INDEX
Negative Logits
edly
-0.08
Bikini
-0.08
opleiding
-0.08
academia
-0.08
-benar
-0.08
resident
-0.08
plant
-0.08
stell
-0.08
tinta
-0.08
civilization
-0.07
POSITIVE LOGITS
(domain
0.08
solves
0.08
ほ
0.07
Nun
0.07
(addr
0.07
Einzel
0.07
.common
0.07
先
0.07
↵
0.07
(chunk
0.07
Activations Density 0.007%