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

    Neuronpedia

    INDEX
    Explanations

    comparative terms expressing superiority

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating the concept of "best" or optimal choices

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    idon
    -0.74
    onut
    -0.73
     Dru
    -0.69
    chy
    -0.67
     Reloaded
    -0.66
     Emer
    -0.65
     Smy
    -0.62
     Manson
    -0.62
     probing
    -0.62
    oidal
    -0.62
    POSITIVE LOGITS
    seller
    1.14
    iary
    1.06
    iaries
    1.03
     suited
    1.00
    sell
    0.97
    ow
    0.94
    imates
    0.87
    rade
    0.86
    ower
    0.84
    hest
    0.79
    Act Density 0.039%

    No Known Activations