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

    Neuronpedia

    INDEX
    Explanations

    Actions involving the body

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    ub
    -0.07
    /fixtures
    -0.07
     indicates
    -0.06
     nk
    -0.06
    �
    -0.06
     адже
    -0.06
    伯
    -0.06
    bi
    -0.06
     Based
    -0.06
    uf
    -0.06
    POSITIVE LOGITS
     programa
    0.06
    �
    0.06
    (tags
    0.06
    _ASCII
    0.06
    reshape
    0.06
    ificação
    0.06
    nero
    0.06
    _ENSURE
    0.06
     спіл
    0.06
    .ImageAlign
    0.06
    Act Density 0.024%

    No Known Activations