INDEX
Explanations
expressions of desire or longing
New Auto-Interp
Negative Logits
shan
-0.22
ader
-0.18
bá»ı
-0.15
elu
-0.15
actively
-0.14
outu
-0.14
Wer
-0.14
isors
-0.14
asures
-0.14
riter
-0.14
POSITIVE LOGITS
ful
0.25
bone
0.24
fully
0.20
FUL
0.19
fulness
0.18
inkle
0.17
omers
0.17
584
0.17
pered
0.16
(es
0.16
Activations Density 0.010%