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

    Neuronpedia

    INDEX
    Explanations

    repeated instances of the word "sit," particularly in phrases indicating a position or action of sitting

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uolo
    -0.81
     Valerio
    -0.73
    ⋙
    -0.72
     neod
    -0.70
     modo
    -0.64
    略
    -0.62
     Kimberly
    -0.61
     Oni
    -0.60
    uey
    -0.60
    zao
    -0.60
    POSITIVE LOGITS
     sit
    1.51
     Sit
    1.47
     SIT
    1.43
    Sit
    1.43
     sits
    1.35
     sitting
    1.35
    SIT
    1.32
     Sitting
    1.32
    sit
    1.23
    Sitting
    1.22
    Act Density 0.049%

    No Known Activations