math_sat_cot.yml

task: gsm8k
dataset_path: mcaleste/sat_multiple_choice_math_may_23
dataset_name: main
output_type: generate_until
fewshot_split: none
test_split: train
doc_to_text: "Question: {{question}}\nAnswer:"
doc_to_target: "({{answer}})"
metric_list:
  - metric: acc
    aggregation: mean
    higher_is_better: true
process_results: !function _utils.process_results
generation_kwargs:
  until:
    - "I hope it is correct."
  do_sample: false
  temperature: 0.0
repeats: 1
num_fewshot: 5
# filter_list:
#   - name: "get-answer"
#     filter:
#       - function: "regex"
#         regex_pattern: "### (\\-?[0-9\\.\\,]+)"
#       - function: "take_first"