INDEX
Explanations
words related to paint
references to painting or related activities
New Auto-Interp
Negative Logits
zbek
-0.76
AMY
-0.76
nesses
-0.75
-|
-0.74
ambers
-0.72
doms
-0.71
WATCHED
-0.69
cgi
-0.69
ITED
-0.68
Kenyan
-0.67
POSITIVE LOGITS
brush
1.13
thinner
1.10
paints
0.99
paint
0.88
painting
0.87
painter
0.84
painted
0.84
acrylic
0.82
Painter
0.80
bats
0.79
Activations Density 0.015%