Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Jailbreaking

Benchmarks

Task NameDataset NameSOTA ResultTrend
Explanation FaithfulnessJailbreaking DAN (test)
Label Change Rate24
8
Explanation FaithfulnessJailbreaking GCG (test)
Rate of Label Changes0
8
JailbreakingJailbreaking
HQ Neuron Count185
2
Showing 3 of 3 rows