INDEX
Explanations
terms related to specific industries or professions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1343
+0.16
0.5%
1842
+0.14
0.4%
674
+0.14
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1842
+0.16
0.06
59
+0.14
0.05
227
+0.14
0.07
Negative Logits
prét
-0.81
unden
-0.79
effe
-0.78
asics
-0.77
casio
-0.77
madden
-0.75
aveug
-0.75
wien
-0.74
Freuden
-0.74
alre
-0.73
POSITIVE LOGITS
Tikang
0.79
ⓧ
0.71
bardziej
0.68
商品説明
0.66
Datuak
0.66
habang
0.66
astéro
0.64
porcent
0.63
jeste
0.63
costumi
0.63
Activations Density 0.428%