INDEX
Explanations
concepts related to routine or consistent activities
New Auto-Interp
Negative Logits
chod
-0.17
ason
-0.16
acen
-0.16
elson
-0.15
Stage
-0.14
oji
-0.14
/ubuntu
-0.14
etr
-0.14
unas
-0.14
Vien
-0.14
POSITIVE LOGITS
mente
0.26
ity
0.23
xuyên
0.21
ly
0.20
ities
0.19
ness
0.19
üstü
0.18
regular
0.17
/month
0.17
ily
0.17
Activations Density 0.050%