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

    Neuronpedia

    INDEX
    Explanations

    phrases and words that highlight uniqueness or originality

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    olella
    -0.55
     cser
    -0.53
    pherds
    -0.51
    utilisons
    -0.51
     Scrolls
    -0.49
    图源
    -0.48
    IOL
    -0.47
     pitä
    -0.47
    harusnya
    -0.47
     нат
    -0.47
    POSITIVE LOGITS
     uniqu
    0.96
     Unique
    0.95
    unique
    0.94
    Unique
    0.94
     UNIQUE
    0.94
     unique
    0.94
     uniques
    0.90
    UNIQUE
    0.84
     uniquely
    0.82
     unik
    0.81
    Act Density 0.073%

    No Known Activations