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

    Neuronpedia

    INDEX
    Explanations

    #include

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Time
    -0.07
    ifferences
    -0.07
     Parties
    -0.07
    Project
    -0.06
     Jones
    -0.06
    _subs
    -0.06
     stadium
    -0.06
     wrongly
    -0.06
    Pop
    -0.06
    phone
    -0.06
    POSITIVE LOGITS
     उच
    0.07
     чемпион
    0.07
     zač
    0.06
    _svg
    0.06
    _posts
    0.06
     дол
    0.06
     никогда
    0.06
    _routes
    0.06
    ktop
    0.06
     топ
    0.06
    Act Density 0.006%

    No Known Activations