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

    Neuronpedia

    INDEX
    Explanations

    expressions of goodwill and positive wishes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hev
    -0.15
    zar
    -0.14
     Certain
    -0.14
    adem
    -0.13
    hone
    -0.13
    oster
    -0.13
    Certain
    -0.13
    ене
    -0.13
    )?.
    -0.13
     --
    -0.13
    POSITIVE LOGITS
     wherever
    0.25
    ainment
    0.17
    inize
    0.17
     whatever
    0.17
     wishes
    0.16
    .Safe
    0.16
     safe
    0.15
    ckill
    0.15
     always
    0.15
     santé
    0.15
    Act Density 0.067%

    No Known Activations