INDEX
Explanations
elements related to visual perception and description
New Auto-Interp
Negative Logits
rrha
-0.15
SOUND
-0.15
sound
-0.14
æ»
-0.14
fragrance
-0.14
âm
-0.14
scent
-0.14
_sound
-0.14
oui
-0.14
Sound
-0.14
POSITIVE LOGITS
view
0.29
viewed
0.26
viewing
0.23
-view
0.23
perspective
0.22
detail
0.22
looking
0.22
viewers
0.22
appearing
0.21
View
0.21
Activations Density 0.584%