this-is-real

mrc-pretrained-roberta-large-1

this-is-real/mrc-pretrained-roberta-large-1

mrc-pretrained-roberta-large-1 is a fill-mask model from this-is-real. Use it when you need the model to fill a missing word. It is set up for transformers.

- model: klue/roberta-large - learning rate: 1e-4 - lr scheduler type: linear - weight decay: 0.01 - epochs: 5 - checkpoint: 2700