INDEX
Explanations
occurrences of time-related phrases and sharing information
New Auto-Interp
Negative Logits
ritt
-0.16
opoulos
-0.16
olon
-0.16
assis
-0.16
Mush
-0.15
Lah
-0.15
abo
-0.15
acher
-0.15
ooter
-0.15
VEL
-0.15
POSITIVE LOGITS
SB
0.19
èŃľ
0.18
SB
0.17
uby
0.16
sb
0.16
æŃ¡
0.16
vik
0.16
_sb
0.15
.sb
0.14
anner
0.14
Activations Density 0.006%