Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Red Teaming against BB-3B on ConvAI2
Loading...
45
RSR
BRT (e+r)
-0.656
11.197
23.05
34.903
May 27, 2023
RSR
Self-BLEU (k)
Updated 4d ago
Evaluation Results
Method
Method
Links
RSR
Self-BLEU (k)
BRT (e+r)
Query limit (NQ)=20,000
2023.05
45
34
BRT (e)
Query limit (NQ)=20,000
2023.05
44
33.8
SL
Backbone=OPT-1.3B, Que...
2023.05
16.4
46.6
SFS
Backbone=OPT-1.3B, Que...
2023.05
13.1
42.7
SFS
Backbone=Bloom, Query...
2023.05
11.3
42.9
Offensive Top-NQ
Query limit (NQ)=20,000
2023.05
4.8
34.4
BRT (s+r)
Query limit (NQ)=20,000
2023.05
4.8
33.7
BRT (s)
Query limit (NQ)=20,000
2023.05
4.3
33.7
Rand
Query limit (NQ)=20,000
2023.05
1.1
34.6
Feedback
Search any
task
Search any
task