INDEX
Explanations
proper names with the word "Tony" in them
mentions of the name "Tony."
New Auto-Interp
Negative Logits
INESS
-0.92
ername
-0.75
cipline
-0.74
baugh
-0.74
ãģ¦
-0.72
20439
-0.70
ebook
-0.70
ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
-0.70
doors
-0.69
IFT
-0.69
POSITIVE LOGITS
Abbott
0.93
Blair
0.89
Sop
0.88
Tony
0.86
Romo
0.86
Hawk
0.80
neau
0.79
Tony
0.78
Cliff
0.77
Stark
0.77
Activations Density 0.014%