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

    Neuronpedia

    INDEX
    Explanations

    Japanese/Korean particles

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     becomes
    -0.07
     уз
    -0.07
    dog
    -0.07
     INPUT
    -0.07
     Solution
    -0.07
     Yourself
    -0.06
     refusing
    -0.06
     Tender
    -0.06
     Introduction
    -0.06
     led
    -0.06
    POSITIVE LOGITS
    は
    0.08
    Grow
    0.07
    では
    0.07
    ูไ
    0.07
     artık
    0.07
    시는
    0.07
    Warn
    0.06
    子は
    0.06
     ePub
    0.06
    agoon
    0.06
    Act Density 0.013%

    No Known Activations