INDEX
Explanations
references to weather conditions, specifically fog
mentions of fog and related atmospheric conditions
New Auto-Interp
Negative Logits
Comments
-0.74
brates
-0.72
Murray
-0.71
Parenthood
-0.71
Union
-0.71
realDonaldTrump
-0.70
Topics
-0.70
brate
-0.69
Students
-0.65
DonaldTrump
-0.65
POSITIVE LOGITS
fog
1.25
Fog
0.97
cloud
0.92
ging
0.90
haze
0.86
clouds
0.84
gy
0.83
curtain
0.83
gers
0.80
rain
0.80
Activations Density 0.005%