INDEX
Explanations
expressions and sentiments related to empathy and personal concern
New Auto-Interp
Negative Logits
themselves
-0.19
allee
-0.16
idual
-0.16
irie
-0.16
igma
-0.16
umes
-0.15
egas
-0.15
nám
-0.15
mey
-0.14
oru
-0.14
POSITIVE LOGITS
myself
0.40
personally
0.24
бÑĥдÑĥ
0.18
zz
0.16
jsem
0.15
zzo
0.15
erior
0.14
ÑħоÑĩÑĥ
0.14
numberWith
0.14
*)((
0.14
Activations Density 0.515%