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

    Neuronpedia

    INDEX
    Explanations

    _sub

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    hc
    -0.07
     Priest
    -0.06
     É
    -0.06
    中学
    -0.06
     mailbox
    -0.06
    ewart
    -0.06
     };
↵
    -0.06
     Над
    -0.06
    ép
    -0.06
    Ocean
    -0.06
    POSITIVE LOGITS
    _add
    0.22
    _sub
    0.18
    _SUB
    0.09
    _ADD
    0.08
    mul
    0.08
    šet
    0.07
     paní
    0.06
     İslâm
    0.06
    sub
    0.06
     Included
    0.06
    Act Density 0.003%

    No Known Activations