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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "correct" and related expressions indicating accuracy or validity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    udd
    -0.18
    /desktop
    -0.17
    icap
    -0.15
    loub
    -0.15
    PHY
    -0.15
    holding
    -0.15
    olt
    -0.15
    Ŀ
    -0.15
    ropping
    -0.15
    ary
    -0.15
    POSITIVE LOGITS
    zza
    0.21
    ives
    0.18
    IVES
    0.16
    itude
    0.15
    mente
    0.15
    fully
    0.15
    backs
    0.14
    respond
    0.14
    _runtime
    0.14
     Chúa
    0.14
    Act Density 0.030%

    No Known Activations