INDEX
Explanations
words related to medical conditions such as "stroke"
occurrences of the word "stroke" and its variations, indicating a focus on health-related themes
New Auto-Interp
Negative Logits
guyen
-0.84
gemony
-0.74
proprietary
-0.73
andro
-0.68
bidden
-0.68
antz
-0.65
ramer
-0.64
pires
-0.64
essor
-0.63
segreg
-0.63
POSITIVE LOGITS
stroke
1.59
strokes
1.29
stroke
1.12
Stro
1.04
manship
0.91
stro
0.73
iovascular
0.70
Canaver
0.69
Newport
0.68
SetFontSize
0.68
Activations Density 0.006%