INDEX
Explanations
references to food, particularly sausage-related content
New Auto-Interp
Negative Logits
Dahl
-0.17
ikki
-0.17
imbus
-0.16
ìķĻ
-0.15
é¾
-0.14
поÑģад
-0.14
orthand
-0.13
ayet
-0.13
modal
-0.13
interp
-0.13
POSITIVE LOGITS
sausage
0.34
cured
0.30
sa
0.27
pork
0.27
sa
0.26
ham
0.26
links
0.25
frank
0.25
bacon
0.24
Cure
0.24
Activations Density 0.061%