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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "admit" and its variations, highlighting themes of acknowledgment and confession

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    olding
    -0.15
    ìĶ
    -0.15
    blo
    -0.15
     Niet
    -0.15
    olds
    -0.14
    pj
    -0.14
    .shutdown
    -0.14
    fu
    -0.14
    ież
    -0.14
    lei
    -0.13
    POSITIVE LOGITS
     defeat
    0.30
     defeats
    0.18
     freely
    0.18
     defeated
    0.18
    ration
    0.17
    ting
    0.16
    thumb
    0.16
     to
    0.16
     feeling
    0.15
     responsibility
    0.15
    Act Density 0.023%

    No Known Activations