Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Toxicity Mitigation on ATTAQ
Loading...
0.122
Average Max Toxicity
M+
0.0996
0.2508
0.402
0.5532
Nov 11, 2025
Average Max Toxicity
Toxic Rate
Perplexity (Quality)
Diversity (Trigram)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Max Toxicity
Toxic Rate
Perplexity (Quality)
Diversity (Trigram)
M+
Backbone=Aya-23-8B, Tr...
2025.11
0.122
0
10.81
23.7
M+
Backbone=Llama-2-7B, T...
2025.11
0.207
5.8
9.29
11.8
M+
Backbone=Llama-3-8B, T...
2025.11
0.331
17.5
11.26
16
M'
Backbone=Aya-23-8B, Tr...
2025.11
0.364
20
8.35
12.9
M'
Backbone=Llama-3-8B, T...
2025.11
0.426
35.8
8.36
12.7
M'
Backbone=Llama-2-7B, T...
2025.11
0.468
41.7
7.63
10
M
Backbone=Llama-3-8B, T...
2025.11
0.643
80.8
7.48
15
M
Backbone=Aya-23-8B, Tr...
2025.11
0.661
75
7.34
13.7
M
Backbone=Llama-2-7B, T...
2025.11
0.682
84.2
7.21
12.3
Feedback
Search any
task
Search any
task