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

    Neuronpedia

    INDEX
    Explanations

    checksums and hashes

    np_max-act · gemini-2.0-flash

    The neuron is tuned to spot long sequences of hexadecimal digits typical of cryptographic hash values.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    c
    -0.08
    cx
    -0.07
    W
    -0.07
    an
    -0.06
     Dark
    -0.06
    /scripts
    -0.06
    &o
    -0.06
     cans
    -0.06
    _kwargs
    -0.06
    ’яз
    -0.06
    POSITIVE LOGITS
     GUIDATA
    0.07
    !=-
    0.07
    ительные
    0.07
    ublik
    0.06
    图
    0.06
     تأثیر
    0.06
    (control
    0.06
     наступ
    0.06
    ejs
    0.06
    (typ
    0.06
    Act Density 0.022%

    No Known Activations