INDEX
Explanations
ways in which food is prepared or cooked
New Auto-Interp
Negative Logits
aneous
-0.71
urers
-0.70
Nadu
-0.66
aries
-0.66
uminati
-0.63
URA
-0.63
repetition
-0.62
uated
-0.62
uate
-0.61
Dull
-0.61
POSITIVE LOGITS
ccoli
1.25
oks
1.19
thel
1.09
dy
1.05
keye
1.03
swer
1.01
chet
0.97
gue
0.95
thren
0.92
oke
0.92
Activations Density 0.062%