INDEX
Explanations
information related to transportation systems, including trains, infrastructure, accidents, and operations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
1842
+0.15
0.4%
609
+0.14
0.4%
1385
+0.09
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
284
+0.15
0.08
1842
+0.14
0.06
2044
+0.09
0.07
Negative Logits
inev
-2.00
volunte
-1.98
emphat
-1.91
thut
-1.89
depic
-1.86
encomp
-1.85
accla
-1.84
fta
-1.83
reluct
-1.82
increa
-1.82
POSITIVE LOGITS
without
1.23
without
1.01
efficiently
0.88
ohne
0.87
while
0.85
WITHOUT
0.85
via
0.83
Without
0.83
safely
0.83
ได้
0.79
Activations Density 0.592%