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

    Neuronpedia

    INDEX
    Explanations

    purification methods

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     ダ
    -0.07
    �
    -0.06
     OMIT
    -0.06
    liqu
    -0.06
    463
    -0.06
     irrelevant
    -0.06
    tě
    -0.06
     Marco
    -0.06
     onCancelled
    -0.06
     Riyadh
    -0.06
    POSITIVE LOGITS
    률
    0.07
    avoid
    0.07
     عبار
    0.06
    .Contact
    0.06
     kus
    0.06
    .pb
    0.06
    extent
    0.06
     submenu
    0.06
    Ensure
    0.06
     cạnh
    0.06
    Act Density 0.018%

    No Known Activations