INDEX
Explanations
mentions of family members who are cousins
mentions of familial relationships, particularly focusing on the term "cousin."
New Auto-Interp
Negative Logits
hner
-0.84
inth
-0.83
ilipp
-0.79
largeDownload
-0.78
yss
-0.76
arching
-0.76
urrent
-0.76
tical
-0.74
inem
-0.73
overe
-0.71
POSITIVE LOGITS
cousin
1.17
cousins
1.10
aunt
1.06
nephew
0.98
niece
0.97
uncle
0.93
uncle
0.79
Cous
0.76
relatives
0.76
daddy
0.75
Activations Density 0.005%