Merge pull request #922 from EleutherAI/mmlu_subgroups

[Refactor] Mmlu subgroups and weight avg

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_management.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_management.yaml

 "dataset_name": "management"
-"description": "The following are multiple choice questions (with answers) about management.\n\n"
+"description": "The following are multiple choice questions (with answers) about management.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_management"
+"task": "mmlu_flan_n_shot_generative_management"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_marketing.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_marketing.yaml

 "dataset_name": "marketing"
-"description": "The following are multiple choice questions (with answers) about marketing.\n\n"
+"description": "The following are multiple choice questions (with answers) about marketing.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_marketing"
+"task": "mmlu_flan_n_shot_generative_marketing"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_medical_genetics.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_medical_genetics.yaml

 "dataset_name": "medical_genetics"
-"description": "The following are multiple choice questions (with answers) about medical genetics.\n\n"
+"description": "The following are multiple choice questions (with answers) about medical\
+  \ genetics.\n\n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_medical_genetics"
+"task": "mmlu_flan_n_shot_generative_medical_genetics"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_miscellaneous.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_miscellaneous.yaml

 "dataset_name": "miscellaneous"
-"description": "The following are multiple choice questions (with answers) about miscellaneous.\n\n"
+"description": "The following are multiple choice questions (with answers) about miscellaneous.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_miscellaneous"
+"task": "mmlu_flan_n_shot_generative_miscellaneous"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_disputes.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_moral_disputes.yaml

 "dataset_name": "moral_disputes"
-"description": "The following are multiple choice questions (with answers) about moral disputes.\n\n"
+"description": "The following are multiple choice questions (with answers) about moral\
+  \ disputes.\n\n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_moral_disputes"
+"task": "mmlu_flan_n_shot_generative_moral_disputes"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_scenarios.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_moral_scenarios.yaml

 "dataset_name": "moral_scenarios"
-"description": "The following are multiple choice questions (with answers) about moral scenarios.\n\n"
+"description": "The following are multiple choice questions (with answers) about moral\
+  \ scenarios.\n\n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_moral_scenarios"
+"task": "mmlu_flan_n_shot_generative_moral_scenarios"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_nutrition.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_nutrition.yaml

 "dataset_name": "nutrition"
-"description": "The following are multiple choice questions (with answers) about nutrition.\n\n"
+"description": "The following are multiple choice questions (with answers) about nutrition.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_nutrition"
+"task": "mmlu_flan_n_shot_generative_nutrition"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_philosophy.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_philosophy.yaml

 "dataset_name": "philosophy"
-"description": "The following are multiple choice questions (with answers) about philosophy.\n\n"
+"description": "The following are multiple choice questions (with answers) about philosophy.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_philosophy"
+"task": "mmlu_flan_n_shot_generative_philosophy"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_prehistory.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_prehistory.yaml

 "dataset_name": "prehistory"
-"description": "The following are multiple choice questions (with answers) about prehistory.\n\n"
+"description": "The following are multiple choice questions (with answers) about prehistory.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_prehistory"
+"task": "mmlu_flan_n_shot_generative_prehistory"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_accounting.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_professional_accounting.yaml

 "dataset_name": "professional_accounting"
-"description": "The following are multiple choice questions (with answers) about professional accounting.\n\n"
+"description": "The following are multiple choice questions (with answers) about professional\
+  \ accounting.\n\n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_professional_accounting"
+"task": "mmlu_flan_n_shot_generative_professional_accounting"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_law.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_professional_law.yaml

 "dataset_name": "professional_law"
-"description": "The following are multiple choice questions (with answers) about professional law.\n\n"
+"description": "The following are multiple choice questions (with answers) about professional\
+  \ law.\n\n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_professional_law"
+"task": "mmlu_flan_n_shot_generative_professional_law"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_medicine.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_professional_medicine.yaml

 "dataset_name": "professional_medicine"
-"description": "The following are multiple choice questions (with answers) about professional medicine.\n\n"
+"description": "The following are multiple choice questions (with answers) about professional\
+  \ medicine.\n\n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_professional_medicine"
+"task": "mmlu_flan_n_shot_generative_professional_medicine"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_psychology.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_professional_psychology.yaml

 "dataset_name": "professional_psychology"
-"description": "The following are multiple choice questions (with answers) about professional psychology.\n\n"
+"description": "The following are multiple choice questions (with answers) about professional\
+  \ psychology.\n\n"
+"group": "mmlu_flan_n_shot_generative_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_professional_psychology"
+"task": "mmlu_flan_n_shot_generative_professional_psychology"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_public_relations.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_public_relations.yaml

 "dataset_name": "public_relations"
-"description": "The following are multiple choice questions (with answers) about public relations.\n\n"
+"description": "The following are multiple choice questions (with answers) about public\
+  \ relations.\n\n"
+"group": "mmlu_flan_n_shot_generative_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_public_relations"
+"task": "mmlu_flan_n_shot_generative_public_relations"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_security_studies.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_security_studies.yaml

 "dataset_name": "security_studies"
-"description": "The following are multiple choice questions (with answers) about security studies.\n\n"
+"description": "The following are multiple choice questions (with answers) about security\
+  \ studies.\n\n"
+"group": "mmlu_flan_n_shot_generative_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_security_studies"
+"task": "mmlu_flan_n_shot_generative_security_studies"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_sociology.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_sociology.yaml

 "dataset_name": "sociology"
-"description": "The following are multiple choice questions (with answers) about sociology.\n\n"
+"description": "The following are multiple choice questions (with answers) about sociology.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_sociology"
+"task": "mmlu_flan_n_shot_generative_sociology"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_us_foreign_policy.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_us_foreign_policy.yaml

 "dataset_name": "us_foreign_policy"
-"description": "The following are multiple choice questions (with answers) about us foreign policy.\n\n"
+"description": "The following are multiple choice questions (with answers) about us\
+  \ foreign policy.\n\n"
+"group": "mmlu_flan_n_shot_generative_social_sciences"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_us_foreign_policy"
+"task": "mmlu_flan_n_shot_generative_us_foreign_policy"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_virology.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_virology.yaml

 "dataset_name": "virology"
-"description": "The following are multiple choice questions (with answers) about virology.\n\n"
+"description": "The following are multiple choice questions (with answers) about virology.\n\
+  \n"
+"group": "mmlu_flan_n_shot_generative_other"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_virology"
+"task": "mmlu_flan_n_shot_generative_virology"

lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_world_religions.yaml → lm_eval/tasks/mmlu/flan_n_shot/loglikelihood/mmlu_world_religions.yaml

 "dataset_name": "world_religions"
-"description": "The following are multiple choice questions (with answers) about world religions.\n\n"
+"description": "The following are multiple choice questions (with answers) about world\
+  \ religions.\n\n"
+"group": "mmlu_flan_n_shot_generative_humanities"
 "include": "_mmlu_flan_loglikelihood_template_yaml"
-"task": "mmlu_flan_n_shot_loglikelihood_world_religions"
+"task": "mmlu_flan_n_shot_generative_world_religions"