INDEX
Explanations
references to poverty and related socioeconomic issues
New Auto-Interp
Negative Logits
åijĺ
-0.17
ä»ĺãģij
-0.16
cha
-0.15
çĸĨ
-0.15
å»Ĭ
-0.14
uild
-0.14
.palette
-0.14
hift
-0.14
commerce
-0.14
Saģlık
-0.14
POSITIVE LOGITS
IPPING
0.16
allev
0.16
AMESPACE
0.15
ards
0.15
ters
0.15
kontakte
0.15
otos
0.15
trap
0.14
Äįet
0.14
/free
0.14
Activations Density 0.023%