for 4 epochs. They were trained on the MNLI.MNLI.mm . MNLI is a pre-trained model obtained from converting Tensorflow checkpoint found in the official Google BERT repository. The following model is a Pytorch pre-training model . It was trained for four epochs with the help of Prajjwal_1.@prajwala.com and Praj,