https://huggingface.co/cknowledge/mlperf-inference-bert-pytorch-fp32-squad-v1.1/blob/79d57e8ee4ea18ebc6febb953c6655109ba3d577/config.json#L3
It finetuned on squad dataset right? why the architecture is BertForMaskedLM?
BertForMaskedLM
· Sign up or log in to comment