INDEX
Explanations
mentions of the location Wu or its dialect, indicating a focus on regional references in the text
New Auto-Interp
Negative Logits
ood
-0.17
yre
-0.16
koli
-0.16
ısız
-0.15
erva
-0.14
EST
-0.14
oust
-0.14
ChangeEvent
-0.14
cura
-0.14
estring
-0.14
POSITIVE LOGITS
amu
0.17
cre
0.15
caz
0.14
Cre
0.14
ensch
0.14
Andrea
0.14
han
0.14
ca
0.14
âĨ
0.13
ca
0.13
Activations Density 0.009%