INDEX
Explanations
prices or numerical values related to products or budget allocations
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
501
+0.14
0.5%
204
+0.13
0.4%
61
+0.12
0.4%
Correlated Neurons
Index
P. Corr.
Cos Sim.
331
+0.14
0.04
382
+0.13
0.04
501
+0.12
0.04
Negative Logits
intersper
-0.87
beaute
-0.80
shenan
-0.77
:—
-0.77
subgoals
-0.75
buddha
-0.74
gaily
-0.74
tsl
-0.73
unspeak
-0.72
intrigu
-0.72
POSITIVE LOGITS
meras
0.66
vermel
0.63
vogli
0.62
alkoh
0.61
PerformLayout
0.61
fú
0.61
viesa
0.60
ló
0.60
limone
0.60
surla
0.59
Activations Density 0.089%