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

    Neuronpedia

    INDEX
    Explanations

    -t

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     cubic
    -0.07
     of
    -0.07
     unsafe
    -0.07
     de
    -0.07
     John
    -0.07
    刑
    -0.07
    _grade
    -0.07
     Constructor
    -0.07
    -0.07
    едак
    -0.07
    POSITIVE LOGITS
    -t
    0.08
    rought
    0.07
    YT
    0.07
    ʷ
    0.07
    tod
    0.07
    zl
    0.07
     comet
    0.07
    /t
    0.07
     Comet
    0.07
    可达
    0.07
    Act Density 0.011%

    No Known Activations