INDEX
Explanations
references to using or mentioning chalk
references to the word "chalk" and variations of "talk."
New Auto-Interp
Negative Logits
vae
-0.69
gio
-0.68
conflic
-0.67
anmar
-0.67
ettel
-0.67
ctic
-0.66
AAF
-0.64
PDATE
-0.64
çĶŁ
-0.63
======
-0.63
POSITIVE LOGITS
ership
1.23
ed
1.07
edIn
0.99
ers
0.98
erness
0.97
ington
0.93
ings
0.91
idy
0.91
edin
0.90
ery
0.87
Activations Density 0.030%