INDEX
Explanations
emotions and states of being
New Auto-Interp
Negative Logits
令人
-0.18
ickness
-0.18
otation
-0.17
ivation
-0.17
itage
-0.17
endency
-0.17
izziness
-0.17
cription
-0.17
uration
-0.17
ellation
-0.17
POSITIVE LOGITS
overwhelmed
0.21
ophobic
0.21
bothered
0.21
struck
0.21
interested
0.20
happy
0.20
azed
0.19
challenged
0.19
whel
0.19
shocked
0.19
Activations Density 0.097%