INDEX
Explanations
instances of the word "been" in various contexts
New Auto-Interp
Negative Logits
éĥİ
-0.19
uhan
-0.16
EGIN
-0.15
phans
-0.14
/TR
-0.14
oux
-0.14
box
-0.14
лÑĥ
-0.14
sten
-0.14
cid
-0.14
POSITIVE LOGITS
/is
0.25
since
0.18
lately
0.18
_since
0.17
COME
0.17
able
0.17
liken
0.15
awhile
0.14
tent
0.14
nt
0.14
Activations Density 0.111%