INDEX
Explanations
The neuron detects mentions of a party choosing or electing to represent themselves (self-representation/pro se representation).
New Auto-Interp
Negative Logits
irradi
-0.07
'<
-0.06
IRS
-0.06
jointly
-0.06
silly
-0.06
iger
-0.06
Mayer
-0.06
september
-0.06
sampler
-0.06
.SUB
-0.06
POSITIVE LOGITS
��
0.06
także
0.06
libc
0.06
sức
0.06
olmayan
0.06
ypical
0.06
,$_
0.06
Чтобы
0.06
userID
0.06
creativecommons
0.06
Activations Density 0.002%