INDEX
Explanations
informative video-related cues in the text
New Auto-Interp
Negative Logits
soever
-0.65
tis
-0.63
REDACTED
-0.61
Runner
-0.60
iments
-0.59
Pod
-0.58
morrow
-0.56
manship
-0.56
phas
-0.55
Cs
-0.55
POSITIVE LOGITS
ARTICLE
0.75
Detail
0.72
iframe
0.64
TABLE
0.63
POLITICO
0.63
âĩ
0.61
VIDEO
0.61
illance
0.61
Videos
0.61
Invalid
0.60
Activations Density 0.059%