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

    Neuronpedia

    INDEX
    Explanations

    presumed

    np_max-act · gemini-2.0-flash

    This neuron activates on words ending in the suffix “-ed.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     filling
    -0.07
    MainThread
    -0.06
     город
    -0.06
    系
    -0.06
    (vars
    -0.06
    eck
    -0.06
    _activation
    -0.06
     touches
    -0.06
     женщин
    -0.06
     Şehir
    -0.06
    POSITIVE LOGITS
    uras
    0.07
    ují
    0.06
    蔵
    0.06
    uckle
    0.06
    backs
    0.06
     случ
    0.06
     $#
    0.06
    뮤
    0.06
    ood
    0.06
    ADING
    0.06
    Act Density 0.433%

    No Known Activations