Merge pull request #922 from EleutherAI/mmlu_subgroups

[Refactor] Mmlu subgroups and weight avg

Merge pull request #922 from EleutherAI/mmlu_subgroups
[Refactor] Mmlu subgroups and weight avg
815f59e6 · Lintang Sutawika · GitHub · 3533e4b9 · 44124d95 · 815f59e6
Unverified Commit 815f59e6 authored Nov 06, 2023 by Lintang Sutawika Committed by GitHub Nov 06, 2023
19 changed files
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_management.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_management.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_marketing.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_marketing.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_medical_genetics.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_medical_genetics.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_miscellaneous.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_miscellaneous.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_disputes.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_disputes.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_scenarios.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_moral_scenarios.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_nutrition.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_nutrition.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_philosophy.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_philosophy.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_prehistory.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_prehistory.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_accounting.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_accounting.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_law.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_law.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_medicine.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_medicine.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_psychology.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_professional_psychology.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_public_relations.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_public_relations.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_security_studies.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_security_studies.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_sociology.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_sociology.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_us_foreign_policy.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_us_foreign_policy.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_virology.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_virology.yaml
--- a/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_world_religions.yaml
+++ b/lm_eval/tasks/mmlu/flan_n_shot/mmlu_log_world_religions.yaml