INDEX
Explanations
This neuron activates on the informal collective term “folks” used to refer to people.
New Auto-Interp
Negative Logits
Maximum
-0.07
ien
-0.07
ain
-0.07
enr
-0.07
projectile
-0.06
Mayer
-0.06
EE
-0.06
stair
-0.06
ACHER
-0.06
limiting
-0.06
POSITIVE LOGITS
folks
0.16
folk
0.09
kidd
0.07
ุคคล
0.07
Folk
0.07
.Theme
0.07
ск
0.07
foreclosure
0.07
gent
0.07
.cls
0.07
Activations Density 0.004%