INDEX
Explanations
expressions of social connection and interaction
New Auto-Interp
Negative Logits
elah
-0.16
/operators
-0.14
:async
-0.14
sons
-0.14
ouble
-0.14
ulton
-0.14
маз
-0.14
èĢħçļĦ
-0.14
_observer
-0.14
#End
-0.14
POSITIVE LOGITS
nhau
0.25
other
0.21
them
0.20
people
0.20
him
0.19
someone
0.19
/about
0.18
somebody
0.17
others
0.17
him
0.17
Activations Density 0.274%