INDEX
Explanations
the word "uck" in the middle of longer words
the repeated occurrence of the syllable "uck"
New Auto-Interp
Negative Logits
senal
-0.81
ccording
-0.75
streng
-0.73
ank
-0.69
warr
-0.65
circ
-0.65
Seym
-0.64
attribution
-0.63
APH
-0.63
cryst
-0.62
POSITIVE LOGITS
erman
1.06
weed
0.95
olds
0.90
eries
0.88
eez
0.88
abee
0.88
lesh
0.84
olls
0.83
ield
0.83
NOW
0.82
Activations Density 0.018%