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

    Neuronpedia

    INDEX
    Explanations

    phrases related to the importance of specific details or characteristics

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating significant emotional or social themes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    igslist
    -0.69
    jri
    -0.68
    eln
    -0.66
    Specifically
    -0.65
    rin
    -0.64
    Ö¼
    -0.64
     Previously
    -0.64
    bis
    -0.64
    bish
    -0.63
    querque
    -0.62
    POSITIVE LOGITS
     whichever
    1.13
     ALWAYS
    0.86
     whoever
    0.85
    whatever
    0.85
     whatever
    0.84
     regardless
    0.83
     whether
    0.79
    depending
    0.78
     invariably
    0.76
     Regardless
    0.75
    Act Density 0.305%

    No Known Activations