INDEX
Explanations
cooking times and instructions in recipe texts
New Auto-Interp
Negative Logits
fas
-0.15
Erot
-0.14
Moh
-0.14
ávánÃŃ
-0.14
eth
-0.14
Invoker
-0.14
.typ
-0.14
icit
-0.14
fol
-0.14
atham
-0.14
POSITIVE LOGITS
.rl
0.17
Tent
0.15
vida
0.14
tdown
0.14
BCM
0.14
Toe
0.14
isman
0.14
lz
0.14
ipi
0.14
iew
0.14
Activations Density 0.009%