INDEX
Explanations
references to various forms of art, particularly digital art and artwork in different mediums
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.12
0.5%
2011
+0.06
0.3%
1974
+0.06
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1051
+0.12
0.05
601
+0.06
0.05
1974
+0.06
0.04
Negative Logits
<bos>
-1.67
,
-0.78
</tbody>
-0.77
.
-0.76
public
-0.75
//
-0.75
ുറ
-0.74
/*
-0.74
<eos>
-0.74
-0.72
POSITIVE LOGITS
increa
2.30
impra
2.21
maneu
2.19
affor
2.16
stockholm
2.15
lidl
2.14
disagre
2.10
ftu
2.05
fta
2.03
inev
2.03
Activations Density 0.153%