INDEX
Explanations
figurative language that expresses strong emotions or vivid imagery
New Auto-Interp
Negative Logits
bland
-0.15
åŁºåľ°
-0.15
-related
-0.14
related
-0.14
nice
-0.14
ï¸
-0.14
iveness
-0.14
بس
-0.14
Nice
-0.13
posed
-0.13
POSITIVE LOGITS
yet
0.19
ims
0.17
occasionally
0.17
çĬ¬
0.16
sometimes
0.15
fern
0.14
slightly
0.14
wounded
0.14
opi
0.14
thorough
0.13
Activations Density 0.179%