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

    Neuronpedia

    INDEX
    Explanations

    Reasons or explanations

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on negative modal constructions (especially “can’t” or similar prohibitions) indicating that the user is unable or forbidden to do something.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Orange
    -0.07
     آباد
    -0.07
    uParam
    -0.07
     силу
    -0.07
    syscall
    -0.07
     Trouble
    -0.07
    OutOfRangeException
    -0.07
    チュ
    -0.06
     creator
    -0.06
     آسی
    -0.06
    POSITIVE LOGITS
     Loans
    0.07
     conditional
    0.06
    (first
    0.06
    359
    0.06
    кам
    0.06
     cock
    0.06
    rowave
    0.06
    ็ต
    0.06
    enders
    0.06
    Α
    0.06
    Act Density 0.094%

    No Known Activations