INDEX
Explanations
descriptions of individuals or objects in a visual context
New Auto-Interp
Negative Logits
ella
-0.16
ibbon
-0.16
pro
-0.15
core
-0.15
aud
-0.14
po
-0.14
_singleton
-0.13
é
-0.13
mascot
-0.13
éģİ
-0.13
POSITIVE LOGITS
&E
0.17
marked
0.16
boru
0.15
marked
0.15
gnore
0.15
икÑĥ
0.14
skou
0.14
umi
0.14
SGlobal
0.14
TestCategory
0.14
Activations Density 0.152%