INDEX
Explanations
nutritional information such as calorie counts and vitamin content in recipes
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
876
+0.16
0.5%
1403
+0.08
0.2%
185
+0.08
0.2%
Correlated Neurons
Index
P. Corr.
Cos Sim.
753
+0.16
0.01
185
+0.08
0.02
1823
+0.08
0.02
Negative Logits
monaster
-0.79
Jurist
-0.67
Czechos
-0.66
gorov
-0.61
LookAnd
-0.61
episcopal
-0.60
gubern
-0.59
Ecclesiastical
-0.59
Tacitus
-0.58
catedral
-0.58
POSITIVE LOGITS
Calories
0.75
calories
0.73
purcha
0.69
pollut
0.69
appels
0.66
Ename
0.66
affor
0.66
Calories
0.66
scrat
0.64
increa
0.64
Activations Density 0.168%