INDEX
Explanations
terms related to presidential candidates and elections
New Auto-Interp
Negative Logits
fak
-0.17
imar
-0.17
ocz
-0.16
Ages
-0.16
laus
-0.15
rous
-0.15
igu
-0.14
olit
-0.14
Venue
-0.14
Truy
-0.14
POSITIVE LOGITS
èά
0.15
ä»ķ
0.15
anza
0.14
¹
0.14
ty
0.14
ÄĽti
0.13
abin
0.13
poly
0.13
teri
0.13
opoly
0.13
Activations Density 0.017%