INDEX
Explanations
references to time durations or deadlines mentioned in hours or days
references to the number "48"
New Auto-Interp
Negative Logits
bub
-0.81
ovie
-0.77
bard
-0.75
Flavoring
-0.73
andise
-0.69
Jinn
-0.67
Afric
-0.66
achu
-0.66
isson
-0.65
Pros
-0.64
POSITIVE LOGITS
eenth
1.03
576
0.96
een
0.93
kHz
0.89
MJ
0.84
00
0.83
teen
0.81
80
0.78
Hours
0.78
hours
0.77
Activations Density 0.041%