INDEX
Explanations
references to specific food dishes and their descriptions
New Auto-Interp
Negative Logits
etermin
-0.21
ernetes
-0.16
eson
-0.16
UGIN
-0.15
meteor
-0.15
Starr
-0.15
isp
-0.15
elson
-0.15
thr
-0.15
smack
-0.14
POSITIVE LOGITS
ANCE
0.21
IFY
0.19
rench
0.19
iosa
0.19
anc
0.19
andi
0.18
RESS
0.18
ishes
0.17
andel
0.17
renched
0.17
Activations Density 0.031%