INDEX
Explanations
founder or co-founder related information
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.18
1.0%
1778
+0.12
0.7%
1068
+0.11
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1778
+0.18
0.04
1194
+0.12
0.03
1068
+0.11
0.03
Negative Logits
<bos>
-2.97
꿔
-0.69
/**
-0.67
/**
-0.65
public
-0.65
المل
-0.64
assist
-0.64
protected
-0.62
springfox
-0.62
///
-0.62
POSITIVE LOGITS
affor
1.74
bandung
1.74
impra
1.70
increa
1.70
unlaw
1.65
lele
1.65
Juf
1.65
maroc
1.64
Minang
1.62
stockholm
1.62
Activations Density 0.134%