INDEX
Explanations
employment/jobs
This neuron activates on mentions of compensation figures and related terms (e.g. salaries, pay amounts, job‐related monetary data).
New Auto-Interp
Negative Logits
-none
-0.07
अत
-0.07
_redis
-0.07
Shaman
-0.07
_consum
-0.06
Hop
-0.06
син
-0.06
vegetation
-0.06
minden
-0.06
buttons
-0.06
POSITIVE LOGITS
Quad
0.07
¸
0.06
-scenes
0.06
exhilar
0.06
dazz
0.06
@
0.06
iến
0.06
운
0.06
]}↵
0.06
ja
0.06
Activations Density 0.092%