LIAMF-USP/roberta-large-finetuned-race

The initial model was roberta large model which was then fine-tuned on RACE dataset . The training procedure was necessary to preprocess the data with a method that is exemplified for a single instance in the How to use section . The used hyperparameters were the following: hyperparameter, threshold, return_overflowing_tokens=True, max_length=MAX_SEQ_LENGTH, padding="max_length", truncation=True . The model was trained with a Tesla VIE VIE-100PC VIE . The data was then preprocessed using a method described in this section . It was then used to train the data,