INDEX
Explanations
instances of the substring "ain" within words
New Auto-Interp
Negative Logits
RESULTS
-0.79
externalToEVAOnly
-0.75
RNA
-0.72
nesday
-0.70
DragonMagazine
-0.69
trop
-0.69
conservancy
-0.69
ODUCT
-0.66
RN
-0.66
kefeller
-0.65
POSITIVE LOGITS
fo
1.15
ese
1.00
ees
0.97
bow
0.95
thood
0.94
ned
0.93
sts
0.93
tenance
0.92
es
0.90
esville
0.90
Activations Density 0.009%