INDEX
Explanations
mentions of relationships or connections between people
references to romantic or passionate relationships
New Auto-Interp
Negative Logits
SPONSORED
-0.80
rudimentary
-0.76
UL
-0.75
é¾
-0.72
enei
-0.72
orio
-0.71
ural
-0.71
ursed
-0.69
issue
-0.68
ulhu
-0.66
POSITIVE LOGITS
lover
1.11
lovers
1.08
affair
0.80
club
0.79
friend
0.78
mistress
0.78
Lover
0.77
atical
0.77
rejoice
0.74
passionately
0.73
Activations Density 0.009%