INDEX
Explanations
mentions of European entities or concepts
New Auto-Interp
Negative Logits
bek
-0.18
owell
-0.16
utdown
-0.15
uring
-0.14
aling
-0.14
iat
-0.14
arella
-0.14
ibox
-0.13
APER
-0.13
ilk
-0.13
POSITIVE LOGITS
-wide
0.20
antz
0.15
ALLY
0.15
μη
0.13
bsolute
0.13
.snap
0.13
oldt
0.13
paged
0.13
/local
0.13
apist
0.13
Activations Density 0.016%