tag:
  - m_mmlu_str
dataset_path: alexandrainst/m_mmlu
test_split: test
fewshot_split: train
fewshot_config:
  sampler: first_n
output_type: generate_until
doc_to_text: "{{instruction.strip()}}\nA. {{option_a}}\nB. {{option_b}}\nC. {{option_c}}\nD. {{option_d}}\nAnswer:"
doc_to_target: "{{answer}}"
filter_list:
  - name: none
    filter:
      - function: remove_whitespace
      - function: take_first
      - function: regex
        regex_pattern: "(.).*"
        group_select: 0
      - function: take_first  
generation_kwargs:
  max_gen_toks: 2
  do_sample: false
  temperature: 0.0
metric_list:
  - metric: exact_match
    aggregation: mean
    higher_is_better: true
metadata:
  version: 0.0
