textattack/roberta-base-CoLA

The model was fine-tuned for 5 epochs with a batch size of 32, a learning rate of 2e-05, and a maximum sequence length of 128 . The best score the model achieved on this task was 0.850431447746884, as measured by the eval set accuracy, found after 1 epoch . TextAttack was trained with a cross-entropy loss function and the glue dataset was loaded using the nlp library . For more information, check out TextAttack on Github and the model is available on the site of the TextAttack Model Card and the data is available in the public on the current version of TextAttack. It is available to download,