INDEX
Explanations
references to geographical locations and landmarks
New Auto-Interp
Negative Logits
ippo
-0.17
orest
-0.16
ught
-0.15
enville
-0.14
ypy
-0.14
ippy
-0.14
odom
-0.14
yz
-0.14
koa
-0.13
isson
-0.13
POSITIVE LOGITS
island
0.56
islands
0.50
Island
0.48
Islands
0.42
оÑģÑĤÑĢов
0.35
å³¶
0.31
å²Ľ
0.31
Äijảo
0.29
à¹Ģà¸ģาะ
0.28
ostrov
0.28
Activations Density 0.177%