Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Activation Steering on DEBATEMIX
Loading...
56
Avg. Activation Change
K-Steer
-1.2
13.65
28.5
43.35
May 30, 2025
Avg. Activation Change
Updated 12d ago
Evaluation Results
Method
Method
Links
Avg. Activation Change
K-Steer
K=1, Model=Olmo
2025.05
56
K-Steer
K=1, Model=Llama
2025.05
56
CAA
K=1, Model=Olmo
2025.05
43
DCT
K=1, Model=Olmo
2025.05
32
DCT
K=1, Model=Llama
2025.05
29
K-Steer
K=2, Model=Olmo
2025.05
25
K-Steer
K=2, Model=Llama
2025.05
25
CAA
K=2, Model=Olmo
2025.05
24
DCT
K=2, Model=Llama
2025.05
20
DCT
K=2, Model=Olmo
2025.05
16
K-Steer
K=3, Model=Olmo
2025.05
9
K-Steer
K=3, Model=Llama
2025.05
9
CAA
K=3, Model=Llama
2025.05
8
CAA
K=2, Model=Llama
2025.05
7
DCT
K=3, Model=Llama
2025.05
7
CAA
K=3, Model=Olmo
2025.05
4
CAA
K=1, Model=Llama
2025.05
2
DCT
K=3, Model=Olmo
2025.05
1
Feedback
Search any
task
Search any
task