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

    Neuronpedia

    INDEX
    Explanations

    system commands, mount, sudo

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     macOS
    0.46
     MacOS
    0.43
    风格
    0.42
    WINDOWS
    0.42
    ʢ
    0.41
    backslash
    0.40
    ␊
    0.40
     OSX
    0.40
    ুঁ
    0.40
    𝒈
    0.39
    POSITIVE LOGITS
     mod
    0.58
     Mod
    0.47
     Mount
    0.45
    bmod
    0.44
    mod
    0.44
    Mount
    0.44
    Mod
    0.42
     Mods
    0.41
     mods
    0.41
     um
    0.40
    Act Density 0.019%

    No Known Activations