INDEX
Explanations
instances of characters expressing strong emotions or actions through physical gestures, particularly raising hands
New Auto-Interp
Negative Logits
ffi
-0.15
akh
-0.15
irl
-0.15
irt
-0.14
uka
-0.14
paired
-0.14
ere
-0.14
yo
-0.14
eor
-0.14
holm
-0.13
POSITIVE LOGITS
owitz
0.16
lian
0.16
/Branch
0.15
Picker
0.15
IGHT
0.15
Fleet
0.15
à¥įà¤
0.15
ights
0.15
_subplot
0.14
fleet
0.14
Activations Density 0.004%