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

    Neuronpedia

    INDEX
    Explanations

    SA

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the “EFSA” acronym and related structured header/citation tokens in EFSA scientific‐opinion documents.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     veteran
    -0.07
    λία
    -0.07
    。↵↵↵↵
    -0.07
    $str
    -0.07
    �
    -0.06
    _units
    -0.06
    ()(
    -0.06
     тран
    -0.06
    .usuario
    -0.06
     축구
    -0.06
    POSITIVE LOGITS
    thic
    0.08
    apid
    0.07
    ，此
    0.07
     authoritarian
    0.06
    avenous
    0.06
     Tradable
    0.06
    Passed
    0.06
    fruit
    0.06
    quee
    0.06
     Erk
    0.06
    Act Density 0.003%

    No Known Activations