INDEX
Explanations
names of famous jazz musicians
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
50
+0.18
0.7%
152
+0.07
0.3%
376
+0.07
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1136
+0.18
0.05
1823
+0.07
0.04
1205
+0.07
0.04
Negative Logits
<bos>
-2.68
contentLoaded
-0.86
expandindo
-0.79
Roskov
-0.78
EndContext
-0.77
HideFlags
-0.76
Datuak
-0.76
Geplaatst
-0.75
లాలు
-0.74
intptr
-0.73
POSITIVE LOGITS
milf
2.17
Juf
2.11
increa
2.09
maneu
2.05
affor
2.01
emphat
2.00
hentai
1.97
inev
1.97
shenan
1.94
accla
1.93
Activations Density 0.364%