INDEX
Explanations
phrases related to windows or window-related actions
New Auto-Interp
Negative Logits
avez
-0.86
currency
-0.79
ghan
-0.79
Haram
-0.76
IGH
-0.67
uca
-0.67
DonaldTrump
-0.65
yon
-0.65
asma
-0.65
icit
-0.65
POSITIVE LOGITS
pane
1.11
sill
1.04
glass
0.96
tint
0.89
windows
0.86
wip
0.84
door
0.82
bars
0.82
windows
0.82
openings
0.81
Activations Density 0.056%