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

    Neuronpedia

    INDEX
    Explanations

    the word "Att" at different activation levels

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "Att" or related terms typically associated with attention or attachment

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    theless
    -0.90
    姫
    -0.85
    enegger
    -0.84
    FTWARE
    -0.83
    assetsadobe
    -0.78
     ãĤµãĥ¼ãĥĨãĤ£ãĥ¯ãĥ³
    -0.78
    ktop
    -0.78
    76561
    -0.77
     Giuliani
    -0.76
    REDACTED
    -0.75
    POSITIVE LOGITS
    anooga
    1.18
    itudes
    1.17
    ributed
    1.16
    ributes
    1.13
    ribute
    1.12
    ainment
    1.11
    achment
    1.09
    itude
    1.08
    ension
    1.07
    ention
    1.03
    Act Density 0.006%

    No Known Activations