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

    Neuronpedia

    INDEX
    Explanations

    Sher

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Treatment
    -0.07
     pipes
    -0.07
    messages
    -0.07
     categoryName
    -0.06
     Ae
    -0.06
     transformed
    -0.06
    pend
    -0.06
     increase
    -0.06
     caso
    -0.06
     Late
    -0.06
    POSITIVE LOGITS
    وة
    0.07
    .HashSet
    0.07
     du
    0.07
    _principal
    0.06
     يح
    0.06
     sher
    0.06
     threesome
    0.06
    Prefab
    0.06
     hookers
    0.06
    /arm
    0.06
    Act Density 0.004%

    No Known Activations