INDEX
Explanations
descriptions of violence or aggressive actions.
The neuron detects violent ripping or tearing actions (verbs depicting something being torn or ripped apart).
New Auto-Interp
Negative Logits
msgstr
-0.07
функци
-0.07
fibonacci
-0.07
Plantae
-0.07
companyId
-0.07
有限公司
-0.07
hvě
-0.06
.addClass
-0.06
Soc
-0.06
\">
-0.06
POSITIVE LOGITS
Rip
0.09
ripping
0.09
torn
0.09
ripped
0.09
tearing
0.09
Tear
0.09
tear
0.09
teardown
0.08
shredded
0.08
rip
0.08
Activations Density 0.008%