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

    Neuronpedia

    INDEX
    Explanations

    words related to forceful actions or processes, especially ones involving breaking

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to the action of breaking

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    affer
    -0.69
    imental
    -0.69
    erva
    -0.67
    gel
    -0.63
    iva
    -0.62
    ikk
    -0.61
    oka
    -0.61
    velength
    -0.61
    accompanied
    -0.60
    erala
    -0.60
    POSITIVE LOGITS
    neck
    1.07
    fast
    0.90
    breakers
    0.87
    down
    0.86
    points
    0.85
    break
    0.80
    downs
    0.79
    breaking
    0.77
    breaks
    0.77
     down
    0.76
    Act Density 0.028%

    No Known Activations