INDEX
Explanations
This neuron flags metadata in sports match reports—specifically tokens like years/seasons, venues, and opposing team names.
New Auto-Interp
Negative Logits
umbotron
-0.07
_DOWNLOAD
-0.06
enums
-0.06
Frozen
-0.06
اتاق
-0.06
ůsob
-0.06
yPos
-0.06
Volley
-0.06
(disposing
-0.06
893
-0.06
POSITIVE LOGITS
служ
0.07
ول
0.07
ED
0.07
gall
0.07
promised
0.06
ewe
0.06
фика
0.06
Gov
0.06
Registers
0.06
ed
0.06
Activations Density 0.012%