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

    Neuronpedia

    INDEX
    Explanations

    details or components being broken down, likely related to explanations or analyses

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "breakdown" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Shar
    -0.77
    eno
    -0.74
    nery
    -0.72
    adena
    -0.70
    alty
    -0.70
    ittee
    -0.69
    qua
    -0.67
    ñ
    -0.65
    gged
    -0.65
    adding
    -0.64
    POSITIVE LOGITS
     breakdown
    0.91
    sie
    0.91
    DOWN
    0.87
    s
    0.86
    neck
    0.72
    opian
    0.71
     conversions
    0.68
    schild
    0.66
    urst
    0.66
    alore
    0.65
    Act Density 0.013%

    No Known Activations