INDEX
Explanations
time-related phrases and durations
New Auto-Interp
Negative Logits
urst
-0.18
ÅĻÃŃm
-0.17
rium
-0.17
ernes
-0.16
gth
-0.15
TECTED
-0.15
Ñģли
-0.15
adar
-0.15
aze
-0.14
unny
-0.14
POSITIVE LOGITS
months
0.19
weeks
0.19
month
0.17
week
0.16
Äįka
0.16
esc
0.16
days
0.15
oute
0.15
ignon
0.15
sitting
0.15
Activations Density 0.092%