INDEX
Explanations
references to law enforcement and administrative statements regarding security and public events
New Auto-Interp
Negative Logits
auffi
-0.94
Monfieur
-0.93
itſelf
-0.90
Jefus
-0.87
iſt
-0.86
Efq
-0.86
reaſon
-0.85
myſelf
-0.85
ſy
-0.85
ſind
-0.84
POSITIVE LOGITS
lawayo
0.61
<eos>
0.59
Zimbabwe
0.57
Zimbabwe
0.56
‘
0.54
0.54
“
0.53
T
0.52
G
0.51
„
0.51
Activations Density 0.087%