INDEX
Explanations
product descriptions of vape flavors
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1343
+0.10
0.3%
453
+0.09
0.3%
1403
+0.09
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
609
+0.10
0.05
1050
+0.09
0.04
1925
+0.09
0.04
Negative Logits
thut
-1.12
vainly
-1.09
impractica
-1.09
reconno
-1.07
encomp
-1.06
intersper
-1.05
impra
-1.05
apprehen
-1.05
depic
-1.04
maneu
-1.04
POSITIVE LOGITS
price
0.80
">$
0.78
Price
0.77
>$
0.74
\$
0.74
price
0.71
$\$
0.70
$/
0.69
{\$0.69
/$
0.67
Activations Density 0.224%