INDEX
Explanations
Biographical information
The neuron activates on verbs that describe assuming or holding official roles or positions in a person’s biography (e.g. “became,” “retained,” “served”).
New Auto-Interp
Negative Logits
옷
-0.06
знач
-0.06
Над
-0.06
닝
-0.06
Zur
-0.06
チェ
-0.06
976
-0.06
تعد
-0.06
452
-0.05
mpz
-0.05
POSITIVE LOGITS
τια
0.07
Multiplicity
0.07
DVD
0.07
TRY
0.06
seasoning
0.06
Россий
0.06
.Peek
0.06
(pointer
0.06
ाहरण
0.06
patterns
0.06
Activations Density 0.042%