INDEX
Explanations
references to decorative and functional window treatments
New Auto-Interp
Negative Logits
humanoid
-0.16
bish
-0.16
jacket
-0.15
hete
-0.14
.getOwnProperty
-0.13
plaster
-0.13
Ãłm
-0.13
Episodes
-0.13
енÑĮ
-0.13
kaps
-0.13
POSITIVE LOGITS
blinds
0.27
window
0.25
Window
0.23
/window
0.23
curtains
0.22
curtain
0.22
-window
0.21
WINDOW
0.21
WINDOW
0.21
.Window
0.20
Activations Density 0.042%