INDEX
Explanations
references to a person's native place or hometown
New Auto-Interp
Negative Logits
,proto
-0.18
ekl
-0.17
iani
-0.16
lean
-0.15
reau
-0.15
abin
-0.15
swer
-0.14
áng
-0.14
airy
-0.14
lli
-0.14
POSITIVE LOGITS
imity
0.15
base
0.14
ÏĥÏįν
0.14
.ps
0.13
åΏ
0.13
feelings
0.13
depreci
0.13
216
0.13
base
0.13
hometown
0.13
Activations Density 0.044%