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

    Neuronpedia

    INDEX
    Explanations

    password hashing/encryption

    np_max-act · gemini-2.0-flash

    The neuron activates on any occurrence of the substring “secure” (including variants like “insecure” or “INSECURE”) in tokens.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Twelve
    -0.06
    .goto
    -0.06
    -treated
    -0.06
     Chim
    -0.06
    Ult
    -0.06
    umont
    -0.06
    eron
    -0.06
     arteries
    -0.06
    Declared
    -0.06
    -*
    -0.06
    POSITIVE LOGITS
     국민
    0.07
    umb
    0.06
     Poss
    0.06
    _Load
    0.06
    ewn
    0.06
     Thom
    0.06
     Garcia
    0.06
     worries
    0.06
    ATER
    0.06
     obsess
    0.06
    Act Density 0.002%

    No Known Activations