INDEX
Explanations
common words
The neuron detects numeric tokens, especially decimal‐point numbers (floating‐point values).
New Auto-Interp
Negative Logits
城
-0.08
########################################################################
-0.06
autobi
-0.06
.Des
-0.06
tingham
-0.06
BIG
-0.06
_builtin
-0.06
-single
-0.06
bases
-0.06
ptime
-0.06
POSITIVE LOGITS
nigeria
0.06
호
0.06
анка
0.06
ynchronously
0.06
iators
0.06
люд
0.06
embourg
0.06
inine
0.06
_BUFFER
0.06
Republican
0.06
Activations Density 0.080%