INDEX
Explanations
This neuron flags mentions of screens (and related terms like “blue light”) in the text.
New Auto-Interp
Negative Logits
slightest
-0.06
개최
-0.06
dục
-0.06
toFloat
-0.06
z
-0.06
ёр
-0.06
Snow
-0.06
Bentley
-0.06
Daw
-0.06
recruits
-0.06
POSITIVE LOGITS
-transitional
0.07
unfinished
0.07
imet
0.07
/*↵
0.06
QueryString
0.06
�
0.06
/use
0.06
nl
0.06
mom
0.06
.Pay
0.06
Activations Density 0.004%