INDEX
Explanations
numerical ages, thirty, forty
New Auto-Interp
Negative Logits
teenagers
0.94
teenage
0.92
sixteen
0.89
teenager
0.87
teen
0.86
teenage
0.83
eighteen
0.81
seventeen
0.80
adolescente
0.80
teen
0.80
POSITIVE LOGITS
Thirty
0.91
Thirty
0.91
三十
0.71
thirty
0.71
thirties
0.70
Forty
0.69
treinta
0.69
Forty
0.65
trent
0.62
thirty
0.61
Activations Density 0.008%