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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "on."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     on
    -0.18
    owl
    -0.17
    oretical
    -0.15
    unami
    -0.14
    disposing
    -0.14
    quarters
    -0.14
    eki
    -0.14
    noon
    -0.14
    OnInit
    -0.14
    arness
    -0.14
    POSITIVE LOGITS
     behalf
    0.50
    /off
    0.30
    shore
    0.29
    -site
    0.27
     occasion
    0.27
    -line
    0.25
    eway
    0.24
    /by
    0.23
    etime
    0.23
    -demand
    0.22
    Act Density 0.507%

    No Known Activations