INDEX
Explanations
video-related words or prompts
references to videos that can be watched
New Auto-Interp
Negative Logits
vironment
-0.88
abil
-0.77
Ĭ±
-0.76
å§«
-0.71
forth
-0.69
itage
-0.68
yo
-0.67
naires
-0.66
essee
-0.66
itiz
-0.66
POSITIVE LOGITS
clip
0.97
Thumbnails
0.93
footage
0.89
clips
0.85
playback
0.79
filmed
0.79
Clip
0.75
clip
0.73
Surveillance
0.73
Video
0.70
Activations Density 0.035%