task: AraDiCE_piqa_lev
dataset_path: QCRI/AraDiCE-PIQA
dataset_name: PIQA-lev
training_split: null
validation_split: null
test_split: test
output_type: multiple_choice
doc_to_text: "سؤال : {{goal}}\nإجابة :"
doc_to_target: label
doc_to_choice: "{{[sol1, sol2]}}"
should_decontaminate: true
doc_to_decontamination_query: goal
metric_list:
  - metric: acc
    aggregation: mean
    higher_is_better: true
  - metric: acc_norm
    aggregation: mean
    higher_is_better: true
  - metric: f1
    higher_is_better: true
    aggregation: !function metrics.micro_f1_score
metadata:
  version: 1.0