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

    Neuronpedia

    INDEX
    Explanations

    the adjective "easy."

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "easy"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    eters
    -0.77
    grave
    -0.73
    raints
    -0.73
    rongh
    -0.72
    hips
    -0.67
     Saud
    -0.62
    orf
    -0.62
     strongly
    -0.62
     Buckingham
    -0.62
    mut
    -0.62
    POSITIVE LOGITS
    Jet
    1.14
    going
    0.95
     prey
    0.76
    coded
    0.73
    Recipe
    0.72
    answer
    0.69
    idious
    0.69
    azon
    0.69
    step
    0.69
    jet
    0.68
    Act Density 0.056%

    No Known Activations