INDEX
Explanations
terms related to dowries or financial exchanges in marriages
New Auto-Interp
Negative Logits
nova
-0.19
WOOD
-0.17
sparse
-0.17
ouser
-0.16
erot
-0.16
notice
-0.16
sburg
-0.15
igne
-0.15
nel
-0.15
oney
-0.15
POSITIVE LOGITS
icz
0.28
ls
0.23
orld
0.21
ksi
0.20
dh
0.20
lers
0.19
arde
0.19
ç´ł
0.19
ww
0.18
-wow
0.18
Activations Density 0.037%