INDEX
Explanations
mentions of the word "man" with varying emphasis
repeated references to the word "man."
New Auto-Interp
Negative Logits
Import
-0.69
PsyNetMessage
-0.69
IVERS
-0.64
Prompt
-0.63
Recommend
-0.60
Girls
-0.60
PT
-0.60
Integ
-0.60
Supplement
-0.59
Democratic
-0.59
POSITIVE LOGITS
hunt
1.54
nered
1.35
uscript
1.29
hood
1.27
gling
1.26
hattan
1.15
osphere
1.12
abase
1.10
orah
1.02
ila
1.01
Activations Density 0.067%