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

    Neuronpedia

    INDEX
    Explanations

    abstract concepts

    np_max-act · gemini-2.0-flash

    The neuron activates on words denoting requirements or necessities (e.g. “needs,” “accessibility,” “needed”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    (conf
    -0.07
    	fclose
    -0.06
     рекоменду
    -0.06
     WLAN
    -0.06
    訴
    -0.06
    (student
    -0.06
    _release
    -0.06
    _One
    -0.06
     fclose
    -0.06
    ิหาร
    -0.06
    POSITIVE LOGITS
    Mailer
    0.06
    ẩ
    0.06
    peon
    0.06
     sit
    0.06
    ิทย
    0.06
     Derm
    0.06
     embark
    0.06
    الد
    0.06
    lol
    0.06
    uiten
    0.06
    Act Density 0.441%

    No Known Activations