INDEX
Explanations
mentions of the name "Leanne"
repeated references to a specific name, particularly "Leanne."
New Auto-Interp
Negative Logits
ODUCT
-0.79
matic
-0.78
UID
-0.77
dfx
-0.73
oted
-0.72
odiac
-0.68
direction
-0.68
izoph
-0.67
oter
-0.67
eleph
-0.67
POSITIVE LOGITS
Marie
0.90
anne
0.88
lette
0.87
Marie
0.84
Barr
0.83
lettuce
0.83
theless
0.82
phrine
0.77
cheese
0.76
apolis
0.76
Activations Density 0.010%