INDEX
Explanations
words related to humor or comedy
instances of the word "funny."
New Auto-Interp
Negative Logits
ainer
-1.01
jri
-0.78
iami
-0.76
ignt
-0.76
eded
-0.75
apers
-0.74
eding
-0.74
ensional
-0.73
aining
-0.73
ilings
-0.73
POSITIVE LOGITS
netflix
0.89
funny
0.88
balls
0.79
GIF
0.78
ness
0.76
comedy
0.76
Funny
0.75
karma
0.74
fun
0.72
Laugh
0.72
Activations Density 0.011%