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

    Neuronpedia

    INDEX
    Explanations

    directory trees

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Bennet
    0.88
     WUE
    0.83
     teks
    0.78
     Perry
    0.77
    Bennett
    0.76
    pek
    0.76
    図柄
    0.76
     kW
    0.76
     Fuel
    0.76
     texte
    0.76
    POSITIVE LOGITS
     directory
    1.32
     Directory
    1.24
    Directory
    1.23
     directories
    1.22
    ディレクトリ
    1.21
    directories
    1.13
     folder
    1.11
     folders
    1.11
    directory
    1.09
     parent
    1.09
    Act Density 1.119%

    No Known Activations