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

    Neuronpedia

    INDEX
    Explanations

    destruction

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (window
    -0.07
     پژوهش
    -0.07
     더욱
    -0.06
    杨
    -0.06
    、これ
    -0.06
    �
    -0.06
    lam
    -0.06
    $row
    -0.06
     Mar
    -0.06
    一定
    -0.06
    POSITIVE LOGITS
     devastation
    0.08
     Việc
    0.07
    λία
    0.07
     Slovak
    0.07
    /Instruction
    0.06
     яким
    0.06
     devast
    0.06
    instructions
    0.06
    ॉ
    0.06
     порядок
    0.06
    Act Density 0.025%

    No Known Activations