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

    Neuronpedia

    INDEX
    Explanations

    speculative, uncertain predictions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵↵
    0.50
     back
    0.48
     and
    0.48
     he
    0.47
     guaranteed
    0.46
     on
    0.45
     ح
    0.45
    /
    0.45
     his
    0.45
     heritage
    0.45
    POSITIVE LOGITS
    თვის
    0.54
     Daftar
    0.53
     यावेळी
    0.53
     സെക്രട്ടറി
    0.52
    kker
    0.52
    )،
    0.51
     Partei
    0.51
    ”।
    0.50
    ambahkan
    0.50
    ürdig
    0.50
    Act Density 0.003%

    No Known Activations