INDEX
Explanations
instances of the name "Scott" with particularly high activation levels
mentions of the name "Scott."
New Auto-Interp
Negative Logits
chid
-0.79
senal
-0.79
pread
-0.78
åĤ
-0.74
ngth
-0.72
netflix
-0.72
ĺħ
-0.70
ften
-0.70
iotic
-0.67
ĵĺ
-0.67
POSITIVE LOGITS
Scott
1.09
Walker
0.97
inelli
0.97
Morrison
0.95
Pilgrim
0.94
ards
0.92
Pruitt
0.92
enson
0.91
inson
0.89
Snyder
0.89
Activations Density 0.007%