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

    Neuronpedia

    INDEX
    Explanations

    Text extraction

    np_max-act · gemini-2.0-flash

    This neuron detects text listing awards (especially film/TV categories) and their winners or nominees.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Arcade
    -0.07
    manent
    -0.07
    Des
    -0.06
     inappropriate
    -0.06
    iding
    -0.06
    fund
    -0.06
     verdict
    -0.06
    .Match
    -0.06
     Kang
    -0.06
     diving
    -0.06
    POSITIVE LOGITS
    _HISTORY
    0.07
     hudeb
    0.07
    _fe
    0.06
    istinguished
    0.06
    ู้
    0.06
    ΑΓ
    0.06
    udad
    0.06
     italiani
    0.06
    щё
    0.06
     mediaPlayer
    0.06
    Act Density 0.013%

    No Known Activations