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

    Neuronpedia

    INDEX
    Explanations

    Flame color and elements

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    greso
    -0.08
    igion
    -0.08
     тщ
    -0.08
    _COMPLE
    -0.08
    mele
    -0.08
     Leta
    -0.08
    ilmiş
    -0.08
     крови
    -0.07
     reorgan
    -0.07
     тщательно
    -0.07
    POSITIVE LOGITS
     screaming
    0.08
     coconut
    0.07
    Opening
    0.07
    écution
    0.07
    When
    0.07
     వచ్చింది
    0.07
     Magnolia
    0.07
    桂
    0.07
    Closing
    0.07
     screams
    0.07
    Act Density 0.002%

    No Known Activations