INDEX
Explanations
famous or many international words
New Auto-Interp
Negative Logits
unless
0.88
non
0.83
gently
0.82
empty
0.82
if
0.80
unspecified
0.79
preferably
0.77
_
0.76
lightly
0.76
should
0.75
POSITIVE LOGITS
famoso
1.25
famosa
1.23
বিখ্যাত
1.19
terkenal
1.15
célèbre
1.14
знамени
1.12
számos
1.12
दुनिया
1.05
세계
1.05
многих
1.05
Activations Density 0.001%