INDEX
Explanations
phrases indicating possession or ownership
New Auto-Interp
Negative Logits
allen
-0.17
éĴŁ
-0.16
uer
-0.15
atee
-0.15
erring
-0.14
ep
-0.14
iphy
-0.14
izando
-0.14
mer
-0.14
recent
-0.14
POSITIVE LOGITS
pé
0.14
ÙĨÙħ
0.14
Tru
0.14
uco
0.14
lac
0.14
roid
0.14
/misc
0.14
kancel
0.13
oppers
0.13
kvin
0.13
Activations Density 0.038%