Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    Problems and challenges

    np_max-act · gemini-2.0-flash

    This neuron fires on first-person self-references (I, me, am) in user requests.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    sentences that express a first‑/second‑person conversational turn asking for help or clarifying questions (i.e., interactive, dialogic requests and responses).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ीण
    -0.07
     traps
    -0.06
     opioid
    -0.06
     �
    -0.06
     odpověd
    -0.06
    _z
    -0.06
    %X
    -0.06
     numb
    -0.06
    ecer
    -0.06
    От
    -0.06
    POSITIVE LOGITS
    chez
    0.07
    Cum
    0.07
     uplifting
    0.06
     Advisor
    0.06
    pired
    0.06
     standby
    0.06
     excessive
    0.06
    portal
    0.06
     cropping
    0.06
     aspect
    0.06
    Act Density 0.131%

    No Known Activations