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

    Neuronpedia

    INDEX
    Explanations

    menu options in code

    np_max-act · gemini-2.0-flash

    The neuron activates on numeric menu‐option markers (digits followed by punctuation) in console/menu text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cassette
    -0.07
    оп
    -0.07
     cũng
    -0.07
    _die
    -0.07
     also
    -0.07
    ampoo
    -0.06
     bip
    -0.06
     �
    -0.06
     mon
    -0.06
     Mang
    -0.06
    POSITIVE LOGITS
    τύ
    0.06
     nicely
    0.06
    judul
    0.06
     yaşam
    0.06
     Суд
    0.06
    .et
    0.06
    .command
    0.06
    "],["
    0.06
    ball
    0.06
    .Players
    0.06
    Act Density 0.015%

    No Known Activations