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

    Neuronpedia

    INDEX
    Explanations

    names of cities

    oai_token-act-pair · gpt-3.5-turbo

    instances of the letter "w"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    uate
    -0.70
     paraly
    -0.66
     mosqu
    -0.66
     distingu
    -0.65
     conscientious
    -0.64
     culp
    -0.63
     retri
    -0.61
     unpre
    -0.61
     puzz
    -0.61
    arial
    -0.61
    POSITIVE LOGITS
    elcome
    1.40
    itness
    1.38
    atts
    1.32
    isdom
    1.20
    ashington
    1.19
    restling
    1.19
    atcher
    1.18
    izard
    1.16
    addle
    1.12
    atson
    1.12
    Act Density 0.036%

    No Known Activations