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

    Neuronpedia

    INDEX
    Explanations

    attends to various tokens that signify types, specifically focusing on specific pairs of tokens related to the law or events

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.15
    1:0.07
    2:0.51
    3:0.03
    4:0.03
    5:0.02
    6:0.09
    7:0.06
    Negative Logits
    Charm
    -0.32
    PANY
    -0.32
    NewUrlParser
    -0.31
     CreateTagHelper
    -0.30
    Portail
    -0.29
    organizzazione
    -0.29
    벳
    -0.29
     Gier
    -0.29
    valently
    -0.29
    Hunter
    -0.28
    POSITIVE LOGITS
    TintMode
    0.32
    clusal
    0.31
     TextAppearance
    0.30
     CWE
    0.30
    السكان
    0.29
    ItemBackground
    0.29
    Rhestr
    0.29
     *);
    0.29
     }));
    0.29
    ];
    0.28
    Act Density 0.057%

    No Known Activations