INDEX
Explanations
references to locations and names related to Brooklyn
New Auto-Interp
Negative Logits
ye
-0.19
si
-0.18
y
-0.18
yp
-0.17
tir
-0.17
screen
-0.17
zi
-0.17
san
-0.16
ingo
-0.15
avan
-0.15
POSITIVE LOGITS
ites
0.24
ite
0.22
ITES
0.20
-based
0.19
igne
0.17
ourd
0.17
ensis
0.16
-bound
0.16
ollo
0.15
-area
0.15
Activations Density 0.022%