INDEX
Explanations
No Explanations Found
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.25
0.8%
1870
+0.06
0.2%
1741
+0.04
0.1%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2
-0.25
0.00
0
-0.06
0.00
1
-0.04
0.00
Negative Logits
-0.77
↵
-0.76
no
-0.76
-0.75
a
-0.74
↵↵
-0.71
not
-0.70
as
-0.70
o
-0.70
an
-0.68
POSITIVE LOGITS
<bos>
8.17
🤣🤣
1.89
exé
1.79
ftu
1.75
»>
1.67
<?
1.67
ⓧ
1.66
whofe
1.66
ftre
1.66
fince
1.66
Activations Density 0.000%
No Known Activations
This feature has no known activations.