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

    Neuronpedia

    INDEX
    Explanations

    phrases related to sudden or abrupt change or action

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating conditionality or potential action

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    untarily
    -0.74
    aced
    -0.73
    oreal
    -0.73
    ortment
    -0.73
    azel
    -0.72
    ugu
    -0.72
    throp
    -0.69
    aught
    -0.69
    ossier
    -0.68
    utsu
    -0.68
    POSITIVE LOGITS
     Gemini
    0.68
     Stupid
    0.68
    thin
    0.67
     righteousness
    0.65
     Axel
    0.65
     yawn
    0.63
     shovel
    0.62
     Heller
    0.61
     torches
    0.61
     Supporters
    0.60
    Act Density 0.923%

    No Known Activations