INDEX
Explanations
references to the concept of "half" or "halfway."
New Auto-Interp
Negative Logits
575
-0.18
lac
-0.15
able
-0.15
usu
-0.15
Datagram
-0.14
ì¶ľìŀ¥
-0.14
XL
-0.14
FD
-0.14
lav
-0.14
lint
-0.13
POSITIVE LOGITS
/full
0.25
dozen
0.24
ords
0.22
heart
0.22
way
0.20
-hearted
0.19
baked
0.19
moon
0.19
wit
0.19
weg
0.18
Activations Density 0.024%