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

    Neuronpedia

    INDEX
    Explanations

    phrases related to generating ideas or solutions to problems

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the process of generating or creating something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    clair
    -0.82
    usa
    -0.82
    externalActionCode
    -0.77
    tenance
    -0.70
    PRESS
    -0.69
    icio
    -0.68
    ersion
    -0.68
    ISTER
    -0.68
    thood
    -0.68
    read
    -0.67
    POSITIVE LOGITS
     excuses
    1.48
     ingenious
    1.24
     catchy
    1.20
     solutions
    1.16
     ideas
    1.13
     ways
    1.12
     witty
    1.09
     inventive
    1.09
     hypotheses
    1.08
     excuse
    1.03
    Act Density 0.120%

    No Known Activations