INDEX
Explanations
references to romance and physical condition or state
themes revolving around love and relational concepts
New Auto-Interp
Negative Logits
aren
-0.73
letes
-0.67
ourge
-0.65
redit
-0.64
aper
-0.63
onite
-0.63
oulder
-0.62
illed
-0.60
aped
-0.59
oute
-0.59
POSITIVE LOGITS
guise
1.02
haste
0.84
fashion
0.81
midst
0.78
hurry
0.76
footsteps
0.72
Ø©
0.71
hands
0.71
respects
0.70
form
0.69
Activations Density 0.459%