INDEX
Explanations
wires, cables
The main thing this neuron does is find mentions of cables, cords, or wires.
New Auto-Interp
Negative Logits
_dd
-0.07
Mana
-0.07
ancora
-0.06
.getP
-0.06
المت
-0.06
перет
-0.06
vekili
-0.06
Mana
-0.06
Tango
-0.06
kil
-0.06
POSITIVE LOGITS
absent
0.07
cord
0.07
connecting
0.07
春
0.06
chased
0.06
Following
0.06
/pub
0.06
0.06
habe
0.06
_response
0.06
Activations Density 0.010%