INDEX
Explanations
instances of hugging or physical affection between characters
New Auto-Interp
Negative Logits
äl
-0.17
è¦
-0.17
Dün
-0.16
ause
-0.14
chia
-0.14
lun
-0.14
outines
-0.14
799
-0.14
trous
-0.14
amework
-0.14
POSITIVE LOGITS
INI
0.17
/back
0.15
δεÏĤ
0.15
otes
0.15
atsby
0.14
icon
0.14
ENS
0.14
cons
0.14
uche
0.14
pile
0.14
Activations Density 0.093%