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

    Neuronpedia

    INDEX
    Explanations

    code structures and URLs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <h3>
    -1.38
    subsection
    -0.80
    ┻
    -0.74
    ")[
    -0.73
    ')),
    -0.73
    concerned
    -0.72
    уме
    -0.68
    ###
    -0.68
    "}},
    -0.68
    ')[
    -0.67
    POSITIVE LOGITS
    ")]
    1.09
    )]
    1.03
    0.92
    ')]
    0.91
    ")]
    0.85
    }]
    0.82
    XR
    0.81
    )]
    0.74
    )]$
    0.73
    俱乐部
    0.72
    Act Density 0.137%

    No Known Activations