INDEX
Explanations
names of people and titles, particularly in a political or sports context
New Auto-Interp
Negative Logits
iquid
-0.15
_:*
-0.15
)prepare
-0.14
abant
-0.14
306
-0.14
itness
-0.14
reau
-0.14
огÑĢам
-0.14
ertas
-0.13
Zus
-0.13
POSITIVE LOGITS
col
0.14
bury
0.14
azes
0.14
gorm
0.14
amble
0.14
Alfred
0.14
al
0.14
yle
0.14
throne
0.14
Os
0.14
Activations Density 0.217%