INDEX
Explanations
elements that evoke vivid sensory imagery and physical sensations
New Auto-Interp
Negative Logits
еви
-0.16
roys
-0.16
leans
-0.15
Rounds
-0.15
.useState
-0.15
estroy
-0.14
hoe
-0.14
entiful
-0.14
ulkan
-0.14
Blick
-0.14
POSITIVE LOGITS
crack
0.25
dance
0.23
dancing
0.22
gl
0.22
dances
0.20
dance
0.20
danced
0.20
Dance
0.19
morph
0.19
tw
0.18
Activations Density 0.339%