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

    Neuronpedia

    INDEX
    Explanations

    realloc

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     donna
    -0.07
     være
    -0.07
    assistant
    -0.06
    ANGO
    -0.06
    谳
    -0.06
    记忆力
    -0.06
    _FORCE
    -0.06
    combat
    -0.06
    表现
    -0.06
    规定的
    -0.06
    POSITIVE LOGITS
    ,\"
    0.07
    acağız
    0.07
    ittal
    0.07
    clusão
    0.07
     באר
    0.07
    风口
    0.07
    _bg
    0.06
     outliers
    0.06
    Jun
    0.06
     collision
    0.06
    Act Density 0.003%

    No Known Activations