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

    Neuronpedia

    INDEX
    Explanations

    at

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    theros
    -0.07
    ********
    -0.06
     rng
    -0.06
     eden
    -0.06
     renewable
    -0.06
     человечес
    -0.06
     Blender
    -0.06
     collagen
    -0.06
    oders
    -0.06
     tuning
    -0.06
    POSITIVE LOGITS
    at
    0.12
    AT
    0.11
    hat
    0.09
    alat
    0.08
    kat
    0.08
    imat
    0.08
    kap
    0.08
    rat
    0.08
    quat
    0.08
    ат
    0.08
    Act Density 0.073%

    No Known Activations