INDEX
Explanations
references to political claims and discussions surrounding political candidates
New Auto-Interp
Negative Logits
uant
-0.18
utzer
-0.17
jee
-0.15
oldt
-0.15
ynet
-0.15
enia
-0.15
ãĥ³ãĤ¬
-0.14
onymous
-0.14
ingleton
-0.14
getExtension
-0.14
POSITIVE LOGITS
party
0.21
aspir
0.19
ticket
0.19
Peoples
0.18
Party
0.17
party
0.16
Party
0.15
contest
0.14
candidate
0.14
ure
0.14
Activations Density 0.017%