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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens indicating causal relationships or conditions from tokens discussing enabling actions or elements

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.12
    1:0.11
    2:0.11
    3:0.38
    4:0.07
    5:0.02
    6:0.07
    7:0.08
    Negative Logits
    RenderAtEndOf
    -0.36
    PerformLayout
    -0.35
     overdo
    -0.34
     improvised
    -0.33
    Tikang
    -0.33
    opha
    -0.32
    はじめに
    -0.32
     cosmo
    -0.32
     thanking
    -0.32
     providence
    -0.32
    POSITIVE LOGITS
     TextInputType
    0.39
    matchCondition
    0.39
     propOrder
    0.37
     gyhoeddwyd
    0.35
     hObject
    0.34
    routeProvider
    0.34
     Grant
    0.31
     Ellis
    0.31
    ViewGroup
    0.31
    hadur
    0.30
    Act Density 1.106%

    No Known Activations