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

    Neuronpedia

    INDEX
    Explanations

    entering through doors

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     befind
    -0.08
     rape
    -0.08
     alli
    -0.08
     recoge
    -0.08
     embryo
    -0.08
    ետ
    -0.08
     recoger
    -0.08
     اِ
    -0.07
     υπό
    -0.07
     deme
    -0.07
    POSITIVE LOGITS
     двери
    0.10
     accueill
    0.10
     തുറ
    0.10
     двер
    0.09
    门
    0.09
    .open
    0.09
     ingresso
    0.09
     doors
    0.09
     дверь
    0.09
     отверст
    0.09
    Act Density 0.026%

    No Known Activations