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

    Neuronpedia

    INDEX
    Explanations

    describing properties/usefulness

    np_max-act · gemini-2.0-flash

    The neuron fires on promotional or evaluative phrasing—particularly the “making it an ideal choice for…” style of superlative/comparative language.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     removeFromSuperview
    -0.07
     Fischer
    -0.06
    Response
    -0.06
    967
    -0.06
     panicked
    -0.06
    zych
    -0.06
    ’i
    -0.06
    throat
    -0.06
    junction
    -0.06
    Compar
    -0.06
    POSITIVE LOGITS
     Tags
    0.07
     STDERR
    0.07
    祭
    0.06
    '><
    0.06
    いつ
    0.06
    资
    0.06
     HOH
    0.06
    isay
    0.06
    도별
    0.06
     totalTime
    0.06
    Act Density 0.055%

    No Known Activations