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

    Neuronpedia

    INDEX
    Explanations

    financial information such as monetary values and market figures

    oai_token-act-pair · gpt-3.5-turbo

    the word "with" indicating relationships or connections in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    hops
    -0.75
    manac
    -0.66
    fair
    -0.66
    kind
    -0.65
    stall
    -0.65
    soft
    -0.64
    hack
    -0.63
    sburg
    -0.63
    chuk
    -0.61
    bus
    -0.60
    POSITIVE LOGITS
     regard
    1.46
     regards
    1.43
    stood
    1.27
     respect
    1.12
    draw
    1.05
    drawn
    1.03
    standing
    0.99
     impunity
    0.97
    holding
    0.81
     apologies
    0.71
    Act Density 0.199%

    No Known Activations