Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Behavior Steering on Attribute Steering Evaluation Open-ended Generation Qwen3-14b based
Loading...
2.32
Wealth Score
SVF
1.9664
2.0582
2.15
2.2418
Feb 2, 2026
Wealth Score
Myopia Score
Corrigibility Score
Hallucination Score (+)
Hallucination Score (-)
Overall Score
Updated 3mo ago
Evaluation Results
Method
Method
Links
Wealth Score
Myopia Score
Corrigibility Score
Hallucination Score (+)
Hallucination Score (-)
Overall Score
SVF
Backbone=Qwen3-14b
2026.02
2.32
3.86
3.73
3.22
2.6
3.11
BiPO
Backbone=Qwen3-14b
2026.02
2.29
3.88
3.69
3.26
2.64
3.1
RED
Backbone=Qwen3-14b
2026.02
2.2
3.74
3.73
3.08
2.71
3.01
CAA
Backbone=Qwen3-14b
2026.02
2.04
3.49
3.56
2.94
2.9
2.79
Base
Backbone=Qwen3-14b
2026.02
1.98
3.42
3.4
2.98
2.98
2.76
Feedback
Search any
task
Search any
task
