INDEX
Explanations
political speeches
This neuron detects mentions of warfare or military conflict.
New Auto-Interp
Negative Logits
前
-0.07
whip
-0.07
vron
-0.07
анні
-0.07
pm
-0.07
heat
-0.06
trước
-0.06
CET
-0.06
زو
-0.06
sitcom
-0.06
POSITIVE LOGITS
monthly
0.07
prostor
0.06
surreal
0.06
nání
0.06
quet
0.06
σκευή
0.06
activate
0.06
.flowLayoutPanel
0.06
SEMB
0.06
checkpoint
0.06
Activations Density 0.063%