Merge branch 'mmlu-pro' of github.com:ysjprojects/lm-evaluation-harness into mmlu-pro

sync with remote

lm_eval/tasks/arabicmmlu/arabicmmlu_high_history.yaml

+"dataset_name": "High History"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_high_history"
+"task_alias": "High History"

lm_eval/tasks/arabicmmlu/arabicmmlu_high_islamic_studies.yaml

+"dataset_name": "High Islamic Studies"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_high_islamic_studies"
+"task_alias": "High Islamic Studies"

lm_eval/tasks/arabicmmlu/arabicmmlu_high_philosophy.yaml

+"dataset_name": "High Philosophy"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_high_philosophy"
+"task_alias": "High Philosophy"

lm_eval/tasks/arabicmmlu/arabicmmlu_high_physics.yaml

+"dataset_name": "High Physics"
+"group": "arabicmmlu_stem"
+"group_alias": "stem"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_high_physics"
+"task_alias": "High Physics"

lm_eval/tasks/arabicmmlu/arabicmmlu_islamic_studies.yaml

+"dataset_name": "Islamic Studies"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_islamic_studies"
+"task_alias": "Islamic Studies"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_arabic_language.yaml

+"dataset_name": "Middle Arabic Language"
+"group": "arabicmmlu_language"
+"group_alias": "language"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_arabic_language"
+"task_alias": "Middle Arabic Language"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_civics.yaml

+"dataset_name": "Middle Civics"
+"group": "arabicmmlu_social_science"
+"group_alias": "social science"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_civics"
+"task_alias": "Middle Civics"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_computer_science.yaml

+"dataset_name": "Middle Computer Science"
+"group": "arabicmmlu_stem"
+"group_alias": "stem"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_computer_science"
+"task_alias": "Middle Computer Science"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_economics.yaml

+"dataset_name": "Middle Economics"
+"group": "arabicmmlu_social_science"
+"group_alias": "social science"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_economics"
+"task_alias": "Middle Economics"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_general_knowledge.yaml

+"dataset_name": "Middle General Knowledge"
+"group": "arabicmmlu_other"
+"group_alias": "other"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_general_knowledge"
+"task_alias": "Middle General Knowledge"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_geography.yaml

+"dataset_name": "Middle Geography"
+"group": "arabicmmlu_social_science"
+"group_alias": "social science"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_geography"
+"task_alias": "Middle Geography"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_history.yaml

+"dataset_name": "Middle History"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_history"
+"task_alias": "Middle History"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_islamic_studies.yaml

+"dataset_name": "Middle Islamic Studies"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_islamic_studies"
+"task_alias": "Middle Islamic Studies"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_natural_science.yaml

+"dataset_name": "Middle Natural Science"
+"group": "arabicmmlu_stem"
+"group_alias": "stem"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_natural_science"
+"task_alias": "Middle Natural Science"

lm_eval/tasks/arabicmmlu/arabicmmlu_middle_social_science.yaml

+"dataset_name": "Middle Social Science"
+"group": "arabicmmlu_social_science"
+"group_alias": "social science"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_middle_social_science"
+"task_alias": "Middle Social Science"

lm_eval/tasks/arabicmmlu/arabicmmlu_primary_arabic_language.yaml

+"dataset_name": "Primary Arabic Language"
+"group": "arabicmmlu_language"
+"group_alias": "language"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_primary_arabic_language"
+"task_alias": "Primary Arabic Language"

lm_eval/tasks/arabicmmlu/arabicmmlu_primary_computer_science.yaml

+"dataset_name": "Primary Computer Science"
+"group": "arabicmmlu_stem"
+"group_alias": "stem"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_primary_computer_science"
+"task_alias": "Primary Computer Science"

lm_eval/tasks/arabicmmlu/arabicmmlu_primary_general_knowledge.yaml

+"dataset_name": "Primary General Knowledge"
+"group": "arabicmmlu_other"
+"group_alias": "other"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_primary_general_knowledge"
+"task_alias": "Primary General Knowledge"

lm_eval/tasks/arabicmmlu/arabicmmlu_primary_geography.yaml

+"dataset_name": "Primary Geography"
+"group": "arabicmmlu_social_science"
+"group_alias": "social science"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_primary_geography"
+"task_alias": "Primary Geography"

lm_eval/tasks/arabicmmlu/arabicmmlu_primary_history.yaml

+"dataset_name": "Primary History"
+"group": "arabicmmlu_humanities"
+"group_alias": "humanities"
+"include": "_default_template_yaml"
+"task": "arabicmmlu_primary_history"
+"task_alias": "Primary History"