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

    Neuronpedia

    INDEX
    Explanations

    Open or uncovered

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     textos
    -0.08
    .activate
    -0.08
    (_.
    -0.08
     состав
    -0.07
     technical
    -0.07
    (anchor
    -0.07
    력
    -0.07
     reviewer
    -0.07
     attorney
    -0.07
    ।
    -0.07
    POSITIVE LOGITS
     bloot
    0.11
    Exposure
    0.10
    无遮
    0.10
     Exposure
    0.10
     exposure
    0.10
     unattended
    0.09
    面对
    0.09
     drying
    0.09
     בפני
    0.09
    开放
    0.08
    Act Density 0.025%

    No Known Activations