INDEX
Explanations
painting surfaces and decorative finishes
New Auto-Interp
Negative Logits
seksual
0.63
misog
0.63
qubits
0.62
amenazas
0.61
citoyens
0.59
هاکي
0.59
громадян
0.58
sadistic
0.57
kidnapping
0.56
habeas
0.55
POSITIVE LOGITS
paint
1.16
wall
1.05
Paint
1.05
stucco
1.02
drywall
1.02
Painting
1.01
Paint
1.00
Decorative
0.98
decorative
0.98
دیوار
0.96
Activations Density 0.050%