INDEX
Explanations
references to individuals named Martin
New Auto-Interp
Negative Logits
.portal
-0.17
SizeMode
-0.16
ixin
-0.16
asley
-0.15
kart
-0.15
ReadWrite
-0.15
ivant
-0.15
ÐŁÐļ
-0.15
cats
-0.15
idas
-0.14
POSITIVE LOGITS
Luther
0.28
ique
0.26
elli
0.23
borough
0.20
IQUE
0.19
ho
0.18
engo
0.17
uzzi
0.17
iqu
0.17
ss
0.16
Activations Density 0.011%