INDEX
Explanations
phrases indicating importance or emphasis
the phrase "it is important to note that."
New Auto-Interp
Negative Logits
raq
-0.70
oses
-0.66
zman
-0.65
Pont
-0.64
agram
-0.63
icum
-0.62
eal
-0.61
aukee
-0.61
pec
-0.61
mouth
-0.60
POSITIVE LOGITS
although
0.85
there
0.80
"[
0.75
chery
0.71
whereas
0.70
we
0.69
they
0.68
despite
0.68
unless
0.66
whilst
0.65
Activations Density 0.182%