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

    Neuronpedia

    INDEX
    Explanations

    references to skipping content or features in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eing
    -0.14
    麼
    -0.14
    /legal
    -0.14
    /ay
    -0.14
    ega
    -0.14
    ë¶Ħ
    -0.13
     upd
    -0.13
    ucht
    -0.13
     sân
    -0.13
    á»Ļ
    -0.13
    POSITIVE LOGITS
     ahead
    0.35
    ahead
    0.33
     Ahead
    0.30
    Ahead
    0.26
     skip
    0.25
     past
    0.24
     Skip
    0.24
    -ahead
    0.24
    cq
    0.22
    per
    0.22
    Act Density 0.034%

    No Known Activations