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

    Neuronpedia

    INDEX
    Explanations

    attends to conditional phrases from preceding tokens that anticipate or suggest a scenario based on earlier actions or conditions

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.10
    2:0.13
    3:0.07
    4:0.04
    5:0.02
    6:0.11
    7:0.40
    Negative Logits
    :]:
    -0.31
     crossorigin
    -0.29
    protoimpl
    -0.28
    Referanser
    -0.25
    })->
    -0.25
     ineff
    -0.24
    //
    -0.24
     tille
    -0.23
    ApiException
    -0.22
    HasBeenSet
    -0.22
    POSITIVE LOGITS
    +#+#
    0.41
     ujednoznacz
    0.35
     instead
    0.30
    زيون
    0.30
    ebenarnya
    0.30
    útbol
    0.29
    もっと
    0.29
    ברס
    0.29
    ênis
    0.28
    __':
    0.28
    Act Density 0.729%

    No Known Activations