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

    Neuronpedia

    INDEX
    Explanations

    say bracketed code

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Flat
    -0.07
    aturally
    -0.07
     العاص
    -0.07
    xAB
    -0.07
    当たり
    -0.07
    atten
    -0.07
    ጥ
    -0.06
    辂
    -0.06
    quiries
    -0.06
     unite
    -0.06
    POSITIVE LOGITS
    _bitmap
    0.07
    报警
    0.07
     binder
    0.07
    _Delay
    0.06
     womens
    0.06
     Tasks
    0.06
     repositories
    0.06
    Sentence
    0.06
    _browser
    0.06
    <>↵
    0.06
    Act Density 0.069%

    No Known Activations