INDEX
Explanations
references to a specific person's name: "Cruz"
mentions of the name "Cruz."
New Auto-Interp
Negative Logits
ebin
-0.72
à¨
-0.70
Seym
-0.70
ochemical
-0.67
unseen
-0.66
OX
-0.65
Finnish
-0.65
Ö¼
-0.64
Bulgarian
-0.63
Pebble
-0.62
POSITIVE LOGITS
Cruz
1.15
Cruz
1.14
omics
0.89
Rubio
0.80
anne
0.78
yne
0.77
ettes
0.75
Rafael
0.74
itus
0.73
aye
0.72
Activations Density 0.006%