INDEX
Explanations
interactions and expressions of emotional states in dialogue
New Auto-Interp
Negative Logits
screamed
-0.20
blinked
-0.19
scream
-0.18
cries
-0.18
screams
-0.17
ummings
-0.17
Dash
-0.16
screaming
-0.16
blink
-0.16
cry
-0.16
POSITIVE LOGITS
purs
0.21
gest
0.20
nod
0.19
cock
0.18
look
0.18
looks
0.18
incl
0.18
looks
0.17
drum
0.17
straight
0.17
Activations Density 0.477%