INDEX
Explanations
political parties/ideologies
This neuron detects mentions of political party names and coalitions.
New Auto-Interp
Negative Logits
ане
-0.07
Crossing
-0.07
SKI
-0.07
says
-0.07
ілі
-0.07
>";↵
-0.06
TESTING
-0.06
('/')↵-0.06
_capability
-0.06
orta
-0.06
POSITIVE LOGITS
needing
0.06
employment
0.06
فکر
0.06
typedef
0.06
0.06
свеж
0.06
.xyz
0.06
nfl
0.05
%H
0.05
ятель
0.05
Activations Density 0.022%