INDEX
Explanations
names of individuals
instances of the name "Emily" and other related names
New Auto-Interp
Negative Logits
lasses
-0.81
ername
-0.80
nings
-0.76
xual
-0.75
ebin
-0.73
tenance
-0.72
nesses
-0.72
DragonMagazine
-0.71
zsche
-0.71
hips
-0.69
POSITIVE LOGITS
Lak
1.28
Dickinson
1.14
issance
0.81
gown
0.78
sburg
0.75
itte
0.73
pton
0.73
gdala
0.72
issions
0.72
endi
0.71
Activations Density 0.026%