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

    Neuronpedia

    INDEX
    Explanations

    instances where actions are being directed towards a specific recipient, often with a call for support or sharing

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and phrases that suggest collective action or shared experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Ahead
    -0.70
    smanship
    -0.65
    iren
    -0.65
    ourse
    -0.59
     flagship
    -0.59
    raph
    -0.58
    duc
    -0.58
     ahead
    -0.57
    rup
    -0.57
    estate
    -0.56
    POSITIVE LOGITS
     selves
    0.79
     Templ
    0.72
    vous
    0.71
    selves
    0.69
    azy
    0.68
    imaru
    0.66
     mates
    0.63
    otent
    0.62
    ortmund
    0.62
     THEN
    0.61
    Act Density 0.527%

    No Known Activations