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

    Neuronpedia

    INDEX
    Explanations

    repeated phrases or themes referred to as mantras

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gne
    -0.08
    ÙĪÙĨد
    -0.07
     arity
    -0.07
    tae
    -0.07
    efs
    -0.07
    ors
    -0.06
    ppard
    -0.06
    zew
    -0.06
    ahy
    -0.06
     Hyp
    -0.06
    POSITIVE LOGITS
    ingleton
    0.07
    -io
    0.06
    plusplus
    0.06
    ãĤ¤ãĥĪ
    0.06
    ict
    0.06
    /theme
    0.06
    ãģıãĤĭ
    0.06
    ufen
    0.06
    .dict
    0.06
    atic
    0.06
    Act Density 0.004%

    No Known Activations