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

    Neuronpedia

    INDEX
    Explanations

    KN

    np_max-act · gemini-2.0-flash

    The neuron is highly sensitive to sentences expressing personal stance or advice—e.g. “I believe…,” “We need…,” “It is important…”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _MAP
    -0.08
     accuses
    -0.06
    ünden
    -0.06
     каль
    -0.06
    uš
    -0.06
     liền
    -0.06
     carr
    -0.06
    стика
    -0.06
    stead
    -0.06
    enk
    -0.06
    POSITIVE LOGITS
     DISTRIBUT
    0.07
    Birthday
    0.07
     television
    0.07
     anarchist
    0.07
    ippy
    0.06
    ）↵
    0.06
     azimuth
    0.06
     contador
    0.06
     Budget
    0.06
     impaired
    0.06
    Act Density 0.083%

    No Known Activations