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

    Neuronpedia

    INDEX
    Explanations

    hothead or shithead

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ０
    0.91
    ő
    0.75
    0
    0.73
    ?’
    0.72
    <h4>
    0.72
    খিক
    0.72
     불구하고
    0.71
    ’
    0.71
     ০
    0.71
    োর্
    0.68
    POSITIVE LOGITS
    ні
    1.30
    на
    1.19
    a
    1.02
    da
    1.01
    ات
    0.93
    to
    0.89
    น
    0.88
    s
    0.87
    in
    0.87
    та
    0.83
    Act Density 0.000%

    No Known Activations