INDEX
Explanations
incidents of physical altercations or brawls
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.30
1.1%
946
+0.13
0.5%
198
+0.08
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
946
+0.30
0.08
736
+0.13
0.06
184
+0.08
0.01
Negative Logits
<bos>
-2.53
///**
-0.71
MessageState
-0.65
{?>-0.65
HasAnnotation
-0.63
bewerken
-0.63
},{
-0.60
脚注の使い方
-0.60
Sucesor
-0.59
ProtoMessage
-0.59
POSITIVE LOGITS
Juf
1.30
Sted
1.20
Vaugh
1.15
Intere
1.14
Bartholo
1.10
Rine
1.08
stockholm
1.07
reluct
1.07
Theile
1.07
accla
1.07
Activations Density 0.638%