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

    Neuronpedia

    INDEX
    Explanations

    phrases related to conformity or adherence to rules, guidelines, or instructions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "according to" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    estern
    -0.72
    utters
    -0.70
    apons
    -0.69
    apo
    -0.68
    IFT
    -0.65
    blast
    -0.65
    adows
    -0.63
    kb
    -0.62
    boa
    -0.62
    alone
    -0.62
    POSITIVE LOGITS
    Ĥİ
    0.79
    edIn
    0.75
    chwitz
    0.71
    tains
    0.70
    Style
    0.70
    format
    0.69
    graded
    0.67
    soType
    0.66
    ortment
    0.66
    Rank
    0.66
    Act Density 0.045%

    No Known Activations