Using BERT-Base with train_allennlp_local.sh

Hello! I am attempting to replicate some of the experiments described in the paper. Do you have any recommendations for using BERT-Base with the `train_allennlp_local.sh` script. The PyTorch implementation of the model that I am using seems to require more files than just `vocab.txt`, `bert_config.json`, and `pytorch_model.bin`. Any help that you could provide would be greatly appreciated!