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

    Neuronpedia

    INDEX
    Explanations

    toxic

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Labor
    -0.07
    ımsız
    -0.06
    �
    -0.06
     plastics
    -0.06
    _trip
    -0.06
     Ub
    -0.06
    Lazy
    -0.06
    lazy
    -0.06
    izar
    -0.06
    ın
    -0.06
    POSITIVE LOGITS
     vriend
    0.08
     Mirage
    0.08
    _buffers
    0.07
    _DISPLAY
    0.07
     ditch
    0.07
    rotch
    0.07
    /start
    0.06
    .Read
    0.06
    ایش
    0.06
     dressing
    0.06
    Act Density 0.000%

    No Known Activations