INDEX
Explanations
recurring mentions of the name "Liz" and its variations
New Auto-Interp
Negative Logits
esin
-0.16
igne
-0.15
ustil
-0.14
tehdy
-0.14
OCKET
-0.14
isted
-0.14
isting
-0.14
esub
-0.14
esini
-0.14
_heat
-0.13
POSITIVE LOGITS
y
0.21
zi
0.19
tro
0.17
yo
0.17
zy
0.17
yor
0.16
irit
0.15
quierda
0.15
ings
0.15
zen
0.15
Activations Density 0.024%