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

    Neuronpedia

    INDEX
    Explanations

    information related to news articles about individuals and their actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of commas in a text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    earable
    -0.67
     commitments
    -0.66
     showc
    -0.65
    ¬¼
    -0.65
    ords
    -0.64
    ights
    -0.62
    earch
    -0.62
     baskets
    -0.62
     executions
    -0.61
     buckets
    -0.61
    POSITIVE LOGITS
    who
    1.23
     whom
    1.17
    whose
    1.16
     who
    1.13
     whose
    1.03
     albeit
    0.81
     huh
    0.78
     namely
    0.76
     aka
    0.75
     nicknamed
    0.73
    Act Density 0.417%

    No Known Activations