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

    Neuronpedia

    INDEX
    Explanations

    statements indicating refusal to comment or share information

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "declined."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    è¦ļéĨĴ
    -0.73
    Ranked
    -0.71
    Offline
    -0.71
    ohyd
    -0.69
    ocratic
    -0.68
    =-=-=-=-=-=-=-=-
    -0.68
    agate
    -0.67
    facts
    -0.67
    stones
    -0.66
    types
    -0.65
    POSITIVE LOGITS
     Krug
    0.74
     miser
    0.73
     unanimously
    0.67
    CLASSIFIED
    0.65
     declines
    0.65
     admission
    0.64
     warr
    0.64
     hes
    0.63
     confir
    0.63
     sharply
    0.62
    Act Density 0.016%

    No Known Activations