INDEX
Explanations
topics related to various events, locations, and names, possibly related to announcements, performances, classes, and general information within a community
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.09
0.4%
1047
+0.07
0.3%
1335
+0.06
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1640
+0.09
0.05
237
+0.07
0.05
1043
+0.06
0.04
Negative Logits
<bos>
-1.05
/*
-0.74
public
-0.71
,
-0.71
.
-0.69
ⓧ
-0.68
namespace
-0.67
void
-0.67
-0.67
EndProject
-0.66
POSITIVE LOGITS
Rock
2.18
ROCK
2.01
Rock
1.97
rock
1.92
maneu
1.76
rock
1.75
ROCK
1.74
increa
1.74
rocher
1.74
Minang
1.67
Activations Density 0.133%