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

    Neuronpedia

    INDEX
    Explanations

    command-line commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    r
    0.61
    <unused1011>
    0.58
    <unused757>
    0.52
    <unused229>
    0.50
    <unused323>
    0.50
    <unused2019>
    0.50
    <unused573>
    0.50
    <unused670>
    0.50
    larni
    0.50
    <unused1057>
    0.50
    POSITIVE LOGITS
    Linux
    0.64
     Linux
    0.61
    시
    0.59
    कु
    0.57
    今日
    0.56
    0
    0.54
    OS
    0.54
    Привет
    0.53
    昨日
    0.51
    linux
    0.50
    Act Density 1.129%

    No Known Activations