INDEX
Explanations
This neuron remains inactive and does not detect any specific tokens or patterns.
New Auto-Interp
Negative Logits
ده
-0.07
.Sum
-0.07
fall
-0.07
deepcopy
-0.06
飾
-0.06
param
-0.06
隨
-0.06
Kut
-0.06
complicated
-0.06
умов
-0.06
POSITIVE LOGITS
roker
0.08
Fields
0.07
rens
0.07
인
0.07
ื
0.06
InstanceOf
0.06
tanggal
0.06
Schools
0.06
emploi
0.06
求购
0.06
Activations Density 0.019%