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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens related to a specific concept or context from tokens of general or technical descriptors

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.04
    1:0.10
    2:0.05
    3:0.04
    4:0.20
    5:0.47
    6:0.03
    7:0.04
    Negative Logits
    InjectAttribute
    -0.40
    Hochspringen
    -0.39
    ագրություններ
    -0.38
    FormTagHelper
    -0.37
     يتيمه
    -0.35
    EndInit
    -0.35
    出版年
    -0.35
     发表于
    -0.34
    󠁴
    -0.33
    httphttps
    -0.33
    POSITIVE LOGITS
    UnknownFieldSet
    0.25
     Católica
    0.23
    classnames
    0.23
     peup
    0.21
     Langen
    0.20
    TestRunner
    0.20
    openzeppelin
    0.19
     voler
    0.19
     Bewer
    0.19
     vuo
    0.19
    Act Density 1.636%

    No Known Activations