INDEX
Explanations
instances of political power struggles and changes in leadership
New Auto-Interp
Negative Logits
المناصب
-0.77
المعيارى
-0.75
<unused41>
-0.75
snippetHide
-0.75
<unused8>
-0.74
<pad>
-0.74
<unused14>
-0.74
<unused43>
-0.74
<unused28>
-0.74
[@BOS@]
-0.74
POSITIVE LOGITS
unsuccessfully
0.36
was
0.32
berat
0.31
Schulte
0.31
caused
0.31
persuaded
0.29
Palacios
0.28
he
0.27
partita
0.26
secretly
0.26
Activations Density 1.081%