INDEX
Explanations
elements related to character relationships and dynamics in stories
New Auto-Interp
Negative Logits
_vlog
-0.15
ôm
-0.15
gmt
-0.15
edik
-0.15
-caret
-0.14
blr
-0.14
θι
-0.14
peq
-0.14
oningen
-0.14
ctp
-0.13
POSITIVE LOGITS
another
0.36
another
0.31
Another
0.27
Another
0.26
åı¦ä¸Ģ
0.23
otro
0.22
otra
0.21
åı¦
0.19
ebenfalls
0.18
ãģĵãģ¡ãĤī
0.17
Activations Density 0.083%