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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions that involve rolling

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "rolled" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     sqor
    -0.86
    iera
    -0.73
    rament
    -0.71
    ique
    -0.68
    emia
    -0.67
    Compan
    -0.67
    iques
    -0.66
    ITION
    -0.66
    Lens
    -0.65
    etta
    -0.65
    POSITIVE LOGITS
    abouts
    0.68
     Gutenberg
    0.67
    gil
    0.65
    terday
    0.65
    ournal
    0.64
     CTR
    0.63
    ndum
    0.63
    sie
    0.62
     Hearthstone
    0.62
    bill
    0.61
    Act Density 0.001%

    No Known Activations