INDEX
Explanations
The neuron fires on mentions of theatrical “play” (and closely related headings/categories) identifying when the text is referring to stage works.
New Auto-Interp
Negative Logits
judges
-0.07
覚
-0.07
ап
-0.06
BOT
-0.06
BOOK
-0.06
SEO
-0.06
Expose
-0.06
IDE
-0.06
synchronization
-0.06
보고
-0.06
POSITIVE LOGITS
broadcasts
0.08
(substr
0.07
)$_
0.07
actors
0.06
ödül
0.06
müda
0.06
icide
0.06
ircular
0.06
~(
0.06
人的
0.06
Activations Density 0.028%