INDEX
Explanations
references to the concept of "phrenology."
New Auto-Interp
Negative Logits
indr
-0.16
گار
-0.16
erotik
-0.16
deniz
-0.15
gw
-0.15
ockey
-0.14
permalink
-0.14
yd
-0.14
rella
-0.14
cf
-0.14
POSITIVE LOGITS
edis
0.17
ypse
0.15
ault
0.15
eci
0.15
pud
0.14
sut
0.14
edio
0.14
677
0.14
amework
0.14
earch
0.14
Activations Density 0.019%