INDEX
Explanations
elements related to strong emotional expressions and personal connections
New Auto-Interp
Negative Logits
agner
-0.19
aura
-0.15
Establishment
-0.15
Preferred
-0.14
amburg
-0.14
ever
-0.14
/sm
-0.14
leich
-0.14
publication
-0.14
423
-0.14
POSITIVE LOGITS
anine
0.16
DRV
0.16
Insensitive
0.15
juan
0.15
ç´¹
0.14
Nad
0.14
Timothy
0.14
atin
0.14
affe
0.14
razier
0.14
Activations Density 0.014%