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

    Neuronpedia

    INDEX
    Explanations

    specific phrases referring to the reason or cause of a statement

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "that" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    venants
    -0.85
    atin
    -0.67
    ledged
    -0.67
    marks
    -0.66
    amaru
    -0.65
    gd
    -0.65
    println
    -0.64
    mates
    -0.63
    arde
    -0.62
    books
    -0.61
    POSITIVE LOGITS
     sake
    1.58
     reason
    1.37
     purpose
    1.30
     purposes
    1.25
     reasons
    1.09
     matter
    0.94
     occasion
    0.90
    é»Ĵ
    0.80
     alone
    0.78
     foreseeable
    0.77
    Act Density 0.065%

    No Known Activations