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

    Neuronpedia

    INDEX
    Explanations

    action words that signify the start of an event or narrative

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the beginning or initiation of events or stories

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    illard
    -0.74
    mented
    -0.72
    cedented
    -0.67
    mens
    -0.64
    ilit
    -0.64
    ivable
    -0.64
    ordinary
    -0.63
    aths
    -0.63
    athed
    -0.63
    arded
    -0.62
    POSITIVE LOGITS
     anew
    0.99
    nings
    0.77
     raining
    0.74
    here
    0.71
     airing
    0.70
    Ń·
    0.70
     BELOW
    0.69
    ablishment
    0.67
    ocrin
    0.67
    rss
    0.66
    Act Density 0.069%

    No Known Activations