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

    Neuronpedia

    INDEX
    Explanations

    )

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     consists
    -0.07
     país
    -0.06
     advocate
    -0.06
     čer
    -0.06
     manifest
    -0.06
     Sü
    -0.06
    DUCTION
    -0.06
     파
    -0.06
    REFER
    -0.06
    w
    -0.06
    POSITIVE LOGITS
    cle
    0.07
    -divider
    0.07
    �
    0.06
    ắm
    0.06
     Xxx
    0.06
    isOpen
    0.06
    <Scalars
    0.06
     (_,
    0.06
    0.06
    dispose
    0.06
    Act Density 0.015%

    No Known Activations