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

    Neuronpedia

    INDEX
    Explanations

    terms

    np_max-act · gemini-2.0-flash

    This neuron responds to formal contract or agreement vocabulary (e.g. “terms,” “conditions,” “contract,” “commission”), i.e. legal‐style clause language.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     onPause
    -0.07
     amplified
    -0.06
    贵
    -0.06
    }{$
    -0.06
    play
    -0.06
     приступ
    -0.06
     افز
    -0.06
    .the
    -0.06
     پروژه
    -0.06
    .par
    -0.06
    POSITIVE LOGITS
     terms
    0.09
    Terms
    0.07
     Terms
    0.07
     serializers
    0.07
     TERMS
    0.07
     haystack
    0.06
    setScale
    0.06
    verse
    0.06
     оформ
    0.06
     Coordinates
    0.06
    Act Density 0.010%

    No Known Activations