INDEX
Explanations
references to poetry and poets
New Auto-Interp
Negative Logits
ató
-0.16
onet
-0.16
angers
-0.15
anger
-0.15
agn
-0.15
al
-0.14
cas
-0.14
¯
-0.14
agen
-0.14
enburg
-0.14
POSITIVE LOGITS
.po
0.22
laure
0.20
(po
0.18
ewe
0.18
ical
0.18
-po
0.18
slam
0.16
Po
0.16
stride
0.15
.cms
0.15
Activations Density 0.026%