INDEX
Explanations
terms related to technical concepts and structures, especially involving data, programming, and technological devices
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
394
+0.14
0.4%
690
+0.11
0.3%
227
+0.11
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
394
+0.14
0.06
1819
+0.11
0.07
227
+0.11
0.07
Negative Logits
sappi
-1.26
vogli
-1.19
mef
-1.12
incess
-1.09
dises
-1.05
alberto
-1.03
socie
-1.01
scopri
-1.00
dico
-0.99
Perci
-0.99
POSITIVE LOGITS
themselves
0.81
are
0.72
themselves
0.60
are
0.60
were
0.57
toppers
0.56
hips
0.55
חיצוניים
0.55
Are
0.55
aren
0.54
Activations Density 0.453%