INDEX
Explanations
references to political parties and their members
New Auto-Interp
Negative Logits
DET
-0.15
.topic
-0.15
ÑĨеп
-0.14
erg
-0.14
u
-0.14
Tin
-0.14
:description
-0.14
VS
-0.14
966
-0.14
åĿĤ
-0.13
POSITIVE LOGITS
andom
0.17
obot
0.16
çŃĴ
0.15
áze
0.14
串
0.14
иной
0.13
Sil
0.13
opr
0.13
splash
0.12
opal
0.12
Activations Density 0.781%