CulturalBench / data /CulturalBench_Hard_per_continent.csv
kellycyy's picture
first init
c76f18f
Models, North America,South America,East Europe,South Europe ,West Europe,Africa,Middle East/ West Asia,South Asia ,Southeast Asia,East Asia,Oceania
gpt-3-5-turbo-1106 ,36.78,26.45,20.21,29.58,37.4,28.37,17.61,36.74,25.33,23.38,36.97
gpt35turbo ,41.78,36.27,30.65,34.03,43.53,34.22,21,46.51,30.28,28.76,36.97
gpt4omini ,63.57,45.54,37.04,50,55.56,48.76,37.87,60.13,50.08,47.44,44.84
gpt4o ,70.72,54.19,54.3,65.69,68.25,67.76,54.22,72.5,64.49,61.38,54.84
gpt-4o-2024-08-06 ,75.72,55.76,52.73,64.03,69.81,65.2,59.05,68.42,61.39,62.49,60.61
gpt-4-0125-preview ,75.72,57.42,56.57,60.14,68.03,67.56,54.94,66.7,59.99,65.22,59.39
gpt-4-1106-preview ,77.85,52.82,53.9,61.25,60.12,66.87,53.23,73.5,59.56,61.55,56.06
haiku ,43.93,30.17,28.36,20.84,33.96,27.64,19.42,33,17.2,20.32,27.88
sonnet3 ,48.93,31.01,33.1,52.08,40.09,34.68,26.3,40.47,26.67,32.63,30.3
opus ,75.35,43.52,47.59,54.86,52.78,58.94,42.45,64.37,58.28,52.95,53.94
sonnet35 ,70.72,40.32,42.9,58.61,59.92,57.56,43.69,63.64,55.25,49.63,65.16
mistralnemo ,48.93,30.09,37.84,46.94,45.66,39.86,29.76,45.78,39.63,34.06,34.84
mistralsmall ,63.22,43.4,39.54,46.25,48,53.2,30.37,60.24,44.62,33.36,36.97
mistral-large-2402 ,63.22,39.53,50.15,52.22,42.3,52.84,40.46,62.88,58.72,44.79,48.48
mistrallarge ,72.85,34.93,44.11,49.58,48.88,42.08,39.13,45.09,51.78,44.22,42.72
llama3-8b ,24.28,20.56,15.56,26.25,14.48,19.67,10.57,26.67,25.43,26.05,19.09
llama3-70b ,27.14,26.2,24.09,23.75,34.52,30.02,27.92,38.54,30.01,33.73,41.52
llama3-1-8b ,51.43,33.81,37.35,44.58,34.3,38.75,28.74,47.43,40.85,37.02,40.61
llama3-1-70b ,75.72,46.18,45.53,56.25,63.69,60.15,50.04,62.72,51.5,57.19,53.94
llama3-1-405b ,73.22,50.27,47.05,57.92,54.34,55.01,42.71,64.46,48.91,54.15,49.39
gemma2-9b ,51.43,37.71,30.69,40.84,40.08,42.89,29.62,54.04,39.62,36.24,16.66
gemma2-27b ,65.72,42.25,37.01,38.2,55.35,50.65,32.82,60.94,36.51,40.82,30.3
mistral-7b-v1 ,31.78,26.2,18.53,22.08,22.39,22.72,11.43,21.71,15.67,16.68,15.76
mistral-7b-v2 ,41.78,28.49,26.45,34.03,36.96,38.46,27.84,47.3,38.88,33.22,33.64
mixtral-8x22B ,68.22,37.08,39.08,44.3,53.01,50.73,37.73,66.94,43.24,38.46,38.18
qwen1-5-72b-chat ,75.35,45.88,40.48,45.84,44.22,49.83,29.83,65.22,44.87,41.15,41.52
qwen2-72b ,75.35,40.44,44.23,50.7,54,49.83,37.52,55.04,49.48,50.72,55.16