INDEX
Explanations
references to "smart" technology and devices
New Auto-Interp
Negative Logits
hma
-0.17
anine
-0.17
olvers
-0.17
htar
-0.16
ymous
-0.16
idis
-0.16
Ïĥι
-0.15
orris
-0.15
hort
-0.15
inous
-0.15
POSITIVE LOGITS
ened
0.29
est
0.28
ening
0.27
phones
0.26
phone
0.25
yp
0.24
PHONE
0.23
watch
0.23
Alec
0.23
ly
0.22
Activations Density 0.009%