INDEX
Explanations
any instance of the word 'clever'
words expressing cleverness or ingenuity
New Auto-Interp
Negative Logits
asta
-0.67
Helsinki
-0.66
avored
-0.65
eds
-0.64
emb
-0.63
eda
-0.63
AST
-0.62
ulhu
-0.62
ental
-0.61
thood
-0.61
POSITIVE LOGITS
ly
1.23
nesses
0.97
enough
0.93
clever
0.82
vier
0.79
glers
0.78
humour
0.78
ness
0.78
tricks
0.77
sonian
0.77
Activations Density 0.009%