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

    Neuronpedia

    INDEX
    Explanations

    attends to states or cities mentioned as locations from earlier tokens referring to individuals or groups associated with those locations

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.09
    1:0.11
    2:0.10
    3:0.06
    4:0.08
    5:0.05
    6:0.19
    7:0.28
    Negative Logits
    choenen
    -0.28
     الوطنيه
    -0.28
    hoeddwyd
    -0.28
    })`
    -0.26
    esticides
    -0.26
    quête
    -0.26
    expandindo
    -0.26
    dataIndex
    -0.26
    ihnachts
    -0.25
    χη
    -0.25
    POSITIVE LOGITS
    ştır
    0.34
    prnewswire
    0.31
    PyExc
    0.31
    pkins
    0.30
    AutoScaleMode
    0.30
    <bos>
    0.30
    works
    0.29
    FieldBuilder
    0.29
     herself
    0.29
     Trost
    0.29
    Act Density 0.044%

    No Known Activations