INDEX
Explanations
references to dog breeding and related activities
New Auto-Interp
Negative Logits
kitty
-0.17
cat
-0.17
Cat
-0.17
ilden
-0.16
Cats
-0.15
burgl
-0.15
Cat
-0.15
-cat
-0.15
nett
-0.15
λμ
-0.15
POSITIVE LOGITS
breeding
0.42
breed
0.40
bre
0.36
-bre
0.36
bre
0.34
Bre
0.33
bred
0.33
Breed
0.33
Bre
0.32
bred
0.29
Activations Density 0.058%