Commit 94334a79 authored by lintang's avatar lintang
Browse files

quick and dirty implementation of m-mmlu

parent c3b40173
dataset_path: openai/MMMLU
dataset_name: by_language
tag: mmmlu
fewshot_config:
sampler: first_n
output_type: multiple_choice
doc_to_text: "{{Question.strip()}}\nA. {{A}}\nB. {{B}}\nC. {{C}}\nD. {{D}}\nAnswer:"
doc_to_choice: ["A", "B", "C", "D"]
doc_to_target: Answer
metric_list:
- metric: acc
aggregation: mean
higher_is_better: true
metadata:
version: 1.0
dataset_kwargs:
trust_remote_code: true
include: _default_template_yaml
task: mmmlu_arabic
test_split: AR_XY
fewshot_split: AR_XY
include: _default_template_yaml
task: mmmlu_bengali
test_split: BN_BD
fewshot_split: BN_BD
include: _default_template_yaml
task: mmmlu_brazilian_portuguese
test_split: PT_BR
fewshot_split: PT_BR
include: _default_template_yaml
task: mmmlu_french
test_split: FR_FR
fewshot_split: FR_FR
include: _default_template_yaml
task: mmmlu_german
test_split: DE_DE
fewshot_split: DE_DE
include: _default_template_yaml
task: mmmlu_hindi
test_split: HI_IN
fewshot_split: HI_IN
include: _default_template_yaml
task: mmmlu_indonesian
test_split: ID_ID
fewshot_split: ID_ID
include: _default_template_yaml
task: mmmlu_italian
test_split: IT_IT
fewshot_split: IT_IT
include: _default_template_yaml
task: mmmlu_japanese
test_split: JA_JP
fewshot_split: JA_JP
include: _default_template_yaml
task: mmmlu_korean
test_split: KO_KR
fewshot_split: KO_KR
include: _default_template_yaml
task: mmmlu_chinese
test_split: ZH_CN
fewshot_split: ZH_CN
include: _default_template_yaml
task: mmmlu_spanish
test_split: ES_LA
fewshot_split: ES_LA
include: _default_template_yaml
task: mmmlu_swahili
test_split: SW_KE
fewshot_split: SW_KE
include: _default_template_yaml
task: mmmlu_yoruba
test_split: YO_NG
fewshot_split: YO_NG
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment