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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens related to collective or plural subjects from singular first-person tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.18
    1:0.19
    2:0.29
    3:0.08
    4:0.03
    5:0.02
    6:0.05
    7:0.13
    Negative Logits
    Portail
    -0.34
    moiselle
    -0.28
    vacy
    -0.27
    Tikang
    -0.27
     voltak
    -0.26
    XmlEnum
    -0.26
     staffers
    -0.26
    oredCriteria
    -0.25
    gge
    -0.25
     staffs
    -0.25
    POSITIVE LOGITS
     Physica
    0.31
    yarnpkg
    0.27
     PARSER
    0.27
    kloped
    0.27
    !*\
    0.26
    Referanser
    0.25
    醐
    0.24
    OSI
    0.24
    #
    0.23
    навли
    0.23
    Act Density 0.124%

    No Known Activations