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

    Neuronpedia

    INDEX
    Explanations

    attends to the term "died" and related expressions of death or killing from tokens indicating the consequence of such actions

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.07
    1:0.11
    2:0.06
    3:0.07
    4:0.44
    5:0.05
    6:0.08
    7:0.08
    Negative Logits
     typelib
    -0.28
     inso
    -0.26
    extAlignment
    -0.25
    GEBURTSDATUM
    -0.25
     hura
    -0.23
     kand
    -0.23
    تل
    -0.23
    PostMapping
    -0.23
    InsertCommand
    -0.22
    abs
    -0.22
    POSITIVE LOGITS
    ynchronously
    0.37
    Amicalement
    0.37
     Dossier
    0.34
    󠁿
    0.34
    serializers
    0.33
     dossier
    0.31
     retrouvé
    0.31
    RAINT
    0.31
     exchangers
    0.31
     enkelt
    0.31
    Act Density 0.507%

    No Known Activations