INDEX
Explanations
Ranks and news titles
The neuron strongly activates on first-person self‐references (e.g. “I,” “my,” “am”), signaling it detects personal pronouns indicating the speaker’s own perspective.
New Auto-Interp
Negative Logits
deaf
-0.07
Connect
-0.06
terminology
-0.06
혀
-0.06
GDPR
-0.06
lasted
-0.06
.getRoot
-0.06
Masters
-0.06
exterity
-0.06
_object
-0.06
POSITIVE LOGITS
онах
0.06
(Profile
0.06
าหล
0.06
Cougar
0.06
return
0.06
�
0.06
ภ
0.06
Lorem
0.06
}}
0.06
Profiler
0.06
Activations Density 0.000%