INDEX
Explanations
dialogues and interactions involving emotions and responses
New Auto-Interp
Negative Logits
screamed
-0.20
scream
-0.19
screams
-0.18
blinked
-0.18
cries
-0.16
blink
-0.16
cry
-0.16
screaming
-0.15
cried
-0.15
shouted
-0.15
POSITIVE LOGITS
gest
0.19
nod
0.16
oor
0.16
çĴ°
0.15
ouser
0.15
resett
0.15
lez
0.15
egan
0.14
incl
0.14
ignum
0.14
Activations Density 0.590%