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

    Neuronpedia

    INDEX
    Explanations

    eye treatments

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Vermont
    -0.07
    _indent
    -0.07
     beard
    -0.06
    (encoding
    -0.06
     nit
    -0.06
    让我
    -0.06
     bol
    -0.06
     Riders
    -0.06
     eBooks
    -0.06
     zm
    -0.06
    POSITIVE LOGITS
    OTH
    0.06
     =========================================================================
    0.06
    kn
    0.06
    _);↵↵
    0.06
    North
    0.06
     Donation
    0.06
    agree
    0.06
     extravagant
    0.06
        ↵    ↵    ↵
    0.06
     convention
    0.06
    Act Density 0.008%

    No Known Activations