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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens with assignment operations from tokens denoting equality or comparisons

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.09
    2:0.08
    3:0.04
    4:0.09
    5:0.45
    6:0.03
    7:0.06
    Negative Logits
     Walkover
    -0.38
    ########.
    -0.34
    uxxxx
    -0.33
    esez
    -0.32
    ⓧ
    -0.31
    IonicModule
    -0.29
    IGI
    -0.29
     noDo
    -0.28
    thâu
    -0.28
    oa̍t
    -0.27
    POSITIVE LOGITS
     uParam
    0.34
     bParam
    0.32
     AssemblyTitle
    0.30
     fup
    0.28
    loroethene
    0.27
    ionage
    0.27
    THISDAY
    0.26
    __(/*!
    0.25
    umente
    0.25
    </h6>
    0.25
    Act Density 0.161%

    No Known Activations