INDEX
Explanations
The neuron activates on mentions of administrative “County” labels in location descriptions.
New Auto-Interp
Negative Logits
Kut
-0.07
FE
-0.07
zig
-0.06
hel
-0.06
astype
-0.06
lion
-0.06
signaling
-0.06
ol
-0.06
astro
-0.06
Polymer
-0.06
POSITIVE LOGITS
래
0.06
و
0.06
=$(
0.06
enact
0.06
بعد
0.06
Pokud
0.06
ECC
0.06
(freq
0.06
险
0.06
-%
0.06
Activations Density 0.002%