INDEX
Explanations
prepositions indicating location or direction
New Auto-Interp
Negative Logits
ksam
-0.16
essel
-0.15
inne
-0.14
tat
-0.14
ingerprint
-0.14
ogl
-0.14
ozor
-0.14
unst
-0.14
jin
-0.14
PIL
-0.14
POSITIVE LOGITS
overe
0.16
oll
0.16
rchive
0.16
eway
0.16
forb
0.15
aland
0.14
ega
0.14
fid
0.14
DTD
0.14
Col
0.14
Activations Density 0.008%