INDEX
Explanations
the phrase "a couple of"
the phrase "a couple of."
New Auto-Interp
Negative Logits
acus
-0.75
=]
-0.67
Painter
-0.67
otype
-0.64
Males
-0.63
aste
-0.63
fair
-0.63
pecting
-0.62
hend
-0.61
ared
-0.61
POSITIVE LOGITS
dozen
0.76
additional
0.70
extra
0.70
announcements
0.68
things
0.68
ricular
0.66
entries
0.65
teen
0.64
hundred
0.64
unanswered
0.63
Activations Density 0.158%