INDEX
Explanations
This neuron does not detect any specific token patterns—it remains inactive for all inputs.
New Auto-Interp
Negative Logits
전
-0.07
considerations
-0.07
__;↵
-0.06
闭
-0.06
証
-0.06
_ob
-0.06
underneath
-0.06
singers
-0.06
.od
-0.06
undergone
-0.06
POSITIVE LOGITS
inplace
0.08
spirituality
0.07
enght
0.07
Tropical
0.07
Комп
0.06
Backpack
0.06
preamble
0.06
Agile
0.06
Pak
0.06
_CAMERA
0.06
Activations Density 0.000%