INDEX
Explanations
phrases related to beauty and skincare products
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.16
0.6%
1385
+0.11
0.4%
690
+0.09
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1385
+0.16
0.12
394
+0.11
0.08
690
+0.09
0.06
Negative Logits
<bos>
-2.92
ⓧ
-0.75
/***
-0.74
IActionResult
-0.72
Vegeu
-0.72
PreAuthorize
-0.71
żdy
-0.70
EndProject
-0.69
lateinit
-0.68
MockBean
-0.68
POSITIVE LOGITS
maneu
1.16
maroc
1.16
jawa
1.14
jaya
1.10
impractica
1.06
kuli
1.06
reluct
1.04
pamph
1.02
bandung
1.02
Leyden
1.01
Activations Density 0.980%