INDEX
Explanations
origins of phrases
This neuron responds to text in explanatory or definitional contexts—e.g. etymology, meaning statements, and attribution phrases (like “this phrase means,” “originated,” “is often used,” etc.).
New Auto-Interp
Negative Logits
Logistics
-0.07
redundancy
-0.07
ledik
-0.06
AMC
-0.06
mpi
-0.06
gl
-0.06
itudes
-0.06
.Dom
-0.06
elerden
-0.06
くと
-0.05
POSITIVE LOGITS
ималь
0.07
日本
0.07
Kingston
0.06
epis
0.06
consulta
0.06
peel
0.06
DELAY
0.06
Liter
0.06
maker
0.06
Folding
0.06
Activations Density 0.038%