INDEX
Explanations
references to graphic design or graphic-related content
New Auto-Interp
Negative Logits
yr
-0.21
wick
-0.17
geb
-0.16
ismic
-0.16
ib
-0.15
bri
-0.15
frey
-0.15
ym
-0.15
yt
-0.15
yer
-0.15
POSITIVE LOGITS
arts
0.20
/graph
0.18
novels
0.17
designer
0.17
/video
0.17
novel
0.17
Arts
0.16
-intensive
0.16
ENE
0.16
Novel
0.16
Activations Density 0.011%