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

    Neuronpedia

    INDEX
    Explanations

    attends to specific protocol tokens from generic tokens indicating types or categories

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.08
    1:0.10
    2:0.10
    3:0.13
    4:0.11
    5:0.07
    6:0.19
    7:0.17
    Negative Logits
    CPtr
    -0.28
    audiovisuel
    -0.26
    зулта
    -0.24
    XmlAccessorType
    -0.23
    AddTagHelper
    -0.23
     CanadaChoose
    -0.23
    bootstrapcdn
    -0.23
    isante
    -0.22
    oubted
    -0.22
    やし
    -0.22
    POSITIVE LOGITS
    tagHelperRunner
    0.40
    FormTagHelper
    0.38
    évaluateur
    0.38
     ainfi
    0.36
     myſelf
    0.35
    interopRequire
    0.35
    onOptions
    0.34
     enfans
    0.33
     reaſon
    0.32
     ſen
    0.32
    Act Density 0.051%

    No Known Activations