INDEX
Explanations
details about male characters in narratives.
The neuron detects occurrences of words indicating the masculine gender (e.g. “maschile,” “masculinos,” etc.).
New Auto-Interp
Negative Logits
xo
-0.07
Ring
-0.07
áno
-0.06
Berry
-0.06
Power
-0.06
zayıf
-0.06
occasion
-0.06
ños
-0.06
.cost
-0.06
requester
-0.06
POSITIVE LOGITS
boys
0.06
-boy
0.06
@section
0.06
,true
0.06
(',');↵0.06
รงเร
0.06
cuts
0.06
.gridx
0.06
localVar
0.06
níku
0.06
Activations Density 0.294%