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

    Neuronpedia

    INDEX
    Explanations

    attends to the conjunctions "or" from tokens that clarify or question the preceding context

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.14
    2:0.13
    3:0.06
    4:0.08
    5:0.01
    6:0.06
    7:0.37
    Negative Logits
    Geplaatst
    -0.30
     onBackPressed
    -0.28
    Erreferentziak
    -0.28
    GraphicsUnit
    -0.28
     onOptions
    -0.27
     {!
    -0.27
    хьтан
    -0.27
     урна
    -0.27
     createState
    -0.27
    Билгалдахарш
    -0.27
    POSITIVE LOGITS
    duff
    0.29
    <h6>
    0.28
     luôn
    0.26
    TintMode
    0.26
    gouv
    0.26
     بتاريخ
    0.25
    ApiProperty
    0.24
    mbe
    0.24
    CardBody
    0.24
    lovakia
    0.24
    Act Density 0.145%

    No Known Activations