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

    Neuronpedia

    INDEX
    Explanations

    instances of emphasis on specific points or ideas, often starting with the phrase "the only thing"

    oai_token-act-pair · gpt-3.5-turbo

    repeated phrases emphasizing the importance of specific actions or concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    inav
    -0.86
    gaard
    -0.71
    choes
    -0.70
    ilings
    -0.69
    onz
    -0.68
    ñ
    -0.67
     largeDownload
    -0.67
    aeper
    -0.66
    ONSORED
    -0.65
    cul
    -0.63
    POSITIVE LOGITS
     happens
    1.04
     happened
    1.00
     Valiant
    0.97
     happening
    0.88
    iverse
    0.84
     happ
    0.81
     transpired
    0.81
     happen
    0.79
     undone
    0.74
     bothers
    0.71
    Act Density 0.041%

    No Known Activations