Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on WMDP multiple-choice QA
Loading...
65.1
Bio Accuracy
Base
29.324
38.612
47.9
57.188
Mar 1, 2026
Bio Accuracy
Cyber Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Bio Accuracy
Cyber Accuracy
Base
Model=Mistral-7B, Zero...
2026.03
65.1
41.5
Base
Model=Zephyr-7B-β, Zer...
2026.03
64.3
44.8
LLMU
Model=Zephyr-7B-β, Zer...
2026.03
59.5
39.5
SSD
Model=Zephyr-7B-β, Zer...
2026.03
50.2
35
SCRUB
Model=Zephyr-7B-β, Zer...
2026.03
43.8
39.3
ASU
Model=Zephyr-7B-β, Zer...
2026.03
32.1
31.7
ASU
Model=Mistral-7B, Zero...
2026.03
31.5
29.5
RMU
Model=Zephyr-7B-β, Zer...
2026.03
31.2
28.2
RMU
Model=Mistral-7B, Zero...
2026.03
30.7
32.3
Feedback
Search any
task
Search any
task