INDEX
Explanations
mentions of gingerbread and ginger-related content
New Auto-Interp
Negative Logits
swire
-0.17
aging
-0.15
PLY
-0.15
uros
-0.15
rious
-0.14
PLE
-0.14
Hugo
-0.14
ucch
-0.14
gages
-0.14
ended
-0.14
POSITIVE LOGITS
bread
0.34
ibre
0.20
ale
0.18
ed
0.17
orman
0.17
ãĥŃãĥ¼
0.16
haired
0.16
Snap
0.16
loff
0.16
ochen
0.15
Activations Density 0.007%