INDEX
Explanations
locations and directions related to travel and navigation
New Auto-Interp
Negative Logits
upstairs
-0.15
ighth
-0.14
à¸Ļà¸ģ
-0.14
vertical
-0.14
ilm
-0.14
Vertical
-0.13
upside
-0.13
egrator
-0.13
elop
-0.13
Zip
-0.13
POSITIVE LOGITS
via
0.46
via
0.44
Via
0.34
Via
0.33
past
0.32
_via
0.30
past
0.28
VIA
0.27
passing
0.26
across
0.25
Activations Density 0.225%