INDEX
Explanations
phrases related to intellectual property and copyright
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1065
+0.15
0.7%
1233
+0.14
0.7%
1363
+0.13
0.6%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1363
+0.15
0.03
1065
+0.14
0.03
1677
+0.13
0.02
Negative Logits
<bos>
-1.44
intersper
-1.10
gratify
-1.04
rouse
-1.02
fernando
-0.95
overcrow
-0.94
endow
-0.92
quitted
-0.88
reconno
-0.86
amass
-0.84
POSITIVE LOGITS
intellectual
1.32
Intellectual
1.21
Intellectual
1.16
intellectual
1.15
intelectual
1.12
lectual
0.91
intellectually
0.78
intellectuals
0.73
intellect
0.71
withRouter
0.67
Activations Density 0.334%