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

    Neuronpedia

    INDEX
    Explanations

    attends to tokens marked as "References" from various tokens with content preceding them

    oai_attention-head · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.19
    1:0.34
    2:0.09
    3:0.08
    4:0.08
    5:0.04
    6:0.05
    7:0.09
    Negative Logits
     "..\..\
    -0.66
     "..\..\..\
    -0.59
     AssemblyTitle
    -0.57
     CreateTagHelper
    -0.56
    ]='\
    -0.56
    SequentialGroup
    -0.55
     }}"></
    -0.54
    Datuak
    -0.54
    fjspx
    -0.53
    Viitteet
    -0.53
    POSITIVE LOGITS
    <eos>
    0.35
     at
    0.32
    hase
    0.30
    ↵↵↵
    0.29
    tritts
    0.28
     mismo
    0.27
    3
    0.27
    ~
    0.27
    </em>
    0.27
    тен
    0.27
    Act Density 0.085%

    No Known Activations