INDEX
Explanations
sir or dame followed by name
New Auto-Interp
Negative Logits
Brook
0.47
ITO
0.43
Communications
0.38
GEORGE
0.38
revanche
0.37
Georges
0.37
sàng
0.37
Sung
0.36
Sung
0.36
Berkeley
0.35
POSITIVE LOGITS
Dame
0.80
Dame
0.79
dame
0.68
Sir
0.59
Sir
0.59
dames
0.50
Judi
0.50
dama
0.49
sir
0.47
SIR
0.45
Activations Density 0.002%