INDEX
Explanations
information related to nuclear power generation, safety concerns, and health effects
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1499
+0.09
0.3%
100
+0.08
0.2%
1867
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1499
+0.09
0.05
844
+0.08
0.04
100
+0.08
0.03
Negative Logits
<bos>
-0.62
انجليز
-0.43
errHandler
-0.41
wixt
-0.41
省市镇
-0.39
Predecesor
-0.38
ennifer
-0.38
RetentionPolicy
-0.38
Reentrant
-0.38
ویکیپدیا
-0.38
POSITIVE LOGITS
unlaw
0.81
fup
0.77
ftu
0.72
fep
0.70
sovere
0.69
felicity
0.69
§.
0.68
perfon
0.68
eyel
0.67
embodi
0.67
Activations Density 0.308%