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

    Neuronpedia

    INDEX
    Explanations

    000

    np_max-act · gemini-2.0-flash

    This neuron detects internal document markup tokens that label sections or numbered positions (e.g. “sec-000,” numeric codes) in academic article source text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    :o
    -0.07
    oten
    -0.06
     ее
    -0.06
    erna
    -0.06
    LocalizedString
    -0.06
    .but
    -0.06
    \t
    -0.06
     Nx
    -0.06
    .oauth
    -0.06
     afraid
    -0.06
    POSITIVE LOGITS
    ulfill
    0.07
     addslashes
    0.07
    負
    0.07
     SHALL
    0.07
    sson
    0.07
     Assembly
    0.06
     Inline
    0.06
     стали
    0.06
     kans
    0.06
    UNKNOWN
    0.06
    Act Density 0.001%

    No Known Activations