INDEX
Explanations
numeric values and age-related information
New Auto-Interp
Negative Logits
ello
-0.17
eriod
-0.16
IEL
-0.15
hai
-0.15
ä¸įå¾Ĺ
-0.15
åı£
-0.14
vang
-0.14
roma
-0.14
ANGO
-0.13
èĮĤ
-0.13
POSITIVE LOGITS
ãģ«ãģĭ
0.17
δÏģο
0.17
ushima
0.17
táºŃn
0.16
udo
0.16
uchar
0.16
-tracking
0.15
alike
0.15
indre
0.15
ãĤ·ãĥ§ãĥ³
0.14
Activations Density 0.159%