INDEX
Explanations
phrases containing the word "Ye."
mentions of a particular location, specifically related to "Yeonpyeong" and similar terms
New Auto-Interp
Negative Logits
ãĥ¼ãĥĨ
-0.94
ctica
-0.71
ITAL
-0.71
ULAR
-0.69
TAIN
-0.67
ãĥĩ
-0.66
Crusher
-0.65
olls
-0.64
ãĥ¥
-0.64
ãĤŃ
-0.64
POSITIVE LOGITS
Ye
1.07
Ye
1.04
erk
0.89
asing
0.81
ldon
0.80
hovah
0.78
ongh
0.75
xt
0.75
eree
0.74
aser
0.74
Activations Density 0.006%