INDEX
Explanations
institutions related to education and healthcare
New Auto-Interp
Negative Logits
zek
-0.14
uco
-0.13
öz
-0.13
224
-0.13
WARE
-0.12
opal
-0.12
lech
-0.12
met
-0.12
edom
-0.12
_INST
-0.12
POSITIVE LOGITS
located
0.24
ợ
0.20
situated
0.17
near
0.17
located
0.17
ubic
0.17
igar
0.17
à¹ĥà¸Ļ
0.16
ä½įæĸ¼
0.16
igure
0.15
Activations Density 0.155%