INDEX
Explanations
The neuron fires on taxonomic classification terms in descriptions—especially the word “camelid.”
New Auto-Interp
Negative Logits
721
-0.06
vertime
-0.06
اطل
-0.06
Hulu
-0.06
conce
-0.06
AWN
-0.06
stagnant
-0.06
Screenshot
-0.06
.population
-0.06
.real
-0.06
POSITIVE LOGITS
comple
0.07
Sosyal
0.07
Advisor
0.06
(inputs
0.06
yleft
0.06
credit
0.06
({↵↵0.06
odore
0.06
процесс
0.06
ád
0.06
Activations Density 0.003%