Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Behavioral Steering on Anthropic Model-Written Evaluations Hallucination (test)
Loading...
39
Avg Token Probability
HiDRA
-0.52
9.74
20
30.26
Jun 13, 2026
Avg Token Probability
Updated 1mo ago
Evaluation Results
Method
Method
Links
Avg Token Probability
HiDRA
System prompt=positive...
2026.06
39
CAA
System prompt=positive...
2026.06
37
No steering
System prompt=positive
2026.06
20
SAS
System prompt=positive...
2026.06
19
SAS
System prompt=positive...
2026.06
19
CAA
System prompt=positive...
2026.06
13
HiDRA
System prompt=positive...
2026.06
1
Feedback
Search any
task
Search any
task
