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

    Neuronpedia

    INDEX
    Explanations

    occurrences of cancellation actions or commands

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     houſe
    -1.16
     Diſ
    -1.11
     Jefus
    -1.11
     Efq
    -1.11
     Chriftian
    -1.09
     faſt
    -1.08
     Houſe
    -1.08
     ſtate
    -1.07
     Inſ
    -1.07
     ſmall
    -1.06
    POSITIVE LOGITS
     em
    0.68
    ↵
    0.64
    }
    0.61
    ers
    0.60
     är
    0.57
    ↵↵
    0.56
    ,
    0.54
    bounds
    0.54
     ""
    0.54
    {}
    0.54
    Act Density 0.104%

    No Known Activations