INDEX
Explanations
words related to the action of painting or color application
references to paint and painting
New Auto-Interp
Negative Logits
nesses
-0.76
AMY
-0.74
cgi
-0.72
irl
-0.70
doms
-0.69
andr
-0.69
ness
-0.66
Zealand
-0.65
wegian
-0.63
Proceedings
-0.62
POSITIVE LOGITS
brush
1.20
paint
1.15
paints
1.10
painting
1.00
painted
0.96
Paint
0.96
painter
0.90
Painting
0.86
thinner
0.83
isans
0.82
Activations Density 0.011%