INDEX
Explanations
names, especially Chinese names containing the word "Wang"
names of people, particularly in a political context
New Auto-Interp
Negative Logits
props
-0.72
Normandy
-0.67
rave
-0.63
empt
-0.62
AMS
-0.62
appropriate
-0.61
recess
-0.61
store
-0.61
Countdown
-0.61
TODAY
-0.61
POSITIVE LOGITS
Jian
1.55
jiang
1.53
Xia
1.53
wei
1.53
Xiang
1.52
Guang
1.46
Jing
1.45
Yi
1.43
jing
1.43
zhou
1.42
Activations Density 0.094%