INDEX
Explanations
instances of the color pink
references to the color pink
New Auto-Interp
Negative Logits
=-=-=-=-
-0.81
ENTS
-0.74
REL
-0.74
UTH
-0.73
Assad
-0.73
Dresden
-0.73
SPONSORED
-0.73
REC
-0.73
Interstitial
-0.72
ETHOD
-0.72
POSITIVE LOGITS
pink
0.98
blob
0.92
erton
0.89
slime
0.89
ribbon
0.86
washing
0.85
stripe
0.84
etooth
0.84
ish
0.83
ishly
0.82
Activations Density 0.007%