INDEX
Explanations
No Explanations Found
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
674
+0.34
1.4%
381
+0.06
0.2%
1960
+0.05
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
2
-0.34
0.00
0
-0.06
0.00
1
-0.05
0.00
Negative Logits
that
-1.30
to
-1.27
for
-1.25
so
-1.24
by
-1.23
in
-1.22
with
-1.22
such
-1.19
where
-1.19
don
-1.19
POSITIVE LOGITS
<bos>
9.98
guarante
4.58
effe
4.57
encomp
4.56
fta
4.55
squa
4.55
affor
4.52
fuf
4.48
secon
4.47
purcha
4.47
Activations Density 0.000%
No Known Activations
This feature has no known activations.