Merge branch 'main' into humaneval

# Conflicts: # lm_eval/api/task.py

Merge branch 'main' into humaneval
# Conflicts: # lm_eval/api/task.py
173b2bc3 · Baber · 74344829 · bb098f13 · 173b2bc3 · 173b2bc3
Commit 173b2bc3 authored Jan 10, 2025 by Baber
20 changed files
--- a/lm_eval/tasks/aradice/ArabicMMLU/EGY/utils.py
+++ b/lm_eval/tasks/aradice/ArabicMMLU/EGY/utils.py
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_history.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_history.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_islamic-studies.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_islamic-studies.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_philosophy.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_humanities_philosophy.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_language_arabic-language.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_language_arabic-language.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_civics.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_civics.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_economics.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_economics.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_geography.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_social-science_geography.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_biology.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_biology.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_computer-science.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_computer-science.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_physics.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_high_stem_physics.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_humanities_history.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_humanities_history.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_humanities_islamic-studies.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_humanities_islamic-studies.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_language_arabic-language.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_language_arabic-language.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_other_general-knowledge.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_other_general-knowledge.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_civics.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_civics.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_economics.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_economics.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_geography.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_geography.yaml
--- a/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_social-science.yaml
+++ b/lm_eval/tasks/aradice/ArabicMMLU/LEV/AraDiCE_ArabicMMLU_middle_social-science_social-science.yaml