INDEX
Explanations
emotion-related actions or reactions, particularly related to strong feelings like excitement, rage, or bursting into tears
New Auto-Interp
Negative Logits
saf
-0.87
shortest
-0.71
narrower
-0.71
osuke
-0.68
laus
-0.68
ryu
-0.67
taker
-0.65
deceased
-0.63
alty
-0.63
bourg
-0.62
POSITIVE LOGITS
frenzy
1.19
adrenaline
1.18
fury
1.15
excitement
1.15
rage
1.09
pandemonium
1.06
indignation
1.03
laughter
1.02
adren
0.98
anticipation
0.98
Activations Density 0.627%