INDEX
Explanations
phrases related to legal proceedings and investigations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1244
+0.10
0.3%
946
+0.08
0.2%
400
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
400
+0.10
0.04
763
+0.08
0.03
1483
+0.08
0.03
Negative Logits
stasia
-0.84
embra
-0.78
palab
-0.77
brille
-0.77
tanga
-0.76
mourut
-0.76
chande
-0.76
gmbh
-0.74
pernic
-0.74
autob
-0.72
POSITIVE LOGITS
wondered
0.52
whether
0.51
Czy
0.50
why
0.50
للاسماء
0.49
wanted
0.49
possibility
0.49
would
0.47
knew
0.46
iesp
0.46
Activations Density 0.158%