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

    Neuronpedia

    INDEX
    Explanations

    passwords and admin credentials

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    :=
    0.39
    ---
    0.38
    Tokenizer
    0.37
     Requires
    0.37
    iert
    0.36
     angesch
    0.36
    另一方面
    0.35
     ঘণ্টায়
    0.35
    Import
    0.35
    IRC
    0.35
    POSITIVE LOGITS
    admin
    0.67
     admin
    0.55
     abc
    0.51
    chang
    0.51
    abcdef
    0.50
     berubah
    0.48
    abcd
    0.47
    change
    0.47
    abc
    0.47
    ADMIN
    0.46
    Act Density 0.025%

    No Known Activations