INDEX
Explanations
occurrences of the word "elected" or its variants in the context of political positions
New Auto-Interp
Negative Logits
ovel
-0.15
³
-0.15
lek
-0.14
erek
-0.14
sson
-0.14
hide
-0.14
TEL
-0.14
acen
-0.14
tend
-0.13
Erotik
-0.13
POSITIVE LOGITS
eer
0.17
lernen
0.14
ayi
0.14
GetMethod
0.14
Antar
0.14
lotte
0.14
à¥Ĥत
0.14
efault
0.14
imir
0.14
Thr
0.14
Activations Density 0.005%