group: - super-glue-lm-eval-v1 task: "default" dataset_path: super_glue dataset_name: cb output_type: multiple_choice training_split: train validation_split: validation doc_to_text: "{{premise}}\nQuestion: {{hypothesis}}. True, False, or Neither?\nAnswer:" doc_to_target: "{{answer_choices[labe]}}" gold_alias: "{{label}}" # this will be cast to an int. template_aliases: "{% set answer_choices = ['True', 'False', 'Neither'] %}" metric_list: - metric: acc - metric: f1 aggregation: !function "aggregate.cb_multi_fi"