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

    Neuronpedia

    INDEX
    Explanations

    phrases beginning with "Unlike"

    oai_token-act-pair · gpt-3.5-turbo

    comparative phrases that highlight differences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    essen
    -0.77
    hiba
    -0.74
    anut
    -0.74
    adel
    -0.70
    gae
    -0.69
    iola
    -0.69
    idates
    -0.68
    eway
    -0.66
    oca
    -0.65
    ells
    -0.64
    POSITIVE LOGITS
    lihood
    1.43
    liest
    1.01
    ly
    0.89
     ours
    0.86
     minded
    0.86
    liness
    0.85
    lier
    0.80
    minded
    0.77
    entimes
    0.71
     ordinary
    0.70
    Act Density 0.012%

    No Known Activations