INDEX
Explanations
proper nouns, specifically related to a person named Herman
mentions of the name "Herman."
New Auto-Interp
Negative Logits
yrim
-0.75
Thumbnails
-0.73
uren
-0.68
aucus
-0.63
uthor
-0.63
ĵĺ
-0.63
ifter
-0.62
imag
-0.61
oward
-0.61
iful
-0.61
POSITIVE LOGITS
n
1.17
nant
0.91
nian
0.88
ns
0.85
ni
0.84
ium
0.83
stein
0.80
tz
0.79
nis
0.78
cil
0.78
Activations Density 0.049%