# This file will be included in the generated language-specific task configs.
# It doesn't have a yaml file extension as it is not meant to be imported directly
# by the harness.
tag: llama3
dataset_path: juletxara/mgsm
dataset_name: null  # Overridden by language-specific config.
output_type: generate_until
training_split: train
test_split: test
target_delimiter: ""
doc_to_text: "{{question}}"
doc_to_target: answers # list
process_results: !function utils.process_results
generation_kwargs:
  until: []
  do_sample: false
  temperature: 0.0
  max_gen_toks: 2048
metric_list:
  - metric: exact_match
    aggregation: mean
    higher_is_better: true
    ignore_case: true
    ignore_punctuation: true
filter_list:
  - name: "strict-match"
    filter:
      - function: "regex"
        regex_pattern: "(?:Answer|Réponse|Antwort|Ответ|Respuesta|答え|Jibu|答案|คำตอบ|సమాధానం|উত্তর): (\\-?[0-9\\.\\,]+)"
      - function: remove_whitespace
      - function: take_first
  - name: "flexible-extract"
    filter:
      - function: regex
        group_select: -1
        regex_pattern: "(?:Answer|Réponse|Antwort|Ответ|Respuesta|答え|Jibu|答案|คำตอบ|సమాధానం|উত্তর): (-?[$0-9.,]{2,})|(-?[0-9]+)"
      - function: remove_whitespace
      - function: take_first
metadata:
  version: 0.0
