INDEX
Explanations
information related to neural networks, specifically focusing on the structure and parameters of convolutional layers and hidden units
New Auto-Interp
Negative Logits
llan
-1.03
astern
-1.00
Seraph
-0.97
Bav
-0.90
Fly
-0.90
itect
-0.88
guyen
-0.87
ategory
-0.87
Piet
-0.87
aign
-0.86
POSITIVE LOGITS
ted
1.57
ting
1.40
ters
1.15
enza
1.10
inputs
1.09
input
1.06
impedance
1.00
ifferent
1.00
output
0.99
flow
0.97
Activations Density 0.575%