INDEX
Explanations
attributes and qualities related to artistic expression
New Auto-Interp
Negative Logits
enjoyment
-0.16
Enjoy
-0.15
aden
-0.15
иÑģÑĮ
-0.14
plr
-0.14
äs
-0.14
anter
-0.14
öh
-0.14
Surround
-0.14
apers
-0.14
POSITIVE LOGITS
deep
0.31
depth
0.26
vulnerability
0.26
deep
0.26
deeper
0.25
depths
0.25
deepest
0.24
depth
0.24
vulner
0.23
vulnerable
0.23
Activations Density 0.012%