INDEX
Explanations
News/Current events
The neuron activates on tokens naming political offices, election candidates, or related presidential‐election terms.
New Auto-Interp
Negative Logits
>New
-0.07
_quotes
-0.06
율
-0.06
Abb
-0.06
Parkway
-0.06
Payload
-0.06
�
-0.06
rg
-0.06
mom
-0.06
time
-0.06
POSITIVE LOGITS
얼
0.06
يناير
0.06
PROF
0.06
Benef
0.06
чет
0.06
можуть
0.06
coercion
0.06
NOAA
0.06
Ike
0.06
seinen
0.06
Activations Density 0.202%