INDEX
Explanations
This neuron responds to the occurrence of “Circuit” in appellate‐court citations.
New Auto-Interp
Negative Logits
.tolist
-0.06
_HIT
-0.06
shines
-0.06
ㅡ
-0.06
Allocate
-0.06
项目
-0.06
-follow
-0.06
_WIDTH
-0.06
-------
-0.06
ิดข
-0.06
POSITIVE LOGITS
customs
0.07
odmít
0.07
Berkeley
0.06
.slim
0.06
widening
0.06
CTIONS
0.06
512
0.06
_PCIE
0.06
sóng
0.06
peace
0.06
Activations Density 0.001%