INDEX
Explanations
the name "Allison" in text
proper nouns, particularly names of individuals
New Auto-Interp
Negative Logits
rum
-0.90
Osw
-0.84
umbn
-0.77
bread
-0.76
keye
-0.75
cent
-0.74
lasses
-0.70
isphere
-0.69
rab
-0.67
ener
-0.66
POSITIVE LOGITS
idian
0.83
naire
0.75
yz
0.75
ombies
0.75
iguous
0.72
Dear
0.71
iesel
0.70
Hail
0.70
ocaly
0.70
tsky
0.69
Activations Density 0.020%