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

    Neuronpedia

    INDEX
    Explanations

    command-line and Command Injection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     KQkq
    0.44
     sama
    0.42
    quirrel
    0.41
    IAN
    0.39
     animales
    0.39
     empresa
    0.39
    IA
    0.38
    доз
    0.38
     Hemat
    0.38
     الشركة
    0.38
    POSITIVE LOGITS
    itaires
    0.65
     命令
    0.57
     cmd
    0.52
    コマンド
    0.52
    命令
    0.52
     comand
    0.52
     Commands
    0.52
     commanded
    0.50
     CMD
    0.50
     Cmd
    0.50
    Act Density 0.015%

    No Known Activations