INDEX
Explanations
proper names
references to the name "Manuel"
New Auto-Interp
Negative Logits
nces
-0.86
rug
-0.81
Scotland
-0.74
orks
-0.74
ramid
-0.74
neys
-0.74
kef
-0.71
older
-0.70
REE
-0.70
rab
-0.69
POSITIVE LOGITS
Manuel
1.30
theless
0.96
Antonio
0.87
Santos
0.80
ity
0.76
Laurent
0.75
Andreas
0.73
Anton
0.71
Carlo
0.71
Carlos
0.69
Activations Density 0.007%