INDEX
Explanations
Christian religious texts
This neuron activates on Bible citation tokens (book names or abbreviations accompanied by chapter and verse numbers).
New Auto-Interp
Negative Logits
/Header
-0.07
Operators
-0.07
했고
-0.06
were
-0.06
hiatus
-0.06
resolutions
-0.06
(keys
-0.06
DAY
-0.06
avis
-0.06
impact
-0.06
POSITIVE LOGITS
และม
0.07
.Groups
0.06
站在
0.06
dealings
0.06
tableName
0.06
を見る
0.06
hears
0.06
unicipio
0.06
λογ
0.06
جات
0.06
Activations Density 0.002%