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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a purpose or intention, particularly using the word "to."

    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
     Advertisement
    -0.69
     Transcript
    -0.65
     drafts
    -0.64
    TION
    -0.62
     Required
    -0.61
    ahime
    -0.61
     disproportionate
    -0.60
    Already
    -0.60
     Description
    -0.60
     copied
    -0.60
    POSITIVE LOGITS
     celebrate
    1.15
    othy
    1.04
     teach
    0.95
     help
    0.94
    ilet
    0.92
     commemorate
    0.92
     defend
    0.92
     promote
    0.89
    ulhu
    0.89
    bernatorial
    0.88
    Act Density 0.144%

    No Known Activations