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

    Neuronpedia

    INDEX
    Explanations

    (

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of U.S. Navy ship designations (e.g. “USS …” plus hull classification and number).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Props
    -0.07
     neutrality
    -0.07
     encode
    -0.06
    match
    -0.06
     idx
    -0.06
    ίνη
    -0.06
    _sections
    -0.06
     Spinner
    -0.06
     şeyi
    -0.06
    女子
    -0.06
    POSITIVE LOGITS
     Func
    0.06
    .robot
    0.06
     »↵
    0.06
     сто
    0.06
     Üy
    0.06
    _PRESS
    0.06
    .setInt
    0.06
     bás
    0.06
    .rs
    0.06
     проведения
    0.06
    Act Density 0.008%

    No Known Activations