INDEX
Explanations
mentions of the name "Harry."
New Auto-Interp
Negative Logits
ermal
-0.15
gb
-0.15
aci
-0.14
ittle
-0.14
ORY
-0.14
yar
-0.14
esh
-0.14
marshall
-0.14
áºŃp
-0.14
ory
-0.14
POSITIVE LOGITS
hausen
0.31
Potter
0.27
pot
0.21
Pot
0.19
Styles
0.18
POT
0.18
.nlm
0.18
styles
0.17
_pot
0.17
uki
0.16
Activations Density 0.005%