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

    Neuronpedia

    INDEX
    Explanations

    Describing device problems

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    -source
    -0.06
    Universal
    -0.06
    ασ
    -0.06
    žel
    -0.06
     createAction
    -0.06
    =a
    -0.06
     замі
    -0.06
     의미
    -0.06
    asına
    -0.06
    (Un
    -0.06
    POSITIVE LOGITS
     qualify
    0.07
     Learn
    0.07
    test
    0.07
    _cust
    0.06
     reminds
    0.06
    ایی
    0.06
    -hooks
    0.06
     inside
    0.06
    Remember
    0.06
     terrorist
    0.06
    Act Density 0.038%

    No Known Activations