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

    Neuronpedia

    INDEX
    Explanations

    systemic issues, barriers, failures

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    يل
    1.03
    quela
    1.02
    زيد
    0.99
     İlç
    0.98
    其他
    0.97
    ير
    0.95
     Це
    0.93
    ב
    0.93
    زي
    0.92
     Ри
    0.91
    POSITIVE LOGITS
    )
    1.16
    ]
    0.96
    {
    0.92
    =
    0.90
    .
    0.89
    0.86
    <0xBF>
    0.85
    5
    0.85
    ak
    0.82
    _
    0.80
    Act Density 0.199%

    No Known Activations