INDEX
Explanations
words related to clothing, particularly frocks
occurrences of the substring "fro."
New Auto-Interp
Negative Logits
UAL
-0.72
animous
-0.71
yrinth
-0.68
ually
-0.67
****************
-0.65
ities
-0.65
itutional
-0.65
Oriental
-0.64
ividual
-0.63
ifying
-0.62
POSITIVE LOGITS
ggy
0.97
zens
0.97
sts
0.91
ogle
0.91
bie
0.89
¸
0.88
cade
0.88
bies
0.86
terness
0.83
lio
0.81
Activations Density 0.007%