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

    Neuronpedia

    INDEX
    Explanations

    words related to visual differences or comparisons

    oai_token-act-pair · gpt-3.5-turbo

    phrases that describe differences or comparisons between elements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Mos
    -0.73
    authorized
    -0.69
    aye
    -0.67
    reditary
    -0.66
    ackers
    -0.66
    rake
    -0.65
    Bay
    -0.64
    aptic
    -0.62
    bey
    -0.61
    kn
    -0.61
    POSITIVE LOGITS
     juxtap
    0.91
     contrasts
    0.81
     contrasting
    0.79
    xual
    0.77
     stark
    0.77
    ibilities
    0.74
    ĸļ
    0.72
     Contrast
    0.72
    lihood
    0.72
     sexes
    0.71
    Act Density 0.012%

    No Known Activations