INDEX
Explanations
expressions related to emotional support and interpersonal connections
New Auto-Interp
Negative Logits
oux
-0.18
chop
-0.15
çķ
-0.15
sof
-0.14
omu
-0.14
stre
-0.14
chaft
-0.14
Voj
-0.14
cke
-0.14
ģn
-0.14
POSITIVE LOGITS
recip
0.15
Å¡tÄĽ
0.14
aghan
0.14
å»ł
0.14
à¥įà¤ķर
0.14
ages
0.14
apan
0.14
èĩº
0.13
ilton
0.13
uar
0.13
Activations Density 1.387%