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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating quantity or numerical relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     которая
    -0.48
     яке
    -0.47
    ècie
    -0.46
     яка
    -0.44
     która
    -0.42
     koja
    -0.41
     joka
    -0.36
     ktorá
    -0.36
     која
    -0.36
     ddelweddau
    -0.35
    POSITIVE LOGITS
     whom
    1.61
    whom
    1.26
     Whom
    1.08
    Whom
    1.02
     whome
    0.83
     kasarigan
    0.81
    rungsseite
    0.71
    ########.
    0.71
     duquel
    0.71
     WH
    0.69
    Act Density 0.216%

    No Known Activations