INDEX
Explanations
submissions marked with "Sub"
references to specific subjects or keywords, particularly related to submissions and categories
New Auto-Interp
Negative Logits
gloves
-0.77
boots
-0.69
hawk
-0.69
ocratic
-0.68
haw
-0.67
ill
-0.66
lasting
-0.65
recollection
-0.65
pont
-0.64
tower
-0.63
POSITIVE LOGITS
Sub
3.81
sub
2.33
Sub
2.20
SUB
1.96
sub
1.61
subs
1.49
Subst
1.30
Interstitial
1.27
SU
1.20
Submit
1.14
Activations Density 0.019%