INDEX
Explanations
references to food-related events and gatherings
New Auto-Interp
Negative Logits
camin
-0.34
Савезне
-0.34
voors
-0.33
preguntar
-0.33
supo
-0.33
begann
-0.32
cesse
-0.32
aliento
-0.32
való
-0.32
caminar
-0.32
POSITIVE LOGITS
unleash
0.79
whipping
0.75
cran
0.75
chuck
0.74
trot
0.74
crank
0.73
strut
0.73
sling
0.72
decked
0.71
pum
0.70
Activations Density 0.697%