BertLanguageModelHead(self, model_name:str) -> None

Loads just the LM head from transformers.BertForMaskedLM. It was easiest to load the entire model before only pulling out the head, so this is a bit slower than it could be, but for practical use in a model, the few seconds of extra loading time is probably not a big deal.