Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Stealthiness Evaluation on LLaMA Guard 7B
Loading...
3.05
Mean Perplexity (PPL)
Ascii
-19.3608
131.9121
283.185
434.4579
Oct 2, 2024
Mean Perplexity (PPL)
Std Dev Perplexity (PPL)
Updated 16d ago
Evaluation Results
Method
Method
Links
Mean Perplexity (PPL)
Std Dev Perplexity (PPL)
Ascii
2024.10
3.05
0.27
ArtPrompt
2024.10
3.36
1
Base64
2024.10
10.14
2.85
Morse Cipher
2024.10
11.18
1.89
ReNeLLM
2024.10
13.33
4.16
Unicode
2024.10
29.37
15.38
UTF-8
2024.10
29.37
15.38
Origin
2024.10
33.44
21.14
Caesar Cipher
2024.10
202.08
139.35
FlipAttack
2024.10
563.32
234.23
Feedback
Search any
task
Search any
task