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

    Neuronpedia

    INDEX
    Explanations

    stripping

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     carving
    -0.07
     sank
    -0.07
     knew
    -0.07
    rait
    -0.07
    die
    -0.06
     Cic
    -0.06
    122
    -0.06
     inherent
    -0.06
     judiciary
    -0.06
     paw
    -0.06
    POSITIVE LOGITS
    Denver
    0.07
     Forget
    0.06
     pg
    0.06
     Lever
    0.06
    .work
    0.06
    ▏
    0.06
     strapon
    0.06
    ilendir
    0.06
    (selected
    0.06
    _CREATE
    0.06
    Act Density 0.000%

    No Known Activations