INDEX
Explanations
the word "Aston"
references to the football club Aston Villa
New Auto-Interp
Negative Logits
labeled
-0.85
labeling
-0.71
circ
-0.68
rice
-0.68
paste
-0.68
chery
-0.65
filter
-0.63
prim
-0.63
file
-0.63
signaling
-0.63
POSITIVE LOGITS
Aston
4.02
aston
1.60
Bentley
1.36
Sunderland
1.30
Wolver
1.23
McLaren
1.23
Everton
1.22
borgh
1.19
Nottingham
1.14
Norwich
1.13
Activations Density 0.028%