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

    Neuronpedia

    INDEX
    Explanations

    instances of knocking or door-related actions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    chain
    -0.17
    åīĤ
    -0.17
    ipo
    -0.15
    amburger
    -0.14
    chains
    -0.14
    å¼
    -0.14
    _ABS
    -0.14
     sø
    -0.14
     Mec
    -0.14
     åĨĨ
    -0.14
    POSITIVE LOGITS
     door
    0.23
     Door
    0.21
    _door
    0.20
    -door
    0.20
    éĸĢ
    0.19
    éŨ
    0.19
    Door
    0.18
     unlocked
    0.17
     knock
    0.16
    door
    0.16
    Act Density 0.109%

    No Known Activations