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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     songs
    -0.07
    ervations
    -0.07
     marsh
    -0.07
     actionTypes
    -0.07
    .RED
    -0.07
     strapped
    -0.07
     rio
    -0.07
     jaw
    -0.07
    瘴
    -0.07
    /test
    -0.07
    POSITIVE LOGITS
    בק
    0.08
     mir
    0.07
    嗯
    0.07
    ysi
    0.07
    Override
    0.07
    orbit
    0.07
    @Override
    0.07
    薇
    0.07
    öff
    0.07
     Cover
    0.07
    Act Density 0.094%

    No Known Activations