INDEX
Explanations
words related to emotional states or feelings
New Auto-Interp
Negative Logits
vester
-0.19
orer
-0.16
alling
-0.15
ainty
-0.14
aldo
-0.14
ulent
-0.14
toi
-0.14
raith
-0.14
Herr
-0.14
arat
-0.14
POSITIVE LOGITS
اسطة
0.18
okt
0.16
konus
0.16
dÅĻÃŃ
0.15
pys
0.15
'ye
0.15
’ye
0.15
purch
0.14
ÑĩеÑģ
0.14
ÑĩаÑĤ
0.14
Activations Density 0.019%