INDEX
Explanations
Sports players and teams
The neuron responds to proper names of sports teams and athletes.
New Auto-Interp
Negative Logits
uur
-0.07
foo
-0.07
]}↵
-0.06
effic
-0.06
tourist
-0.06
""↵
-0.06
перс
-0.06
MAR
-0.06
/signup
-0.06
dit
-0.06
POSITIVE LOGITS
_admin
0.07
于
0.07
Vig
0.07
eyond
0.07
(layout
0.07
ustil
0.06
●●
0.06
fellow
0.06
Guaranteed
0.06
비밀
0.06
Activations Density 0.017%