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

    Neuronpedia

    INDEX
    Explanations

    Okay, let's break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     숫자
    1.86
    Aslamualaikum
    1.85
    Ⲙ
    1.85
    <unused1940>
    1.83
    十伍章
    1.83
    HTree
    1.83
     threx
    1.82
    wallepics
    1.82
    十肆章
    1.81
    dwm
    1.81
    POSITIVE LOGITS
     both
    1.20
     these
    0.96
     the
    0.87
    0.87
    -
    0.86
     them
    0.85
    both
    0.82
     above
    0.81
     even
    0.81
     just
    0.79
    Act Density 0.183%

    No Known Activations