INDEX
Explanations
musician names
This neuron detects mentions of individual musician names (artists) in the album personnel sections.
New Auto-Interp
Negative Logits
tud
-0.07
privile
-0.06
slump
-0.06
yaşam
-0.06
سیاست
-0.06
าณ
-0.06
chopping
-0.06
相信
-0.06
saddle
-0.06
Three
-0.06
POSITIVE LOGITS
(ListNode
0.07
vera
0.07
Ib
0.06
reflections
0.06
_EMIT
0.06
(ast
0.06
vyšší
0.06
instincts
0.06
Ability
0.06
Eid
0.06
Activations Density 0.013%