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

    Neuronpedia

    INDEX
    Explanations

    attends to "ul" tokens from non-specific tokens

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.16
    1:0.16
    2:0.08
    3:0.15
    4:0.10
    5:0.14
    6:0.07
    7:0.12
    Negative Logits
     '\\;'
    -0.51
    expandindo
    -0.46
     ProtoMessage
    -0.44
    Хьажоргаш
    -0.40
     "..\..\..\
    -0.40
     onAnimation
    -0.37
    UnusedPrivate
    -0.37
    Datuak
    -0.37
    BeginContext
    -0.36
    NameInMap
    -0.35
    POSITIVE LOGITS
     CURIAM
    0.37
    enumii
    0.36
     ujednoznacz
    0.35
    ̈́
    0.32
    積
    0.31
     isinstance
    0.31
     ngoài
    0.29
    积
    0.29
    segue
    0.29
     defStyleAttr
    0.28
    Act Density 0.023%

    No Known Activations