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

    Neuronpedia

    INDEX
    Explanations

    attends to objects or entities being referred to from a corresponding pronoun or descriptor that points back to them

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.11
    1:0.20
    2:0.15
    3:0.11
    4:0.09
    5:0.07
    6:0.06
    7:0.18
    Negative Logits
    aarrggbb
    -0.54
    NameInMap
    -0.52
     AssemblyTitle
    -0.46
     ProtoMessage
    -0.46
    verwijspagina
    -0.45
    minecraft
    -0.41
    principalTable
    -0.41
    parsedMessage
    -0.40
    ukone
    -0.40
    getStatusCode
    -0.40
    POSITIVE LOGITS
    帖最后由
    0.39
     genel
    0.36
    ēl
    0.35
    il
    0.33
    AxisAlignment
    0.32
     yeast
    0.32
    Šalt
    0.30
     saluti
    0.30
     Outros
    0.30
    IL
    0.30
    Act Density 0.133%

    No Known Activations