INDEX
Explanations
references to locations and institutions in Oakland
New Auto-Interp
Negative Logits
Bend
-0.17
.eulerAngles
-0.16
ekk
-0.16
λλι
-0.15
uta
-0.15
åĸ
-0.15
citiz
-0.14
.sb
-0.14
.gb
-0.14
uten
-0.14
POSITIVE LOGITS
Oakland
0.40
Oak
0.33
Berkeley
0.28
oak
0.28
Oak
0.25
Hayward
0.24
510
0.24
Pied
0.22
Liver
0.21
Walnut
0.21
Activations Density 0.047%