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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a purpose or intention to do something specific

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating purpose or intention

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Leaves
    -0.64
     Sources
    -0.64
     Classification
    -0.64
    Ī
    -0.62
     Width
    -0.61
     airs
    -0.60
    CNN
    -0.60
    Already
    -0.59
     NCT
    -0.59
     Accounts
    -0.58
    POSITIVE LOGITS
     celebrate
    0.99
    brate
    0.92
     defend
    0.87
     conserve
    0.85
     help
    0.83
     promote
    0.82
     nurture
    0.80
     assist
    0.79
     uphold
    0.79
    bernatorial
    0.78
    Act Density 0.133%

    No Known Activations