INDEX
Explanations
references to publishing details, such as ISBNs and page numbers
New Auto-Interp
Negative Logits
osit
-0.15
unist
-0.14
anonym
-0.14
alm
-0.14
branded
-0.13
onn
-0.13
Weiss
-0.13
hangi
-0.13
ached
-0.13
Raw
-0.13
POSITIVE LOGITS
//{{0.25
ISBN
0.25
ISBN
0.24
cloth
0.21
cloth
0.20
isbn
0.20
paper
0.20
isbn
0.20
hard
0.19
pb
0.18
Activations Density 0.057%