INDEX
Explanations
needs, benefits, or states to improve
New Auto-Interp
Negative Logits
wollte
0.50
believes
0.46
believe
0.45
believing
0.44
awọn
0.43
sellest
0.43
wanted
0.42
quieres
0.42
võib
0.40
udało
0.40
POSITIVE LOGITS
its
0.63
Its
0.56
Its
0.54
它的
0.47
izarse
0.47
использоваться
0.45
быть
0.44
因为它
0.44
적용
0.43
être
0.43
Activations Density 0.099%