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/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Mauritania.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Mauritania.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Mesopotamia_civilization.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Mesopotamia_civilization.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Morocco.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Morocco.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Oman.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Oman.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Palestine.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Palestine.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Qatar.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Qatar.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Saudi_Arabia.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Saudi_Arabia.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Somalia.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Somalia.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Sudan.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Sudan.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Syria.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Syria.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Tunisia.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Tunisia.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_United_Arab_Emirates.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_United_Arab_Emirates.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Yemen.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_Yemen.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_communication.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_communication.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_computer_and_phone.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_computer_and_phone.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_daily_life.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_daily_life.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_entertainment.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/arabic_leaderboard_acva_entertainment.yaml
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/utils.py
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_avca/utils.py
--- a/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_complete.yaml
+++ b/lm_eval/tasks/arabic_leaderboard_complete/arabic_leaderboard_complete.yaml
--- a/lm_eval/tasks/arabic_leaderboard_light/README.md
+++ b/lm_eval/tasks/arabic_leaderboard_light/README.md