INDEX
Explanations
mentions of guests and authors in various contexts, likely related to their appearances or contributions
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.18
1.0%
871
+0.10
0.6%
1573
+0.09
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
260
+0.18
0.03
484
+0.10
0.03
851
+0.09
0.03
Negative Logits
<bos>
-3.15
ⓧ
-0.85
/***
-0.75
/**
-0.74
lateinit
-0.70
//{
-0.68
Enllaços
-0.64
public
-0.64
///**
-0.63
/*++
-0.62
POSITIVE LOGITS
maroc
1.21
Minang
1.19
stockholm
1.19
bandung
1.18
wien
1.16
maneu
1.16
Khart
1.14
riva
1.14
lidl
1.13
tucson
1.12
Activations Density 0.087%