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

    Neuronpedia

    INDEX
    Explanations

    numerical values related to performance metrics

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Kris
    -0.17
     _↵
    -0.16
    áy
    -0.15
     exc
    -0.15
    ary
    -0.14
    azar
    -0.14
    ugi
    -0.14
    regation
    -0.14
    ú
    -0.14
     Thr
    -0.14
    POSITIVE LOGITS
    istrovstvÃŃ
    0.17
    gons
    0.15
    presso
    0.15
    ambi
    0.14
     sounding
    0.14
    åħ¸
    0.14
    agina
    0.13
    phins
    0.13
    amat
    0.13
    /misc
    0.13
    Act Density 0.100%

    No Known Activations