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

    Neuronpedia

    INDEX
    Explanations

    phrases related to instructional or educational content

    oai_token-act-pair · gpt-3.5-turbo

    instructions or phrases starting with "How to."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Coff
    -0.74
    prem
    -0.71
    court
    -0.69
     Stanley
    -0.65
    listed
    -0.65
    lights
    -0.65
    pict
    -0.63
    years
    -0.63
     Appears
    -0.62
    cart
    -0.60
    POSITIVE LOGITS
     navigate
    1.19
     proceed
    1.19
     behave
    1.12
     interpret
    1.12
     maximize
    1.09
     treat
    1.09
     communicate
    1.09
     achieve
    1.08
     incorporate
    1.08
     handle
    1.08
    Act Density 0.045%

    No Known Activations