INDEX
Explanations
mentions of political events and locations, especially related to caucuses
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
67
+0.15
0.6%
538
+0.14
0.6%
479
+0.12
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
67
+0.15
0.03
200
+0.14
0.02
648
+0.12
0.02
Negative Logits
FT
-0.46
rit
-0.45
RUnlock
-0.44
Greer
-0.44
parsedMessage
-0.43
rin
-0.42
Touchable
-0.42
san
-0.42
subscription
-0.42
مرئيه
-0.41
POSITIVE LOGITS
Iowa
1.36
Iowa
1.28
IOWA
1.25
lowa
1.10
Hawkeye
0.90
lowa
0.81
Moines
0.79
iowa
0.77
pavillon
0.69
consommate
0.68
Activations Density 0.113%