INDEX
Explanations
Creation of political entities
This neuron detects numeric tokens—especially years, dates, and other numbers indicating when something was created, abolished, or took place.
New Auto-Interp
Negative Logits
학교
-0.07
ypi
-0.06
^n
-0.06
От
-0.06
місце
-0.06
كي
-0.06
tracts
-0.06
lifting
-0.06
Apache
-0.06
watershed
-0.06
POSITIVE LOGITS
jab
0.08
cabo
0.07
pars
0.07
lush
0.06
(Result
0.06
založ
0.06
податков
0.06
lá
0.06
ngoài
0.06
�
0.06
Activations Density 0.026%