INDEX
Explanations
mentions of a specific person named "Friedman"
mentions of the name "Friedman."
New Auto-Interp
Negative Logits
carry
-0.68
reek
-0.67
query
-0.67
ropolitan
-0.67
xy
-0.66
Hop
-0.65
Flavoring
-0.65
quer
-0.64
origin
-0.64
pipe
-0.64
POSITIVE LOGITS
Friedman
1.48
Shapiro
0.87
enegger
0.80
Becker
0.79
enstein
0.77
Sachs
0.76
Herz
0.75
stadt
0.73
Haas
0.72
slic
0.72
Activations Density 0.008%