INDEX
Explanations
instances of structural features and spatial arrangements in the text
New Auto-Interp
Negative Logits
cou
-0.16
olist
-0.15
ocking
-0.15
zie
-0.14
Benz
-0.14
/html
-0.13
acer
-0.13
ulus
-0.13
desar
-0.13
olib
-0.13
POSITIVE LOGITS
enson
0.15
-linux
0.15
.firebaseapp
0.14
zcze
0.14
aan
0.14
anye
0.14
ordo
0.14
udden
0.14
_PKG
0.13
ìłĪ
0.13
Activations Density 0.242%