INDEX
Explanations
references to political entities with variations in form
occurrences of the term "republic" and its variations
New Auto-Interp
Negative Logits
Oracle
-0.83
Event
-0.82
Kut
-0.72
Aid
-0.71
Dialogue
-0.71
Visual
-0.71
Ack
-0.70
Bench
-0.70
Technical
-0.69
Cube
-0.69
POSITIVE LOGITS
republic
1.16
rats
0.95
Seym
0.89
edom
0.83
republican
0.82
elector
0.80
monarch
0.80
hess
0.79
monarchy
0.77
inally
0.77
Activations Density 0.013%