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

    Neuronpedia

    INDEX
    Explanations

    as to why/which/whether/what

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     There
    1.16
    ت
    1.13
    us
    1.07
    ل
    1.04
    т
    0.96
    νες
    0.94
    ()
    0.93
     These
    0.92
    ic
    0.91
    ر
    0.91
    POSITIVE LOGITS
    '
    1.10
    ի
    0.97
    も
    0.96
    ノ
    0.91
    of
    0.90
    インド
    0.88
    ’
    0.87
    くる
    0.84
    지
    0.82
    luğ
    0.81
    Act Density 0.006%

    No Known Activations