INDEX
Explanations
geographic locations and directions
New Auto-Interp
Negative Logits
bes
-0.15
ãĥ¼ãĤ¸
-0.14
orts
-0.13
ltra
-0.13
/src
-0.13
.URI
-0.13
/native
-0.12
Forward
-0.12
ucid
-0.12
صÙĨ
-0.12
POSITIVE LOGITS
north
0.48
south
0.48
west
0.47
sou
0.45
nor
0.44
NW
0.44
east
0.43
nor
0.42
northwest
0.41
south
0.41
Activations Density 0.315%