INDEX
Explanations
references to classic literature and cultural allusions
New Auto-Interp
Negative Logits
جÙĪ
-0.16
uru
-0.16
ISO
-0.15
lio
-0.15
enos
-0.15
hy
-0.14
rir
-0.14
olon
-0.14
lit
-0.14
Technical
-0.13
POSITIVE LOGITS
-esque
0.24
-like
0.21
-type
0.19
-style
0.18
èά
0.17
sorts
0.17
stype
0.16
ä¼¼çļĦ
0.16
ÅŁer
0.16
ëĵ¯
0.16
Activations Density 0.085%