INDEX
Explanations
descriptions related to the taste or qualities of food and drinks
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
31
+0.10
0.3%
1961
+0.09
0.3%
228
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
1013
+0.10
0.03
12
+0.09
0.02
57
+0.08
0.02
Negative Logits
setDo
-0.55
accepte
-0.54
componentWill
-0.52
inter
-0.51
fikation
-0.51
лю
-0.50
lig
-0.49
di
-0.49
الحق
-0.49
enchymal
-0.49
POSITIVE LOGITS
toast
2.81
Toast
2.37
toast
2.16
toasted
1.89
Toast
1.76
wien
1.41
stockholm
1.35
toaster
1.34
fte
1.32
seoul
1.32
Activations Density 0.219%