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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     gearbox
    -0.07
    /alert
    -0.07
    _baseline
    -0.07
    _Key
    -0.07
     Phon
    -0.07
    载
    -0.06
    _categorical
    -0.06
    _msgs
    -0.06
    _subnet
    -0.06
     Transform
    -0.06
    POSITIVE LOGITS
     dood
    0.06
     λ
    0.06
     camouflage
    0.06
    lambda
    0.06
     неб
    0.06
    these
    0.06
     discovering
    0.06
     Homepage
    0.06
     第
    0.06
    派
    0.06
    Act Density 0.235%

    No Known Activations