Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
LLM Robustness Evaluation on Handcrafted prompt
Loading...
92
BAR
RA-LLM
91.68
93.84
96
98.16
Sep 18, 2023
BAR
ASR
Updated 4d ago
Evaluation Results
Method
Method
Links
BAR
ASR
RA-LLM
Model=Guanaco-7B-HF
2023.09
92
9.3
Original LLM
Model=Guanaco-7B-HF
2023.09
95.3
94.7
Perplexity Defense
Model=Vicuna-7B-chat-HF
2023.09
98
100
RA-LLM
Model=Vicuna-7B-chat-HF
2023.09
98.7
12
Original LLM
Model=Vicuna-7B-chat-HF
2023.09
99.3
98.7
Perplexity Defense
Model=Guanaco-7B-HF
2023.09
100
100
Feedback
Search any
task
Search any
task