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

    Neuronpedia

    INDEX
    Explanations

    first item or lesson

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ,
    1.08
     or
    0.97
     and
    0.92
     other
    0.88
     various
    0.88
     different
    0.81
     the
    0.81
     also
    0.80
    ↵↵
    0.80
    ↵
    0.77
    POSITIVE LOGITS
     Öncelikle
    1.01
    首先
    1.00
     首先
    0.99
     Özellikle
    0.97
     Loại
    0.96
    Firstly
    0.96
    ⇰
    0.94
     먼저
    0.93
    ěru
    0.93
    Cliente
    0.92
    Act Density 4.047%

    No Known Activations