INDEX
Explanations
mentions of ginger and gingerbread
New Auto-Interp
Negative Logits
swire
-0.20
vel
-0.16
FD
-0.15
rious
-0.15
PLY
-0.14
ravel
-0.14
ucch
-0.14
.cc
-0.14
ÃŃst
-0.14
ientos
-0.13
POSITIVE LOGITS
bread
0.38
ale
0.21
ibre
0.20
Ale
0.19
Snap
0.18
root
0.18
snaps
0.18
root
0.18
snap
0.17
ochen
0.17
Activations Density 0.006%