Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Safety and Cultural Bias Evaluation on KoBBQ
Loading...
89.18
Accuracy
Kanana-1.5-8B-Instruct
51.8544
61.5447
71.235
80.9253
Jun 5, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Kanana-1.5-8B-Instruct
Stage=Base
2026.06
89.18
Kanana-1.5-8B-Instruct
Stage=Post-trained
2026.06
88.9
EXAONE-3.5-7.8B-Instruct
Stage=Post-trained
2026.06
85.87
EXAONE-3.5-7.8B-Instruct
Stage=Base
2026.06
85.78
Qwen-2.5-7B-Instruct
Stage=Post-trained
2026.06
83.29
Qwen-2.5-7B-Instruct
Stage=Base
2026.06
82.99
A.X-4.0-Light
Stage=Post-trained
2026.06
77.39
Llama-3.1-8B-Instruct
Stage=Post-trained
2026.06
75.74
A.X-4.0-Light
Stage=Base
2026.06
72.14
Llama-3.1-8B-Instruct
Stage=Base
2026.06
71.33
Gemma-3-4B-IT
Stage=Post-trained
2026.06
53.36
Gemma-3-4B-IT
Stage=Base
2026.06
53.29
Feedback
Search any
task
Search any
task