INDEX
Explanations
phrases related to mental health and medical symptoms
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1842
+0.39
1.6%
1967
+0.25
1.1%
674
+0.25
1.1%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1842
+0.39
0.14
184
+0.25
0.06
394
+0.25
0.09
Negative Logits
impra
-1.15
indestru
-1.03
uninten
-1.03
unden
-1.00
reluct
-0.99
shenan
-0.97
suscep
-0.97
impractica
-0.97
seclu
-0.97
resear
-0.95
POSITIVE LOGITS
rrggbb
0.46
peq
0.45
kloped
0.44
BoxFit
0.44
onResume
0.43
оригіналу
0.43
GraphicsUnit
0.42
Bourgoin
0.42
跳转至
0.42
onPause
0.42
Activations Density 1.419%