INDEX
Explanations
names of political candidates and their attributes
New Auto-Interp
Negative Logits
unders
-0.16
.inst
-0.16
elman
-0.15
+offset
-0.15
ottage
-0.15
uner
-0.15
.lt
-0.14
Ïħν
-0.14
UNDER
-0.14
strand
-0.14
POSITIVE LOGITS
pires
0.16
Harr
0.15
{text0.15
hare
0.15
Sky
0.15
pit
0.14
Vend
0.14
sky
0.14
sky
0.14
Seg
0.14
Activations Density 0.011%