Spaces:
Build error
Build error
Update gemini mmlu
#6
by JessicaOjo - opened
community_results/New Results - June2025.csv
ADDED
|
@@ -0,0 +1,65 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
model,task,prompt,eng,amh,hau,ibo,kin,lin,lug,orm,sna,sot,swa,xho,yor,zul,avg,prompt_num
|
| 2 |
+
Gemma 3 27b,injongointent,4.0,79.58,71.56,69.53,56.56,53.44,38.12,37.97,31.25,56.88,47.66,81.09,63.59,49.53,60.78,55.2,
|
| 3 |
+
Gemma 3 27b,afrixnli,2.0,65.5,59.5,55.5,53.83,46.67,35.17,48.17,50.17,52.67,51.83,55.67,53.83,48.17,54.83,51.2,
|
| 4 |
+
Gemma 3 27b,afrimmlu,1.0,74.2,52.6,45.2,47.0,44.6,36.8,34.0,34.2,45.2,44.8,60.2,44.8,40.2,47.0,44.4,
|
| 5 |
+
Gemma 3 27b,afrimgsm,2.0,87.6,61.2,61.2,36.8,54.0,18.4,36.4,32.0,52.0,51.6,78.0,48.8,36.8,50.8,47.5,
|
| 6 |
+
Gemma 3 27b,belebele,5.0,93.44,76.44,71.44,59.22,69.0,38.44,44.67,52.11,68.78,62.0,84.7,67.33,49.56,67.0,62.4,
|
| 7 |
+
Gemma 3 27b,sib,3.0,87.25,83.33,80.4,81.86,82.84,59.8,67.16,58.33,76.47,72.55,72.06,77.45,71.08,80.88,74.2,
|
| 8 |
+
Gemma 3 27b,flores - en_xx,,68.9,23.8,43.9,35.2,35.6,24.8,21.6,17.5,35.1,38.1,57.0,35.8,21.0,40.8,33.1,
|
| 9 |
+
Gemini-2.0 Flash,injongointent,4.0,88.9,84.8,80.5,78.8,63.8,65.3,63.0,62.3,77.0,59.4,91.4,84.2,75.3,74.5,73.9,
|
| 10 |
+
Gemini-2.0 Flash,sib,3.0,86.8,87.3,85.3,82.8,87.3,84.8,81.9,80.4,85.8,82.8,87.8,86.3,84.3,87.3,84.9,
|
| 11 |
+
Gemini-2.0 Flash,afrixnli,3.0,87.0,74.2,72.0,68.7,62.5,32.2,67.5,70.0,71.5,68.7,74.0,71.8,63.2,68.3,66.5,
|
| 12 |
+
Gemini-2.0 Flash,belebele,4.0,63.0,66.0,49.2,48.7,59.3,51.9,54.3,52.0,63.4,50.8,69.8,61.8,49.9,61.3,56.8,
|
| 13 |
+
Gemini-2.0 Flash,afrimmlu,5.0,80.8,65.0,61.2,57.4,54.4,52.8,49.4,51.8,57.0,55.8,71.4,60.6,57.4,56.8,57.8,
|
| 14 |
+
Gemini-2.0 Flash,afrimgsm,2.0,92.8,76.0,70.4,61.2,64.4,63.6,59.2,73.6,63.6,65.2,84.0,60.8,70.8,65.2,67.5,
|
| 15 |
+
Gemini-2.0 Flash,flores - en_xx,2.0,71.5,66.0,60.9,57.8,59.2,51.7,52.2,54.7,54.8,63.6,70.3,62.4,50.5,64.2,59.1,
|
| 16 |
+
Gemini-2.0 Flash,flores - xx_en,2.0,73.1,40.9,54.3,45.2,53.0,51.5,44.0,44.2,50.2,52.0,65.2,54.3,31.6,58.4,49.6,
|
| 17 |
+
GPT-4.1 (April),injongointent,1.0,87.8,86.6,93.3,86.6,77.2,82.8,81.3,73.8,90.2,75.0,93.9,89.5,85.6,81.9,84.4,
|
| 18 |
+
GPT-4.1 (April),sib,1.0,89.7,85.8,84.8,85.8,87.3,84.3,82.4,84.3,86.3,86.3,81.4,87.7,80.4,85.8,84.8,
|
| 19 |
+
GPT-4.1 (April),afrixnli,1.0,88.5,70.8,72.7,69.0,69.3,32.0,67.0,68.0,73.7,71.5,72.8,74.0,67.2,69.8,67.5,
|
| 20 |
+
GPT-4.1 (April),belebele,1.0,73.9,72.0,48.0,37.7,76.4,61.7,54.0,58.3,77.6,60.8,90.0,78.0,52.2,76.1,64.8,
|
| 21 |
+
GPT-4.1 (April),afrimmlu,1.0,71.4,55.2,55.4,56.2,63.8,60.2,50.2,53.0,65.2,57.0,76.8,66.6,53.6,69.8,60.2,
|
| 22 |
+
GPT-4.1 (April),afrimgsm,1.0,82.4,57.2,57.2,54.0,60.4,49.6,54.4,64.4,60.0,64.4,77.2,51.2,65.2,58.0,59.5,
|
| 23 |
+
GPT-4.1 (April),flores - en_xx,1.0,73.1,32.5,53.1,42.0,53.2,48.7,41.3,41.6,49.6,50.6,65.5,53.6,27.5,55.7,47.3,
|
| 24 |
+
LLaMa 4 405B,injongointent,4.0,88.9,84.8,80.5,78.8,63.8,65.3,63.0,62.3,77.0,59.4,91.4,84.2,75.3,74.5,73.9,
|
| 25 |
+
LLaMa 4 405B,sib,3.0,84.8,86.8,80.9,82.4,74.5,84.3,79.4,71.6,77.5,77.9,86.8,84.8,78.9,81.9,80.6,
|
| 26 |
+
LLaMa 4 405B,afrixnli,2.0,49.2,56.8,44.8,53.3,43.0,35.3,42.5,45.0,46.5,43.7,48.5,47.0,41.8,43.0,45.5,
|
| 27 |
+
LLaMa 4 405B,belebele,5.0,25.0,24.78,24.11,24.44,24.67,25.22,24.78,24.22,24.67,25.22,24.56,24.78,23.89,24.33,24.6,
|
| 28 |
+
LLaMa 4 405B,afrimmlu,1.0,11.2,18.4,16.2,18.0,14.2,10.8,14.8,16.0,15.2,16.0,15.6,16.4,16.8,17.0,15.8,
|
| 29 |
+
LLaMa 4 405B,afrimgsm,4.0,97.6,83.2,68.8,64.4,60.0,52.0,46.8,58.8,67.2,65.2,89.2,61.2,64.8,64.0,65.0,
|
| 30 |
+
LLaMa 4 405B,flores - en_xx,3.0,73.0,37.64,50.34,43.73,39.05,41.76,34.11,36.48,40.53,40.07,61.96,50.25,28.0,53.07,42.8,
|
| 31 |
+
LLaMa 4 405B,flores - xx_en,3.0,71.55,62.57,56.47,53.5,54.28,43.12,44.89,46.96,50.36,56.32,67.83,58.03,44.82,60.64,53.8,
|
| 32 |
+
Lugha-Llama 8B,injongointent,1.0,16.7,3.1,5.8,4.2,3.0,8.0,2.3,2.5,2.5,5.0,5.6,2.7,5.8,2.5,4.1,
|
| 33 |
+
Lugha-Llama 8B,sib,1.0,43.6,43.6,32.4,30.9,41.7,24.5,34.8,42.2,29.4,30.4,42.6,27.0,43.1,21.1,34.1,
|
| 34 |
+
Lugha-Llama 8B,afrixnli,1.0,46.8,37.0,37.3,37.2,35.5,33.3,34.2,36.2,35.8,35.8,39.7,37.2,38.7,38.7,36.7,
|
| 35 |
+
Lugha-Llama 8B,belebele,1.0,22.4,23.1,22.9,22.9,23.2,22.9,23.7,22.1,23.2,22.7,23.4,22.8,22.9,23.1,23.0,
|
| 36 |
+
Lugha-Llama 8B,afrimmlu,1.0,31.8,24.6,24.8,23.8,24.8,26.4,24.6,23.2,26.2,25.4,27.8,26.6,23.6,25.6,25.2,
|
| 37 |
+
Lugha-Llama 8B,afrimgsm,1.0,6.4,2.4,0.8,2.0,1.2,0.8,2.4,1.6,0.8,2.0,2.0,2.0,2.4,3.6,1.8,
|
| 38 |
+
Lugha-Llama 8B,flores - en_xx,1.0,51.3,7.9,19.3,25.5,29.2,9.6,15.3,5.7,30.3,25.9,40.9,31.3,16.9,29.4,22.1,
|
| 39 |
+
Gemini-2.5 Flash,afrixnli,,86.2,76.3,71.8,76.2,68.8,32.8,69.2,74.2,76.2,71.3,74.0,76.8,67.7,73.2,69.9,3.0
|
| 40 |
+
Gemini-2.5 Flash,afrimgsm,,86.8,78.0,72.0,71.6,69.2,58.8,60.8,74.0,72.4,68.0,87.2,62.0,77.6,66.4,70.6,2.0
|
| 41 |
+
Gemini-2.5 Flash,flores - en_xx,,67.8,41.5,50.1,42.8,50.4,46.4,41.0,43.0,46.7,48.4,60.6,50.9,28.2,54.4,46.5,3.0
|
| 42 |
+
Gemini-2.5 Flash,afrimmlu,,83.6,74.4,60.6,62.4,62.2,70.0,71.2,72.6,69.4,61.4,56.6,73.8,72.0,73.8,67.7,1.0
|
| 43 |
+
Gemini-2.5 Flash,injongointent,,89.1,92.8,95.0,91.4,80.5,86.7,85.5,82.8,92.3,76.9,95.0,89.8,89.8,83.8,87.9,4.0
|
| 44 |
+
Gemini-2.5 Flash,sib,,90.2,87.7,86.8,85.3,85.8,90.2,85.8,87.3,85.3,87.3,89.2,88.7,86.8,87.7,87.2,3.0
|
| 45 |
+
Gemini-2.5 Flash,belebele,,74.1,47.9,33.7,28.3,44.1,41.7,37.4,38.1,49.2,42.6,53.6,48.6,36.3,46.6,42.2,5.0
|
| 46 |
+
Claude 4.0 Sonnet,afrixnli,,90.2,80.3,72.3,74.0,66.3,32.0,62.7,68.3,70.8,71.2,73.5,74.8,67.2,71.3,68.1,1.0
|
| 47 |
+
Claude 4.0 Sonnet,afrimgsm,,94.0,76.4,68.8,63.6,63.2,58.8,55.6,64.8,68.0,66.8,86.0,59.6,71.2,67.2,66.9,4.0
|
| 48 |
+
Claude 4.0 Sonnet,flores - en_xx,,69.8,41.1,51.1,43.5,47.8,47.6,41.0,41.2,44.9,48.8,60.0,51.6,27.6,52.4,46.0,3.0
|
| 49 |
+
Claude 4.0 Sonnet,afrimmlu,,90.0,83.2,74.4,80.8,73.2,73.6,63.2,72.6,74.0,75.6,84.8,79.6,72.6,74.2,75.5,1.0
|
| 50 |
+
Claude 4.0 Sonnet,injongointent,,89.7,90.6,88.3,84.7,69.1,77.0,74.5,67.8,84.8,68.0,92.5,87.0,81.6,78.8,80.4,4.0
|
| 51 |
+
Claude 4.0 Sonnet,sib,,88.2,88.7,84.8,83.8,85.3,81.4,76.0,76.0,78.9,83.3,87.7,84.8,86.3,84.8,83.2,5.0
|
| 52 |
+
Claude 4.0 Sonnet,belebele,,94.7,86.0,77.9,72.4,77.9,70.2,64.4,68.4,77.8,77.9,89.6,81.9,68.3,78.1,76.2,5.0
|
| 53 |
+
Claude 3.7 Sonnet,afrixnli,,83.2,69.3,60.5,60.8,60.0,35.0,58.5,62.2,61.5,63.8,63.7,63.7,55.5,63.3,59.8,1.0
|
| 54 |
+
Claude 3.7 Sonnet,afrimgsm,,60.4,46.0,30.4,30.4,37.2,26.8,30.8,32.8,30.8,36.0,53.2,32.0,38.4,33.2,35.2,4.0
|
| 55 |
+
Claude 3.7 Sonnet,flores - en_xx,,69.9,37.6,45.6,39.1,46.8,45.4,40.4,38.9,40.9,50.9,60.5,47.9,27.7,50.3,44.0,3.0
|
| 56 |
+
Claude 3.7 Sonnet,afrimmlu,,77.8,71.0,64.2,67.8,64.0,62.0,62.8,65.6,63.6,67.6,76.4,68.6,66.4,67.6,66.7,1.0
|
| 57 |
+
Claude 3.7 Sonnet,injongointent,,84.1,82.0,79.2,72.3,66.4,65.3,71.4,59.8,76.4,64.2,88.8,82.2,73.8,71.9,73.4,4.0
|
| 58 |
+
Claude 3.7 Sonnet,sib,,86.8,86.8,86.3,83.8,86.3,85.8,81.4,82.4,83.8,83.8,85.8,87.7,85.8,83.8,84.9,5.0
|
| 59 |
+
Claude 3.7 Sonnet,belebele,,74.0,69.4,60.1,59.0,68.7,64.7,62.6,62.2,70.4,65.0,74.8,66.8,57.6,65.2,65.1,5.0
|
| 60 |
+
Claude 4.5 Sonnet,afrixnli,,90.8,83.2,75.2,76.8,70.5,32.5,70.8,74.3,78.2,72.8,76.7,78.3,70.0,72.5,71.7,1.0
|
| 61 |
+
Claude 4.5 Sonnet,afrimgsm,,96.4,80.8,73.2,72.8,71.2,65.2,63.6,74.8,68.4,73.6,91.6,68.4,77.6,69.6,73.1,4.0
|
| 62 |
+
Claude 4.5 Sonnet,afrimmlu,,89.6,65.0,56.0,57.2,53.4,51.8,57.8,58.0,62.4,59.4,58.8,60.6,64.0,57.4,58.6,1.0
|
| 63 |
+
Claude 4.5 Sonnet,injongointent,,88.3,90.6,89.2,83.8,70.8,78.6,75.9,71.9,86.1,70.3,93.6,87.3,83.4,79.8,81.6,4.0
|
| 64 |
+
Claude 4.5 Sonnet,sib,,87.7,89.7,86.8,85.3,87.3,79.4,76.5,83.3,80.4,84.8,88.7,85.8,82.8,84.3,84.2,5.0
|
| 65 |
+
Claude 4.5 Sonnet,belebele,,83.9,81.8,73.4,68.7,73.2,70.1,66.3,71.6,75.2,75.3,87.7,79.4,72.2,77.0,74.8,5.0
|
leaderboard_json/afrobench_lite.json
ADDED
|
@@ -0,0 +1,176 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"NLI": {
|
| 3 |
+
"afrixnli": {
|
| 4 |
+
"Aya-101 13B": 52.4,
|
| 5 |
+
"Gemma1.1 7b": 34.6,
|
| 6 |
+
"LLaMa2 7b": 33.7,
|
| 7 |
+
"LLaMa3 8B": 35.6,
|
| 8 |
+
"LLaMa3.1 8B": 36.4,
|
| 9 |
+
"LLaMAX3 8B": 41.4,
|
| 10 |
+
"Gemma2 9b": 40.9,
|
| 11 |
+
"Gemma2 27b": 43.7,
|
| 12 |
+
"LLaMa3.1 70B": 38.4,
|
| 13 |
+
"AfroLLaMa 8B": 36.3,
|
| 14 |
+
"Gemini 1.5 pro": 63.6,
|
| 15 |
+
"GPT-4o (Aug)": 66.3,
|
| 16 |
+
"Gemma 3 27b": 51.2,
|
| 17 |
+
"Gemini-2.0 Flash": 66.5,
|
| 18 |
+
"GPT-4.1 (April)": 67.5,
|
| 19 |
+
"LLaMa 4 405B": 45.5,
|
| 20 |
+
"Lugha-Llama 8B": 36.7,
|
| 21 |
+
"Gemini-2.5 Flash": 69.9,
|
| 22 |
+
"Claude 4.0 Sonnet": 68.1,
|
| 23 |
+
"Claude 3.7 Sonnet": 59.8,
|
| 24 |
+
"Claude 4.5 Sonnet": 71.7
|
| 25 |
+
}
|
| 26 |
+
},
|
| 27 |
+
"Intent": {
|
| 28 |
+
"injongointent": {
|
| 29 |
+
"AfroLLaMa 8B": 0.9,
|
| 30 |
+
"LLaMAX3 8B": 6.3,
|
| 31 |
+
"LLaMa2 7b": 1.4,
|
| 32 |
+
"LLaMa3.1 70B": 36.9,
|
| 33 |
+
"LLaMa3.1 8B": 6.3,
|
| 34 |
+
"LLaMa3 8B": 2.3,
|
| 35 |
+
"Gemma1.1 7b": 10.2,
|
| 36 |
+
"Gemma2 27b": 37.8,
|
| 37 |
+
"Gemma2 9b": 34.1,
|
| 38 |
+
"Aya-101 13B": 48.2,
|
| 39 |
+
"Gemini 1.5 pro": 76.3,
|
| 40 |
+
"GPT-4o (Aug)": 78.3,
|
| 41 |
+
"Gemma 3 27b": 55.2,
|
| 42 |
+
"Gemini-2.0 Flash": 73.9,
|
| 43 |
+
"GPT-4.1 (April)": 84.4,
|
| 44 |
+
"LLaMa 4 405B": 73.9,
|
| 45 |
+
"Lugha-Llama 8B": 4.1,
|
| 46 |
+
"Gemini-2.5 Flash": 87.9,
|
| 47 |
+
"Claude 4.0 Sonnet": 80.4,
|
| 48 |
+
"Claude 3.7 Sonnet": 73.4,
|
| 49 |
+
"Claude 4.5 Sonnet": 81.6
|
| 50 |
+
}
|
| 51 |
+
},
|
| 52 |
+
"MT(en/fr-xx)": {
|
| 53 |
+
"flores - en_xx": {
|
| 54 |
+
"AfroLLaMa 8B": 10.7,
|
| 55 |
+
"LLaMAX3 8B": 27.7,
|
| 56 |
+
"LLaMa2 7b": 9.2,
|
| 57 |
+
"LLaMa3.1 70B": 27.4,
|
| 58 |
+
"LLaMa3.1 8B": 14.5,
|
| 59 |
+
"LLaMa3 8B": 11.8,
|
| 60 |
+
"Aya-101 13B": 31.1,
|
| 61 |
+
"Gemma1.1 7b": 11.7,
|
| 62 |
+
"Gemma2 27b": 38.4,
|
| 63 |
+
"Gemma2 9b": 36.5,
|
| 64 |
+
"Gemini 1.5 pro": 43.0,
|
| 65 |
+
"GPT-4o (Aug)": 42.4,
|
| 66 |
+
"Gemma 3 27b": 33.1,
|
| 67 |
+
"Gemini-2.0 Flash": 59.1,
|
| 68 |
+
"GPT-4.1 (April)": 47.3,
|
| 69 |
+
"LLaMa 4 405B": 42.8,
|
| 70 |
+
"Lugha-Llama 8B": 22.1,
|
| 71 |
+
"Gemini-2.5 Flash": 46.5,
|
| 72 |
+
"Claude 4.0 Sonnet": 46.0,
|
| 73 |
+
"Claude 3.7 Sonnet": 44.0
|
| 74 |
+
}
|
| 75 |
+
},
|
| 76 |
+
"MMLU": {
|
| 77 |
+
"afrimmlu": {
|
| 78 |
+
"AfroLLaMa 8B": 25.8,
|
| 79 |
+
"Aya-101 13B": 30.2,
|
| 80 |
+
"Gemma1.1 7b": 27.3,
|
| 81 |
+
"LLaMa2 7b": 25.9,
|
| 82 |
+
"LLaMa3 8B": 28.4,
|
| 83 |
+
"LLaMAX3 8B": 29.6,
|
| 84 |
+
"Gemma2 9b": 36.1,
|
| 85 |
+
"Gemma2 27b": 40.8,
|
| 86 |
+
"LLaMa3.1 8B": 31.6,
|
| 87 |
+
"LLaMa3.1 70B": 40.6,
|
| 88 |
+
"GPT-4o (Aug)": 63.1,
|
| 89 |
+
"Gemini 1.5 pro": 62.6,
|
| 90 |
+
"Gemma 3 27b": 44.4,
|
| 91 |
+
"Gemini-2.0 Flash": 57.8,
|
| 92 |
+
"GPT-4.1 (April)": 60.2,
|
| 93 |
+
"LLaMa 4 405B": 15.8,
|
| 94 |
+
"Lugha-Llama 8B": 25.2,
|
| 95 |
+
"Gemini-2.5 Flash": 67.7,
|
| 96 |
+
"Claude 4.0 Sonnet": 75.5,
|
| 97 |
+
"Claude 3.7 Sonnet": 66.7,
|
| 98 |
+
"Claude 4.5 Sonnet": 58.6
|
| 99 |
+
}
|
| 100 |
+
},
|
| 101 |
+
"Math": {
|
| 102 |
+
"afrimgsm": {
|
| 103 |
+
"AfroLLaMa 8B": 0.3,
|
| 104 |
+
"Aya-101 13B": 4.9,
|
| 105 |
+
"Gemma1.1 7b": 5.0,
|
| 106 |
+
"LLaMa2 7b": 1.9,
|
| 107 |
+
"LLaMa3 8B": 5.7,
|
| 108 |
+
"LLaMa3.1 8B": 7.8,
|
| 109 |
+
"LLaMAX3 8B": 5.2,
|
| 110 |
+
"Gemma2 9b": 21.7,
|
| 111 |
+
"Gemma2 27b": 31.1,
|
| 112 |
+
"LLaMa3.1 70B": 26.5,
|
| 113 |
+
"GPT-4o (Aug)": 57.3,
|
| 114 |
+
"Gemini 1.5 pro": 57.7,
|
| 115 |
+
"Gemma 3 27b": 47.5,
|
| 116 |
+
"Gemini-2.0 Flash": 67.5,
|
| 117 |
+
"GPT-4.1 (April)": 59.5,
|
| 118 |
+
"LLaMa 4 405B": 65.0,
|
| 119 |
+
"Lugha-Llama 8B": 1.8,
|
| 120 |
+
"Gemini-2.5 Flash": 70.6,
|
| 121 |
+
"Claude 4.0 Sonnet": 66.9,
|
| 122 |
+
"Claude 3.7 Sonnet": 35.2,
|
| 123 |
+
"Claude 4.5 Sonnet": 73.1
|
| 124 |
+
}
|
| 125 |
+
},
|
| 126 |
+
"Topic": {
|
| 127 |
+
"sib": {
|
| 128 |
+
"AfroLLaMa 8B": 13.9,
|
| 129 |
+
"LLaMAX3 8B": 62.6,
|
| 130 |
+
"LLaMa2 7b": 18.0,
|
| 131 |
+
"LLaMa3.1 70B": 64.3,
|
| 132 |
+
"LLaMa3.1 8B": 48.0,
|
| 133 |
+
"LLaMa3 8B": 40.1,
|
| 134 |
+
"Aya-101 13B": 76.6,
|
| 135 |
+
"Gemma1.1 7b": 42.7,
|
| 136 |
+
"Gemma2 27b": 68.8,
|
| 137 |
+
"Gemma2 9b": 65.7,
|
| 138 |
+
"Gemini 1.5 pro": 81.3,
|
| 139 |
+
"GPT-4o (Aug)": 82.8,
|
| 140 |
+
"Gemma 3 27b": 74.2,
|
| 141 |
+
"Gemini-2.0 Flash": 84.9,
|
| 142 |
+
"GPT-4.1 (April)": 84.8,
|
| 143 |
+
"LLaMa 4 405B": 80.6,
|
| 144 |
+
"Lugha-Llama 8B": 34.1,
|
| 145 |
+
"Gemini-2.5 Flash": 87.2,
|
| 146 |
+
"Claude 4.0 Sonnet": 83.2,
|
| 147 |
+
"Claude 3.7 Sonnet": 84.9,
|
| 148 |
+
"Claude 4.5 Sonnet": 84.2
|
| 149 |
+
}
|
| 150 |
+
},
|
| 151 |
+
"RC": {
|
| 152 |
+
"belebele": {
|
| 153 |
+
"AfroLLaMa 8B": 24.7,
|
| 154 |
+
"LLaMAX3 8B": 28.9,
|
| 155 |
+
"LLaMa2 7b": 23.5,
|
| 156 |
+
"LLaMa3.1 8B": 36.2,
|
| 157 |
+
"LLaMa3.1 70B": 45.9,
|
| 158 |
+
"LLaMa3 8B": 26.2,
|
| 159 |
+
"Aya-101 13B": 60.3,
|
| 160 |
+
"Gemma1.1 7b": 34.3,
|
| 161 |
+
"Gemma2 27b": 53.5,
|
| 162 |
+
"Gemma2 9b": 50.5,
|
| 163 |
+
"Gemini 1.5 pro": 55.0,
|
| 164 |
+
"GPT-4o (Aug)": 70.4,
|
| 165 |
+
"Gemma 3 27b": 62.4,
|
| 166 |
+
"Gemini-2.0 Flash": 56.8,
|
| 167 |
+
"GPT-4.1 (April)": 64.8,
|
| 168 |
+
"LLaMa 4 405B": 24.6,
|
| 169 |
+
"Lugha-Llama 8B": 23.0,
|
| 170 |
+
"Gemini-2.5 Flash": 42.2,
|
| 171 |
+
"Claude 4.0 Sonnet": 76.2,
|
| 172 |
+
"Claude 3.7 Sonnet": 65.1,
|
| 173 |
+
"Claude 4.5 Sonnet": 74.8
|
| 174 |
+
}
|
| 175 |
+
}
|
| 176 |
+
}
|