INDEX
Explanations
testicles
This neuron detects explicit references to ball-busting or testicle-focused sexual content (e.g., squeezing, crushing, or otherwise torturing the testicles).
New Auto-Interp
Negative Logits
Uniform
-0.07
infection
-0.06
:]↵
-0.06
-major
-0.06
Rif
-0.06
Light
-0.06
Scoped
-0.06
速
-0.06
('.')↵-0.06
=\""
-0.06
POSITIVE LOGITS
lops
0.07
Hin
0.06
níkem
0.06
cez
0.06
calculator
0.06
lla
0.06
Sala
0.06
regul
0.06
UserID
0.06
jak
0.06
Activations Density 0.004%