Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Toxicity Reduction on Toxicity
Loading...
0.21
Final Toxicity
LoRA (rank = 16)
0.2088
0.2169
0.225
0.2331
Nov 11, 2025
Final Toxicity
Toxicity Reduction (%)
Updated 1mo ago
Evaluation Results
Method
Method
Links
Final Toxicity
Toxicity Reduction (%)
LoRA (rank = 16)
Trainable Params=40.0M...
2025.11
0.21
73.08
LoRA (rank = 1)
Trainable Params=2.5 M...
2025.11
0.24
69.23
Policy Patch
Trainable Params=0.2M...
2025.11
0.24
69.23
Feedback
Search any
task
Search any
task