INDEX
Explanations
names and proper nouns associated with locations and organizations
New Auto-Interp
Negative Logits
inka
-0.17
erdem
-0.14
-div
-0.14
meis
-0.14
[]=$
-0.14
زاÙĨ
-0.14
hani
-0.14
è£ı
-0.14
lien
-0.14
_bias
-0.14
POSITIVE LOGITS
ows
0.15
utto
0.15
owitz
0.14
Pearce
0.14
æĸĹ
0.13
outfit
0.13
iggers
0.13
Consumer
0.13
velt
0.13
Scre
0.13
Activations Density 0.062%