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

    Neuronpedia

    INDEX
    Explanations

    require

    np_max-act · gemini-2.0-flash

    The neuron activates on verbs and phrases that signal legal or regulatory obligations (e.g. “required to,” “requires,” “are required”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _cred
    -0.07
     tainted
    -0.07
    _inicio
    -0.06
    โลย
    -0.06
     gravy
    -0.06
     rigged
    -0.06
    емого
    -0.06
     ал
    -0.06
     declining
    -0.06
    लब
    -0.06
    POSITIVE LOGITS
    WindowSize
    0.07
     先
    0.06
     varlık
    0.06
     SAF
    0.06
     genitals
    0.06
     ORIGINAL
    0.06
     앞
    0.06
    !’
    0.06
    [__
    0.06
     maxY
    0.06
    Act Density 0.024%

    No Known Activations