INDEX
Explanations
expressions of strong emotions or frustrations
New Auto-Interp
Negative Logits
lum
-0.15
Smile
-0.15
INFO
-0.15
ylum
-0.15
[color
-0.14
OUCH
-0.14
Ãĥ
-0.14
Zot
-0.14
heim
-0.14
hookers
-0.14
POSITIVE LOGITS
gif
0.24
gifs
0.23
GIF
0.21
gif
0.20
rn
0.18
icorn
0.17
-prepend
0.17
tb
0.17
ternet
0.16
squad
0.16
Activations Density 0.440%