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

    Neuronpedia

    INDEX
    Explanations

    sensitive

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    surface
    -0.08
    -0.07
     bitmap
    -0.07
     사실
    -0.07
     tie
    -0.07
     миров
    -0.07
    José
    -0.07
     impossible
    -0.07
     obrá
    -0.07
     पहले
    -0.07
    POSITIVE LOGITS
     banget
    0.10
     للغاية
    0.09
    舒
    0.08
     souhaite
    0.08
     kokoa
    0.08
     irritated
    0.08
     Strauss
    0.08
    人士
    0.08
    =password
    0.08
     preferred
    0.08
    Act Density 0.004%
    No Known Activations