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

    Neuronpedia

    INDEX
    Explanations

    text indicating something is full or filled with something

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe something being completely occupied or filled

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rf
    -0.75
    appro
    -0.66
    aji
    -0.66
    practice
    -0.66
    recogn
    -0.65
    conduct
    -0.65
    hawks
    -0.65
     Surveillance
    -0.62
    isation
    -0.62
    nature
    -0.61
    POSITIVE LOGITS
     vacancies
    0.79
     fills
    0.77
    ummer
    0.75
    ups
    0.75
     coffers
    0.72
    outs
    0.71
     buckets
    0.69
    udicrous
    0.68
     rooms
    0.68
    erton
    0.68
    Act Density 0.027%

    No Known Activations