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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on personal pronouns (especially “I” and “you”) in conversational or dialogue‐style text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     GO
    -0.07
    ころ
    -0.06
     Man
    -0.06
     Negro
    -0.06
     Applicants
    -0.06
     section
    -0.06
     taxonomy
    -0.06
    มาร
    -0.06
     Livingston
    -0.06
    _IDENTIFIER
    -0.06
    POSITIVE LOGITS
    เหม
    0.07
    	points
    0.06
     rookies
    0.06
     prázd
    0.06
    잡담
    0.06
    pb
    0.06
     pública
    0.06
     المج
    0.06
     geçti
    0.06
     olmam
    0.06
    Act Density 0.007%

    No Known Activations