INDEX
Explanations
references to people and their interactions with children and families
New Auto-Interp
Negative Logits
lamaz
-0.20
inia
-0.15
ó
-0.15
Mund
-0.14
azzi
-0.14
gider
-0.14
æµİ
-0.14
erif
-0.14
Winvalid
-0.14
obia
-0.14
POSITIVE LOGITS
Sri
0.33
Lanka
0.32
à¶
0.31
à·
0.31
à
0.26
Sinh
0.25
Diy
0.22
anka
0.20
imal
0.20
سرÛĮ
0.20
Activations Density 0.086%