INDEX
Explanations
The neuron selectively activates on occurrences of the word “player” or “players” (mostly in category listings).
New Auto-Interp
Negative Logits
GNU
-0.07
ertoire
-0.06
90
-0.06
ningen
-0.06
ets
-0.06
ka
-0.06
enne
-0.06
erspective
-0.06
foregoing
-0.06
Bent
-0.06
POSITIVE LOGITS
w
0.07
vigor
0.07
.workspace
0.07
disgrace
0.07
.Btn
0.06
Boutique
0.06
itemView
0.06
.Drop
0.06
(dllexport
0.06
каш
0.06
Activations Density 0.004%