INDEX
Explanations
instances of strong sentiments or emotional expressions
New Auto-Interp
Negative Logits
a
-0.15
lag
-0.15
627
-0.15
"
-0.15
iu
-0.15
o
-0.14
oph
-0.14
l
-0.14
stand
-0.14
scan
-0.14
POSITIVE LOGITS
ÐĶÐļ
0.15
drm
0.15
oreach
0.15
bette
0.15
grav
0.15
.byId
0.14
__$
0.14
porr
0.14
mtx
0.14
ripe
0.14
Activations Density 0.038%