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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating clarification or explanation

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "clarify" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    MET
    -0.75
    pes
    -0.67
    die
    -0.66
    aden
    -0.65
    cgi
    -0.65
    ren
    -0.64
    jay
    -0.63
    @#&
    -0.63
    atri
    -0.63
    ADS
    -0.62
    POSITIVE LOGITS
     clar
    1.08
     clarify
    0.97
     clarified
    0.92
     clarification
    0.90
     wording
    0.85
    ifications
    0.82
    ifying
    0.79
    ifies
    0.79
     deline
    0.75
     misunderstand
    0.72
    Act Density 0.022%

    No Known Activations