INDEX
Explanations
husband or wife relationships
New Auto-Interp
Negative Logits
Friendship
0.54
Cerc
0.47
amistad
0.47
friendship
0.44
Zu
0.41
afford
0.40
wort
0.40
friend
0.40
汩
0.39
assi
0.39
POSITIVE LOGITS
husband
0.99
স্বামীর
0.97
wife
0.96
spouse
0.93
manžel
0.91
супру
0.90
spouses
0.86
husbands
0.86
wives
0.85
স্ত্রীর
0.85
Activations Density 0.147%