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

    Neuronpedia

    INDEX
    Explanations

    phrases related to abstract concepts or emotions

    oai_token-act-pair · gpt-3.5-turbo

    expressions related to perception or understanding of experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    raphic
    -0.75
    sites
    -0.71
    astered
    -0.70
    annis
    -0.69
     helicop
    -0.68
    essee
    -0.64
    script
    -0.63
    skill
    -0.62
     silent
    -0.62
    aunting
    -0.62
    POSITIVE LOGITS
     sense
    0.98
    terday
    0.92
     Sense
    0.83
    sense
    0.81
    ibly
    0.80
    omatic
    0.78
    uitive
    0.76
    ively
    0.76
    ible
    0.75
    ibility
    0.74
    Act Density 0.014%

    No Known Activations