INDEX
Explanations
food-related information, specifically focusing on watermelon and its benefits
New Auto-Interp
Neuron Alignment
Index
Value
% of L₁
776
+0.11
0.3%
678
+0.11
0.3%
1403
+0.10
0.3%
Correlated Neurons
Index
P. Corr.
Cos Sim.
678
+0.11
0.05
1705
+0.11
0.05
776
+0.10
0.05
Negative Logits
négociations
-1.01
vœux
-1.00
déclarations
-0.97
débats
-0.95
Secrétaire
-0.90
complètes
-0.89
merve
-0.88
larmes
-0.88
leçons
-0.87
vérit
-0.85
POSITIVE LOGITS
examples
0.97
ways
0.95
reasons
0.77
suggestions
0.76
highlights
0.75
example
0.74
tips
0.73
possible
0.72
possibilities
0.72
options
0.71
Activations Density 0.369%