INDEX
Explanations
references to the TV series "Mad Men"
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1602
+0.15
0.7%
1472
+0.14
0.6%
976
+0.12
0.5%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1472
+0.15
0.03
1602
+0.14
0.02
1516
+0.12
0.02
Negative Logits
acakt
-0.53
alım
-0.48
mayacak
-0.48
liminaries
-0.47
ColumnHeaders
-0.47
Allister
-0.45
WRENCE
-0.45
UnitTesting
-0.44
ıyoruz
-0.44
odkazy
-0.42
POSITIVE LOGITS
Mad
1.37
Mad
1.29
MAD
1.28
mad
1.13
MAD
1.11
Madd
1.05
mad
1.05
Madsen
0.97
Maddie
0.96
madison
0.89
Activations Density 0.090%